Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Confidential draft: not for distribution, dissemination, or publication 1 Evidence-based practice guideline: Disease-modifying Therapies for Multiple Sclerosis 2 Report of the Guideline Development, Dissemination, and Implementation Subcommittee 3 (GDDI) of the American Academy of Neurology 4 Alexander Rae-Grant, MD, FRCP(C)1; Gregory Day, MD, MSc, FRCPC2; Ruth Ann Marrie, 6 MD, PhD3; Alejandro Rabinstein, MD, FAAN4; Bruce Cree, MD, PhD, MCR5; Gary Gronseth, 7 MD, FAAN6; Michael Haboubi, DO7; June Halper, MSN, APN-C, MSCN, FAAN8; Jonathan P. 8 Hosey, MD, FAAN9; David E. Jones, MD10; Annette Langer-Gould, MD, PhD11; Robert Lisak, 9 FRCP (UK), FAAN, FANA12; Daniel Pelletier, MD13; Sonja Potrebic, MD, PhD, FAAN14; 10 Cynthia Sitcov15; Rick Sommers16; Julie Stachowiak, PhD17; Thomas S.D. Getchius18; Shannon 11 A. Merillat, MLIS19; Tamara Pringsheim, MD20 AF T 5 12 1. Cleveland Clinic, Cleveland, OH 14 2. Knight Alzheimer Disease Research Center, Washington University in St. Louis, Saint Louis, MO D 15 R 13 16 3. University of Manitoba, Winnipeg, Manitoba 17 4. Mayo Clinic, Rochester, MN 18 5. Multiple Sclerosis Center, University of California, San Francisco, CA 19 6. Kansas University Medical Center, Kansas City, KS 20 7. University of Louisville, Louisville, KY 21 8. Consortium of Multiple Sclerosis Centers, Hackensack, NJ 22 9. St. Luke’s University Health Network, Bethlehem, Pa 23 10. University of Virginia, Charlottesville, VA 1 Confidential draft: not for distribution, dissemination, or publication 1 11. Kaiser Permanente Southern California, Pasadena, CA 2 12. Consortium of Multiple Sclerosis Centers, Hackensack, NJ; Wayne State University, Detroit, MI 3 13. University of Southern California, Los Angeles, CA 5 14. Kaiser Permanente Southern California, Los Angeles, CA 6 15. National Multiple Sclerosis Society, Arlington, VA 7 16. National Multiple Sclerosis Society, New York, NY 8 17. National Multiple Sclerosis Society, Santa Fe, NM 9 18. American Academy of Neurology, Minneapolis, MN 10 19. American Academy of Neurology, Minneapolis, MN 11 20. University of Calgary, Calgary, Alberta AF T 4 12 Acknowledgement 14 The authors would like to acknowledge the North American Research Committee on Multiple 15 Sclerosis (NARCOMS) Registry for their assistance in administering an outcomes survey, the 16 results of which were included in this guideline. NARCOMS is supported in part by the 17 Consortium of MS Centers (CMSC) and its Foundation. D R 13 18 19 Address correspondence and reprint requests to 20 American Academy of Neurology: 21 [email protected] 22 Title character count: 115 23 Abstract word count: [xxx] 2 Confidential draft: not for distribution, dissemination, or publication 1 Manuscript word count: [XXXXX] 2 STUDY FUNDING This practice guideline protocol was developed with financial support from the American Academy of Neurology. Authors who serve as AAN subcommittee members or methodologists (A.R.-G., A.R., G.S.D., M.H., A.L.G., S.P., T.P.) were reimbursed by the AAN for expenses related to travel to subcommittee meetings where drafts of manuscripts were reviewed. All authors on the panel were reimbursed by the AAN for expenses related to travel to two in-person T meetings. AF 3 DISCLOSURE 5 Alexander Rae-Grant is a local PI for one multiple sclerosis (MS) treatment study in progressive 6 MS with Novartis and Biogen IDEC, for which he does not receive any intellectual property, 7 time, or financial reimbursement; receives royalties from two textbooks he has published, one on 8 neurology and one on MS; and organizes and receives honoraria for ground rounds and 9 neurology review courses. D R 4 10 11 Alejandro Rabinstein serves as an associate editor for the Neurocritical Care journal; receives 12 publishing royalties from Elsevier, Oxford, and UptoDate; and has served as an external safety 13 monitor for the trial ALIAS, and as a member of the event adjudication team for the PREVAIL 14 trial. His institution receives financial compensation from DJO Global. 15 3 Confidential draft: not for distribution, dissemination, or publication 1 Bruce Cree has received compensation for consulting from Abbvie, Biogen IDEC, EMD Serono, 2 MedImmune, Novartis, Genzyme/Sanofi Aventis, and Teva Neurosciences; has received 3 research support from Acorda, Avanir, Biogen IDEC, EMD Serono, Hoffman La Roche, 4 MedImmune, and Teva Neurosciences; has given expert testimony and prepared an affidavit for 5 medical malpractice cases (1 or 2 per year) within his area of expertise; and has acted as 6 consultant in a legal proceeding for Biogen IDEC and Acorda. 7 Gregory Day serves as the clinical director of the Anti-NMDA Receptor Encephalitis 9 Foundation, Inc. Canada (the Foundation is supported by private donor contributions) but does T 8 not receive financial reimbursement for his appointment or service; received the Future Leader in 11 Dementia award in October 2013 (including travel support to a Canadian conference for 12 dementia), sponsored by Pfizer, Canada; has received research support from the Weston Brain 13 Foundation, as a recipient of the Eugene M Johnson Jr. Weston Brain Institute Postdoctoral 14 Fellowship in Dementia Research; and holds stock in ANI Pharmaceuticals. R AF 10 D 15 16 Michael Haboubi has received travel reimbursement and honoraria for grand rounds 17 presentations in Madisonville, Kentucky. 18 19 June Halper has no relevant disclosures to report. 20 21 Jonathan P. Hosey has no relevant disclosures to report. 22 4 Confidential draft: not for distribution, dissemination, or publication 1 David E. Jones authored an MS Coalition consensus paper on disease-modifying therapies 2 (DMTs) in MS and served as co-chair of the advocacy committee of the Consortium of Multiple 3 Sclerosis Centers (CMSC), but did not receive financial compensation for either service; receives 4 institutional and personal compensation for consulting with Biogen IDEC, Genzyme, and 5 Novartis Pharmaceuticals; receives travel reimbursement from Biogen IDEC, Genzyme, CMSC, 6 and Multiple Sclerosis Association of America (MSAA) for travel to meetings; has received 7 honoraria for projects with the CMSC and MSAA; is employed by the University of Virginia; 8 and has given expert testimony on a malpractice case involving a patient with neurosarcoidosis. T 9 Annette Langer-Gould is involved in the National Multiple Sclerosis Society’s (NMSS) MS and 11 Pregnancy Study, and the MS Prevalence Working Group, but does not receive financial 12 compensation for her involvement; has received travel reimbursement from the Mayo Clinic, 13 Arizona, where she spoke as an invited speaker; from the Institute of Medicine, Washington, DC, 14 for participating on a study panel; and from the National Institute for Neurological Disorders and 15 Stroke (NINDS), to participate in a study section; receives research support from Biogen IDEC 16 for her service as a local PI for two clinical trials, from Hoffman LaRoche for her service as a 17 local PI for a clinical trial, and from the NINDS as R01 grant funding. D R AF 10 18 19 Robert Lisak serves as the President of the CMSC and as a member of the Board of the DMC 20 Foundation; has served on scientific advisory boards for Syntimmune and Mallinckrodt; has 21 received funding for travel from the CMSC, the NMSS, the GBS/CIDP Foundation, and the 22 International Syntimmune for travel to consultants meetings; has served as a journal editor for 23 Clinical and Experimental Neuroimmunology and for Clinical Neuropharmacology; has received 5 Confidential draft: not for distribution, dissemination, or publication 1 publishing royalties from Willey for International Neurology, A Clinical Approach; has received 2 honoraria from Syntimmune, Mallinckrodt, and Teva Pharmaceuticals; has served on a speakers 3 bureau for Teva Pharmaceuticals for non-drug-related talks; has received research support from 4 Mallinckrodt for investigator-initiated wet bench studies, and from Teva, Biogen IDEC, 5 Novartis, Avanir, and Acorda for serving as a site investigator in multicenter trials; has received 6 financial compensation from the NMSS; has given expert testimony, prepared an affidavit, and 7 acted as witness for Teva Pharmaceuticals; and has acted as a witness for Acorda. 8 Ruth Ann Marrie has served as site principal investigator for Sanofi-Aventis clinical trials; T 9 receives research grants from nonprofit organizations, including CIHR, NMSS, MS Society of 11 Canada, MS Scientific Research Foundation, CMSC, Research Manitoba, and Rx & D Health 12 Research Foundation; and serves on the editorial board of Neurology. AF 10 13 Daniel Pelletier has served on scientific advisory boards for Biogen-IDEC, Genzyme, Hoffman 15 LaRoche, Acorda, Novartis, and EMD-Serono; has received research support for Biogen-IDEC, 16 Genzyme, Hoffman LaRoche, and the NIH NINDS; and has received honoraria for providing 17 consulting services at scientific advisory board meetings from Biogen-IDEC, Genzyme, 18 Hoffman LaRoche, Acorda, Novartis, and EMD-Serono. D R 14 19 20 Sonja Potrebic receives travel reimbursement from the AAN for the RITE work group and 21 Registry Committee meetings. 22 23 Rick Sommers reports no relevant disclosures. 6 Confidential draft: not for distribution, dissemination, or publication 1 2 Cynthia Sitcov reports no relevant disclosures. 3 4 Julie Stachowiak reports no relevant disclosures. 5 Thomas S.D. Getchius has received financial compensation for travel for speaking at the 7 University of Louisville mTBI conference; has been serving as the vice-chair of the Council of 8 Medical Specialty Societies Clinical Practice Guideline Component Group from November 2013 9 to present, and has received research support from the CDC as a grant for muscular dystrophy guideline development, dissemination, and implementation. AF 10 T 6 11 12 Shannon A. Merillat reports no relevant disclosures. 13 Tamara Pringsheim has received research support from Shire Canada Inc., and the Canadian 15 Institute for Health Research. D R 14 16 17 ABSTRACT 18 Maximum of 250 words; should summarize the guideline as follows: 19 Objective: Summary of clinical focus 20 Methods: Description of process 21 Results: Status, quality, and content of evidence 22 Recommendations: Summary of recommendations 23 7 Confidential draft: not for distribution, dissemination, or publication INTRODUCTION 2 Multiple sclerosis (MS) affects approximately 400,000 people in the United States and 2.3 3 million patients worldwide (Browne 2014). Annual direct costs are estimated to be $24,000 4 higher for people living with MS, than those without MS (Campbell 2014). In many patients MS 5 is ultimately disabling, although the course varies markedly (Confavreaux 2006). Since 1993, 6 multiple disease-modifying therapies (DMTs) have been approved in the United States for the 7 treatment of relapsing forms of MS; most of these therapies are approved for use in other 8 industrialized nations. In addition, many other medications have been used off-label for disease 9 modification of MS. All of these medications may have adverse effects, which may vary from bothersome to life-threatening. AF 10 T 1 11 Controversies persist concerning who should receive these medications (e.g., with respect to 13 disease course and disease activity). It is sometimes unclear when or whether therapy should be 14 initiated, and whether there are sufficient data to comment on the comparative efficacy of agents 15 (Bruck 2013, Wingerchuk 2014). In addition, changes in the diagnostic criteria for MS in 2010 16 (Polman 2011), as well as modification of the classification scheme for MS subtypes in 2014 17 (Lublin 2014), have complicated the extension of efficacy data from clinical trials. Before 18 recommending a specific therapy, the clinician must navigate these complexities while carefully 19 balancing the potential for therapeutic benefits of a medication with patient preferences, 20 monitoring recommendations, and drug- and individual-specific risk factors. The recent 21 proliferation of new DMTs with differing mechanisms of action, risk profiles and monitoring 22 requirements has advanced this discussion to the forefront of clinical care, increasing its 23 relevance to clinicians and patients. AAN members and leadership have articulated a strong need D R 12 8 Confidential draft: not for distribution, dissemination, or publication 1 for an updated guideline in this topic area. Given these complexities, an up-to-date guideline 2 focusing on these medications is critical to delivering optimal care to patients with MS. 3 The prior 2002 AAN clinical practice guideline on DMTs in MS (Goodin 2002) systematically 5 reviewed injectable medications approved for use in patients with MS at that time, including 6 interferon-beta 1b (Interferon Beta 1b SQ alternate day; Betaseron), interferon beta1a IM 7 (Interferon beta 1a intramuscular weekly; Avonex), interferon beta 1a SQ (Interferon beta 1a SQ 8 three times a week; Rebif) and glatiramer acetate (Copaxone). Medications commonly prescribed 9 off-label for the treatment of MS were not reviewed (e.g., azathioprine, cyclophosphamide, mycofenolate mofetil) (Goodin 2002). AF 10 T 4 11 The treatment landscape has changed considerably over the past 14 years, with more than 16 13 medications currently approved and widely prescribed for the treatment of MS in the United 14 States, and other agents nearing commercial approval (e.g., ocreluzimab). Recognizing these 15 remarkable therapeutic advances and persistent knowledge gaps, an updated systematic review 16 and practice guideline was prioritized by the AAN GDDI and AAN leadership in 2015. The 17 general membership of the American Academy of Neurology and other clinicians involved in 18 care of people with MS are the target users of this guideline. In addition, this guideline may be 19 informative to patients and their families, clinical and institutional leaders, and administrators. D R 12 20 21 The updated guideline reviews indications for the use of DMT; counseling prior to and during 22 use of DMTs; patient preferences regarding DMT use; stopping DMTs; and indications for use in 23 different types of MS (relapsing remitting MS [RRMS], secondary progressive MS [SPMS], 9 Confidential draft: not for distribution, dissemination, or publication 1 primary progressive MS [PPMS]). We reviewed adverse effects related to DMT use including 2 those affecting tolerability and less common but important risks and adverse effects. When 3 available, data concerning patient preference and patient-prioritized outcomes for treatment were 4 expressly analyzed. In addition, we solicited opinions concerning outcome measures of 5 importance from guideline panelists and patients with MS (via the NARCOMS registry) using a 6 formalized process. Perceived health benefits, adverse effects, and risks were formally 7 considered when developing recommendations. 8 The clinical questions for the guideline were developed by the panel and disseminated for public T 9 review in our guideline protocol. No substantive changes were made based on public review. 11 After review of the available evidence, adequate evidence was available to address the following 12 questions: 13 1. 14 annualized relapse rates? 15 2. 16 disease activity as measured by new T2 lesion burden or atrophy measures? 17 3. 18 progression as measured by in-study disease progression measures? 19 4. 20 different DMT superior to continuing present DMT in terms of relapse rate and MRI T2 or 21 gadolinium lesion activity? 22 5. In people with progressive MS, are DMTs superior to placebo or other DMTs as measured by 23 relapse rate or in-study disease progression? AF 10 R In people with RRMS, are DMTs superior to placebo or other DMTs as measured by D In people with RRMS, are DMTs superior to placebo or other DMTs in reducing MRI new In people with RRMS, are DMTs superior to placebo or other DMTs in preventing disease In people with RRMS who experience disease activity while on a DMT, is changing to a 10 Confidential draft: not for distribution, dissemination, or publication 1 6. What are the adverse effects of DMTs in patients with MS compared with placebo (adverse 2 event related discontinuation, and serious or life threatening adverse events)? 3 7. In people with clinically isolated syndromes, are DMTs superior to placebo in decreasing the 4 risk of conversion to clinically definite MS? 5 DESCRIPTION OF THE ANALYTIC PROCESS 7 In May 2015, a multidisciplinary panel consisting of 12 AAN physician members, 2 8 representative members from the Consortium of Multiple Sclerosis Centers (CMSC), and 3 9 patient representatives were recruited to develop this guideline. Two AAN staff representatives T 6 were also appointed to the panel. The physicians include content experts (B.C., D.E.J., D.P., 11 R.A.M., J.H.); a methodology expert (T.P.); Guideline Development, Dissemination, and 12 Implementation Subcommittee (GDDI) members (A.R., A.R.G., G.S.P., A.L.G.); and CMSC 13 representatives (J.H., R.L.). The patient representatives (J.S., R.S., C.S.) are 3 adults with an MS 14 diagnosis. All panel members were required to submit online conflict of interest (COI) forms and 15 copies of their curriculum vitae (CV). The panel leadership, consisting of the lead author 16 (A.R.G.), the AAN methodologist (T.P.), and the AAN staff persons (T.G., S.M.), reviewed the 17 COI forms and CVs for financial and intellectual COI. These documents were specifically 18 screened to exclude individuals with a clear financial conflict as well as those whose professional 19 and intellectual bias might diminish the perceived credibility of the review. In accordance with 20 AAN policy, the lead author (A.R.G.) has no COI. Four of the 20 authors were determined to 21 have COI, which were judged to be not significant enough to preclude them from authorship 22 (D.E.J., B.C., D.P., R.L.). All authors determined to have COI were not permitted to review or 23 rate the evidence. These individuals were utilized in an advisory capacity to help validate key D R AF 10 11 Confidential draft: not for distribution, dissemination, or publication 1 questions, assess the scope of the literature search, and identify seminal articles to validate the 2 literature search, and participated in the recommendation development process. AAN GDDI 3 leadership provided final approval of the author panel. This panel was solely responsible for 4 decisions concerning the design, analysis and reporting of the proposed systematic review, which 5 was then submitted for approval to the AAN GDDI. 6 This evidence-based guideline follows the methodologies outlined in the 2011 edition of the 8 AAN’s guideline development process manual. We summarize the process here and provide a 9 detailed description in the appendices referenced below. This process is compliant with Institute 10 of Medicine recommendations for guideline development (Graham 2011). Over the course of its 11 development, the public and experts had an opportunity to review the draft protocol during a 30- 12 day public comment period, during which the document was posted on the AAN website. During 13 this period, AAN staff sent invitations to review and comment on the guideline to key stake- 14 holders, including all AAN section members, and pertinent external physician and patient 15 organizations, including the National Multiple Sclerosis Society, the Consortium for MS Centers, 16 and (others to be added). The guideline was reviewed by the GDDI subcommittee before and 17 after the public comment period. D R AF T 7 18 19 Panel members developed the clinical questions and the data extraction template. The guideline 20 panel defined DMTs as medications that aim to affect the clinical course of MS by decreasing 21 relapses and/or slowing disease progression. We limited our search for relevant literature to 22 medications that have been approved by the US Food and Drug Administration (FDA), Health 23 Canada or the European Medicines Agency, or medications historically used for disease 12 Confidential draft: not for distribution, dissemination, or publication modification in MS but which were not licensed for this purpose. When disputes arose 2 concerning whether a medication should be included, the panel erred on the side of inclusion 3 (e.g., corticosteroids for disease modification and cladribine). In addition, agents which 4 potentially would receive licensure within the time frame of the guideline process were included 5 (e.g., daclizumab, approved by FDA May 2016; and ocrelizumab). Using this strategy, 20 6 medications were identified and included in this review: methotrexate, cyclophosphamide, 7 pulsed corticosteroids for disease modification, interferon-beta (four types; interferon beta 1b SQ 8 alt day, interferon beta 1a IM SQ, pegylated interferon SQ q alt week, interferon beta 1a SQ 9 TIW), glatiramer acetate, natalizumab, azathioprine, teriflunomide, mycophenolate mofetil, T 1 rituximab, daclizumab, mitoxantrone, alemtuzumab, fingolimod, dimethyl fumarate, intravenous 11 immunoglobulin for disease modification, and cladribine. To find relevant studies, the project 12 methodologist (T.P.) searched the Cochrane Library for Cochrane Systematic Reviews, including 13 any of the above mentioned medications for disease modification in individuals with MS. All 14 Cochrane reviews were evaluated by a second reviewer and graded using the AMSTAR tool—a 15 rating instrument of systematic review quality (ref to be added). Cochrane reviews were updated 16 by repeating the outlined search strategies in MEDLINE and CENTRAL and/or EMBASE from 17 the date last searched in the review to the present (XXXX). When there was a DMT for which no 18 previous Cochrane review had been published, we performed a de novo systematic review 19 following the 2011 AAN guideline development process manual (Neurology 2011). The search 20 was restricted to peer reviewed English-language articles on human subjects. A secondary search 21 of references of all Cochrane reviews and polling of all panel members was conducted until 22 March 2016 to identify any articles missed in the initial search. For efficacy outcomes, we D R AF 10 13 Confidential draft: not for distribution, dissemination, or publication 1 considered data from randomized controlled trials. For harms, we considered data from 2 randomized controlled trials and cohort studies, case reports or case series. 3 Two non-conflicted panel members rated the class of evidence for each article according to the 5 AAN scheme for classification of therapeutic articles. Disagreements were resolved by a third 6 panel member. Outcome data from included studies were extracted by the guideline 7 methodologist and verified by a second panel member. Data for different disease types (RRMS, 8 SPMS, PPMS) were analyzed separately where possible. Studies comparing different DMTs 9 were included in this analysis. Prior to data analysis the panel completed an anonymous survey 10 on what would be considered a minimal clinically meaningful differences for various measures 11 of DMT efficacy and adverse effects to use in the analytic portion of the guideline. For relapses, 12 a relative risk reduction of ≥5% was considered clinically important, and a relative risk reduction 13 of ≤2% was considered clinically unimportant. For disability progression, a relative risk 14 reduction of ≥2% was considered clinically important, and a relative risk reduction of ≤1% was 15 considered clinically unimportant. For adverse event related discontinuation, a risk difference of 16 ≥10% was considered clinically important, and a risk difference of ≤5% was considered 17 clinically unimportant. For serious or life threatening adverse events, a risk difference of ≥0.1% 18 was considered clinically important, and a risk difference of ≤0.01% was considered clinically 19 unimportant. D R AF T 4 20 21 A modified GRADE process was utilized to develop conclusions (Guyatt 2011). In this process 22 the evidence is analyzed based on a variety of parameters of risk of bias (multiple types), 14 Confidential draft: not for distribution, dissemination, or publication 1 consistency, directness, precision, and publication bias. Using this process, class of evidence can 2 be modified up or down transparently (Guyatt 2011). 3 The panel formulated practice recommendations on the basis of the strength of evidence and 5 other factors, including axiomatic principles of care, the magnitude of anticipated health benefits 6 relative to harms, financial burden, availability of interventions, and patient preferences. The 7 panel assigned levels of obligation (A, B, C, U, R) to the recommendations using a modified 8 Delphi process. Considerations for future research and recommendations were also developed 9 during the process of this guideline development. T 4 AF 10 11 This guideline will be reassessed over time for currency and need for updating according to the 12 published AAN Guideline Development, Dissemination and Implementation process (Neurology 13 2011). R 14 RESULTS OF PHYSICIAN AND PATIENT SURVEY ON DESIRED OUTCOMES 16 Although studies have independently considered how MS experts [1, 2] and patients [3-8] with 17 MS weigh factors in therapeutic decision-making, few studies have investigated how closely 18 aligned prescriber- and patient-priorities are throughout this process. As AAN guidelines are 19 developed primarily to improve the ability of physicians to meet patient needs, we sought to 20 compare how outcomes potentially affected by DMTs were rank-ordered by panel experts (N = 21 18) and patients with MS (N = 2156). Data from 5 outcomes of interest were consistently 22 reported in high-quality (Class I or II) studies of medications for patients with RRMS: relapse 23 rate, disability progression, discontinuation of treatment due to adverse events, neuroimaging D 15 15 Confidential draft: not for distribution, dissemination, or publication changes and serious adverse events (threatening life or organ). Three additional outcomes of 2 interest were identified through review of studies focused on patient-specified outcome measures 3 [3, 7, 8], including the effect of treatment on cognition, MS symptoms (e.g., fatigue, pain, 4 urinary incontinence), and quality of life measures. MS experts and patients with MS were asked 5 to rank all outcomes “in order of importance to you when choosing a MS treatment.” The MS 6 Guideline Development guideline panel was electronically surveyed using Google Surveys 7 (December 18, 2015 to January 11, 2016). Patients with MS participating in the North American 8 Research Committee on Multiple Sclerosis (NARCOMS) registry were invited to participate via 9 email. MS patient responses were accrued for a 7-day period (January 5–11, 2016). Survey T 1 questions (Supplement 1) were developed in consultation with members of the MS guideline 11 development and NARCOMS executive committees. AF 10 12 All 18 members of the MS Guideline Development panel (100% response rate) and 2156 of 5487 14 actively engaged NARCOMS registry members (39.3% response rate) responded to the survey 15 request. The percentage of respondents designating an outcome as first-, second- or third-highest 16 priority is shown in Figure X. There was good agreement between MS experts and patients in the 17 rank-ordering of outcome measures of greatest importance to therapeutic decision making. The 18 ability of MS DMTs to reduce disability progression and relapse rate were identified as the two 19 most important factors when choosing a MS treatment. The potential for serious adverse events 20 associated with a treatment was the third most important factor when selecting a MS DMT. 21 Physicians were more likely than patients to rank relapse rate as a top priority (Fisher’s exact 22 test, two-tailed: p = 0.02). No significant differences were observed between respondents 23 concerning other outcomes (Fisher’s exact test, two-tailed: p > 0.05). Of interest, a cumulative D R 13 16 Confidential draft: not for distribution, dissemination, or publication 1 total of 20% of MS experts and 18% of patients with MS identified selection of therapies most 2 likely to lead to improvements in quality of life measures. MS symptoms and preservation of 3 cognition was their top priority. Data concerning these measures are not consistently reported in 4 Class I or II studies of MS DMTs, which identifies a potentially important shortcoming in the 5 literature concerning meaningful outcomes that may be addressed through future studies. 6 ANALYSIS OF EVIDENCE 8 A total of 20 Cochrane systematic reviews were identified and used in the evidence review 9 process. These systematic reviews included 65 randomized controlled trials, which were 10 included in our evidence synthesis. For the update of the Cochrane reviews and de novo 11 systematic reviews, the combined searches yielded 3123 abstracts. Each abstract was reviewed 12 for relevance by at least 2 panel members. Of these, 235 were deemed relevant. The 13 corresponding articles were obtained for full-text review by 2 panelists working independently. 14 61 additional articles were identified for data extraction. R AF T 7 D 15 16 All trials included individuals with MS ≥18 years of age. The maximum age of participants 17 varied across trials, but it was generally between 50 to 60 years. The majority of studies were 2 18 years in length (range between six months and three years). Trials occurred in multiple countries 19 worldwide. 20 21 Therapeutic question: In people with RRMS, are DMTs superior to placebo or other DMTs 22 as measured by annualized relapse rates? 17 Confidential draft: not for distribution, dissemination, or publication 1 The efficacy of DMTs for preventing relapses was assessed in most trials by measuring the 2 proportion of patients with relapses compared to placebo over 2 years. Annualized relapse rates 3 (ARR) were derived from this information. Results are reported by drug, in alphabetical order. 4 Alemtuzumab 6 One class I (CARE-MS 1) and one class 2 (CAMMS223) study evaluated the proportion of 7 patients with RRMS with at least one relapse at two years with alemtuzumab treatment compared 8 to interferon beta 1a SQ three times a week. Meta-analysis of data from 914 participants 9 revealed reveals a risk ratio of 0.43 (0.29, 0.61), favoring alemtuzumab. Meta-analysis of data T 5 from the same two studies revealed a raw mean difference in annualized relapse rate of 0.26 11 (0.22, 0.29), favoring alemtuzumab. 12 AF 10 Conclusions 14 For individuals with RRMS, alemtuzumab is more effective than interferon beta 1a SQ 3 times a 15 week in reducing the risk of relapse at 2 years (high confidence, 1 Class 1 study, 1 Class 2 study; 16 confidence upgraded due to magnitude of effect). For individuals with RRMS, alemtuzumab is 17 more effective than interferon beta 1a SQ 3 times a week in reducing the ARR (high confidence, 18 1 Class 1 study, 1 Class 2 study; confidence upgraded due to magnitude of effect). D R 13 19 20 21 Azathioprine 22 One class II study (Goodkin 1991) evaluated the proportion of individuals with at least 1 relapse 23 at 2 years with azathioprine vs placebo. This study found a risk ratio of 0.74 (0.50, 1.07), 18 Confidential draft: not for distribution, dissemination, or publication 1 favoring treatment. The same study reported the ARR after 2 years of treatment; the raw mean 2 difference was 0.49 (0.07, 0.91), favoring treatment. 3 There are 2 Class II studies (Massacesi 2014; Etemadifar 2007) comparing azathioprine to beta 5 interferons in 244 individuals with RRMS for the ARR. Meta-analysis of data from these studies 6 shows a rate ratio of 0.64 (0.44, 0.92), favoring azathioprine. One of these Class II studies 7 (Etemadifar 2007) also evaluated the proportion of individuals with RRMS with relapses at 1 8 year. The RR was 0.75 (0.55, 0.99), favoring azathioprine. 9 Conclusions T 4 For individuals with RRMS, there is insufficient evidence to determine the efficacy of 11 azathioprine compared to placebo in reducing the risk of relapse over 24 months (very low 12 confidence, 1 Class II study; confidence in evidence downgraded because of imprecision). 13 Azathioprine is probably more effective than placebo in reducing the mean number of relapses 14 over 2 years (moderate confidence, 1 Class II study; confidence upgraded due to magnitude of 15 effect). Azathioprine is more effective than beta interferons in reducing the ARR (high 16 confidence, 2 Class II studies; confidence upgraded due to magnitude of effect). Azathioprine is 17 probably more effective than beta interferon in reducing the risk of relapse over 12 months 18 (moderate confidence, 1 Class II study; confidence upgraded due to magnitude of effect). D R AF 10 19 20 Cladribine 21 Two Class II studies (Romine 1999, Giovannoni 2010) evaluated cladribine vs placebo in 1376 22 individuals with RRMS, with 1 study (Giovannoni 2010) evaluating 2 different doses of oral 23 cladribine, and the other (Romine 1999) evaluating subcutaneous cladribine. The raw mean 19 Confidential draft: not for distribution, dissemination, or publication 1 difference in the ARR was 0.19 (0.14, 0.24). The proportion of patients with at least 1 relapse at 2 2 years was studied in 1 Class II study of 1326 individuals treated with oral cladribine 3.5 mg/kg 3 or 5.2 mg/kg. The risk ratio was 0.53 (0.45, 0.63) in individuals treated with either dose of 4 cladribine relative to placebo. 5 Conclusions 7 For individuals with RRMS, cladribine is more effective than placebo in reducing the ARR (high 8 confidence, 2 Class II studies; confidence upgraded due to magnitude of effect). Cladribine is 9 probably more effective than placebo in reducing the risk of relapse over 2 years (moderate confidence, 1 Class II study; confidence upgraded due to magnitude of effect). 11 AF 10 T 6 Corticosteroids 13 One Class IV study of 81 individuals with RRMS (Zivadinov 2001) evaluated the effect of 14 methylprednisone 1 g/d for 5 days every 4 months vs placebo on the risk of relapse at 2 years, 15 and the ARR. The risk ratio for at least 1 relapse at two years was 1.16 (0.851, 1.591); the raw 16 mean difference in the annualized relapse rate was 0.0 (-0.238, 0.238). D R 12 17 18 Two Class II studies (Sorensen 2009, Ravnborg 2010) evaluated monthly pulsed 19 methylprednisolone added to interferon-beta treatment compared to placebo added to interferon 20 treatment in 471 individuals with RRMS. The risk ratio for at least 1 relapse at 2 years was 0.33 21 (0.20, 0.54) based on data from the Sorensen study, and the risk ratio for at least 1 relapse at 3 22 years was 0.78 (0.62, 0.98) based on data from the Ravnborg study. The raw mean difference in 23 the ARR based on data from both studies was 0.24 (0.02, 0.47). 20 Confidential draft: not for distribution, dissemination, or publication 1 Conclusions 3 For individuals with RRMS, there is insufficient evidence to determine the efficacy of pulsed 4 corticosteroids alone compared to placebo in reducing the risk of relapse over 2 years or the 5 ARR (very low confidence, 1 Class IV study). Monthly pulsed corticosteroids added to 6 interferon-beta is probably more effective than placebo added to interferon in reducing the risk of 7 relapse over 2 years and 3 years (moderate confidence, 1 Class II study, confidence upgraded 8 due to magnitude of effect). Monthly pulsed corticosteroids added to interferon-beta is probably 9 more effective than placebo added to interferon-beta in reducing the ARR (moderate confidence, 2 Class II studies). 11 AF 10 T 2 Cyclophosphamide 13 One Class III study (Killian 1988) evaluated cyclophosphamide vs placebo in 14 individuals 14 with RRMS. The risk ratio for at least 1 relapse at 1 year was 0.67 (0.25, 1.44). The raw mean 15 difference between cyclophosphamide and placebo for the AAR was 1.80 (0.56, 3.04). D R 12 16 17 Conclusions 18 For individuals with RRMS, there is insufficient evidence to determine the efficacy of 19 cyclophosphamide compared to placebo in reducing the risk of relapse over 12 months (very low 20 confidence, 1 Class III study). Cyclophosphamide is possibly more effective than placebo in 21 reducing the ARR (low confidence, 1 Class III study; confidence upgraded due to magnitude of 22 effect). 23 21 Confidential draft: not for distribution, dissemination, or publication 1 Daclizumab High Yield Process (HYP) 2 One Class I study (Gold 2013) evaluated the effect of daclizumab HYP vs placebo in 397 3 individuals with RRMS. The risk ratio for relapse at 1 year was 0.54 (0.38, 0.75), and the RMD 4 in the ARR was 0.25 (0.13, 0.37). One Class II (Kappos 2015) study compared daclizumab HYP 5 to interferon beta 1a in 1841 individuals with RRMS over 3 years. The risk ratio for at least 1 6 relapse at 3 years was 0.67 (0.60, 0.75), and the RMD in the ARR was 0.17 (0.12, 0.22), favoring 7 daclizumab. 8 Conclusions T 9 For individuals with RRMS, daclizumab is more effective than placebo in decreasing the risk of 11 at least 1 relapse at 1 year (high confidence, 1 class I study, confidence in evidence upgraded due 12 to magnitude of effect). Daclizumab is more effective than placebo in decreasing the ARR (high 13 confidence, 1 Class I study, confidence in evidence upgraded due to magnitude of effect). 14 Daclizumab is probably more effective than interferon beta 1a in decreasing the risk of at least 1 15 relapse at 3 years (moderate confidence, 1 Class II study; confidence in evidence upgraded due 16 to magnitude of effect). Daclizumab is probably more effective than interferon beta 1a in 17 decreasing the ARR at 3 years (moderate confidence, 1 Class II study; confidence in evidence 18 upgraded due to magnitude of effect). D R AF 10 19 20 Dimethyl Fumarate 21 One Class I (Fox 2012) and 1 Class II (Gold 2012) study evaluated the efficacy of dimethyl 22 fumarate vs placebo on relapses in 1540 individuals with RRMS. The risk ratio for the 22 Confidential draft: not for distribution, dissemination, or publication 1 proportion of individuals with at least 1 relapse at 2 years with dimethyl fumarate compared to 2 placebo was 0.64 (0.54, 0.77), and the raw mean difference in the ARR was 0.19 (0.13, 0.25). 3 Conclusions 5 In individuals with RRMS, dimethyl fumarate is more effective than placebo in decreasing the 6 risk of at least 1 relapse at 2 years (high confidence, 1 Cclass I and 1 Class II study; confidence 7 in evidence upgraded due to magnitude of effect). Dimethyl fumarate is more effective than 8 placebo in decreasing the ARR (high confidence, 1 Class I and 1 Class II study; confidence in 9 evidence upgraded due to magnitude of effect). AF 10 T 4 Fingolimod 12 One Class I (Kappos 2010) and 1 Class II (Calabresi 2014) study compared fingolimod to 13 placebo for the risk of relapse at 2 years in 1556 individuals with RRMS. The risk ratio for the 14 proportion of individuals with at least 1 relapse at 2 years was 0.57 (0.50, 0.65), favoring 15 fingolimod. Two Class I studies (Kappos 2010, Saida 2012) and 1 Class II (Calabresi 2014) 16 study compared fingolimod to placebo for the ARR in 1670 individuals with RRMS. The raw 17 mean difference between fingolimod and placebo was 0.21 (0.16, 0.26), favoring fingolimod. 18 One Class I study (Cohen 2010) compared fingolimod to interferon beta 1a in 860 individuals 19 with RRMS. The risk ratio for at least 1 relapse at 12 months was 0.58 (0.46, 0.75), and the 20 RMD in the ARR was 0.17 (0.08, 0.26), favoring fingolimod. D R 11 21 22 Conclusions 23 Confidential draft: not for distribution, dissemination, or publication 1 For individuals with RRMS, fingolimod is more effective than placebo in reducing the risk of 2 relapse over 2 years (high confidence, 1 Class I and 1 Class 2 study; confidence upgraded due to 3 magnitude of effect). Fingolimod is more effective than placebo in reducing the ARR (high 4 confidence, 2 Class I studies and 1 Class 2 study). Fingolimod is more effective than interferon 5 beta 1a in reducing the risk of relapse at 1 year (high confidence, 1 Class I study; confidence in 6 evidence upgraded due to magnitude of effect). Fingolimod is more effective than interferon beta 7 1a in reducing the ARR (high confidence, 1 Class I study; confidence in evidence upgraded due 8 to magnitude of effect). T 9 Glatiramer Acetate 11 Two Class I studies (Fox 2012, Khan 2013) and 1 Class II (Johnson 1995) study evaluated the 12 ARR with glatiramer compared to placebo in 2368 individuals with RRMS. The RMD was 0.18 13 (0.09, 0.28), favoring glatiramer. There is 1 Class I study (Fox 2012) and 2 Class II (Bornstein 14 1987, Johnson 1995) studies in 1012 individuals with RRMS evaluating the proportion of 15 individuals with at least 1 relapse at 2 years. The risk ratio was 0.82 (0.69, 0.97), favoring 16 glatiramer acetate. One Class II study (Mikol 2008) compared glatiramer to Interferon beta 1a 17 SQ 3 times a week in 764 individuals with RRMS. There was no difference between the 18 proportion of patients with at least 1 relapse at 2 years, with a risk ratio of 0.93 (0.77, 1.14). Two 19 Class II studies (Cavadid 2009, O'Connor 2009) compared glatiramer to Interferon Beta 1b SQ 20 alternate day (250 micrograms) in 1420 individuals with RRMS. There was no difference 21 between the proportion of patients with at least 1 relapse at 2 years, with a risk ratio of 1.19 22 (0.75, 1.90). D R AF 10 23 24 Confidential draft: not for distribution, dissemination, or publication Conclusions 2 For individuals with RRMS, glatiramer is more effective than placebo in reducing the ARR (high 3 confidence, 2 Class I and 1 Class II studies). Glatiramer is probably more effective than placebo 4 in reducing the risk of relapse at 2 years (moderate confidence, 1 Class I and 2 Class II studies). 5 Interferon beta 1a SQ 3 times a week is possibly no more effective than glatiramer in reducing 6 the risk of at least 1 relapse at 2 years (low confidence, 1 Class II study). Interferon Beta 1b SQ 7 alternate day is possibly no more effective than glatiramer in reducing the risk of at least 1 8 relapse at 2 years (low confidence, 2 Class II studies; confidence in evidence downgraded due to 9 imprecision). T 1 AF 10 Immunoglobulins 12 Two Class I (Fazekas 1997, Fazekas 2008) and 2 Class II (Achiron 1998, Lewanska 2002) 13 studies compared immunoglobulins to placebo in 460 individuals with RRMS. Meta-analysis of 14 study results revealed no difference between treatments, with a RMD of 0.37 (-0.21, 0.94) in the 15 ARR. One Class I (Fazekas 1997) and 1 Class II (Achiron 1998) study reported the proportion of 16 190 individuals with at least 1 relapse at 2 years. The risk ratio was 0.74 (0.61, 0.87), favoring 17 immunoglobulins. D R 11 18 19 Conclusions 20 For individuals with RRMS, there is insufficient evidence to determine the efficacy of 21 immunoglobulins compared to placebo in reducing the ARR (very low confidence, 2 Class I 22 study and 2 Class II studies; confidence in evidence downgraded due to imprecision). 23 Immunoglobulins are more effective than placebo in reducing the risk of at least 1 relapse 25 Confidential draft: not for distribution, dissemination, or publication 1 over2two years (high confidence, 1 Class I study and 1 Class II study; confidence in evidence 2 upgraded due to magnitude of effect). 3 Interferon beta 1a intramuscular weekly 5 One Class 1 (Vollmer 2014) and 1 Class 2 (Jacobs 1996) study evaluated the proportion of 6 individuals with RRMS with at least 1 relapse at 2 years compared to placebo. Meta-analysis of 7 data from 1198 individuals revealed a risk ratio of 0.79 (0.68, 0.92), favoring treatment. One 8 Class I study (Vollmer 2014) found that interferon beta 1a intramuscular weekly was more 9 effective than placebo in reducing the ARR, with a RMD1 of 0.08 (0.01, 0.15). T 4 AF 10 Conclusions 12 For individuals with RRMS, interferon beta 1a intramuscular weekly is more effective than 13 placebo in reducing the risk of relapse over 24 months (high confidence, 1 Class I study, 1 Class 14 II study; confidence upgraded due to magnitude of effect). Interferon beta 1a intramuscular 15 weekly is probably more effective than placebo in reducing the ARR (moderate confidence, 1 16 Class I study). D R 11 17 18 Interferon beta 1a SQ three times a week 19 One Class I study (PRISMS 1998) compared Interferon beta 1a SQ 3 times a week to placebo in 20 560 individuals with RRMS. The risk of relapse at 2 years was significantly lower with 21 Interferon beta 1a SQ 3 times a week, with a risk ratio of 0.84 (0.77, 0.92). One Class II study 22 (Panitch 2002) compared Interferon beta 1a SQ 3 times a week to Interferon beta 1a 1 The raw mean difference is used when all studies use the same outcome measure to assess an outcome. The raw mean difference is the mean difference between the treatment groups for the specified outcome. 26 Confidential draft: not for distribution, dissemination, or publication 1 intramuscular weekly in 677 individuals with RRMS. The risk of relapse at one year was 2 significantly lower with Interferon beta 1a SQ 3 times a week compared to Interferon beta 1a 3 intramuscular weekly, with a risk ratio of 0.84 (0.72, 0.99). 4 Conclusions 6 For individuals with RRMS, Interferon beta 1a SQ 3 times a week is more effective than placebo 7 in reducing the risk of relapse at 2 years (high confidence, 1 Class I study; confidence in 8 evidence upgraded due to magnitude of effect). Interferon beta 1a SQ 3 times a week is probably 9 more effective than Interferon beta 1a intramuscular weekly in reducing the risk of relapse at 1 year (moderate confidence, 1 Class II study). AF 10 T 5 11 Interferon Beta 1b SQ alternate day 13 One Class II study (Group 1993) including 227 individuals with RRMS evaluated the effect of 14 Interferon Beta 1b SQ alternate day vs placebo on relapses. This study found a risk ratio of 0.82 15 (0.70, 0.95) for the proportion of individuals with at least 1 relapse at 2 years, and a RMD of 16 0.43 (0.24, 0.62) in the relapse rate over 2 years. D R 12 17 18 Conclusions 19 For individuals with RRMS, Interferon Beta 1b SQ alternate day is probably more effective than 20 placebo in reducing the risk of relapse over 2 years (moderate confidence, 1 Class II study; 21 confidence upgraded due to magnitude of effect). Interferon Beta 1b SQ alternate day is probably 22 more effective than placebo in reducing the relapse rate over 2 years (moderate confidence, 1 23 Class II study; confidence upgraded due to magnitude of effect). 27 Confidential draft: not for distribution, dissemination, or publication 1 2 Methotrexate 3 One Class III study (Currier 1993) compared methotrexate to placebo in 20 individuals with 4 RRMS. There was no significant difference in the risk of relapse over 18 months, with a risk 5 ratio of 0.35 (0.10, 1.04). One Class II study (ref to be added) compared methotrexate to 6 interferon beta 1a in 80 individuals with RRMS. The difference between treatments in the 7 number of relapses over 12 months was not statistically significant, although the RMD of 0.39 (- 8 0.01, 0.79) favored interferon beta 1a over methotrexate. T 9 Conclusions 11 In individuals with RRMS, there is insufficient evidence to determine the efficacy of 12 methotrexate compared to placebo in decreasing the proportion of patients who relapsed over 18 13 months (very low confidence, 1 Class III study). There is insufficient evidence to determine the 14 efficacy of methotrexate compared to interferon beta 1a in decreasing the number of relapses 15 over 12 months (very low confidence, 1 Class III study). D R AF 10 16 17 Mitoxantrone 18 One Class I study (Millefiorini 1997) compared mitoxantrone to placebo in 51 individuals with 19 RRMS. The risk ratio for the proportion of individuals with a relapse over 24 months was 0.47 20 (0.27, 0.77), favoring mitoxantrone. 21 22 Conclusion 28 Confidential draft: not for distribution, dissemination, or publication 1 For individuals with RRMS, mitoxantrone is more effective than placebo in decreasing the risk 2 of relapse at 2 years (high confidence, 1 Class I study; confidence in evidence upgraded due to 3 magnitude of effect). 4 Mycophenolate Mofetil 6 Two Class II studies (Etemadifar 2010, Remington 2010) compared mycophenolate mofetil plus 7 interferon beta 1a to placebo plus interferon beta 1a in 50 individuals with RRMS. No significant 8 difference was observed between treatments in the risk of relapse at 1 year, with a risk ratio of 9 0.63 (0.18, 2.23). One Class III study (Frohman 2010) compared mycophenolate mofetil to T 5 interferon beta 1a in 35 individuals with RRMS. There was no significant difference in the risk 11 of relapse at 6 months, with a risk ratio of 1.18 (0.22, 6.16). 12 AF 10 Conclusions 14 For individuals with RRMS, mycophenolate plus interferon beta 1a is possibly no more effective 15 than placebo plus interferon beta 1a in decreasing the risk of relapse at 1 year (low confidence, 2 16 Class II studies; confidence in evidence downgraded due to imprecision). There is insufficient 17 evidence to determine the efficacy of mycophenolate mofetil compared to interferon beta 1a in 18 decreasing the risk of relapse at 6 months (very low confidence, 1 Class III study). D R 13 19 20 Natalizumab 21 One Class I study (Polman 2006) compared natalizumab to placebo in 942 individuals with 22 RRMS. The risk of at least 1 relapse at 2 years was significantly lower with natalizumab than 29 Confidential draft: not for distribution, dissemination, or publication 1 placebo, with a risk ratio of 0.56 (0.49, 0.64). The annualized relapse rate was significantly lower 2 with natalizumab, with a RMD of 0.41 (0.31, 0.51). 3 4 Conclusion 5 For individuals with RRMS, natalizumab is more effective than placebo in reducing the risk of 6 relapse at 2 years, and reducing the ARR (high confidence, 1 Class I study; confidence in 7 evidence upgraded due to magnitude of effect). 8 Pegylated Interferon T 9 One Class I study (Calabresi 2014) compared pegylated interferon 125 mcg every 2 weeks to 11 placebo in 1012 individuals with RRMS. The risk of at least 1 relapse at 1 year was significantly 12 lower with pegylated interferon, with a risk ratio of 0.62 (0.49, 0.78). The ARR was significantly 13 lower with pegylated interferon, with a RMD of 0.11 (0.01, 0.21). 14 R AF 10 Conclusion 16 For individuals with RRMS, pegylated interferon is more effective than placebo in reducing the 17 risk of relapse and the ARR at 1 year (high confidence, 1 Class I study; confidence in evidence 18 upgraded due to magnitude of effect). D 15 19 20 Rituximab 21 One Class II study compared rituximab to placebo in 104 individuals with RRMS (Hauser 2008). 22 The risk of relapse was significantly lower with rituximab compared to placebo, with a risk ratio 30 Confidential draft: not for distribution, dissemination, or publication 1 of 0.51 (0.28, 0.94). The ARR at 1 year was not significantly different between treatments with a 2 RMD 0.30 (-0.07, 0.67). 3 Conclusions 5 For individuals with RRMS, rituximab is probably more effective than placebo in decreasing the 6 risk of relapse at 1 year (moderate confidence, 1 Class II study; confidence in evidence upgraded 7 due to magnitude of effect). There is insufficient evidence to determine the efficacy of rituximab 8 compared to placebo in decreasing the ARR at 1 year (very low confidence, 1 Class II study; 9 confidence in evidence downgraded due to imprecision). T 4 AF 10 Teriflunomide 12 One Class II study (O'Connor 2011) compared teriflunomide to placebo in 1088 individuals with 13 RRMS. The risk of at least one relapse at 2 years was significantly lower with teriflunomide, 14 with a risk ratio of 0.88 (0.79, 0.98). Three Class II studies (O'Connor 2006, O'Connor 2011, 15 Confavreux 2014) compared teriflunomide to placebo in 1597 individuals with RRMS. The ARR 16 was significantly lower with teriflunomide compared to placebo, with a RMD of 0.18 (0.11, 17 0.25). D R 11 18 19 Conclusions 20 For individuals with RRMS, teriflunomide is probably more effective than placebo in decreasing 21 the risk of relapse at 2 years (moderate confidence, 1 Class II study; confidence in evidence 22 upgraded due to magnitude of effect). Teriflunomide is more effective than placebo in decreasing 31 Confidential draft: not for distribution, dissemination, or publication 1 the ARR (high confidence, 3 Class II studies; confidence in evidence upgraded due to magnitude 2 of effect). 3 Therapeutic question: In people with RRMS, are DMTs superior to placebo or other DMTs 5 in reducing MRI new disease activity as measured by new T2 lesion burden or atrophy 6 measures? 7 Various measures of the effect of DMTs on MRI disease activity were performed, including 8 mean differences in volume or number of T2 lesions, the proportion of patients with new or 9 enlarging T2 lesions, or changes in whole brain volume compared to baseline. T 4 AF 10 Alemtuzumab 12 One Class I study in 563 relapsing patients with MS (Cohen 2012) comparing alemtuzumab to 13 interferon beta 1a reported a decrease in T2 lesion volume from baseline to 2 years in, with a 14 standardized mean difference (SMD)2 of 0.18 (0.01, 0.36). The same study reported a lower 15 proportion of patients with new or enlarging T2 lesions at 2 years in those treated with 16 alemtuzumab compared to interferon beta 1a, with a risk ratio of 0.84 (0.71, 0.99). One Class II 17 study (Coles 2012) found no difference between alemtuzumab and interferon beta 1a concerning 18 the T2 lesion load from baseline to 3 years, with an SMD 0.10 (-0.20, 0.39). The same Class II 19 study (Coles 2012) found a smaller decrease in brain volume on T1 from baseline to 3 years with 20 alemtuzumab compared to interferon beta 1a, with an SMD of 0.37 (0.07, 0.67). D R 11 21 2 The standardized mean difference is used as a summary statistic in meta‐analysis when the studies all assess the same outcome but measure it in a variety of ways. In this circumstance it is necessary to standardize the results of the studies to a uniform scale before they can be combined. The standardized mean difference expresses the size of the intervention effect in each study relative to the variability observed in that study. 32 Confidential draft: not for distribution, dissemination, or publication Conclusions 2 For individuals with RRMS, alemtuzumab is probably more effective than interferon beta 1a in 3 decreasing the volume of T2 lesions from baseline to 2 years (moderate confidence, 1 Class I 4 study). Alemtuzumab is probably more effective than interferon beta 1a in reducing the risk of 5 new or enlarging T2 lesions at 2 years (moderate confidence, 1 Class I study). There is 6 insufficient evidence to determine the efficacy of alemtuzumab compared to interferon beta 1a in 7 decreasing the T2 lesion load from baseline to 3 years (very low confidence, 1 Class II study; 8 confidence in evidence downgraded due to imprecision). Alemtuzumab is probably more 9 effective than interferon beta 1a in decreasing the change in brain volume on T1 from baseline to T 1 3 years (moderate confidence, 1 Class II study, confidence in evidence upgraded due to 11 magnitude of effect). 12 AF 10 Azathioprine 14 One Class II study comparing azathioprine to beta interferons (Massacesi 2014) evaluated the 15 proportion of individuals with RRMS with no new T2 lesions at 2 years. There was no 16 significant difference between treatments, with a RR of 1.21 (0.81, 1.82). D R 13 17 18 Conclusion 19 For individuals with RRMS, there is insufficient evidence to determine the efficacy of 20 azathioprine compared to beta interferon in reducing the risk of new T2 lesions at 2 years (very 21 low confidence, 1 Class II study; confidence in evidence downgraded due to imprecision). 22 23 Cladribine 33 Confidential draft: not for distribution, dissemination, or publication 1 One Class II study (Giovannoni 2010) evaluated the effect of 2 different doses of oral cladribine 2 (3.5 mg vs 5.25 mg) in 1326 individuals with RRMS on the proportion of patients with at least 1 3 active T2 lesion at 2 years. This study demonstrated a risk ratio of 0.529 (0.478, 0.587) with 4 either dose of cladribine compared to placebo. 5 Conclusion 7 For patients with RRMS, cladribine is probably more effective than placebo in reducing the risk 8 of at least 1 active T2 lesion at 2 years (moderate confidence, 1 Class II study; confidence in the 9 evidence upgraded due to magnitude of effect). AF 10 T 6 Corticosteroids 12 One Class II study (Zivadinov 2001) compared pulsed corticosteroids to placebo in 81 13 individuals with RRMS on MRI T2 lesion volume and brain parenchymal volume at 5 years. 14 This study demonstrated a SMD of 0.203 (-0.234, 0.640) for T2 lesion volume and a SMD of 15 0.622 (0.175, 1.068) for brain parenchymal volume at 5 years. D R 11 16 17 Conclusions 18 For patients with RRMS, there is insufficient evidence to determine the efficacy of pulsed 19 corticosteroids compared to placebo in reducing T2 lesion volume at 5 years (very low 20 confidence, 1 Class II study, confidence in evidence downgraded due to imprecision). Pulsed 21 corticosteroids are probably more effective than placebo in reducing the loss of parenchymal 22 volume at 5 years (moderate confidence, 1 Class II study, confidence upgraded due to magnitude 23 of effect). 34 Confidential draft: not for distribution, dissemination, or publication 1 2 Daclizumab HYP 3 One Class I study (Gold 2013) evaluated the effect of daclizumab HYP vs placebo on the mean 4 number of new or newly enlarging lesions at 1 year in 397 individuals with RRMS. This study 5 found a SMD of 0.94 (0.74, 1.15) favoring daclizumab HYP. One Class II study (Kappos 2015) 6 compared daclizumab HYP to interferon beta 1a in 1841 individuals with RRMS. The SMD in 7 the mean number of new or newly enlarging T2 hyperintense lesions over 2 years was 0.42 8 (0.33, 0.52), favoring daclizumab HYP. T 9 Conclusions 11 For individuals with RRMS, daclizumab HYP is more effective than placebo in preventing new 12 or newly enlarging T2 lesions at 1 year (high confidence, 1 Class I study; confidence in evidence 13 upgraded due to magnitude of effect). Daclizumab HYP is probably more effective than 14 interferon beta 1a in preventing new or newly enlarging T2 lesions at 2 years (moderate 15 confidence, 1 Class II study, confidence in evidence upgraded due to magnitude of effect). D R AF 10 16 17 Dimethyl Fumarate 18 One Class I (Fox 2012) and 1 Class II (Gold 2012) study compared the effect of dimethyl 19 fumarate to placebo on the mean number of new or enlarging T2 lesions at 2 years in 1097 20 individuals with RRMS. The SMD was 0.49 (0.29, 0.69), favoring dimethyl fumarate. 21 22 Conclusion 35 Confidential draft: not for distribution, dissemination, or publication 1 In individuals with RRMS, dimethyl fumarate is more effective than placebo in decreasing the 2 number of new or enlarging T2 lesions at 2 years (high confidence, 1 Class I and 1 Class II 3 study; confidence in evidence upgraded due to magnitude of effect). 4 Fingolimod 6 One Class I (Kappos 2010) and 1 Class II (Calabresi 2014) study compared the proportion of 7 individuals with new or enlarging T2 lesions at 2 years with fingolimod vs placebo in 1224 8 persons with RRMS. The risk ratio was 0.65 (0.59, 0.71), favoring fingolimod. One Class I study 9 (Cohen 2010) in 733 individuals with RRMS compared fingolimod to interferon beta 1a. The 10 risk ratio for the proportion of individuals with new or enlarged T2 lesions at 12 months was 11 0.83 (0.72, 0.96). AF T 5 12 Conclusions 14 In individuals with RRMS, fingolimod is more effective than placebo in reducing the risk of new 15 or enlarging T2 lesions at 2 years (high confidence, 1 Class I and 1 Class II study; confidence in 16 evidence upgraded due to magnitude of effect). Fingolimod is probably more effective than 17 interferon beta 1a in reducing the risk of new or enlarging T2 lesions over 1 year (moderate 18 confidence, 1 Class I study). D R 13 19 20 Glatiramer 21 One Class I study (Fox 2012) compared the number of new or enlarging T2 lesions at 2 years in 22 292 individuals with RRMS receiving glatiramer vs placebo. The SMD between treatments was 23 0.46 (0.23, 0.69), favoring glatiramer. One Class II study (Mikol 2008) compared the proportion 36 Confidential draft: not for distribution, dissemination, or publication 1 of patients with active T2 lesions at 2 years with glatiramer vs interferon beta 1a SQ 3 times a 2 week in 460 individuals with RRMS. There was no difference between treatments, with a risk 3 ratio of 0.95 (0.82, 1.11). 4 Conclusions 6 For individuals with RRMS, glatiramer is more effective than placebo in decreasing the number 7 of new or enlarging T2 lesions at 2 years (high confidence, 1 Class I study; confidence in 8 evidence upgraded due to magnitude of effect). Interferon beta 1a SQ 3 times a week is possibly 9 no more effective than glatiramer in decreasing the risk of active T2 lesions at 2 years (low confidence, 1 Class II study). AF 10 T 5 11 Interferon beta 1a intramuscular weekly 13 One Class I study (Vollmer 2014) evaluated the mean change in brain volume from baseline to 2 14 years and the cumulative number of new/enlarged T2 lesions at months 12 and 24 with interferon 15 beta 1a intramuscular weekly compared to placebo. No significant difference in brain volume 16 was detected (SMD 0.003 [-0.128, 0.134]); however, the cumulative number of new or enlarged 17 T2 lesions was lower in the interferon beta 1a intramuscular weekly group (SMD 0.37, 95% CI 18 0.23, 0.50). D R 12 19 20 Conclusions 21 For individuals with RRMS, there is insufficient evidence to determine the efficacy of interferon 22 beta 1a intramuscular weekly compared to placebo in reducing the change in brain volume at 2 23 years (very low confidence, 1 Class I study; confidence in evidence downgraded due to 37 Confidential draft: not for distribution, dissemination, or publication 1 imprecision). Interferon beta 1a intramuscular weekly is more effective than placebo in reducing 2 the cumulative number of new or enlarged T2 lesions at months 12 and 24 (high confidence, 1 3 Class 1 study; confidence in evidence upgraded due to magnitude of effect). 4 Interferon beta 1a SQ 3 times a week 6 One Class I study (PRISMS 1998) evaluated new T2 activity over 2 years in 366 individuals 7 with RRMS treated with interferon beta 1a SQ 3 times a week compared to placebo. The risk 8 ratio for new T2 activity was 0.75 (0.67, 0.83), favoring interferon beta 1a SQ 3 times a week. 1 9 Class II study (Panitch 2002) compared interferon beta 1a SQ 3 times a week to interferon beta T 5 1a intramuscular weekly in 677 individuals with RRMS. The proportion of individuals with new 11 or enlarging T2 lesions at 1 year was significantly lower with interferon beta 1a SQ 3 times a 12 week, with a risk ratio of 0.62 (0.50, 0.76). 13 AF 10 Conclusions 15 For individuals with RRMS, interferon beta 1a SQ 3 times a week is more effective than placebo 16 in reducing the risk of new or enlarging T2 lesions at 2 years (high confidence, 1 Class I study; 17 confidence in evidence upgraded due to magnitude of effect). Interferon beta 1a SQ 3 times a 18 week is probably more effective than interferon beta 1a intramuscular weekly in reducing the 19 risk of new or enlarging T2 lesions at 1 year (moderate confidence, 1 Class II study; confidence 20 in evidence upgraded due to magnitude of effect). D R 14 21 22 Mitoxantrone 38 Confidential draft: not for distribution, dissemination, or publication 1 One Class I study (Millefiorini 1997) of mitoxantrone vs placebo evaluated the number of new 2 lesions on T2 from baseline to 2 years in 42 individuals with RRMS. The SMD between 3 treatments was 0.64 (0.02, 1.26), favoring mitoxantrone. 4 5 Conclusion 6 In individuals with RRMS, mitoxantrone is more effective than placebo in decreasing the 7 number of new lesions on T2 at 2 years (high confidence, 1 Class I study; confidence in evidence 8 upgraded due to magnitude of effect). T 9 Mycophenolate Mofetil 11 One Class II study (Remington 2010) compared mycophenolate mofetil plus interferon beta 1a to 12 placebo plus interferon beta 1a in 24 individuals with RRMS. There was no significant 13 difference in T2 lesion volume percent change or brain volume percent change at 1 year between 14 treatments. The SMD for T2 lesion volume percent change was 0.52 (-0.29, 1.34) and for brain 15 volume percent change was 0.72 (-0.11, 1.55). One Class II study (add ref) compared 16 mycophenolate mofetil to interferon beta 1a in the mean number of new T2 lesions at 6 months 17 in 35 individuals with RRMS. There was no significant difference between treatments, with an 18 SMD of 0.38 (-0.29, 1.06). D R AF 10 19 20 Conclusions 21 In individuals with RRMS, there is insufficient evidence to determine the efficacy of 22 mycophenolate mofetil plus interferon beta 1a compared to placebo plus interferon beta 1a in 23 decreasing the percentage of T2 lesion volume change or brain volume change at 1 year (very 39 Confidential draft: not for distribution, dissemination, or publication 1 low confidence, 1 Class II study; confidence in evidence downgraded due to imprecision). There 2 is insufficient evidence to determine the efficacy of mycophenolate mofetil compared to 3 interferon beta 1a in decreasing the mean number of new T2 lesions at 6 months (very low 4 confidence, 1 Class II study; confidence in evidence downgraded due to imprecision). 5 Natalizumab 7 One Cass I study (Polman 2006) compared natalizumab to placebo in 942 individuals with 8 RRMS. The risk of having at least 1 new or enlarging T2 lesion at 1 year was significantly lower 9 with natalizumab than placebo, with a risk ratio of 0.51 (0.45, 0.57). The MRI T2 lesion load at 2 10 years was significantly less with natalizumab than with placebo with a SMD of 0.28 (0.14, 0.42); 11 individuals treated with natalizumab had a mean decrease in lesion load, whereas individuals 12 treated with placebo had a mean increase in lesion load. AF T 6 13 Conclusion 15 In individuals with RRMS, natalizumab is more effective than placebo in reducing the risk of at 16 least 1 new or enlarging T2 lesion at 1 year and in reducing the MRI T2 lesion load at 2 years 17 (high confidence, 1 Class I study; confidence in evidence upgraded due to magnitude of effect). D R 14 18 19 Pegylated Interferon 20 One Class I study (Calabresi 2014) compared pegylated interferon 125 mcg every 2 weeks to 21 placebo in 1012 individuals with RRMS. The number of new or newly enlarging T2 lesions at 1 22 year was significantly reduced with pegylated interferon, with a SMD of 0.60 (0.48, 0.73). 23 40 Confidential draft: not for distribution, dissemination, or publication 1 Conclusion 2 For individuals with RRMS, pegylated interferon is more effective than placebo in reducing the 3 number of new or newly enlarging T2 lesions at 1 year (high confidence, 1 Class I study, 4 confidence in evidence upgraded due to magnitude of effect). 5 Rituximab 7 One Class II study (Hauser 2008) compared rituximab to placebo in 104 individuals with RRMS. 8 The mean decrease in volume of lesions on T2 MRI from baseline to week 36 was significantly 9 greater with rituximab compared to placebo, with a SMD of 0.48 (0.07, 0.90). T 6 AF 10 Conclusion 12 For individuals with RRMS, rituximab is probably more effective than placebo in decreasing the 13 volume of T2 lesions from baseline to week 36 (moderate confidence, 1 Class II study, 14 confidence in evidence upgraded due to magnitude of effect). R 11 D 15 16 Teriflunomide 17 One Class II study (O'Connor 2011) compared teriflunomide 14 mg to placebo in 721 18 individuals with RRMS. Teriflunomide was more effective than placebo in decreasing the 19 volume of MRI T2 lesions at 2 years than placebo, with a SMD of 0.19 (0.05, 0.34). 20 21 Conclusion 41 Confidential draft: not for distribution, dissemination, or publication 1 For individuals with RRMS, teriflunomide is probably more effective than placebo in decreasing 2 the volume of MRI T2 lesions at 2 years (moderate quality, 1 Class II study; confidence in 3 evidence upgraded due to magnitude of effect). 4 Therapeutic question: In people with RRMS, are DMTs superior to placebo or other DMTs 6 in preventing disease progression as measured by in-study disease progression measures? 7 The most consistently reported measure for in-study disability progression was the proportion of 8 patients with disability progression, defined as an increase in the Expanded Disability Status 9 Scale (EDSS) of 1 point in those with a baseline EDSS less than or equal to 5.0, or an increase of 10 0.5 points in those with a baseline EDSS greater than or equal to 5.5, sustained for 3 or 6 months, 11 which was detected over a 2-year study period. 12 AF T 5 Alemtuzumab 14 One Class 1 (Cohen 2012) and 1 Class II (Coles 2012) study evaluated the proportion of patients 15 with disability progression sustained for 6 months over 2 years with alemtuzumab compared to 16 interferon beta 1a, including a total of 914 patients. Meta-analysis of data demonstrated a risk 17 ratio of 0.44 (0.28, 0.70), favoring alemtuzumab. D R 13 18 19 Conclusion 20 For individuals with RRMS, alemtuzumab is more effective than interferon beta 1a in reducing 21 the risk of disability progression over 2 years (high confidence, 1 Class I and 1 Class II study; 22 confidence upgraded due to magnitude of effect). 23 42 Confidential draft: not for distribution, dissemination, or publication 1 Azathioprine 2 One Class II study (add reference) evaluated the proportion of individuals with disability 3 progression sustained for at least 2 months with azathioprine vs placebo, reporting a risk ratio of 4 0.58 (0.23, 1.46), favoring azathioprine. 5 Conclusion 7 For individuals with RRMS, there is insufficient evidence to determine the efficacy of 8 azathioprine compared to placebo in reducing the risk of disability progression over 24 months 9 (very low confidence, 1 Class II study; confidence in evidence downgraded due to imprecision). T 6 AF 10 Cladribine 12 One Class II study (Giovannoni 2010) evaluated the proportion of patients with a 3 month 13 sustained change in EDSS at 2 years with 2 doses of oral cladribine in 1326 individuals with 14 RRMS. Relative to placebo, the risk ratio for either dose of cladribine was 0.72 (0.56, 0.91), 15 favoring cladribine. D R 11 16 17 Conclusion 18 For individuals with RRMS, cladribine is probably more effective than placebo in reducing the 19 risk of disability progression over 2 years (moderate confidence, 1 Class II study; confidence in 20 evidence upgraded due to magnitude of effect). 21 22 Corticosteroids 43 Confidential draft: not for distribution, dissemination, or publication 1 One Class IV study (Zivadinov 2001) compared every 4 months pulsed intravenous 2 corticosteroids (methylprednisolone 1 g for 5 days with oral prednisone taper) to placebo for 3 disability progression (sustained changed in EDSS for 4 months) at 2 years in 81 individuals 4 with RRMS. The risk ratio was 0.14 (0.04, 0.49), favoring pulsed corticosteroids. One Class II 5 study (Ravnborg 2010) evaluated the proportion of patients with disability progression sustained 6 for 6 months with pulsed corticosteroid added to interferon compared to placebo added to 7 interferon in 341 individuals with RRMS over 3 years. The risk ratio was 0.93 (0.66, 1.33), 8 favoring pulsed corticosteroids. T 9 Conclusions 11 For individuals with RRMS, pulsed corticosteroids are possibly more effective than placebo in 12 reducing the risk of disability progression at 2 years (low confidence, 1 Class IV study; 13 confidence in evidence upgraded due to magnitude of effect). Pulsed corticosteroids added to 14 interferon are possibly no more effective than placebo added to interferon in reducing the risk of 15 disability progression over 3 years (low confidence, 1 Class II study). D R AF 10 16 17 Daclizumab HYP 18 One Class I study (Gold 2013) evaluated the proportion of individuals with disease progression 19 sustained for 3 months with daclizumab HYP compared to placebo after 1 year in 397 20 individuals with RRMS. The risk ratio was 0.43 (0.22, 0.84), favoring daclizumab HYP. One 21 Class II study (Kappos 2015) compared daclizumab HYP to interferon beta 1a in 1841 22 individuals with RRMS. The risk ratio for disability progression sustained for 3 months over 3 23 years was 0.80 (0.66, 0.98), favoring daclizumab HYP. 44 Confidential draft: not for distribution, dissemination, or publication 1 2 Conclusion 3 In individuals with RRMS, daclizumab HYP is more effective than placebo in decreasing the 4 risk of disability progression at 1 year (high confidence, 1 Class I study; confidence in evidence 5 upgraded due to magnitude of effect). Daclizumab HYP is probably more effective than 6 interferon beta 1a in decreasing the risk of disability progression at 3 years (moderate 7 confidence, 1 Class II study; confidence in evidence upgraded due to magnitude of effect). 8 Dimethyl Fumarate T 9 One Class I (Fox 2012) and 1 Class II (Gold 2012) study evaluated dimethyl fumarate vs placebo 11 in 1539 individuals with RRMS. The risk ratio for disability progression sustained for 3 months 12 over 2 years was 0.65 (0.53, 0.81), favoring dimethyl fumarate. 13 AF 10 Conclusion. 15 In individuals with RRMS, dimethyl fumarate is more effective than placebo in decreasing the 16 risk of disability progression over 2 years (high confidence, 1 Class I and 1 Class II study; 17 confidence in evidence upgraded due to magnitude of effect). D R 14 18 19 Fingolimod 20 One Class I (Kappos 2010) and 1 Class II (Calabresi 2014) study compared fingolimod to 21 placebo in 1556 individuals with RRMS. The risk ratio for disability progression sustained for 3 22 months over 2 years was 0.81 (0.68, 0.96), favoring fingolimod. One Class I study (Cohen 2010) 45 Confidential draft: not for distribution, dissemination, or publication 1 compared fingolimod to interferon beta 1a in 860 individuals with RRMS. The risk ratio for 2 disability progression sustained for 3 months over 1 year was 0.74 (0.45, 1.21). 3 Conclusion 5 In individuals with RRMS, fingolimod is more effective than placebo in reducing the risk of 6 disability progression over 2 years (high confidence, 1 Class I and 1 Class II study; confidence in 7 evidence upgraded due to magnitude of effect). Fingolimod is possibly no more effective than 8 interferon beta 1a in reducing the risk of disability progression over 1 year (low confidence, 1 9 Class I study; confidence in evidence downgraded due to imprecision). T 4 AF 10 Glatiramer Acetate 12 One Class I (Fox 2012) and 2 Class II (Bornstein 1987, Johnson 1995) studies compared 13 glatiramer acetate to placebo in 1024 individuals with RRMS. The risk ratio for the proportion of 14 individuals with disease progression sustained for at least 3 months over 2 years was 0.76 (0.53, 15 1.08). One Class II study (O'Connor 2009) compared glatiramer to interferon beta 1b SQ 16 alternate day (250 mg) in 1345 individuals with RRMS. A greater proportion of individuals 17 taking interferon beta 1b SQ alternate day had confirmed disease progression compared to those 18 taking glatiramer, with a risk ratio of 1.32 (1.08, 1.64). D R 11 19 20 Conclusion 21 For individuals with RRMS, glatiramer is possibly no more effective than placebo in reducing 22 the risk of disability progression over 2 years (low confidence, 1 Class I and 2 Class II studies; 23 confidence in evidence downgraded due to imprecision). Interferon beta-1b SQ alternate day is 46 Confidential draft: not for distribution, dissemination, or publication 1 probably less effective than glatiramer in decreasing the risk of disease progression over 2 years 2 (moderate confidence, 1 Class II study; confidence in evidence upgraded due to magnitude of 3 effect). 4 Immunoglobulins 6 One Class I (Fazekas 1997) and 1 Class II (Achiron 1998) study compared immunoglobulins to 7 placebo, measuring the risk of disability progression over 2 years in 190 people with RRMS. 8 There was no difference between immunoglobulins and placebo, with a risk ratio of 0.70 (0.39, 9 1.24). T 5 AF 10 Conclusion 12 For individuals with RRMS, there is insufficient evidence to determine the efficacy of 13 immunoglobulins compared to placebo in reducing the risk of disability progression at 2 years 14 (very low confidence, 1 Class I and 1 Class II study; confidence in evidence downgraded due to 15 imprecision). D R 11 16 17 Interferon beta 1a intramuscular weekly 18 One Class I (Vollmer 2014) and 1 Class II (Jacobs 1996) study evaluated the proportion of 19 individuals with disability progression sustained for 3 or 6 months over the 2-year study period 20 and included a total of 1198 patients. Meta-analysis of data demonstrated a risk ratio of 0.71 21 (0.52, 0.97), favoring interferon beta 1a intramuscular weekly. 22 23 Conclusion 47 Confidential draft: not for distribution, dissemination, or publication 1 For individuals with RRMS, interferon beta 1a intramuscular weekly is more effective than 2 placebo in reducing the risk of disability progression over 24 months (high confidence, 1 Class I 3 and 1 Class II study; confidence upgraded due to magnitude of effect). 4 5 Interferon beta 1b SQ alternate day 6 One Class II study (Group 1993) evaluated the proportion of individuals with RRMS with 7 disability progression sustained for 3 months in a 3–year study of interferon beta 1b SQ alternate 8 day vs placebo. This study demonstrated a risk ratio of 0.73 (0.47, 1.14). T 9 Conclusion 11 For individuals with RRMS, interferon beta 1b SQ alternate day is possibly no more effective 12 than placebo in reducing the risk of disability progression over 36 months (low confidence, 1 13 Class II study; confidence in evidence downgraded due to imprecision). AF 10 R 14 Interferon beta 1a SQ 3 times a week 16 One Class I study (PRISMS 1998) compared interferon beta 1a SQ 3 times a week to placebo in 17 560 individuals with RRMS. The proportion of individuals with disability progression sustained 18 for 3 months was significantly lower with interferon beta 1a SQ 3 times a week compared to 19 placebo, with a risk ratio of 0.77 (0.61, 0.96). D 15 20 21 Conclusion 48 Confidential draft: not for distribution, dissemination, or publication 1 For patients with RRMS, interferon beta 1a SQ 3 times a week is more effective than placebo in 2 reducing the risk of disability progression over 2 years (high confidence, 1 Class I study; 3 confidence in evidence upgraded due to magnitude of effect). 4 5 Methotrexate 6 One Class III study (Currier 1993) compared methotrexate to placebo in 20 individuals with 7 RRMS. There was no significant difference between treatments in the risk of disability 8 progression over 18 months, with a risk ratio of 0.73 (0.24, 2.00). T 9 Conclusion 11 There is insufficient evidence to determine the efficacy of methotrexate compared to placebo in 12 the risk of disability progression over 18 months (very low confidence, 1 Class III study). 13 AF 10 Mitoxantrone 15 One Class I study (Millefiorini 1997) compared mitoxantrone to placebo in 51 individuals with 16 RRMS. The risk ratio for disability progression from baseline to endpoint at 2 years was 0.20 17 (0.05, 0.73), favoring mitoxantrone. D R 14 18 19 Conclusion 20 For patients with RRMS, mitoxantrone is more effective than placebo in decreasing the risk of 21 disability progression at 2 years (high confidence, 1 Class I study, confidence in evidence 22 upgraded due to magnitude of effect). 23 49 Confidential draft: not for distribution, dissemination, or publication 1 Natalizumab 2 One Class I study (Polman 2006) compared natalizumab to placebo in 942 individuals with 3 RRMS. The risk of disability progression sustained for 12 weeks over 2 years was significantly 4 lower with natalizumab, with a risk ratio of 0.64 (0.52, 0.80). 5 Conclusion 7 For patients with RRMS, natalizumab is more effective than placebo in decreasing the risk of 8 disability progression at 2 years (high confidence, 1 Class I study, confidence in evidence 9 upgraded due to magnitude of effect). AF 10 T 6 Pegylated Interferon 12 One Class I study (Calabresi 2014) compared pegylated interferon 250 mcg every 2 weeks to 13 placebo in 1012 individuals with RRMS. The risk ratio for disease progression sustained for 3 14 months over 1 year was 0.61 (0.40, 0.93). R 11 D 15 16 Conclusion: 17 For patients with RRMS, pegylated interferon is more effective than placebo in decreasing the 18 risk of disability progression at 1 year (high confidence, 1 Class I study, confidence in evidence 19 upgraded due to magnitude of effect). 20 21 Teriflunomide 50 Confidential draft: not for distribution, dissemination, or publication 1 Two Class II studies (O'Connor 2011, Confavreux 2014) compared teriflunomide 14 mg to 2 placebo in 1479 individuals with RRMS. Teriflunomide was associated with a lower risk of 3 disability progression sustained for 12 weeks than placebo, with a risk ratio of 0.76 (0.62, 0.93). 4 5 Conclusion 6 In individuals with RRMS, teriflunomide 14 mg is more effective than placebo in reducing the 7 risk of disability progression over 2 years (high confidence, 2 Class II studies; confidence in 8 evidence upgraded due to magnitude of effect). T 9 Therapeutic question: In people with RRMS who experience disease activity while on a 11 DMT, is changing to a different DMT superior to continuing present DMT in terms of 12 relapse rate and MRI T2 activity? 13 Alemtuzumab 14 One Class I study (Coles 2012) compared alemtuzumab to interferon beta 1a in individuals with 15 RRMS who experienced a relapse in the past year while taking interferon (any formulation) or 16 glatiramer. Alemtuzumab was more effective than interferon beta 1a in reducing the risk of 17 relapse at 2 years (risk ratio 0.59, 95%CI 0.51, 0.69), reducing the ARR (RMD 0.26, 95% CI 18 0.13, 0.39), reducing the proportion of patients with disability progression over 2 years (RR 0.38, 19 95% CI 0.28, 0.50), and decreasing the proportion of individuals with new or enlarging T2 20 hyperintense lesions at 24 months (RR 0.68, 95%CI 0.58, 0.78). D R AF 10 21 22 Conclusion 51 Confidential draft: not for distribution, dissemination, or publication 1 For individuals with RRMS who experienced a relapse on interferon or glatiramer, alemtuzumab 2 is more effective than interferon beta 1a in reducing the reducing the ARR, reducing the risk of 3 relapse, disability progression, and risk of new or enlarging T2 lesions over 2 years (high 4 confidence, 1 Class 1 study; confidence upgraded due to magnitude of effect). 5 Natalizumab 7 One Class I study (Rudick 2006) compared interferon beta 1a plus natalizumab (“natalizumab 8 add-on therapy”) to interferon beta 1a plus placebo in 1171 individuals with RRMS who 9 experienced at least one relapse in the past 12 months on interferon. Natalizumab add-on therapy T 6 was more effective than adding placebo for all clinical outcome measures, with a risk ratio for 11 relapse at 2 years of 0.62 (0.55, 0.70), a RMD in ARR of 0.41 (0.36, 0.59), and a risk ratio of 12 0.80 (0.69, 0.93) for disability progression sustained for 12 weeks over 2 years. Natalizumab 13 add-on therapy was also more effective than adding placebo for all MRI outcomes, including the 14 risk of having at least 1 new or enlarging T2 lesion at 1 year with a risk ratio of 0.51 (0.44, 0.59). R AF 10 D 15 16 One Class II study (Goodman 2009) compared glatiramer plus natalizumab to glatiramer plus 17 placebo in 110 individuals with RRMS who relapsed at least once in the past 12 months on 18 glatiramer. There was no difference between natalizumab add-on therapy and placebo in the 19 number of individuals who relapsed over 6 months, with a risk ratio of 0.80 (0.42, 1.52). 20 Natalizumab add-on therapy was more effective than placebo in decreasing the mean number of 21 new or enlarging T2 lesions at 6 months, with a SMD of 0.48 (0.10, 0.86). 22 23 Conclusions 52 Confidential draft: not for distribution, dissemination, or publication In individuals with RRMS who experienced one or more relapses in the past 12 months on 2 interferon, adding natalizumab is more effective than adding placebo in decreasing the risk of 3 relapse over 2 years, decreasing the ARR, decreasing the risk of disability progression over 2 4 years, and decreasing the risk of new or enlarging T2 lesions at 1 year (high confidence, 1 Class I 5 study; confidence in evidence upgraded due to magnitude of effect). In individuals with RRMS 6 who experienced one or more relapses in the past 12 months on glatiramer, there is insufficient 7 evidence to determine the efficacy of natalizumab added to glatiramer compared to placebo 8 added to glatiramer in decreasing the risk of relapse at 6 months (very low confidence, 1 Class II 9 study; confidence in evidence downgraded due to imprecision). Natalizumab added to glatiramer 10 is probably more effective than placebo added to glatiramer in decreasing the cumulative number 11 of new or enlarging T2 lesions at 6 months (moderate quality evidence, 1 Class II study; 12 confidence in evidence upgraded due to magnitude of effect). AF T 1 13 Therapeutic question: In people with progressive MS, are DMTs superior to placebo or 15 other DMTs as measured by relapse rate or in-study disease progression? 16 Azathioprine 17 One Class II study (Ellison 1989) compared azathioprine to placebo in 64 people with 18 progressive forms of MS. This study demonstrated a risk ratio of 0.53 (0.25, 1.10) for relapses at 19 2 years, and a RMD of 0.31 (0.01, 0.61) in the number of relapses at 2 years. The study also 20 evaluated the proportion of individuals with disability progression from baseline to endpoint over 21 3 years and demonstrated a risk ratio of 0.64 (0.32, 1.25), favoring azathioprine. D R 14 22 23 Conclusions 53 Confidential draft: not for distribution, dissemination, or publication 1 For patients with progressive forms of MS, there is insufficient evidence to determine the 2 efficacy of azathioprine compared to placebo in reducing the risk of relapse over 2 years (very 3 low confidence, 1 Class II study; confidence in evidence downgraded due to imprecision). 4 Azathioprine is probably more effective than placebo in reducing the mean number of relapses 5 over 2 years (moderate confidence, 1 Class II study; confidence in evidence upgraded due to 6 magnitude of effect). There is insufficient evidence to determine the efficacy of azathioprine 7 compared to placebo in reducing the risk of disability progression over 3 years (very low 8 confidence, 1 Class II study; confidence in evidence downgraded due to imprecision). T 9 Cladribine 11 One Class II study (Rice 2000) evaluated the proportion of patients with disability progression at 12 1 year in 111 individuals with SPMS treated with subcutaneous cladribine compared to placebo. 13 The risk ratio was 0.78 (0.44, 1.42). 14 R AF 10 Conclusion 16 For individuals with SPMS, cladribine is possibly no more effective than placebo in reducing the 17 risk of disability progression at 1 year (low confidence, 1 Class II study). D 15 18 19 Corticosteroids 20 One Class II study compared high-dose corticosteroids (bimonthly pulses of 500 mg IV for 3 21 days) to low-dose corticosteroids (bimonthly pulses of 10 mg IV for 3 days) in 108 individuals 22 with SPMS (Goodkin, Kinkel et al. 1998). The risk ratio for the proportion of individuals with 54 Confidential draft: not for distribution, dissemination, or publication 1 relapses at 2 years was 0.33 (0.08, 1.44). The risk ratio for the proportion of patients with 2 worsening of disability from baseline to endpoint was 1.50 (0.29, 7.54). 3 Conclusion: 5 In individuals with SPMS, high-dose pulsed steroids are possibly no more effective than low- 6 dose pulsed corticosteroids in reducing the proportion of individuals with relapses at 2 years (low 7 confidence, 1 Class II study). High-dose pulsed steroids are possibly no more effective than low- 8 dose pulsed corticosteroids in reducing the proportion of individuals with SPMS with disability 9 progression at 2 years (low confidence, 1 Class II study). T 4 AF 10 Cyclophosphamide 12 Two Class II studies (CCMSG 1991, Likosky 1991) evaluated the proportion of 97 individuals 13 with progressive forms of MS with disability progression over 2 years. The risk ratio for 14 disability progression was 1.37 (0.88, 2.13). R 11 D 15 16 Conclusion 17 For individuals with progressive forms of MS, there is insufficient evidence to determine the 18 efficacy of cyclophosphamide compared to placebo in reducing the risk of disability progression 19 over 2 years (very low confidence, 2 Class II studies; confidence in evidence downgraded due to 20 imprecision). 21 22 Glatiramer Acetate 55 Confidential draft: not for distribution, dissemination, or publication 1 One Class II study (Bornstein 1991) compared glatiramer to placebo in 106 individuals with 2 progressive forms of MS. The risk ratio for the proportion of individuals with confirmed 3 disability progression sustained for 3 months over 2 years was 0.69 (0.33, 1.46). One Class II 4 study (Wolinsky, Narayana et al. 2007) compared glatiramer to placebo in 943 individuals with 5 PPMS. The risk ratio for confirmed disability progression sustained for 3 months over 2 years 6 was 0.99 (0.86, 1.14). 7 Conclusion 9 For patients with progressive forms of MS, glatiramer is possibly no more effective than placebo 10 in reducing the risk of disability progression over 2 years (low confidence, 1 Class II study). For 11 patients with PPMS, glatiramer is possibly no more effective than placebo in reducing the risk of 12 disability progression over 2 years (low confidence, 1 Class II study). 13 AF T 8 Immunoglobulins 15 There are 2 Class II studies (Hommes 2004, Pohlau 2007) comparing immunoglobulins to 16 placebo in 515 individuals with progressive MS. There was no difference between 17 immunoglobulins and placebo in the risk of at least 1 relapse at 2 years, with a risk ratio of 0.96 18 (0.79, 1.16), or in the risk of disability progression over 2 years, with a risk ratio of 0.96 (0.74, 19 1.25). D R 14 20 21 Conclusion 56 Confidential draft: not for distribution, dissemination, or publication 1 For individuals with progressive MS, there is insufficient evidence to determine the efficacy of 2 immunoglobulins compared to placebo in reducing the risk of relapse or disability progression 3 over 2 years (very low confidence, 2 Class II studies). 4 Interferon beta 1a intramuscular weekly 6 One Class II study (Cohen 2002) evaluated the efficacy of interferon beta 1a intramuscular 7 weekly for relapse prevention and clinical disease progression in 436 individuals with SPMS. 8 Individuals treated with interferon beta 1a intramuscular weekly had a lower risk of relapse at 2 9 years, with a risk ratio of 0.72 (0.54, 0.95). The proportion of individuals with disability T 5 progression sustained for three months was not different between interferon beta 1a 11 intramuscular weekly and placebo, with a risk ratio of 0.85 (0.64, 1.12). One Class I study (Leary 12 2003) in 50 individuals with PPMS found that the proportion of individuals with disability 13 progression sustained for 3 months was not different between interferon beta 1a intramuscular 14 weekly and placebo, with a risk ratio of 1.19 (0.78, 1.21). R AF 10 D 15 16 Conclusion 17 For individuals with SPMS, interferon beta 1a intramuscular weekly is probably more effective 18 than placebo in reducing the risk of relapse over 2 years (moderate confidence, 1 Class II study; 19 confidence upgraded due to magnitude of effect). For individuals with SPMS, interferon beta 1a 20 intramuscular weekly is possibly no more effective than placebo in reducing the risk of disability 21 progression over 24 months (low confidence, 1 Class II study). For individuals with PPMS, 22 interferon beta 1a intramuscular weekly is possibly no more effective than placebo in reducing 23 the risk of disability progression over 2 years (low confidence, 1 Class I study). 57 Confidential draft: not for distribution, dissemination, or publication 1 Interferon beta 1a SQ 3 times a week 3 There is 1 Class I study (SPECTRIMS 2001) comparing interferon beta 1a SQ 22 mcg or 44 mcg 4 3 times a week to placebo in 618 individuals with SPMS. Both doses of interferon beta 1a SQ 3 5 times a week were more effective than placebo in decreasing the mean number of exacerbations 6 per year. The RMD between interferon beta 1a SQ 3 times a week of 22 mcg and placebo was 7 0.21 (0.14, 0.28), and the RMD between interferon beta 1a SQ 3 times a week of 44 mcg and 8 placebo was 0.21 (0.15, 0.27), favoring interferon beta 1a SQ 3 times a week. The same study 9 also evaluated the risk of disability progression sustained for 3 months over 2 years. There was 10 no difference between either dose of interferon beta 1a SQ 3 times a week and placebo, with a 11 risk ratio of 0.93 (0.79, 1.10). 12 AF T 2 Conclusion 14 In individuals with SPMS, interferon beta 1a SQ 3 times a week is more effective than placebo in 15 decreasing the mean number of exacerbations per year (high confidence, 1 Class I study; 16 confidence in evidence upgraded due to magnitude of effect). Interferon beta 1a SQ 3 times a 17 week is possibly no more effective than placebo in reducing the risk of disability progression 18 over 2 years (low confidence, 1 Class I study; confidence in evidence downgraded due to 19 imprecision). D R 13 20 21 Interferon Beta 1b SQ alternate day 22 Two Class II studies evaluated the efficacy of interferon beta 1b SQ alternate day compared to 23 placebo on the risk of relapse, ARR and disability progression in 1333 individuals with SPMS 58 Confidential draft: not for distribution, dissemination, or publication 1 (Kappos 1998, NASG 2004). The risk ratio for at least 1 relapse over 3 years was 0.84 (0.75, 2 0.93), and the RMD in ARR was 0.16 (0.02, 0.29), favoring interferon beta 1b SQ alternate day. 3 The risk ratio for the proportion of patients with disability progression over 3 years was 0.85 4 (0.71, 1.02), favoring interferon beta 1b SQ alternate day. One Class II study compared 5 interferon beta 1b SQ alternate day vs placebo in 73 individuals with PPMS (Montalban 2009). 6 The risk ratio for the proportion of patients with disability progression sustained for 3 months 7 over a 2 year period was 0.69 (0.32, 1.43). 8 Conclusion T 9 For patients with SPMS, interferon beta 1b SQ alternate day is more effective than placebo in 11 reducing the risk of relapse over 3 years (high confidence, 2 Class II studies; confidence 12 upgraded due to magnitude of effect). Interferon beta 1b SQ alternate day is more effective than 13 placebo in reducing the ARR (high confidence, 2 Class II studies, confidence upgraded due to 14 magnitude of effect). Interferon beta 1b SQ alternate day is possibly no more effective than 15 placebo in reducing the risk of disability progression over 36 months (low confidence, 2 Class II 16 studies; confidence in evidence downgrade due to imprecision). For patients with PPMS, 17 interferon beta 1b SQ alternate day is possibly no more effective than placebo in reducing the 18 risk of disability progression over 2 years (low confidence, 1 Class II study). D R AF 10 19 20 Methotrexate 21 There is 1 Class II study (Goodkin 1995) comparing methotrexate to placebo in 60 individuals 22 with chronic progressive MS. The risk of relapse over 2 years was not significant between 59 Confidential draft: not for distribution, dissemination, or publication 1 treatments, with a risk ratio of 1.12 (0.41, 3.13). The risk of disability progression sustained for 2 2 months over 2 years was not significant between treatments, with a risk ratio of 0.81 (0.47, 1.38). 3 4 Conclusion 5 For individuals with chronic progressive MS, there is insufficient evidence to determine the 6 efficacy of methotrexate compared to placebo in reducing the risk of relapse or disability 7 progression over 2 years (very low confidence, 1 Class II study; confidence in evidence 8 downgraded due to imprecision). T 9 Mitoxantrone 11 There is 1 Class II study (Hartung 2002) comparing mitoxantrone (12 mg/m2 and 5 mg/m2) to 12 placebo in 124 individuals with worsening RRMS and SPMS. The risk of relapse over 2 years 13 was lower with mitoxantrone, with a risk ratio of 0.68 (0.48, 0.94). The risk ratio for disability 14 progression from baseline to endpoint over 2 years was 0.33 (0.13, 0.82), also favoring 15 mitoxantrone. D R AF 10 16 17 Conclusion 18 For individuals with worsening RRMS and SPMS, mitoxantrone is probably more effective than 19 placebo in decreasing the risk of relapse and disability progression over 2 years (moderate 20 confidence, 1 Class II study; confidence in evidence upgraded due to magnitude of effect). 21 22 Rituximab 60 Confidential draft: not for distribution, dissemination, or publication 1 There is 1 Class II study of rituximab vs placebo in 439 individuals with PPMS (Hawker 2009). 2 There was no difference between treatments in the proportion of individuals with confirmed 3 disease progression sustained for 3 months after 2 years of treatment, with a risk ratio of 0.78 4 (0.60, 1.02). 5 6 Conclusion 7 In individuals with PPMS, rituximab is possibly no more effective than placebo in decreasing the 8 risk of disease progression over 2 years (low confidence, 1 Class II study). T 9 Therapeutic question: What are the adverse effects of DMTs in patients with MS compared 11 with placebo (adverse event related discontinuation, and serious or life threatening adverse 12 events)? 13 Alemtuzumab 14 Meta-analysis of the 2 Class 1 (Cohen 2012, Coles 2012) and 1 Class 2 (Investigators, Coles et 15 al. 2008) studies found a risk difference of -3.8% (-6.4, -1.3) for adverse event related 16 discontinuation, with more individuals in the interferon beta 1a treatment group discontinuing 17 due to adverse events than the alemtuzumab group. There was no significant difference in the 18 risk of cancer, death, liver toxicity, serious infection, immune thrombocytopenic purpura 19 between treated groups across 24–month follow-up. There was a higher risk of “thyroid- 20 associated” events (e.g., hyperthyroidism, hypothyroidism, thyroiditis, goiter and thyroid cyst) in 21 individuals treated with alemtuzumab compared to interferon beta 1a, with a risk difference of 22 13.9% (9.3, 18.6). D R AF 10 23 61 Confidential draft: not for distribution, dissemination, or publication 1 Conclusions 2 Alemtuzumab possibly has a lower risk than interferon beta 1a of adverse event related 3 discontinuation at 24 months (low confidence, 2 Class I and 1 Class II studies). There is 4 insufficient evidence to determine the difference between alemtuzumab and interferon beta 1a in 5 the risk of cancer, death, liver toxicity, serious infection, and immune thrombocytopenic purpura 6 (very low confidence, 2 Class I and 1 Class II studies; confidence in evidence downgraded due to 7 imprecision). Alemtuzumab has a higher risk than interferon beta 1a of a thyroid-associated 8 event (high confidence, 2 Class II and 1 Class II studies). T 9 Azathioprine 11 One Class II study (Goodkin 1991) of 52 individuals with RRMS found a risk difference of 12 16.69% (-2.26, 34.69) between azathioprine and placebo for the proportion of individuals with 13 adverse event related discontinuation. 14 R AF 10 Conclusion 16 For individuals with RRMS, there is insufficient evidence to determine the risk difference 17 between azathioprine and placebo for adverse event related discontinuation at 24 months (very 18 low confidence, 1 Class II study; confidence in evidence downgraded due to imprecision). D 15 19 20 Cladribine 21 One Cass II study (Giovannoni 2010) of 1326 individual with RRMS found a risk difference of 22 0.43% (-1.20, 1.67) between cladribine and placebo for the proportion of individuals who 23 discontinued due to adverse events at 2 years. The same study found a risk difference of 1.13% 62 Confidential draft: not for distribution, dissemination, or publication 1 (0.12, 2.07) for neoplasms and a risk difference of 24.86% (21.55, 28.00) for lymphocytopenia. 2 Neoplasms in the cladribine group included uterine leiomyomas, melanoma, cervical carcinoma, 3 carcinoma of pancreas, ovarian carcinoma; no tumors were seen in the placebo group. 4 Conclusions 6 For individuals with RRMS, cladribine is possibly not different from placebo in the risk of 7 adverse event related discontinuation at 2 years (low confidence, 1 Class II study). Cladribine 8 possibly has a higher risk of neoplasms than placebo (low confidence, 1 Class II study). 9 Cladribine probably has a higher risk of lymphocytopenia than placebo (moderate confidence, 1 Class II study, confidence in evidence upgraded due to large effect size). 11 AF 10 T 5 Corticosteroids 13 One Class II study (Sorensen 2009) evaluated the number of individuals with RRMS 14 discontinuing treatment due to adverse effects with pulsed corticosteroids plus interferon vs 15 placebo plus interferon over 2 years. The risk difference was 24.15% (12.18, 36.13), with more 16 individuals receiving corticosteroids discontinuing treatment due to adverse effects. Data from 17 the same study found a risk difference between pulsed corticosteroid plus interferon and placebo 18 plus interferon of 14.82% (1.39, 27.74) for psychiatric symptoms. Data from 2 Class II studies 19 (Sorensen 2009, Ravnborg, Sorensen et al. 2010) found a risk difference of 21.9% (15.5, 28.3) 20 for sleep disturbances with corticosteroids plus interferon vs placebo plus interferon. D R 12 21 22 Conclusion 63 Confidential draft: not for distribution, dissemination, or publication 1 For individuals with RRMS, pulsed corticosteroids plus interferon probably increases the risk of 2 adverse event related discontinuation at 2 years compared to placebo plus interferon (moderate 3 confidence, 1 Class II study, confidence in evidence upgraded due to magnitude of effect). 4 Pulsed corticosteroids plus interferon possibly increases the risk of psychiatric symptoms 5 compared to placebo plus interferon (low confidence, 1 Class II study). Pulsed corticosteroids 6 plus interferon has a higher risk of sleep disturbances compared to placebo plus interferon (high 7 confidence, 2 Class II studies; confidence upgraded due to magnitude of effect). 8 Cyclophosphamide T 9 One Class III study (Killian 1988) of cyclophosphamide vs placebo evaluated the risk of adverse 11 event related discontinuation at 1 year in 14 patients with RRMS. The risk difference was - 12 12.5% (-47.09, 27.86). 13 AF 10 Conclusion 15 For individuals with RRMS, there is insufficient evidence to determine the risk difference 16 between cyclophosphamide and placebo for adverse event related discontinuation at 1 year (very 17 low confidence, 1 Class III study). D R 14 18 19 Daclizumab HYP 20 One Class I study (Gold 2013) of daclizumab HYP vs placebo evaluated the risk difference in 21 serious infections over 1 year. The risk difference was 2.88% (0.47, 6.15), with more individuals 22 with daclizumab HYP experiencing serious infections. The same study evaluated the risk 23 difference for elevation of ALT or AST beyond 5 times the normal upper limit and found a risk 64 Confidential draft: not for distribution, dissemination, or publication 1 difference of 3.84% (0.82, 7.55) with daclizumab HYP. One Class II study (Kappos 2015) 2 compared daclizumab HYP to interferon beta 1a over 3 years. The risk difference for adverse 3 event related discontinuation was 3.3% (0.001, 6.46), with more individuals in daclizumab HYP 4 than interferon discontinuing treatment. 5 Conclusions 7 For individuals with RRMS, daclizumab HYP has a higher risk of serious infection and 8 pronounced AST/ALT elevation than placebo (high confidence, 1 Class I study; confidence in 9 evidence upgraded due to magnitude of effect). Daclizumab HYP possibly importantly increases 10 the risk of adverse event related discontinuation compared to interferon (low confidence, 1 Class 11 II study). 12 AF T 6 Dimethyl Fumarate 14 One Class I (Fox 2012) and 1 Class II (Gold 2012) study evaluated the number of patients who 15 discontinued dimethyl fumarate vs placebo for adverse effects (excluding relapses) at 2 years in 16 1540 individuals with RRMS. The risk difference was 6.8% (2.8, 10.7). Data from the same 2 17 studies found a risk difference for serious infections at 2 years of 0.8% (-0.7, 2.2). D R 13 18 19 Conclusion 20 For individuals with RRMS, dimethyl fumarate possibly importantly increases the risk of 21 adverse event related discontinuation compared to placebo (low confidence, 1 Class I and 1 22 Class II study). There is insufficient evidence to determine the risk difference between dimethyl 65 Confidential draft: not for distribution, dissemination, or publication 1 fumarate and placebo for serious infections (very low confidence, 1 Class I and 1 Class II study; 2 confidence in evidence downgraded due to imprecision). 3 Fingolimod 5 There is 1 Class I (Kappos 2010) and 1 Class II (Calabresi 2014) study comparing fingolimod to 6 placebo in 1556 individuals with RRMS. The risk difference for adverse event related 7 discontinuation at 2 years was 3.6% (-4.4, 11.5). The risk difference for lower respiratory tract or 8 lung infection was 3.0% (0.2, 5.9), and the risk difference for having an ALT 5 times normal was 9 1.1% (0.94, 3.33). The risk difference for death was -0.4% (-1.73, 0.48) and neoplasms was T 4 1.23% (-1.23, 4.1). There is 1 Class I study (Cohen 2010) comparing fingolimod to interferon 11 beta 1a in 860 individuals with RRMS. The risk difference for adverse event related 12 discontinuation at 1 year was 1.88% (-0.99, 4.84), and was 1.63% for neoplasms (0.22, 3.41). 13 AF 10 Conclusion 15 In individuals with RRMS, fingolimod is possibly no different from placebo in the risk of 16 adverse event related discontinuation at 2 years (low confidence, 1 Class I and 1 Class II study; 17 confidence in evidence downgraded due to imprecision). Fingolimod increases the risk of lower 18 respiratory tract or lung infection compared to placebo (high confidence, 1 class I and 1 Class II 19 study; confidence in evidence upgraded due to magnitude of effect). Fingolimod possibly 20 increases the risk of having an ALT five times normal compared to placebo (low confidence, 1 21 Class II study). There is insufficient evidence to determine the risk difference between 22 fingolimod and placebo for death and neoplasms (very low confidence, 1 Cass II study; 23 confidence in evidence downgraded due to imprecision). Fingolimod is probably not different D R 14 66 Confidential draft: not for distribution, dissemination, or publication 1 from interferon beta 1a in the risk of adverse event related discontinuation at 1 year (moderate 2 confidence, 1 class I study). Fingolimod has a higher risk than interferon beta 1a of neoplasm 3 (high confidence, 1 Class I study, confidence in evidence upgraded due to magnitude of effect). 4 Glatiramer 6 There are 1 Class I (Fox 2012) and 2 Class II (Bornstein 1987, Johnson 1995) studies comparing 7 glatiramer to placebo for adverse event related discontinuation over 2 years in 1015 individuals 8 with RRMS. The difference in treatment discontinuation was not significant, with a risk 9 difference of 2.2% (-1.4, 5.8). There is 1 Class II study of 756 individuals with RRMS T 5 comparing adverse event related discontinuation with glatiramer vs Interferon beta 1a SQ 3 times 11 a week (Mikol 2008). There was no significant difference between treatments in discontinuation, 12 with a risk difference of 0.97% (-2.39, 4.35). There is 1 Class II study of 1333 individuals with 13 RRMS comparing injection site reactions and influenza-like illness with glatiramer vs interferon 14 beta 1b SQ alternate day (250 ug) (O'Connor 2009). The risk of injection site reactions was 15 higher with glatiramer than interferon beta 1b SQ alternate day, with a risk difference of 10.12% 16 (4.44, 15.67). The risk of influenza-like illness was higher with interferon beta 1b SQ alternate 17 day that glatiramer, with a risk difference of 34.81% (30.74, 38.53). D R AF 10 18 19 Conclusions 20 For individuals with RRMS, glatiramer is probably not different from placebo in the risk of 21 adverse event related discontinuation at 2 years (moderate confidence, 1 Class I and 2 Class II 22 studies). Glatiramer is possibly not different from interferon beta 1a SQ 3 times a week in the 23 risk of adverse event related discontinuation (low confidence, 1 Class II study). Interferon beta 67 Confidential draft: not for distribution, dissemination, or publication 1 1b SQ alternate day possibly importantly has a lower risk of injection site reactions than 2 glatiramer (low confidence, 1 Class II study). Interferon beta 1b SQ alternate day has a higher 3 risk of influenza-like illness than glatiramer (moderate confidence, 1 Class II study; confidence 4 in evidence upgraded due to magnitude of effect). 5 Immunoglobulins 7 There is 1 Class I study (Fazekas 1997) evaluating the risk of adverse event related 8 discontinuation at 24 months with immunoglobulin vs placebo in 150 individuals with RRMS. 9 The risk difference was 2.67% (-3.74, 9.86). AF 10 T 6 Conclusion 12 In individuals with RRMS, immunoglobulins are probably no different from placebo in the risk 13 of adverse event related discontinuation at 2 years (moderate confidence, 1 Class I study). R 11 14 Interferon beta 1a intramuscular weekly 16 Meta-analysis of the 1 Class I (Vollmer 2014) and one class 2 (Jacobs 1996) studies found that 17 interferon beta 1a intramuscular weekly was not associated with a higher risk of adverse event 18 related discontinuation than placebo at 2 years, with a risk difference of 3.0% (-1.4, 7.4). D 15 19 20 Conclusion 21 For individuals with RRMS, Interferon beta 1a intramuscular weekly is probably not different 22 from placebo in the risk of adverse event related discontinuation (moderate confidence, 1 Class I 23 and 1 Class II study). 68 Confidential draft: not for distribution, dissemination, or publication 1 Interferon beta 1a SQ 3 times a week 3 There is 1 Class I study (PRISMS 1998) comparing interferon beta 1a SQ 3 times a week to 4 placebo in 560 individuals with RRMS. The risk difference in adverse event related 5 discontinuation at 2 years was 2.95% (-0.21, 5.58). There is 1 Class I study (SPECTRIMS 2001) 6 comparing interferon beta 1a SQ 3 times a week to placebo in 619 individuals with SPMS. The 7 risk difference in adverse event related discontinuation over 3 years was 5.55% (1.69, 8.88). 8 There is 1 Class II study (Panitch 2002) comparing interferon beta 1a SQ 3 times a week to 9 interferon beta 1a intramuscular weekly in 677 individuals with RRMS. The risk difference in 10 adverse event related discontinuation over 1 year was 0.28% (-3.27, 3.83). The risk of injection 11 site reactions was higher with interferon beta 1a SQ 3 times a week than interferon beta 1a 12 intramuscular weekly, with a risk difference of 15.31% (9.67, 20.90). 13 AF T 2 Conclusion 15 In patients with RRMS, interferon beta 1a SQ 3 times a week is possibly no different from 16 placebo in the risk of adverse event related discontinuation at 2 years (low confidence, 1 Class I 17 study; confidence in evidence downgraded due to imprecision). In patients with SPMS, 18 interferon beta 1a SQ 3 times a week possibly importantly increases the risk of adverse event 19 related discontinuation over 3 years (low confidence, 1 Class I study). In patients with RRMS, 20 interferon beta 1a SQ 3 times a week is possibly no different from interferon beta 1a 21 intramuscular weekly in the risk of adverse event related discontinuation at 1 year (low 22 confidence, 1 Class II study). In patients with RRMS, interferon beta 1a SQ 3 times a week D R 14 69 Confidential draft: not for distribution, dissemination, or publication 1 possibly importantly increases the risk of injection site reaction compared to interferon beta 1a 2 intramuscular weekly (low confidence, 1 Class II study). 3 Interferon Beta 1b SQ alternate day 5 One Class II study (Group 1993) of 227 individuals with RRMS found a risk difference of 7.25% 6 (2.11, 13.43) between interferon Beta 1b SQ alternate day and placebo for the proportion of 7 individuals with adverse event related discontinuation. One Class II study of 718 individuals 8 with SPMS found a risk difference of 8.42% (4.47, 12.56) between interferon beta 1b SQ 9 alternate day and placebo for the proportion of individuals with adverse event related T 4 discontinuation (Kappos 1998). The risk of injection site reactions was higher in all studies of 11 individuals with RRMS, SPMS and PPMS, with a risk difference of 63.66% (53.41, 71.64), 12 38.09% (31.1, 44.52) and 77.78% (59.34, 88.28) respectively. 13 AF 10 Conclusion 15 In patients with RRMS and SPMS, interferon beta 1b SQ alternate day possibly importantly 16 increases the risk of adverse event related discontinuation (low confidence, 1 Class II study for 17 each MS subtype). In patients with RRMS, SPMS, and PPMS, interferon beta 1b SQ alternate 18 day probably has a higher risk of injection site reactions than placebo (moderate confidence, 1 19 Class II study for each MS subtype; confidence in evidence upgraded due to magnitude of 20 effect). D R 14 21 22 Mitoxantrone 70 Confidential draft: not for distribution, dissemination, or publication 1 There is 1 class II study (Hartung 2002) assessing the risk of adverse effects with mitoxantrone 2 compared to placebo over 2 years in 126 individuals with relapsing progressive MS. 3 Mitoxantrone use was associated with a higher risk of amenorrhea (risk difference 28.00%; 4 10.40, 47.58), nausea and vomiting (risk difference 55.49%; 38.97, 67.52), alopecia (risk 5 difference, 30.04%; 12.65, 45.01), urinary tract infections (19.76%; 5.21, 33.52), and leukopenia 6 (19.35%; 9.62, 30.85). 7 Conclusion 9 In individuals with relapsing progressive MS, mitoxantrone probably importantly increases the 10 risk of amenorrhea, nausea and vomiting, and alopecia (moderate confidence, 1 Class II study, 11 confidence in evidence upgraded due to magnitude of effect). Mitoxantrone possibly importantly 12 increases the risk of urinary tract infections and leukopenia (low confidence, 1 Class II study). 13 AF T 8 Natalizumab 15 There is 1 Class I study (Polman 2006) comparing natalizumab to placebo in 942 individuals 16 with RRMS. There was no difference between natalizumab and placebo in the risk of adverse 17 event related discontinuation, with a risk difference of 3.2% (-0.6, 6.5). No significant risk 18 difference was reported in neoplasms (0.5%; -1.1, 1.6), or death (0.3%; -0.9, 1.2). D R 14 19 20 Conclusion 21 In individuals with RRMS, natalizumab is possibly no different from placebo in the risk of 22 adverse event related discontinuation over 2 years (low confidence, 1 Class I study). There is 23 insufficient evidence to determine the difference between natalizumab and placebo in the risk of 71 Confidential draft: not for distribution, dissemination, or publication 1 neoplasm or death (very low confidence, 1 Class I study; confidence in evidence downgraded 2 due to imprecision). 3 Pegylated Interferon 5 There is 1 Class I study (Calabresi 2014) comparing pegylated interferon 125 mcg every 2 weeks 6 to placebo in 1012 individuals with RRMS. The risk difference in adverse event related 7 discontinuation at 1 year was 3.48% (1.35, 5.82). The risk difference for severe adverse events, 8 defined as symptoms that cause severe discomfort, incapacitation, or a significant effect on daily 9 life was 6.98% (2.69, 11.25). T 4 AF 10 Conclusions 12 In patients with RRMS, pegylated interferon possibly importantly increases the risk of adverse 13 event related discontinuation at 1 year (low confidence, 1 Class I study). In patients with RRMS, 14 pegylated interferon importantly increases the risk of severe adverse events (high confidence, 1 15 Class I study, confidence in evidence upgraded due to magnitude of effect). D R 11 16 17 Rituximab 18 There is 1 Class II study comparing rituximab to placebo in 104 individuals with RRMS (Hauser, 19 Waubant et al. 2008). The risk difference in adverse event related discontinuation from the study 20 was -1.37% (-14.57, 7.35) and for an infection-associated serious adverse event was 2.82% (- 21 15.88, 5.37). 22 23 Conclusion 72 Confidential draft: not for distribution, dissemination, or publication 1 In individuals with RRMS, there is insufficient evidence to determine the risk difference between 2 rituximab and placebo for adverse event related withdrawal and infection associated serious 3 adverse events (very low confidence, 1 Class II study; confidence in evidence downgraded due 4 to imprecision). 5 Teriflunomide 7 There is 1 Class II study (O'Connor 2011) comparing teriflunomide to placebo in 1088 8 individuals with RRMS for adverse event related discontinuation. The risk difference in adverse 9 event related discontinuation over 2 years was 2.36% (-1.47, 5.75). There is 1 Class II study 10 (Confavreux 2014) comparing teriflunomide to placebo in 1165 individuals with RRMS for 11 serious infections. The risk difference for serious infections was 0.35% (-2.07, 2.29). 12 AF T 6 Conclusion 14 For individuals with RRMS, teriflunomide is possibly no different from placebo in the risk of 15 adverse event related discontinuation (low confidence, 1 Class II study). There is insufficient 16 evidence that teriflunomide is different from placebo in the risk of serious infections (very low 17 confidence, 1 Class II study; confidence in evidence downgraded due to imprecision). D R 13 18 19 Therapeutic Question: In people with clinically isolated syndromes (CIS), are DMTs 20 superior to placebo in decreasing the risk of conversion to clinically definite MS (CDMS)? 21 Cladribine 73 Confidential draft: not for distribution, dissemination, or publication 1 There is 1 Class II study (Leist 2014) comparing oral cladribine 5.25 mg/kg to placebo in 410 2 individuals with CIS. The risk of conversion to CDMS over 2 years was significantly lower with 3 cladribine, with a risk ratio of 0.43 (0.29, 0.62). 4 5 Conclusion 6 For individuals with CIS, cladribine is probably more effective than placebo in reducing the 7 proportion of patients converting to CDMS over 3 years (moderate confidence, 1 Class II study; 8 confidence in evidence upgraded due to magnitude of effect). T 9 Glatiramer 11 There is 1 Class I study (Comi, Martinelli et al. 2009) comparing glatiramer to placebo in 481 12 individuals with CIS. The risk of conversion to CDMS over 3 years was significantly lower with 13 glatiramer, with a risk ratio of 0.58 (0.44, 0.75). 14 R AF 10 Conclusion 16 For individuals with CIS, glatiramer is more effective than placebo in reducing the proportion of 17 patients converting to CDMS over 3 years (high confidence, 1 Class I study; confidence in 18 evidence upgraded due to magnitude of effect). D 15 19 20 Immunoglobulins 21 There is 1 Class II study (Achiron, Kishner et al. 2004) comparing IVIG to placebo in 90 22 individuals with CIS. The risk of conversion to CDMS over 1 year was significantly lower with 23 IVIG, with a risk ratio of 0.50 (0.28, 0.88). 74 Confidential draft: not for distribution, dissemination, or publication 1 2 Conclusion 3 For individuals with CIS, immunoglobulin is probably more effective than placebo in reducing 4 the proportion of patients converting to CDMS over 1 year (moderate confidence, 1 Class II 5 study; confidence in evidence upgraded due to magnitude of effect). 6 Interferon beta 1a intramuscular weekly 8 There is 1 Class II study (Jacobs, Beck et al. 2000) comparing interferon beta 1a intramuscular 9 weekly to placebo in 383 individuals with CIS. The risk of conversion to CDMS over 3 years 10 was significantly lower with interferon beta 1a intramuscular weekly, with a risk ratio of 0.71 11 (0.56, 0.89). 12 AF T 7 Conclusion 14 For individuals with CIS, interferon beta 1a intramuscular weekly is probably more effective 15 than placebo in reducing the proportion of patients converting to CDMS over 3 years (moderate 16 confidence, 1 Class II study; confidence in evidence upgraded due to magnitude of effect). D R 13 17 18 Interferon beta 1a SQ 3 times a week 19 There is 1 Class I study (Comi, Filippi et al. 2001) comparing interferon beta 1a SQ 44 mcg 3 20 times per week to placebo in 342 individuals with CIS. The risk of conversion to CDMS over 2 21 years was significantly lower with interferon beta 1a SQ 3 times a week, with a risk ratio of 0.55 22 (0.38, 0.78). 23 75 Confidential draft: not for distribution, dissemination, or publication 1 Conclusion 2 For individuals with CIS, interferon beta 1a SQ 3 times a week is more effective than placebo in 3 reducing the proportion of patients converting to CDMS over 2 years (high confidence, 1 Class I 4 study; confidence in evidence upgraded due to magnitude of effect). 5 Interferon Beta 1b SQ alternate day 7 There is 1 Class II study (Kappos, Polman et al. 2006) comparing interferon beta 1b SQ alternate 8 day to placebo in 468 individuals with CIS. The risk of conversion to CDMS over 2 years was 9 significantly lower with interferon beta 1b SQ alternate day, with a risk ratio of 0.59 (0.46, 0.76). T 6 AF 10 Conclusion: 12 For individuals with CIS, interferon beta 1b SQ alternate day is probably more effective than 13 placebo in reducing the proportion of patients converting to CDMS over 2 years (moderate 14 confidence, 1 Class II study; confidence in evidence upgraded due to magnitude of effect). R 11 D 15 16 Teriflunomide 17 There is 1 Class II study comparing teriflunomide 14 mg to placebo in 411 individuals with CIS 18 (Miller 2014). The risk of conversion to CDMS over 2 years was significantly lower with 19 teriflunomide, with a risk ratio of 0.64 (0.44, 0.91). 20 21 Conclusion 76 Confidential draft: not for distribution, dissemination, or publication 1 For individuals with CIS, teriflunomide is probably more effective than placebo in reducing the 2 proportion of patients converting to CDMS over 2 years (moderate confidence, 1 Class II study; 3 confidence in evidence upgraded due to magnitude of effect). 4 PRACTICE RECOMMENDATIONS 6 This section translates the conclusions into recommendations, or action statements. The 7 recommendations are statements that result from a modified Delphi process performed by the 8 panel members. The panel responds to key questions that capture key elements, including the 9 available evidence, strong evidence from closely related conditions, inferences from one or more T 5 of the other statements, and established principles of care. The recommendations are rated to 11 indicate their levels of strength (Level A, B, C, U, and R). Recommendations should not be 12 broader or narrower than the clinical question. Each recommendation or set of 13 recommendations is preceded by a clinical context section that delineates the justification for the 14 recommendations and their corresponding levels. R AF 10 D 15 16 Starting DMT Recommendations 17 Starting Recommendation 1 18 Rationale/Clinical Context: Receiving the diagnosis of multiple sclerosis (MS) is a stressful life 19 event (PRIN) (Schofield et al. 2003; Johnson 2003). People receiving major diagnoses may not 20 recall much of the information given to them at the time (Ptacek 1996). Providing information 21 about disease modifying therapy (DMT) at a follow-up interaction is likely to allow a better 22 understanding of these medications and their risks and benefits (PRIN). 23 77 Confidential draft: not for distribution, dissemination, or publication 1 Statement 1 2 Clinicians should counsel patients just diagnosed with MS about specific treatment options with 3 DMT at a dedicated treatment visit. 4 5 Starting Recommendation 2 6 Rationale/Clinical Context: Respecting patient preferences is an important component of care for 7 chronic conditions (PRIN). Given the variety of DMTs available (EVID), evaluating patient 8 preferences may improve acceptance of and adherence to DMT (INFER). T 9 Statement 2a 11 Clinicians must ascertain and incorporate/review preferences in terms of safety, route of 12 administration, lifestyle, cost, efficacy, common side effects and tolerability in the choice of 13 DMT in patients with MS being considered for DMT. 14 R AF 10 Statement 2b 16 Clinicians must engage in an ongoing dialogue regarding treatment decisions throughout the 17 course of the disease with patients with MS. D 15 18 19 Starting Recommendation 3 20 Rationale/Clinical Context: DMTs reduce but do not eliminate relapses and MRI activity 21 (EVID). Educating patients about realistic expectations regarding DMT effects is important 22 (PRIN) (Mohr 1996). Clinicians should inform patients that they may still need symptomatic 78 Confidential draft: not for distribution, dissemination, or publication 1 treatment in addition to DMT (Henze, Rieckmann, Toyka, & Multiple Sclerosis Therapy 2 Consensus Group of the German Multiple Sclerosis 2006)(PRIN). 3 4 Statement 3 5 Clinicians should counsel that DMTs are prescribed to reduce relapses and new MRI lesion 6 activity. DMTs are not prescribed for symptom improvement in patients with MS. 7 Starting Recommendation 4 9 Rationale/Clinical Context: Given that DMT use requires commitment to ongoing therapy and an T 8 understanding of side effects (PRIN) (EVID), readiness to initiate DMT and factors causing 11 reluctance may have an impact on adherence to DMT use (INFER). AF 10 12 Statement 4 14 Clinicians should evaluate readiness or reluctance to initiate DMT and counsel on its importance 15 in patients with MS who are candidates to initiate DMT. D R 13 16 17 Starting Recommendation 5 18 Rationale/Clinical Context: In patients with MS, comorbid disease and adverse health behaviors 19 (e.g., depression and anxiety, vascular risk factors, physical inactivity, smoking) are associated 20 with worse outcomes (Marrie & Horwitz 2010; Tettey et al. 2016) (RELA). Similarly, addressing 21 depression prior to initiating DMT may improve decision making and adherence to DMT 22 (INFER). Concomitant medications may have important interactions with DMTs (R. H. Thomas 23 & Wakefield 2015) (RELA). 79 Confidential draft: not for distribution, dissemination, or publication 1 2 Statement 5 3 Clinicians should counsel about comorbid disease and adverse health behaviors and potential 4 interactions of the DMT with concomitant medications when patients with MS initiate DMTs. 5 Starting Recommendation 6 7 Rationale/Clinical Context: Because DMT requires adherence to treatment to provide full 8 efficacy (INFER), and because that adherence to treatment may be an issue for patients with MS 9 (Katsarava et al. 2015; Irwin 2015) (RELA), discussing adherence issues prior to initiating DMT T 6 is good clinical practice (INFER). Efforts to increase adherence may improve outcomes 11 (INFER). AF 10 12 Statement 6a 14 Clinicians should evaluate barriers to adherence to DMT in patients with MS. R 13 D 15 16 Statement 6b 17 Clinicians should counsel on the importance of adherence to DMT when patients with MS 18 initiate DMTs. 19 20 Starting Recommendation 7 21 Rationale/Clinical Context: Evidence from multiple Class I and II trials confirms that DMTs are 22 associated with a significant delay in second clinical relapse or new brain MRI lesions in patients 23 with a first demyelinating event (EVID). There is insufficient evidence concerning the 80 Confidential draft: not for distribution, dissemination, or publication comparative efficacy of specific DMTs for this purpose. Decisions concerning selection of 2 specific DMTs for patients presenting with a first demyelinating event should abide by 3 prescribing principles espoused in other recommendations. Patients presenting with a first 4 demyelinating event and who do not meet the 2010 International Criteria for MS are commonly 5 encountered in clinical practice. Multiple prospective observational trials have consistently 6 confirmed that patients with a single clinical demyelinating event with two or more brain or 7 spinal cord lesions remain at increased risk of a future diagnosis of MS, with the highest risk 8 incurred within 5 years of the sentinel event (Miller, Barkhof, Montalban et al., 2005; Optic 9 2008) (RELA). T 1 AF 10 Statement 7 12 Clinicians may offer disease modifying therapy for patients with a single clinical demyelinating 13 event with two or more brain or spinal cord lesions that have imaging characteristics consistent 14 with MS. D 15 R 11 16 Starting Recommendation 8 17 Rationale/ Clinical context: In currently untreated patients with clinically isolated syndromes or 18 relapsing forms of MS who have not had relapses in two or more years and do not have active 19 new MRI lesion activity on a recent MRI, the benefit of initiating DMT has not been studied 20 (EVID). In such patients, the harm of initiating DMTs including side effects, major adverse 21 events, and burden of taking a long-term medication that may not be beneficial may outweigh the 22 potential benefit of reducing relapse rate (INFER). 23 81 Confidential draft: not for distribution, dissemination, or publication 1 Statement 8 2 Clinicians may recommend serial imaging and close follow-up rather than initiating DMT in 3 patients with CIS or relapsing forms of MS not on DMT who have not had relapses in the past 2 4 years and who do not have active new MRI lesion activity on a recent MRI. 5 Starting Recommendation 9 7 Rationale/Clinical Context: Multiple studies of disease modifying agents in patients with 8 relapsing forms of MS who have had recent relapses and or MRI activity (EVID) have shown 9 benefit of DMT in terms of reducing relapses and reducing MRI activity (EVID). This includes T 6 patients with a single clinical episode who meet 2010 International Criteria (EVID) (Comi, 11 DeStefano, Freedman et al. 2012; Miller, Wolinsky, Kappos et al. 2014). AF 10 12 Statement 9 14 Clinicians should offer DMTs to patients with relapsing forms of MS with recent clinical 15 relapses or MRI activity. D R 13 16 17 Starting Recommendation 10 18 Rationale/Clinical Context: Lack of adherence to treatment of chronic diseases is a wide-ranging 19 problem (RELA). The result of poor adherence is reduced effectiveness and increased healthcare 20 costs (RELA) (Singer, 2008; Brandes 2013; Girouard 2011; Oleen-Burkey 2011; Thomas, 21 Curkendall, Farr et al. 2016) (RELA). Regular interactions and assessments by clinicians 22 facilitate prompt identification and treatment of side effects, increased tolerability of the 23 medication, and safety monitoring (RELA) (Bender 2002; Mohr 1996; Treadaway 2009). Some 82 Confidential draft: not for distribution, dissemination, or publication 1 MS DMTs have specific risk evaluation and mitigation strategies with recommendations for 2 follow-up frequency (RELA) (FDA, 2013, 2014, 2015b). 3 4 Statement 10a 5 Clinicians should monitor for medication adherence, side effects, tolerability, safety and 6 effectiveness of the therapy in patients with MS on DMTs. 7 Statement 10b 9 Clinicians should follow up either annually or according to medication-specific risk evaluation and mitigation strategies (REMS) in patients with MS on DMTs. AF 10 T 8 11 Starting Recommendation 11 13 Rationale/Clinical Context: DMT have potential risks in pregnant women (Bove et al. 2014) 14 (RELA) to varying degrees. Discussing pregnancy with patients with MS before initiating DMT 15 is good clinical practice (INFER). If patients with MS are planning pregnancy soon, DMT may 16 need to be deferred until after pregnancy (Coyle 2016) (RELA). In addition, since DMTs vary in 17 terms of pregnancy risks from Category B to Category X (Bove et al. 2014) (RELA), the choice 18 of DMT may depend on plans for pregnancy (INFER). D R 12 19 20 Statement 11 21 Clinicians should monitor patient’s reproductive plans and counsel on reproductive risks and on 22 use of birth control while on a DMT in women of childbearing years with MS. 23 83 Confidential draft: not for distribution, dissemination, or publication 1 Starting Recommendation 12 2 Rationale/clinical context: Chemotherapy such as cyclophosphamide may affect male fertility 3 (Leroy et al. 2015) (RELA); thus, male patients starting cyclophosphamide should be counseled 4 about the risk of infertility. With teriflunomide treatment, there may be a risk of teratogenicity 5 from male sperm, which could extend for 2 years after treatment cessation if the patient is not 6 treated with chelation therapy (Cree 2013) (RELA). 7 Statement 12 9 Clinicians must counsel men with MS on their reproductive plans regarding treatment implications before initiating teriflunomide or cyclophosphamide. AF 10 T 8 11 Starting Recommendation 13 13 Rationale/clinical context: Post-approval, new evidence has shown a high risk of 14 cardiomyopathy, ovarian failure, male infertility, chromosomal aberrations and promyelocytic 15 leukemia (RELA) (Ellis, Boggild 2009; Ellis, Brown, Boggild 2014; Fleischer 2014; Le Page, 16 Leray, Edan, & French Mitoxantrone Safety Group 2011) associated with the use of 17 mitoxantrone. Other effective and lower risk medications, which we not available at the time of 18 FDA–approval for mitoxantrone (RELA), are available for use in MS. D R 12 19 20 Statement 13 21 Clinicians should not prescribe mitoxantrone to patients for MS. 22 23 Starting Recommendation 14 84 Confidential draft: not for distribution, dissemination, or publication Rationale/Clinical Context: MS is a heterogeneous disease and is characterized by highly 2 variable degrees of disease activity in the relapsing phase as well as varying rates of worsening 3 during the progressive phases (Confavreux & Vukusic 2006; Lucchinetti et al. 2000; Lucchinetti 4 2000) (RELA). Definitions of very active MS vary and can include measures of relapsing 5 activity and MRI markers of disease activity such as numbers of contrast enhancing lesions 6 (Schmidt 2011) (RELA). Subgroup analyses from phase III pivotal trials of alemtuzumab, 7 fingolimod, and natalizumab showed a reduction in relapses and MRI measures in patients with 8 very active disease (EVID) (RELA) (Coles, Fox, Vladic et al. 2011; Devonshire, Havdrova, 9 Radue et al. 2012; Hutchinson 2009). Compared with interferon therapy, treatment with these T 1 therapies resulted in more favorable outcomes in the subgroup of patients with very active 11 disease (RELA) (Cohen, Barkhof, Comi et al. 2010; Cohen, Coles, Arnold et al. 2012; Coles, 12 Twyman, Arnold et al. 2012; Rudick 2006). However, the risks and benefits of each treatment 13 strategy need to be considered on a patient-by-patient basis (PRIN). 14 R AF 10 Statement 14 16 Clinicians should prescribe alemtuzumab, fingolimod or natalizumab for patients with very 17 active MS. D 15 18 19 Starting Recommendation 15 20 Rationale/Clinical Context: DMTs should be available to all patients with relapsing forms of MS 21 (PRIN). Due to disparities in health care provision in different settings (RELA) (MSIF 2013, 22 Browne 2014), there may be situations where approved DMTs are not available to an individual. 23 In this situation, there may be support from the pharmaceutical industry or organizations, such as 85 Confidential draft: not for distribution, dissemination, or publication 1 the National Organization of Rare Diseases (NORD), county organizations, or government 2 organizations, to obtain DMTs. If this is not available, certain medications have a lower cost and 3 may become a choice for care. Azathioprine has mixed result and low confidence evidence to 4 support efficacy in relapsing forms of MS (EVID). Cladribine has evidence for benefit for the 5 both the oral and parental formulations but currently only the parenteral formulations are 6 available (EVID). Teriflunomide is the active metabolite of leflunomide (RELA) (Claussen 7 2012). Teriflunomide is approved for the treatment of relapsing forms of MS (EVID). 8 T 10 Statement 15a Clinicians may direct patients with MS who are candidates for DMTs to support programs. 11 AF 9 Statement 15b 13 Clinicians should recommend one of the following DMTs (azathioprine, cladribine, minocycline, 14 leflunomide) for patients with relapsing forms of MS who do not have access to approved DMTs R 12 D 15 16 Starting Recommendation 16 17 Rationale/Clinical Context: Patients with a positive John Cunningham (JC) virus antibody have a 18 higher risk of developing progressive multifocal leukoencephalopathy (PML) while on 19 natalizumab, particularly after the second year on treatment or in patients who have had prior 20 immunosuppressive treatment (EVID). There are now other highly effective treatments that may 21 be used that have not been shown to have a similar risk of PML (EVID). 22 23 Statement 16 86 Confidential draft: not for distribution, dissemination, or publication 1 Clinicians should not initiate natalizumab treatment in patients with positive JC virus antibodies 2 in serum unless there are overwhelming reasons to do so. 3 Starting Recommendation 17 5 Rationale/Clinical Context: Rationale/Clinical Context: No currently approved DMTs have 6 demonstrated efficacy in PPMS (EVID). Interferon-beta, glatiramer acetate and rituximab are 7 possibly ineffective (EVID). The use of ineffective therapy may pose harms to the affected 8 individual, society and the health system (PRIN). Newer agents are being studied and ongoing 9 clinical trials may alter these findings (INFER). T 4 AF 10 11 Statement 17 12 For patients with PPMS clinicians should not initiate currently approved DMTs. 13 Switching DMT Recommendations (21 statements) 15 Switching Recommendation 1 16 Clinical Context. Ongoing disease activity, measured either by clinical relapses or new MRI 17 lesions (including unequivocally new T2 or new contrast-enhancing lesions), could lead to 18 physical or cognitive worsening over time (Fusniku et al.; Rocca et al.; Rio et al.; Prosperini et 19 al.) (RELA). Now that several DMTs are available and have demonstrated efficacy for the 20 prevention of clinical relapses and new MRI lesions (EVID), physicians and patients often face 21 the decision of switching from one DMT to another due to a perceived lack of efficacy. Such 22 lack of response to a DMT has been difficult to define, as most patients are not free of all disease 23 activity, and investigators have considered using the number of clinical attacks or new MRI D R 14 87 Confidential draft: not for distribution, dissemination, or publication 1 lesions in the last 12 months (RELA) (Rio et al.; Prosperini et al.). DMTs take a variable amount 2 of time to become clinically active, and new lesion formation may occur after initiation but 3 before the time of full efficacy, confounding interpretation of follow-up MRIs (RELA) (Cohen, 4 Barkhof, Comi et al. 2010; Coles, Twyman, Arnold et al. 2012; Comi, Filippi, Wolinksy et al. 5 2001; Confavreaux 2014; Johnson, Brooks, Cohen et al. 1995; PRISMS 1998). 6 Statement 1a 8 Clinicians should periodically monitor MRI disease activity during the first 5 years from the 9 clinical onset of disease to detect the accumulation of new lesions in order to inform treatment decisions in patients with MS on DMTs. 11 AF 10 T 7 Statement 1b 13 Clinicians should recognize that relapses or new MRI lesions may develop after initiation of a 14 DMT and before the treatment becomes effective in patients with MS on DMTs. R 12 D 15 16 Statement 1c 17 Clinicians should discuss switching from one DMT to another in patients who have been on a 18 DMT long enough to take full effect and are adherent to their therapy when a patient has 19 experienced one or more relapses, two or more unequivocally new MRI lesions, or increased 20 disability on examination, over a 1-year period on a DMT. 21 22 Switching Recommendation 2 88 Confidential draft: not for distribution, dissemination, or publication 1 Rationale/Clinical Context: None of the available DMTs are completely effective against 2 relapses and MRI activity (EVID). When a patient shows breakthrough disease (continued 3 relapses or MRI activity), trying a medication with a different mechanism or efficacy profile may 4 be beneficial (INFER). Although all possible clinical scenarios cannot be answered by drug 5 trials, current evidence supports higher efficacy of alemtuzumab, natalizumab, and fingolimod, 6 compared to previously approved self-injectable DMTs (EVID high). Tolerability and likelihood 7 of adherence are other factors that are important in decisions on switching DMTs (PRIN). 8 Physician judgment and patient preferences are critical in this process (PRIN). T 9 Statement 2 11 Clinicians may evaluate the amount of disease activity, adherence, adverse event profiles, and 12 mechanism of action of DMTs when switching DMTs in patients with breakthrough disease on 13 DMTs. AF 10 R 14 Switching Recommendation 3 16 Rationale/Clinical Context: Multiple DMTs are available for MS (EVID). Switching therapies 17 may be appropriate in patients with MS who are experiencing side effects or complications with 18 a DMT (INFER). Adherence to injectable DMTs is often incomplete (Devonshire et al. 2011) 19 (RELA). Injection fatigue (physical or emotional) or injection-related pain or discomfort (EVID) 20 may be a common reason for poor adherence (INFER). D 15 21 22 Statement 3 89 Confidential draft: not for distribution, dissemination, or publication 1 Clinicians should discuss a change to a non-injectable or less frequently injectable DMT in 2 patients who report pain with the injections or in those who report “injection fatigue” on 3 injectable DMTs. 4 5 Switching Recommendation 4 6 Rationale/Clinical Context: Adherence to a DMT may also be affected by medication adverse 7 effects (RELA) (Irwin 2015; Treadaway 2009). All DMTs have common adverse effects that 8 may affect adherence (refer to table) (EVID). T 9 Statement 4a 11 Clinicians should inquire about medication adverse effects with patients with MS who are taking 12 a DMT and attempt to manage these adverse effects, as appropriate. 13 AF 10 Statement 4b 15 Clinicians should discuss a medication switch with patients for whom these adverse effects 16 negatively influence adherence. D R 14 17 18 Switching Recommendation 5 19 Rationale/Clinical Context: Persistent laboratory abnormalities such as elevated liver enzymes 20 and decreased white cell counts may prompt a discussion on switching DMT (refer to table) 21 (EVID). 22 23 Statement 5a 90 Confidential draft: not for distribution, dissemination, or publication 1 Clinicians should monitor laboratory abnormalities found on requisite laboratory surveillance (as 2 outlined in the medication's package insert) in patients with MS who are on a DMT. 3 4 Statement 5b 5 Clinicians should discuss switching DMT or reducing dose or frequency (where there is data on 6 different doses [e.g., interferons, teriflunomide, azathioprine) when there are persistent 7 laboratory abnormalities. 9 Switching Recommendation 6 T 8 Rationale/Clinical Context: PML is a serious safety concern (Clifford et al. 2010) (RELA) that 11 may affect compliance and necessitate consideration of a treatment switch (INFER). The risk of 12 PML is estimated at 4/1000 overall with natalizumab (Biogen 2016), however, the presence of 13 JCV antibodies, longer duration use, and prior immunosuppression increase the risk of PML with 14 natalizumab even further (RELA) (Clifford et al. 2010). There are rare reports of this infection 15 with the use of both fingolimod and dimethyl fumarate (RELA) (FDA 2015a) (Rosenkranz, 16 Novas, & Terborg, 2015) (Nieuwkamp et al. 2015). D R AF 10 17 18 Statement 6a 19 Clinicians should counsel patients considering natalizumab, fingolimod, and dimethyl fumarate 20 about the risk of PML associated with these agents. 21 22 Statement 6b 91 Confidential draft: not for distribution, dissemination, or publication 1 Clinicians should discuss switching to a DMT with a lower risk of PML in patients taking 2 natalizumab who are JC virus antibody positive or become JC virus antibody positive while on 3 therapy. 4 Switching Recommendation 7 6 Rationale/Clinical Context: Immunosuppressive medications may come with a risk of 7 opportunistic infection and malignancy, especially with prolonged use of the agent (EVID, 8 PRIN). These risks are often undefined with newer medication (INFER). Cases of cryptococcal 9 infections have been reported with the use of fingolimod (Huang 2015) (RELA). Herpes family 10 virus infections have been reported with fingolimod and natalizumab (Pelletier & Hafler 2012; 11 Ratchford, Costello, Reich & Calabresi 2012; Fine, Sorbello, Kortepeter & Scarazzini 2013) 12 (RELA). A potential increased risk of basal cell carcinoma was recently added to the product 13 label of fingolimod (RELA) (FDA 2016). 14 R AF T 5 Statement 7a 16 Clinicians should counsel that new DMTs without long-term safety data have an undefined risk 17 of malignancy and infection in patients starting or using new DMTs. D 15 18 19 Statement 7b 20 If a patient with MS develops a malignancy while on a DMT, clinicians should promptly discuss 21 switching to an alternate DMT, especially for patients on azathioprine, methotrexate, 22 mycophenolate, cyclophosphamide, fingolimod, teriflunomide, alemtuzumab, or dimethyl 23 fumarate. 92 Confidential draft: not for distribution, dissemination, or publication 1 2 Statement 7c 3 Patients with serious infections potentially linked to their DMT should switch DMTs (note this 4 does not pertain to management of PML in patients on DMT). 5 Switching Recommendation 8 7 Rationale/Clinical Context: Neutralizing antibodies may be produced against natalizumab and 8 have been associated with allergic reactions (Vennegoor et al. 2013; Krumbholz et al. 2007) 9 (RELA) (Vennegoor 2013; Krumbholz 2007). These antibodies may reduce the efficacy of the drug (PRIN), especially if they are persistent. 11 AF 10 T 6 Statement 8a 13 Clinicians should check for natalizumab antibodies in patients who have infusion reactions prior 14 to subsequent infusions, or in patients who experience breakthrough disease on natalizumab. R 12 D 15 16 Statement 8b 17 Clinicians should switch DMTs in patients who have persistent natalizumab antibodies. 18 19 Switching Recommendation 9 20 Rationale/Clinical Context: Patients with MS taking natalizumab may discontinue natalizumab 21 because of fear of PML or for pregnancy planning (INFER). Discontinuation of natalizumab 22 increases the risk of MS relapse or MRI disease activity (RELA) within 6 months of 23 discontinuation (Prosperini 2015). Data assessing the appropriate choice of an alternate DMT 93 Confidential draft: not for distribution, dissemination, or publication 1 after natalizumab are limited. There is evidence that initiating fingolimod after 8 weeks reduces 2 new MRI lesions compared to initiation after 12 and 16 weeks, whereas initiating fingolimod 3 after 8 to 12 weeks increases the proportion of patients relapse-free compared to initiation after 4 16 weeks (RELA) (Comi 2015, Iaffaldano 2015, Kappos 2015). 5 6 Statement 9a 7 Physicians must counsel patients considering discontinuation of natalizumab that there is an 8 increased risk of MS relapse or MRI disease activity within 6 months of discontinuation. T 9 Statement 9b 11 Physicians and patients choosing to switch from natalizumab to fingolimod should initiate 12 treatment within 8 weeks after discontinuation of natalizumab (for reasons other than pregnancy 13 or pregnancy planning) to diminish the return of disease activity. AF 10 R 14 Recommendation 10 16 Rationale/Clinical Context: Relapse risk is reduced during pregnancy and increases in the post- 17 partum period (RELA) (Finkelsztejn et al. 2011). Pregnancy exposure to DMTs have potential 18 risks to the fetus to varying degrees (RELA). The risks to the fetus vary from severe 19 malformations (category X) to no major increased risk of malformations, but unknown risks of 20 important early life outcomes such as infections, vaccination responses, asthma, and 21 neurocognitive disorders (category B). FDA approved medications are either Category B 22 (glatiramer acetate), Category C (interferons, dimethylfumarate, fingolimod, natalizumab), 23 Category D (mitoxantrone), or Category X (teriflunomide) (RELA). Discussing these potential D 15 94 Confidential draft: not for distribution, dissemination, or publication 1 risks and how best to minimize them is good clinical practice (PRIN). The majority of human 2 safety data for pregnancy exposure to DMTs is derived from accidental exposure early in 3 pregnancy. There is a paucity of safety information with second and third trimester exposure (Lu 4 et al. 2012) (RELA). 5 6 Statement 10a 7 Clinicians should counsel women to stop their DMT prior to conception. 8 Statement 10b T 9 Clinicians should discontinue DMTs during pregnancy if accidental exposure occurs, unless the 11 risk of MS activity during pregnancy outweighs the risk of the specific DMT during pregnancy. 12 AF 10 Statement 10c 14 Clinicians should not initiate DMTs during pregnancy. R 13 D 15 16 Stopping DMT Recommendations 17 Stopping Recommendation 1 18 Rationale/Clinical Context: No randomized controlled trials have directly addressed the question 19 of whether, when or why to discontinue DMTs in an individual with RRMS who has no evidence 20 of relapses, disability progression and stable brain imaging (EVID). The natural history of 21 untreated RRMS is for relapses and disability accumulation to occur. Most individuals with 22 RRMS ultimately transition to SPMS if observed for long enough intervals although disease 23 course is highly variable (Weinshenker et al. 1989) (RELA). Patients who are stable on DMTs 95 Confidential draft: not for distribution, dissemination, or publication 1 may question the continued value of using DMTs (INFER). If patients on DMTs stop these 2 medications, continued monitoring may show subclinical disease activity or relapse activity that 3 would indicate a possible need for resumption of treatment (INFER). Patients who are on DMTs 4 with no evidence of ongoing disease activity may be benefiting from their DMT with disease 5 suppression (INFER). There are presently no biological markers of medication efficacy that can 6 guide decision making in this area (EVID). 7 Statement 1a 9 In patients with relapsing remitting MS who are stable on DMT and wish to discontinue therapy, 10 clinicians should counsel patients regarding the uncertain risks of stopping therapy and the need 11 for ongoing follow-up and periodic re-evaluation of the decision to discontinue DMT. 12 AF T 8 Statement 1b 14 Clinicians should advocate that people with MS who are stable (that is, no relapses, no disability 15 progression, stable imaging) on DMT should be allowed to continue on their current DMT. D R 13 16 17 Stopping Recommendation 2 18 Rationale/Clinical Context: No randomized controlled trials have directly addressed the question 19 of whether or when to discontinue disease-modifying therapies in patients with SPMS (EVID). 20 Clinical trials have not evaluated the benefits of DMT in non-ambulatory individuals with SPMS 21 with respect to other clinically relevant domains including cognition and upper limb function 22 (EVID). Relapses are associated with more rapid disability progression in SPMS but tend to 23 occur in those at younger ages (< 55 years) (Paz Soldan et al. 2015; Tremlett, Zhao, Joseph, 96 Confidential draft: not for distribution, dissemination, or publication 1 Devonshire, & Neurologists 2008) (RELA). In patients with SPMS who are ambulatory with or 2 without assistance, interferon beta reduces the risk of relapse but does not delay disability 3 progression as measured by the EDSS—a measure that emphasizes ambulation (EVID). The 4 benefits of therapy should outweigh the risks (PRIN). The use of ineffective therapy may pose 5 harms to the affected individual, society and the health system (PRIN). 6 Statement 2a 8 Clinicians should advise discontinuation of DMT in patients with SPMS who do not have 9 ongoing relapses (or gadolinium enhancing lesions on MRI), and in addition who have not been ambulatory (EDSS ≥7) for at least 2 years. 11 AF 10 T 7 Statement 2b 13 Clinicians should assess the likelihood of future relapse by assessing patient age and relapse 14 history (e.g., frequency, severity, time since most recent relapse or gadolinium enhancing lesion). R 12 D 15 16 Stopping Recommendation 3 17 Rationale/Clinical Context: No DMTs have demonstrated efficacy in PPMS (EVID). Interferon- 18 beta, glatiramer acetate and rituximab are possibly ineffective (EVID). As such, the issue of 19 discontinuing therapies should not be relevant to this subgroup (INFER), as they should not have 20 been initiated because the use of ineffective therapy may pose harms to the affected individual, 21 society and the health system (PRIN). Ongoing clinical trials may alter these findings. 22 23 Statement 3 97 Confidential draft: not for distribution, dissemination, or publication 1 Clinicians should discuss discontinuation of currently approved DMTs in patients with PPMS 2 with no MRI disease activity currently receiving such treatments. 3 Stopping Recommendation 4 5 Rationale/Clinical Context: Patients with clinically isolated syndromes of demyelination as 6 defined in the international criteria of 2010 (Polman et al. 2011) may not progress to MS 7 (EVID). Patients with CIS who are on DMTs may at some point question the value of continuing 8 DMTs indefinitely (INFER). There remains a gap in knowledge about stopping DMTs in CIS 9 patients (INFER). Discussing the risks of continuing DMT vs the risks of their use being unnecessary as part of ongoing treatment is good clinical practice (PRIN). 11 AF 10 T 4 Statement 4 13 Clinician should review the risk of continuing DMTs vs the risk of stopping DMT in patients 14 with CIS on DMTs. 15 16 RECOMMENDATIONS FOR FUTURE RESEARCH 17 This section presents the identified gaps in the literature. The challenge remains of choosing 18 which MS medication to use in which patient and when due to the limited real world efficacy 19 and safety data as reviewed herein. Below we present key gaps in the literature that if addressed, 20 would greatly improve informed decision-making. D R 12 21 22 1. Comparative effectiveness and safety studies of MS DMTs in clinical practice that 23 examine outcomes important to patients, including long–term physical or cognitive 24 disability, preservation of cognition, MS symptoms, quality of life and efficacy on quality 98 Confidential draft: not for distribution, dissemination, or publication 1 of life measures. It is particularly important to compare affordable to unaffordable 2 treatments, as many patients and insurance carriers in the USA can no longer afford to 3 purchase many of the FDA–approved MS DMTs 4 a. Transparent reporting and comparison of the efficacy of MS DMTs among different 5 MS subpopulations including those who have had continued relapses and/or MRI 6 disease activity on previous MS DMTs, those with highly active disease by these 7 measures, those with clinically isolated syndrome and those in primarily progressive, 8 non-relapsing phase of the disease b. Transparent reporting and comparison of long–term safety of MS DMTs among T 9 patients with comorbidities including diabetes, obesity, cardiovascular disease and 11 physical disability AF 10 2. Research into biomarkers which can predict DMT efficacy in different patient 12 subpopulations. Identification of stronger surrogate clinical or biomarkers for long–term 14 physical or cognitive disability and quality of life R 13 3. More research on matching disease severity to specific DMT types and outcomes of such D 15 methodologies 16 4. More research on pregnancy and DMTs—particularly dissemination of information from 17 pregnancy registries 18 5. No randomized controlled trials have directly addressed the question of whether, when or 19 20 why to discontinue DMTs in an individual with RRMS who has no evidence of relapses, 21 disability progression and stable brain imaging. More studies are needed to inform in 22 which patients discontinuation of DMTs pose little to no harm. 99 Confidential draft: not for distribution, dissemination, or publication 6. 1 In patients with SPMS, no randomized controlled trials have directly addressed the question of whether or when to discontinue DMTs 2 7. 3 Clinical trials have not evaluated the benefits of DMT in non-ambulatory individuals 4 with SPMS with respect to other clinically relevant domains including cognition and 5 upper limb function 8. 6 Comparative efficacy of specific DMTs for treating persons with clinically isolated syndrome is unknown 7 9. Comparative efficacy of highly active DMTs is unknown 9 10. Optimal definition of highly active MS is lacking—uniform definition needed to enhance T 8 comparability between studies and to guide treatment decisions and prognostication AF 10 11 11. What is the optimal definition of response when considering a switch in therapy? 12 12. Need for Class I/II evidence concerning the effect of MS-DMTs on outcomes deemed important by MS experts and patients when selecting MS therapies; namely, efficacy of 14 drugs on quality of life, effect on preservation of cognition and effect on MS symptoms 15 (e.g., fatigue, urinary urgency, pain, etc.) D R 13 13. Good data is lacking for the designation of therapies as “highly active” and the potential 16 to match therapies to individualized disease activity 17 18 14. More information is needed on what to do concerning pregnancy and MS (as our review 19 of the evidence has uncovered): when to stop? Are some agents safer than others? Might 20 some agents be safe enough to continue through conception / pregnancy in patients with 21 active disease? 22 15. Pregnancy and Lactation. Relapsing MS predominantly affects women of childbearing 23 age and unplanned pregnancies while on an MS DMT is common. To minimize risk to 100 Confidential draft: not for distribution, dissemination, or publication 1 the mother and child more studies are needed particularly of newer agents to address the 2 following: a. Determine the safety of in utero or lactation exposure to DMTs, on both major birth 3 outcomes, neurocognitive development, infections and response to vaccines of the infant. 4 b. The risk of return of disease activity during pregnancy or the postpartum period on the 5 6 mother’s long term risk of disability and quality of life with pre-conception/early 7 pregnancy discontinuation of DMT or withholding treatments during lactation 9 Tables/Figures Evidence tables. See appendices 11 Forest Plots. See appendices 12 Survey data figure D R AF 10 T 8 13 14 Figure X: Outcomes of greatest importance to MS experts and patients with MS when selecting a 15 MS disease-modifying therapy. The percentage of physicians (N=18) and patients with MS 16 (N=2156) designating an outcome as first, second or third-highest priority when selecting a MS 17 disease-modifying therapy is depicted. Outcomes measures were similarly ranked by physicians 101 Confidential draft: not for distribution, dissemination, or publication 1 and patients. Within category differences were considered, with significant differences indicated 2 (*=p<0.05). MRI = brain magnetic resonance imaging; AE = adverse events; QoL = quality of 3 life 4 5 Side effect/major risk tables 6 Table 1: Considerations/populations which would suggest caution with specific FDA approved 7 MS therapies 8 Special Considerations/Populations T Medication Interferons AF Leukopenia (Betaseron®/Betaferon®/Extavia®, Liver Disease Avonex®, Rebif®) Depression Significant Spasticity Cosmetic concerns (Skin induration) Natalizumab (Tysabri®) Liver Disease D R Glatiramer acetate (Copaxone®,Glatopa®) History of Immunosuppression JCV Seropositivity Prolonged use >2 years Fingolimod (Gilenya®) Leukopenia Liver Disease Uveitis, Macular Edema Diabetes VZV Seronegativity 102 Confidential draft: not for distribution, dissemination, or publication Atrioventricular Conduction Block Medications that May Affect Cardiac Conduction Basal Cell Carcinoma Teriflunomide (Aubagio®) Liver Disease Leukopenia Hypertension Short-term Plans for Pregnancy Leukopenia AF Dimethyl Fumarate (Tecfidera®) T History of Tuberculosis Alemtuzumab (Lemtrada®) Thyroid disease Daclizumab (Zinbryta®) Hepatic disease Autoimmune disease Cardiac disease Neutropenia <1,500 cells/mm3 D R Mitoxantrone (Novantrone®) VZV = varicella zoster virus. 1 2 3 4 103 Confidential draft: not for distribution, dissemination, or publication 1 Table 2: Risks of MS therapies and potential mitigation strategies. Drug Risks Mitigation Strategies Glatiramer Skin reactions Education, proper injection technique Acetate Immediate post-injection (Copaxone) systemic reaction Education to minimize unnecessary testing/ED visits Interferon beta Flu-like symptoms (Rebif, Betaseron, Elevated liver enzymes Avonex) Leukopenia Baseline and monitoring CBC Headache Baseline headache screen Depression Screen and monitor for depression T NSAIDs, hydration Baseline and monitoring LFT AF R Natalizumab PML D 2 (Tysabri) Neutralizing antibodies as needed JCV testing q6 months; For JCV+ pts, MRI q6m, screen for PML symptoms, limit to 24 infusions, avoid if prior immunosuppression Allergic reaction Increased risk with natalizumab Ab+ or hx of anaphylaxis Elevated liver enzymes Baseline and monitoring LFT Leukopenia Baseline and monitoring CBC 104 Confidential draft: not for distribution, dissemination, or publication Fingolimod (Gilenya) Cardiac events on 1st dose Baseline EKG, first dose Bradycardia, AV block observation for 6 or 24 hours, Cardiac arrest, arrhythmias contraindicated for new or existing cardiac disease or on antiarrhythmic; cardiology consult if abnormal EKG Herpes infection VZV serology screen; vaccination if negative Baseline and 3 month CBC PML None currently available R Lymphopenia Contraindicated in pregnancy, Liver dysfunction Baseline and q6 month LFT; Teratogenesis D (Aubagio) Baseline and 3 month LFT AF Liver enzyme derangements Teriflunomide Baseline and 3 month OCT T Macular edema ensure contraception contraindicated for existing liver disease Reactivation of latent TB Screen for TB (e.g. PPD) Alopecia Dimethyl Lymphopenia CBC at baseline, 3 mo, 6 mo fumarate Flushing Education, non-enteric ASA (Tecfidera) GI symptoms 105 Confidential draft: not for distribution, dissemination, or publication Take with food; symptomatic PML treatment; increased risk for pts with GERD/gastritis Lymphocyte monitoring every six months 1 2 3 Alemtuzumab Autoimmune events: 5 (Lemtrada) Hyperthyroidism/Grave’s Monthly CBC and CMP T 4 AF TSH every 3 months. Immune thrombocytopenia Platelets monthly 7 Immune glomerulonephritis Renal function check 8 Skin cancer, lymphoproliferative Skin check annually 9 disorders R 6 11 D 10 ______________________________________________________________________________ 12 13 Daclizumab Hypersensitivity reactions Prophylaxis prior to infusion 14 (Zinbryta) Severe liver injury Hepatic monitoring Immune mediated disorders Monitoring Cardiomyopathy Baseline EKG, history exam 15 16 17 18 Mitoxantone 106 Confidential draft: not for distribution, dissemination, or publication 1 (Novantrone) LVEF assessment baseline and prior 2 to each dose, max dose lifetime 3 140mg/m2 Treatment related Leukemia 4 Monitoring 5 ______________________________________________________________________________ 7 AV: Atrioventricular 8 CBC: Complete blood count 9 CMP: Complete metabolic profile T 6 ED:Emergency Department 11 EKG: Electrocardiogram 12 GERD: Gastroesophageal reflux 13 JCV: John Cunningham Virus 14 LFT: Liver function tests 15 LVEF: left ventricular ejection fraction 16 NSAIDS: non-steroidal anti-inflammatory drugs 17 OCT: Ocular Coherence Tomography 18 PML: Progressive multifocal leukoencephalopathy 19 PPD: Purified protein derivative (PPD) skin test 20 TB: Tuberculosis 21 TSH: Thyroid stimulating hormone 22 VZV: Varicella zoster virus 23 Adapted from Onteneda et al, MSARD 2014 D R AF 10 107 Confidential draft: not for distribution, dissemination, or publication DISCLAIMER 2 Practice guidelines, practice advisories, comprehensive systematic reviews, focused systematic 3 reviews and other guidance published by the American Academy of Neurology and its affiliates 4 are assessments of current scientific and clinical information provided as an educational service. 5 The information: 1) should not be considered inclusive of all proper treatments, methods of care, 6 or as a statement of the standard of care; 2) is not continually updated and may not reflect the 7 most recent evidence (new evidence may emerge between the time information is developed and 8 when it is published or read); 3) addresses only the question(s) specifically identified; 4) does not 9 mandate any particular course of medical care; and 5) is not intended to substitute for the T 1 independent professional judgment of the treating provider, as the information does not account 11 for individual variation among patients. In all cases, the selected course of action should be 12 considered by the treating provider in the context of treating the individual patient. Use of the 13 information is voluntary. AAN provides this information on an “as is” basis, and makes no 14 warranty, expressed or implied, regarding the information. AAN specifically disclaims any 15 warranties of merchantability or fitness for a particular use or purpose. AAN assumes no 16 responsibility for any injury or damage to persons or property arising out of or related to any use 17 of this information or for any errors or omissions. D R AF 10 18 19 CONFLICT OF INTEREST STATEMENT 20 The American Academy of Neurology is committed to producing independent, critical, and 21 truthful clinical practice guidelines (CPGs). Significant efforts are made to minimize the 22 potential for conflicts of interest to influence the recommendations of this CPG. To the extent 23 possible, the AAN keeps separate those who have a financial stake in the success or failure of the 108 Confidential draft: not for distribution, dissemination, or publication products appraised in the CPGs and the developers of the guidelines. Conflict of interest forms 2 were obtained from all authors and reviewed by an oversight committee prior to project 3 initiation. AAN limits the participation of authors with substantial conflicts of interest. The AAN 4 forbids commercial participation in, or funding of, guideline projects. Drafts of the guideline 5 have been reviewed by at least three AAN committees, a network of neurologists, Neurology 6 peer reviewers, and representatives from related fields. The AAN Guideline Author Conflict of 7 Interest Policy can be viewed at www.aan.com. For complete information on this process, access 8 the 2011 AAN process manual, as amended 9 (https://www.aan.com/Guidelines/Home/Development). Acknowledgments 11 AF 10 T 1 APPENDICES 13 Appendices will include GDDI member roster and mission statement, the complete search 14 strategy employed, the schemes for classification of evidence, rules for determining confidence 15 in evidence, the steps and rules for formulating recommendations, and the clinical contextual 16 profiles. This section can be populated by AAN staff and the methodologist assigned to the 17 project. All appendices are included in the complete manuscript, available as an online data 18 supplement to the main article. D R 12 19 20 Starting DMT Recommendations 21 Statement 1 22 23 Clinicians should counsel patients just diagnosed with MS about specific treatment options with DMT at a dedicated treatment visit. 109 Confidential draft: not for distribution, dissemination, or publication Domain Rating Rationale is logical 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 0 Mildly 10 3 Moderate 3 Yes 4 Yes Always 10 4 Yes Moderate 1 Small 9 B 4 Yes T C 10 Yes Minimal 10 1 1 10 Yes 7 Yes Usually Large Yes Critically Important 9 1 Occasionally 10 Yes Benefit >>> Harm Modest 0 10 Yes 7 Very 0 R/U 10 Benefit >> Harm 1 1 Strength of recommendation 1 Consensus < 50% A Statement 2a 3 4 5 Clinicians must ascertain and incorporate/review preferences in terms of safety, route of administration, lifestyle, cost, efficacy, common side effects and tolerability in the choice of DMT in patients with MS being considered for DMT. Domain AF 2 Rating < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Rationale is logical D Axioms true R Evidence statements accurate Strength of recommendation 6 Consensus Benefit > Harm 0 1 Mildly 0 1 Moderate 0 6 Modest 1 Occasionally 0 3 Usually 1 Large 0 R/U 4 Very 4 Moderate 1 C 4 B 10 Yes 10 Yes 10 Yes 10 Yes 10 Yes 10 10 Critically Important 8 Minimal 11 Always 10 Small 10 Yes Yes Yes Yes Yes A 7 Statement 2b 8 9 Clinicians must engage in an ongoing dialogue regarding treatment decisions throughout the course of the disease with patients with MS. 110 Confidential draft: not for distribution, dissemination, or publication Domain Rating Rationale is logical 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 0 0 Mildly 1 Very 0 0 Moderate 5 Modest 1 0 Occasionally 0 3 Usually 0 Large 4 Moderate 0 0 R/U C 3 B 10 Yes 10 Yes 10 Yes 10 Yes 10 Yes 10 14 Critically Important 10 Minimal 11 Always 11 Small 12 T Strength of recommendation 1 Consensus < 50% Yes Yes Yes Yes Yes A Statement 3 3 4 Clinicians should counsel that DMTs are prescribed to reduce relapses and new MRI lesion activity. DMTs are not prescribed for symptom improvement in patients with MS. Domain AF 2 Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% 100% R Evidence statements accurate 10 100% < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large D Axioms true Strength of recommendation 5 Consensus Benefit > Harm 0 0 Mildly 0 0 0 0 3 Usually 0 Large 0 R/U 7 Modest Occasionally 3 Moderate 0 C 2 B 10 Yes 10 Yes 10 Yes 10 Yes Benefit >>> Harm 0 Very Moderate 1 10 Benefit >> Harm Yes 15 Yes Critically Important 8 Yes Minimal 11 Yes Always 12 Yes Small 13 Yes A 6 7 Statement 4 8 9 Clinicians should evaluate readiness or reluctance to initiate DMT and counsel on its importance in patients with MS who are candidates to initiate DMT. 111 Confidential draft: not for distribution, dissemination, or publication Domain Rating Rationale is logical 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 1 Mildly 10 0 Moderate 6 Modest 1 1 Occasionally 4 Usually 0 Large 3 Moderate 0 0 Yes 10 Yes 10 Yes 10 Yes 3 B 14 Critically Important 9 Minimal 9 Always 12 Small 12 T C 10 Yes Benefit >>> Harm 0 Very 0 R/U 10 Benefit >> Harm 0 0 Strength of recommendation 1 Consensus < 50% Yes Yes Yes Yes Yes A 3 Statement 5 4 5 Clinicians should counsel about comorbid disease and adverse health behaviors and potential interactions of the DMT with concomitant medications when patients with MS initiate DMTs. AF 2 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Domain Rating < 50% 50% to < 80% D R Rationale is logical Strength of recommendation 6 Benefit > Harm 0 Mildly 7 Modest 0 Occasionally 6 Usually 1 Large 0 4 Moderate 0 C 10 Yes 10 Yes 10 Yes 10 Yes Yes High Benefit >>> Harm 2 0 0 10 Very Moderate 0 10 Benefit >> Harm 0 0 R/U 80% to < 100% Consensus 4 B 7 Statement 6a 8 Clinicians should evaluate barriers to adherence to DMT in patients with MS. 13 Critically Important 8 Minimal 9 Always 10 Small 11 A Yes Yes Yes Yes Yes 112 Confidential draft: not for distribution, dissemination, or publication Domain Rating Rationale is logical 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 0 0 Mildly 1 Very 0 0 Moderate 10 Modest 0 0 Occasionally 0 3 Usually 0 Large 5 Moderate 0 0 R/U C 2 B 10 Yes 10 Yes 10 Yes 10 Yes 10 Yes 10 14 Critically Important 5 Minimal 12 Always 10 Small 13 T Strength of recommendation 1 Consensus < 50% Yes Yes Yes Yes Yes A Statement 6b 3 Clinicians should counsel on the importance of adherence to DMT when patients with MS initiate DMTs. AF 2 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Domain Rating Rationale is logical Evidence statements accurate < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% D R Axioms true Strength of recommendation 4 Consensus Benefit > Harm 0 0 Mildly 0 0 Moderate 0 10 Modest 0 Occasionally 0 3 Usually 0 Large 0 R/U 1 Very 5 Moderate 0 C 2 B 10 Yes 10 Yes 10 Yes 10 Yes 10 Yes 10 14 Critically Important 5 Minimal 12 Always 10 Small 13 Yes Yes Yes Yes Yes A 5 6 Statement 7 7 8 Clinicians may offer disease modifying therapy for patients with a single clinical demyelinating event with two or more brain or spinal cord lesions that have imaging characteristics consistent with MS. 113 Confidential draft: not for distribution, dissemination, or publication Domain Rating Rationale is logical 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 1 Mildly 2 Yes 2 Yes Minimal 1 Yes Always 13 1 Yes Moderate 9 Small 5 B 0 Yes T C 10 Yes 3 1 1 10 Yes Critically Important Usually Large 10 Yes 12 10 Occasionally 10 Yes Benefit >>> Harm Modest 1 Yes 12 1 Moderate R/U 10 Very 0 100% 10 Benefit >> Harm 0 0 Strength of recommendation 1 Consensus < 50% A 3 4 5 Statement 8 6 7 8 Clinicians may recommend serial imaging and close follow‐up rather than initiating DMT in patients with CIS or relapsing forms of MS not on DMT who have not had relapses in the past 2 years and who do not have active new MRI lesion activity on a recent MRI. R Domain AF 2 Consensus < 50% 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% N/A Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% N/A < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large D Rationale is logical Strength of recommendation 9 Rating Benefit > Harm 1 2 Mildly 0 0 1 R/U 5 0 0 1 Yes Always 7 Moderate 2 C 2 Yes Minimal Usually Large 10 2 No 12 7 1 10 Yes Critically Important Modest Occasionally 10 Yes 9 Very Moderate 10 Yes 6 Yes Small 9 B 3 Yes A 114 Confidential draft: not for distribution, dissemination, or publication 1 2 Statement 9 3 4 Clinicians should offer DMTs to patients with relapsing forms of MS with recent clinical relapses or MRI activity. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 1 0 0 4 Very 0 Moderate 0 6 Modest 2 10 Usually AF Occasionally 0 Large 1 R/U 10 Yes 10 Yes 10 Yes 10 Yes 10 Yes 10 11 Yes Critically Important 10 Yes Minimal 4 Yes Always 2 Yes Small 1 Yes T Mildly 0 Strength of recommendation 5 Consensus 6 C 14 Moderate B 8 A 7 Statement 10a 8 9 Clinicians should monitor for medication adherence, side effects, tolerability, safety and effectiveness of the therapy in patients with MS on DMTs. D R 6 115 Confidential draft: not for distribution, dissemination, or publication Domain Rating Rationale is logical 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 0 Mildly 8 Modest 0 1 Occasionally 3 Usually 0 Large 4 Moderate 0 0 10 Yes 10 Yes 10 Yes 10 Yes 5 B 13 Critically Important 7 Minimal 11 Always 11 Small 10 T C Yes Benefit >>> Harm 2 0 Moderate R/U 10 Very 0 100% 10 Benefit >> Harm 0 0 Strength of recommendation 1 Consensus < 50% Yes Yes Yes Yes Yes A Statement 10b 3 4 Clinicians should follow up either annually or according to medication‐specific risk evaluation and mitigation strategies (REMS) in patients with MS on DMTs. Domain AF 2 Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% R Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large D Axioms true Strength of recommendation 5 Consensus Benefit > Harm 0 Mildly 0 2 0 7 Usually 0 Large 0 R/U 8 Modest Occasionally 7 Moderate 0 C 8 B 10 Yes 10 Yes Yes 10 Yes Benefit >>> Harm 4 Very Moderate 0 10 Benefit >> Harm 0 0 10 10 Yes 11 Yes Critically Important 7 Yes Minimal 6 Yes Always 8 Yes Small 7 Yes A 6 Statement 11 7 8 Clinicians should monitor patient’s reproductive plans and counsel on reproductive risks and on use of birth control while on a DMT in women of childbearing years with MS. 116 Confidential draft: not for distribution, dissemination, or publication Domain Rating Rationale is logical 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 0 0 9 Moderate Modest 0 3 7 Occasionally Usually 0 6 Large Moderate 0 0 3 10 Yes 10 Yes 10 Yes 10 Yes B 10 Critically Important 6 Minimal 5 Always 9 Small 12 T C Yes Benefit >>> Harm Very 0 R/U 10 5 Mildly 100% 10 Benefit >> Harm 0 0 Strength of recommendation 1 Consensus < 50% Yes Yes Yes Yes Yes A 3 Statement 12 4 5 Clinicians must counsel men with MS on their reproductive plans regarding treatment implications before initiating teriflunomide or cyclophosphamide. AF 2 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Domain Rating < 50% 50% to < 80% D R Rationale is logical Strength of recommendation 6 Benefit > Harm 0 1 Mildly 0 2 Very 1 Moderate 0 5 Modest 2 Occasionally 0 2 Usually 0 Large 0 R/U 80% to < 100% Consensus 3 Moderate 0 C 7 Statement 13 8 Clinicians should not prescribe mitoxantrone to patients for MS. 4 B 10 Yes 10 Yes 10 Yes 10 Yes 10 Yes 10 12 Critically Important 9 Minimal 11 Always 12 Small 11 A Yes Yes Yes Yes Yes 117 Confidential draft: not for distribution, dissemination, or publication Domain Rating Rationale is logical 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 1 Mildly 4 Modest 0 0 Occasionally 3 Usually 1 Large 0 Moderate 0 0 10 Yes 10 Yes 10 Yes 10 Yes 3 B 13 Critically Important 11 Minimal 12 Always 14 Small 12 T C Yes Benefit >>> Harm 1 0 Moderate R/U 10 Very 0 100% 10 Benefit >> Harm 0 0 Strength of recommendation 1 Consensus < 50% Yes Yes Yes Yes Yes A 3 Statement 14 4 Clinicians should prescribe alemtuzumab, fingolimod or natalizumab for patients with very active MS. Domain AF 2 Rating < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Rationale is logical D Axioms true R Evidence statements accurate Strength of recommendation 5 Consensus Benefit > Harm 0 Mildly 0 R/U 10 Yes 6 Yes 0 Yes Always 12 Moderate 5 C 10 Yes 4 Yes 10 1 1 10 Yes Minimal Usually Large 10 Yes Critically Important 9 4 0 Yes Benefit >>> Harm Modest Occasionally 100% 11 Very Moderate 1 10 Benefit >> Harm 0 0 10 2 Yes Small 8 B 1 Yes A 6 7 8 Statement 15a 9 Clinicians may direct patients with MS who are candidates for DMTs to support programs. 118 Confidential draft: not for distribution, dissemination, or publication Domain Rating Rationale is logical 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 0 2 Mildly 0 0 Moderate 2 Yes 7 Yes Always 7 0 Yes Small 6 B 8 Yes T C 10 Minimal Moderate 1 R/U 10 Yes 5 8 0 10 Yes 9 Yes Usually Large 10 Yes 13 3 Occasionally 10 Yes Critically Important Modest 0 10 Yes 4 Very 0 Strength of recommendation 1 Consensus < 50% A Statement 15b 3 4 Clinicians should recommend one of the following DMTs (azathioprine, cladribine, minocycline, leflunomide) for patients with relapsing forms of MS who do not have access to approved DMTs Domain AF 2 Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% 100% R Evidence statements accurate 10 100% < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large D Axioms true Strength of recommendation 5 Consensus 6 7 8 Statement 16 Benefit > Harm 0 2 Mildly 1 2 R/U 1 Yes 0 Yes Always 11 Moderate 3 C 2 Yes 9 3 0 10 Yes Minimal Usually Large 10 Yes Critically Important 12 5 0 10 Yes Benefit >>> Harm Modest Occasionally 10 Yes 11 Very Moderate 1 10 Benefit >> Harm Yes 1 Yes Small 9 B 3 Yes A 119 Confidential draft: not for distribution, dissemination, or publication 1 2 Clinicians should not initiate natalizumab treatment in patients with positive JC virus antibodies in serum unless there are overwhelming reasons to do so. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% N/A Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% N/A < 50% 50% to < 80% 80% to < 100% 100% N/A Confidence in Inference Very low Low Moderate Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 1 Mildly Benefit >>> Harm 6 2 Occasionally 10 1 Yes Usually 8 T 2 Always Large 1 Moderate 1 C B 4 Yes A 5 Statement 17 6 For patients with PPMS clinicians should not initiate currently approved DMTs. R Rating Consensus < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Rationale is logical AF R/U 4 Yes Small 8 D Evidence statements accurate Strength of recommendation 7 7 Yes Minimal 4 Domain 7 Yes Critically Important Modest 1 10 Yes 5 1 Moderate Yes High 10 Very 0 100% 10 Benefit >> Harm 1 0 Strength of recommendation 3 Consensus Benefit > Harm 3 0 Mildly 0 0 Moderate 0 14 Modest 5 Occasionally 0 10 Usually 2 Large 2 R/U 4 Very 3 Moderate 1 C 4 B 8 9 Switching DMT Recommendations (21 statements) 10 Yes 10 Yes 10 Yes N/A 10 Yes 10 11 Critically Important 4 Minimal 3 Always 13 Small 11 A Yes Yes Yes Yes Yes 120 Confidential draft: not for distribution, dissemination, or publication 1 Statement 1a 2 3 4 Clinicians should periodically monitor MRI disease activity during the first 5 years from the clinical onset of disease to detect the accumulation of new lesions in order to inform treatment decisions in patients with MS on DMTs. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 1 Mildly 10 10 8 Yes 5 Yes AF C 1 Yes Always 12 Moderate 1 R/U Minimal 12 Usually 2 1 10 Yes 9 1 Large Yes Critically Important Modest Occasionally 10 Yes Benefit >>> Harm T 1 Yes 6 1 Moderate Yes 100% 10 Very 0 100% 10 Benefit >> Harm 0 0 Strength of recommendation 5 Consensus 1 Yes Small 13 B 0 Yes A Statement 1b 7 8 Clinicians should recognize that relapses or new MRI lesions may develop after initiation of a DMT and before the treatment becomes effective in patients with MS on DMTs. Domain Rationale is logical R 6 < 50% 50% to < 80% 80% to < 100% 10 100% < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large D 10 Consensus Evidence statements accurate Strength of recommendation 9 Rating Benefit > Harm 0 Mildly 0 R/U 5 Yes 5 Yes Always 8 Moderate 1 C 9 Yes 9 0 0 10 Yes Minimal Usually Large Yes Critically Important 10 1 1 10 Yes Benefit >>> Harm Modest Occasionally 10 Yes 6 Very Moderate 0 10 Benefit >> Harm 0 0 10 Yes 7 Yes Small 7 B 7 Yes A Statement 1c 121 Confidential draft: not for distribution, dissemination, or publication 1 2 3 4 Clinicians should discuss switching from one DMT to another in patients who have been on a DMT long enough to take full effect and are adherent to their therapy when a patient has experienced one or more relapses, two or more unequivocally new MRI lesions, or increased disability on examination, over a 1‐year period on a DMT. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 1 Mildly 10 Yes 3 Yes Minimal 0 Yes Always 12 AF Moderate 3 C Yes 4 Yes 10 Usually 2 0 10 Yes Critically Important 11 5 Large Yes Benefit >>> Harm Modest Occasionally 10 Yes 10 1 R/U 10 Very Moderate 0 10 T 0 10 Benefit >> Harm 0 0 Strength of recommendation 5 Consensus B 1 Yes Small 12 0 Yes A 7 Statement 2 8 9 Clinicians may evaluate the amount of disease activity, adherence, adverse event profiles, and mechanism of action of DMTs when switching DMTs in patients with breakthrough disease on DMTs. D 10 R 6 122 Confidential draft: not for distribution, dissemination, or publication Domain Rating Rationale is logical < 50% 50% to < 80% Evidence statements accurate < 50% Axioms true 100% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large 10 10 Benefit > Harm 0 Mildly 0 Moderate 8 Modest 1 3 Occasionally 9 Usually 1 Large 11 Moderate 0 2 10 Yes Yes 10 Yes 10 Yes 6 B 10 Critically Important 7 Minimal 2 Always 3 Small 7 T C No Benefit >>> Harm 5 Very 0 R/U 10 Benefit >> Harm 0 0 Strength of recommendation 1 Consensus 80% to < 100% Yes Yes Yes Yes Yes A Statement 3 3 4 Clinicians should discuss a change to a non‐injectable or less frequently injectable DMT in patients who report pain with the injections or in those who report “injection fatigue” on injectable DMTs. Domain AF 2 Rating Rationale is logical 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% R Evidence statements accurate < 50% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large 7 8 9 Benefit > Harm 0 0 Mildly 0 1 R/U 3 Yes 2 Yes Always 16 Moderate 3 C 7 Yes 13 0 0 10 Yes Minimal Usually Large 10 Yes Critically Important 14 2 0 N/A Benefit >>> Harm Modest Occasionally Yes 11 Very Moderate 1 10 Benefit >> Harm Yes 100% 80% to < 100% D 10 100% 50% to < 80% Strength of recommendation 6 10 < 50% Axioms true 5 Consensus 2 Yes Small 12 B 3 Yes A 123 Confidential draft: not for distribution, dissemination, or publication 1 Statement 4a 2 3 Clinicians should inquire about medication adverse effects with patients with MS who are taking a DMT and attempt to manage these adverse effects, as appropriate. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 0 0 Mildly 2 Very 0 0 11 Modest 10 Yes 10 Yes 10 Yes 10 Yes 10 Yes 10 13 Critically Important 4 Minimal 10 Always 9 Small 9 T Moderate 0 1 Occasionally 0 4 Usually 1 Large 0 5 Moderate 1 AF Strength of recommendation 4 Consensus R/U C 5 B Yes Yes Yes Yes Yes A Statement 4b 6 7 Clinicians should discuss a medication switch with patients for whom these adverse effects negatively influence adherence. Domain R 5 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large D 9 Consensus < 50% Rationale is logical Strength of recommendation 8 Rating Benefit > Harm 0 0 Mildly 0 0 1 R/U 6 Yes Always 6 Moderate 2 C 10 4 Yes 7 2 0 10 Yes Minimal Usually Large 10 Yes 9 Yes 10 0 0 10 Yes Critically Important Modest Occasionally 10 Yes 5 Very Moderate 10 Yes 6 Yes Small 8 B 4 Yes A 124 Confidential draft: not for distribution, dissemination, or publication 1 Statement 5a 2 3 Clinicians should monitor laboratory abnormalities found on requisite laboratory surveillance (as outlined in the medication's package insert) in patients with MS who are on a DMT. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 0 Mildly 10 Yes 10 Yes 10 Yes 10 Yes Benefit >>> Harm 2 0 9 Modest 13 Critically Important 6 Minimal 7 Always 10 Small 6 T Moderate Yes 100% 10 Very 0 100% 10 Benefit >> Harm 0 0 0 Occasionally 0 8 Usually 0 Large 0 5 Moderate 0 AF Strength of recommendation 4 Consensus R/U C 9 B Yes Yes Yes Yes Yes A Statement 5b 6 7 8 Clinicians should discuss switching DMT or reducing dose or frequency (where there is data on different doses [e.g., interferons, teriflunomide, azathioprine) when there are persistent laboratory abnormalities. Domain < 50% D Rationale is logical R 5 10 Consensus 50% to < 80% 80% to < 100% 10 100% Yes Evidence statements accurate < 50% 50% to < 80% 80% to < 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% N/A Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% N/A < 50% 50% to < 80% 80% to < 100% 100% N/A Confidence in Inference Very low Low Moderate Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Strength of recommendation 9 Rating Benefit > Harm 1 Mildly 2 R/U 8 0 2 3 No Always 10 Moderate 0 C 5 Yes Minimal Usually Large 9 Yes Critically Important 7 2 0 Benefit >>> Harm Modest Occasionally 4 Yes Small 9 B Yes High 3 Very Moderate 1 10 Benefit >> Harm 1 0 10 100% 3 Yes A 125 Confidential draft: not for distribution, dissemination, or publication 1 Statement 6a 2 3 Clinicians should counsel patients considering natalizumab, fingolimod, and dimethyl fumarate about the risk of PML associated with these agents. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 2 Mildly 10 0 10 Yes 10 Yes 10 Yes 10 Yes 4 Modest 10 Critically Important 11 Minimal 11 Always 13 Small 13 T Moderate Yes Benefit >>> Harm 3 Very 0 100% 10 Benefit >> Harm 0 1 1 Occasionally 0 2 Usually 1 Large 1 1 Moderate 0 AF Strength of recommendation 4 Consensus R/U C 1 B Yes Yes Yes Yes Yes A Statement 6b 6 7 Clinicians should discuss switching to a DMT with a lower risk of PML in patients taking natalizumab who are JC virus antibody positive or become JC virus antibody positive while on therapy. Domain R 5 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large D 9 Consensus < 50% Rationale is logical Strength of recommendation 8 Rating Benefit > Harm 0 0 Mildly 0 0 Moderate 0 6 Modest 1 Occasionally 0 9 Usually 1 Large 0 R/U 5 Very 6 Moderate 0 C 9 B 10 Yes 10 Yes 10 Yes 10 Yes 10 Yes 10 10 Critically Important 9 Minimal 5 Always 8 Small 6 A Yes Yes Yes Yes Yes 126 Confidential draft: not for distribution, dissemination, or publication 1 Statement 7a 2 3 Clinicians should counsel that new DMTs without long‐term safety data have an undefined risk of malignancy and infection in patients starting or using new DMTs. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 0 10 Mildly 10 1 Yes Yes 10 Yes 8 Modest 13 Critically Important 6 Minimal 9 Always 12 Small 12 T Moderate 10 Yes Benefit >>> Harm 0 Very 1 10 Yes 100% 10 Benefit >> Harm 2 0 1 Occasionally 0 5 Usually 1 Large 0 2 Moderate 0 AF Strength of recommendation 4 Consensus R/U C 3 B Yes Yes Yes Yes Yes A Statement 7b 6 7 8 If a patient with MS develops a malignancy while on a DMT, clinicians should promptly discuss switching to an alternate DMT, especially for patients on azathioprine, methotrexate, mycophenolate, cyclophosphamide, fingolimod, teriflunomide, alemtuzumab, or dimethyl fumarate. Domain < 50% D Rationale is logical R 5 10 Consensus 50% to < 80% 80% to < 100% 10 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% Axioms true < 50% 50% to < 80% 80% to < 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Strength of recommendation 9 Rating Benefit > Harm 1 Mildly R/U 10 Yes 9 Yes 8 Yes Minimal 6 Usually 1 6 Yes Always 6 Moderate 1 C Yes Critically Important 6 1 Large Yes 100% Benefit >>> Harm Modest Occasionally 1 10 Yes 100% 5 0 0 10 Very Moderate 2 10 Benefit >> Harm 0 1 10 Yes 100% 8 Yes Small 6 B 7 Yes A Statement 7c 127 Confidential draft: not for distribution, dissemination, or publication 1 2 Patients with serious infections potentially linked to their DMT should switch DMTs (note this does not pertain to management of PML in patients on DMT). Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% Axioms true < 50% 50% to < 80% 80% to < 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 0 Mildly 10 5 Usually 0 5 Moderate 1 C 7 B Yes Yes Yes Yes Yes A AF R/U 10 Yes 10 Critically Important 7 Minimal 7 Always 9 Small 6 T 0 Large Yes Benefit >>> Harm 7 1 Occasionally Yes 100% 10 Modest 1 Yes 100% 10 3 0 Moderate Yes 100% 10 Very 0 100% 10 Benefit >> Harm 1 0 Strength of recommendation 3 Consensus 5 Statement 8a 6 7 Clinicians should check for natalizumab antibodies in patients who have infusion reactions prior to subsequent infusions, or in patients who experience breakthrough disease on natalizumab. Domain R 4 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large D 9 Consensus < 50% Rationale is logical Strength of recommendation 8 Rating Benefit > Harm 0 Mildly 0 8 Usually 2 Large 0 R/U 9 Modest Occasionally 8 Moderate 1 C 10 B 10 Yes Yes 100% Yes 10 Yes Benefit >>> Harm 4 0 0 10 Very Moderate 0 10 Benefit >> Harm 0 0 10 10 Yes 11 Critically Important 6 Minimal 7 Always 5 Small 4 A Yes Yes Yes Yes Yes 128 Confidential draft: not for distribution, dissemination, or publication 1 Statement 8b 2 Clinicians should switch DMTs in patients who have persistent natalizumab antibodies. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 1 Mildly 10 10 4 Yes 2 Yes 8 4 Yes T 1 0 5 Yes Small 9 B AF C Always 8 Moderate 3 R/U 10 Yes Minimal Usually Large Yes Critically Important 12 1 Occasionally Yes Benefit >>> Harm Modest 1 10 Yes 9 0 Moderate Yes 100% 10 Very 1 100% 10 Benefit >> Harm 0 0 Strength of recommendation 3 Consensus 2 Yes A 4 Statement 9a 5 6 Physicians must counsel patients considering discontinuation of natalizumab that there is an increased risk of MS relapse or MRI disease activity within 6 months of discontinuation. 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Benefit >> Harm Benefit >>> Harm Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large D Evidence statements accurate 8 Consensus < 50% Rationale is logical Strength of recommendation 7 Rating R Domain Benefit > Harm 0 0 Mildly 0 0 Moderate 0 5 Modest 1 Occasionally 0 2 Usually 0 Large 1 R/U 6 Very 2 Moderate 0 C 2 B 10 Yes 10 Yes 10 Yes 10 Yes 10 Yes 10 9 Critically Important 10 Minimal 12 Always 13 Small 12 A Yes Yes Yes Yes Yes Statement 9b 129 Confidential draft: not for distribution, dissemination, or publication 1 2 3 Physicians and patients choosing to switch from natalizumab to fingolimod should initiate treatment within 8 weeks after discontinuation of natalizumab (for reasons other than pregnancy or pregnancy planning) to diminish the return of disease activity. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 0 Mildly T AF B 6 Clinicians should counsel women to stop their DMT prior to conception. 7 R Statement 10a 3 Yes A Consensus 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% Confidence in Inference Very low Low Moderate Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Strength of recommendation Benefit > Harm 1 Mildly 9 Usually 0 Large 7 Moderate 1 C 9 B Yes N/A 10 Yes Yes N/A High Benefit >>> Harm 10 1 1 10 Modest Occasionally 100% 100% 6 0 0 10 Very Moderate 2 10 Benefit >> Harm 1 0 R/U Rating 50% to < 80% D 9 1 Yes Small < 50% Rationale is logical 8 4 Yes Always 9 5 Domain 6 Yes 14 Moderate C 10 Yes 7 Yes 8 3 R/U 10 Yes Minimal Usually 0 0 10 Yes Critically Important 9 2 Large 10 Yes Benefit >>> Harm Modest Occasionally Yes 8 0 Moderate 1 10 Very 0 100% 10 Benefit >> Harm 0 0 Strength of recommendation 4 Consensus 10 Critically Important 8 Minimal 6 Always 11 Small 7 A Yes Yes Yes Yes Yes 130 Confidential draft: not for distribution, dissemination, or publication 1 Statement 10b 2 3 Clinicians should discontinue DMTs during pregnancy if accidental exposure occurs, unless the risk of MS activity during pregnancy outweighs the risk of the specific DMT during pregnancy. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 1 Mildly AF C 6 Clinicians should not initiate DMTs during pregnancy. 9 Yes Always 12 Yes Small 2 B 10 Yes A R Rating Consensus < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Rationale is logical D Evidence statements accurate 8 9 Yes Minimal 0 Moderate 0 R/U Yes 6 Yes 2 1 1 10 Yes Critically Important Usually Large 10 Yes Benefit >>> Harm 4 1 Statement 10c Strength of recommendation 7 10 Modest Occasionally 10 Yes High T 1 Yes 6 0 Moderate 5 Domain 10 Very 0 100% 10 Benefit >> Harm 0 0 Strength of recommendation 4 Consensus Benefit > Harm 1 Mildly 0 1 0 7 Usually 0 Large 1 R/U 7 Modest Occasionally 6 Moderate 0 C 7 B Yes N/A 10 Yes 10 Yes N/A High Benefit >>> Harm 6 Very Moderate 0 10 Benefit >> Harm 0 0 10 100% 11 Critically Important 11 Minimal 10 Always 12 Small 10 A Yes Yes Yes Yes Yes 9 131 Confidential draft: not for distribution, dissemination, or publication 1 Stopping DMT Recommendations 2 Statement 1a 3 4 5 In patients with relapsing remitting MS who are stable on DMT and wish to discontinue therapy, clinicians should counsel patients regarding the uncertain risks of stopping therapy and the need for ongoing follow‐up and periodic re‐evaluation of the decision to discontinue DMT. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large 10 10 10 Benefit >> Harm 10 Yes Yes 10 Yes Yes 10 Yes Benefit >>> Harm T Benefit > Harm 0 1 Mildly 0 4 Very 1 Moderate 2 AF 0 10 Modest Occasionally 0 Strength of recommendation 6 Consensus 0 Large 4 Moderate 0 1 R/U 9 Usually C 7 B 10 Critically Important 4 Minimal 4 Always 11 Small 7 Yes Yes Yes Yes Yes A Statement 1b 8 9 Clinicians should advocate that people with MS who are stable (that is, no relapses, no disability progression, stable imaging) on DMT should be allowed to continue on their current DMT. D Domain R 7 Consensus Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% Axioms true < 50% 50% to < 80% 80% to < 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% Confidence in Inference Very low Low Moderate Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Strength of recommendation 10 Rating Benefit > Harm 0 Mildly R/U 10 100% 100% N/A Yes Yes 9 Yes Critically Important 6 Yes Minimal 4 No Always 10 Moderate 7 Yes Small 13 B Yes Benefit >>> Harm 10 2 Yes High 11 1 C 10 Usually Large 0 10 Modest 4 0 100% 9 1 Occasionally 100% 100% Very Moderate 0 10 Benefit >> Harm 0 0 10 3 Yes A 132 Confidential draft: not for distribution, dissemination, or publication 1 2 Statement 2a 3 4 5 Clinicians should advise discontinuation of DMT in patients with SPMS who do not have ongoing relapses (or gadolinium enhancing lesions on MRI), and in addition who have not been ambulatory (EDSS ≥7) for at least 2 years. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large 100% 10 10 10 Benefit >> Harm 1 2 Mildly 0 4 0 1 R/U C 5 Yes 1 Yes Minimal 1 Yes Always 2 Moderate 1 10 Yes 8 Usually Large 10 Yes 10 4 Occasionally Yes Critically Important Modest AF 2 10 Yes 7 Very Moderate Yes Benefit >>> Harm T Benefit > Harm 0 Strength of recommendation 6 Consensus 13 Yes Small 5 B 8 Yes A Statement 2b 8 9 Clinicians should assess the likelihood of future relapse by assessing patient age and relapse history (e.g., frequency, severity, time since most recent relapse or gadolinium enhancing lesion). D Domain R 7 Consensus Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% < 50% 50% to < 80% 80% to < 100% Confidence in Inference Very low Low Moderate Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Strength of recommendation 10 Rating Benefit > Harm 0 Mildly R/U Yes N/A Benefit >>> Harm 5 No Critically Important 1 Yes Minimal 1 Yes Always 8 Moderate 4 Yes Small 10 B Yes High 10 1 C 10 Usually 1 0 10 Yes 100% 10 2 Large Yes 100% Modest Occasionally 100% 6 3 1 10 Very Moderate 1 10 Benefit >> Harm 3 0 10 3 Yes A 133 Confidential draft: not for distribution, dissemination, or publication 1 Statement 3 2 3 Clinicians should discuss discontinuation of currently approved DMTs in patients with PPMS with no MRI disease activity currently receiving such treatments. Domain Rating Rationale is logical < 50% 50% to < 80% 80% to < 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% Axioms true < 50% 50% to < 80% 80% to < 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% 100% Confidence in Inference Very low Low Moderate High Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large Benefit > Harm 1 Mildly 10 10 5 Yes 3 Yes AF C 2 Yes Always 3 Moderate 0 R/U Minimal 9 Usually 0 1 10 Yes 12 4 Large 10 Yes Critically Important Modest Occasionally Yes Benefit >>> Harm T 0 Yes 100% 7 0 Moderate Yes 100% 10 Very 0 100% 10 Benefit >> Harm 2 0 Strength of recommendation 4 Consensus 12 Yes Small 6 B 8 Yes A Statement 4 6 7 Clinician should review the risk of continuing DMTs vs the risk of stopping DMT in patients with CIS on DMTs. Domain R 5 50% to < 80% 80% to < 100% 100% Evidence statements accurate < 50% 50% to < 80% 80% to < 100% 100% Axioms true < 50% 50% to < 80% 80% to < 100% 100% Related evidence strong & applicable Internal inferences logically follow < 50% 50% to < 80% 80% to < 100% 100% < 50% 50% to < 80% 80% to < 100% Confidence in Inference Very low Low Moderate Benefit relative to Harm Harm > Benefit Importance of outcomes Not Important or Variation in preferences Large Feasible Rarely Cost relative to net benefit Very Large D 9 Consensus < 50% Rationale is logical Strength of recommendation 8 Rating Benefit > Harm 1 Mildly R/U C 9 Yes 4 Yes Minimal 4 Yes Always 8 Moderate 9 Yes Small 13 B Yes High 12 1 N/A Critically Important Usually 1 0 100% 14 1 Large 10 Yes Benefit >>> Harm Modest Occasionally Yes 8 0 0 10 Very Moderate 1 10 Benefit >> Harm 0 0 10 10 Yes 4 Yes A 134 Confidential draft: not for distribution, dissemination, or publication 1 REFERENCES 2 Achiron, A., Gabbay,U, Gilad,R, et al (1998). "Intravenous immunoglobulin treatment in 3 multiple sclerosis, effect on relapses." Neurology 50: 398-402. 4 5 Achiron, A., et al. (2004). "Intravenous immunoglobulin treatment following the first 6 demyelinating event suggestive of multiple sclerosis: A randomized, double-blind, placebo- 7 controlled trial." Archives of Neurology 61(10): 1515-1520. 8 multiple sclerosis." NEJM 317: 408-414. T 10 Bornstein, M., Miller, A, Slagle,S, et al (1987). "A pilot trial of COP 1 in exacerbating-remitting AF 9 11 12 Bornstein, M., Miller,A, Slagle,S, et al (1991). "A placebo‐controlled, double‐blind, randomized, 13 two‐center, pilot trial of Cop 1 in chronic progressive multiple sclerosis." Neurology 41(4): 533. R 14 Browne, P., Chandraratna,D, Angood, C, et al (2014). "Atlas of multiple sclerosis 2013: a 16 growing global problem with widespread inequality." Neurology 83: 1022-1024. D 15 17 18 Bruck, W., Gold,R, Lund, BT, et al (2013). "Therapeutic decisions in multiple sclerosis: moving 19 beyond efficacy." JAMA Neurol 70(10): 1315-1324. 20 21 Calabresi, P., Kieseier,BC, Arnold,DL, et al (2014). "Pegylated interferon beta-1a for relapsing- 22 remitting 23 multiple sclerosis (ADVANCE): a randomised, phase 3, 135 Confidential draft: not for distribution, dissemination, or publication 1 double-blind study." Lancet Neurology 13: 657-665. 2 3 Calabresi, P., Radue,E-W, Goodin,D, et al (2014). "Safety and effi cacy of fi ngolimod in 4 patients with 5 relapsing-remitting multiple sclerosis (FREEDOMS II): 6 a double-blind, randomised, placebo-controlled, phase 3 trial." Lancet Neurology. 7 Campbell, J., Ghushchyan,V, Brett McQueen R, et al (2014). "Burden of multiple sclerosis on 9 direct, indirect costs and quality fo live: National US estimates." Multiple Sclerosis and Related Disorders 3: 227-236. 11 AF 10 T 8 Cavadid, D., Wolanski,LJ, Skurnick,J, et al (2009). "Efficacy of treatment of MS with IFNβ-1b 13 or glatiramer acetate by monthly brain MRI in the BECOME study." Neurology 72(23): 1876- 14 1983. D 15 R 12 16 CCMSG (1991). "The Canadian cooperative trial of cyclophosphamide and plasma exchange in 17 progressive multiple sclerosis." Lancet 337(8739): 441-446. 18 19 Cohen, J., Barkhof,F, Comi,G, et al (2010). "Oral Fingolimod or Intramuscular Interferon 20 for Relapsing Multiple Sclerosis." NEJM 362: 402-415. 21 22 Cohen, J., Coles, AJ, Arnold,DL, et al (2012). "Alemtuzumab versus interferon beta 1a as fi rst- 23 line 136 Confidential draft: not for distribution, dissemination, or publication 1 treatment for patients with relapsing-remitting multiple 2 sclerosis: a randomised controlled phase 3 trial." Lancet. 3 4 Cohen, J., Cutter, GR, Fischer,JS, et al (2002). "Benefit of interferon -1a on MSFC 5 progression in secondary progressive MS." Neurology 59: 679-687. 6 Coles, A., Twyman,CL, Arnold,DL, et al (2012). "Alemtuzumab for patients with relapsing 8 multiple sclerosis 9 after disease-modifying therapy: a randomised controlled phase 3 trial." Lancet. AF 10 T 7 11 12 Comi, G., et al. (2001). "Effect of early interferon treatment on conversion to definite multiple 13 sclerosis: a randomised study." Lancet 357(9268): 1576-1582. R 14 Comi, G., et al. (2009). "Effect of glatiramer acetate on conversion to clinically definite multiple 16 sclerosis in patients with clinically isolated syndrome (PreCISe study): a randomised, double- 17 blind, placebo-controlled trial.[Erratum appears in Lancet. 2010 Apr 24;375(9724):1436]." 18 Lancet 374(9700): 1503-1511. D 15 19 20 Confavreaux, C., Vukusic, S (2006). "Age at disability milestones in multiple sclerosis." Brain 21 129: 595-605. 22 137 Confidential draft: not for distribution, dissemination, or publication 1 Confavreux, C., O'Connor, P, Comi,G, et al (2014). "Oral terifl unomide for patients with 2 relapsing multiple 3 sclerosis (TOWER): a randomised, double-blind, placebocontrolled, 4 phase 3 trial." Lancet Neurology 13: 247-256. 5 6 Currier, R., Haerer,AR, Meydrech,EF (1993). "Low dose oral methotrexate treatment of multiple 7 sclerosis: a pilot study." JNNP 56(11): 1217-1218. 8 Ellison, G., Myers,LW, Mickey,MR, et al (1989). "A placebo‐controlled, randomized, double‐ 10 masked, variable dosage, clinical trial of azathioprine with and without methylprednisolone in 11 multiple sclerosis." Neurology 39(8): 1018. AF T 9 12 Etemadifar, M., et al (2010). "Mycophenolate mofetil in combination with interferon beta-1a in 14 the treatment of relapsing-remitting multiple sclerosis: a preliminary study." J Res Med Sci 15 16(1): 1-5. D R 13 16 17 Fazekas, F., Deisenhammer,F, Stasser-Fuchs,S, et al (1997). "Randomised placebo-controlled 18 trial of monthly intravenous 19 immunoglobulin therapy in relapsing-remitting multiple sclerosis." Lancet 349: 598-593. 20 21 Fazekas, F., Lublin,FD, Li,D, et al (2008). "Intravenous immunoglobulin in relapsing remitting 22 multiple sclerosis 23 A dose-finding trial." Neurology 71: 265-271. 138 Confidential draft: not for distribution, dissemination, or publication 1 2 Fox, R., Miller,DH, Phillips,JT, et al (2012). "Placebo-Controlled Phase 3 Study of Oral BG-12 3 or Glatiramer in Multiple Sclerosis." NEJM 367: 1087-1097. 4 5 Fox, R., Miller,DH, Phillips,JT, et al (2012). "Placebo-Controlled Phase 3 Study of Oral BG-12 6 or Glatiramer in Multiple Sclerosis." NEJM 367(12): 1087-1097. 7 Frohman, E., et al (2010). "A randomized, blinded, parallel-group, pilot trial of mycophenolate 9 mofetil (CellCept) compared with interferon beta-1a (Avonex) in patients with relapsingremitting multiple sclerosis." Therapeutic Advances in Neurological Disorders 3(1): 15-28. AF 10 T 8 11 12 Giovannoni, G., Comi,G, Cook, S, et al (2010). "A Placebo-Controlled Trial of Oral 13 Cladribine for Relapsing Multiple Sclerosis." NEJM 362: 416-426. R 14 Gold, R., Giovannoni,G, Selmaj,K, et al (2013). "Daclizumab high-yield process in relapsing- 16 remitting multiple sclerosis (SELECT): a randomised, double-blind, placebo-controlled trial." 17 Lancet 381(9884): 22-28. D 15 18 19 Gold, R., Kappos,L, Arnold,DL, et al (2012). "Placebo-Controlled Phase 3 Study of Oral 20 BG-12 for Relapsing Multiple Sclerosis." NEJM 367: 1098-1107. 21 139 Confidential draft: not for distribution, dissemination, or publication 1 Goodin, D., Frohman,EM, Garmany Jr, GP, et al (2002). "Disease modifying therapies in 2 multiple sclerosis: Subcommittee of the American Academy of Neurology and the MS Council 3 for Clinical Practice Guidelines." Neurology 58: 169-178. 4 5 Goodkin, D., Bailly,RC, Teetzen,ML, et al (1991). "The efficacy of azathioprine in relapsing- 6 remitting multiple sclerosis." Neurology 41(1): 20-25. 7 Goodkin, D., Rudick,RA, Medendorp,SV, et al (1995). "Low-dose (7.5 mg) oral methotrexate 9 reduces the rate of progression in chronic progressive multiple sclerosis." Annals of Neurology 37(1): 30-40. AF 10 T 8 11 12 Goodkin, D. E., et al. (1998). "A phase II study of IV methylprednisolone in secondary- 13 progressive multiple sclerosis." Neurology 51(1): 239-245. R 14 Goodman, A., Rossman,H, Bar-Or,A, et al (2009). "GLANCE; Results of a phase 2, randomized, 16 double-blind, placebo-controlled study." Neurology 72(9): 806-812. D 15 17 18 Graham, R., Mancher, M, Wolman DM et al (2011). Clinical Practice Guidelines we can trust/ 19 Committee on standards for developing trustworthy clinical practice guidelines. I. o. M. o. t. N. 20 A. Board on Health Care Services. Washington DC. 21 22 Group, I. M. S. S. (1993). "Interferon beta-1b is effective in relapsing-remitting multiple 23 sclerosis." Neurology 43: 655-661. 140 Confidential draft: not for distribution, dissemination, or publication 1 2 Guyatt, G., Oxman AD, Schunemann, HJ, et al (2011). "GRADE guidelines: A new series of 3 articles in the Journal of Clinical Epidemiology." J Clin Epidemiology 64: 380-382. 4 5 Guyatt, G., Oxman, AD, Akl, EA, et al (2011). "GRADE guidelines: 1. Introduction-GRADE 6 evidence profiles and summary of findings tables." J Clin Epidemiology 64: 383-394. 7 Hartung, H., Gonsette,R, Konig,N, et al (2002). "Mitoxantrone in progressive multiple sclerosis: 9 a placebo-controlled, double-blind, randomised, multicentre trial " Lancet 360(9350): 20182025. AF 10 T 8 11 12 Hauser, S., Waubant, E, Arnold,DL, et al (2008). "B-Cell Depletion with Rituximab in 13 Relapsing–Remitting Multiple Sclerosis." NEJM 358: 676-688. R 14 Hauser, S. L., et al. (2008). "B-cell depletion with rituximab in relapsing-remitting multiple 16 sclerosis." The New England journal of medicine 358(7): 676-688. D 15 17 18 Hawker, K., O'Connor,P, Freedman,MS, et al (2009). "Rituximab in Patients with Primary 19 Progressive Multiple Sclerosis 20 Results of a Randomized Double-Blind Placebo-Controlled 21 Multicenter Trial." Annals of Neurology 66: 460-471. 22 141 Confidential draft: not for distribution, dissemination, or publication 1 Hommes, O., Sorensen,PS, Fazekas,F, et al (2004). "Intravenous immunoglobulin in secondary 2 progressive 3 multiple sclerosis: randomised placebo-controlled trial." Lancet 364: 1149-1156. 4 5 Investigators, C. T., et al. (2008). "Alemtuzumab vs. interferon beta-1a in early multiple 6 sclerosis." New England Journal of Medicine 359(17): 1786-1801. 7 Jacobs, L., Cookfair,DL, Rudick,RA, et al (1996). "Intramuscular interferon beta-1a for disease 9 progression in relapsing multiple sclerosis." Annals of Neurology 39: 285-294. T 8 AF 10 11 Jacobs, L. D., et al. (2000). "Intramuscular interferon beta-1a therapy initiated during a first 12 demyelinating event in multiple sclerosis. CHAMPS Study Group." New England Journal of 13 Medicine 343(13): 898-904. R 14 Johnson, K., Brooks,BR, Cohen,JA, et al (1995). "Copolymer 1 reduces relapse rate and 16 improves disability in relapsing-remitting multiple sclerosis." Neurology 45: 1268-1276. D 15 17 18 Kappos, L. (1998). "Placebo-controlled multicentre randomised trial of interferon β-1b in 19 treatment of secondary progressive multiple sclerosis " Lancet 352(9139): 1486-1487. 20 21 Kappos, L., et al. (2006). "Treatment with interferon beta-1b delays conversion to clinically 22 definite and McDonald MS in patients with clinically isolated syndromes." Neurology 67(7): 23 1242-1249. 142 Confidential draft: not for distribution, dissemination, or publication 1 2 Kappos, L., Radue,E-W, O'Connor,P, et al (2010). "A Placebo-Controlled Trial of Oral 3 Fingolimod 4 in Relapsing Multiple Sclerosis." NEJM 362(5): 387-401. 5 6 Kappos, L., Wiendl,H, Selmaj,K, et al (2015). "Daclizumab HYP versus Interferon Beta-1a 7 in Relapsing Multiple Sclerosis." NEJM 373: 1418-1428. 8 T 10 Khan, O., Rieckmann,P, Boyke,A, et al (2013). "Three times weekly glatiramer acetate in relapsing–remitting multiple sclerosis." Annals of Neurology 73(6): 705-713. AF 9 11 12 Killian, J., Bressler,RB, Armstrong,RM, et al (1988). "Controlled pilot trial of monthly 13 intravenous cyclophosphamide in multiple sclerosis." Archives of Neurology 45: 27-30. R 14 Leary, S., Miller,D, Stevenson,VL, et al (2003). "Interferon β-1a in primary progressive MS." 16 Neurology 60: 144-151. D 15 17 18 Leist, T., Comi,G, Cree, BAC, et al (2014). "Effect of oral cladribine on time to conversion to 19 clinically 20 definite multiple sclerosis in patients with a first 21 demyelinating event (ORACLE MS): a phase 3 22 randomised trial." Lancet Neurology 13(257-267). 23 143 Confidential draft: not for distribution, dissemination, or publication 1 Lewanska, M., Siger-Zajdel,M, Selmaj,K (2002). "No difference in efficacy of two different 2 doses of intravenous immunoglobulins in MS: clinical and MRI assessment." " European Journal 3 of Neurology 9(6565-572). 4 5 Likosky, W., Fireman,B, Elmore,R, et al (1991). "Intense immunosuppression in chronic 6 progressive multiple sclerosis: The Kaiser study." JNNP 54: 1055-1060. 7 Lublin, F., Reingold, SC, Cohen, JA, et al (2014). "Defining the clinical course of multiple 9 sclerosis: the 2013 revisions." Neurology 83: 278-286. T 8 AF 10 Mikol, D., Barkhof,F, Chang, P, et al (2008). "Comparison of subcutaneous interferon beta-1a 12 with glatiramer acetate in patients with relapsing multiple sclerosis (the REbif vs Glatiramer 13 Acetate in Relapsing MS Disease [REGARD] study): a multicentre, randomised, parallel, open- 14 label trial." Lancet Neurology 7(10): 903-914. R 11 D 15 16 Millefiorini, E., Gasperini,C, Pozzilli,C, et al (1997). "Randomized placebocontrolled 17 trial of mitoxantrone in relapsing-remitting 18 multiple sclerosis: 24-month clinical and MRI outcome." J Neurology 19 244: 153-159. 20 21 Miller, A., Wolinsky,JS, Kappos,L, et al (2014). "Oral teriflunomide for patients with a first 22 clinical episode 23 suggestive of multiple sclerosis (TOPIC): a randomised, 144 Confidential draft: not for distribution, dissemination, or publication 1 double-blind, placebo-controlled, phase 3 trial." Lancet Neurology 13: 977-986. 2 3 Montalban, X., Sastre-Garriga,J, Tintore,M, et al. (2009). "A single-center, randomized, double- 4 blind, placebo-controlled study of interferon beta-1b on primary progressive and transitional 5 multiple sclerosis." Multiple Sclerosis Journal 15(10): 1195-1205. 6 7 NASG (2004). "Interferon beta-1b in secondary progressive MS 8 Results from a 3-year controlled study." Neurology 63(10): 1788-1795. T 9 AF 10 11 Neurology, A. A. o. (2011). Clinical Practice Guideline Process Manual. AAN. A. A. o. 12 Neurology. St. Paul, Minnesota. 13 O'Connor, P., Filippi,M, Arnason,B, et al (2009). "250 μg or 500 μg interferon beta-1b versus 20 15 mg glatiramer acetate in relapsing-remitting multiple sclerosis: a prospective, randomised, 16 multicentre study." Lancet Neurology 8(10): 889-897. D R 14 17 18 O'Connor, P., Freedman,MS, Bar-Or,A, et al (2006). "A Phase II study of the safety and efficacy 19 of teriflunomide in multiple sclerosis with relapses." Neurology 66(6): 894-900. 20 21 O'Connor, P., Wolinsky,JS, Confavreux, C, et al (2011). "Randomized Trial of Oral 22 Teriflunomide 23 for Relapsing Multiple Sclerosis." NEJM 365: 1293-1303. 145 Confidential draft: not for distribution, dissemination, or publication 1 2 Panitch, H., Goodin,DS, Francis,G, et al (2002). "Randomized, comparative study of 3 interferon -1a treatment regimens in MS 4 The EVIDENCE Trial." Neurology 59: 1496-1506. 5 6 Pohlau, D., Przuntek,H, Sailer,M, et al (2007). "Intravenous immunoglobulin in primary and 7 secondary chronic progressive multiple sclerosis: a randomized placebo controlled multicentre 8 study." Multiple Sclerosis 13(9): 1107-1117. T 9 Polman, C., et al (2011). "Diagnostic criteria for multiple sclerosis: 2010 revisions to the 11 McDonald criteria." Ann Neurol 69(2): 292-302. 12 AF 10 Polman, C., O'Connor,PW, Havdrova,E, et al (2006). "A Randomized, Placebo-Controlled Trial 14 of Natalizumab 15 for Relapsing Multiple Sclerosis." NEJM 354(9): 899-910. D R 13 16 17 PRISMS (1998). "Randomised double-blind placebo-controlled study of 18 interferon -1a in relapsing/remitting multiple sclerosis." Lancet 352: 1498-1504. 19 20 Ravnborg, M., et al. (2010). "Methylprednisolone in combination with interferon beta-1a for 21 relapsing-remitting multiple sclerosis (MECOMBIN study): a multicentre, double-blind, 22 randomised, placebo-controlled, parallel-group trial." The Lancet Neurology 9(7): 672-680. 23 146 Confidential draft: not for distribution, dissemination, or publication 1 Ravnborg, M., Sorensen,PS, Andersson,M, et al (2010). "Methylprednisolone in combination 2 with interferon beta-1a for relapsing-remitting multiple sclerosis (MECOMBIN study): a 3 multicentre, double-blind, randomised, placebo-controlled, parallel-group trial." Lancet 4 Neurology 9(7): 672-680. 5 Remington, G., et al (2010). "A one-year prospective, randomized, placebo-controlled, 7 quadruple-blinded, phase II safety pilot trial of combination therapy with interferon beta-1a and 8 mycophenolate mofetil in early relapsing—remitting multiple sclerosis (TIME MS). ." 9 Therapeutic Advances in Neurological Disorders 3(3-13). T 6 AF 10 11 Rice, G., Filippi,M, Comi,G (2000). "Cladribine and progressive MS Clinical and MRI outcomes 12 of a multicenter controlled trial." Neurology 54(5): 1145-1155. 13 Romine, J., Sipe, JC, Koziol,JA, et al (1999). "A Double-Blind, Placebo-Controlled, 15 Randomized Trial of Cladribine in Relapsing-Remitting Multiple Sclerosis." PNAS 111(1): 35- 16 44. 17 D R 14 18 Rudick, R., Stuart,WH, Calabresi,PA, et al (2006). "Natalizumab plus Interferon Beta-1a 19 for Relapsing Multiple Sclerosis." NEJM 354: 911-923. 20 21 Saida, T., Kikuchi,S, Itoyama,Y, et al (2012). "A randomized, controlled trial of fingolimod 22 (FTY720) in Japanese patients with multiple sclerosis." Multiple Sclerosis. 23 147 Confidential draft: not for distribution, dissemination, or publication 1 Sorensen, P., Mellgren,SI, Svenningsson,A, et al (2009). "NORdic trial of oral 2 Methylprednisolone as add-on therapy to Interferon beta-1a for treatment of relapsing-remitting 3 Multiple Sclerosis (NORMIMS study): a randomised, placebo-controlled trial." Lancet 4 Neurology 8(6): 519-529. 5 6 SPECTRIMS (2001). "Randomized controlled trial of interferon- beta-1a in secondary 7 progressive MS." Neurology 56(11): 1496-1504. 8 T 10 Vollmer, T., Sorensen,PS, Selmaj,K, et al (2014). "A randomized placebo-controlled phase III trial of oral laquinimod for multiple sclerosis." J. Neurol 261(4): 773-783. AF 9 11 12 Wingerchuk, D., Carter, JL (2014). "Multiple sclerosis: current and emerging disease-modifying 13 therapies and treatment strategies. ." Mayo Clinic Proceedings 89(2): 225-240. R 14 Wolinsky, J. S., et al. (2007). "Glatiramer acetate in primary progressive multiple sclerosis: 16 results of a multinational, multicenter, double-blind, placebo-controlled trial." Annals of 17 Neurology 61(1): 14-24. D 15 18 19 Zivadinov, R., Rudick,RA, De Masi,R, et al (2001). "Effects of IV methylprednisolone on brain 20 atrophy in relapsing-remitting MS." Neurology 5: 1239-1247. 21 22 ADDITIONAL REFERENCES (survey data) 23 References 148 Confidential draft: not for distribution, dissemination, or publication 1 2 Achiron, A., Gabbay,U, Gilad,R, et al (1998). "Intravenous immunoglobulin treatment in multiple sclerosis, effect on relapses." Neurology 50: 398‐402. 3 4 5 6 Achiron, A., et al. (2004). "Intravenous immunoglobulin treatment following the first demyelinating event suggestive of multiple sclerosis: A randomized, double‐blind, placebo‐controlled trial." Archives of Neurology 61(10): 1515‐1520. 7 8 9 Bornstein, M., Miller, A, Slagle,S, et al (1987). "A pilot trial of COP 1 in exacerbating‐remitting multiple sclerosis." NEJM 317: 408‐414. Bornstein, M., Miller,A, Slagle,S, et al (1991). "A placebo‐controlled, double‐blind, randomized, two‐ center, pilot trial of Cop 1 in chronic progressive multiple sclerosis." Neurology 41(4): 533. 13 14 15 Browne, P., Chandraratna,D, Angood, C, et al (2014). "Atlas of multiple sclerosis 2013: a growing global problem with widespread inequality." Neurology 83: 1022‐1024. 16 17 18 Browne, P., Chandraratna,D, Angood,C, et al (2014). "Atlas of Multiple Sclerosis 2013: A growing global problem with widespread inequity." Neurology 83(11): 1022‐1024. 19 20 21 Bruck, W., Gold,R, Lund, BT, et al (2013). "Therapeutic decisions in multiple sclerosis: moving beyond efficacy." JAMA Neurol 70(10): 1315‐1324. 22 23 Calabresi, P., Kieseier,BC, Arnold,DL, et al (2014). "Pegylated interferon beta‐1a for relapsing‐remitting 24 multiple sclerosis (ADVANCE): a randomised, phase 3, 25 double‐blind study." Lancet Neurology 13: 657‐665. 26 27 Calabresi, P., Radue,E‐W, Goodin,D, et al (2014). "Safety and effi cacy of fi ngolimod in patients with 28 relapsing‐remitting multiple sclerosis (FREEDOMS II): 29 a double‐blind, randomised, placebo‐controlled, phase 3 trial." Lancet Neurology. 30 31 32 33 Campbell, J., Ghushchyan,V, Brett McQueen R, et al (2014). "Burden of multiple sclerosis on direct, indirect costs and quality fo live: National US estimates." Multiple Sclerosis and Related Disorders 3: 227‐236. 34 35 36 Cavadid, D., Wolanski,LJ, Skurnick,J, et al (2009). "Efficacy of treatment of MS with IFNβ‐1b or glatiramer acetate by monthly brain MRI in the BECOME study." Neurology 72(23): 1876‐1983. 37 38 39 CCMSG (1991). "The Canadian cooperative trial of cyclophosphamide and plasma exchange in progressive multiple sclerosis." Lancet 337(8739): 441‐446. D R AF T 10 11 12 149 Confidential draft: not for distribution, dissemination, or publication 1 2 Cohen, J., Barkhof,F, Comi,G, et al (2010). "Oral Fingolimod or Intramuscular Interferon 3 for Relapsing Multiple Sclerosis." NEJM 362: 402‐415. 4 5 Cohen, J., Coles, AJ, Arnold,DL, et al (2012). "Alemtuzumab versus interferon beta 1a as fi rst‐line 6 treatment for patients with relapsing‐remitting multiple 7 sclerosis: a randomised controlled phase 3 trial." Lancet. 8 9 Cohen, J., Cutter, GR, Fischer,JS, et al (2002). "Benefit of interferon ‐1a on MSFC progression in secondary progressive MS." Neurology 59: 679‐687. 11 12 13 Coles, A., Twyman,CL, Arnold,DL, et al (2012). "Alemtuzumab for patients with relapsing multiple sclerosis 14 after disease‐modifying therapy: a randomised controlled 15 phase 3 trial." Lancet. 16 17 18 Comi, G., et al. (2001). "Effect of early interferon treatment on conversion to definite multiple sclerosis: a randomised study." Lancet 357(9268): 1576‐1582. 19 20 21 Comi, G., Gold,R, Dahlke,F, et al (2015). "Relapses in patients treated with fingolimod after previous exposure to natalizumab." Multiple Sclerosis 21(6): 786‐790. 22 23 24 25 26 Comi, G., et al. (2009). "Effect of glatiramer acetate on conversion to clinically definite multiple sclerosis in patients with clinically isolated syndrome (PreCISe study): a randomised, double‐blind, placebo‐ controlled trial.[Erratum appears in Lancet. 2010 Apr 24;375(9724):1436]." Lancet 374(9700): 1503‐ 1511. 27 28 29 Confavreaux, C., Vukusic, S (2006). "Age at disability milestones in multiple sclerosis." Brain 129: 595‐ 605. 30 31 32 Confavreux, C., O'Connor, P, Comi,G, et al (2014). "Oral terifl unomide for patients with relapsing multiple 33 sclerosis (TOWER): a randomised, double‐blind, placebocontrolled, 34 phase 3 trial." Lancet Neurology 13: 247‐256. 35 36 37 Currier, R., Haerer,AR, Meydrech,EF (1993). "Low dose oral methotrexate treatment of multiple sclerosis: a pilot study." JNNP 56(11): 1217‐1218. D R AF T 10 150 Confidential draft: not for distribution, dissemination, or publication Ellison, G., Myers,LW, Mickey,MR, et al (1989). "A placebo‐controlled, randomized, double‐masked, variable dosage, clinical trial of azathioprine with and without methylprednisolone in multiple sclerosis." Neurology 39(8): 1018. 5 6 7 Etemadifar, M., et al (2010). "Mycophenolate mofetil in combination with interferon beta‐1a in the treatment of relapsing‐remitting multiple sclerosis: a preliminary study." J Res Med Sci 16(1): 1‐5. 8 9 10 Fazekas, F., Deisenhammer,F, Stasser‐Fuchs,S, et al (1997). "Randomised placebo‐controlled trial of monthly intravenous 11 immunoglobulin therapy in relapsing‐remitting multiple sclerosis." Lancet 349: 598‐593. 12 13 Fazekas, F., Lublin,FD, Li,D, et al (2008). "Intravenous immunoglobulin in relapsing remitting 14 multiple sclerosis 15 A dose‐finding trial." Neurology 71: 265‐271. 16 17 18 Fox, R., Miller,DH, Phillips,JT, et al (2012). "Placebo‐Controlled Phase 3 Study of Oral BG‐12 or Glatiramer in Multiple Sclerosis." NEJM 367: 1087‐1097. 19 20 Fox, R., Miller,DH, Phillips,JT, et al (2012). "Placebo‐Controlled Phase 3 Study of Oral BG‐12 21 or Glatiramer in Multiple Sclerosis." NEJM 367(12): 1087‐1097. 22 23 24 25 Frohman, E., et al (2010). "A randomized, blinded, parallel‐group, pilot trial of mycophenolate mofetil (CellCept) compared with interferon beta‐1a (Avonex) in patients with relapsing‐remitting multiple sclerosis." Therapeutic Advances in Neurological Disorders 3(1): 15‐28. 26 27 Giovannoni, G., Comi,G, Cook, S, et al (2010). "A Placebo‐Controlled Trial of Oral 28 Cladribine for Relapsing Multiple Sclerosis." NEJM 362: 416‐426. 29 30 31 32 Gold, R., Giovannoni,G, Selmaj,K, et al (2013). "Daclizumab high‐yield process in relapsing‐remitting multiple sclerosis (SELECT): a randomised, double‐blind, placebo‐controlled trial." Lancet 381(9884): 22‐ 28. 33 34 Gold, R., Kappos,L, Arnold,DL, et al (2012). "Placebo‐Controlled Phase 3 Study of Oral 35 BG‐12 for Relapsing Multiple Sclerosis." NEJM 367: 1098‐1107. 36 D R AF T 1 2 3 4 151 Confidential draft: not for distribution, dissemination, or publication 1 2 3 Goodin, D., Frohman,EM, Garmany Jr, GP, et al (2002). "Disease modifying therapies in multiple sclerosis: Subcommittee of the American Academy of Neurology and the MS Council for Clinical Practice Guidelines." Neurology 58: 169‐178. 4 5 6 Goodkin, D., Bailly,RC, Teetzen,ML, et al (1991). "The efficacy of azathioprine in relapsing‐remitting multiple sclerosis." Neurology 41(1): 20‐25. 7 8 9 Goodkin, D., Rudick,RA, Medendorp,SV, et al (1995). "Low‐dose (7.5 mg) oral methotrexate reduces the rate of progression in chronic progressive multiple sclerosis." Annals of Neurology 37(1): 30‐40. Goodkin, D. E., et al. (1998). "A phase II study of IV methylprednisolone in secondary‐progressive multiple sclerosis." Neurology 51(1): 239‐245. 13 14 15 Goodman, A., Rossman,H, Bar‐Or,A, et al (2009). "GLANCE; Results of a phase 2, randomized, double‐ blind, placebo‐controlled study." Neurology 72(9): 806‐812. 16 17 18 19 Graham, R., Mancher, M, Wolman DM et al (2011). Clinical Practice Guidelines we can trust/ Committee on standards for developing trustworthy clinical practice guidelines. I. o. M. o. t. N. A. Board on Health Care Services. Washington DC. 20 21 22 Group, I. M. S. S. (1993). "Interferon beta‐1b is effective in relapsing‐remitting multiple sclerosis." Neurology 43: 655‐661. 23 24 25 Guyatt, G., Oxman AD, Schunemann, HJ, et al (2011). "GRADE guidelines: A new series of articles in the Journal of Clinical Epidemiology." J Clin Epidemiology 64: 380‐382. 26 27 28 Guyatt, G., Oxman, AD, Akl, EA, et al (2011). "GRADE guidelines: 1. Introduction‐GRADE evidence profiles and summary of findings tables." J Clin Epidemiology 64: 383‐394. 29 30 31 Hartung, H., Gonsette,R, Konig,N, et al (2002). "Mitoxantrone in progressive multiple sclerosis: a placebo‐controlled, double‐blind, randomised, multicentre trial " Lancet 360(9350): 2018‐2025. 32 33 Hauser, S., Waubant, E, Arnold,DL, et al (2008). "B‐Cell Depletion with Rituximab in 34 Relapsing–Remitting Multiple Sclerosis." NEJM 358: 676‐688. 35 36 37 Hauser, S. L., et al. (2008). "B‐cell depletion with rituximab in relapsing‐remitting multiple sclerosis." The New England journal of medicine 358(7): 676‐688. 38 39 Hawker, K., O'Connor,P, Freedman,MS, et al (2009). "Rituximab in Patients with Primary D R AF T 10 11 12 152 Confidential draft: not for distribution, dissemination, or publication 1 Progressive Multiple Sclerosis 2 Results of a Randomized Double‐Blind Placebo‐Controlled 3 Multicenter Trial." Annals of Neurology 66: 460‐471. 4 5 6 Hommes, O., Sorensen,PS, Fazekas,F, et al (2004). "Intravenous immunoglobulin in secondary progressive 7 multiple sclerosis: randomised placebo‐controlled trial." Lancet 364: 1149‐1156. 8 9 Iaffaldano, P., Lucisano,G, Pozzilli,C, et al (2015). "Fingolimod versus interferon beta/glatiramer acetate after natalizumab suspension in 11 multiple sclerosis." Brain 138: 3275‐3286. 12 13 14 Investigators, C. T., et al. (2008). "Alemtuzumab vs. interferon beta‐1a in early multiple sclerosis." New England Journal of Medicine 359(17): 1786‐1801. 15 16 17 Jacobs, L., Cookfair,DL, Rudick,RA, et al (1996). "Intramuscular interferon beta‐1a for disease progression in relapsing multiple sclerosis." Annals of Neurology 39: 285‐294. 18 19 20 21 Jacobs, L. D., et al. (2000). "Intramuscular interferon beta‐1a therapy initiated during a first demyelinating event in multiple sclerosis. CHAMPS Study Group." New England Journal of Medicine 343(13): 898‐904. 22 23 24 Johnson, K., Brooks,BR, Cohen,JA, et al (1995). "Copolymer 1 reduces relapse rate and improves disability in relapsing‐remitting multiple sclerosis." Neurology 45: 1268‐1276. 25 26 27 Kappos, L. (1998). "Placebo‐controlled multicentre randomised trial of interferon β‐1b in treatment of secondary progressive multiple sclerosis " Lancet 352(9139): 1486‐1487. 28 29 30 Kappos, L., et al. (2006). "Treatment with interferon beta‐1b delays conversion to clinically definite and McDonald MS in patients with clinically isolated syndromes." Neurology 67(7): 1242‐1249. 31 32 Kappos, L., Radue,E‐W, Comi,G, et al (2015). "Switching from natalizumab to fingolimod 33 A randomized, placebo‐controlled study in RRMS." Neurology 85: 29‐39. 34 35 Kappos, L., Radue,E‐W, O'Connor,P, et al (2010). "A Placebo‐Controlled Trial of Oral Fingolimod 36 in Relapsing Multiple Sclerosis." NEJM 362(5): 387‐401. 37 D R AF T 10 153 Confidential draft: not for distribution, dissemination, or publication 1 Kappos, L., Wiendl,H, Selmaj,K, et al (2015). "Daclizumab HYP versus Interferon Beta‐1a 2 in Relapsing Multiple Sclerosis." NEJM 373: 1418‐1428. 3 4 5 Khan, O., Rieckmann,P, Boyke,A, et al (2013). "Three times weekly glatiramer acetate in relapsing– remitting multiple sclerosis." Annals of Neurology 73(6): 705‐713. 6 7 8 Killian, J., Bressler,RB, Armstrong,RM, et al (1988). "Controlled pilot trial of monthly intravenous cyclophosphamide in multiple sclerosis." Archives of Neurology 45: 27‐30. Leary, S., Miller,D, Stevenson,VL, et al (2003). "Interferon β‐1a in primary progressive MS." Neurology 60: 144‐151. 12 13 Leist, T., Comi,G, Cree, BAC, et al (2014). "Effect of oral cladribine on time to conversion to clinically 14 definite multiple sclerosis in patients with a first 15 demyelinating event (ORACLE MS): a phase 3 16 randomised trial." Lancet Neurology 13(257‐267). 17 18 19 20 Lewanska, M., Siger‐Zajdel,M, Selmaj,K (2002). "No difference in efficacy of two different doses of intravenous immunoglobulins in MS: clinical and MRI assessment." " European Journal of Neurology 9(6565‐572). 21 22 23 Likosky, W., Fireman,B, Elmore,R, et al (1991). "Intense immunosuppression in chronic progressive multiple sclerosis: The Kaiser study." JNNP 54: 1055‐1060. 24 25 26 Lublin, F., Reingold, SC, Cohen, JA, et al (2014). "Defining the clinical course of multiple sclerosis: the 2013 revisions." Neurology 83: 278‐286. 27 28 29 30 31 Mikol, D., Barkhof,F, Chang, P, et al (2008). "Comparison of subcutaneous interferon beta‐1a with glatiramer acetate in patients with relapsing multiple sclerosis (the REbif vs Glatiramer Acetate in Relapsing MS Disease [REGARD] study): a multicentre, randomised, parallel, open‐label trial." Lancet Neurology 7(10): 903‐914. 32 33 Millefiorini, E., Gasperini,C, Pozzilli,C, et al (1997). "Randomized placebocontrolled 34 trial of mitoxantrone in relapsing‐remitting 35 multiple sclerosis: 24‐month clinical and MRI outcome." J Neurology 36 244: 153‐159. 37 38 Miller, A., Wolinsky,JS, Kappos,L, et al (2014). "Oral teriflunomide for patients with a first clinical episode D R AF T 9 10 11 154 Confidential draft: not for distribution, dissemination, or publication suggestive of multiple sclerosis (TOPIC): a randomised, 2 double‐blind, placebo‐controlled, phase 3 trial." Lancet Neurology 13: 977‐986. 3 4 5 6 Montalban, X., Sastre‐Garriga,J, Tintore,M, et al. (2009). "A single‐center, randomized, double‐blind, placebo‐controlled study of interferon beta‐1b on primary progressive and transitional multiple sclerosis." Multiple Sclerosis Journal 15(10): 1195‐1205. 7 8 9 MSIF (2013). "Atlas of MS." from https://www.msif.org/about‐ us/advocacy/atlas/?gclid=CKuDwICjls4CFUQvgQodbuEOeQ. 10 11 NASG (2004). "Interferon beta‐1b in secondary progressive MS 12 13 Results from a 3‐year controlled study." Neurology 63(10): 1788‐1795. 14 15 16 Neurology, A. A. o. (2011). Clinical Practice Guideline Process Manual. AAN. A. A. o. Neurology. St. Paul, Minnesota. 17 18 19 20 O'Connor, P., Filippi,M, Arnason,B, et al (2009). "250 μg or 500 μg interferon beta‐1b versus 20 mg glatiramer acetate in relapsing‐remitting multiple sclerosis: a prospective, randomised, multicentre study." Lancet Neurology 8(10): 889‐897. 21 22 23 O'Connor, P., Freedman,MS, Bar‐Or,A, et al (2006). "A Phase II study of the safety and efficacy of teriflunomide in multiple sclerosis with relapses." Neurology 66(6): 894‐900. 24 25 O'Connor, P., Wolinsky,JS, Confavreux, C, et al (2011). "Randomized Trial of Oral Teriflunomide 26 for Relapsing Multiple Sclerosis." NEJM 365: 1293‐1303. 27 28 Panitch, H., Goodin,DS, Francis,G, et al (2002). "Randomized, comparative study of 29 interferon ‐1a treatment regimens in MS 30 The EVIDENCE Trial." Neurology 59: 1496‐1506. 31 32 33 34 Pohlau, D., Przuntek,H, Sailer,M, et al (2007). "Intravenous immunoglobulin in primary and secondary chronic progressive multiple sclerosis: a randomized placebo controlled multicentre study." Multiple Sclerosis 13(9): 1107‐1117. 35 36 37 Polman, C., et al (2011). "Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria." Ann Neurol 69(2): 292‐302. 38 D R AF T 1 155 Confidential draft: not for distribution, dissemination, or publication 1 2 Polman, C., O'Connor,PW, Havdrova,E, et al (2006). "A Randomized, Placebo‐Controlled Trial of Natalizumab 3 for Relapsing Multiple Sclerosis." NEJM 354(9): 899‐910. 4 5 PRISMS (1998). "Randomised double‐blind placebo‐controlled study of 6 interferon ‐1a in relapsing/remitting multiple sclerosis." Lancet 352: 1498‐1504. 7 8 Prosperini, L., Annovazzi,P, Capobianco,M, et al (2015). "Natalizumab discontinuation in patients with 9 multiple sclerosis: Profiling risk and benefits at therapeutic crossroads." Multiple Sclerosis Journal. 11 12 13 14 Ravnborg, M., et al. (2010). "Methylprednisolone in combination with interferon beta‐1a for relapsing‐ remitting multiple sclerosis (MECOMBIN study): a multicentre, double‐blind, randomised, placebo‐ controlled, parallel‐group trial." The Lancet Neurology 9(7): 672‐680. 15 16 17 18 Ravnborg, M., Sorensen,PS, Andersson,M, et al (2010). "Methylprednisolone in combination with interferon beta‐1a for relapsing‐remitting multiple sclerosis (MECOMBIN study): a multicentre, double‐ blind, randomised, placebo‐controlled, parallel‐group trial." Lancet Neurology 9(7): 672‐680. 19 20 21 22 23 Remington, G., et al (2010). "A one‐year prospective, randomized, placebo‐controlled, quadruple‐ blinded, phase II safety pilot trial of combination therapy with interferon beta‐1a and mycophenolate mofetil in early relapsing—remitting multiple sclerosis (TIME MS). ." Therapeutic Advances in Neurological Disorders 3(3‐13). 24 25 26 Rice, G., Filippi,M, Comi,G (2000). "Cladribine and progressive MS Clinical and MRI outcomes of a multicenter controlled trial." Neurology 54(5): 1145‐1155. 27 28 29 Romine, J., Sipe, JC, Koziol,JA, et al (1999). "A Double‐Blind, Placebo‐Controlled, Randomized Trial of Cladribine in Relapsing‐Remitting Multiple Sclerosis." PNAS 111(1): 35‐44. 30 31 Rudick, R., Stuart,WH, Calabresi,PA, et al (2006). "Natalizumab plus Interferon Beta‐1a 32 for Relapsing Multiple Sclerosis." NEJM 354: 911‐923. 33 34 35 Saida, T., Kikuchi,S, Itoyama,Y, et al (2012). "A randomized, controlled trial of fingolimod (FTY720) in Japanese patients with multiple sclerosis." Multiple Sclerosis. 36 37 38 39 Sorensen, P., Mellgren,SI, Svenningsson,A, et al (2009). "NORdic trial of oral Methylprednisolone as add‐ on therapy to Interferon beta‐1a for treatment of relapsing‐remitting Multiple Sclerosis (NORMIMS study): a randomised, placebo‐controlled trial." Lancet Neurology 8(6): 519‐529. D R AF T 10 156 Confidential draft: not for distribution, dissemination, or publication 1 2 3 SPECTRIMS (2001). "Randomized controlled trial of interferon‐ beta‐1a in secondary progressive MS." Neurology 56(11): 1496‐1504. 4 5 6 Vollmer, T., Sorensen,PS, Selmaj,K, et al (2014). "A randomized placebo‐controlled phase III trial of oral laquinimod for multiple sclerosis." J. Neurol 261(4): 773‐783. 7 8 9 Wingerchuk, D., Carter, JL (2014). "Multiple sclerosis: current and emerging disease‐modifying therapies and treatment strategies. ." Mayo Clinic Proceedings 89(2): 225‐240. Wolinsky, J. S., et al. (2007). "Glatiramer acetate in primary progressive multiple sclerosis: results of a multinational, multicenter, double‐blind, placebo‐controlled trial." Annals of Neurology 61(1): 14‐24. 13 14 15 Zivadinov, R., Rudick,RA, De Masi,R, et al (2001). "Effects of IV methylprednisolone on brain atrophy in relapsing‐remitting MS." Neurology 5: 1239‐1247. 16 AF 17 T 10 11 12 D R 18 157 Confidential draft: not for distribution, dissemination, or publication 1 2 1. Saposnik, G., et al., Decision making under uncertainty, therapeutic inertia, and 3 physicians' risk preferences in the management of multiple sclerosis (DIScUTIR MS). BMC 4 Neurol, 2016. 16(1): p. 58. 5 2. 6 Challenges reported internationally with emerging treatment complexity. Mult Scler Relat 7 Disord, 2015. 4(4): p. 320-8. 8 3. 9 in patients with multiple sclerosis. Mult Scler, 2004. 10(6): p. 643-50. Ng, P., S. Murray, and S.M. Hayes, Clinical decision-making in multiple sclerosis: T Heesen, C., et al., Decisional role preferences, risk knowledge and information interests 4. Kasper, J., et al., Informed shared decision making about immunotherapy for patients 11 with multiple sclerosis (ISDIMS): a randomized controlled trial. Eur J Neurol, 2008. 15(12): p. 12 1345-52. 13 5. 14 injectable multiple sclerosis treatments in Germany. Ther Adv Neurol Disord, 2016. 9(2): p. 95- 15 104. 16 6. 17 therapy: a choice-based conjoint analysis. Ther Adv Neurol Disord, 2014. 7(6): p. 263-75. 18 7. 19 modifying therapies: development and results of a ratings-based conjoint analysis. Int J MS Care, 20 2015. 17(2): p. 74-82. 21 8. 22 multiple sclerosis patients. Qual Life Res, 2015. 24(9): p. 2173-82. AF 10 D R Poulos, C., et al., A discrete-choice experiment to determine patient preferences for Utz, K.S., et al., Patient preferences for disease-modifying drugs in multiple sclerosis Wilson, L.S., et al., Patient preferences for attributes of multiple sclerosis disease- Rosato, R., et al., Quality of life and patient preferences: identification of subgroups of 23 158 Confidential draft: not for distribution, dissemination, or publication 1 2 3 4 5 Browne, P., Chandraratna,D, Angood,C, et al. (2014). Atlas of Multiple Sclerosis 2013: A growing global problem with widespread inequity. Neurology, 83(11), 1022‐1024. Comi, G., Gold,R, Dahlke,F, et al. (2015). Relapses in patients treated with fingolimod after previous exposure to natalizumab. Multiple Sclerosis, 21(6), 786‐790. Iaffaldano, P., Lucisano,G, Pozzilli,C, et al. (2015). Fingolimod versus interferon beta/glatiramer 6 acetate after natalizumab suspension in 7 8 multiple sclerosis. Brain, 138, 3275‐3286. Kappos, L., Radue,E‐W, Comi,G, et al. (2015). Switching from natalizumab to fingolimod A randomized, placebo‐controlled study in RRMS. Neurology, 85, 29‐39. MSIF. (2013). Atlas of MS. from https://www.msif.org/about‐ us/advocacy/atlas/?gclid=CKuDwICjls4CFUQvgQodbuEOeQ Prosperini, L., Annovazzi,P, Capobianco,M, et al. (2015). Natalizumab discontinuation in patients with 13 multiple sclerosis: Profiling risk and benefits at 14 therapeutic crossroads. Multiple Sclerosis Journal. doi: DOI: 10.1177/ 15 1352458515570768 T 9 10 11 12 D R AF 16 159