Download Leflunomide-associated tuberculosis?

Rheumatology 2006;45:228–243
Letters to the Editor
Leflunomide-associated tuberculosis?
SIR, Tuberculosis (TB) is still one of the world’s most frequently
occurring infectious diseases. The incidence of active TB in Slovenia
is 14.7 patients per 105 inhabitants [1]. In Spain, the risk of TB
infection in rheumatoid arthritis (RA) patients is increased fourfold
compared with the general Spanish population [2], which is not the
case in the USA, a country with a low incidence of TB [3]. The major
risk factor for active disease seems to be immunosuppressive
therapy. Following the introduction of TNF--blocking agents in
order to control RA, a growing number of active TB cases have
been reported, leading to renewed interest in this disease.
One of the widely used and efficacious disease-modifying
antirheumatic drugs (DMARDs) is leflunomide. We used leflunomide from early 1990 as part of early phase II studies [3] and later
in phase III clinical trials [4]. Among 2000 randomly selected
RA patients in Slovenia (with a population of 1.5 million adults
and a prevalence of RA of 1.2%), 13% have been treated with
leflunomide.
Adverse effects associated with leflunomide have been closely
followed and, as documented, are generally mild and only seldom
life-threatening even in a 5-yr follow up study [6]. We present five
cases of active pulmonary TB that developed during treatment
with leflunomide. Major characteristics of our patients are
presented in Table 1.
Cases of active TB following RA treatment with corticosteroids,
methotrexate and TNF--blocking agents have already been
reported. So far, to our knowledge, active TB in association with
leflunomide treatment has not been described. Nevertheless,
leflunomide might increase susceptibility to reactivation TB disease.
Two possible explanations exist: general immunosuppression
and/or specific anti-TNF activity of leflunomide.
Leflunomide has anti-inflammatory and immunomodulatory
properties. Its active metabolite mediates the suppression of B and
T cells, inhibiting the de novo synthesis of pyrimidine nucleotides.
Higher doses also inhibit tyrosine kinases, responsible for T and B
cell signalling, with subsequent reduction in proinflammatory
cytokines [7].
The suppressive effect of leflunomide on TNF- signalling
might be another possible mechanism for effectiveness in RA [8, 9].
It was demonstrated that leflunomide inhibits TNF--induced
NF-B activation, TNF-mediated cytotoxicity and the generation of TNF-induced reactive oxygen intermediates and lipid
peroxidation [10].
Optimal activity of TNF- and other cytokines (IFN-, IL-12,
IL-15) is crucial in host defence against bacteria, including
Mycobacterium tuberculosis. The essential role of TNF- has
been shown in studies on animal models indicating that inhibition
of TNF- increases the frequency and severity of reactivation TB
disease. This cytokine acts in granuloma formation [11].
An analogy can be drawn with the use of biological anti-TNF-
agents, where increased TB risk and frequency has been reported
[12]. However, in contrast to reactivation TB disease in patients
treated with TNF- blockers, where presentation is frequently
atypical, disseminated and extrapulmonary, the pattern of
TB disease in our cases was classical pulmonary. This difference
also exists in the period between introduction of the immunosuppressive therapy and the occurrence of TB, which in our cases
was longer (average duration 66 weeks) than in TB cases
associated with infliximab, where the median duration reported
was 11 weeks [13].
In four out of five patients, leflunomide was the only DMARD
used at the time of TB manifestation, in three cases it was
combined with corticosteroids (methylprednisolone 8 mg daily),
and in one case with methotrexate (7.5 mg weekly). In the patients
treated with a combination of leflunomide and methotrexate
or corticosteroids, we cannot exclude the contribution of both
concomitant drugs to active TB disease. However, it is less
likely since methotrexate is a standard treatment and the most
commonly prescribed drug for RA patients at our department, but
until now we are not aware of any TB case connected with either
methotrexate or corticosteroids in the last 5 yr.
In conclusion, it is unclear whether leflunomide therapy
increases the risk of reactivation TB beyond the elevated rates of
reactivation TB documented for RA patients. However, temporal
association of reactivation TB disease and leflunomide treatment
may suggest this possibility. Reported cases underline the necessity
of careful monitoring for reactivation TB disease during treatment
with leflunomide.
Key messages
Rheumatology
Rheumatology 2006;45:228–229
doi:10.1093/rheumatology/kei173
Advance Access publication 30 November 2005
Patients treated with leflunomide may
develop tuberculosis.
The authors have declared no conflicts of interest. B.R. has
participated as a site investigator in phase II and III clinical trials
of leflunomide and S.P. and M.T. as co-investigators.
TABLE 1. Patient characteristics
Patient
Sex
Diagnosis
Age/disease
duration (yr)
Previous DMARDs
Corticosteroids
Introduction of
leflunomide (date)
1
2
3
4
5
F
M
F
F
F
RA
RA
RA
RA
JCA
58/9
50/12
71/25
33/2
24/23
MTX, SSZ, Au, Cy, CQ
MTX, SSZ, Au, CQ
Au
None
SSZ, MTX CY, INFa
Yes
Yes
No
No
Yes
Monotherapy (April 2000)
Monotherapy (April 2000)
Monotherapy (March 2003)
Monotherapy (July 2003)
Combination with MTX (April 2003)
Diagnosis of active
pulmonary TB
January 2002
September 2002
November 2003
March 2004
December 2004
F, female; M, male; RA, rheumatoid arthritis; JCA, juvenile chronic arthritis; MTX, methotrexate, SSZ, sulphasalazine; CQ, chloroquine;
Au, parenteral gold; Cy, cyclosporin; INF, infliximab. aIneffective treatment with infliximab from November 2002 to February 2003.
ß The Author 2005. Published by Oxford University Press on behalf of the British Society for Rheumatology. All rights reserved. For Permissions, please email: [email protected]
Letters to the Editor
A. HOČEVAR, B. ROZMAN, S. PRAPROTNIK, B. LESTAN,
D. ERŽEN1, V. PETRIC2, M. TOMŠIČ
Department of Rheumatology, University Medical Centre,
Ljubljana, 1University Clinic of Respiratory and Allergic
Diseases, Golnik and 2Department of Internal Medicine, General
Hospital, Murska Sobota, Slovenia
Accepted 23 September 2005
Correspondence
to:
M.
Tomšič,
Department
of
Rheumatology, University Medical Centre, Vodnikova 62, 1000
Ljubljana, Slovenia. E-mail: [email protected]
1. Praprotnik S, Rozman B, Tomsic M. New issues in tuberculosis.
Ann Rheum Dis 2005;64:795.
2. Carmona L, Hernandez-Garcia C, Vadillo C et al. Increased risk
of tuberculosis in patients with rheumatoid arthritis. J Rheumatol
2003;30:1436–9.
3. Wolfe F, Michaud K, Anderson J, Urbansky K. Tuberculosis
infection in patients with rheumatoid arthritis and the effect of
infliximab therapy. Arthritis Rheum 2004;50:372–9.
4. Mladenovic V, Domljan Z, Rozman B et al. Safety and effectiveness
of leflunomide in the treatment of patients with active rheumatoid
arthritis. Results of a randomized, placebo-controlled, phase II
study. Arthritis Rheum 1995;38:1595–603.
5. Smolen JS, Kalden JR, Scott DL et al. Efficacy and safety of
leflunomide compared with placebo and sulphasalazine in active
rheumatoid arthritis: a double-blind randomised, multicentre trial.
European Leflunomide Study Group. Lancet 1999;353:259–66.
6. Kalden JR, Schattenkirchner M, Sorensen H et al. The efficacy and
safety of leflunomide in patients with active rheumatoid arthritis:
a five-year followup study. Arthritis Rheum 2003;48:1513–20.
7. Cannon GW, Kremer JM. Leflunomide. Rheum Dis Clin North Am
2004;30:295–309.
8. Alldred A, Emery P. Leflunomide: a novel DMARD for the
treatment of rheumatoid arthritis. Expert Opin Pharmacother.
2001;2:125–37.
9. Miceli-Richard C, Dougados M. Leflunomide for the treatment of rheumatoid arthritis. Expert Opin Pharmacother 2003;
4:987–97.
10. Manna SK, Mukhopadhyay A, Aggarwal BB. Leflunomide suppresses TNF-induced cellular responses: effects on NF-B, activator
protein-1, c-Jun N-terminal protein kinase and apoptosis. J
Immunol 2000;165:5962–9.
11. Bieber J, Kavanaugh A. Consideration of the risk and treatment
of tuberculosis in patients who have rheumatoid arthritis and
receive biologic treatments. Rheum Dis Clin North Am 2004;
30:257–70.
12. Wallis RS, Broder MS, Wong JY, Hanson ME, Beenhouwer DO.
Granulomatous infectious diseases associated with tumor necrosis
factor antagonists. Clin Infect Dis 2004;38:1261–5.
13. Keane J. Tumor necrosis factor blockers and reactivation of latent
tuberculosis. Clin Infect Dis 2004:39:300–2.
Rheumatology 2006;45:229–230
doi:10.1093/rheumatology/kei196
Advance Access publication 13 December 2005
Sequential synovial fluid sampling suggests plasma
and synovial fluid IL-6 vary independently in
rheumatoid arthritis
SIR, Inflammatory cytokines, including interleukin-6 (IL-6), are
raised in rheumatoid arthritis (RA). Single paired daytime samples
show that the IL-6 and TNF- concentrations in synovial fluid
exceed blood concentrations during the day [1–3], and these and
229
other data suggest that joints are the source. Recent data have
shown that plasma IL-6 in RA follows a circadian pattern with
a peak at 6 a.m. [4]. If IL-6 is generated inside joints and
diffuses into plasma, it might be expected that the synovial fluid
IL-6 concentration would also undergo a large diurnal variation.
Such investigations necessitate the development of a safe and
acceptable synovial catheter allowing sequential synovial
fluid sampling. Following the approval of the United Bristol
Healthcare Trust Research Ethics Committee, we undertook a
daytime pilot study for this purpose. In addition we obtained
preliminary data for IL-6 variations in paired synovial fluid and
plasma samples.
We recruited six volunteers aged 36–64 yr with RA according
to the criteria of the American College of Rheumatology. Each
volunteer had active disease (at least three swollen and three tender
joints) and a large knee effusion. After establishing the technique,
we obtained paired blood and synovial fluid samples in three
patients. One patient was taking prednisolone 7.5 mg daily, but
no other patient was taking glucocorticoids and none had had
intra-articular or intra-muscular glucocorticoids for more than
10 weeks. After we had obtained written consent, each patient was
admitted and remained on bed rest between 8.30 a.m. and 5.30 p.m.
With the patient on the bed, we used an aseptic non-touch
technique including sterile gown and gloves to insert a synovial
catheter. The knee was exposed, a large area of skin carefully
cleaned with iodine 1% antiseptic solution, and sterile drapes were
used to produce an aseptic field. Up to 2 ml bupivacaine 0.5% was
infiltrated under the skin adjacent to the lateral parapatellar joint
space and towards the synovial membrane. After local anaesthesia
was obtained, a sterile disposable intravenous (i.v.) cannula (14
gauge BD Venflon; Becton Dickinson Infusion Therapy, Sweden)
was inserted into the parapatellar joint space. The cannula was
connected to a three-way tap with flexible extension (BD Connecta
Plus 3; Becton Dickinson Infusion Therapy). Sterile film wound
dressings (C-View; Unomedical, UK) enabled the synovial cannula
to remain perpendicular to the skin. After taking microbiological
and orthopaedic advice, we chose not to use prophylactic
antibiotics. A standard i.v. cannula was used for sequential
blood sampling. No samples were taken for at least 1 h after
cannulae insertion. The i.v. cannula was flushed with 10 ml 0.9%
saline after insertion and before and after each blood sample. The
first 2 ml of each sample taken from the knee and the first 4 ml of
each sample taken from the i.v. cannula were discarded. Paired
blood and synovial fluid samples were taken through the day.
Plasma and synovial fluid samples were stored at 80 C and
analysed using solid-phase sandwich ELISA (enzyme-linked
immunosorbent assay) kits (Diaclone Research, detection limit
2 pg/ml, inter- and intra-assay variation <6%). A colleague not
directly involved in the project asked each volunteer for comments
about their experience.
The synovial catheter allowed sequential synovial fluid sampling
over 8 h. From the independent interviews, the procedure was
acceptable to volunteers. The only adverse effect noted was
slight knee stiffness in two volunteers, and there were no other
adverse effects. In addition we obtained a small number of paired
synovial fluid and blood samples (Fig. 1). Plasma IL-6 decreased
exponentially by 95%, from a mean [95% confidence interval (CI)]
of 19.1 (34.1, 4.1) pg/ml at 10 a.m. to <2 (<2, <2) pg/ml at 5 p.m.
(paired t-test, P<0.05). These results are similar to previously
published data [4]. Synovial fluid IL-6 changed little during the 8 h
(non-significant decrease of 30%), with 10 a.m. IL-6 mean (95%
CI) 12.0 (17.5, 6.5) ng/ml and 5 p.m. mean 7.9 (11.1, 4.7) ng/ml.
The patient with the highest synovial fluid IL-6 concentration had
the lowest plasma IL-6 concentration.
Developing a safe and acceptable synovial catheter was the
prime reason for this study. We found that sequential synovial fluid
samples can be taken safely through a synovial catheter over 8 h by
a technique that is acceptable to volunteers and avoids repeated
arthrocentesis. Our very preliminary values are similar to previous