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
Clinical article Bone marrow response as a potential biomarker of outcomes in glioblastoma patients *Eugene J. Vaios, AB,1,5 Brian V. Nahed, MD, MSc,1,5 Alona Muzikansky, MA,2 Amir T. Fathi, MD,1,3 and Jorg Dietrich, MD, PhD1,4 Harvard Medical School; and Departments of 2Biostatistics, 3Hematology-Oncology, 4Neuro-Oncology, and 5Neurosurgery, Massachusetts General Hospital, Boston, Massachusetts 1 Objective Glioblastoma (GBM) is a highly aggressive malignancy that requires a multidisciplinary therapeutic approach of surgery, chemotherapy, and radiation therapy, but therapy is frequently limited by side effects. The most common adverse effect of chemotherapy with temozolomide (TMZ) is myelosuppression. It remains unclear whether the degree of bone-marrow suppression might serve as a biomarker for treatment outcome. The aim of the current study was to investigate whether the degree of bone-marrow toxicity in patients treated with TMZ correlates with overall survival (OS) and MRI-based time to progression (progression-free survival [PFS]). Methods Complete blood counts and clinical and imaging information were collected retrospectively from 86 cases involving GBM patients who had completed both radiation therapy and at least 6 monthly cycles of chemotherapy with TMZ. Results Using a multivariate Cox proportional hazard model, it was observed that MGMT promoter methylation, wild-type EGFR, younger patient age at diagnosis, and treatment-induced decreases in white blood cell counts were associated with improved OS. The 2-year survival rate was 25% and 58% for patients with increases and decreases, respectively, in white blood cell counts from baseline over 6 months of TMZ treatment. Consistent with the literature, IDH mutation and MGMT promoter methylation were associated with better PFS and OS. IDH mutation and MGMT promoter methylation were not correlated with changes in peripheral red blood cell or white blood cell counts. Conclusions Decreases in white blood cell counts might serve as a potential biomarker for OS and PFS in malignant glioma patients treated with radiation therapy and TMZ. It remains unclear whether treatment-induced changes in white blood cell counts correlate with drug-induced antitumor activity or represent an independent factor of the altered local and systemic tumor environment. Additional studies will be needed to determine dose dependence for chemotherapy based upon peripheral blood counts. http://thejns.org/doi/abs/10.3171/2016.7.JNS16609 Key Words glioblastoma; temozolomide; bone marrow; biomarker; overall survival; oncology G (GBM) is the most common and most aggressive primary brain tumor, with few therapeutic advances over the last 20 years.1 The standard of care includes surgery with a goal of maximal safe resection, followed by a combination of radiation and chemotherapy.16 Chemoradiation includes radiotherapy 5 lioblastoma days per week over 6 weeks in combination with daily temozolomide (TMZ), followed by at least 6 monthly cycles of TMZ administered on 5 consecutive days per 28-day cycle.8,16 Patients may continue on TMZ for up to 12 months at some institutions unless there are intolerable side effects such as myelosuppression. Despite standard Abbreviations EGFR = epidermal growth factor receptor; GBM = glioblastoma; OS = overall survival; PFS = progression-free survival; TMZ = temozolomide. SUBMITTED March 18, 2016. ACCEPTED July 25, 2016. include when citing Published online October 14, 2016; DOI: 10.3171/2016.7.JNS16609. * Drs. Nahed and Dietrich contributed equally to this work. ©AANS, 2016 J Neurosurg October 14, 2016 1 E. J. Vaios et al. treatment, GBM remains incurable, with a median survival of less than 19 months and only a 30% probability of survival 2 years after diagnosis.1,19 The genetic heterogeneity of GBM and the emergence of various resistance mechanisms are proposed limitations of standardized therapies.1,18 However, MGMT promoter methylation and IDH1 mutation have been identified in genome-wide association studies as robust genetic markers for improved clinical outcomes in patients treated with standard chemoradiation.1,18 While IDH status is considered an independent biomarker for clinical outcomes, MGMT promoter methylation is associated with improved response to TMZ and radiation and thus serves as a predictive biomarker of more favorable treatment outcomes. This observed relationship between the molecular-genetic tumor signature and treatment response has elevated the importance of noninvasive biomarkers that enable realtime selection and adjustment of personalized therapies. The unique role of the local and systemic tumor environment is increasingly recognized in overall tumor biology and treatment outcome.18,23 The identification of a peripheral biomarker for treatment response and overall survival (OS) would improve patient management and perhaps alter chemotherapy (e.g., TMZ) dosing protocols, which currently do not account for individual pharmacogenetics, pharmacokinetics, or pharmacodynamics.2 The use of circulating blood counts as a marker of drug activity and clinical outcomes is an emerging area of investigation. The current understanding of the effect of chemotherapeutic agents suggests that greater toxicity to various organ systems, such as the bone marrow, might reflect increased potency and, conversely, be associated with a more favorable antitumor profile.11,12 For instance, in patients with renal cell carcinoma, leukopenia induced by the tyrosine kinase inhibitor sunitinib was identified as an independent and prognostic marker for improved response rate and progression-free survival (PFS).4 This finding was further supported by work from Zhu et al.,24 suggesting that sunitinib-induced decreases in neutrophils, monocytes, and platelets were associated with improved progression and survival outcomes in patients with hepatocellular carcinoma. TMZ therapy, in combination with radiotherapy, significantly improves OS in GBM patients.15 However, dosing is based solely on patient body surface area and does not account for variability in resistance mechanisms and drug metabolism, raising the possibility that some patients may be dosed subtherapeutically. Inadequate dosing may partially contribute to the observed resistance to cytotoxic treatment and ultimately tumor recurrence. One of the most common adverse effects of chemotherapy with TMZ is myelosuppression, including thrombocytopenia and leukopenia. It remains unclear whether patients who do not show a notable decrease in blood counts in response to TMZ are treated subtherapeutically. Therefore, we hypothesized that changes in circulating blood counts may be predictive of clinical outcomes, with alterations in blood counts indicative of myelosuppression being associated with improved OS and PFS. 2 J Neurosurg October 14, 2016 Methods We conducted a retrospective, chart-review analysis of clinical and demographic data from patients who previously underwent surgery and treatment for primary GBM at the Massachusetts General Hospital between 2007 and 2014. Patient data were obtained from a Massachusetts General Hospital institutional database. This time interval allowed enough time for diagnosis and to follow a complete blood count throughout the patient’s disease course. This study received institutional review board approval from Massachusetts General Hospital for all activities. Eligibility All patients were treated at the Massachusetts General Hospital and met the following eligibility criteria: newly diagnosed with GBM (WHO Grade IV) between January 1, 2007, and July 30, 2014; 18 years of age or older at the time of diagnosis; surgical biopsy/resection after initial presentation; and treatment with at least 6 cycles of monthly TMZ. Patients who did not complete 6 months of TMZ therapy for any reason were excluded from the study. Variables Descriptive information, including age, sex, and steroid use, was collected. Steroid use was defined as exposure to steroids (e.g., dexamethasone) at any given time during the course of chemotherapy. Genetic information including chromosomal abnormalities, point mutations, and gene methylation was recorded. This included known prognostic markers for gliomas such as epidermal growth factor receptor (EGFR) amplification, MGMT promoter methylation, IDH mutation, and 1p/19q co-deletion. The genetic characteristics of the sample are reported as the percentage of those patients for whom that genetic variable was tested. Absolute peripheral blood platelet, red blood cell, white blood cell, eosinophil, basophil, lymphocyte, neutrophil, and monocyte count measurements were recorded at a maximum of 15 discrete time points during the course of treatment. Time points included before surgery, after surgery, before chemoradiation, and before each monthly TMZ treatment cycle. PFS and OS were also assessed. Statistics The primary outcome measure was OS, which was defined as the length of time from the date of initial diagnosis to time of death or last date known to be alive for those who were censored. The secondary outcome measure was PFS, which was defined as the length of time from initial diagnosis to the time of first progression based on radiology report and clinician notes indicating a switch in therapy or last date known to be progression free for censored patients. The effect of changes in peripheral blood counts on clinical outcomes was assessed during the interval between baseline measurement (before chemoradiation) and Cycle 6 of monthly TMZ, as this time interval had the greatest sample size and was controlled for the immunosuppressive effect of steroid use during surgery and chemoradiation. Baseline and 6-month blood counts were Bone marrow response as a potential biomarker performed closest to the time of chemoradiation or TMZ initiation, but no earlier than 2 weeks prior. Changes in blood counts are reported as the percentage change from baseline to Cycle 6 of monthly TMZ. Genetic variables known to be strong prognostic markers were compared with clinical outcomes using logistic regression or Pearson’s chi-square test. Spearman or Pearson correlation coefficients were estimated to measure the relation between baseline demographic variables and clinical outcome measures. Wilcoxon rank-sum tests and Kruskal-Wallis tests by ranks were used to examine differences in mean blood count changes between groups of patients stratified by IDH mutation and MGMT promoter methylation. Univariate and multivariate Cox proportional hazards models were used to evaluate variables for association with PFS and OS. Variables were chosen for multivariate analysis using the backward selection method based on statistical significance in univariate analysis. Percentage changes from baseline in white blood cell and red blood cell counts were included in the multivariate model as dichotomized variables, indicating either an increase or a decrease in that blood count from baseline. IDH mutation was excluded from the multivariate analysis due to a sample size less than 10. A Wilcoxon rank-sum test was performed to assess the association between changes in neutrophils and steroid use. Neutrophil counts at Cycle 6 of TMZ were compared between patients with and without steroid treatment using a 2-sample t-test. Changes in neutrophil counts were excluded from the multivariate analysis as they were confounded by steroid use. Survival probabilities were compared between patient groups, stratified by the dichotomous white blood cell variable, using the log-rank test. In subgroup analysis, patients with a decreased peripheral white blood cell count from baseline were subdivided into groups based on the degree of change, using either quartiles or the median decrease. Here too the log-rank test was used to compare survival curves between these patient subgroups. All reported p values were 2-sided, and statistical significance was considered as p < 0.05. Results Descriptive Data Analysis In total, 86 patients diagnosed with GBM were included in this study. Their median age at diagnosis was 55 years; 32 patients (37%) were women and 54 (63%) were men. Nineteen patients (22%) were still alive at the time of data cutoff for analysis. The median OS for the entire group was 800 days, and the median PFS was 453 days. Baseline and 6-month peripheral blood counts are listed in Table 1. Mutation frequencies in the patient cohort are reported in Table 2. Univariate and Multivariate Analysis of Biomarker Impact on OS By univariate analysis, MGMT promoter methylation and IDH mutation were associated with better OS and PFS, consistent with the literature (Table 3 and Supplemental Table 1A). EGFR amplification, changes in neutro- phil counts, changes in peripheral red and white blood cell parameters, steroid use, and patient age at diagnosis were also associated with OS (Table 3). As predicted, patients receiving steroids during TMZ therapy had worse OS, with a 2-year survival rate of 51% compared with 71% in patients who were not treated with steroids (p = 0.0051). Additionally, changes in neutrophil counts differed between patients with and without steroid use (p = 0.0032), and patients receiving steroids had significantly greater neutrophil counts at 6 months of TMZ compared with those who never received steroids (p = 0.0031). Changes in neutrophil counts were not associated with OS in patients who never received steroids (p = 0.300). On multivariate analysis, MGMT promoter methylation, a decreased white blood cell count from baseline, and wild-type EGFR status (i.e., EGFR not amplified) were significantly associated with improved OS (Table 4). These associations remained significant on multivariate analysis that incorporated patient age and steroid use, despite our observation that steroid use was associated with increased white blood cells counts (p = 0.0585). On subgroup analysis, decreases in white blood cell counts remained associated with improved OS with hazard ratios of 0.410 (p = 0.053) and 0.109 (p = 0.066) in patients with and without steroid use. Association of MGMT and IDH With Alterations in Circulating Blood Cell Counts MGMT promoter methylation and IDH1 mutation are known to be robust prognostic markers for OS in GBM patients.1,18 Regarding the association of these genetic markers with alterations in circulating biomarkers that were significant on univariate analysis, we report that MGMT promoter methylation was not correlated with changes in peripheral red blood cell or white blood cell (p = 0.4492) counts. Similarly, IDH mutation was not correlated with changes in peripheral red blood cell (p = 0.2198) or white blood cell (p = 0.3447) counts during the course of treatment. Association of Changes in White Blood Cell Counts With OS The Kaplan-Meier estimated 2-year survival rate for patients with a decrease in white blood cell counts from baseline was 58% and that for patients with an increase relative to baseline was 25% (p = 0.0019; Fig. 1). Patients with decreases in white blood cell counts had a median OS of 850 days (95% CI 691–1097 days) compared with 627 days (95% CI 454–745 days) for those with increases in white blood cell counts from baseline. We found no significant differences in OS between subgroups of patients with decreased white blood cell counts during treatment, when the data were stratified by quartile or median decrease. Discussion We here demonstrate that treatment-associated myelosuppression, as manifested by a decrease in circulating white blood cell counts from baseline during adjuvant J Neurosurg October 14, 2016 3 E. J. Vaios et al. TABLE 1. Patient blood counts Hematology Baseline Platelets (×10 9 /L) Mean (SD) Range Red blood cells (×1012 /L) Mean (SD) Range White blood cells (×10 9 /L) Mean (SD) Range Eosinophils (×10 9 /L) Mean (SD) Range Basophils (×10 9 /L) Mean (SD) Range Lymphocytes (×10 9 /L) Mean (SD) Range Neutrophils (×10 9 /L) Mean (SD) Range Monocytes (×10 9 /L) Mean (SD) Range 6-mo adjuvant TMZ* Platelets (×10 9 /L) Mean (SD) Range Red blood cells (×1012 /L) Mean (SD) Range White blood cells (×10 9 /L) Mean (SD) Range Eosinophils (×10 9 /L) Mean (SD) Range Basophils (×10 9 /L) Mean (SD) Range Lymphocytes (×10 9 /L) Mean (SD) Range Neutrophils (×10 9 /L) Mean (SD) Range Monocytes (×10 9 /L) Mean (SD) Range TABLE 2. Summary of patient characteristics Value 279.28 (97.59) 95.00–589.00 4.28 (0.43) 3.27–5.16 8.69 (3.29) 3.27–20.00 0.11 (0.11) 0.00–0.58 0.03 (0.03) 0.00–0.20 1.67 (0.73) 0.52–3.93 6.21 (2.90) 2.16–15.43 0.46 (0.24) 0.14–1.51 188.74 (65.76) 80.00–400.00 4.13 (0.45) 3.24–5.29 5.68 (2.50) 3.00–14.10 0.12 (0.10) 0.00–0.45 0.02 (0.02) 0.00–0.15 0.99 (0.41) 0.35–2.10 3.97 (2.21) 0.02–11.30 0.40 (0.18) 0.02–1.02 * Blood counts obtained at the time of the 6th monthly cycle of TMZ treatment. 4 J Neurosurg October 14, 2016 Characteristic Sex Male Female Age at diagnosis, yrs Mean (SD) Range OS, days Mean (SD) Range PFS, days Mean (SD) Range Genetic mutation EGFR MGMT IDH Steroid use Deceased Value 54 (63%) 32 (37%) 55.37 (12.55) 18.00–80.00 915.09 (476.37) 324.00–2660.00 622.10 (487.02) 12.00–2660.00 37 (50.00%) 39 (54.17%) 6 (8.96%) 62 (72.09%) 67 (77.91%) Values represent n (%) unless otherwise indicated. TMZ therapy, might serve as a potential prognostic marker for clinical outcomes in patients with GBM. Our serial assessment of peripheral blood counts and additional laboratory and radiographic data found that a decrease in white blood cells from baseline during adjuvant TMZ therapy predicts significantly improved OS. Our institutional survival data for patients with decreases in white blood cells compares favorably with data from the original EORTC/NCIC trial, which reported a median survival of 14.6 months and a 2-year survival rate of 26.5% for patients receiving radiotherapy plus TMZ.16 The present findings are consistent with findings of previous studies demonstrating an association of MGMT promoter methylation and IDH1 mutation with improved clinical outcomes.3,5,6,17,21 Interestingly, we found that MGMT promoter methylation and IDH mutation did not correlate with changes in white blood cell counts, suggesting that these changes may serve as an independent prognostic factor. Despite these robust findings, our study is limited by its modest sample size and retrospective nature. Nevertheless, our exploratory analysis of hematological parameters identified that treatment-related changes in white blood cell counts are associated with OS, with decreases in white blood cell counts from baseline serving as a biomarker for improved OS. This association was maintained on multivariate analysis, even after controlling for other known prognostic variables, including age at the time of diagnosis, steroid use, EGFR amplification status, and MGMT promoter methylation status. On subgroup analysis, decreases in white blood cell counts from baseline maintained a strong association with improved OS regardless of whether a patient used steroids during TMZ therapy. Notably, an increase in white blood cell counts from baseline was also considered an important biomarker for worse OS, suggesting that changes in white blood cell counts play an Bone marrow response as a potential biomarker TABLE 3. Univariate analysis for OS Covariate Sex Male Female Age Genetic mutations EGFR MGMT IDH Percent change in Platelets Red blood cells White blood cells Eosinophils Basophils Lymphocytes Neutrophils Monocytes Steroids Used Not used TABLE 4. Multivariate analysis for OS HR 95% CI p Value* 0.685 — 1.021 0.418–1.124 — 1.000–1.042 0.1346 — 0.0470 1.818 0.353 0.100 1.074–3.078 0.203–0.615 0.014–0.726 0.0261 0.0002 0.0228 1.163 0.074 3.133 0.979 0.938 1.330 1.856 1.407 0.443–3.051 0.007–0.787 1.489–6.591 0.895–1.070 0.678–1.296 0.675–2.621 1.218–2.828 0.873–2.270 0.7588 0.0308 0.0026 0.6340 0.6973 0.4102 0.0040 0.1610 2.286 — 1.262–4.142 0.0064 — * Based on log-rank test. important biological role in the tumor microenvironment independent of chemotherapy. We were unable to identify an association between changes in the peripheral platelet count and clinical outcomes, as described by Williams et al.22 This finding suggests that decreases in white blood cell counts, rather than general “bone marrow suppression,” might be a predictor of improved OS. However, we did not observe a statistically significant association between white blood cell changes and PFS in our multivariate model, when controlling for steroid use and other markers that were significant on univariate analysis (Supplemental Data). Our findings, however, indicate that patient age and MGMT promoter methylation are important predictors of PFS. Changes in neutrophils were also significantly associated with OS and PFS, with neutrophil increases from baseline predicting worse outcomes. However, given the association of neutrophil counts with medications (e.g., steroid use), infections, and environmental factors, these findings remain more challenging to interpret. Neutrophil counts were likely confounded by steroid use, perhaps indicative of more aggressive tumor growth necessitating steroid administration for the management of edema. Consistent with the literature, our study found that steroid use was associated with elevated neutrophil counts and worse outcomes. In patients who did not receive steroids, we did not observe an association between neutrophil counts and OS. Therefore, this hematological parameter was excluded from our multivariate analysis. The observed relationship between a decrease in white blood cell counts and clinical outcomes is possibly due to higher in vivo drug concentrations of TMZ, resulting from differences in drug metabolism. TMZ activity is dose and schedule dependent and induces cytotoxicity primarily by Covariate HR 95% CI p Value* Age MGMT EGFR White blood cells Increase Decrease Red blood cells Increase Decrease Steroids Used Not used 1.064 0.106 1.779 1.033–1.096 0.043–0.259 0.978–3.247 <0.0001 <0.0001 0.0594 3.040 — 1.245–7.407 — 0.0147 — 1.065 — 0.578–1.961 — 0.8390 — 1.196 — 0.575–2.494 — 0.6317 — IDH was excluded due to inadequate sample size. * Based on log-rank test. causing O6-meG lesions, which deplete the repair enzyme MGMT, leading to double-strand DNA breaks and tumor cell apoptosis.7,13,14,25 TMZ is administered orally and has a half-life of 1.8 hours, reaching concentrations in the CSF that are 30%–40% of plasma concentrations. The clinical efficacy of TMZ and its active metabolite depends on MGMT activity; the integrity of the mismatch repair system, which recognizes O6-meG lesions in template DNA strands; and function of the base excision repair system, which corrects highly lethal N3-meA lesions via poly (ADP-ribose) polymerase.9,10,20 There are no guidelines for dose adjustment based on the tumor genetic signature or in the context of severe renal or hepatic impairment. Since Fig. 1. Kaplan-Meier analysis of OS in patients stratified by increase (dotted line) or decrease (solid line) in white blood cell counts from baseline. A significant survival benefit is noted for patients with a decrease in white blood cells relative to baseline (log-rank p = 0.0019). J Neurosurg October 14, 2016 5 E. J. Vaios et al. current dosing guidelines do not factor interpatient differences in resistance mechanisms or patient-specific drug metabolism, it is possible that some patients are treated subtherapeutically. Given the routine and reliable assessment of peripheral blood counts in GBM patients receiving conventional therapies, white blood cell counts could serve as a valuable biomarker of treatment response and for predicting clinical outcomes. Future prospective studies should address whether blood cell counts could serve as a correlate biomarker for in vivo TMZ levels and drug activity as well as a predictor of clinical outcomes, which in turn could help to optimize dosing and scheduling of chemotherapy by accounting for variability in drug metabolism. Conclusions We report a temporal relationship between changes in peripheral white blood cell counts during adjuvant TMZ treatment and clinical outcomes. Specifically, depression of white blood cell counts appears to be an independent prognostic factor and was associated with improved OS. This relationship may be a reflection of plasma TMZ levels and, in time, may serve as a surrogate marker of therapeutic efficacy. These findings warrant further investigation in prospective studies, including correlations with the degree of change in white blood cell counts and pharmacokinetics of TMZ in individual patients. It also remains unclear whether treatment-associated changes in white blood cell counts correlate with drug-induced antitumor activity or represent an independent factor of the altered local and systemic tumor environment. Acknowledgments This work was supported by the 2015 Neurosurgical Research and Education Foundation (NREF) Medical Student Summer Research Fellowship (Eugene J. Vaios) and the Harvard Medical School Scholars in Medicine Office (Eugene J. Vaios). Jorg Dietrich received support from the American Academy of Neurology, the American Cancer Society, and generous gifts from the family foundations of Bryan Lockwood, Ronald Tawil, and Sheila McPhee. References 1. Bastien JI, McNeill KA, Fine HA: Molecular characterizations of glioblastoma, targeted therapy, and clinical results to date. Cancer 121:502–516, 2015 2. Ellingson BM, Wen PY, van den Bent MJ, Cloughesy TF: Pros and cons of current brain tumor imaging. Neuro Oncol 16 (Suppl 7):vii2–vii11, 2014 3. Esteller M, Garcia-Foncillas J, Andion E, Goodman SN, Hidalgo OF, Vanaclocha V, et al: Inactivation of the DNArepair gene MGMT and the clinical response of gliomas to alkylating agents. N Engl J Med 343:1350–1354, 2000 4. Fujita T, Wakatabe Y, Matsumoto K, Tabata K, Yoshida K, Iwamura M: Leukopenia as a biomarker of sunitinib outcome in advanced renal cell carcinoma. Anticancer Res 34:3781– 3787, 2014 5. Hegi ME, Diserens AC, Gorlia T, Hamou MF, de Tribolet N, Weller M, et al: MGMT gene silencing and benefit from temozolomide in glioblastoma. N Engl J Med 352:997–1003, 2005 6. Idbaih A, Omuro A, Ducray F, Hoang-Xuan K: Molecular 6 J Neurosurg October 14, 2016 genetic markers as predictors of response to chemotherapy in gliomas. Curr Opin Oncol 19:606–611, 2007 7. Kanzawa T, Bedwell J, Kondo Y, Kondo S, Germano IM: Inhibition of DNA repair for sensitizing resistant glioma cells to temozolomide. J Neurosurg 99:1047–1052, 2003 8. Lonardi S, Tosoni A, Brandes AA: Adjuvant chemotherapy in the treatment of high grade gliomas. Cancer Treat Rev 31:79–89, 2005 9. Marchesi F, Turriziani M, Tortorelli G, Avvisati G, Torino F, De Vecchis L: Triazene compounds: mechanism of action and related DNA repair systems. Pharmacol Res 56:275– 287, 2007 10. Miknyoczki S, Chang H, Grobelny J, Pritchard S, Worrell C, McGann N, et al: The selective poly(ADP-ribose) polymerase-1(2) inhibitor, CEP-8983, increases the sensitivity of chemoresistant tumor cells to temozolomide and irinotecan but does not potentiate myelotoxicity. Mol Cancer Ther 6:2290–2302, 2007 11. Motzer RJ, Hutson TE, Tomczak P, Michaelson MD, Bukowski RM, Oudard S, et al: Overall survival and updated results for sunitinib compared with interferon alfa in patients with metastatic renal cell carcinoma. J Clin Oncol 27:3584– 3590, 2009 12. Motzer RJ, Hutson TE, Tomczak P, Michaelson MD, Bukowski RM, Rixe O, et al: Sunitinib versus interferon alfa in metastatic renal-cell carcinoma. N Engl J Med 356:115– 124, 2007 13. Newlands ES, Stevens MF, Wedge SR, Wheelhouse RT, Brock C: Temozolomide: a review of its discovery, chemical properties, pre-clinical development and clinical trials. Cancer Treat Rev 23:35–61, 1997 14. Roos WP, Batista LF, Naumann SC, Wick W, Weller M, Menck CF, et al: Apoptosis in malignant glioma cells triggered by the temozolomide-induced DNA lesion O6methylguanine. Oncogene 26:186–197, 2007 15. Stupp R, Hegi ME, Mason WP, van den Bent MJ, Taphoorn MJ, Janzer RC, et al: Effects of radiotherapy with concomitant and adjuvant temozolomide versus radiotherapy alone on survival in glioblastoma in a randomised phase III study: 5-year analysis of the EORTC-NCIC trial. Lancet Oncol 10:459–466, 2009 16. Stupp R, Mason WP, van den Bent MJ, Weller M, Fisher B, Taphoorn MJ, et al: Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 352:987–996, 2005 17. Tabatabai G, Stupp R, van den Bent MJ, Hegi ME, Tonn JC, Wick W, et al: Molecular diagnostics of gliomas: the clinical perspective. Acta Neuropathol 120:585–592, 2010 18. Tanaka S, Louis DN, Curry WT, Batchelor TT, Dietrich J: Diagnostic and therapeutic avenues for glioblastoma: no longer a dead end? Nat Rev Clin Oncol 10:14–26, 2013 19. Thomas AA, Brennan CW, DeAngelis LM, Omuro AM: Emerging therapies for glioblastoma. JAMA Neurol 71:1437–1444, 2014 20. Villano JL, Seery TE, Bressler LR: Temozolomide in malignant gliomas: current use and future targets. Cancer Chemother Pharmacol 64:647–655, 2009 21. von Deimling A, Korshunov A, Hartmann C: The next generation of glioma biomarkers: MGMT methylation, BRAF fusions and IDH1 mutations. Brain Pathol 21:74–87, 2011 22. Williams M, Liu ZW, Woolf D, Hargreaves S, Michalarea V, Menashy R, et al: Change in platelet levels during radiotherapy with concurrent and adjuvant temozolomide for the treatment of glioblastoma: a novel prognostic factor for survival. J Cancer Res Clin Oncol 138:1683–1688, 2012 23. Wilson TA, Karajannis MA, Harter DH: Glioblastoma multiforme: State of the art and future therapeutics. Surg Neurol Int 5:64, 2014 Bone marrow response as a potential biomarker 24. Zhu AX, Duda DG, Ancukiewicz M, di Tomaso E, Clark JW, Miksad R, et al: Exploratory analysis of early toxicity of sunitinib in advanced hepatocellular carcinoma patients: kinetics and potential biomarker value. Clin Cancer Res 17:918–927, 2011 25. Ziegler DS, Kung AL, Kieran MW: Anti-apoptosis mechanisms in malignant gliomas. J Clin Oncol 26:493– 500, 2008 Disclosures The authors report no conflict of interest concerning the materials or methods used in this study or the findings specified in this paper. Author Contributions Conception and design: Vaios, Nahed, Dietrich. Acquisition of data: Vaios. Analysis and interpretation of data: Vaios, Nahed, Muzikansky, Dietrich. Drafting the article: Vaios. Critically revising the article: Vaios, Nahed, Fathi, Dietrich. Reviewed submitted version of manuscript: all authors. Approved the final version of the manuscript on behalf of all authors: Vaios. Statistical analysis: Vaios, Muzikansky. Administrative/technical/material support: Nahed, Dietrich. Study supervision: Nahed, Dietrich. Supplemental Information Online-Only Content Supplemental material is available with the online version of the article. Supplemental Tables 1A and B. http://thejns.org/doi/ suppl/0.3171/2016.7.JNS16609. Correspondence Eugene John Vaios, Vanderbilt Hall Box 099, 107 Ave. Louis Pasteur, Boston, MA 02115. email: [email protected]. edu. J Neurosurg October 14, 2016 7