Journal Logo

Research Article: Observational Study

Survival analysis of patients with glioblastoma treated by long-term administration of temozolomide

Quan, Rencui PhDa; Zhang, Huaqing PhDb; Li, Zihuang MDa,∗; Li, Xianming MDa,∗

Section Editor(s): Ding., Jianxun

Author Information
doi: 10.1097/MD.0000000000018591
  • Open


1 Introduction

Glioblastoma is the most common and aggressive form of primary brain tumors with a median survival of fewer than 2 years.[1–3] According to the Central Brain Tumor Registry of the United States Statistical Report, glioma accounts for about 27% of all tumors in the central nervous system and 80% of malignant tumors. Among the primary malignant tumors of the central nervous system, the incidence of glioblastoma is 3.20 per 100,000 populations uppermost, followed by the diffuse astrocytoma with the incidence of 0.53 per 100,000 populations. The incidence increases with age, among the predilection age ranging 75 to 84 years and the median age of 64 years.[4] The incidence of glioblastoma in the People's Republic of China is 1 to 4 per 100,000 populations.[5] The safe and feasible extent of resection followed by radiotherapy with concomitant and adjuvant temozolomide (TMZ, 6 cycles) is the landmark treatment protocol for the ones with newly diagnosed glioblastoma.[1,3]

TMZ converts to the active metabolite 5-(3-methyltriazen-1-yl) imidazole-4-carboxamide by nonenzymatic chemical conversion, which causes cell cycle arrest and ultimately cell death by breaking the DNA double strands.[6,7] TMZ is well tolerated even in elderly patients with glioblastoma.[8] The fatality rate is less than 10% due to toxicity.[3,8] The common toxicities include mild hematotoxicity, fatigue, and nausea.

The mainstay of TMZ in glioblastoma has been consistently admitted, and 6 cycles of adjuvant TMZ is pervasive treatment. However, the most suitable cycles of TMZ still remain controversial. Some retrospective studies and a meta-analysis conferred the prolonged progression-free survival (PFS) and overall survival (OS) of patients with glioblastoma treated with long-term administration of more than 6 cycles of TMZ.[9–14] The OS of patients receiving fewer than 7 cycles of TMZ ranging from 8 to 20 months compared with that of patients receiving at least 7 cycles ranging from 21 to 30 months, while PFS ranged from 4 to 18 months compared with 17 to 28 months, respectively, in the aforementioned 2 groups.[9,14] However, some other studies queried the improvement in extended adjuvant TMZ in glioblastoma.[15,16] They considered that neither OS nor PFS of patients with newly diagnosed glioblastoma was superior, albeit they received more than 6 cycles of TMZ. Moreover, a secondary analysis of EORTC and NRG Oncology/RTOG deemed that the number of TMZ cycles contributed to somewhat improved PFS but not OS.[17,18]

In the United States, patients commonly receive up to 12 or even more cycles of maintenance TMZ following chemoradiation, but the role of extending TMZ use is still controversial.[19] It is worth mentioning that the extended administration of TMZ has been universally accepted as safe and feasible.[18] In China, glioma occurs at a younger age with a median age less than 50 years according to the large sample analysis, and the treatment of TMZ in glioblastoma is not standardized.[20,21]

Therefore, this analysis was performed to investigate whether the long-term administration of TMZ claimed a survival advantage for patients with glioblastoma in China.

2 Patients and methods

2.1 Patients

This is a retrospective study and the ethical approval was not necessary. Patients firstly diagnosed with World Health Organization grade IV glioblastoma[22] between August 1, 2008, and August 1, 2016, were selected. Only patients initially diagnosed with glioblastoma received radiation and without other kinds of malignant tumors were eligible. All patients were followed up telephonically until they died or lost to follow-up, or till August 1, 2018. The characteristics of these patients were reviewed from the records in Shenzhen People's Hospital. All patients received intracranial surgery. The extent of resection was determined by the magnetic resonance imaging (MRI) scans after the surgery and operation records. The conventional, fractionated, and conformal radiations at a daily dose of 2 Gy mostly began within 6 weeks after surgery. Treatment was carried out 5 days a week within 6 weeks to a total dose of 60 Gy with concomitant TMZ [patients received a daily dose of 75 mg/(m2 day) for 7 days a week in the entire duration of radiotherapy]. After 1 month's interval from the end of concomitant TMZ, the patients began adjuvant TMZ of the standard regimen [150–200 mg/(m2 day) for 5 consecutive days in each 28-day cycle] until disease progression or were unwilling to continued chemotherapy. The patients received TMZ at a starting dose of 150 mg/m2 for 5 consecutive days of a 28-day cycle, and TMZ was increased for subsequent cycles to 200 mg/m2 if no unacceptable treatment-related adverse events were noted. The dose, schedule, and maintenance cycles of TMZ in patients with glioblastoma were ensured from the records and their prescribing physicians. Response was evaluated by computed tomography or MRI using the standardized McDonald criteria. The patients were identified into 2 groups according to their cycles of adjuvant of TMZ.

2.2 Statistical analysis

This study aimed to evaluate the OS and PFS of patients in the 2 groups. A propensity-matched analysis was performed between the groups to balance the 4 known independent prognostic factors listed in Table 1. To balance selection biases and other confounding items, we used 1:1 matching on propensity scores and logistic regression and nearest neighbor algorithm within caliper width of 0.01. The absolute standardized differences were used to assess the imbalance of each of the variables. Briefly, the software automatically calculated the propensity score of each value and completed the matching by logistic regression analysis between 2 groups.[23–26] The Student t test and normality test were performed to evaluate quantitative variables. The chi-square and Fisher exact tests were performed for analyzing nominal variables. All statistical tests were 2 tailed. OS was calculated from the surgery date to death or loss of follow-up. PFS was defined from the day of surgery to progression, death, or deadline of follow-up. OS and PFS curves were depicted by the Kaplan–Meier method. Univariate and multivariable analyses were compared using the log-rank test and Cox proportional hazards model. A value of P < .05 was defined as statistically significant. All statistical analyses were performed with SPSS Statistics version 22.0 (IBM, NY).

Table 1
Table 1:
Characteristics of patients.

3 Results

3.1 Patient characteristics

Altogether 75 patients with initially diagnosed glioblastoma were identified. Their characteristics are summarized in Table 1. The patients were divided into 2 groups according to different cycles of TMZ: fewer than 7 cycles (group A = TMZ for less than or equal to 6 cycles, n = 55) and beyond 6 cycles (group B = TMZ more than 6 cycles, n = 20). The median age was 47 years (range 18–70) and 45 years (range 29–66), respectively (P = .279). Table 1 shows that both groups were balanced for age (P = .143), sex (P = .834), type of surgery (P = .600), but not methylguanine methyltransferase (MGMT) status (P = .001). After matching by propensity analysis, the MGMT promoter methylation status of tumor was balanced (P = .558). In the case of more than 6 cycles of TMZ, patients received median TMZ cycles of 12 (range 7–30). In the case of 6 cycles of TMZ, 43 patients (78.2%) experienced gross total resection; 15 patients (85.0%) received more than 6 cycles of TMZ.

3.2 OS and PFS

The last time of follow-up was on August 1, 2018. Forty-three patients (78.2%) demised and 47 patients (85.5%) relapsed in group A. Ten patients (50.0%) demised and 15 patients (75.0%) relapsed in group B. The median OS was 22.0 months (95% CI 18.8–25.2), and the median PFS was 14.8 months (95% CI 11.5–18.1) in the overall population. The median OS was 20.6 months (95% CI 17.9–23.2) in group A compared with 47.0 months (95% CI 20.0–24.5) in group B (P = .014). The median PFS was 14.2 months (95% CI 11.8–16.6) in group A compared with 17.0 months (95% CI 10.1–24.5) in group B (P = .133). After propensity analysis, the median OS was 20.0 months (P = .024) and the median PFS was 13.0 months (P = .086) in groups A and B. These data demonstrated that the superior OS derived from the different therapy rather than the number of people (Fig. 1).

Figure 1
Figure 1:
Association of TMZ exposure with the outcome. Overall survival and progression-free survival in patients with initially diagnosed glioblastoma by the number of TMZ cycles (A and B) Before matching: ≤6 cycles (n = 55) and >6 cycles (n = 20). (C and D) After matching: ≤6 cycles (n = 20) and >6 cycles (n = 20). TMZ = temozolomide.

Univariate analysis was performed to evaluate the clinical factors including age, sex, type of surgery, MGMT status, and TMZ cycles. The model identified the MGMT promoter methylation status, and the number of TMZ cycles was associated with OS (Table 2). The prolonged number of TMZ cycles was a positive factor for OS (HR = 0.43, 95% CI 0.22–0.86, P = .017). The MGMT unmethylated status was a significant risk factor for disease progression and OS (HR = 2.84, 95% CI 1.30–6.20, P = .009 and HR = 4.65, 95% CI 1.57–13.76, P = .005, respectively). The prolonged cycles of TMZ, whether before or after matching, were favorable factors for OS but not PFS (Table 2). In particular, the multivariate analysis also supported that the methylation of MGMT status (HR = 4.17, 95% CI 1.34–12.90, P = .014) improved the OS of patients. The multivariate analysis revealed that extended TMZ treatment for more than 6 cycles (HR = 0.67, 95% CI 0.32–1.39, P = .277) might not be an independent factor associated with survival when adjusted for known prognostic factors in glioblastoma (age, sex, MGMT promoter methylated, and type of surgery).

Table 2
Table 2:
Univariate analysis of PFS and OS in 75 patients.

4 Discussion

The number of adjuvant TMZ cycles in patients with glioblastoma was generally recognized as 6, followed by RT/TMZ–TMZ treatment since 2005.[1] Some researches attempted to improve the benefits by prolonging the adjuvant TMZ cycles or using dose-dense TMZ therapy.[2,27] Although the role of prolonging the adjuvant TMZ cycles still remains controversial, it has become a common practice not only in the United States but also in China, especially among the population well tolerant to TMZ.

In the present study, 75 Chinese patients diagnosed with glioblastoma who received adjuvant TMZ therapy were identified. Patient characteristics in both groups were well balanced (Table 1). The patients receiving more than 6 cycles of TMZ seemed to be superior survivors with median OS reaching to 47.0 months (95% CI 20.0–24.5, P = .014). The prolonged OS was worth noting. Gloria B[11] also supported that patients in whom adjuvant chemotherapy was stopped at cycle 6 experienced a median survival of 16.5 months, whereas, those who received more than 6 cycles survived for 24.6 months (P = .031). Extended adjuvant therapy was not associated with increased toxicity. Prolonged administration of TMZ after radio-chemotherapy in patients with GBM was feasible and seemed to be well-tolerated. There is a growing number of data, including the present study, which suggests a benefit of this strategy on PFS and OS.[12,14] The extended TMZ treatment failed to show the survival improvement in multivariate analyses may associate the limit object and observe time.

The univariate and multivariate analyses of PFS and OS showed that MGMT promoter methylation was a strong and predictive biomarker when patients diagnosed with glioblastoma received TMZ treatment. In patients with a methylated MGMT promoter status, the prolonged administration of TMZ was associated with increasing PFS and OS (Tables 2 and 3). This was similar to the preliminary analysis.[15,16] Though the present study did not demonstrate that the PFS of patients in group B was superior to that of in group A, which was contrary to other studies.[17,18] We had observed an interesting thing some patients obtained survival benefit from prolonged TMZ treatment even they suffered from progression during the treatment. Maybe this could help to understand the OS improvement. The superior OS might be also due to the median age of participants in this study, who were younger than those in Gramatzki's study.[15] Valduvieco demonstrated the impact of delay in initiation of RT on the survival of patients with glioblastoma, indicating that the initiation of radiation therapy within 6 weeks after surgery was a favorable and independent prognostic factor for survival.[28] Most of the patients in this study received radiation therapy within 6 weeks after the surgery. However, the proportion of MGMT status was significantly lower than that in Gramatzki's study, which was 40.0% vs 56.7%, leading to inferior PFS.

Table 3
Table 3:
Multivariate analysis of PFS and OS.

Nowadays the effect of extended maintenance of TMZ varies. Some studies hypothesize that the extended treatment of TMZ may confer resistance to tumor cells and be detrimental. Prolonged exposure to TMZ therapy results in mutational changes, which may portend resistance to oral alkylating medicine and aggravate the cumulative toxicities, whereas no obvious mechanism of resistance was identified in glioblastoma cells.[29,30] Hence, the potential possibility that prolonged administration of TMZ accelerated the development of alkylating drug resistance and generated counterproductive result was inconclusive but the patients benefited from treatment was obvious. Parallelly, recent studies had showed that maintenance chemotherapy with pemetrexed chemotherapy or maintenance treatment with fluoropyrimidine plus bevacizumab after first-line therapy demonstrated prolonged survival advantages only in patients with non-small cell lung cancer and those with metastatic colorectal cancer.[31–33] Furthermore, a phase II rescue study suggested that conventional adjuvant TMZ doses or a treatment-free interval might benefit from salvage TMZ therapy.[34]

Since this was a retrospective study, some limitations were unavoidable including nonrandom nature, small size of population, and some missing data of characteristics including the MGMT status of patients, which is an important prognostic factor of survival. However, the study still suggested physicians to consider prescribing the long-term administration of TMZ for Chinese patients.

Author contributions

Conceptualization: ZiHuang Li, Xianming Li.

Data curation: Rencui Quan, Huaqing Zhang.

Formal analysis: Rencui Quan, Xianming Li.

Methodology: Rencui Quan.

Supervision: Xianming Li.

Writing – original draft: Rencui Quan.

Rencui Quan orcid: 0000-0002-5953-8142.


[1]. Stupp R, Mason WP, van den Bent MJ, et al. Radiotherapy plus concomitant and adjuvant temozolomide for glioblastoma. N Engl J Med 2005;352:987–96.
[2]. Clarke JL, Iwamoto FM, Sul J, et al. Randomized phase II trial of chemoradiotherapy followed by either dose-dense or metronomic temozolomide for newly diagnosed glioblastoma. J Clin Oncol 2009;27:3861–7.
[3]. Stupp R, Hegi ME, Mason WP, 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 2009;10:459–66.
[4]. Ostrom QT, Gittleman H, Liao P, et al. CBTRUS Statistical Report: primary brain and other central nervous system tumors diagnosed in the United States in 2010–2014. Neuro-Oncology 2017;19:v1–88.
[5]. Guidelines for the diagnosis and treatment of glioma of central nervous system in China (2015). Chin Med J 2016;485–509.
[6]. Thomas A, Tanaka M, Trepel J, et al. Temozolomide in the era of precision medicine. Cancer Res 2017;77:823–6.
[7]. Lee CY. Strategies of temozolomide in future glioblastoma treatment. Onco Targets Ther 2017;10:265–70.
[8]. Behm T, Horowski A, Schneider S, et al. Concomitant and adjuvant temozolomide of newly diagnosed glioblastoma in elderly patients. Clin Neurol Neurosurg 2013;115:2142–6.
[9]. Seiz M, Krafft U, Freyschlag CF, et al. Long-term adjuvant administration of temozolomide in patients with glioblastoma multiforme: experience of a single institution. J Cancer Res Clin 2010;136:1691–5.
[10]. Freyschlag CF, Smolczyk DR, Janzen E, et al. Prolonged administration of temozolomide in adult patients with anaplastic glioma. Anticancer Res 2011;31:3873.
[11]. Roldán Urgoiti GB, Singh AD, Easaw JC. Extended adjuvant temozolomide for treatment of newly diagnosed glioblastoma multiforme. J Neuro-Oncol 2012;108:173–7.
[12]. Darlix A, Baumann C, Lorgis V, et al. Prolonged administration of adjuvant temozolomide improves survival in adult patients with glioblastoma. Anticancer Res 2013;33:3467–74.
[13]. Barbagallo GMV, Paratore S, Caltabiano R, et al. Long-term therapy with temozolomide is a feasible option for newly diagnosed glioblastoma: a single-institution experience with as many as 101 temozolomide cycles. Neurosurg Focus 2014;37:E4.
[14]. Xu W, Li T, Gao L, et al. Efficacy and safety of long-term therapy for high-grade glioma with temozolomide: a meta-analysis. Oncotarget 2017;8:51758–65.
[15]. Gramatzki D, Kickingereder P, Hentschel B, et al. Limited role for extended maintenance temozolomide for newly diagnosed glioblastoma. Neurology 2017;88:1422–30.
[16]. Gilbert MR, Wang M, Aldape KD, et al. Dose-dense temozolomide for newly diagnosed glioblastoma: a randomized phase III clinical trial. J Clin Oncol 2013;31:4085–91.
[17]. Blumenthal DT, Gorlia T, Gilbert MR, et al. Is more better? The impact of extended adjuvant temozolomide in newly diagnosed glioblastoma: a secondary analysis of EORTC and NRG Oncology/RTOG. Neuro Oncol 2017;19:1119–26.
[18]. Hau P, Koch D, Hundsberger T, et al. Safety and feasibility of long-term temozolomide treatment in patients with high-grade glioma. Neurology 2007;68:688–90.
[19]. Nabors LB, Portnow J, Ammirati M, et al. Central nervous system cancers, Version 1.2015. J Natl Compr Canc Netw 2015;13:1191–202.
[20]. Yang P, Wang Y, Peng X, et al. Management and survival rates in patients with glioma in China (2004–2010): a retrospective study from a single-institution. J Neurooncol 2013;113:259–66.
[21]. Chang L, Su J, Jia X, et al. Treating malignant glioma in Chinese patients: update on temozolomide. Onco Targets Ther 2014;7:235–44.
[22]. Louis DN, Ohgaki H, Wiestler OD, et al. The 2007 WHO classification of tumours of the central nervous system. Acta Neuropathol 2007;114:97–109.
[23]. Robins JM, Mark SD, Newey WK. Estimating exposure effects by modelling the expectation of exposure conditional on confounders. Biometrics 1992;48:479–95.
[24]. Cao C, Zhu Z, Yan TD, et al. Video-assisted thoracic surgery versus open thoracotomy for non-small-cell lung cancer: a propensity score analysis based on a multi-institutional registry. Eur J Cardio-Thorac 2013;44:849–54.
[25]. D’Agostino RB Jr. Propensity score methods for bias reduction in the comparison of a treatment to a non-randomized control group. Stat Med 1998;17:2265–81.
[26]. Austin PC. A critical appraisal of propensity-score matching in the medical literature between 1996 and 2003. Stat Med 2008;27:2037–49.
[27]. Weiler M, Hartmann C, Wiewrodt D, et al. Chemoradiotherapy of newly diagnosed glioblastoma with intensified temozolomide. Int J Radiat Oncol Biol Phys 2010;77:670–6.
[28]. Valduvieco I, Verger E, Bruna J, et al. Impact of radiotherapy delay on survival in glioblastoma. Clin Transl Oncol 2013;15:278–82.
[29]. Happold C, Roth P, Wick W, et al. Distinct molecular mechanisms of acquired resistance to temozolomide in glioblastoma cells. J Neurochem 2012;122:444–55.
[30]. Johannessen TC, Bjerkvig R. Molecular mechanisms of temozolomide resistance in glioblastoma multiforme. Expert Rev Anticancer Ther 2012;12:635–42.
[31]. Ciuleanu T, Brodowicz T, Zielinski C, et al. Maintenance pemetrexed plus best supportive care versus placebo plus best supportive care for non-small-cell lung cancer: a randomised, double-blind, phase 3 study. Lancet 2009;374:1432–40.
[32]. Simkens LH, van Tinteren H, May A, et al. Maintenance treatment with capecitabine and bevacizumab in metastatic colorectal cancer (CAIRO3): a phase 3 randomised controlled trial of the Dutch Colorectal Cancer Group. Lancet 2015;385:1843–52.
[33]. Hegewisch-Becker S, Graeven U, Lerchenmuller CA, et al. Maintenance strategies after first-line oxaliplatin plus fluoropyrimidine plus bevacizumab for patients with metastatic colorectal cancer (AIO 0207): a randomised, non-inferiority, open-label, phase 3 trial. Lancet Oncol 2015;16:1355–69.
[34]. Perry JR, Belanger K, Mason WP, et al. Phase II trial of continuous dose-intense temozolomide in recurrent malignant glioma: RESCUE study. J Clin Oncol 2010;28:2051–7.

glioblastoma; long-term administration; overall survival; progression-free survival; temozolomide

Copyright © 2020 the Author(s). Published by Wolters Kluwer Health, Inc.