Can Extended Chemotherapy Improve Glioblastoma Outcomes? A Retrospective Analysis of Survival in Real-World Patients

A combined analysis of two prospective randomised studies exploring the impact of extended post-radiation temozolomide on survival outcomes in newly diagnosed glioblastoma

PURPOSE

The optimal duration of post-radiation temozolomide in newly diagnosed glioblastoma remains unclear, with no published phase III randomised trials. Standard-of-care stipulates 6 months. However, in routine care, it is often extended to 12 months, despite lacking robust supporting data.

METHODS

GEINO14-01 (Spain) and EX-TEM (Australia) studies enrolled glioblastoma patients without progression at the end of 6 months post-radiation temozolomide. Participants were randomised 1:1 to six additional months of temozolomide or observation. Primary endpoint was 6-month progression free survival from date of randomisation (6mPFS). Secondary endpoints included overall survival (OS) and toxicity. 204 patients were required to detect an improvement in 6mPFS from 50 to 60% (80% power). Neither study recruited sufficient patients. We performed a combined analysis of individual patient data.

RESULTS

205 patients were recruited: 159 in GEINO14-01 (2014-2018) and 46 in EX-TEM (2019-2022). Median follow-up was 20.0 and 14.5 months. Baseline characteristics were balanced. There was no significant improvement in 6mPFS (57.2% vs 64.0%, OR0.75, p = 0.4), nor across any subgroups, including MGMT methylated; PFS (HR0.92, p = 0.59, median 7.8 vs 9.7 months); or OS (HR1.03, p = 0.87, median 20.1 vs 19.4 months). During treatment extension, 64% experienced any grade adverse event, mainly fatigue and gastrointestinal (both 54%). Only a minority required treatment changes: 4.5% dose delay, 7.5% dose reduction, 1.5% temozolomide discontinuation.

CONCLUSION

For glioblastoma patients, extending post-radiation temozolomide from 6 to 12 months is well tolerated but does not improve 6mPFS. We could not identify any subset that benefitted from extended treatment. Six months should remain standard-of-care.

Updated systematic review and meta-analysis of extended adjuvant temozolomide in patients with newly diagnosed glioblastoma

Background

Contemporary standard-of-care for newly diagnosed glioblastoma (GBM) is maximal safe resection followed by postoperative focal conformal radiotherapy (RT) plus concurrent temozolomide (TMZ) and 6-cycles of adjuvant TMZ (Stupp regimen). However, many patients continue to receive extended adjuvant TMZ (beyond 6-cycles) without solid scientific evidence. This review pools data from nonrandomized studies and randomized controlled trials (RCTs) comparing extended adjuvant TMZ (>6-cycles) to standard adjuvant TMZ (6-cycles) in patients with newly diagnosed GBM for updated evidence-synthesis.

Methods

This systematic review and meta-analysis was carried out in accordance with the Cochrane methodology including quality assessment of primary studies. Primary outcome of interest was comparative efficacy defined as progression-free survival (PFS) and overall survival (OS). Hazard ratios (HRs) for PFS and OS with corresponding 95% confidence interval (CIs) were extracted/computed from individual primary studies and pooled using random-effects model. Any p-value <0.05 was considered statistically significant.

Results

Evidence-synthesis was based on pooling of data from 2578 patients enrolled in 16 nonrandomized comparative studies and 5 RCTs. Overall, extended adjuvant TMZ was associated with statistically significant reduction in the risk of progression (HR = 0.72, 95%CI: 0.60-0.87; p = 0.007) and death (HR = 0.71, 95%CI: 0.57-0.90; p = 0.004) compared to standard adjuvant TMZ. However, on subgroup analysis, survival benefit of extended adjuvant TMZ was limited to data synthesized from retrospective nonrandomized comparative studies with no statistically significant difference in outcomes seen after pooling of data from RCTs only.

Conclusion

Apparent survival benefit of extended adjuvant TMZ in newly diagnosed GBM is largely driven by nonrandomized comparative studies with high inherent potential for multiple biases.