RTOG 0825

Background

RTOG 0825 was a phase III randomized double-blind placebo-controlled trial evaluating bevacizumab (anti-VEGF monoclonal antibody) added to the standard Stupp regimen (temozolomide + radiation, then adjuvant TMZ maintenance) in newly diagnosed GBM. Bevacizumab had shown activity in recurrent GBM (BRAIN trial) and was FDA-approved in 2009 for recurrent GBM. RTOG 0825 (conducted in parallel with the AVAGLIO trial) asked whether first-line bevacizumab could improve PFS and OS in unselected newly diagnosed GBM.

Interventions and follow up

Arm A: Standard Stupp + placebo — radiation 60 Gy + concurrent TMZ 75 mg/m² → adjuvant TMZ × 6–12 cycles + placebo q2w
Arm B: Standard Stupp + bevacizumab 10 mg/kg q2w (concurrent and maintenance)
Primary endpoint: OS and PFS (co-primary)
mFollow-up: 20.5 months

Results

mOS: 15.7 (bev) vs 16.1 months (placebo) (HR 1.13, 95% CI 0.93–1.37, P=.21 — not significant)
mPFS: 10.7 vs 7.3 months (HR 0.79, 95% CI 0.66–0.94, P=.007 — significant)
ORR: 67.5% vs 39.5%
MGMT-methylated mOS: 23.2 vs 26.1 months (NS); neurocognitive decline and symptom burden higher with bevacizumab over time

Adverse events

Overall: Grade ≥3 AEs 67% vs 57%; Grade ≥3 fatigue 20% vs 15%
Vascular: Grade ≥3 hypertension 11.2% vs 2.5%; Grade ≥3 thromboembolic events 8.2% vs 4.7%; intracranial hemorrhage and wound-healing complications increased with bevacizumab
Other: bevacizumab-related proteinuria, rare GI perforation, wound dehiscence; neurocognitive decline accelerated in the bevacizumab arm at late timepoints

Conclusions

Bevacizumab added to standard first-line chemoradiation significantly improved PFS (10.7 vs 7.3 months) but did not improve OS in newly diagnosed GBM, and was associated with worse neurocognitive function and symptom burden over time, collectively demonstrating that first-line bevacizumab does not benefit unselected GBM patients and should not be incorporated into standard frontline treatment.

Key Limitations

Pseudoprogression may have inflated PFS benefit in the placebo arm and deflated it in the bev arm (MRI changes with bev mask true progression); neurocognitive decline in the bevacizumab arm undermines the rationale for PFS as a patient-centered benefit endpoint; no biomarker selection for VEGF-responsive tumors; the parallel AVAGLIO trial showed similar results (improved PFS, no OS benefit, but better QoL — contradicting RTOG 0825 QoL findings); no biomarker (VEGF, VEGFR, MGMT) predicted bevacizumab benefit; bevacizumab continuation beyond progression confounds second-line therapy options; trial predates TTFields (EF-14) — the treatment landscape has changed.

Clinical Context

RTOG 0825, together with the European AVAGLIO trial (Chinot OL, NEJM 2014), established that first-line bevacizumab does not improve OS in unselected newly diagnosed GBM — a major negative result that shaped subsequent GBM treatment strategy. Bevacizumab remains FDA-approved for recurrent GBM but is not standard in the first-line setting; ASCO-SNO and EANO guidance do not recommend first-line bevacizumab for newly diagnosed GBM outside clinical trials. Bevacizumab is, however, widely used for symptom management (radiation necrosis, cerebral edema reduction) and recurrent disease. The discordant QoL findings between RTOG 0825 (worse with bev) and AVAGLIO (better with bev) highlight methodological challenges in neuro-oncology trials. Steroid-sparing with bevacizumab remains clinically useful for peritumoral edema.

References