Optimal treatment of recurrent glioblastoma multiforme (rGBM) in elderly and/or frail patients remains virtually unexplored, the best supportive care (BSC) only is routinely administered due to the fatal prognosis. We evaluated the impact of different treatment methods on post-progression survival (PPS) and overall survival (OS) of such patients. Data from 98 elderly and/or frail rGBM patients, treated initially with 1-week or 3-week radiotherapy (RT) within the phase III IAEA study (2010-2013), were analyzed. KPS at relapse and salvage treatment methods were recorded. Kaplan-Meier method was used to estimate PPS and OS for different treatment modalities. Eighty-four patients experienced recurrence: 47 (56%) received BSC, 21 (25%)-chemotherapy (CHT), 8 (9.5%)-surgery, 3 (3.5%)-RT, for 5 (6%) the data was unavailable. Median OS from randomization for all 84 patients was 35 weeks: 55 versus 30 weeks for any treatment versus BSC, p < 0.0001. Median PPS was 15 weeks: 23 weeks with any treatment versus 9 weeks with BSC, p < 0.0001. For local treatment (surgery and/or RT) median PPS was 51 versus 21 weeks for CHT, p = 0.36. In patients with poor KPS (≤60) at relapse median PPS was 9 weeks with BSC versus 21 weeks with any treatment, p = 0.014. In poor KPS patients median PPS for local treatment was 14 weeks versus 21 weeks with CHT, p = 0.88. An active therapeutic approach may be beneficial for selected elderly and/or frail rGBM patients. Poor KPS patients may also benefit from active treatment, but there is no benefit of local treatment over CHT.
Keywords: Elderly patients; Frail patients; Recurrent glioblastoma multiforme; Salvage therapy.