Nivolumab fails to deliver OS benefit for recurrent glioblastoma

The PD-1 inhibitor nivolumab did not improve overall survival (OS) in patients with recurrent glioblastoma compared with bevacizumab, the phase III CheckMate 143* trial has shown, highlighting the need for new therapeutic interventions for this disease.

Glioblastoma has a poor prognosis, with nearly all patients experiencing recurrence following standard-of-care treatment (ie, resection, radiotherapy, and temozolomide). [Lancet Oncol 2009;10:459-466; JAMA 2013;310:1842-1850; Neuro Oncol 2013;15:4-27] “Treatment options at recurrence are limited, and no therapy has prolonged OS in this setting,” said the researchers.

Researchers randomized 369 individuals 1:1 to receive IV nivolumab 3 mg/kg or bevacizumab 10 mg/kg Q2W. Treatment ran until confirmed disease progression, onset of toxic effects, or death. [JAMA Oncol 2020;doi:10.1001/jamaoncol.2020.1024]

At a median follow-up of 9.5 months, median OS was comparable between the nivolumab and the bevacizumab arms (9.8 vs 10 months; hazard ratio [HR], 1.04; p=0.76).

Grade 3/4 treatment-related adverse events (TRAEs) were similar between nivolumab and bevacizumab recipients (18 percent vs 15 percent). No unexpected neurologic TRAEs or TRAE-related deaths were reported. The safety profile of nivolumab observed in this study is consistent with those found in other tumour types, noted the researchers.

Corticosteroid, MGMT promoter influence

Baseline corticosteroid use in nivolumab recipients is apparently associated with unfavourable outcomes. [Brain 2016;139:1458-1471] The subgroup analysis appeared to correlate with this, as reflected by the inferior OS among nivolumab recipients with baseline corticosteroid use (HR, 1.41). Conversely, a trend towards an OS benefit was observed among those who had no prior corticosteroid use (HR, 0.84).

“Patients requiring corticosteroids to treat symptomatic cerebral oedema may have more rapidly progressive disease and may not have sufficient time to derive benefit from immunotherapy,” explained the researchers. Moreover, the direct effects of corticosteroids on T-cell function may abrogate immune system priming, consequently suppressing immune response, they added. [Aging 2015;7:521-522; Neuro Oncol 2010;12:631-644]

Nearly a quarter of participants (23 percent in each arm) had tumours with a methylated MGMT promoter, which is a well-known prognostic factor for glioblastoma. Subgroup analysis appeared to favour nivolumab over bevacizumab in terms of median OS in individuals with methylated tumours (HR, 0.92). Median OS was inferior among nivolumab recipients with unmethylated MGMT promoter tumours (HR, 1.34).

The post hoc subgroup analyses reinforced these exploratory results, showing OS trends favouring nivolumab-treated patients who had no baseline corticosteroid use (HR, 0.59) and had methylated MGMT promoter status (HR, 0.47).

Taken together, these findings imply that individuals with no baseline corticosteroid dependence and with methylated MGMT promoter glioblastoma have a greater likelihood of obtaining benefit from immune checkpoint inhibition, the researchers underlined.

Considering the small subgroup samples however, larger trials should probe further into MGMT promoter methylation status to elucidate its role in this setting. It would also be worth looking into the impact of corticosteroid use to validate the subgroup findings. Evaluating archival tissue collected at initial diagnosis is also recommended for future biomarker analyses, they said.

A combination of nivolumab, radiotherapy, and temozolomide in patients with newly diagnosed glioblastoma with methylated MGMT promoter is under investigation, they added.