Nivolumab Does Not Boost Recurrent Glioblastoma Overall Survival

Author: By Lynda Williams, Senior medwireNews Reporter 

medwireNews: The CheckMate 143 trial investigators have not found a significant overall survival (OS) benefit with nivolumab versus bevacizumab for patients with recurrent glioblastoma. 

The phase III study recruited 369 patients with their first recurrence of grade IV glioblastoma or gliosarcoma after radiotherapy and temozolomide treatment, say David Reardon, from Dana-Farber-/Harvard Cancer Center in Boston, Massachusetts, USA, and co-authors in JAMA Oncology. 

Toxicity was “consistent with the known safety profiles” of the two agents, they write, with grade 3–4 treatment-related adverse events reported for 18.1% of the nivolumab arm and 15.2% of the bevacizumab group, including neurological events at this severity in 4.4% and 1.2% of the participants, respectively. Immune-mediated adverse events in the nivolumab and bevacizumab groups included diarrhoea (14.8 vs 7.9%), elevated alanine aminotransferase (8.2 vs 5.5%) and rash (9.3 vs 4.2%). 

After a median 9.5 months of follow-up, median OS was 9.8 months for the 184 patients randomly assigned to receive open-label nivolumab 3 mg/kg every 2 weeks, comparable to the 10.0 months achieved by the 185 patients who instead were given bevacizumab 10 mg/kg on a 2-week schedule. 

And the nivolumab and bevacizumab arms had comparable rates of OS at 6 (72.3 vs 78.2%), 12 (41.8 vs 42.0%) and 18 months (21.7 vs 21.6%). 

However, median progression-free survival was significantly shorter with nivolumab than bevacizumab (1.5 vs 3.5 months, hazard ratio=1.97) and this was reflected in the rates of PFS at 6 (15.7 vs 29.6%), 12 (10.5 vs 17.4%) and 18 months (5.8 vs 8.9%). 

Furthermore, the objective response rate was lower with nivolumab, at 7.8% versus 23.1% with bevacizumab, although the researchers note that “the responses were numerically more durable with nivolumab”, at a median of 11.1 months versus 5.3 months with bevacizumab. 

A hypothesis-generating multivariable analysis indicated that nivolumab-treated patients had significantly longer OS if they were not using corticosteroids at the start of the study than if they were taking corticosteroids (HR=0.59) and if they had a methylated MGMT promoter gene – a known prognostic factor in this patient population – than if they did not (HR=0.47). 

Among the bevacizumab-treated patients, OS was not significantly associated with baseline corticosteroid use but methylated MGMT promoter status did predict a significantly better outcome (HR=0.54). 

And when the two factors were combined, nivolumab-treated patients without baseline corticosteroid use and a methylated MGMT promoter had significantly longer OS than their bevacizumab-treated counterparts, at 17.0 versus 10.1 months and a HR of 0.58. 

“Patients requiring corticosteroids to treat symptomatic cerebral edema may have more rapidly progressive disease and may not have sufficient time to derive benefit from immunotherapy”, David Reardon et al hypothesize.

“Furthermore, direct effects of corticosteroids on T-cell function might abrogate activation or priming of the immune system”, they say.

The authors of an accompanying editorial highlight the small number of patients in each of the MGMT subgroups assessed but suggest that the higher rate of somatic variations in MGMT-methylated disease might contribute to immunotherapy efficacy in this subgroup.

“If it is validated that MGMT methylation is a biomarker of improved response to immune-based therapies, then the current practice of designing clinical trials of experimental immunotherapeutic strategies to exclude this subgroup might be counterproductive”, caution Linda Liau and Yagmur Muftuoglu, from the University of California, Los Angeles in the USA. 

References 

Reardon DA, Brandes AA, Omuro A, et al. Effect of nivolumab vs bevacizumab in patients with recurrent glioblastoma. The CheckMate 143 phase 3 randomized clinical trial. JAMA Oncol; Advance online publication 21 May 2020. doi:10.1001/jamaoncol.2020.1024

Muftuoglu Y, Liau LM. Results from the CheckMate 143 clinical trial. Stalemate or new game strategy for glioblastoma immunotherapy? JAMA Oncol; Advance online publication 21 May 2020. doi:10.1001/jamaoncol.2020.0857