Overall survival with first-line atezolizumab in combination with vemurafenib and cobimetinib in BRAFV600 mutation-positive advanced melanoma (IMspire150): second interim analysis of a multicentre, randomised, phase 3 study.

BACKGROUND Primary analysis of the phase 3 IMspire150 study showed improved investigator-assessed progression-free survival with first-line atezolizumab, vemurafenib, and cobimetinib (atezolizumab group) versus placebo, vemurafenib, and cobimetinib (control group) in patients with BRAFV600 mutation-positive melanoma. With a median follow-up of 18·9 months (IQR 10·4-23·8) at the primary analysis, overall survival data were immature. Here, we report the results from the second, prespecified, interim overall survival analysis. METHODS The multicentre, double-blind, placebo-controlled, randomised, phase 3 IMspire150 study was done at 108 academic and community hospitals in 20 countries. Patients aged 18 years or older with previously untreated unresectable stage IIIc or stage IV melanoma and an Eastern Cooperative Oncology Group performance status of 0 or 1 were eligible for inclusion. Patients were randomly assigned (1:1) to receive either atezolizumab (840 mg intravenously on day 1 and 15) or placebo plus vemurafenib (960 mg or 720 mg twice daily orally) and cobimetinib (60 mg once daily orally; 21 days on and 7 days off) in 28-day cycles. Atezolizumab and placebo were added to treatment regimens from cycle two onwards. Randomisation was done centrally (Durham, NC, USA) based on a permuted block randomisation scheme (block size of 4) using an interactive web-based response system and was stratified by geographical region and baseline lactate dehydrogenase concentration. Overall survival was analysed in the intention-to-treat population and safety was analysed in all patients who received at least one dose of study drug according to actual treatment received. The primary endpoint was investigator-assessed progression-free survival, which was previously reported. Here, we report the second, prespecified, interim overall survival analysis, which was planned after about 270 overall survival events had occurred. The trial is ongoing, but is no longer enrolling patients, and it is registered with ClinicalTrials.gov, NCT02908672. FINDINGS Between Jan 13, 2017, and April 26, 2018, 514 patients (median age 54 years [IQR 43-63]; 299 [58%] men and 215 [42%] women) were enrolled in the trial and randomly assigned to the atezolizumab group (256 [50%] patients) or the control group (258 [50%] patients). At the data cutoff (Sept 8, 2021), 273 patients had died (126 in the atezolizumab group and 147 in the control group). Median follow-up was 29·1 months (IQR 10·1-45·4) for the atezolizumab group versus 22·8 months (10·6-44·1) for the control group. Median overall survival was 39·0 months (95% CI 29·9-not estimable) in the atezolizumab group versus 25·8 months (22·0-34·6) in the control group (HR 0·84 [95% CI 0·66-1·06]; p=0·14). The most common adverse events of any grade in the atezolizumab group were blood creatine phosphokinase increased (123 [53%] of 231 patients), diarrhoea (116 [50%]), and pyrexia (115 [50%]). The most common adverse events of any grade in the control group were diarrhoea (157 [56%] of 280 patients), blood creatine phosphokinase increased (135 [48%]), and rash (119 [43%]). The most common grade 3-4 adverse events were increased lipase (54 [23%] of 231 patients in the atezolizumab group vs 62 [22%] of 280 patients in the control group), increased blood creatine phosphokinase (51 [22%] vs 50 [18%]), and increased alanine aminotransferase (32 [14%] vs 26 [9%]). Serious adverse events were reported in 112 (48%) patients in the atezolizumab group and 117 (42%) patients in the control group. Grade 5 adverse events were reported in eight (3%) patients in the atezolizumab group versus six (2%) patients in the control group. Two grade 5 adverse events (hepatitis fulminant and hepatic failure) in the atezolizumab group were considered to be associated with the triplet combination, and one event in the control group (pulmonary haemorrhage) was considered to be associated with cobimetinib. INTERPRETATION Additional follow-up of the IMspire150 trial showed that overall survival was not significantly improved with atezolizumab, vemurafenib, and cobimetinib compared with placebo, vemurafenib, and cobimetinib in patients with BRAFV600 mutation-positive advanced melanoma. Results of the final analysis are awaited to establish whether a significant improvement in overall survival can be achieved with long-term treatment with this triplet combination versus vemurafenib plus cobimetinib. FUNDING F Hoffmann-La Roche.