Phase Ib trial results of durvalumab plus tremelimumab suggest that the combination has a manageable safety profile and shows activity in pretreated patients with advanced non-small-cell lung cancer (NSCLC).
Our findings support “the continued study of the combination” in this patient population, said presenting author Naiyer Rizvi, from Columbia University Hospital in New York, USA, at the European Society for Medical Oncology Asia Congress in Singapore.
The dose escalation phase of the trial included 102 stage III or IV NSCLC patients who were given the anti-programmed death-ligand 1 (PD-L1) antibody durvalumab plus the cytotoxic T-lymphocyte-associated protein 4 blocker tremelimumab in varying dose combinations every 2 or 4 weeks for a total of 1 year. Durvalumab doses ranged from 3 to 20 mg/kg while tremelimumab was given at 1, 3 or 10 mg/kg.
Altogether 42% of study participants experienced at least one side effect of grade 3 or 4, with the most common being diarrhoea and colitis, observed in 11% and 9% of patients, respectively. Drug-related toxicity led to treatment discontinuation in 28% of the cohort and to mortality in 3%.
Toxicity increased with escalating doses of tremelimumab, clearly highlighting a tolerability issue, said Naiyer Rizvi. The incidence of grade 3 or 4 adverse events increased from 30% in the group that received tremelimumab 1 mg/kg to 56% and 78% for the tremelimumab 3 mg/kg and 10 mg/kg groups, respectively. And the corresponding rates of adverse event-related discontinuation were 16%, 44% and 56%.
However, the increasing doses did not correlate with added efficacy, but “fortunately tremelimumab at a dose of 1 mg/kg was enough to drive a T-cell response and anti-tumour activity”, Naiyer Rizvi reported.
After a median follow-up of 18.8 weeks, 25% of 84 evaluable patients achieved an objective response, a rate that was not dissimilar to the 28% objective response rate for the 39 patients who received tremelimumab at a dose of 1 mg/kg.
When all patients were stratified by PD-L1 status, the objective response rate was 35% for patients with PD-L1 levels of at least 25% and 22% for those with levels below this cutoff. And again, the rates were comparable for patients in the tremelimumab 1 mg/kg group, at 33% and 26%, respectively.
Although the discontinuation rates in the study, even at the lowest tremelimumab dose, are challenging, said the presenting author, there are ways to manage it, with one potential being continuation on single agent durvalumab.
He concluded that on the basis of their safety and efficacy findings, a phase III dose of durvalumab 20 mg/kg and tremelimumab 1 mg/kg, where both are given once every 4 weeks, has been selected and various phase III trials – such as the ARCTIC, MYSTIC and NEPTUNE – are recruiting participants.
The discussant Johan Vansteenkiste, from University Hospitals KU Leuven in Belgium, welcomed the choice of the phase III dose, especially from a clinical point of view as patients would only need to visit the clinic once in 4 weeks.
Comparing the trial findings with those of other checkpoint inhibitors, such as nivolumab, pembrolizumab and atezolizumab, he noted that the combination of durvalumab and tremelimumab has higher efficacy, but that the rates of grade 3–4 toxicity and related discontinuation are not minor.
Johan Vansteenkiste concluded that the long-term value of the combination, albeit an interesting and rational concept, still needs to be established and he awaits the results of longer follow-up with interest.
Rizvi N, Balmanoukian A, Goldberg SB, et al. Phase 1b study of the safety and antitumour activity of durvalumab (MEDI4736) + tremelimumab in advanced NSCLC. Presented at: ESMO Asia 2015 Congress. Singapore; 18–21 December 2015; 418O