Durvalumab ± tremelimumab by cisplatin eligibility in metastatic urothelial carcinoma

The randomised, phase 3 DANUBE trial did not achieve its co-primary endpoints. Therefore, data presented at EAU21 were exploratory. Durvalumab showed activity in the cisplatin-ineligible, PD-L1^high population, which is in line with results for other immune checkpoint inhibitors in this population. Tremelimumab appeared to improve the activity of durvalumab when given in combination [1].

Based on single-arm trials, atezolizumab and pembrolizumab are currently licensed in cisplatin-ineligible, PD-L1-positive patients in the first-line setting. Randomised phase 3 trials have not yet shown overall survival (OS) benefit for these immune checkpoint inhibitors over chemotherapy. There is little prospective data on the comparative efficacy of cisplatin and carboplatin in patients with a good performance status, but carboplatin is considered inferior to cisplatin.

The DANUBE trial (NCT02516241) evaluated treatment with durvalumab ± tremelimumab in patients with metastatic urothelial carcinoma. Previously published results demonstrated that DANUBE did not meet its primary endpoint [2]. There was no significant benefit of durvalumab ± tremelimumab in median OS compared with standard-of-care (SOC) chemotherapy (gemcitabine/cisplatin or gemcitabine/carboplatin, at the discretion of the treating physician). Concerning progression-free survival, no significant difference was observed between the different regimens in the control arm (median PFS 5.7 months for carboplatin vs 7.0 months for cisplatin; P=0.31). Median OS trended towards improvement with cisplatin (14.2 months) versus carboplatin (10.6 months; P=0.09). This suggests that carboplatin-treated patients do better than historical controls with good performance status.

Prof. Thomas Powles (Barts Cancer Centre, London, UK) shared results of an exploratory post-hoc analysis of durvalumab ± tremelimumab by cisplatin eligibility and in the first-line carboplatin-treated population [1]. In the intention-to-treat (ITT) population, no significant difference was observed between durvalumab and SOC in OS rates, regardless of cisplatin eligibility (see Figure). Even in the PD-L1^high cohort, there was no difference between durvalumab (HR 0.91; 95% CI 0.67–1.22) and SOC (HR 0.87; 95% CI 0.62–1.21). Nonetheless, in the cisplatin-ineligible, PD-L1^high cohort, response rates were similar to other immune checkpoint inhibitors. Adding tremelimumab to durvalumab appeared to improve activity of durvalumab in the PD-L1^high population.

Figure. OS by cisplatin eligibility in the ITT population for durvalumab versus SOC [1]



D, durvalumab; HR, hazard ratio; OS, overall survival; SOC, standard of care.
Figure kindly provided by Prof. Powles.

Biomarker analyses based on tumour mutational burden (TMB) demonstrated that OS favoured durvalumab plus tremelimumab over SOC at prespecified blood TMB cut-off ≥24 mut/Mb and tissue TMB cut-off ≥10 mut/Mb. These biomarker analyses suggest that components of immune and tumour expression may be relevant in determining outcomes.

The combination of durvalumab plus tremelimumab will be further explored in the randomised, phase 3 NILE and VOLGA trials

1. Powles T, et al. DANUBE post-hoc analysis: Outcomes for durvalumab with or without tremelimumab by cisplatin eligibility and PD-L1 biomarker status in metastatic urothelial carcinoma. Game changing session 3, EAU21 Virtual, 8–12 July 2021.
2. Powles T, et al. Lancet Oncol. 2020;21(12):1574–88.

 

Copyright ©2021 Medicom Medical Publishers



Posted on 31 August 202117 May 2024