As reported in The Lancet Oncology by Thomas Powles, MD, of Barts Cancer Institute, Queen Mary University of London, and colleagues, the phase III DANUBE trial showed that durvalumab monotherapy did not prolong overall survival vs standard chemotherapy in previously untreated patients with metastatic urothelial carcinoma with high PD-L1 expression and that durvalumab plus the CTLA-4 inhibitor tremelimumab did not improve overall survival vs chemotherapy in the intent-to-treat population.1
Study Details
In the open-label trial, 1,032 patients from 224 sites in 23 countries with unresectable locally advanced or metastatic urothelial carcinoma were randomly assigned 1:1:1 between November 2015 and March 2017 to receive durvalumab (n = 346), durvalumab plus tremelimumab (n = 342), or chemotherapy (n = 344). Treatment consisted of: durvalumab at 1,500 mg every 4 weeks; durvalumab 1,500 mg plus tremelimumab at 75 mg every 4 weeks for up to four doses, followed by durvalumab maintenance at 1,500 mg every 4 weeks; or standard-of-care chemotherapy with gemcitabine plus either cisplatin or carboplatin, depending on cisplatin eligibility, for up to six cycles. Randomization was stratified by cisplatin eligibility, PD-L1 status, and the presence or absence of liver or lung metastases at baseline. The co-primary endpoints were overall survival with durvalumab monotherapy vs chemotherapy in patients with high PD-L1 expression and with durvalumab/tremelimumab vs chemotherapy in the intent-to-treat population.
Thomas Powles, MD
High PD-L1 expression was defined as at least 25% of tumor cells with membrane staining or at least 25% of immune cells staining for PD-L1 at any intensity if more than 1% of the tumor area contained immune cells, or 100% of immune cells staining for PD-L1 at any intensity if 1% of the tumor area contained immune cells. Totals of 209 of 346 patients in the durvalumab group (60%), 205 of 342 in the durvalumab/tremelimumab group (60%), and 207 of 344 in the chemotherapy group (60%) had high PD-L1 expression. Overall, 57% of patients in the durvalumab group, 57% in the durvalumab/tremelimumab group, and 56% in the chemotherapy group were eligible for cisplatin. Disease was metastatic in 97%, 96%, and 94% of patients, respectively.
Overall Survival
At data cutoff for final analysis of overall survival (January 2020), median follow-up for survival was 41.2 months. In the high–PD-L1 population, median overall survival was 14.4 months (95% confidence interval [CI] = 10.4–17.3 months) in the durvalumab monotherapy group vs 12.1 months (95% CI = 10.4–15.0 months) in the chemotherapy group (hazard ratio [HR] = 0.89, 95% CI = 0.71–1.11, P = .30). Outcomes were consistent across most prespecified subgroups. For stratification factors, hazard ratios were 0.91 (95% CI = 0.67–1.22) among cisplatin-eligible patients and 0.87 (95% CI = 0.62–1.21) among cisplatin-ineligible patients; they were 0.85 (95% CI = 0.63–1.14) among patients with lung or liver metastases at baseline and 0.93 (95% CI = 0.66–1.31) among those without lung or liver metastases at baseline.
In the intent-to-treat population, median overall survival was 15.1 months (95% CI = 13.1–18.0 months) in the durvalumab/tremelimumab group vs 12.1 months (95% CI = 10.9–14.0 months) in the chemotherapy group (HR = 0.85, 95% CI = 0.72–1.02, P = .075). Outcomes were consistent across most prespecified subgroups. For stratification factors, hazard ratios were 0.86 (95% CI = 0.68–1.08) among cisplatin-eligible patients and 0.86 (95% CI = 0.67–1.11) among cisplatin-ineligible patients; they were 0.77 (95% CI = 0.61–0.96) among patients with lung or liver metastases at baseline and 0.91 (95% CI = 0.70–1.19) among those without lung or liver metastases at baseline.
In secondary analyses, median overall survival was 13.2 months (95% CI = 10.3–15.0 months) in the durvalumab group in the intent-to-treat population (HR vs chemotherapy = 0.99, 95% CI = 0.83–1.17) and 17.9 months (95% CI = 14.8–24.2 months) in the durvalumab-plus-tremelimumab group in the high–PD-L1 population (HR vs chemotherapy = 0.74, 95% CI = 0.59–0.93). In the low–PD-L1 population, median overall survival was 10.9 months (95% CI = 8.0–14.8 months) in the durvalumab group vs 12.2 months (95% CI = 10.4–14.0 months) in the chemotherapy group (HR = 1.14, 95% CI = 0.87–1.49). Median overall survival was 11.8 months (95% CI = 8.9–15.8 months) in the durvalumab/tremelimumab group (HR vs chemotherapy = 1.04, 95% CI = 0.80–1.36).
In the intention-to-treat population, median progression-free survival was 2.3 months in the durvalumab group, 3.7 months in the durvalumab/tremelimumab group, and 6.7 months in the chemotherapy group. In the high–PD-L1 population, median progression-free survival was 2.4 months, 4.1 months, and 5.8 months, respectively.
Adverse Events
Grade 3 or 4 adverse events regardless of causal attribution occurred in 52% of the durvalumab group, 62% of the durvalumab/tremelimumab group, and 74% of the chemotherapy group. Serious adverse events regardless of attribution occurred in 40%, 53%, and 40%, respectively.
Treatment-related deaths occurred in two patients in the durvalumab group (acute hepatic failure and hepatitis), two in the durvalumab/tremelimumab group (septic shock and pneumonitis), and one in the chemotherapy group (acute kidney injury).
The investigators concluded: “This study did not meet either of its co-primary endpoints. Further research to identify the patients with previously untreated metastatic urothelial carcinoma who benefit from treatment with immune checkpoint inhibitors, either alone or in combination regimens, is warranted.”
DISCLOSURE: The study was funded by AstraZeneca. Dr. Powles has received honoraria from AstraZeneca, Bristol Myers Squibb, Ferring, GLG Group, Janssen Research & Development, Merck, Novartis, Pfizer, Roche/Genentech, and Seattle Genetics/Astellas; has served as a consultant or advisor to AstraZeneca, Bristol Myers Squibb, Genentech/Roche, Incyte, Ipsen, Merck, Novartis, Pfizer, and Seattle Genetics; has received research funding from AstraZeneca/MedImmune and Roche/Genentech; and has been reimbursed for travel, accommodations, or other expenses by AstraZeneca, Bristol Myers Squibb, Ferring, MSD, Novartis/Ipsen, Pfizer, Research to Practice, and Roche/Genentech.
REFERENCE
1. Powles T, van der Heijden MS, Castellano D, et al: Durvalumab alone and durvalumab plus tremelimumab vs chemotherapy in previously untreated patients with unresectable, locally advanced or metastatic urothelial carcinoma (DANUBE): A randomised, open-label, multicentre, phase III trial. Lancet Oncol S1470-2045:30541-30546, 2020.
