Atezolizumab vs. chemotherapy in cisplatin-ineligible patients with advanced or metastatic urothelial carcinoma highly expressing PD-L1

Previously, IMvigor130 showed a significant improvement in progression-free survival (PFS) with atezolizumab (atezo) plus chemotherapy (chemo) vs. chemo alone as first-line treatment of advanced or metastatic urothelial carcinoma. However, the combination failed to demonstrate an improvement in overall survival (OS). Updated results now show a survival benefit and better tolerability with atezo monotherapy for the subgroup of patients that are cisplatin-ineligible and whose tumours highly express PD-L1.

The phase III IMvigor130 trial evaluated the safety and efficacy of the anti-PD-L1 antibody atezolizumab (atezo), alone or in combination with chemotherapy (chemo), compared to chemo, as first-line treatment for patients with advanced or metastatic urothelial carcinoma (UC). The primary analysis showed a statistically significant improvement in progression-free survival (PFS) with atezo + chemo vs. chemo. However, the first and second interim analyses failed to show an overall survival (OS) benefit with atezo + chemo. At ASCO 2023, the final OS analysis for atezo monotherapy vs. chemo was presented, including the results for cisplatin(cis)-ineligible patients by PD-L1 status. Since the second co-primary endpoint, the OS comparison between atezo + chemo vs. chemo alone was not met, the analyses here presented are exploratory.

Study design

The phase III IMvigor130 trial enrolled platinum-eligible patients with locally advanced or metastatic UC who had not received prior systemic therapy in the metastatic setting. Patients were randomly assigned 1:1:1 to arm A (atezo + platinum-gemcitabine chemotherapy), B (atezo monotherapy) or C (placebo + platinum-gemcitabine chemotherapy). Patients were stratified by PD-L1 expression on tumour-infiltrating immune cells (IC) (IC0 vs. IC1 vs. IC2/3) and Bajorin risk factor score (0 vs. 1 vs. 2 and/or liver metastases). The first co-primary endpoints were PFS and OS in arms A vs. C in the intention-to-treat (ITT) population. The second co-primary endpoint was OS in arms B vs. C in the ITT population and PD-L1 IC2/3 populations. Due to the statistical testing hierarchy, no formal comparison of OS in arm B vs. C was performed in ITT and IC2/3 patients. Secondary endpoints included overall response rate (ORR), duration of response (DOR), and safety.

Results

The median survival follow-up was 13.4 months, and the time since the last patient was randomised was 49 months. There was no difference in OS between arms B and C in the ITT population (median OS 15.2 vs. 13.3 months, respectively; HR: 0.98). There was no difference in survival between the arms in patients who express no or very low levels of PD-L1 (median OS 13.5 vs. 12.9 months in arms B and C, respectively; HR: 1.03). In patients whose tumours highly express PD-L1 (IC2/3), there was a trend favouring OS in the atezo arm (median OS 27.5 vs. 16.7 months in arms B and C, respectively; HR: 0.70[0.48-1.03]). This suggested benefit was more pronounced when the analysis was restricted to cis-ineligible patients. In patients with high PD-L1 expression, the median OS was 18.6 vs. 10.0 months in arms B and C (HR: 0.56). No difference was observed in patients with low or non-PD-L1 expression (11.2 vs. 11.8 months; HR: 1.14). Higher ORR was observed in arm C in the ITT population (24.2% vs. 44.4% in arms B and C, respectively). When this was restricted to IC2/3 cis-ineligible patients, the responses were 40.0% vs. 32.6%). DOR was longer for atezo in both the ITT (median DOR of 29.6 vs. 8.1 months for arms B and C) and the IC2/3 cis-ineligible population (median DOR not estimable vs. 6.2 months).

No new safety signals regarding atezo monotherapy were observed.  In fact, treatment-related adverse events (AEs) of all grades and of grades ≥3 were lower in the atezolizumab monotherapy arm than in the chemotherapy arm. Looking at the AEs of special interest, the most frequent were hepatitis, rash, thyroid dysfunction, and pneumonitis, which were more frequent in the atezolizumab arm. Of note, grade 3/4 events of special interest were encountered only in 10% of patients. In the atezo group, 9% of patients had to discontinue the treatment, as opposed to 34% in the chemotherapy arm.

Conclusions

The final OS analysis was consistent with previous results, showing that atezo monotherapy did not improve OS over chemotherapy for patients with advanced or metastatic UC. However, the exploratory efficacy data suggest there is a clinical benefit with atezo monotherapy as first-line treatment for those patients who are cis-ineligible and whose tumours express PD-L1 in high levels. No new safety signals were observed and tolerability showed to be better with atezo monotherapy vs. chemo. Taking these observations together, the benefit-risk ratio of atezo vs. chemo supports atezo as a first-line treatment for cis-ineligible patients with tumours highly expressing PD-1.

Reference

Bamias A, et al. Final overall survival (OS) analysis of atezolizumab (atezo) monotherapy vs chemotherapy (chemo) in untreated locally advanced or metastatic urothelial carcinoma (mUC) from the Phase 3 IMvigor130 study. Presented at ASCO GU; Abstract LBA441.