The impact of single agent PD-1 or PD-L1 inhibition on advanced endometrial cancers: meta-analysis.

Background: Immune checkpoint inhibitor (ICI) therapy is an emerging option for advanced endometrial cancer (EC). Mismatch repair (MMR) status is widely regarded as a biomarker predictive of response to ICIs. The predictive value of MMR based on small, single-arm trials, however, is conflicting. In this meta-analysis, we aimed to assess the activity of single-agent ICI in advanced EC, and compared the magnitude of treatment benefit in MMR deficient (dMMR) and MMR proficient (pMMR) EC.

Methods: We carried out an electronic search to identify prospective trials of single-agent ICI in advanced EC. Data on objective response rate (ORR) and progression-free survival (PFS) were extracted and pooled. ORR was estimated using the inverse variance method and subgroup difference by MMR status was examined. PFS difference according to MMR status was summarized using the Kaplan-Meier approach.

Results: From eight trials with 492 women, the pooled ORR was 19% [95% confidence interval (CI) 16% to 22%]. ORR was significantly greater in dMMR (n = 281) than pMMR EC (n = 211) (dMMR: 46%, pMMR: 8%; risk ratio 5.74, 95% CI 3.58-9.21; interaction P < 0.001). Complete response was 11% and 0.05% and median PFS was 8.3 and 2.1 months in dMMR and pMMR EC, respectively (hazard ratio PFS 0.58, 95% CI 0.38-0.89; P = 0.01). The 12-month PFS rates were 42.0% and 20.7%, respectively.

Conclusion: Single-agent ICI is associated with a 5.74 times greater objective response and 42% reduction in risk of disease progression or death in dMMR compared with pMMR EC. MMR status should be determined prospectively and be used as a stratification factor in future trials of advanced EC. Further translational analysis is urgently required to identify the cause of dMMR and allow subclassification of EC into different dMMR molecular subtypes.