Impact of Programmed Death-ligand 1 Expression on Oncological Outcomes in Patients with Muscle-invasive Bladder Cancer Treated with Radiation-based Therapy

Gautier Marcq; Gertruda Evaristo; Ronald Kool; Surashri Shinde-Jadhav; Rodrigo Skowronski; José João Mansure; Luis Souhami; Fabio Cury; Fadi Brimo; Wassim Kassouf

doi:10.1016/j.euros.2022.06.009

Impact of Programmed Death-ligand 1 Expression on Oncological Outcomes in Patients with Muscle-invasive Bladder Cancer Treated with Radiation-based Therapy

Eur Urol Open Sci. 2022 Jul 13:43:14-21. doi: 10.1016/j.euros.2022.06.009. eCollection 2022 Sep.

Authors

Affiliations

¹ Division of Urology, McGill University Health Centre, McGill University, Montreal, Quebec, Canada.
² Urology department, Claude Huriez Hospital, CHU Lille, Lille, France.
³ CNRS, Inserm, CHU Lille, Institut Pasteur de Lille, UMR9020-U1277 - CANTHER - Cancer Heterogeneity Plasticity and Resistance to Therapies, University of Lille, Lille, France.
⁴ Division of Pathology, McGill University Health Centre, McGill University, Montreal, Quebec, Canada.
⁵ Division of Radiation Oncology, McGill University Health Centre, McGill University, Montreal, Quebec, Canada.

Abstract

Background: No biomarkers are recommended for patients undergoing radiation-based therapy (RT) for muscle-invasive bladder cancer (MIBC).

Objective: We aim to evaluate the predictive role of programmed death-ligand 1 (PD-L1) expression on the oncological outcomes of patients treated with RT for MIBC.

Design setting and participants: A single-center retrospective analysis of tumor specimens collected through transurethral resection (TURBT) from 104 MIBC patients, implemented in a tissue microarray and stained with the SP263 PD-L1 clone (Ventana Medical Systems, Tucson, AZ, USA), was conducted. Two reviewers measured the PD-L1 H-score for tumor and immune cells.

Intervention: RT (maximal TURBT followed by radiation and concurrent chemotherapy when eligible).

Outcome measurements and statistical analysis: Logistic and Cox regression models were used to predict 3-mo complete response (CR) and overall survival (OS) after RT, respectively.

Results and limitations: A total of 88 (85%) patients had cT2 disease and 39 (37.5%) had high immune cell PD-L1 expression. A CR was achieved in 68 (65%) patients. On the multivariable analysis (MVA), a higher clinical stage (p = 0.02) and a low immune cell PD-L1 H-score (p = 0.02) were associated with a decreased CR after RT. The median time to death was 43 mo (95% confidence interval 20-66). On Cox MVA, a high immune cell PD-L1 H-score (p = 0.0017) was associated with better OS, independently of performance status (p = 0.0005) or tumor stage (p = 0.0013). A high tumor cell PD-L1 H-score was not an independent predictor of CR or OS. Limitations of the study include the retrospective design.

Conclusions: MIBC patients with high PD-L1 expression on immune cells appear to have better oncological outcomes following RT. Our results may aid in patient stratification for future clinical trial design.

Patient summary: In this report, we evaluated the role of programmed death-ligand 1 (PD-L1) expressed on tumor and immune cells in the tumor microenvironment for patients treated with a bladder-sparing regimen. We found that PD-L1 overexpression on immune cells is able to predict a better response to radiation-based therapy.

Keywords: Bladder-sparing approach; Muscle-invasive bladder cancer; Programmed Death-Ligand 1; Radiation-based therapy; Tumor microenvironment.