The estrogen response pathway as a putative predictive biomarker of neoadjuvant pembrolizumab benefit in patients with muscle-invasive bladder carcinoma (MIBC).

Abstract

843 Background: Sex-specific differences in incidence, stage, and prognosis of have raised interest in the role of sex steroid hormones in pathogenesis and treatment response of MIBC. Emerging evidence supports the involvement of both androgen and estrogen pathways in MIBC. We have previously identified the androgen receptor pathway as a negative predictive biomarker for neoadjuvant immunotherapy (Tateo et al. GU-ASCO24). However, little is known about the role of the estrogen receptor (ER) pathway. Methods: In this unplanned post-hoc analysis of PURE-01 study, we evaluated transcriptome-wide expression with the Decipher assay in 102 transurethral resection of the bladder (TURB) samples of cT2-4N0M0 MIBC patients (pts) treated with neoadjuvant pembrolizumab and radical cystectomy (RC). Here, we focused on ER gene expression and estrogen response signatures (ERS-early and ERS-late), assessing the expression of 200 estrogen pathway-related genes (PMID: 26771021). Moreover, we analyzed the correlation between Immune-190 signature (PMID: 28740126) and ERS-early and -late scores. Pts and tumor characteristics were compared using Wilcoxon rank-sum tests, and Kaplan-Meier analysis assessed the impact of gene expression and signature scores on relapse-free survival (RFS), defined as the time from the first pembrolizumab dose to relapse. Results: From 2017 to 2022, 87 (85.3%) males and 15 (14.7%) females received pembrolizumab and RC. Of note, we did not find significant differences in tumor-based ER gene expression or signature scores between females and males. ER gene expression did not vary across principal molecular subtyping models (TCGA, Consensus, and GSC classifications) although ERS–early and –late pathway activity scores were consistently elevated in luminal tumors. Only ERS-late score differed significantly according to the response to neoadjuvant therapy, showing significantly lower levels in patients who achieved a pathological complete response (p = 0.019). There were no significant differences in RFS based on ER gene expression, while low ERS–early and -late pathway activity scores were significantly associated with better RFS (p = 0.039 and p = 0.009, respectively). Scatter plots revealed a significant, moderate negative correlation between the Immune-190 signature and both ERS-early (cor = -0.64, p = < 0.001) and ERS-late (cor = -0.6, p < 0.001) scores. Conclusions: Estrogen response pathway emerged to be a putative biomarker of immunotherapy benefit in MIBC, with particularly higher scores observed for luminal molecular subtypes, which are related to worse outcomes following neoadjuvant checkpoint inhibition. These findings require further validation in prospective studies and underscore the importance of molecular subtype-driven approaches, potentially facilitating novel combination therapies in MIBC.