Molecular Characterization of Residual Muscle-Invasive Bladder Cancer Identifies a Scar-Like Transcriptomic Profile with Favorable Prognosis after Neoadjuvant Therapy

Abstract Purpose: Many patients have residual disease after neoadjuvant therapy, but their prognosis and need for adjuvant therapy are unclear. This study evaluates patient prognosis based on the molecular profiling of residual disease at cystectomy. Experimental Design: RNA sequencing data from TURBT samples were available from N = 84 ABACUS patients, of whom N = 64 had matched radical cystectomy (RC) samples, including 10 patients with pathologic complete response (ypT0N0). Pre- and post-atezolizumab tumor gene expression data were classified into molecular subtypes using the consensus subtyping model as a benchmark. Unsupervised consensus clustering was performed to categorize RC samples de novo, and each cluster was characterized using gene expression signatures. Two RC cohorts (neoadjuvant chemotherapy, N = 133 and University of Texas Southwestern, N = 94) known to harbor a scar-like biologic cluster were used for training and testing of a single-sample transcriptomic classifier that was validated in two independent (PURE-01, N = 26 ad ABACUS, N = 64) RC cohorts after neoadjuvant immunotherapy. Results: Unsupervised consensus clustering revealed four distinct post-atezolizumab clusters (scar-like, basal, luminal–stromal, and luminal). The scar-like cluster was present in 25% (16/64) of the post-atezolizumab samples and expressed genes associated with wound healing/scarring. A transcriptomic classifier trained to identify a favorable scar-like transcriptomic profile in residual bladder tumors showed robust performance in two validation cohorts, indicating a patient subgroup with favorable prognosis among neoadjuvant chemotherapy–treated, pembrolizumab-treated, and atezolizumab-treated patients with residual bladder cancer. Conclusions: This study contributes to the framework for defining molecular subtypes at RC. Residual bladder cancer with a scar-like transcriptomic profile may predict favorable patient prognosis after neoadjuvant chemotherapy and immunotherapy, identifying potential candidates for treatment deintensification.