Molecular Characterization of Residual Muscle-Invasive Bladder Cancer Identifies a Scar-Like Transcriptomic Profile...

Area of Science:

• Oncology
• Molecular Biology
• Genomics

Background:

• Prognosis for patients with residual bladder cancer post-neoadjuvant therapy remains unclear.
• Determining the need for adjuvant therapy in these patients is challenging.
• Molecular profiling of residual disease at cystectomy offers insights into patient outcomes.

Purpose of the Study:

• To evaluate patient prognosis based on molecular profiling of residual bladder cancer.
• To identify distinct molecular subtypes in residual tumors after neoadjuvant immunotherapy.
• To develop and validate a transcriptomic classifier for predicting prognosis.

Main Methods:

• RNA-sequencing (RNA-seq) data from transurethral resection of bladder tumor (TURBT) and radical cystectomy (RC) samples were analyzed.
• Unsupervised consensus clustering (CC) categorized RC samples into molecular subtypes.
• A single-sample transcriptomic classifier (TC) was trained and validated in independent cohorts.

Main Results:

• Four molecular subtypes were identified post-atezolizumab treatment: scar-like, basal, luminal-stromal, and luminal.
• A scar-like cluster, associated with wound healing/scarring, was found in 25% of samples.
• The TC demonstrated robust performance in identifying a favorable scar-like transcriptomic profile, predicting good prognosis.

Conclusions:

• This study establishes a framework for defining molecular subtypes in residual bladder cancer at radical cystectomy.
• A scar-like transcriptomic profile in residual disease may predict favorable prognosis following neoadjuvant chemo- and immunotherapy.
• These findings can inform treatment strategies for bladder cancer patients with residual disease.