Ten years ago TCGA and two private groups reported that muscle-invasive bladder cancers (MIBCs) could be grouped into basal and luminal molecular subtypes that shared features with the corresponding subtypes of breast cancer. These and earlier studies by the group at Lund University demonstrated that molecular subtype membership was prognostic: patients whose tumors were assigned to basal/squamous subtypes had shorter survival than did patients whose tumors were assigned to the luminal subtypes, and these relationships were confirmed in subsequent studies. In addition, one of the studies included tumors from patients treated with neoadjuvant cisplatin-based combination chemotherapy (NAC), where patients whose tumors were infiltrated with smooth muscle cells and fibroblasts (the MD Anderson p53-like subtype) had the lowest response rates. Parallel comparisons of patients who either did or did not receive NAC concluded that patients with basal tumors derived the most clinical benefit in terms of disease-specific or overall survival. Thus, in addition to being prognostic, there was a suggestion that molecular subtypes might also be used to predict resistance to NAC. This work led to the design of two large collaborative projects that were designed to validate these observations and the conclusion that inactivating mutations in several DNA damage and repair (DDR) genes were also associated with response. One leveraged the S1314 Phase-2 “CoXEN” trial that was designed to prospectively validate another gene expression-based biomarker, and the other exploited pretreatment samples from the CALGB/Alliance 90601 Phase-3 trial of gemcitabine/cisplatin with or without bevacizumab in patients with advanced or metastatic disease. Comparisons of response rates as a function of molecular subtype membership in S1314 failed to validate the original link between the MD Anderson p53-like subtype and resistance, and patients with basal/squamous tumors in 90601 did not have better outcomes than did patients with luminal tumors. Thus, while the prognostic value of molecular subtype membership was confirmed, the predictive value was not. Developing biomarkers to predict response to chemotherapy remains a challenge not only for bladder cancer but also for other solid and hematopoietic malignancies. It seems likely that success will require the integration of measurements using multiple platforms, including liquid biopsies and others. The exciting results of the EV-302 clinical trial of enfortumab vedotin plus pembrolizumab have shifted the focus of biomarker work to these and other novel agents.

Citation Format: David J McConkey. Molecular subtypes and clinical benefit from neoadjuvant chemotherapy: A ten-year update [abstract]. In: Proceedings of the AACR Special Conference on Bladder Cancer: Transforming the Field; 2024 May 17-20; Charlotte, NC. Philadelphia (PA): AACR; Clin Cancer Res 2024;30(10_Suppl):Abstract nr IA003.