Molecular subtyping of advanced bladder cancer patients and patient-derived organoids based on a 3-marker immunohistochemistry approach to evaluate chemotherapy sensitivity

Abstract

Background Bladder cancer is among the top ten most common cancers globally and one of the most expensive to treat, primarily due to high recurrence rates stemming from significant heterogeneity and mutational rates. The standard treatment, cisplatin-based combination chemotherapy, yields only 40–56% clinical responses, highlighting the need for improved patient stratification to enable precision treatments. Currently, in general practice patient stratification is based on physical fitness and immunohistopathological characterization. While molecular subtyping has traditionally relied on RNA-sequencing, this method is clinically impractical due to cost and often lacking direct sensitivity correlation. Therefore, we aim to develop an immunohistochemistry-based classification system for the major bladder cancer subtypes. Methods and Patient Summary We used the log10 IHC ratios of 3 markers to classify major bladder cancer subtypes. We applied this subtyping in 3 patient cohorts. First on a tissue microarray cohort (n = 53) from cystectomy treated MIBC patients. Furthermore, a cisplatin-based neoadjuvant chemotherapy cohort with pre- and post-treatment tissues (n = 18). Finally, patient derived bladder cancer organoid cohort (n = 8) for direct subtype specific drug sensitivity testing. Key Findings and Limitations Our findings reveal that the classification system offers a valuable approach for subtyping patients individually, in cohorts and in patient-derived organoids (n = 8 distinct organoid lines). Second, this classification system allows detection of subtype changing in patients undergoing neoadjuvant chemotherapy. Furthermore, the classification system effectively distinguishes between significantly (P = 4.07e−10) cisplatin-sensitive (basal-like) and cisplatin-resistant (luminal-like) subtypes in bladder cancer organoids and is hypothesis generating. Conclusions and Clinical Implications Our findings suggest that our 3-marker classification system could be valuable as method in future BC patient subtyping for clinical stratification of patients undergoing cisplatin-based treatment.