Predictive models of response to neoadjuvant chemotherapy in muscle-invasive bladder cancer using nuclear morphology and tissue architecture

Characterizing likelihood of response to neoadjuvant chemotherapy (NAC) in muscle-invasive bladder cancer (MIBC) is an important yet unmet challenge. In this study, a machine-learning framework is developed using imaging of biopsy pathology specimens to generate models of likelihood of NAC response. Developed using cross-validation (evaluable N = 66) and an independent validation cohort (evaluable N = 56), our models achieve promising results (65%-73% accuracy). Interestingly, one model-using features derived from hematoxylin and eosin (H&E)-stained tissues in conjunction with clinico-demographic features-is able to stratify the cohort into likely responders in cross-validation and the validation cohort (response rate of 65% for predicted responder compared with the 41% baseline response rate in the validation cohort). The results suggest that computational approaches applied to routine pathology specimens of MIBC can capture differences between responders and non-responders to NAC and should therefore be considered in the future design of precision oncology for MIBC.

Automated summary

A machine-learning framework was developed to predict the response to neoadjuvant chemotherapy in muscle-invasive bladder cancer using imaging of biopsy pathology specimens, achieving promising results through cross-validation and an independent validation cohort. One model was able to stratify the cohort into likely responders based on features derived from stained tissues and clinico-demographic features.