Validation of Prognostic Scores in Patients With Metastatic Urothelial Cancer Enrolling in Phase I Targeted Therapy or Next Generation Immunotherapy Trials

Introduction: Enrolling patients with metastatic urothelial carcinoma (mUC) in phase I trials provides an opportunity to identify biological drug activity. Developing prognostic scores may aid in patient selection for phase 1 trials. Patients and Methods:: We analyzed records of patients with mUC who participated in targeted therapy and immunotherapy phase I clinical trials at MD Anderson Cancer Center (MDACC). The Bellmunt and Bajorin scores were calculated as bladder cancer-specific prognostic scores. The Royal Marsden Hospital (RMH) and MDACC scores were calculated as phase I prognostic scores. Hazard ratios (HR) were calculated using the Cox proportional hazard model. The prognostic value of the Bellmunt, Bajorin, RMH, and MDACC scores were assessed using the Likelihood ratio (LR) χ2 test and the c-index. Results: Between 2015 and 2019, 43 patients were enrolled in phase I trials and 12 were enrolled in >I trial leading to a total of 57 trial participants (TPs). Ninty-seven percent of TPs received prior platinum therapy and 60% received a prior checkpoint inhibitor. Median overall survival (OS) and progression-free survival (PFS) were significantly shorter with increasing Bajorin, RMH, or MDACC scores, but not with increasing Bellmunt score. The RMH (c-index=0.658, LR χ2=11.8, P=.008) and MDACC scores (c-index =0.66, LR χ2=12.76, P=.01) outperformed the Bajorin score (c-index=0.522, LR χ2=1.22, P=.5) and the Bellmunt score (c-index=0.537, LR χ2=0.36, P=.9) in predicting overall survivalover. The Bajorin, RMH, and MDACC scores, but not the Bellmunt score, were also predictive of progression-free survival (PFS)prog. The RMH and MDACC scores again outperformed the Bajorin scoreand the Bellmunt score for predicting PFS. Conclusion: The RMH and MDACC phase I prognostic scores accurately predicted survival in patients with mUC and outperformed the bladder cancer-specific scores at time of enrollment on phase 1 clinical trials. The RMH and MDACC scores could optimize selection of patients with mUC for phase I clinical trials.