Integrating evolutionary dynamics into treatment of metastatic castrate-resistant prostate cancer (mCRPC): Updated analysis of the adaptive abiraterone (abi) study (NCT02415621).

Abstract

5041 Background: To achieve better prostate cancer control and to delay the emergency of treatment resistance, we developed an evolutionary game theory model using Lotka-Volterra equations with three competing prostate cancer "species": T+, Tp, and T-. T+ prostate cancer cells depend on exogenous androgen; Tp cells express CYP17A1, produce and depend on androgen; and T- cells are androgen-independent and abi-resistant. We applied this model to guide the on and off treatment cycles with abi for mCRPC. At the first interim analysis with 11 patients, this approach was shown to prolong the time to cancer progression with less than 50% drug usage compared to the conventional continuous Abi ( Nat Commun. 2017). Here we present the updated data of this phase 2 study. Methods: Men with asymptomatic or minimal symptomatic mCRPC were enrolled after they achieved > 50% PSA reduction with abi as a frontline therapy for mCRPC. The primary objective is feasibility and is measured by the percentage of abi responsive men who remain to be responsive to abi (defined as > 50% decline of the pre Abi PSA) after completing 2 adaptive treatment cycles. The secondary objective is to assess the clinical benefits by comparing the radiographic progression free survival (rPFS) in men undergoing adaptive Abi therapy to the historical AA 302 trial. Results: At the data cut off in Jan 2019, the study has completed enrollment for the non-African American cohort. 15 enrolled men had > 11 months of follow up. All 15 men were off Abi for at least 3 months before abi was restarted for PSA progression at cycle 1. Seven out of the 15 men had completed at least 2 adaptive therapy cycles. Four of the rest 8 men remained on study and have not reached cycle 2. Six men were off study due to radiographic progression at month 11, 20.4, 30, 30.5, 38 and 53 from their first dose of Abi. Compare to the 16.2 months median rPFS in the AA 302 trial, the median rPFS of the 15 men would be no less than 30 months (p = 0.0068, Fisher’s exact test). Their average usage of Abi was 49% of the continuous Abi. Conclusions: Adaptive Abi therapy is feasible in men who responded to Abi as a frontline therapy for mCRPC. The updated data are consistent with our initial finding that our adaptive therapy approach can prolong the time to cancer progression with less than 50% drug usage compared to the conventional continuous Abi. Clinical trial information: NCT02415621.