28 Background: Predictive biomarkers are needed to help guide Lu-PSMA vs. taxane use in post-ARPI mCRPC. Randomized trials comparing Lu-PSMA to taxanes did not include extensive tissue NGS profiling. Kwan et al 2025 observed strong associations with ctDNA tumor fraction and differential benefit from Lu-PSMA and cabazitaxel in the TheraP trial. However, additional genomic assessments were limited by small numbers. TP53 is commonly mutated in advanced prostate cancer, affecting downstream responses to radiotherapy in preclinical models. Making use of a large real-world cohort, we sought to perform comparative effectiveness of these agents defined by TP53 status. Methods: This study used the nationwide (US-based) de-identified Flatiron Health-Foundation Medicine Prostate Cancer clinico-genomic database (FH-FMI CGDB), originating from approximately 280 US cancer clinics (~800 sites of care). Patients with metastatic prostate cancer treated with docetaxel or Lu-PSMA in the post-ARPI setting with tissue tumor genomic testing were eligible for analysis. PSA growth rate (g-rate) was calculated with R package ‘tumrg.’ Time to next treatment (TTNT) and overall survival (OS) were adjusted for between-group imbalances and for factors influencing their treatment assignment in routine practice via propensity weighting. Results: 561 patients were eligible for the cohort. Receiving docetaxel: 144 TP53mut(+) & 274 TP53mut(-). Receiving Lu-PSMA: 50 TP53mut(+) & 93 TP53mut(-). Patients receiving docetaxel vs. Lu-PSMA had faster PSA doubling times prior to treatment assignment, less prior lines of therapy, less prior docetaxel, and higher baseline PSA values. After propensity weighting, residual imbalances were largely mitigated (less than 0.1 SMD). Patients with TP53mut(-) had more favorable TTNT on Lu-PSMA vs. docetaxel (HR: 0.47, 95% CI 0.28 – 0.81, p = 0.006) while this was not observed in TP53mut(+) (HR: 1.09, 95%CI 0.61 – 1.93, p = 0.77). Similar associations were observed for OS. Significant treatment interactions were observed for Lu-PSMA vs. docetaxel by TP53mut status for TTNT (p = 0.034) and OS (p = 0.00070). Extensive sensitivity analyses observed similar results. Conclusions: In a large routine practice cohort, making use of best practices in causal inference, we observed the majority of clinical benefit from Lu-PSMA to be in the non-TP53 mutated group. Compared to similar patients who received docetaxel, we observe more favorable outcomes on Lu-PSMA only in the non-TP53 mutated group. Treatment assignments were consistent with more aggressive getting docetaxel and initial later line Lu-PSMA approval. However, now with earlier approval, TP53 status is a strong candidate predictive biomarker to guide Lu-PSMA vs. docetaxel decisions in mCRPC, warranting future studies.
Leuva et al. (Sun,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: