Structural classification of <i>FOXA1</i> alterations and their association with clinical outcomes and transcriptional profiles in prostate cancer patients.

Abstract

e17071 Background: FOXA1 alterations occur in 10-15% of prostate cancers; however, due to variability in mutational mechanism, clinical outcomes associated with FOXA1 alterations have been challenging to decipher. A recent classification system for FOXA1 alterations defined by mutational mechanism and orientation with respect to the Forkhead domain has been developed to inform clinical management for prostate cancer (Hwang J. CCR 2025). Here, we have consolidated multi-institutional patient data to determine detailed outcomes and gene signatures associated with FOXA1 alteration classes in prostate cancer patients. Methods: Clinical outcomes from 62 prostate cancer patients across 3 institutions were analyzed by FOXA1 alteration class. Bivariate analyses vs. FOXA1 were examined using Chi-square test. For survival analysis, Kaplan-Meier plot with log-rank test was conducted between FOXA1 alteration class across time. Uni- and multi-variate logit regression models were used to explore the association between FOXA1 class and treatment outcomes. RNA-sequencing of prostate tumors was available for 43/62 patients; gene expression patterns by FOXA1 class were analyzed using GSEA and IPA. Results: In our integrated patient cohort, FOX1A alteration classes included 2 patients with Class 1A, 38 with Class 1B, 2 with Class 2, 4 with Class 3A, 10 with Class 3B, and 6 with Class 4 alterations. Comparing FOXA1 class to disease progression, patients with Class 1A, 2, and 4 FOXA1 alterations had shorter progressive disease-free survival (p=0.03), time on AR directed therapy (26.3 vs. 33.4 months), and time to metastasis (11.6 vs. 39.4 months) vs. Class 1B. Interestingly, Class 3A/B patients had the shortest time on taxanes (2.9 vs. 4.1 months) and the longest time to PSA nadir (34.7 vs. 18.6 months) vs. all others. Genomically, TMPRSS/ERG fusions were significantly enriched in Class 3B (40%) and 4 (34%) vs. all other classes (0-2.6%) (p=0.012), and TP53/RB genomic alterations were significantly lower in Class 1B (13.2%) and 3B (0%) vs. all other classes (34-50%) (p=0.04). Bioinformatics analyses revealed gene signatures specific to each alteration class, which could be used to stratify patients based on FOXA1 to inform clinical management of prostate disease. Conclusions: Detailed patient outcomes (n=62) were assessed between classes of FOXA1 alterations in prostate cancer. Patients with Class 1B FOXA1 alterations had improved clinical outcomes compared to patients with Class 1A, 2, and 4 (shorter progression-free survival, time to metastasis, time on AR-targeted therapy), suggesting these alterations necessitate more aggressive therapeutic approaches. These findings emphasize the importance of structural classification of FOXA1 variants, rather than just their presence or absence, for prognostication and therapeutic selection for patients with prostate cancer.