Abstract CT102: Exploring the tumor molecular landscape associated with<i>TMPRSS2-ERG</i>: A post hoc analysis from the phase 3 TALAPRO-2 study of first-line (1L) talazoparib (TALA) + enzalutamide (ENZA) vs placebo (PBO) + ENZA in metastatic castrati

Abstract Background: In TALAPRO-2 (NCT03395197), patients (pts) unselected for homologous recombination repair gene alterations received TALA + ENZA or PBO + ENZA in 1L mCRPC. TALA + ENZA significantly improved radiographic progression-free survival and overall survival. We found enhanced efficacy with TALA + ENZA vs PBO + ENZA in pts bearing ctDNA TMPRSS2-ERG. TMPRSS2-ERG fusions are common in advanced prostate cancer: 2341/17634 (13%) in ctDNA and 9949/26415 (38%) in tumor tissue, based on the Foundation Medicine database, which is comparable to TALAPRO-2 (98/687 [14%] and 143/484 [30%], respectively). Methods: TALAPRO-2 datasets for this analysis included prospectively collected/retrospectively analyzed plasma ctDNA (clinical trial assay based on FoundationOne®Liquid CDx; N = 681 informative records for the safety population used in all genomic analyses, except for alteration prevalence which used the intent-to-treat population, N = 687 informative records). Alterations were defined as known/likely pathogenic variants. Archival/baseline tumor transcriptomic data were generated using HTG’s Oncology Biomarker Panel (N = 304). Data cutoff Aug 16, 2022. Results: Multiple genes implicated in prostate cancer pathobiology showed imbalances in alteration status in TMPRSS2-ERG vs non-TMPRS2-ERG ctDNA: TP53 (60/98 [61%] vs 202/589 [34%]; p &amp;lt; 0.00001, Fisher’s exact test); PTEN (33/98 [34%] vs 77/589 [13%]; p &amp;lt; 0.00001); SPOP (0/98 vs 43/589 [7%]; p = 0.002); AR (54/98 [55%] vs 178/589 [30%]; p &amp;lt; 0.00001); CDK12 (1/98 [1%] vs 39/589 [7%]; p = 0.032); BRCA2 (12/98 [12%] vs 34/589 [6%]; p = 0.0269). ctDNA burden was higher with TMPRSS2-ERG, consistent with poor prognosis; tumor mutational burden was modestly higher. The tumor transcript most elevated in TMPRSS2-ERG vs non-TMPRS2-ERG was ERG (p = 5.10E-15 using logistic regression). Using gene expression signature analysis, Notch and Wnt-beta-catenin pathways were differentially activated in TMPRSS2-ERG tumors. Limitations of these retrospective exploratory analyses include low ctDNA in some pts and not correcting for multiple statistical comparisons. Conclusions: In TALAPRO-2 the presence of TMPRSS2-ERG fusions in ctDNA is associated with a distinct tumor molecular landscape, including positive association with TP53 alterations, negative association with SPOP alterations, and Notch and Wnt-beta-catenin pathway activation. These associations are consistent with those reported recently (e.g., Ding et al, PMID: 37537199; Kron et al, PMID: 28783165; Nakasawa et al, PMID: 34783071). The enhanced efficacy reported for TALA + ENZA vs PBO + ENZA (Piulats et al, AACR 2024) in pts bearing TMPRSS2-ERG ctDNA is striking considering the high prevalence of TP53 alterations in this subgroup. Citation Format: Xinmeng Jasmine Mu, Josep M. Piulats, A. Douglas Laird, Arun A. Azad, Glenn Liu, Steven M. Yip, Stefanie Zschäbitz, Joan Carles, André P. Fay, Nobuaki Matsubara, Robert J. Jones, Ugo De Giorgi, Whijae Roh, Michelle Saul, Jijumon Chelliserry, Dana Kennedy, Matko Kalac, David Fabrizio, Karim Fizazi, Neeraj Agarwal. Exploring the tumor molecular landscape associated withTMPRSS2-ERG: A post hoc analysis from the phase 3 TALAPRO-2 study of first-line (1L) talazoparib (TALA) + enzalutamide (ENZA) vs placebo (PBO) + ENZA in metastatic castration-resistant prostate cancer (mCRPC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr CT102.