Abstract BACKGROUND: In early-stage triple negative breast cancer (TNBC), the addition of carboplatin (CBDCA) to neoadjuvant chemotherapy (CT) increases pathologic complete response (pCR) and relapsed-free survival. However, it is unclear whether CBDCA improves overall survival (OS). In addition, the prognostic and/or predictive role of pathogenic germline variants (PGV) in BRCA1/2 genes and other cancer risk in this setting is not fully understood. Here, we assessed the efficacy of (neo)adjuvant CBDCA and the prognostic and predictive role of a panel of 14 genes in patients (pts) with eTNBC. METHODS: This is a retrospective study on 117 pts diagnosed with early-stage TNBC between 2000-2021 at Hospital Clinic of Barcelona. Eighty-one pts (69%) were candidates for PGV testing. Overall, 14 genes (PGV) (BRCA1, BRCA2, PALB2, BRIP1, CHEK2, TP53, ATM, RAD51C, RAD51D, BARD1, MLH1, MSH2, MSH6 and PMS2) were assessed using the TruSight hereditary cancer panel (Illumina MySeq platform) according to local guidelines. Univariable and multivariable logistic regression and Cox regression analyses were performed to identify clinical and molecular predictors of pCR and relapse-free survival (RFS), respectively. Chi-squared or Fisher’s exact tests were used to assess characteristics’ distribution as appropriate. RESULTS: Of 117 pts, 83 (71%) received CT in the neoadjuvant setting and 28 (24%) in the adjuvant setting. CBDCA was added to standard CT in 68 pts (82%) in the neoadjuvant cohort and 7 pts (6%) in the adjuvant cohort. Among pts with germline testing, 32/81 (39%) harbored PGV. BRCA1 was the most frequently mutated gene (18/32, 56%), followed by BRCA2 (5/32, 16%), PALB2 (4/32, 13%), BRIP1 (2/32, 6%), CHEK2, TP53 and PMS2 (3/32, 6%). Percentages of pts receiving CBDCA were similar between patients with and without PGV (14/16,87% and wild type (53/67,79%)(p=0.120). In the neoadjuvant cohort (n=83), CBDCA was the only variable significantly associated with pCR at both univariate (pCR rates of 58.5% with CBCDA and 14.3% without CBCDA; odds ratio [OR]=8.2 [95% CIs 2.0-55.7], p=0.008) and remained statistically significant after adjusting for PGV, tumor size and nodal status (OR=6.9 [95% CIs 1.4-53.0], p=0.028). pCR rates according to PGV are reported in Table 1. In terms of RFS, addition of CBDCA to neoadjuvant therapy (hazard ratio [HR]=0.2 [0.1-0.45], p< 0.001), PGV (HR=0.2 [0.05-0.9], p=0.048), pCR (HR=0.2 [0.1-0.6], p=0.004) and nodal status (HR=5.1 [1.7-15.2], p< 0.003) were significantly associated with RFS in univariate analyses. In a multivariable model, CBDCA remained an independent predictor of improved RFS along with pCR. When the neoadjuvant and adjuvant cohort were pooled together (n=117), platinum-based CT remained significantly associated with better RFS (HR=0.2 [0.14-0.86], p=0.021) regardless of time of administration (i.e., neoadjuvant or adjuvant). With a median follow-up of 5 years, CBCDA use was not found associated with OS (HR=0.7 [0.3-1.8], p= 0.560). CONCLUSIONS: The addition of CBDCA to standard CT was significantly associated with pCR and RFS but not OS, consistent with the phase III data. The benefit in terms of RFS was independent of the presence and the type of pathogenic germline alterations. Table 1. pCR rates according to PGV Citation Format: Adela Rodríguez Hernández, Benedetta Conte, Laia Fernández, Fara Brasó-Maristany, Belén Pastor, Miriam Potrony, Lorena Moreno, Elia Grau, Joan Antón Puig-Butillé, Aurora Sánchez, Blanca González-Farré, Esther Sanfeliu, Olga Martínez-Sáez, Claudette Falato, Maria Vidal, Nuria Chic, Tomás Pascual, Francesco Schettini, Montserrat Muñoz, Francesc Balaguer, Aleix Prat, Barbara Adamo. Efficacy of platinum-based chemotherapy and germline mutational status in early-stage triple-negative breast cancer: a unicenter retrospective analysis with long-term follow-up [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P6-01-09.