Abstract only 1090 Background: In the EMBRACA trial (NCT01945775) of pts with germline BRCA1/2-mutated HER2-negative locally advanced/metastatic breast cancer (ABC), the poly(ADP-ribose) polymerase (PARP) inhibitor TALA significantly improved progression-free survival (PFS) vs PCT (8.6 vs 5.6 mo; HR 95% CI 0.54 0.41-0.71; P < 0.0001). Patient-reported outcomes favored TALA, and most common adverse events included anemia, fatigue, and nausea. Previous subgroup analyses found that pts with a history of CNS metastases had improved PFS for TALA vs PCT (HR 95% CI 0.32 0.15-0.68; P = 0.0016) and improved objective response rate (ORR) 63.2% vs 15.8%, respectively (odds ratio 95% CI 8.95 1.86-52.26; P = 0.0013). This retrospective subgroup analysis further explored the clinical characteristics and outcomes in pts with a history of CNS metastases in EMBRACA. Methods: Pts were randomized 2:1 to TALA or PCT. Pts with adequately treated and stable CNS metastases not requiring corticosteroids were included. This analysis assessed intracranial ORR and best overall response (BOR) based on investigator assessment per RECIST 1.1 in pts with intracranial disease at baseline (data cutoff 15-Sep-17), and overall survival (OS; data cutoff 30-Sep-19). Results: In the intent-to-treat (ITT) population, 63 pts (43/287 15.0% TALA and 20/144 13.9% PCT) had a history of CNS metastases, of which 33 (11.5%) pts (TALA) and 15 (10.4%) pts (PCT) had intracranial disease at baseline. Additional baseline characteristics are shown in the table. Intracranial ORR in pts with intracranial disease at baseline and unconfirmed complete or partial response was 18.2% (TALA) vs 20.0% (PCT) (odds ratio 95% CI 0.78 0.13-5.80; P = 0.765). In pts with intracranial disease at baseline, an intracranial BOR of stable disease was 69.7% for TALA vs 33.3% for PCT. Median OS in pts with a history of CNS metastases was 12.9 mo (95% CI 9.4-15.6) for TALA and 13.4 mo (95% CI 8.8-17.6) for PCT (HR 95% CI 0.67 0.37-1.2; P = 0.1936 stratified log-rank test). In the safety population (n = 43, TALA; n = 19, PCT), median treatment duration (range) with TALA was 5.0 (0.1-36.0) mo compared with 2.1 (0.4-6.9) mo for PCT. Conclusions: In this subgroup analysis, baseline characteristics between pts with a history of CNS metastases treated with TALA or PCT were comparable. More pts with intracranial disease at baseline treated with TALA vs PCT experienced stable disease. Intracranial ORR in pts with intracranial disease was 18.2% for TALA vs 20.0% for PCT. Treatment options for pts with a history of CNS metastases are limited and further investigation in larger data sets is warranted. Clinical trial information: NCT01945775 .Table: see text