Background: Mutations of genes involved in the PI3K signalling pathway occur frequently in BC. We performed a phase II study of BYL719, a selective PI3Kα inhibitor, in BC to evaluate its efficacy and identify biomarkers that correlate with tumor response. Methods: Eligible patients had advanced ER+/HER2- BC and documented genetic alteration of the PI3K pathway detected in either tumor or plasma. Patients received BYL719 350mg orally daily. The primary end point was RECIST objective response rate (ORR). Secondary endpoints were clinical benefit rate (CBR) including stable disease for ≥24 weeks, progression-free survival (PFS) and efficacy according to baseline and week 8 change in ctDNA levels. ctDNA analysis was performed through droplet digital PCR. Results: Of the total 17 patients treated, 16 (94%) had a PIK3CA mutation (mt) and 1 had a PTEN deletion. Of the PI3KCA mutant patients, 8 (47%) had a kinase domain mt, 7 (41%) had a helical domain mt and 1 (6%) patient had 2 PIK3CA mts. 16/17 (94%) patients had their alteration detected in plasma at baseline; concordance between plasma and tumor was 94% (15/16). Median age was 60 years (45-77). Patients had received a median of 3 prior treatment lines (1-7) in the metastatic setting; 16 (94%) had received prior endocrine therapy and 14 (82%) prior chemotherapy. 15 (88%) had visceral disease. The ORR (centrally reviewed) was 41% (7/17) and CBR was 59% (10/17). Median PFS (mPFS) was 5.49 months (mo) (95% CI, 3.63 – 13.78). The most common grade ≥ 3 adverse event was hyperglycaemia (5/17; 29%). Co-existent MAP3K1 or ESR1 mt (N = 7, 41%) had a longer clinical benefit (mPFS 11.02 mo (95%CI, 5.44-NR) with 3 (18%) ≥1 yr) compared with co-existent TP53 mt, CCND1 or FGFR1 amplification (N = 10, 59%) (mPFS 3.66 mo (95%CI, 3.63-NR)). Patients who achieved clinical benefit had a significantly greater decrease in ctDNA levels at week 8 from baseline (median 97.34% vs 9.14% decrease, p = 0.04). Conclusions: BYL719 in previously treated advanced ER+/HER2- BC with PIK3CA mt detectable in plasma at baseline demonstrated robust clinical benefit. This biomarker should be considered in the Phase III setting. Clinical trial information: NCT02506556