Abstract Background: The phosphatidylinositol 3-kinase (PI3K) pathway is frequently dysregulated in hormone receptor (HR)-positive breast cancer (BC), with activating mutations of PIK3CA detected in ~35–45% of patients (pts). Acquired mutations in the ESR1 gene, which encodes estrogen receptor α, may be associated with resistance to aromatase inhibitor (AI) therapy. Taselisib is a potent and selective PI3K inhibitor, with greater selectivity against mutant PI3Kα isoforms than wild-type (WT) via a unique mechanism. In phase I studies, taselisib plus fulvestrant had clinical activity and manageable tolerability in pts with HR-positive BC. We report exploratory analyses of PIK3CA and ESR1 from circulating tumor DNA (ctDNA). Methods: In this phase II, open-label, single-arm study (PMT4979g; NCT01296555), pts were postmenopausal with HER2-negative, HR-positive locally advanced or metastatic BC and progression or non-response to ≥1 prior endocrine therapy in the adjuvant or metastatic setting. Pts received taselisib (6 mg capsule orally, daily) plus fulvestrant (500 mg intramuscular on Days 1 and 15 of Cycle 1, then Day 1 of each 28-day cycle) until disease progression or unacceptable toxicity. PIK3CA-mutation testing on archival tumor tissue used the cobas® PIK3CA Mutation Test. The Sysmex Inostics' BEAMing Digital PCR platform was used for ctDNA analysis of ESR1 and PIK3CA mutations (pre-dose on Cycle 1, Day 1). Primary endpoints were objective response rate (ORR) and clinical benefit rate (CBR) in all pts and those with PIK3CA mutations. ORR was confirmed complete response (cCR) and confirmed partial response (cPR). CBR was cCR, cPR, or stable disease for ≥6 months. Secondary endpoints included safety, efficacy, pharmacokinetics, and exploratory biomarker analysis. Results: 60 pts were enrolled. Median age was 61.5 years (range 31–82). In the metastatic setting, pts had received prior chemotherapy (21.7%) and prior hormonal therapy (50.0%). 86.7% of pts had received prior treatment with an AI. 45 pts had PIK3CA mutation status from archival tumor tissue and ctDNA testing; concordance was 86.7% (39/45). ctDNA analysis, vs archival tumor tissue testing, identified 4 pts and 9 pts with PIK3CA mutations from pts with WT and unknown PIK3CA mutation status, respectively. Based on ctDNA analysis (N=60), 13 pts (21.7%) had mutations in both ESR1 and PIK3CA, 21 pts (35.0%) were 'mutation not detected' (MND) for both genes, 8 (13.3%) had ESR1 mutations and PIK3CA MND, and 18 (30.0%) had ESR1 MND and PIK3CA mutations. In pts with measurable disease at baseline, confirmed responses (all partial) were: PIK3CA mutation, 38.1% (8/21); PIK3CA MND, 8.7% (2/23); all pts, 22.7% (10/44). CBRs were: PIK3CA mutation, 42.9%; PIK3CA MND, 17.4%; all pts, 29.5%. ORR and CBR from ctDNA analyses were similar to archival tumor tissue data. Conclusions: ctDNA analysis identified PIK3CA mutations in pts with previously unknown or WT mutation status from archival tumor tissue; ORR and CBR were similar to those from archival tumor tissue suggesting that PIK3CA mutation testing from ctDNA may be used as a surrogate when tissue is unavailable. 21.7% of pts had mutations in both ESR1 and PIK3CA. Citation Format: Dickler MN, Saura C, Oliveira M, Richards DA, Krop IE, Cervantes A, Stout TJ, Jin H, Savage HM, Wilson TR, Baselga J. Phase II study of taselisib (GDC-0032) plus fulvestrant in HER2-negative, hormone receptor-positive advanced breast cancer: Analysis by PIK3CA and ESR1 mutation status from circulating tumor DNA [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P6-12-01.
Read full abstract