Assessment of FOXC1 expression as a predictor of response to neoadjuvant taxane plus platinum regimens in primary triple-negative breast cancer: Retrospective analysis of three clinical trial cohorts

569

Background: Taxane and platinum (TP) NAC regimens, e.g. Carboplatin and Docetaxel (CbD), in TNBC are currently of great interest, having good pathologic complete response (pCR) rates but with a significantly more manageable toxicity profile compared to anthracycline-based NAC regimens. Forkhead Box C1 (FOXC1), a transcriptional driver of cell plasticity/partial EMT/metastasis is an established mesenchymal marker diagnostic of basal-like breast cancer having proven prognostic value, but of uncertain predictive value. We sought to evaluate the potential of FOXC1 in predicting pCR to neoadjuvant TP regimens in patients diagnosed with TNBC. Methods: Pre-NAC tumor biopsy FOXC1 mRNA expression status was correlated with rate of pCR in a pooled, ambispective cohort (prospective cohort GEICAM/2006-03, NCT00432172 pooled with multi-institutional retrospective cohort, n = 119). A specific FOXC1 mRNA expression cutoff value was derived to maximize Negative Predictive Value (NPV) and Sensitivity for pCR prediction. The pCR-predictive ability of FOXC1 mRNA expression was then assessed in two validation cohorts of evaluable patients who had been enrolled in prospective clinical trials (UCONN/FIOCRUZ, n = 222, HGUGM, NCT01560663, n = 221). All evaluated patients had been diagnosed with TNBC and had received a Taxane plus Platinum-based NAC regimen. Results: FOXC1 mRNA expression was associated with pCR in CbD/TP treated TNBC patients with pCR rates of 43.48%, 47.89% and 52.73% observed in the discovery and two validation cohorts (two tailed T-test p-values of 0.0005, 0.002, 0.009, respectively). FOXC1 expression above the pre-determined cutoff value was associated with pCR to CbD/TP NAC in patients diagnosed with TNBC in both validation cohorts (OR 4.894, 1.504-15.924; p = 0.004 and OR 2.293, 1.208-4.352; p = 0.006). Conclusions: We report the retrospective validation of pre-NAC breast cancer biopsy FOXC1 mRNA expression for predicting efficacy of CbD/TP NAC in two independent, prospectively accrued TNBC patient cohorts. The described strategy may be acceptable for patient stratification to guide CbD/TP NAC recommendations in TNBC. FOXC1 mRNA or protein expression, assessed using qRT-PCR or routine immunohistochemistry (IHC), respectively, could potentially be utilized in future fixed-arm/adaptive clinical trials to further optimize NAC efficacy, in terms of achieved pCR rates, and to extend disease-free survival in patients diagnosed with TNBC.

A phase Ib study to evaluate the oral selective estrogen receptor degrader GDC-9545 alone or combined with palbociclib in metastatic ER-positive HER2-negative breast cancer

1023

Background: GDC-9545 is a potent, orally available, selective estrogen receptor degrader developed for the treatment of ER-positive (ER+) breast cancer alone or combined with CDK4/6 inhibitors. A first-in-human study evaluated 10-250 mg GDC-9545; tolerability, pharmacokinetic (PK), pharmacodynamic (PD), and clinical results support expansion cohorts at ≥30 mg (Jhaveri et al., 2019). Methods: This study evaluated PK, PD, and efficacy of GDC-9545 alone and combined with palbociclib, ± LHRH agonist. Eligible patients (pts) had ER+ (HER2-) metastatic breast cancer (MBC) with ≤ 2 prior therapies in the advanced or metastatic setting. No prior treatment with CDK4/6 inhibitor was allowed in pts receiving palbociclib. Results: Eight-five pts were enrolled in 2 cohorts: GDC-9545 100 mg given once daily ± LHRH agonist (Cohort A), and GDC-9545 100 mg +125 mg palbociclib on a 21 day on/7 day off schedule ± LHRH agonist (Cohort B). Of the 39 pts in Cohort A, adverse events (AE) occurring in ≥10% of pts were fatigue, cough, back pain, pain in extremity, and arthralgia. Related AEs were generally Grade (G) 1-2; there were 3 related G3 AEs of fatigue, transaminase increased, and diarrhea. Two pts had GDC-9545 reduced, one due to G3 diarrhea and another due to G3 transaminitis. Of the 46 pts in Cohort B, AEs in ≥10% of pts were neutropenia, fatigue, bradycardia, diarrhea, constipation, dizziness, nausea, anemia, asthenia, thrombocytopenia, pruritus, and visual impairment. Twenty-six (57%) pts had G≥3 AEs. G≥3 neutropenia was reported in 23 (50%) pts. One pt had palbociclib reduced due to G3 febrile neutropenia. Eleven (13%) of 85 pts had G1 asymptomatic bradycardia considered related to GDC-9545. No pts in either cohort discontinued study treatment due to AEs. PK analysis and clinical data demonstrate no clinically relevant drug-drug interactions between GDC-9545 and palbociclib. Reduced ER, PR, and Ki67 levels, and an ER activity signature, were observed in paired pre- and on-treatment biopsies (n = 12). Eighteen of 33 pts in Cohort A had either confirmed partial responses or were on study 24 weeks (clinical benefit rate 55%). Clinical benefit was observed in pts with prior fulvestrant treatment and with detectable ESR1 mutations at enrollment. Clinical benefit data for both cohorts are anticipated to be mature in April 2020. Conclusions: GDC-9545 was well-tolerated as a single agent and in combination with palbociclib with encouraging PK, PD, and anti-tumor activity in ER+ MBC to support Phase III development. Clinical trial information: NCT03332797 .