Association of Germline Variant Status With Therapy Response in High-risk Early-Stage Breast Cancer: A Secondary Analysis of the GeparOcto Randomized Clinical Trial

Abstract

Importance: The GeparOcto randomized clinical trial compared the efficacy of 2 neoadjuvant breast cancer (BC) treatment regimens: sequential intense dose-dense epirubicin, paclitaxel, and cyclophosphamide (iddEPC) vs weekly paclitaxel and nonpegylated liposomal doxorubicin (PM) in patients with different biological BC subtypes. Patients with triple-negative BC (TNBC) randomized to the PM arm received additional carboplatin (PMCb). Overall, no difference in pathologic complete response (pCR) rates was observed between study arms. It remained elusive whether the germline variant status of BRCA1/2 and further BC predisposition genes are associated with treatment outcome.

Objective: To determine treatment outcome for BC according to germline variant status.

Design, setting, and participants: This retrospective biomarker study is a secondary analysis of the GeparOcto multicenter prospective randomized clinical trial conducted between December 2014 and June 2016. Genetic analyses assessing for variants in BRCA1/2 and 16 other BC predisposition genes in 914 of 945 women were performed at the Center for Familial Breast and Ovarian Cancer, Cologne, Germany, from August 2017 through December 2018.

Main outcomes and measures: Proportion of patients who achieved pCR (ypT0/is ypN0 definition) after neoadjuvant treatment according to germline variant status.

Results: In the study sample of 914 women with different BC subtypes with a mean (range) age at BC diagnosis of 48 (21-76) years, overall higher pCR rates were observed in patients with BRCA1/2 variants than in patients without (60.4% vs 46.7%; odds ratio [OR], 1.74; 95% CI, 1.13-2.68; P = .01); variants in non-BRCA1/2 BC predisposition genes were not associated with therapy response. Patients with TNBC with BRCA1/2 variants achieved highest pCR rates. In the TNBC subgroup, a positive BRCA1/2 variant status was associated with therapy response in both the PMCb arm (74.3% vs 47.0% without BRCA1/2 variant; OR, 3.26; 95% CI, 1.44-7.39; P = .005) and the iddEPC arm (64.7% vs 45.0%; OR, 2.24; 95% CI, 1.04-4.84; P = .04). A positive BRCA1/2 variant status was also associated with elevated pCR rates in patients with ERBB2-negative, hormone receptor-positive BC (31.8% vs 11.9%; OR, 3.44; 95% CI, 1.22-9.72; P = .02).

Conclusions and relevance: Effective chemotherapy for BRCA1/2-mutated TNBC is commonly suggested to be platinum based. With a pCR rate of 64.7%, iddEPC may also be effective in these patients, though further prospective studies are needed. The elevated pCR rate in BRCA1/2-mutated ERBB2-negative, hormone receptor-positive BC suggests that germline BRCA1/2 testing should be considered prior to treatment start.

Trial registration: ClinicalTrials.gov Identifier: NCT02125344.

Conflict of interest statement

Conflict of Interest Disclosures: Dr Loibl reported receiving grants for the GeparOcto Study from Roche, Amgen, and Teva Pharmaceutical Industries during the conduct of the study; grants and honoraria for lectures and advisory boards paid to institution from AbbVie, Amgen, AstraZeneca, Celgene, Novartis, Pfizer, and Roche; honoraria for lectures and advisory boards paid to institution from Seattle Genetics and Samsung; honoraria for lectures paid to institution from PRIME/Medscape; personal fees from Chugai; grants from Teva and Vifor Pharma; grants and honoraria paid to institution from Daiichi Sankyo; honoraria for advisory boards paid to institution from Lilly; and advisor fees paid to institution from Eirgenix outside the submitted work; in addition, Dr Loibl reported a pending patent to EP14153692.0. Dr Möbus received honoraria and lecture fees from Amgen, AstraZeneca, Celgene, Myelo Therapeutics, Novartis, Roche, and Clovis outside the submitted work. Dr Denkert reported grants from GBG Forschungs GmbH during the conduct of the study; grants for advisory or consultancy roles from MSD Oncology, Amgen, and Daiichi Sankyo; honoraria from Teva Pharmaceutical Industries, Novartis, Pfizer, Roche, and Amgen; and holding stock and other ownership interests with Sividon Diagnostics outside the submitted work; and patents, royalties, and intellectual property from VMscope digital pathology software (patent applications for EP18209672 [cancer immunotherapy] and EP20150702464 [therapy response]). Dr Fasching reported receiving grants and personal fees from Novartis, Amgen, and Celgene; grants from BioNTech and Cepheid; personal fees from Roche, Pfizer, Celgene, Daiichi Sankyo, Teva Pharmaceutical Industries, AstraZeneca, Merck Sharp & Dohme, Myelo Therapeutics, Macrogenics, Eisai, Puma, and Lilly during the conduct of the study. Dr Hanusch reported receiving grants for advisory or consultancy roles from Roche, Pfizer, Novartis, and AstraZeneca and honoraria from Roche, Pfizer, Genentech, AstraZeneca, Novartis, Eisai, Lilly, Celgene, and Amgen outside the submitted work. Dr Tesch reported receiving honoraria, travel grants, and grants for advisory or consultancy role from Novartis and Roche outside the submitted work. Dr Müller reported receiving grants, payment to institution for clinical trials, speaker honoraria, and consultancy honoraria from Roche and Pfizer; payment to institution for clinical trials, research grant to institution, speaker honoraria, and consultancy honoraria from Novartis; consultancy honoraria from Amgen, AstraZeneca, Genomic Health, Pierre Fabre, Merck Sharp & Dohme, Lilly; speaker and consultancy honoraria from Daiichi Sankyo, Eisai, Teva Pharmaceutical Industries, Tesaro, Hexal, and Novartis; speaker honoraria from Celgene; consultancy honoraria and payment to institution for clinical trials from Nektar; and research support from Novartis, Roche, Seattle Genetics, and Genentech outside the submitted work. Dr Untch reported serving on advisory boards for Amgen, Myriad Genetics, Celgene, and Roche and receiving travel support from Celgene and Roche during the conduct of the study; serving on advisory boards and receiving travel support for AbbVie, AstraZeneca, Bristol-Myers Squibb, Daiichi Sankyo, Eisai, Janssen, Cilag, Johnson & Johnson, Lilly, Merck Sharp & Dohme, Mundipharma, Odonate Therapeutics, and Pfizer; serving on advisory boards for Riemser, Sanofi-Aventis, and Teva Pharmaceutical Industries; and receiving honoraria from AbbVie, AdBoard, Amgen, AstraZeneca, Celgene, Clovis, Daiichi Sankyo, Eisai, Lilly, Merck Sharp & Dohme, Myriad Genetics, Novartis, Odonate Therapeutics, Pfizer, Pierre Fabre, Roche, Sanofi-Aventis, and Teva Pharmaceutical Industries outside the submitted work. Dr Lübbe reported receiving payment for documentation of patient data from the German Breast Group during the conduct of the study; receiving grants for advisory or consultancy from Roche and Novartis; and receiving honoraria from Pfizer, Lilly, and Genomic Health outside the submitted work. Dr Rhiem reported receiving personal fees from AstraZeneca, Pfizer, and Tesaro outside the submitted work. Dr Jackisch reported receiving travel grants from Bristol-Myers Squibb and Amgen and personal fees from Roche and AstraZeneca outside the submitted work. Drs Schmutzler and Hahnen reported receiving honoraria from AstraZeneca outside the submitted work. Dr Schneeweiss reported receiving research grants from Roche and Celgene; honoraria from Amgen, AstraZeneca, Celgene, Lilly, Merck Sharp & Dohme, Novartis, Pfizer, and Tesaro; and medical writing grants from Roche outside the submitted work. No other disclosures were reported.

Figures

Figure 1.. CONSORT Flow Diagram

Between December 2014 and June 2016, 1204 patients were screened for eligibility, 961 patients were randomized at 57 sites, 945 started treatment (intention-to-treat population), and 938 underwent surgery. Of these 938 patients, 24 patients were not included in this secondary analysis because no DNA sample was available. BC indicates breast cancer; CONSORT, Consolidated Standards of Reporting Trials; HR, hormone receptor; iddEPC, intense dose-dense epirubicin, paclitaxel, and cyclophosphamide; PM(Cb), paclitaxel and nonpegylated liposomal doxorubicin with or without carboplatin; TNBC, triple-negative BC.

Figure 2.. Germline Variant Prevalence and Pathologic Complete Response (pCR) Rates According to Variant Status and Study Arm

A, Number of patients with germline BRCA1/2 (gBRCA1/2) variants in the overall study sample (n = 914) and according to biological subtype (triple-negative breast cancer [TNBC]; ERBB2-positive breast cancer; ERBB2-negative, hormone receptor [HR]–positive breast cancer). B, Number of patients with germline variants in non-BRCA1/2 genes among the subgroup of patients without gBRCA1/2 variants (n = 818) overall and according to the biological subtype. C and D, Rates of pCR according to gBRCA1/2 variant status in the TNBC subgroup (C) and in the ERBB2-negative, HR-positive subgroup (D), overall and by study arm. For the comparison of pCR rates according to gBRCA1/2 variant status, odds ratios (95% CI) are shown (univariate logistic regression). Abbreviations: iddEPC, intense dose-dense epirubicin, paclitaxel, and cyclophosphamide; PM, paclitaxel and nonpegylated liposomal doxorubicin; PMCb, PM with carboplatin.