Background This study evaluated the feasibility of achieving high response rates in stage II or III breast cancer by tailoring neoadjuvant therapy using clinical and histopathological features and the Oncotype DX Breast Recurrence Score. Genomic determinants of response and resistance were also explored. Patients and outcome measures Fifty-one patients were enrolled. The primary cohort comprised 40 patients: 15 human epidermal growth factor receptor type 2 (HER2)-amplified; 15 triple-negative (TNBC); and ten hormone receptor (HR)-positive, HER2-non-amplified tumours; with recurrence scores ≥25. Patients were treated with epirubicin and cyclophosphamide, followed by nab-paclitaxel, with the addition of trastuzumab if HER2-amplified. The primary endpoint was pathological complete response (pCR) in the breast. Pre- and post-treatment tumour samples underwent variant burden, gene and gene pathway, mutational signature profile and clonal evolution analyses. Results The pCR rates were: overall 55% ( n = 22), HER2-amplified 80% ( n = 12), triple-negative 46% ( n = 7) and HR-positive, HER2-non-amplified 30% ( n = 3). Grade 3 or 4 adverse events included febrile neutropenia (8%), neutropenia (18%), sensory neuropathy (5%), deranged transaminases (5%), fatigue (2%), diarrhoea (2%), and pneumothorax (2%). Molecular analyses demonstrated strong similarities between residual disease and matched primary tumour. ATM signalling pathway alterations and the presence of a COSMIC Signature 3 implied the majority of tumours contained some form of homologous repair deficiency. ATM pathway alterations were identified in the subset of TNBC patients who did not achieve pCR; Signature 3 was present in both pCR and non-pCR subgroups. Clonal evolution analyses demonstrated both persistence and emergence of chemoresistant clones. Conclusions This treatment regime resulted in a high rate of pCR, demonstrating that tailored neoadjuvant therapy using a genomic recurrence score is feasible and warrants further investigation. Molecular analysis revealed few commonalities between patients. For TNBC future clinical gains will require precision medicine, potentially using DNA sequencing to identify specific targets for individuals with resistant disease. Trial registration Clinicaltrials.gov NCT01830244

MRI appears more accurate for predicting pCR in HER2+ disease than other subtypes and in cancers with Ki-67 ≥ 15% compared to those with Ki-67 < 15%. Accuracy of MRI in our HR+, RS ≥ 25 cohort is comparable to previous reports of unselected HR+ disease. MRI post-NST should be interpreted in conjunction with HER2 status and Ki-67 index of the primary.

No abstract