Application of capture-based sequencing in predicting pathologic response to neoadjuvant therapy in HER2-positive breast cancer.

Abstract

e24073 Background: Pathologic complete response (pCR) to neoadjuvant therapy correlates with significantly improved survival in early-stage HER2-positive breast cancer (BC). Early indicators of pCR could conceivably be used to optimize treatment. This study aimed to reveal pCR-related genomic characteristics in both tissues and circulating tumor DNA (ctDNA). Methods: We enrolled 31 hormonal receptor (HR)-negative HER2-positive BC patients treated with neoadjuvant chemotherapy with or without trastuzumab from a single-center prospective study between January 2014 and June 2017 (ClinicalTrials.gov, NCT02041338). Twenty-two treatment naïve tumor tissues were obtained. Baseline plasma samples were collected from all 31 patients, among whom 22 plasma after 2 cycles of neoadjuvant treatment were sampled. Target-capture deep sequencing was performed to identify somatic genomic alterations. Clonal population structures were identified using PyClone. Results: TP53 and PIK3CA were the most frequently mutated genes in tumor tissues, detected in 77.3% (17/22) and 50% (11/22) of patients, respectively. The pCR rate of PIK3CA-mutated patients (18.2%, 2/11) was significantly lower than that of PIK3CA wild-type patients (72.7%, 8/11), (P = 0.03). Among patients treated with chemotherapy plus trastuzumab, all 8 patients with clonal PIK3CA-mutated did not achieve pCR. CtDNA was successfully detected in 26 of the 31 patients (83.9%) in whom somatic genomic alterations were identified. TP53 mutation, HER2 gene amplification and PIK3CA mutation were the most frequently somatic variants in the baseline plasma samples, detected in 54.8% (17/31), 41.9% (13/31) and 32.3% (10/31) of patients, respectively. Further analysis of the post-treatment plasmas showed that the pCR rate of ctDNA-negative patients after 2-cycle therapy was 87.5%, significantly higher than that in ctDNA-positive patients (18.2%; P = 0.005). Conclusions: In this study, the PIK3CA-mutation in tumor tissues associates with an unfavorable pCR rate to neoadjuvant chemotherapy plus trastuzumab. Analysis of post-treatment ctDNA offers a promising way to predict pCR in HER2-positive BC patients.