Biomarker evolution after neoadjuvant systemic therapy in breast cancer.

Abstract

e12615 Background: Tumor heterogeneity is one of the hallmarks of breast cancer (BC) and has major implications for treatment response and resistance. Currently, there are no specific guidelines regarding biomarker reassessment on residual tumor post-neoadjuvant systemic therapy (NAST). In this study, we compared biomarker discordance between pre- and post-NAST tumors at our institution and determined if actionable changes were observed. We also report the impact of the detected biomarkers differences on the choice of adjuvant treatment. Methods: We conducted a retrospective review of BC patients who received NAST followed by definitive surgery from 2004 to 2020. The initial tissue biopsies were tested for estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) status per ASCO/CAP guidelines and repeated on the residual tumor tissue after NAST. Descriptive statistics was used to compare clinicopathologic characteristics. The biomarkers were compared using Wilcoxon’s t test and McNemar’s test. Concordance analysis of intrinsic subtypes was evaluated by Cohen’s kappa method. Results: Out of 434 patients who undergone surgical resection following NAST, 95 (22%) patients were identified to have repeat biomarker testing of residual invasive BC post NAST. 30 out of the 96 patients were found to discordant receptor status on repeat biomarker testing. In this cohort of thirty patients with discordant biomarkers, median age was 51.5 (29-70 years), 33.3% cases had stage II, and 56.7% had stage III BC. 66.75 patients were ER positive and/or PR positive, 20% patients were HER2 positive with one triple positive, and 13.3% patients were triple negative on pre-NAST tumors. Majority of the retesting was performed on breast tissue (90%, n = 27). The biomarker change was noted in 11 out of 30 (36.7%) for ER, 12 out of 30 (40%) for PR, and 13 out of 30 (43.3%) for HER2 marker. Concordance analysis demonstrated that HER2 was the most prone to change (Cohen K = -0.003) from pre-NAST to post-NAST specimen, whereas ER (Cohen k = 0.26) and PR (Cohen k = 0.18) were less subject to change. Kappa assumes theoretical maximum value of 1 when pre- and post-NAST are in agreement, a negative value indicates a poor concordance. One third of patients with discordant biomarkers had additional adjuvant treatment. 8 out of 30 patients (26.7%) were started on adjuvant anti-HER2 therapy due to HER2 positivity on post-NAST tumors. After ER positivity was detected on residual BC tumors, 2 initial triple negative patients (6.7%) were initiated on adjuvant endocrine therapy. Conclusions: Biomarker discordance due to tumor heterogeneity is an increasingly recognized phenomenon. Given the prognostic and therapeutic implications of biomarker driven BC management, retesting biomarkers routinely after NAST on residual invasive tumor would be beneficial in this patient population.