Aim: The DETECT study program evaluates whether treatment efficacy in women with metastatic breast cancer (MBC) is increased by taking into account the molecular characteristics of circulating tumor cells (CTCs). Here, we present data on both prevalence and HER2 phenotype of CTCs in patients with HER2 negative MBC screened within the DETECT program. The aim of this study is to evaluate the rate of discordance in HER2 phenotype between primary tumor and CTCs and to analyze whether primary tumor and/or patient characteristics can predict discordance in HER2-status.

Methods: The number of CTCs in 7.5 ml of peripheral blood (using the FDA-cleared CellSearch® System; Janssen Diagnostics, LLC) and their HER2 status were evaluated. Patients were defined as having a positive HER2-status on CTCs if at least 1 CTC with a strong (+++) immunocytochemical HER2 staining intensity was found. To assess which factors predict discordance of HER2 phenotype between primary tumor and CTCs, we used a multivariate binary logistic regression model with backward selection procedure. Patient and primary tumor characteristics included as independent factors were patient age, time since primary diagnosis, tumor stage, nodal stage, grading, histological type, and hormone receptor status (HRS).

Results: 1123 women with HER2-negative MBC were screened for CTCs. Based on a cutoff of ≥ 1 CTC, 711 (63.3%) of 1123 screened patients were positive for CTCs, while 412 (36.7%) of the 1123 screened patients were categorized as CTC positive if a cutoff of ≥ 5 CTCs was used.

At least one HER2-positive CTC was found in 134 of the 711 HER2-negative MBC patients with one or more CTCs (median 2 HER2-positive CTCs, range 1 – 80), indicating a discordance between primary tumor and CTCs with regard to HER2-status in 18.8% of patients. If the analysis was restricted to the 412 patients with 5 or more CTCs, at least one HER2-positive CTC was found in 121 patients, resulting in 29.4% discordance rate.

A multivariate logistic regression with discordance in HER2 phenotype (yes/no) as binary response variable, including number of CTCs (1 – 4 CTCs vs. 5 or more CTCs) to account for the difference in discordance rate observed at different cutoff values for CTC positivity, showed that histological type (lobular vs. ductal, odds ratio OR 2.66, 95% confidence interval CI 1.62 – 4.37, p < 0.001), HRS (positive vs. negative, OR 2.89, 95% CI 1.16 – 7.19, p = 0.022) and CTC number (5 or more CTCs vs. 1 – 4 CTCs, OR 7.57, 95% CI 3.93 – 14.89, p < 0.001) significantly predicted discordance in HER2 phenotype between primary tumor and CTCs.

Conclusion: Our data revealed discordance in HER2 status between primary tumor and CTCs in 19% to 29% of patients with HER2 negative MBC. Discordance in HER2 status was predicted by histological type and HRS of the primary tumor, as well as by the number of CTCs detected. Individualized breast cancer treatment based on CTC phenotype is currently investigated in Phase III trials and not part of clinical routine yet. However, the knowledge of factors associated with discordance in HER2 status may be incorporated in today's clinical practice by guiding the decision process for performing a biopsy to characterize a metastatic relapse.

Citation Format: Janni W, Schramm A, Friedl TWP, Schochter F, Huober J, Rack B, Alunni-Fabbroni M, Fasching PA, Taran F-A, Hartkopf A, Schneeweiss A, Müller V, Aktas B, Krawczyk N, Pantel K, Fehm T. Predictors for discordance in HER2 phenotype between primary tumor and circulating tumor cells in women with metastatic breast cancer. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P2-02-02.