Abstract P1-08-06: Low tumor CD68 mRNA content (intratumoral macrophages) is predictive for benefit from adjuvant trastuzumab in HER2-positive breast cancer: An analysis of the FinHER trial

Abstract

Abstract Background: Tumor immune cell infiltrates influence prognosis of node negative breast cancer (BC), and intratumoral B-cells and T-cells are of importance for achieving response to chemotherapy in triple negative BC. The role of tumor infiltrating macrophages remains unclear, but they might promote tumorigenesis. We investigated the prognostic value of CD68 mRNA levels within the luminal, HER2-positive and triple negative subtypes in the FinHer trial patient population, and evaluated their predictive value on survival outcomes achieved with adjuvant trastuzumab and chemotherapy in early breast cancer. Methods: RNA was extracted from formalin-fixed paraffin-embedded (FFPE) tumor tissue of 917 (90.8%) patients out of the 1010 patients who participated in the FinHer trial. Intratumoral macrophage infiltration was assessed by measuring breast tumor CD68 mRNA content using RT-qPCR from representative FFPE tissue samples. Breast cancer molecular subtypes (luminal, HER2 positive and triple-negative) were approximated using immunohistochemistry (IHC) and central CISH testing data obtained from the FinHer trial datafile. Prognostic significance of CD68 on distant disease-free survival (DDFS) was assessed using Kaplan-Meyer analysis and log-rank test. Results: Tumor CD68 mRNA expression was normally distributed with median expression of 33.56 (dCt). CD68 mRNA levels correlated weakly with estrogen receptor (ER) mRNA expression and ER protein levels (r = 0.11 and r = 0.15, respectively; p<0.0001), but not with tumor HER2 mRNA level or IHC/CISH status (r = -0.06, p = 0.12; and r = 0.01, p = 0.92, respectively). The median tumor CD68 mRNA content was not prognostic for DDFS in the subset of luminal cancers (ER+, HER2-; n = 533, 5-year DDFS 88% versus 88%; p = 0.92), but a lower than the median CD68 level tended to be associated with favorable DDFS in triple negative cancer (n = 143; 81% vs. 70%; p = 0.11) and was significantly associated with favorable DDFS in HER2-positive cancer (n = 191; 86% vs. 67%; p = 0.001). In the subset of HER2-positive cancer with tumor CD68 mRNA expression level lower than the median, patients treated with trastuzumab benefitted from addition of trastuzumab to chemotherapy (5-year DDFS 93% with trastuzumab vs. 79% without trastuzumab; p = 0.02), whereas no benefit from trastuzumab was observed when tumor CD68 content was higher than the median (67% vs. 68%; p = 0.92). Conclusions: The study validates tumor CD68 concentration as a prognostic biomarker in HER2-positive early breast cancer. Patients with HER2-positive cancer and few tumor macrophages (low tumor CD68 mRNA content) benefitted from addition of trastuzumab to chemotherapy, whereas patients with HER2-positive BC with high tumor macrophage content derived no benefit from adjuvant trastuzumab. Trastuzumab may be effective only for macrophage-poor HER2-positive cancers that are prone to antibody-dependent cellular cytotoxicity (ADCC), whereas it may have little efficacy for cancers that progress despite or due to high intratumoral macrophage content that interferes with ADCC. Other agents, such as T-DM1, might work better than trastuzumab in the subset of women who have HER2-positive BC with a high tumor macrophage content. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-08-06.