Elucidating the immune active state of HR+HER2- MammaPrint High 2 early breast cancer.

506 Background: The 70-gene signature, MammaPrint (MP), classifies patients (pts) with early-stage breast cancer (EBC) as having an UltraLow, Low, High 1 (H1), or High 2 (H2) risk of distant recurrence. I-SPY 2 showed that pts with MP H2, HR+HER2- tumors have significantly higher response rates to neoadjuvant chemotherapy + immunotherapy relative to pts with MP H1 tumors. Expression-based immune deconvolution provides more detailed information about immune cell function beyond conventional methods enumerating tumor infiltrating lymphocytes. To elucidate the underlying biology that mediates immune therapy response, we performed in silico analysis of full transcriptome data to characterize immune cell frequencies and antigen presentation in HR+HER2-, MP High Risk EBC from pts enrolled in FLEX. Methods: The prospective, observational FLEX Study (NCT03053193) includes stage I-III pts with EBC who received MP testing and consented to full transcriptome and clinical data collection. Women with HR+HER2-, and MP High Risk invasive ductal EBC were included (n = 2916). Tumors were stratified into MP H1 or MP H2. The R-package Limma was used to preprocess gene expression data. The gene signature-based method xCell was used to determine immune cell abundances for each group based on enrichment score. Expression of genes involved in “antigen processing and presentation KEGG pathway”, and immune regulation, were evaluated between groups. Differences in immune cell frequency and gene expression were determined by a t-test with an adjusted p-value 10%. Nearly 30% of H2 tumors were Grade 1 or 2, whereas 22% of H1 tumors were Grade 3. H2 tumors had significantly higher frequency of antigen presenting cells (APCs) including activated dendritic cells and macrophages, CD4+ memory T cells, CD8+ T cells, memory B cells and plasma cells relative to H1 tumors. The genes expressing PD-1 and PDL-1 and genes involved in antigen processing, including B2Mand TAP1/2, and presentation, i.e., major histocompatibility (MHC) class I ( HLA-A, -B, -F) and class II molecules ( HLA-DM, -DQ), were significantly upregulated in H2 vs H1. Conclusions: H2 tumors exhibited a heighted immune active state compared to H1 tumors among HR+HER2- EBC pts. The presence of APCs and increased antigen presentation, which are critical in eliciting T- and B-cell activation may explain the improved response rates to immunotherapy observed in H2 tumors. Selecting pts based on grade or ER% alone may exclude pts likely to benefit from immune therapy. These data support ongoing trials evaluating the response and benefit of adding immunotherapy to standard of care treatment regimens in MP H2 tumors. Clinical trial information: NCT03053193 .