Use of non-steroidal anti-inflammatory drugs, acetaminophen and somatic immune profiles in breast tumors

Background. Tumor-infiltrating lymphocytes (TILs) represent a prognostic biomarker in several cancers, including breast cancer. Enrichment of TILs may help to identify highly immunogenic and immune-vulnerable tumors that may be more responsive to immune-mediated mechanisms and are associated with improved outcomes. Recent data suggest that the use of acetaminophen may have immunosuppressive effects. In this study, we investigated the relationship between the use of common analgesics and immune profiles in breast tumors from patients within a population-based prospective cohort study. Methods. This analysis was conducted among females with invasive breast cancer enrolled in the Cancer Prevention Study 3 (2006-2013). Use of non-steroidal anti-inflammatory drugs (NSAIDs), and acetaminophen was assessed at enrollment and follow-up surveys (2015 and 2019). Analgesic information came from the survey prior to, but closest in time to, their cancer diagnosis. Eligible patients were required to have formalin-fixed paraffin-embedded breast tumor samples available. Immune profiles were generated from tumor RNA using a previously published method to derive ten immune cell-type-specific scores plus a cytotoxic cell score from breast tumors. Linear regression was used to estimate the associations between each type of analgesic use and the immune scores, adjusting for age at diagnosis, stage, and estrogen receptor (ER) status. Multiple testing was addressed using the false discovery rate (FDR), with FDR<0.05 considered statistically significant. Results. This analysis included 887 breast cancer patients with a median age at diagnosis of 58 years (interquartile range: 51, 63). At the time of diagnosis, most of the women were post-menopausal (61%), had ER positive tumors (80%) and localized disease (65%). Among the breast cancer survivors included in the analysis, 450 (51%) reported NSAID use and 137 (15%) acetaminophen use. Patients who reported use of NSAID had similar immune profiles to non-users (FDR>0.05). However, acetaminophen users had lower scores for T-cells, T helper cells, Treg cells, natural killer (NK) cells, neutrophils, eosinophils and cytotoxic T cells compared with non-users (FDR<0.05). Results among patients with ER-positive tumors were similar to the overall results with the exception that the CD8+ T cell score was also significantly lower among acetaminophen users (FDR<0.05). Among women with ER-negative tumors, there was no difference in immune profiles between NSAID users and non-users. Only NK cell, neutrophil, and eosinophil scores were significantly lower among acetaminophen users compared with non-users (FDR<0.05). Conclusions. NSAID use does not appear to influence the breast tumor immune microenvironment. In contrast, we observed that acetaminophen use lowers the scores of multiple immune cell types, including those related to improved prognosis.