Abstract 1903: Application of lipidomics to a sulindac intervention of aromatase inhibitor-induced pain

Abstract

Abstract (a) Introduction. The regular use of non-steroidal anti-inflammatory drugs (NSAIDs) has been associated with a lower risk for epithelial cancers including breast. NSAIDs block cyclooxygenase (COX)-1 and -2 enzyme metabolism of Ω-6 polyunsaturated fatty acids (PUFA) to prostaglandins (PGs). Additionally, Ω-6 and Ω-3 PUFA are metabolized by lipoxygenases (LOX) and cytochrome P450 (CYP450) enzyme families to produce over 100 metabolites known as ‘oxylipins’. Oxylipins exhibit a wide spectrum of biological activity including mediators of pain and inflammation. The overarching objective of the work is to determine if sulindac (an NSAID) intervention changes oxylipin profiles in breast cancer patients, and if individual oxylipins are related to pain caused by aromatase inhibitors (AI). (b) Methods. This work takes place within the context of an R01-funded single arm, open-label clinical trial of sulindac (N = 100). Breast cancer patients that are stable on AIs complete pain and quality of life questionnaires, and provide blood and urine samples after a 4-week NSAID washout, after a 3 month observation period, at baseline and 3, 6 and 12 months on sulindac intervention (150 mg bid). Oxylipins were quantified in blood and urine (n = 10) at the end of washout, baseline, and after 3 months on sulindac using liquid chromatography (LC) mass spectrometry (MS)-based methods. (c) Results. We quantified 62 total oxylipins in plasma and 87 in urine. Thus far, our analysis has focused on plasma oxylipins. Wilcoxon signed-rank tests were used to compare oxylipin levels at baseline and after 3 months of sulindac. COX metabolites of both Ω-6 and Ω-3 PUFA were non-significantly decreased. In terms of Ω-6 metabolites, 7 LOX products were significantly decreased 11-HETE (P = 0.028), 15-HETE (P = 0.037), 9,12,13-TriHOME (P = 0.017), 9,10,13-TriHOME (P = 0.012), 8-HETE (P = 0.028), 9-HETE (P = 0.028), and 12-HETE (P = 0.047). Significantly decreased Ω-6 CYP450 metabolites were 15(S)-HETrE (P = 0.028), and 5,6-DiHETrE (P = 0.028). The Ω-3 metabolites that significantly decreased were the LOX-derived 5-HEPE (P = 0.022), and 15-HEPE (P = 0.028), and CYP450-derived 5,15-DiHETE (P = 0.047), 5,6-DiHETE (P = 0.009), and 15,16-DiHODE (P = 0.012). Thus far 53% of participants reported pain other than “every day types of pain” post-washout. The Ω-6 CYP450 metabolite 11(12)-EpETrE was significantly correlated with both “worst pain” (ρ = -0.669; P = 0.034) and severity (ρ = -0.778; P = 0.008). The Ω-6 CYP450 metabolite 20-HETE was significantly correlated with severity (ρ = -0.714; P = 0.047). (d) Conclusions. Our preliminary data indicate that oxylipin profiles change in response to sulindac and may be related to pain. Success in shifting plasma oxylipin profiles toward an anti-inflammatory/anti-thromobotic/cardioprotective profile would maximize efficacy and to reduce potential toxicities of NSAIDs within the context of for prevention of breast and other cancers. Citation Format: Jessica Anne Miller, Betsy Werthiem, Jun Yang, Bruce Hammock, Denise Roe, Alison Stopeck, Patricia A. Thompson. Application of lipidomics to a sulindac intervention of aromatase inhibitor-induced pain. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1903. doi:10.1158/1538-7445.AM2015-1903