Prediagnostic Serum Levels of Fatty Acid Metabolites and Risk of Ovarian Cancer in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial.

The participants for the study were predominantly white (92.3%) ( Table 1 ). Cases used menopausal hormonal therapy for longer durations (6+ years) than controls (39.5% in cases vs. 28.8% in controls). The median number of years from blood draw to the year of cancer diagnosis was 8.2 years [interquartile range (IQR): 5.2–10.8]. The age adjusted geometric means of the free fatty acid metabolites were similar between cases and control ( Table 2 ). Out of 31 metabolites analyzed for an association with ovarian cancer, five (8-HETE, 12,13-DHOME, 13-HODE, 9-HODE, and 9,12,13-THOME) were positively associated with ovarian cancer risk ( Table 3 ): 8-HETE [Tertile 3 vs. 1: OR 2.53 (95% CI 1.18–5.39), p-trend 0.02], 12,13-DHOME [2.49 (95% CI 1.29–4.81), 0.01], 13-HODE [2.47 (1.32–4.60), 0.005], 9-HODE [1.97 (1.06–3.68), 0.03], 9,12,13-THOME [2.25 (1.20–4.21), 0.01]. After accounting for multiple comparison, three metabolites (12,13-DHOME, 13-HODE, and 9,12,13-THOME) remained associated with ovarian cancer risk at FDR <0.10. Increased ovarian cancer risk with the five metabolites remained after adjusting for the potential mediating effects of other inflammatory markers (e.g. CRP, TNF-alpha, IL-8; Supplemental Table 2 ). In analyses by subtype ( Table 3 ), heterogeneity was suggested for 8-HETE [serous OR (95%CI): 2.53 (1.18–5.39) vs. non-serous OR (95%CI): 1.15 (0.56–2.36), p-het 0.10] and 12,13-EpOME [1.95 (0.90–4.22) vs. 0.82 (0.39–1.73), p-het 0.05]. In pathway analyses, the linoleic acid-COX derived [summary OR (95% CI), p-value: 0.95 (0.76,1.19), 0.004], linoleic acid-LOX derived [1.52 (1.22–1.90), <0.001], and linoleic acid-cytochrome P450 derived [(1.21 (1.04,1.41), 0.02] pathways were associated with increased risk of ovarian cancer ( Figure 2 ). In analyses by subtype ( Supplemental Figure 1 ), linoleic acid-COX derived [serous OR (95%CI), p-value: 1.38 (1.04–1.84),0.03 vs. non-serous OR (95%CI), p-value: 1.31(0.99–1.73), 0.06] and linoleic acid-LOX derived [1.49 (1.13–1.97), 0.005 vs. 1.54 (1.16–2.04), 0.003)] pathways were associated with increased risk of both serous and non-serous ovarian cancers. Arachidonic acid-LOX derived [1.31 (1.11–1.54), 0.001 vs. 1.11 (0.94–1.32), 0.22)] and arachidonic acid–cytochrome P450 derived [1.18 (1.02–1.36), 0.03 vs. 1.00 (0.86–1.15), 0.95)] pathways were associated with increased risk of serous ovarian cancers but not non-serous ovarian cancers. While the alpha-linoleic acid derived [0.95 (0.76–1.19), 0.66 vs. 1.27 (1.01–1.59), 0.04)] pathway was elevated for non-serous ovarian cancers but not for serous ovarian cancers. In sensitivity analyses, free fatty acid metabolites associations did not vary by time between blood draw and diagnosis ( Supplemental Table 3 ). The results were not significantly modified by BMI (normal vs. overweight/obese) ( Supplemental Table 4 ).

We conducted a nested case-control study within the screening arm of the PLCO Cancer Screening Trial, a randomized two-arm screening trial of men and women, ranging in age 55–74 years was conducted between 1993 and 2001 ( 15 ). Briefly, 78,216 women recruited from ten centers across the United States (Birmingham, AL; Denver, CO; Detroit, MI; Honolulu, HI; Marshfield, WI; Minneapolis, MN; Pittsburgh, PA; Salt Lake City, UT; St. Louis, MO; and Washington, DC). Participants in the screening arm provided a blood sample at baseline and during five follow-up medical examinations and were stored at −70 o C ( 16 ). All participants completed a self-administered baseline questionnaire. Cancer cases were identified by annual mailed questionnaires which were then confirmed by linking to population-based registries and the National Death Index. Medical and pathological records were obtained when possible. To avoid the possibility of reverse causality, cases for our study were 157 individuals diagnosed between two to fourteen years after blood collection with follow-up through December 2010. To ensure a relatively equal distribution of specimens between 2–14 years prior to diagnosis, 10.9% of samples selected were measured at baseline and the remaining at follow-up visits (16.3% year 1, 25.9% year 2, 13.7% year 3, and 33.2% year 4). For inclusion in our study, serum samples were selected from a pre-diagnostic blood draw given the availability of unthawed serum, consent to biochemical studies, completion of the baseline questionnaire and no history of cancer (except non-melanoma skin cancer). There were 160 cases identified out of which 3 were excluded from the study because the sample volume was not sufficient. Controls were women with no history of oophorectomy at the time of diagnosis of their matched case and were matched on: age at blood collection (55–59, 60–64, 65–69, 70+ years), race (white, black, other), study center, and time (a.m. and p.m.) and date (three-month categories) of blood collection. Serum specimens from a single visit were measured for each study subject. All the study participants provided written informed consent and the study received approval from the Institutional Review Board of the involved study centers and from the National Cancer Institute. Levels of 34 free fatty acid metabolites generated through 3 different pathways—COX pathway, cytochrome P450 pathway and LOX pathway—were measured in serum. All free fatty acid metabolites were separated by electrospray ionization liquid chromatography (LC, Agilent 1200 series capillary HPLC) and quantified using tandem mass spectrometry (MS/MS, MDS Sciex API 3000, negative ion mode) with a Turbolon Spray and scheduled multiple reaction. The details of the assay have been published previously ( 17 – 19 ). Concentrations (pg/ml) were quantified with Analyst software (v 1.5, Applied Biosystem) using metabolite and internal standard peaks for each sample. To evaluate assay performance, we included duplicate quality control (QC) samples in each batch and calculated the within-batch and between-batch coefficients of variation (CVs). We also calculated the intraclass correlation coefficient (ICC) for each analyte. 51.6% of CVs were <10%, 25.8% were 10–15%, and 22.5% were 15–20% ( Supplemental Table 1 ). We excluded from evaluation metabolite 5,6-EET given known limitations in its detection via LC-MS/MS, we further excluded two metabolites (19-HETE and 20-HETE) that were undetectable in nearly all samples. Thus, after exclusions, 31 metabolites were evaluated ( 17 , 20 , 21 ). We used generalized linear models adjusted for age to calculate geometric mean concentrations of free fatty acid metabolites by case/control status and by other factors (e.g., smokers vs. non-smokers, normal BMI vs. overweight/obese BMI, and aspirin/Ibuprofen users vs. non-users). P-values were calculated using a Wald test. We used unconditional logistic regression models to calculate odds ratios (OR) and 95% confidence intervals (CI) for the association between pre-diagnostic serum levels of free fatty acid metabolites and risk of ovarian cancer. All models were adjusted for matching factors and potential confounders: parity (nulliparous/parous), duration of oral contraceptive use (never, 1–5 years, 6+ years), duration of menopausal hormone therapy use (never, 1–5 years, 6+years), cigarette smoking status (never, former, current), and body mass index (BMI; <25, 25–29.9, 30+ kg/m 2 ). The risk of ovarian cancer was examined across tertile categories of the metabolites based on the distribution among controls. Metabolites with 90% or more individuals with non-detectable levels were categorized into two groups (detectable vs. undetectable). A Wald test using tertile categories as a continuous variable was used to evaluate trend. For markers associated with ovarian cancer risk, we further adjusted for the inflammatory markers (CRP, TNF-alpha, and IL-8) that were positively associated with ovarian cancer risk in a previous analysis ( 22 ). We also explored the association between free fatty acid metabolites and risk of ovarian cancer by histologic subtypes (serous vs. non-serous) and number of years from blood collection to cancer diagnosis (cancer diagnosis 2–5 years vs. > 5 years from blood draw). P-values for heterogeneity were based on the Wald test for the ordinal metabolite variable from a case-only model with serous histology or the shortest time (2–5 years) between blood draw and diagnosis as the reference group. We used fixed effect meta-analysis to calculate summary ORs of the association of fatty acid metabolite pathways ( Figure 1 ) with the risk of ovarian cancer overall and by histologic subtypes. We evaluated effect modification by BMI (normal vs overweight/obese) and use of medication (aspirin/Ibuprofen users vs. aspirin/Ibuprofen non-users) using stratified models. Likelihood ratio tests were calculated to test for interaction using the cross-product term. Due to a limited number of current smokers, we could not evaluate effect modification by smoking status, but we conducted a sensitivity analysis restricted to never/former smokers. As a secondary analysis, we evaluated the association between free fatty acid metabolites and the risk of ovarian cancer adjusting for aspirin use. To account for multiple comparisons, we applied the false discovery rate (FDR) for the primary free fatty acid metabolites associations with the ovarian cancer. All other analyses were considered exploratory and not corrected for multiple comparisons. All tests were two-sided and statistical significance was defined using a p-value <0.05. Analyses were conducted in SAS version 9.3 (SAS Inc, Cary, NC).

In this prospective study exploring the association between free fatty acids and ovarian cancer, we report that five out of 31 metabolites (8-HETE, 12,13-DHOME, 13-HODE, 9-HODE, and 9,12,13-THOME) were associated with increased ovarian cancer risk. 8-HETE is produced by the metabolism of arachidonic acid and four metabolites: 12,13-DHOME, 13-HODE, 9-HODE, and 9,12,13-THOME, are derived from linoleic acid oxidation. Further supporting the individual metabolite associations, we observed increased ovarian cancer risk with the linoleic acid pathways based on meta-analysis. Free fatty acid metabolism by the LOX pathway produces 8-HETE, 13-HODE, 9-HODE, and 9,12,13-THOME. Although the role of the LOX pathway in ovarian cancer has not been extensively evaluated, few studies support a pivotal role of the LOX pathway in ovarian cancer prognosis. Specifically, a tissue microarray study of 245 paraffin-embedded epithelial ovarian cancer samples demonstrated that strong expression of LOX receptors was indicative of worse ovarian cancer prognosis ( 23 ). An immunohistochemical analysis comparing expression of 12-LOX in a serous ovarian cancer cell line and normal ovarian epithelium demonstrated that expression of 12-LOX was higher in serous ovarian carcinoma compared to normal ovarian epithelium and regulated the cell growth through ERK and MAPK signaling ( 24 ). The LOX pathway metabolites (8-HETE ,13-HODE, 9-HODE, and 9,12,13-THOME) are also known as peroxisome proliferator-activated receptors (PPAR-α/γ) ligands, a lipid-activated transcription factor. PPAR-α promotes lipid uptake ( 25 ) whereas PPAR-γ is mostly implicated in lipid storage and adipocyte differentiation ( 26 ). Experimental data suggests that PPAR isotypes have a role in tumor suppression and/or progression, but the mechanism is unclear ( 22 – 24 ). Therefore, the metabolites produced by the LOX pathway may have a role in ovarian carcinogenesis via PPAR α/γ activation. Linoleic acid oxidation by cytochrome P450 monoxygenase produces 12,13-DHOME (isoleukotoxin diol). A study of ovarian tumors (n=167 cases) reported overexpression of the cytochrome P450 allelic variant ( CYP1B1 ) ( 27 ). Several studies have demonstrated the role of 12,13-DHOME in suppression of the immune system, increased adipocyte differentiation, cell proliferation, and apoptosis ( 28 , 29 ), and as a ligand for PPAR-γ ( 30 ). The role of 12,13-DHOME in several cellular functions might contribute to the increased risk of ovarian cancer. Furthermore, the positive association of 12,13-DHOME with risk of ovarian cancer may be related to increased production of 12,13-DHOME by CYP1B1 . To our knowledge no prior study has explored the association between pre-diagnostic levels of circulating fatty acid metabolites from pathways involving COX, LOX, and cytochrome P450 enzymes and ovarian cancer risk. The strengths of our study include a well-designed study nested within the large prospective PLCO study, with detailed information on established ovarian cancer risk factors enabling careful control for confounding and multiple comparisons. All the analytes were measured by sensitive tandem mass spectrometry. We matched cases and controls on time of day and date of blood collection to minimize any effects due to variability in storage conditions. To limit the impact of reverse causation, we excluded cases diagnosed between within 2 years of blood draw. Our study also included limitations. Although we included all available cases, the number of cases included in the current study was limited. There is a possibility that the single time point measurement of these metabolites may not reflect their concentration over a longer period. To our knowledge, there are no studies on temporal variability of serum concentrations of metabolites within an individual but a systematic review looking at the change in tissue arachidonic acid after the consumption of western diet showed that decreasing the intake of linoleic acid had no effect on circulating levels of arachidonic acid ( 31 ). Thus, validation in an independent cohort is needed. In conclusion, our data provide evidence that out of 31 fatty acid metabolites studied, five metabolites (8-HETE, 12,13-DHOME, 13-HODE, 9-HODE, and 9,12,13-THOME), generated via the LOX/cytochrome P450 pathway linoleic acid metabolite pathway, were associated with the increased risk of ovarian cancer. In pathway analysis, ovarian cancer risk was associated with the all three linoleic acid pathways (COX derived, LOX derived, and cytochrome P450 derived). As inflammation is a well-established risk factor for ovarian cancer, these precursors of inflammatory markers associated with ovarian cancer might be important to understand potential mechanism of the etiology of ovarian cancer.

Inflammation is thought to play a role in the development and progression of ovarian cancer. Usually inflammation is beneficial such that it activates the immune process and protects the body from infections and diseases, but chronic inflammation induces prolonged exposure of the cells to the mediators of inflammation ( 1 , 2 ). These mediators can then promote tumorigenesis. Inflammation during ovulation or inflammatory chronic disease that affects the fallopian tubes and/or ovary (e.g., endometriosis and pelvic inflammatory disease) is associated with increased risk of ovarian cancer ( 3 – 5 ). The use of nonsteroidal anti-inflammatory drugs (NSAIDs), such as aspirin, is associated with reductions in risk ( 6 ). Though a specific mechanism is yet to be determined, reduced risk of ovarian cancer due to aspirin might be due to the medication’s anti-inflammatory effect. Aspirin and other NSAIDs block the synthesis of proinflammatory prostaglandin synthesis by inhibiting the cyclooxygenase (COX) enzyme ( 7 ). In addition to prostaglandin synthesis, the inflammatory response involves biologically active lipid molecules, arachidonic acid and an essential fatty acid linoleic acid and its metabolites ( Figure 1 ) ( 8 – 10 ). Essential fatty acids mainly comprise 2 groups: omega-6 fatty acids (linoleic acid) and omega-3 fatty acids (alpha-linoleic acid). Omega-6 fatty acids are converted to arachidonic acid, and omega-3 fatty acids are converted to docosahexaenoic acid and eventually to arachidonic acid ( 11 , 12 ). In addition to essential fatty acid metabolism to arachidonic acid, arachidonic acid is also released from the phospholipids of the cell membrane during inflammation in response to various stimuli like cytokines, hormones, and stress ( 7 ). Arachidonic acid is acted upon by three enzymes: COX, lipoxygenase (LOX), and cytochrome P450. The COX activity on arachidonic acid produces PGG2 and PGH2 which is later converted to prostanoids including prostaglandins (PGD2, PGE2, and PGF2a), prostacyclins (PG12), and thromboxanes (TXA2 and TXB2). COX exists in 2 isoforms, COX-1 and COX-2. Three forms of lipoxygenases (5-LOX, 12-LOX, and 15-LOX) acts on arachidonic acid to produce hydroxyperoxyeicosatetraenoic acids (HPETE), HETEs, and leukotrienes (LT). The enzyme cytochrome P450 produces HETEs and epoxides. Free fatty acid (arachidonic acid/linoleic acid/alpha-linoleic acid) metabolites are implicated in various signaling pathways involved in physiological processes like cell differentiation, cell migration, platelet aggregation, angiogenesis, and regulation of immune functions ( 8 , 13 ), and thus may play a role in cancer initiation and/or promotion. Due to a potentially tumorigenic role of arachidonic acid, the inhibitors of arachidonic acid metabolites have been widely studied for their potential role in cancer treatment but the precise molecular mechanism by which these metabolites drive tumorigenesis is unknown ( 14 ). Epidemiological studies investigating the role of free fatty acid metabolites in ovarian cancer are limited. To understand the role of arachidonic acid and linoleic acid pathway-induced inflammation in the etiology of ovarian cancer, we assessed several free fatty acid metabolites in a nested case-control study within the Prostate, Lung, Colorectal, and Ovarian (PLCO) cancer screening trial. As ovarian cancer subtypes are known to have a heterogeneous etiology, we further evaluated associations by serous/non-serous subtype and time between blood draw and cancer diagnosis.