Plasma omega-3 polyunsaturated fatty acids and recurrence of endometrial cancer

Abstract Background Omega-3 polyunsaturated fatty acids (PUFAs) were proposed to have potential effects against inflammation and cancer. However, results from epidemiology studies remain inconsistent. We aimed to explore the associations of plasma PUFAs with EC recurrence and all-cause mortality. Method Women diagnosed with endometrial cancer (EC) between 2008 and 2013 and underwent surgery at Fudan University Shanghai Cancer Center of China were recruited. Survival status was followed up through September 2017. EC recurrence and total cause deaths were identified through medical record and telephone interview. In total, 202 patients with enough plasma samples at time of surgery were included. There were 195 patients who provided baseline plasma and survival information included in the current study. Plasma omega-3 PUFAs were measured by GC-FID. Cox Proportional Hazard model adjusted for potential cofounders was used to estimate HRs and 95% CIs. Results Median follow-up time for patients was 58 months after surgery. A total of 13 recurrences and 11 all-cause deaths, of which, 2 deaths from EC, were identified. Level of plasma EPA was higher in recurrent patients than total patients (0.78% vs 0.51%, P = 0.015). Higher plasma eicosapentaenoic acid (EPA) level trended to have positive association with EC recurrence (P-trend = 0.04), although comparing to the lowest tertile, the highest tertile of EPA level was not significantly associated with increased risk of EC recurrence (HRT3vsT1 = 6.02; 95%CI = 0.7–52.06). The association between total omega-3 PUFA and EC recurrence tended to be stronger among patients with deeper myometrial invasion (OR = 3.41; 95%CI = 1.06–10.95; P-interaction = 0.04). Conclusions Higher plasma EPA level was significantly associated with EC recurrence. Further studies are warranted to confirm these findings. Trial registration ChiCTR1900025418 ; Retrospectively registered (26 August 2019); Chinses Clinical Trial Registry.