Gastric cancer is the third most-common cancer and the second leading cause of cancer-related death in the world (1,2). D2 gastrectomy combined with chemotherapy is a standard treatment for gastric cancer; however, almost half of the patients had recurrence, even after modern multidisciplinary treatment (3–5). To further improve the patient's survival, approaches than conventional chemotherapy should be considered.

The administration of eicosapentaenoic acid (EPA), a long-chain polyunsaturated fatty acid of the omega-3 (n-3) family, may be an attractive approach (6,7). Many previous studies have clarified that EPA could be selectively cytotoxic to various tumor cells in vitro and in vivo (8–10). Moreover, EPA was reported to suppress pro-inflammatory cytokines, such as interleukin (IL)-1 and tumor necrosis factors (TNF), which is released by surgical stress (11,12). Recent studies clarified that pro-inflammatory cytokines play a critical roles in tumors (e.g., survival, proliferation, metastasis, and resistance to chemotherapy) (13,14). Thus, the perioperative administration of EPA could improve the prognosis of gastric cancer patients.

However, there is little evidence to support that EPA is associated with a clear survival benefit in cancer patients (15–17). Cockbain et al examined the efficacy of EPA in a small randomized study targeting colorectal cancer patients with liver metastasis who had undergone liver surgery (15). In their study, the EPA group showed better survival than a placebo group; however, the difference was not statistically significant. In the multivariate analysis, they only showed that EPA supplementation was an independent prognostic factor for overall survival (P=0.05). Thus, an apparent clinical benefit of EPA was not demonstrated. Moreover, the efficacy of EPA on long-term outcomes in other human malignancies has not been examined.

Previously, we conducted a prospective multicenter randomized trial to evaluate whether perioperative EPA-enriched nutritional supplementation can improve the short-term and long-term outcomes of patients with localized gastric cancer who require total gastrectomy as curative treatment (18,19). In the primary analysis, we evaluated whether body weight loss caused by surgical stress and gastrectomy was prevented by EPA supplementation. Although perioperative EPA can be safely administered, postoperative weight loss was not prevented. We herein report the long-term oncological outcomes, which reflect antitumor activity, a key secondary endpoint, in this randomized study.

The aim of the present analysis was to explore whether perioperative eicosapentaenoic acid (EPA) supplementation could improve the survival of patients with localized gastric cancer that required curative total gastrectomy. This is the first randomized study to demonstrate a prognostic effect of EPA in patients with gastric cancer. The major finding of this study was that a clear survival benefit of perioperative EPA was not observed in gastric cancer patients.

Previously, Cockbain et al showed that EPA therapy had a marginal survival in patients with liver metastasis of colorectal cancer (15). There were some differences between the present study and the study by Cockbain et al. The first difference was the median cumulative dose of EPA, which was 38.5 g in this study but 60 g (calculated from their results) in their study. Thus, the cumulative EPA dose of their study was higher than that of the present study. Second, the tumor stage was different. In the present study, patients with localized tumors were targeted, while their study targeted patients with metastatic tumors. In addition to the marginal survival benefit in their study, they also demonstrated that EPA had anti-angiogenesis effects using biopsy specimens. Angiogenesis plays an important role in metastatic tumors but would not play an important role in micro-metastatic tumor cells. Several pivotal phase III studies demonstrated that the patient survival was significantly improved by bevacizumab and ramucirumab (vascular endothelial growth factor inhibitors) in metastatic colorectal cancer and by ramucirumab in metastatic gastric cancer (24–26). On the other hand, the efficacy of bevacizumab was not confirmed in the phase III trials, not only for resectable gastric cancer but also for colorectal cancer (27,28). Whether the target is micrometastatic or metastatic disease might affect the efficacy of EPA.

Although the OS rates were similar between the two groups, the hazard ratio was slightly lower than 1.0 for OS, which suggests some clinical efficacy of EPA in gastric cancer. In the subgroup analyses for OS, the hazard ratio was 0.39 for OS in patients who received NAC and 0.57 in patients who had nodal metastasis. In the present cohort, patients who received NAC were limited to patients with cT3 in the case of junctional cancer, scirrhous type, or giant type 3, cT4, para-aortic nodal metastasis, and/or bulky nodal metastasis around the major branched artery. Thus, patients showing a low hazard ratio were considered to have a relatively poor prognosis. It would be interesting to investigate whether EPA is effective for such unfavorable advanced gastric cancer in a future study. On the other hand, the hazard ratio was almost 1.0 in patients who did not receive the NAC and in patients who had no nodal metastasis. EPA would not be effective for these patients.

The present study was associated with some limitations. First, the primary endpoint of the present randomized study was not survival. The sample size was relatively small and was not set to investigate differences in survival. Thus, the results were not confirmatory. Second, EPA could be included in the dietary supplements. The clinical trial using the dietary supplements had bias that the patients buy the corresponding dietary supplements outside of the clinical trial without letting their physicians know. Thus, the negative results may be the result of a bias that the placebo group took the dietary supplement without permission and thereby falsified the end results. We con not exclude that their negative outcome is the result of such a bias. Third, we used oral nutritional supplementation including EPA in this study. The differences between the two groups were not limited to EPA, there were also differences in other nutrients. Thus, we cannot deny the effects of other nutrients.

In conclusion, the present study could not demonstrate a clear survival benefit of EPA supplementation in patients who received curative gastrectomy for gastric cancer. The value of EPA should be further tested in a future study by selecting patients with unfavorable advanced gastric cancer.