Background: Endocannabinoids (ECs) are one type of bioactive endogenous neuroinflammatory mediator derived from polyunsaturated fatty acids (PUFAs), which may regulate the emotional processes. Here, we assessed the effect of ω–3 PUFAs on EC levels, which may be the novel targets for the ω–3 PUFAs’ antidepressive effects. Methods: We conducted a 12-week double-blind, nonplacebo, randomized controlled trial. Eighty-eight major depressive disorder (MDD) participants were randomly assigned to receive eicosapentaenoic acid (EPA, 3.0 g/day), docosahexaenoic acid (DHA, 1.4 g/day), or a combination of EPA (1.5 g/d) and DHA (0.7 g/day). Eighty-five participants completed the trial, and their clinical remission and plasma PUFA-derived EC levels (pmol/mL) were measured. Results: The cumulative rates of clinical remission were significantly higher in the EPA and EPA + DHA groups than the DHA group (51.85 and 53.84 vs. 34.37%; p =0.027 and p =0.024, respectively). EPA and EPA + DHA treatments increased the eicosapentaenoylethanolamide (EPEA) levels compared to DHA treatment (0.33 ± 0.18 and 0.35 ± 0.24 vs. 0.08 ± 0.12; p =0.002 and p =0.001, respectively), while EPA + DHA treatment increased the docosahexaenoylethanolamide levels more than EPA treatment (1.34 ± 2.09 vs. 0.01 ± 1.79; p =0.006). Plasma EPEA levels were positively correlated with rates of clinical remission (hazard ratio: 1.60, 95% confidence interval: 1.08–2.39). Conclusions: Treatments enriched with EPA increased plasma EPEA levels, which was positively associated with clinical remission. This finding may suggest that levels of plasma EPEA play a potential novel endogenous therapeutic target in MDD.