Background & aims: Early-onset dementia (EOD, defined as diagnosis < age 65) imposes a high socio-economic burden. It is less prevalent and less investigated than late-onset dementia (LOD). Observational data indicate that many EOD cases are associated with potentially modifiable risk factors, yet the relationship between diet and EOD has been under-explored. Omega-3 fatty acids are promising dietary factors for dementia prevention; however, existing research has primarily focused on cohorts aged >65. We examined the associations between omega-3 blood levels (which objectively reflect dietary intake) and incident EOD by leveraging data from the UK Biobank cohort.

Methods: We included participants aged 40-64, free of dementia at baseline and for whom plasma omega-3 levels and relevant covariates were available. We modeled the relationships between the three omega-3 exposures (total omega-3, DHA, and non-DHA omega-3) and incident EOD with quintiles (Q) and continuous linear relationships. We constructed Cox proportional hazards adjusting for sex, age at baseline and APOE-ε4 allele load, besides other lifestyle variables reported to relate to incident EOD. We also assessed the interaction between each exposure of interest and APOE-ε4 allele load.

Results: The study included 217,122 participants. During the mean follow-up of 8.3 years, 325 incident EOD cases were ascertained. Compared to participants at Q1 of total omega-3, those at Q4 and Q5 showed a statistically significantly lower risk of EOD (Q4, hazard ratio [95 % confidence interval] = 0.62 [0.43, 0.89]; Q5, 0.60 [0.42, 0.86]). A statistically significant inverse association was also observed for total omega-3 as a continuous variable. Compared to participants at Q1 of DHA, those at Q5 of non-DHA showed a significant lower risk of EOD. A statistically significant lower risk was observed in Q3, Q4 and Q5 of non-DHA omega-3. Finally, we observed no evidence of interaction omega-3 × APOE-ε4 allele load.

Conclusions: This study expands the evidence of a beneficial association of omega-3 and LOD to EOD as well. These findings suggest that an increased intake of omega-3 fatty acids earlier in life may slow the development of EOD. Additional research is needed to confirm our findings, particularly in more diverse populations.