A 3-year randomized clinical trial found that omega-3 polyunsaturated fatty acid treatment did not significantly reduce white matter lesion (WML) progression or neuronal integrity breakdown in older adults at risk for dementia. However, the treatment showed potential benefits for carriers of the apolipoprotein E ε4 allele (APOE*E4).

The study, published in JAMA Network Open, included 102 participants aged 75 years and older with existing WMLs measuring ≥5 cm³ and suboptimal omega-3 status. The study population included 62 women (60.8%) and had a mean age of 81 years. 

The trial was conducted at Oregon Health & Science University and enrolled participants with plasma omega-3 polyunsaturated fatty acid (PUFA) <5.5 weight percentage of total fatty acids. Exclusion criteria included dementia, defined as Clinical Dementia Rating >0.5 or Mini-Mental State Examination (MMSE) score ≤23.

Participants were randomized to receive either 1.65 g of omega-3 PUFA (975 mg eicosapentaenoic acid [EPA] and 650 mg docosahexaenoic acid [DHA]) or a soybean oil placebo daily for 3 years.

Brain MRI scans were performed at baseline and annually for 3 years using a 3-T MRI instrument (Siemens Total Imaging Matrix Trio, upgraded to Prisma during the study). Sequences included T1 magnetization prepared rapid gradient echo (MPRAGE), fluid-attenuated inversion recovery (FLAIR), and diffusion tensor imaging (DTI). FLAIR parameters included repetition time 6000 ms, echo time 388 ms, slice thickness 1 mm, and native resolution 0.5 × 0.5 × 1 mm.

Primary outcome measures included WML progression and DTI diffusion of fractional anisotropy (DTI-FA). Secondary outcomes included medial temporal lobe gray matter volume, total brain gray matter volume, and cognitive function. Linear mixed-effects models were used to estimate mean annual change between groups, with covariates including baseline age, sex, history of vascular disease, history of hypertension, history of depression, and MRI system used.

Results

Twenty-eight participants (28%) were APOE*E4 carriers. The mean MMSE score was 28.0 (standard deviation [SD] = 1.7), and the mean Montreal Cognitive Assessment score was 24.0 (SD = 3.2). In the omega-3 group, 76.4% of patients had a history of hypertension and 15.7% had a history of vascular disease, compared to 68.6% and 13.7% in the placebo group, respectively.

Mean baseline plasma omega-3 PUFA (EPA + DHA) levels were 7.135 (SD = 2.448) mg/dL in the omega-3 group and 6.662 (SD = 2.176) mg/dL in the placebo group.

Annual WML accumulation was 1.19 cm³ (95% confidence interval [CI], 0.64-1.74 cm3) in the omega-3 group vs 1.34 cm³ (95% CI, 0.80-1.88 cm3) in the placebo group (P = .30).

Annual decline in DTI-FA, representing neuronal integrity breakdown, was –0.0014 mm2/s (95% CI, –0.0027 to –0.0002 mm²/s) in the omega-3 group vs –0.0027 mm2/s (95% CI, –0.0041 to –0.0014 mm²/s) in the placebo group (P = .07).

Among APOE*E4 carriers, annual DTI-FA decline was significantly lower in the omega-3 group (–0.0016 mm²/s, 95% CI, –0.0032 to 0.0020 mm²/s) compared to the placebo group (–0.0047 mm²/s, 95% CI, –0.0067 to –0.0025 mm²/s; P = .04).

While the omega-3 group showed less annual WML accumulation and DTI-FA decline compared to placebo, these differences did not reach statistical significance in the overall population. However, APOE*E4 carriers treated with omega-3 demonstrated significantly less neuronal integrity breakdown.

No significant differences were observed between groups in medial temporal lobe atrophy (–0.37 cm³ [95% CI, –0.48 to 0.25] in omega-3 vs -0.36 cm³ [95% CI, –0.47 to 0.24] in placebo; P = .90) or total brain volume change.

Exploratory outcomes revealed:

• An annual periventricular WML increase of 0.09 cm³ (95% CI, 0.05–0.13 cm³) in the omega-3 group vs 0.11 cm³ (95% CI, 0.07–0.15 cm³) in the placebo group (P = .28)
• An annual subcortical WML increase of 0.03 cm³ (95% CI, –0.04 to 0.09 cm³) in the omega-3 group vs 0.04 cm³ (95% CI, –0.02 to 0.11 cm³) in the placebo group (P = .72)
• An annual DTI radial diffusivity increase of 3.143 × 10-6 mm²/s (95% CI, 1.449 × 10-6 to 4.836 × 10-6 mm²/s) in the omega-3 group vs 4.894 × 10-6 mm²/s (95% CI, 3.280 × 10-6 to 6.500 × 10-6 mm²/s) in the placebo group (P = .06)
• An annual DTI mean diffusivity increase of 2.882 × 10-6 mm²/s (95% CI, 1.377 × 10-6 to 4.386 × 10-6 mm²/s) in the omega-3 group vs 4.174 × 10-6 mm²/s (95% CI, 2.736 × 10-6 to 5.602 × 10-6 mm²/s) in the placebo group (P = .10).

In APOE*E4 carriers, annual WML increase was 0.04 cm³ (95% CI, –0.04 to 0.12 cm³) in the omega-3 group vs 0.11 cm³ (95% CI, 0.04–0.17 cm³) in the placebo group (P = .15). For noncarriers, the increase was 0.09 cm³ (95% CI, 0.05 to 0.13 cm³) in the omega-3 group vs 0.10 cm³ (95% CI, 0.05 to 0.14 cm³) in the placebo group (P = .79).

Safety and Adherence

Adverse events were similar between groups. Serious adverse events occurred in 16 participants (31.4%) in each group. The most common adverse events were injurious falls (30.3%), musculoskeletal disorders (25.4%), and gastrointestinal disorders (24.5%).

Hospitalization prevalence was 25.4% (26 participants) over the course of the study (12 in the omega-3 group, 14 in the placebo group). There were five deaths in total (four in the omega-3 group, one in the placebo group). Dropout rates due to adverse events were 11.7% (six participants) in the omega-3 group vs 5.9% (three participants) in the placebo group (P = .49).

The annual attrition rate was 8.5%, with no significant difference between groups (P = .32). Cumulative attrition was 11.7% by the end of year 1, 20.5% by the end of year 2, and 23.5% by the end of year 3. Pill counts measuring less than 80% adherent triggered further contact to resolve compliance issues.

The study authors noted several limitations, including a demographically and geographically homogeneous population, potential underpowering due to less-than-anticipated WML accumulation, and lack of power to directly compare effects between APOE*E4 carriers and noncarriers.

The study authors concluded, "In this study, omega-3 treatment was safe and well-tolerated but failed to reach significant reduction in WML progression and neuronal integrity breakdown among all participants at risk for dementia. However, the significant reduction in neuronal integrity breakdown among APOE*E4 carriers suggests that the effects of omega-3 may be amplified for these individuals. These results will enable improved study design and sample size calculations for future efforts of a relatively cheap, safe, and well-tolerated therapy for primary and secondary dementia prevention."

Conflict of interest disclosures can be found in the study.