Background
In the LILAC pilot trial, 12 weeks of metformin reduced colon CD4+ T-cell infiltration, mTOR activation, and potentially HIV transcription in a subset of ART-suppressed PWH. We investigated whether long-term metformin could affect HIV reservoirs in non-diabetic PWH.
Methods
The MetforAging trial (EudraCT: 2021-003299-15) is a 96-week, randomized, double-blind, placebo-controlled study designed to assess the effects of metformin on epigenetic aging in non-diabetic, ART-suppressed PWH ≥50 years. Participants had CD4 >500 cells/µL, HOMA-IR ≤2.6, VL <50 copies/mL for ≥1 year, and stable ART for ≥3 months. Participants were randomized (1:1) to receive metformin (850 mg BID; 850 mg QD if on dolutegravir) or placebo for 96 weeks. In this substudy, CD4+ T cells isolated at baseline and week 96 were used to quantify total, intact, and defective HIV proviruses and unspliced cell-associated HIV RNA (CA-RNA). Wilcoxon matched-pairs signed-rank test was used for within-group comparisons, and Mann-Whitney test for between-group change differences (week 96 – baseline).
Results
Among 40 participants, 35 (17 metformin, 18 placebo) completed 96 weeks of treatment. All participants were on ART for ≥ 8 years. At week 96, VL was <50 copies/ml. In the metformin arm, total, intact and defective HIV DNA showed no significant differences over time. Placebo recipients showed a significant decline in total HIV DNA (median change [Week 96 – Baseline]: -68.2 copies/106 cells; p=0.024) and in 5′ defective proviruses (-32.6 copies/106 cells; p=0.034). Between-group comparisons showed no differences for total, intact and 3´defective HIV proviruses, while 5′ defective proviruses had significantly greater decline in the placebo arm (metformin vs placebo median change [IQR]: 29.8 [-19.6‒119.6] vs -32.6 [-135‒3.2] copies/106 cells; p=0.027). CA-RNA significantly decreased in both arms (median change: metformin -72.8 [-379.4 to -0.4], p=0.048; placebo -299.5 [-711.4‒33.8] copies/106 cells, p=0.011). CA-RNA/total HIV DNA ratio also decreased in both arms, although the reduction was not statistically significant (median change: metformin -0.07 [-0.5‒0.01], p=0.083; placebo -0.31 [-0.7‒0.02], p=0.051). Between-group comparisons showed no significant differences for CA-RNA or the CA-RNA/total HIV DNA ratio.
Conclusions
Over 2 years, metformin did not alter the size or transcriptional activity of the HIV reservoir, with reservoir dynamics comparable between metformin and placebo in non-diabetic ART-treated PWH ≥ 50 years.