Effect of the vaginal live biotherapeutic LACTIN-V (Lactobacillus crispatus CTV-05) on vaginal microbiota and genital tract inflammation among women at high risk of HIV acquisition in South Africa: a phase 2, randomised, placebo-controlled trial.

Background: Absence of vaginal lactobacilli and accompanying genital inflammation is associated with HIV acquisition. We aimed to assess how a vaginal live biotherapeutic containing Lactobacillus crispatus affects cervicovaginal microbiota and markers of HIV susceptibility in South African women.

Methods: This randomised, placebo-controlled, phase 2 trial evaluated LACTIN-V (L crispatus CTV-05), a vaginal live biotherapeutic, compared with placebo in cisgender women in South Africa, aged 18-23 years, recruited at a community-based research clinic. Eligible participants with a Nugent score of 4-10 (indicating intermediate vaginal microbiota or bacterial vaginosis) completed 7 days of oral metronidazole and were randomly assigned (2:1) to LACTIN-V (2 × 10⁹ colony forming units per dose) or placebo (the substrate alone) via an independently generated randomisation sequence. Pharmacists, participants, and investigators were masked to treatment assignment. The study product (or placebo) was dosed daily for 5 days in week 1, then twice per week for an additional 3 weeks. Adverse events were evaluated 4 weeks and 8 weeks after starting the study product. Vaginal swabs (for 16S rRNA sequencing of the vaginal microbiome) and cervicovaginal lavage (for Luminex analysis of immune markers) were collected before metronidazole treatment, before study product (or placebo) administration, and at the week 4 and week 8 follow-up visits. An endocervical cytobrush for flow cytometry analysis of immune cell populations (including CD3⁺CD4⁺ T cells, and presence of CCR5 and the activation markers CD38 or HLA-DR) was collected before study product use and at 4 weeks and 8 weeks after study product use. The coprimary outcomes for the trial were (1) safety and acceptability of LACTIN-V, as measured by number of adverse events and a validated questionnaire; (2) presence of a Lactobacillus-dominant vaginal microbial community by 16S rRNA gene sequencing at week 4 and week 8; and (3) comparison of change in genital tract inflammatory markers from before metronidazole treatment to week 4 and week 8 between groups. Safety analyses were done in the intention-to-treat population and efficacy analyses in a modified intent-to-treat population (ie, excluding one person assigned placebo who erroneously received LACTIN-V). This trial is completed and registered on ClinicalTrials.gov (NCT05022212).

Findings: 45 Black South African women were randomly assigned to receive LACTIN-V (n=32) or placebo (n=13). One woman in each group discontinued the trial during the intervention and two women discontinued during the follow-up. No severe or serious adverse events were observed. Solicited adverse events occurred in 35 (78%) of 45 participants with no significant difference by group (risk ratio 1·17, 95% CI 0·79-1·75; p=0·44). All local solicited adverse events were mild. 32 (71%) of 45 participants strongly agreed or agreed they would use the product again. L crispatus dominant microbiomes were identified in 13 (41%) of 32 participants in the LACTIN-V group at week 4 and eight (26%) of 31 at week 8, compared with none at week 4 and one (9%) of 11 in week 8 in the placebo group (week 4 p=0·0088; week 8 p=0·40). The proportion of activated endocervical HIV target cells out of total T cells increased from after metronidazole treatment to week 4 in the placebo group (median log₂ fold change 1·891, IQR 1·731-4·018) but not in the LACTIN-V group (1·062, 0·449-1·424; p=0·016). Changes in the concentrations of 13 immune markers from before metronidazole treatment to week 4 or week 8 were not significantly different by group.

Interpretation: The use of LACTIN-V after metronidazole significantly increased vaginal L crispatus colonisation during 4 weeks of use, although this increase was transient, and women in the placebo group had an increase in endocervical CD4⁺ HIV target cells during recovery compared with the LACTIN-V group. These results show that vaginal colonisation with an L crispatus live biotherapeutic is possible in an African context, and that optimisation of this strategy might be a way to decrease risk for HIV.

Funding: US National Institute of Child Health and Human Development and US National Institute of Allergy and Infectious Diseases.