P838 Associations of the vaginal microbiota with HPV infection and cervical dysplasia in south african women living with HIV. (14th July 2019)

• Record Type: Journal Article
• Authors: Wijgert, Janneke Van De; Gill, Allessandra; Darby, Alistair; Kelly, Helen; Chikandiwa, Admire; Delany-Moretlwe, Sinead; Verwijs, Marijn; Francis, Suzanna; Mayaud, Philippe
• Is Part Of: Sexually transmitted infections. Volume 95(2019)Supplement 1
• Page Start: A352
• Page End: A352
• Publication Date: 2019-07-14
• Subjects: HPV; Sexually transmitted diseases -- Periodicals; HIV infections -- Periodicals; 616.951005
• DOI: 10.1136/sextrans-2019-sti.883
• Languages: English
• ISSNs: 1368-4973

Abstract

Abstract : Background: Fifteen longitudinal studies have shown associations between bacterial vaginosis and high risk human papillomavirus (hrHPV) acquisition and/or persistence, and/or cervical dysplasia. However, few studies assessed the vaginal microbiota (VMB) comprehensively, and none controlled the dysplasia association for persistent hrHPV. Methods: 623 women attending HIV outpatient clinics in Johannesburg, South Africa, were examined for hrHPV (InnoLipA HPV Genotyping Extra Assay), cervical dysplasia (histology), and vaginal microbiota (VMB; V3-V4 Illumina HiSeq 2x300bp with Swarm OTU-picking) at baseline and endline, a median of 16 months after baseline. VMB research questions were addressed in two nested case-control designs. Results: Hierarchical clustering resulted in seven VMB types: L. iners -dominated (Li; n=214 samples), Lactobacillus crispatus or L. jensenii -dominated (Lcj; n=68), Bifidobacterium -dominated (BD; n=2), lactobacilli + bacterial vaginosis (BV)-anaerobes (L+A; n=208), BV-like (BV; n=303); BV-anaerobe dominated (AD; n=56); and pathobiont-characterised (PB; n=19). Women with new or persistent hrHPV during follow-up were less likely to have an Lcj VMB type (compared to Li) at endline, and persistent hrHPV was associated with vaginal anaerobic dysbiosis at baseline (decreased lactobacilli, increased BV-anaerobes, and increased Nugent score). Women who developed CIN2+, compared to women with persistent hrHPV but no CIN2+, were more likely to have vaginal anaerobic dysbiosis at endline (decreased lactobacilli, increased BV-anaerobes, and increased diversity), but not at baseline. These associations persisted after controlling for age, hormonal contraception, and CD4+ count; several additional potential confounders (HIV plasma viral load, antiretroviral therapy, sexually transmitted infections, sexual risk-taking, among others) were evaluated. Conclusion: Frequent hrHPV exposure (and/or increased sexual risk-taking) likely causes vaginal dysbiosis, but a bilateral relationship cannot be ruled out. Women with vaginal dysbiosis are not at increased risk of CIN2+ development when hrHPV status is taken into account, but vaginal dysbiosis does develop when CIN2+ lesions develop. These results should be confirmed in even larger longitudinal studies. Disclosure: No significant relationships.