Multiple High-Risk HPV Types Contribute to Cervical Dysplasia in Ugandan Women Living With HIV on Antiretroviral Therapy

Abstract

Background: Cervical cancer mortality remains high in sub-Saharan Africa, especially among women living with HIV (WLWH). Characterization of prevalent high-risk human papillomavirus (hrHPV) types and immune function in WLWH with cervical abnormalities despite antiretroviral therapy (ART) can inform prevention strategies.

Setting: Kampala, Uganda.

Methods: From 2017 to 2020, we enrolled Ugandan women with cervical dysplasia detected with visual inspection with acetic acid (VIA). WLWH were required to be on ART >3 months with plasma HIV RNA <1000 copies/mL. Biopsies from VIA-positive lesions underwent histopathologic grading and cervical swab specimens were tested for hrHPV. Clinical correlations were evaluated with Poisson regression to estimate adjusted prevalence ratios (aPR).

Results: One hundred eighty-eight WLWH and 116 HIV-seronegative women participated. Among WLWH, median ART duration was 6 years and median CD4 667 cells/µL. Cervical intraepithelial neoplasia (CIN) grade 2/3 was found in 29% of WLWH versus 9% of HIV-seronegative women. In women with CIN1 or without histopathology-confirmed dysplasia, hrHPV (aPR [95% confidence interval]: 2.17 [1.43 to 3.29]) and multiple hrHPV (aPR 3.73 [1.07 to 13.1]) were more common in WLWH, as were vaccine-targeted and vaccine-untargeted hrHPVtypes. Differences in hrHPV prevalence by HIV serostatus were not observed in women with CIN2/3 (interaction P < 0.01). Among WLWH, low CD4/8 ratio was associated with hrHPV while detectable plasma HIV RNA (20-1000 copies/mL) was associated with CIN2/3 or invasive cancer.

Conclusion: Despite ART, WLWH with cervical VIA abnormalities remain at elevated risk for multiple hrHPV and high-grade dysplasia. Cervical cancer prevention and research tailored for WLWH are warranted in the ART era.

Figure 1.. Study design.

Schema of screening of women and study participant enrollment from community HIV clinics and the Uganda Cancer Institute gynecology clinical service.

Figure 2.. High-risk HPV (hrHPV) associations.

Prevalence ratio (PR) and 95% confidence interval (95% CI) for hrHPV comparing HIV-seropositive to HIV-seronegative participants, stratified by cervical biopsy pathology findings and adjusted for age, education, male condom use, and hormonal contraceptive use. Arrows indicate 95% CI limit bound is outside of the displayed range (0.1 to 10.0). P-value of interaction term comparing aPRs between strata: *, 0.05 > P > 0.01; ** 0.01 > P > 0.001

Figure 3.. Risk factors for HPV in women living with HIV.

Prevalence ratios (PRs) and 95% confidence intervals (95% CI) of categories of high-risk HPV (hrHPV) associated with viral load, T-cell lymphocyte counts, antiretroviral therapy (ART), and personal characteristics among women living with HIV on ART with abnormal cervical findings by visual inspection with acetic acid. Black circles are univariable (unadjusted) PR; blue triangles represent PRs adjusted for all variables in the figure; red square indicates reference category.

Figure 4.. Risk factors for high grade cervical epithelial neoplasia in women living with HIV.

Prevalence ratios (PR) of high-grade cervical dysplasia (CIN2/3 or ICC) among women living with HIV on ART with abnormal cervical findings by visual inspection with acetic acid, and with detectable cervical high-risk HPV (hrHPV). Black circles are univariable (unadjusted) PR; blue triangles represent PRs adjusted for all variables in the figure; red square indicates reference category.