CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

284LB REDUCED COVERAGE OF HPV VACCINE TYPES IN CERVICAL PRECANCER IN HIV INFECTION Christina Carlander 1 , Camilla Lagheden 1 , Carina Eklund 1 , Sara Nordqvist Kleppe 1 , Philippe Wagner 2 , Aylin Yilmaz 3 , Kristina Elfgren 4 , Anders Sönnerborg 1 , Pär Sparén 1 , Joakim Dillner 1 1 Karolinska Institute, Stockholm, Sweden, 2 Uppsala University, Uppsala, Sweden, 3 Sahlgrenska Academy at the University of Gothenburg, Gothenburg, Sweden, 4 Karolinska University Hospital, Stockholm, Sweden Background: Our aimwas to study to which extent cervical precancer in women living with HIV (WLWH) is associated with HPV types targeted by vaccination. The Swedish National HIV registry (InfCareHIV) includes all WLWH in Sweden and the women in this cohort were found to have an increased risk of cervical precancer (Carlander et al. IJC 2016) and an increased risk of treatment failure of cervical precancer (Carlander et al. AIDS 2018). We requested all tissue blocks from cervical precancer in this cohort and subjected them to HPV genotyping. Methods: By linking InfCareHIV, the Swedish Population Registry and the Swedish National Cervical Screening Registry we identified all WLWH, mainly migrants (70%), living in Stockholm or Gothenburg sometime between 1983 and 2014, with high-grade cervical precancer (CIN2+). For each WLWH we randomly selected two HIV-negative control women (HNW), living in the same counties and also diagnosed with CIN2+, matched for country of birth. We retrieved and HPV genotyped the archived cervical tissue blocks. Type-specific HPV prevalence was compared using prevalence ratios (PR), calculated with Poisson regression analysis. All models were adjusted for age, grade of cervical lesion and region of birth. Results: 108 WLWH and 183 HNW had valid HPV genotype results (100 [93%] WLWH and 164 [90%] HNWwere HPV-positive). WLWH were less likely to be infected with HPV16 (PR: 0.6, 95% CI: 0.4-0.9) than HNW. HPV35 (not included in the 9-valent HPV vaccine) was the second most common HPV type in WLWH and three times more common than in HNW (PR: 3.1, 95% CI 1.3-7.4). WLWH were more likely to be infected with multiple HPVs (30 vs. 20%; PR: 1.5, 95% CI: 1.0-2.4). HPV types targeted by the 9-valent HPV vaccine were significantly less common in WLWH (57%) compared to HNW (80%) (PR = 0.7, 95% CI 0.6-0.9). Conclusion: This national population-based cohort study, controlled for country of birth of migrants, found that cervical precancer in WLWH contained HPV types targeted by vaccination to a lower extent than in HNW, implying that cervical screening remains highly important in WLWH, even if HPV vaccinated.

283 PRIMARY HPV SCREENING IN WOMEN LIVING WITH HIV

Howard D. Strickler 1 , Nancy A. Hessol 2 , Isam-Eldin Eltoum 3 , Mark Einstein 4 , Philip E. Castle 1 , L. Stewart Massad 5 , Lisa Flowers 6 , Lisa Rahangdale 7 , Adaora Adimora 7 , Igho Ofotokun 6 , Meghan Huchko 8 , Gypsyamber D’Souza 9 , Joel Palefsky 2 , Robert Burk 1 , for the WIHS HPV Screening Study 1 Albert Einstein College of Medicine, Bronx, NY, USA, 2 University of California San Francisco, San Francisco, CA, USA, 3 University of Alabama at Birmingham, Birmingham, AL, USA, 4 Rutgers University, Newark, NJ, USA, 5 Washington University in St Louis, St Louis, MO, USA, 6 Emory University, Atlanta, GA, USA, 7 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 8 Duke University, Durham, NC, USA, 9 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA Background: Primary HPV Screening (PHS) utilizes an oncogenic human papillomavirus (oncHPV) assay as the initial cervical cancer screening test (instead of a Pap test). Recent US guidelines support PHS in women in the general population (30-65 years of age), and suggest oncHPV- women should be retested in 5 years. For oncHPV+ women, reflex HPV genotyping is recommended and possibly an additional test (e.g., Pap test) in order to determine appropriate follow-up or triage to colposcopy. However, the current study is, to our knowledge, the first to assess PHS in WLWH. Cervical cancer screening strategies for WLWH need improvement as these women have elevated risk of cervical cancer relative to the general population, but current strategies result in a high rate of colposcopy that does not reflect clinically relevant cervical disease. Methods: The study enrolled 865 WLWH comprised of 323 new enrollees in the Women’s Interagency HIV Study (WIHS) and 542 WLWH enrolled through colposcopy clinics affiliated with WIHS. Newly enrolled WIHS women represented WLWH undergoing routine screening. Colposcopy patients represented WLWH who had a recent abnormal screening test (e.g., ASC-US+). WIHS participants underwent routine screening using liquid-based Pap tests (ThinPrep) and were tested for oncHPV by the FDA-approved Cobas test. WIHS enrollees with a positive oncHPV or ASC-US+ received colposcopy, as did 15% of women with negative oncHPV and Pap results. Like WIHS enrollees, the WLWH enrolled at colposcopy had oncHPV and Pap tests. All Pap/histology was centrally reviewed by two expert pathologists. Results: Mean age was 47 years for both WIHS enrollees and colposcopy patients, and most were Hispanic or non-Hispanic African American. Median (IQR) CD4 count was 560 (342-843) and 631 (362-849), respectively, with 97% and 83% reporting cART use. There was a total of 70 CIN2+ of which 23 were CIN3+ (precancer). The Table shows sensitivity, specificity, and positive predictive value (PPV) to detect CIN3+, as well as the % of WLWH who would be triaged to colposcopy following several strategies. Results for PHS, as estimated by oncHPV with reflex Pap, had the highest PPV and lowest colposcopy rate; Co-Testing had moderately higher sensitivity and colposcopy triage rates. Conclusion: The results show that PHS with reflex Pap testing is a potentially useful cervical cancer screening strategy in WLWH, but may sacrifice a moderate level of sensitivity for a moderate reduction in the rate of triage to colposcopy.

Poster Abstracts

CROI 2019 103