CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

electrosurgical excisional procedure (LEEP) for treatment of cervical intraepithelial neoplasia grade 2 or 3 (CIN2/3). Methods: From June 2011 to July 2014, HIV-infected women enrolled at the Coptic Hope Center in Nairobi, Kenya with CIN2/3 were randomized to receive cryotherapy or LEEP and followed for 2 years with a Pap smear and HPV cervical swab every 6 months. hrHPV was defined as a positive result on at least one of 13 types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68) identified by the Roche Linear Array Genotyping Test. Clearance was defined as testing negative for the same hrHPV type/s detected at baseline on ≥2 consecutive visits ≥6 months apart. Time to clearance or duration of hrHPV infection was defined as the time elapsed from intervention to the date of the first negative hrHPV test. Reinfection was defined as new hrHPV infection after clearance. Outcomes were compared between arms using Chi-square tests and log-binomial regression. Results: Of 400 women randomized to cryotherapy or LEEP, 95% (189 per arm) had baseline hrHPV results. Median age was 37 years [interquartile range (IQR): 31-43], median CD4 count was 380 cells/μl (IQR: 211-525), and median plasma HIV RNA viral load was 1.5 log10/mL (IQR: 1.5-2.8). The majority (88%) of women were on antiretroviral treatment (ART) at baseline, of whom 40%were on ART for ≥2 years. Baseline hrHPV prevalence was 93% in the cryotherapy arm and 92% in the LEEP arm (P=0.83). Clearance of hrHPV was significantly higher in LEEP than cryotherapy both at 6 months following intervention (36% vs 24%; P=0.015) and over two-year follow up (50% vs 39%; P=0.040). Median time to clearance was 6 months in each arm (P=0.16). Those who underwent LEEP were 50% (95% confidence interval (CI), 1.1-2.1; P=0.017) more likely to clear hrHPV than those receiving cryotherapy. The difference in reinfection with hrHPV following clearance of hrHPV in women with LEEP vs cryotherapy was not statistically significant (Relative risk=0.67, 95% CI, 0.4-1.1; P=0.089). Conclusion: Clearance of hrHPV in HIV-infected women after cervical treatment was limited; 40% experienced hrHPV reinfection within 2 years. However, women receiving LEEP were more likely to clear hrHPV than those receiving cryotherapy adding a reason to consider expanding LEEP in resource- limited settings. 279 HPV DNA TESTS FOR CERVICAL CANCER SCREENING OF HIV-INFECTED WOMEN Louise Kuhn 1 , Rakiya Saidu 2 , Ana I. Tergas 1 , Rosalind Boa 2 , Jennifer Moodley 2 , Cecilia Svanholm-Barrie 3 , David Persing 4 , Scott Campbell 4 , Thomas C. Wright 1 , Lynette Denny 2 1 Columbia University Medical Center, New York, NY, USA, 2 University of Cape Town, Cape Town, South Africa, 3 Cepheid, Solna, Sweden, 4 Cepheid, Sunnyvale, CA, USA Background: HPV DNA testing has excellent sensitivity but poor specificity for cervical cancer screening among HIV-infected women. We evaluated whether the point-of-care test, Xpert™ HPV, could be adapted to improve performance characteristics for screening HIV-infected women. Methods: A clinical study of 586 HIV-uninfected and 535 HIV-infected women, aged 30-65 years, was conducted in Cape Town, South Africa. All women had a cervical sample collected that was tested on-site with Xpert HPV which is a cartridge-based PCR assay that detects HPV DNA in 5 channels: HPV 16, HPV 18,45, HPV 31,33,35,52,58, HPV 51,59, and HPV 39,56,66,68. For each channel a cycle threshold (CT) value is generated and values below pre-determined CT cut-offs are defined as positive. All women underwent colposcopy with histological sampling. Cervical intraepithelial neoplasia grade 2,3 or cancer (CIN2+) was diagnosed based on consensus pathology review. Sensitivity, specificity, positive and negative predictive values were calculated based on logistic regression and receiver operating characteristic curves. Results: Almost half (49.2%) of HIV-infected women tested positive for HPV DNA whereas 16.2% of uninfected women did (p<0.001). The prevalence of histology-confirmed CIN2+ was higher in HIV-infected women (17.0%) than in uninfected women (5.3%) (p<0.001). Sensitivity of detecting CIN2+ at the manufacturer-defined CT cut-off was 93.6% in HIV-infected women with a specificity of 59.9%. If screen-positive was limited to the 3 channels detecting HPV 16, HPV 18,45 and HPV 31,33,35,52,58, sensitivity remained high (90.7%) and specificity improved (67.5%). Shifting the CT values from these 3 channels such that sensitivity was set at 85%, resulted in improvements in specificity (77.0%). If sensitivity was set at 80%, specificity improved further (83.2%). At these CT cut-offs, positive predictive value was 49.4% and the proportion screen-positive was 27.4%. Conclusion: Adapting Xpert HPV by restricting the definition of screen- positive to a limited number of high risk HPV types and making CT cut-offs

more stringent (i.e. requiring higher levels of HPV DNA) can greatly improve performance characteristics of HPV DNA testing for cervical cancer screening in HIV-positive women. Making these adaptations limits the number of HIV-infected women who require further follow-up or treatment for cancer precursor lesions.

280 RELATION OF CIN2 PROGRESSION WITH SERIAL CERVICOVAGINAL CYTOKINE/CHEMOKINE LEVELS Kate G. Michel 1 , Christine Colie 1 , L. Stewart Massad 2 , Cuiwei Wang 1 , Allison G. Doyle 1 , Gypsyamber D’Souza 3 , Lisa Rahangdale 4 , Lisa Flowers 5 , Joel Milam 6 , Joel Palefsky 7 , Howard Minkoff 8 , Howard D. Strickler 9 , Seble Kassaye 1 1 Georgetown University, Washington, DC, USA, 2 Washington University in St Louis, St Louis, MO, USA, 3 Johns Hopkins University, Baltimore, MD, USA, 4 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 5 Emory University, Atlanta, GA, USA, 6 University of Southern California, Los Angeles, CA, USA, 7 University of California San Francisco, San Francisco, CA, USA, 8 Maimonides Medical Center, Brooklyn, NY, USA, 9 Albert Einstein College of Medicine, Bronx, NY, USA Background: Women living with HIV are high-risk for cervical intraepithelial neoplasia-2 (CIN2), a potentially pre-cancerous diagnosis. We characterized longitudinal variations in cervicovaginal lavage (CVL) chemokine/cytokine levels and their temporal relation with CIN2 and its progression. Methods: N=104 HIV-positive women (age < 46) were selected from the Women’s Interagency HIV Study (WIHS). To avoid misclassification, we excluded women CIN3+ found within 6 months post-CIN2 diagnosis. CVLs were analyzed at 4 time points: (1) 0.5-1 year prior to CIN2, (2) at CIN2 diagnosis, (3) at progression/regression event, and (4) visit between CIN2 and progression/ regression. We used Luminex assays to measure 34 cytokines/chemokines (normalized by total protein and log transformed). We used Bonferoni-adjusted T tests and logistic regression for analysis; multivariate analysis included adjustment for cervical treatment, smoking status, CD4 and HIV viral load. Results: Of the 104 women, 12 progressed to CIN3 (median 2.7 years), 63 regressed to CIN1/normal (median 1.5 years), and 29 had a subsequent CIN2 diagnosis (median 1.5 years). CIN2 treatment was received by 7 (58%) of CIN2 progressors, 20 (32%) of CIN2 regressors, and 13 (45%) of CIN2 stable women. The majority of women were African American (55%), current smokers at baseline (53%), with a median age of 34 years (no significant differences between groups). Anti-inflammatory cytokines were significantly lower in CIN2 regressors and CIN2 stable women at the CIN2 visit compared to CIN2 progressors (Figure 1). IL-27 was significantly elevated at the pre-CIN2 visit in progressors vs. regressors (p=0.03), no difference in IL-27 at visit where CIN2 was diagnosed. At CIN2 visit, IL-6 (p=0.05), IL-1a (p=0.03) and IL-10 (p=0.05) were elevated in progressors vs. regressors. Controlling for CIN2 treatment, smoking status, CD4 count, and HIV viral load, the odds of CIN2 progression was associated with higher IL-27 (aOR 1.41, 95% CI 1.23, 1.60), and IL-9 (aOR 1.39, 95% CI 1.08, 1.79), along with lowered IL-10 (aOR 0.53, 95% CI 0.34, 0.83) and IL-21 (aOR 0.83, 95% CI 0.71, 0.97), compared to CIN2 regressors. Conclusion: We identified a group of cytokines, present at the time of CIN2 diagnosis, which may influence risk of CIN2 progression in HIV+ women. While IL-1a is strongly pro-inflammatory, IL-6, IL-9, IL-10, and IL-27 have inflammatory and anti-inflammatory activity-suggesting that a nuanced immune response to CIN2 may be key to its progression in HIV+ women.

Poster Abstracts

CROI 2019 101