CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

Poster and Themed Discussion Abstracts

595 SCREENING FOR PRECANCEROUS ANAL LESIONS WITH P16/KI67 DUAL STAIN CYTOLOGY IN HIV Sergio Serrano-Villar 1 , Beatriz Hernández-Novoa 2 , Amparo de Benito 1 , Jorge Romero 3 , Antonio Ocampo 4 , Laura Pérez-Martínez 5 , Mar Masia 6 , Elena Sendagorta 7 , Gonzalo Sanz 8 , José Antonio Pérez-Molina 1 1 Hosp Univ Ramón y Cajal, Madrid, Spain, 2 Brystol-Myers-Squibb, Madrid, Spain, 3 Centro Sandoval, Madrid, Spain, 4 Hosp Xeral de Vigo, Vigo, Pontevedra, 5 Hosp San Pedro, Logroño, Spain, 6 Hosp General Univ de Elche, Elche, Spain, 7 Hosp Univ La Paz, Madrid, Spain, 8 Hosp Clínico San Carlos, Madrid, Spain Background: Anal cancer is among the most prevalent neoplasias in HIV-infected MSM. Screening with anal citology results yields to high rates of false positive results and elevated burden of high-resolution anoscopies (HRA) with anal biopsies. High risk HPV up-regulate p16 expression and increase proliferation (Ki67 expression) in epithelial cells. We assessed the usefulness of P16/Ki-67 dual staining cytology for the diagnosis of precancerous anal lesions. Methods: Prospective multi-center cohort study in 6 HIV clinics. Concomitant anal liquid cytology with p16/Ki-67 dual staining (CINtec® PLUS, Roche) and HRA with biopsy of acetowhite lugol-negative lesions was performed. We compared the diagnostic performance of an abnormal anal cytology (atypical squamous cells [ASC], LSIL or HSIL) and p16/ Ki-67 dual positivity relative to HRA-guided biopsy. We calculated the independent predictive values of anal cytology and p16/Ki-67 positivity in multivariate logistic regression models adjusted by potential confounders. Results: A total of 328 HIV-infected MSM underwent 386 full examinations. Mean age was 39±10 years, median nadir CD4 367 (258-510) cells/uL, 57% had detectable plasma HIV RNA and 30% reported unprotected anal sex in the prior 3 months. Sixty-three % of anal cytologies were abnormal: 24(6.2%) ASC, 143(37%) LSIL and 74(19.2%) HSIL. HSIL was histologically diagnosed in 80 subjects (24.1%), and 2 (0.6%) were diagnosed with anal cancer, in whom the cytology had showed LSIL and HSIL, respectively. An abnormal citology showed the following statistics for the diagnosis of biopsy proven HSIL: sensitivity 95.6% (CI95%, 91.2-99.9), specificity 58.8% (CI95%, 52.2-65.4), positive predictive value 39.8% (CI95%, 33.2-46.4), negative predictive value 95.8% (CI95%, 91.6-99.9). P16/Ki67 dual positivity was not associated with higher rates of biopsy-proven HSIL (Table 1). After adjustment by potential confounders (age, nadir CD4, detectable HIV RNA, tobacco use), an abnormal anal citology, but not a positive P16/Ki67 stain, was an independent predictor of HSIL (OR, 12.1; CI95% 3.5-41.3 and OR, 1.2; CI95% 0.6-2.5, respectively). Conclusion: P16/Ki67 dual staining does not improve the diagnostic accuracy of anal cytology, which shows a high sensitivity yet poor specificity. Other approaches aimed at improving the diagnostic accuracy of current techniques for the diagnostic of precancerous anal lesions are warranted.

596 IMPACT OF ART COVERAGE AND SCREENING ON ANAL CANCER IN HIV+ MEN WHO HAVE SEX WITH MEN Nello Blaser 1 , Barbara Bertisch 1 , Manuel Koller 1 , Marcel Stoeckle 2 , Matthias Cavassini 3 , Roger Kouyos 4 , Janne Estill 1 , Matthias Egger 1 , Olivia Keiser 1 , for the Swiss HIV Cohort Study 1 Univ of Bern, Bern, Switzerland, 2 Univ Hosp Basel, Basel, Switzerland, 3 Univ of Lausanne, Lausanne, Switzerland, 4 Univ of Zurich, Zurich, Switzerland Background: In HIV-positive men who have sex with men (MSM), the incidence of anal cancer is 60-190 times higher than in the general population. The incidence has been rising for many years, but has declined in recent years. In this population, where we assume that almost everybody had a history of contact with oncogenic Human Papillomaviruses (HPV), the most important risk factor for anal cancer is a low CD4 cell count. Little is known about the likely impact of increased antiretroviral therapy (ART) coverage and screening on anal cancer incidence. Methods: We developed a mathematical model to estimate the incidence of anal cancer in HIV-positive MSM in Switzerland up to 2030. We considered two scenarios of future ART coverage: “stable” (same cART coverage between 2016 and 2030 as between 2010 and 2015 and no screening for AIN2/3) and “100% cART coverage” (all individuals diagnosed with HIV are on ART); and four screening scenarios: “no screening”, “yearly cytology”, “yearly anoscopy” and “CD4-dependent anoscopy” (people were screened with anoscopy five years after their CD4 cell count dropped to below 200 cells/µl). We parameterized the model with data from the Swiss HIV Cohort Study (SHCS) and the literature. We considered CD4 cell count trajectory the main predictor of anal cancer.

CROI 2017 251