CROI 2015 Program and Abstracts

Abstract Listing

Oral Abstracts

Conclusions: At one year follow-up, HIV-infected women who were VIA positive at baseline were more likely to progress from negative to CIN 1+ and from CIN 1 to CIN 2+. Of women with negative baseline VIA, 4% progressed to CIN 2+ in a year. HIV-infected women with a positive VIA or HPV should be screened at one year for possible progression.

90 High Rate of HSIL on HRA in HIV+Women Not Meeting Criteria for Anal Cancer Screening Michael M. Gaisa; Fanny Ita-Nagy ; Gabriela Rodriguez Caprio; Michael Mullen; Judith Aberg; Michelle Cespedes Icahn School of Medicine at Mount Sinai, New York, NY, US

Background: HIV-infected patients have a higher burden of HPV associated high-grade squamous intraepithelial lesions (HSIL) and anal cancer. Guidelines for anal cancer screening in HIV-infected women in New York State include a history of abnormal cervical and/or vulvar histology or history of anogenital warts. The HIV Medical Association of IDSA (HIVMA) screening guidelines include these and add “or a history of receptive anal intercourse” to the criteria as another indication. Best practices for this population are poorly defined. Here we report outcomes and associated risk factors for HSIL in HIV-infected women over five years after implementation of a programwhich offers anal cancer screening by means of anal cytology to all HIV-infected women regardless of previous HPV disease or sexual behavior. Methods: Data fromwomen who underwent high resolution anoscopy (HRA) following abnormal anal cytology from April 2009 to July 2014 were reviewed. All HRA were performed by a single operator. Routine clinical data included cervical PAP history, demographics, behavioral data, and HRA results. Chi square tests were used for comparisons. Results: 306 HIV-infected women underwent HRA, median age was 47, mean CD4 was 537 cells/mm 3 , 67% had HIV viral loads <50, 72% had a history of abnormal cervical PAP, and 66% reported a history of anal receptive intercourse. HSIL was found in 28% of anal biopsies. 49% of the 55 women who did not meet criteria for anal cancer screening according to NYS guidelines had anal dysplasia on biopsy. 13 of the 55 women (24%) had HSIL requiring treatment, including one subject with minimally invasive carcinoma. Using HIVMA recommendations, an additional 35 patients met criteria for anal cancer screening. Four of the 20 women (20%) who did not meet screening criteria by either guideline had HSIL on biopsy. Neither meeting NYS criteria, HIVMA criteria, nor a history of receptive anal intercourse as the sole criterion were significantly associated with a diagnosis of HSIL ( p =0.079, 0.403, and 0.093 respectively). A history of smoking was the only factor associated with HSIL on biopsy ( p =0.002).

Oral Abstracts

Conclusions: Anal HSIL was commonly found in HIV-infected women in this cohort. A high rate of dysplasia on biopsy was found even among women who did not meet criteria for routine screening. These results support that all HIV infected women may benefit from anal cancer screening regardless of their cervical / genital HPV history or sex practices. 91 Xpert MTB/RIF Ultra: A New Near-Patient TB Test With Sensitivity Equal to Culture David Alland 1 ; Mazhgan Rowneki 1 ; Laura Smith 1 ; Jamie Ryan 2 ; Mitchell Chancellor 2 ; Ann Marie Simmons 2 ; David Persing 2 ; Robert Kwiatkowski 2 ; Martin Jones 2 ; Soumitesh Chakravorty 1 1 Rutgers New Jersey Medical School, Newark, NJ, US; 2 Cepheid, Sunnyvale, CA, US Background: The Xpert MTB/RIF (Xpert) assay, the first near-patient test for tuberculosis (TB), detects Mycobacterium tuberculosis ( Mtb ) with a limit of detection (LOD) of 133 CFU/ ml sputum, and detects mutations in the Mtb rpoB gene which cause Rifampicin resistance (RIF-R). However, despite high sensitivity overall, Xpert sensitivity is only 60 – 80% in smear-negative TB. Also, small numbers of false RIF-R have been reported recently, usually due to abnormal real-time PCR curves or miss-identification of RIF-susceptible (RIF-S) synonymous rpoB mutants as RIF-R. A newmolecular TB test with an LOD equivalent to the 10 – 100 CFU/ml LOD of liquid culture and improved RIF-R detection is needed. Methods: For the Xpert MTB/RIF Ultra (Ultra) assay, we developed a new sample processing cartridge that doubled the amount of purified DNA delivered to the PCR reaction. Four newly designed probes that detected mutations in rpoB gene replaced the five Xpert real-time probes. Real-time Mtb detecting probes targeting IS6110 and IS1081 were added. Cartridge fluidics and PCR cycling were optimized. Assay LODs were tested by spiking Mtb H37Rv and BCG cells into sputum samples, treating with Sample Reagent, splitting samples, and testing with Xpert and Ultra. RIF-R detection was tested with a panel of 30 different RIF-R Mtb rpoB mutants. LOD was defined as the lowest CFU that could be detected in at least 19/20 (95%) tests. Results: Ultra was significantly more sensitive than Xpert. In sputum samples spiked with Mtb H37Rv, Ultra had an LOD of 5 CFU/ml compared to an LOD of 50 CFU/ml for Xpert (p=0.001). In sputum samples spiked with BCG, Ultra had an LOD of 25 CFU/ml compared to an LOD of 165 CFU/ml for Xpert (p=0.01). Ultra detected 30 different RIF-R Mtb rpoB mutants as RIF-R (sensitivity 100%). None of the 25 RIF-S rpoB wild type samples and none of the 3 RIF-S synonymous rpoB Q513Q (1) and F514F (2) mutant samples were detected as RIF-R (specificity 100%). Ease of use was identical for Xpert and Ultra. Conclusions: The new Ultra assay is much more sensitive than Xpert, and is likely to be as sensitive as liquid TB culture. Ultra detects RIF-R as efficiently as Xpert; but the specificity of Ultra RIF-R is likely to be higher due to improvements in assay design. The Ultra assay should significantly increase TB detection in smear-negative patients and provide more reliable RIF-R detection.

130

CROI 2015