CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

(BIA) were performed at recruitment and after 6 and 12 weeks of ART, and serum CRP was measured at recruitment and 6 weeks. The relationships between CRP and changes in body composition measurements were assessed using linear regression models adjusted for age, sex, CD4 count, hemoglobin, trial arm, study site, tuberculosis treatment at ART initiation, and baseline CRP and anthropometric values. Results: 838 trial participants who had baseline and 6 week CRP measurements were included in this analysis. Median age was 35 years, 51%were female, and median CD4 count was 135 cells/ m L. Median CRP was high at recruitment (61 mg/l; IQR 14, 160) and lower, although still abnormally elevated, after 6 weeks of ART (34 mg/l; IQR 12, 94). A one-log reduction in CRP at 6 weeks was associated with increased mid-upper-arm circumference (0.45 cm, p<0.001), calf circumference (0.38 cm, p<0.001), waist circumference (0.98 cm, p<0.001) and BIA fat-free mass (0.58 kg, p<0.001), but CRP was not associated with BIA fat mass. Trial armwas not a mediating variable in any of these relationships. The anthropometric and fat-free mass gains persisted after completion of the supplement, and the largest increases in lean mass at 12 weeks were observed in patients with moderate-to-high baseline CRP and lower 6 week CRP. Conclusions: Larger reductions in CRP shortly after ART initiation were associated with greater increases in lean body mass, which is a marker of nutritional rehabilitation and may impact future cardiometabolic disease risk. Further studies are needed to understand the directionality of the observed the relationships. 763 Smoking and Obesity May Partially Explain the Inflammation and Morbidity Association Supriya Krishnan 1 ; Ronald Bosch 1 ; Benigno Rodriguez 2 ; PeterW. Hunt 3 ; Cara C.Wilson 4 ; Steven Deeks 3 ; Michael M Lederman 2 ; Alan Landay 5 ; Carey Lumeng 6 ; AllanTenorio 7 1 Center for Biostatistics in AIDS Research, Harvard School of Public Health, Boston, MA, US; 2 Division of Infectious Diseases and HIV Medicine, Case Western Reserve University School of Medicine, Cleveland, OH, US; 3 Positive Health Program, Dept of Medicine, San Francisco General Hospital, San Francisco, CA, US; 4 Division of Infectious Diseases, Univ of Colorado Hospital, Aurora, CO, US; 5 Department of Immunology and Microbiology, Rush Univ Medical Center, Chicago, IL, US; 6 Univ of Michigan Medical School, Department of Pediatrics, Ann Arbor, MI, US; 7 Department of Medicine, Rush University Medical Center, Chicago, IL, US Background: As in the general population, increased central obesity and smoking are correlated with higher levels of soluble inflammatory markers in HIV+ virally suppressed patients. We previously reported that high levels of IL-6, sTNFR-1 and 2 and D-dimer were associated with an occurrence of a non-AIDS event. Understanding the pathophysiology of inflammation in HIV and biologic pathways between inflammation and non-AIDS morbidities may inform therapeutic strategies for HIV-infected populations. Methods: A previously described case-control study (143 cases with non-AIDS morbidities including MI/stroke/non-AIDS malignancy/death, 315 matched controls) examined HIV+ adults from the ACTG-ALLRT cohort who were ART naïve at entry, received a modern ART regimen, and were virally suppressed (<400 cp/ml) at year 1 of ART. Stored plasma at year 1 was tested for IL-6, IP-10, sTNFR-I & –2, sCD14 and D-dimer. Conditional logistic regression was used to examine if each soluble marker was independently associated with non-AIDS morbidities adjusting for other soluble markers (singly), waist circumference (WC, cm) and smoking (# of cigs/day) at year 1. WC and smoking were included because they correlated with the soluble markers in a previous analysis and are known risk factors for the outcome in the general population. Results: IL-6 was independently associated with non-AIDS outcomes even after separately adjusting for sTNFR-1 and 2, D-dimer, IP-10 and sCD14 (all p ≤ 0.001). In contrast, the associations for D-dimer and sCD14 were no longer significant when IL-6 was included in the model. Both sTNFR-1 and sTNFR-2 remained independently associated with non-AIDS outcomes even after adjusting for IL-6. In a model that included WC and smoking, the OR for IL-6 was attenuated from 1.82 to 1.54 (Table 1). In contrast, adjusting for WC and smoking only slightly altered the OR for sTNFR-1 and 2 (Table 1).

Poster Abstracts

Conclusions: Among the six soluble markers evaluated, IL-6, sTNFR-1 and sTNFR-2 measured during viral suppression appear to be robust predictors of and may be part of distinct pathways leading to subsequent non-AIDS outcomes. The IL-6 association was independent of obesity and smoking; however the effect was attenuated by 15% suggesting that obesity and smoking may play a role in some IL-6 related pathways leading to an increased risk of non-AIDS outcomes. These findings further support efforts to identify the effects of reducing obesity and smoking on residual inflammation. 764 Infectious and Noninfectious Multimorbidity Among HIV Clinic Clients in the African Cohort Study Julie Ake 1 ; Jonah Maswai 2 ; Francis Kiweewa 3 ; Lucas Maganga 4 ; Milton Omondi 5 ; Babajide Keshinro 6 ; Lindsay Hughes 1 ;Victor G.Valcour 7 ; Christina Polyak 1 RV 329 AFRICOS StudyTeam 1 US Military HIV Research Program, Bethesda, MD, US; 2 KEMRI/Walter Reed Project, Kericho, Kenya; 3 Makerere University Walter Reed Project, Kampala, Uganda; 4 Mbeya Medical Research Centre, Mbeya, United Republic of Tanzania; 5 KEMRI/Walter Reed Project, Kisumu, Kenya; 6 Walter Reed Program - Nigeria, Abuja, Nigeria; 7 University of California San Francisco (UCSF), San Francisco, CA, US Background: The frequency and complexity of multimorbidity in the increasingly treated African HIV epidemic is not comprehensively described. Methods: The African Cohort Study (AFRICOS) prospectively enrolls adult HIV clinic clients and uninfected adults at 10 PEPFAR supported facilities in Kayunga, Uganda; South Rift Valley and Kisumu West, Kenya; Mbeya, Tanzania and Abuja, Nigeria. We evaluated infectious comorbidities (ICOs) at baseline: HBV, HCV and syphilis by screening serologies, pulmonary TB by Xpert MTB/RIF, and cryptococcosis by serum CrAg if CD4 ≤ 200. Baseline noninfectious comorbidities (NICOs) included elevated blood pressure (SBP>138 or DBP>79), impaired fasting glucose (IFG) (>99 mg/dl), renal insufficiency (GFR<60 mL/min), hypercholesterolemia (total cholesterol >199 mg/dl), anemia (hemoglobin <11 mg/ dl), and cognitive impairment (International HIV Dementia Scale < 1.5 SD below the mean). We compared those with no comorbidities to those with ≥ 2 comorbidities using descriptive statistics and investigated determinants of NICOs employing multivariate Poisson regression. Results: From January 2013 to August 2014, 899 HIV infected adults enrolled in Kayunga (22%), South Rift Valley (41%), Kisumu West (15%), Mbeya (14%), and Abuja (7%). Participants were 60% female with mean age 40.5 years and mean CD4 count 419 cells/mm 3 . Among the 73% taking antiretroviral therapy (ART), mean ART duration was 4.1 years, 68%were suppressed to < 50 copies/mL, and 78% had exposure to thymidine analogues. Rates of smoking and IDU were low (3.4% and 0.2%). Reactive serologies for hepatitis B, C, and syphilis were identified in 5.2%, 4.4% and 1.3%. Sixteen (1.8%) had positive Xpert assays and 2 (0.7%) had cryptococcal antigenemia. The most common NICOs

468

CROI 2015