CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

Methods: Cross-sectional, conveniently sampled non-diabetic HIV-infected adults (≥18 years) who received HAART≥1 year at a large public sector ART clinic in Pune, India were included. Participants whose Quantitative Insulin Sensitivity Check Index (QUICKI) <0.339 were classified as insulin resistant. Logistic regression was used to identify associations between fasting GIP, GLP-1, leptin, ghrelin, C-peptide, PAI-1 (plasminogen activator inhibitor-1), resistin, visfatin, IL-6, IL-10, TNF-α, MCP-1 and IR. Results: Of 350 adults included, 115 (33%) were insulin resistant. Median age was 40 years (IQR:35-45), 157 (45%) were male, median time updated CD4 was 471 cells/µL (IQR:298 – 655), 221 (63%) had undetectable viral loads, 73 (21%) were receiving PI based regimens, median fasting insulin was 42 pmol/L (IQR:26-65). When compared to insulin sensitive individuals, those with IR had higher BMI (p<0.001), longer duration of ART use (p=0.008), a higher proportion were hypertensive (p=0.02) and dyslipidemic (p<0.001), had higher GIP (p<0.001), GLP-1 (p<0.001), C-peptide (p<0.001). In multivariable models adjusted for age, sex, BMI, CD4 count, viral load, HAART duration, dyslipidemia, PI use; C-peptide (Odds Ratio (OR), 95% Confidence Interval (95% CI):11.4, 4.9 – 26.3, p<0.001) and GLP-1 (OR, 95% CI: 1.004 , 1.002 – 1.006, p<0.001) were associated with IR. We found no association between GIP, IL-6, IL-10, TNF-α, MCP-1 and IR. Conclusion: GLP-1 but not GIP was associated with IR. Furthermore, we found a strong association between C-peptide and IR. We found no association between inflammatory markers and IR in our population of non-diabetic PLWH. Elucidation of the pathways regulating IR in PLWH require further investigation. 743 T CELL SUBSETS ASSOCIATED WITH INCIDENT DIABETES RISK IN HIV+ AND HIV− VETERANS Wyatt J. McDonnell 1 , Kathleen McGinnis 2 , Kaku So-Armah 3 , Melissa Wellons 1 , Glory Koerbel 1 , Jeffrey Blume 1 , Margaret F. Doyle 4 , Nels C. Olson 4 , Russell Tracy 4 , Amy C. Justice 5 , Simon Mallal 1 , Matthew Freiberg 1 , John R. Koethe 1 1 Vanderbilt University, Nashville, TN, USA, 2 VA Pittsburgh Healthcare System, Pittsburgh, PA, USA, 3 Boston University, Boston, MA, USA, 4 University of Vermont, Colchester, VT, USA, 5 Yale University, New Haven, CT, USA Background: There is a growing burden of diabetes in the HIV+ population. In the general population an increased percentage of memory (CD45RO+) T cells is associated with higher diabetes risk, but there are no similar data for HIV+ persons. CD4+ and CD8+ subsets of pro-inflammatory effector memory (TEM) and effector memory RA+ (TEMRA) are described as expanded in HIV+ individuals, often in part in response to human cytomegalovirus. We assessed whether the proportional size of the TEM and TEMRA compartments are associated with an increased risk of incident diabetes in HIV+ and HIV− persons. Methods: We analyzed data on 601 HIV+ and 322 HIV− subjects from the Veterans Aging Cohort Study – Biomarker Cohort, a longitudinal study of HIV+ veterans and matched HIV− veterans. We measured the proportion of 8 classes of CD4+ and CD8+ T cells: memory CD45RO+; TEM cells CD45RO+CD28−, TEMRA cells CD45RA+CD28−CD57+; and CD57+. Incident diabetes was ascertained by a computer algorithm incorporating ICD-9 codes, serum glucose, hemoglobin A1c%, and prescribed medications. Subjects with diabetes at baseline were excluded. We compared the median baseline proportions of T cell subsets among subjects who developed diabetes versus who did not, stratified by HIV status, using Mann-Whitney U tests. Results: Subjects were predominantly male (95%) and African-American (72%). Of the selected T cell subsets, no single phenotype was significantly associated with incident diabetes in both the HIV+ and HIV− persons. As reported in prior studies, HIV- subjects who developed diabetes had higher proportions of CD4+ CD45RO+memory cells. In contrast, HIV+ subjects who developed diabetes had higher proportions of CD8+CD57+ T cells and CD8+ TEMRA cells (see table). We did not adjust for multiple comparisons. Findings were similar in regression models adjusting for age, sex, race, alcohol use, and BMI. Conclusion: In a preliminary analysis, higher proportions of CD8+CD57+ T cells and expansion of the CD8+ TEMRA compartment are associated with increased risk of developing diabetes in HIV+ persons. This differs from the CD4+memory cell association observed in HIV- participants in our cohort and similar studies. The mechanisms through which these two pro-inflammatory CD8+ T cell subsets contribute to glucose intolerance, and the potential role of hepatic dysfunction and hepatitis C, in the context of HIV warrant further investigation.

741 RECOMMENDED ANTIRETROVIRAL REGIMENS AND DIABETES MELLITUS IN THAI HIV-INFECTED ADULTS Ninutcha Paengsai 1 , Gonzague Jourdain 2 , Jean-Yves Mary 3 , Nicolas Salvadori 2 , Tim R. Cressey 2 , Apichat Tantraworasin 1 , Romanee Chaiwarith 1 , Chureeratana Bowonwatanuwong 4 , Sorakij Bhakeecheep 5 , Natapong Kosachunhanun 1 1 Chiang Mai University, Chiang Mai, Thailand, 2 IRD, Chiang Mai, Thailand, 3 INSERM, Paris, France, 4 Mahidol University, Bangkok, Thailand, 5 National Health Security Office, Chiang Mai, Thailand Background: Individual antiretroviral (ARV) drugs such as stavudine and didanosine have been associated with a higher risk of type 2 diabetes mellitus (T2DM) in HIV-infected patients. However, the risk of T2DM associated with different ARV drug combinations remains unclear. We investigated the risk of T2DM in Thai adults using the WHO recommended ARV drug combinations. Methods: We searched all records of HIV infected adults who received antiretroviral therapy within the Thai National AIDS Program from October 1, 2006 to September 30, 2013. Evidence of T2DM diagnosis was defined as either fasting plasma glucose ≥126 mg/dl (following the 2013 American Diabetes Association criteria), or the 2010 WHO ICD10 criteria codes E11-E14, or treatment with anti-diabetic drugs. The T2DM incidence rate was estimated by the number of new diagnoses divided by the total number of person-years of follow-up (PYFU). We analyzed the association between the risk of T2DM and those first line combinations currently recommended by the 2016 WHO HIV treatment guidelines (zidovudine [AZT], lamivudine [3TC], tenofovir [TDF], nevirapine [NVP], efavirenz [EFV] and lopinavir/ritonavir [LPV/r]). We used competing risks survival regression (Fine-Gray), with death consider as a competing event, to identify risk factors of T2DM in univariable and multivariable analyses adjusting for sex, age, history of pancreatitis at baseline and time-updated existence of previous fasting plasma glucose records. Results: Data of 504,027 PYFU from 130,950 patients, 5,878 cases were diagnosed with T2DM. The overall incidence was of 11.7 per 1,000 PYFU (95% confidence interval [CI] 11.4 to 12.0) and 15.4 per 1,000 PYFU (95% CI 14.9 to 15.9) in patients aged 35 to 60 years. 35,731 patients received one of the recommended WHO combinations. In multivariable analysis, where AZT+3TC+NVP was the reference combination, the following combination were associated with a higher risk of T2DM: TDF+3TC+EFV (adjusted sub- distribution hazard ratio 1.6; 95% CI 1.3 to 2.0), AZT+3TC+LPV/r (1.7, 1.2 to 2.4), AZT+3TC+EFV (2.1, 1.7 to 2.5), and TDF+3TC+LPV/r (2.6, 1.7 to 3.9). The risk was not associated with the use of TDF+3TC+NVP 1.0 (1.0, 0.6 to 1.5). Conclusion: The incidence of T2DM was higher than in the general population aged 35 to 60 years (7.8-11.4 per 1,000 PYFU). Several of the WHO recommended ARV regimens may associated with an increased risk of T2DM in Thai adults. 742 GLP-1 AND C-PEPTIDE ARE ASSOCIATED WITH INSULIN RESISTANCE AMONG PLWH IN INDIA Ivan Marbaniang 1 , Shashikala Sangle 2 , Dileep Kadam 2 , Vandana Kulkarni 1 , Dhananjay Shere 1 , Nikhil Gupte 1 , Amita Gupta 3 , Vidya Mave 1 1 BJGMC Clinical Trials Unit, Pune, India, 2 Byramjee Jeejeebhoy Government Medical College, Pune, India, 3 Johns Hopkins University, Baltimore, MD, USA Background: Incretins (Glucagon like peptide-1 [GLP-1] and glucose- dependent insulinotropic peptide [GIP]) regulate glucose homeostasis in HIV- uninfected individuals. Inflammation has been implicated as a cause for insulin resistance (IR) among PLWH. Little is known about IR among Asian Indian PLWH, even though India has the highest burden of diabetes and third highest number of PLWH globally. We aimed to assess the relationship between different hormonal and inflammatory markers with IR in Indian PLWH.

Poster Abstracts

CROI 2018 277