The Shifting Picture of HIV Treatment, Comorbidity and Substance use Among US MSM Living With HIV

G D'Souza; L Benning; V Stosor; MD Witt; J Johnson; M Friedman; AG Abraham


HIV Medicine. 2021;22(7):538-546. 

In This Article


The Multicenter AIDS Cohort Study (MACS) is an observational cohort study of MSM, including those living with HIV as well as HIV-uninfected (at risk) MSM, in cities across the US (Baltimore, MD; Washington, DC; Chicago, IL; Los Angeles, CA; Columbus OH; and Pittsburgh, PA),[14] now part of the MACS-WIHS Combined Cohort Study ( Participants report data on health, behaviour and HIV treatment history and undergo physical examinations at semi-annual visits. The MACS collects data on HIV, including clinical measures of HIV disease and ART initiation and use, allowing comparison of these measures among more recent initiators and those who initiated ART under earlier treatment guidelines, and regimens.

The MACS has had several waves of new enrolment since it began in 1984, including enrolment in 2001–03 and 2010–2018. Men were recruited from medical centres as well as community-based and AIDS-service organizations, healthcare organizations, academic institutions and local and state government clinics. Enrolment in 2010–18 targeted recent HIV seroconverters who either had not yet initiated therapy or had recently initiated therapy. Given the 'test and treat' policy during this time, many were enrolled after recent therapy initiation, and enrolment criteria for these included: (1) first started therapy after 1 January 2011 (to ensure they were treated with newer ART regimens); (2) had a known date of initiation; and (3) had documented CD4 cell count and HIV viral load at initiation. Each participant gave informed consent, and each local institutional review board has reviewed and approved the study. There were four trans women in the analysis, all others were cis-gendered men.

Effective ART Definition

The definition of ART was guided by the DHHS/Kaiser Panel[2] and included regimens with three or more unboosted ART drugs, or a boosted agent (cobisistat or boosted ritonavir) with two additional ARTs consisting of two protease inhibitors (PIs) or at least one of the following: nonnucleoside reverse transcriptase inhibitor (NNRTI), NRTI, integrase inhibitor (II) or entry inhibitor (EI; including fusion inhibitors).

Inclusion Criteria and Matching Algorithm

The study sample of recent initiators was drawn from men in the MACS who initiated ART between 2010 and 2018. These recent initiators were matched by age (± 5 years) and race (black/white/other) to men at the same study site who initiated ART during the previous decade, 2000–2009 (early initiators). To be included, men had to have a CD4 cell count and HIV viral load measurement at or before ART initiation [two recent ART initiators without this pre-HAART (highly active antiretroviral therapy) information were excluded from analysis]. To maximize the sample, recent initiators were matched to all eligible early initiators at the same study site who were in the same race and age category, so matching was not one to one.

Covariate Measurements

CD4 cell count and HIV viral load values were drawn from the visit at treatment initiation, when available, otherwise from the closest semi-annual visit before initiation. Other covariates were taken from the closest visit before or after ART initiation. Race, smoking, alcohol use, drug use, quality of life and sexual behaviours were ascertained from self-reported data collected using audio computer self-interview (ACASI). Body mass index was recorded from the physical examination. HIV clinical characteristics were obtained through centralized blood testing including CD4 lymphocyte counts measured using standardized three-colour flow cytometry, and plasma HIV-1 RNA levels measured using Roche assays (Hoffman-LaRoche, Nutley, NJ, USA).

The prevalence of several comorbidities was also evaluated. Diabetes was defined as a fasting glucose ≥ 126 mg/dL or self-reported diagnosis with use of anti-diabetic medication. Kidney disease was defined as estimated glomerular filtration rate < 60 mL/min/1.73 m2 (modification of diet in renal disease) or urine protein ≥ 200 mg/g of creatinine. Hypertension was defined as having systolic blood pressure ≥ 140 mmHg, diastolic blood pressure ≥ 90 mmHg or self-reported diagnosis and documented use of anti-hypertensive medication. Liver disease was defined as having alamine amino transferase or aspartate amino transferase > 150 IU/L. Metabolic syndrome was defined as the presence of three or more of the following five diagnostic features: (1) waist circumference ≥ 102 cm, (2) fasting triglicerides ≥ 150 mg/dL, (3) diabetes, (4) high-density lipoprotein < 40 mg/dL, and (5) hypertension.

Frailty-related phenotype was defined as the presence of three out of the following four self-reported conditions: (1) unintentional weight loss of 4.54 kg (10 lb) or more, (2) difficulty with work or activity due to physical health, (3) inability to walk several blocks, limited by health, and (4) inability to participate in vigorous activities, limited by health.

Hepatitis B status was defined as negative if hepatitis B surface antigen (HBsAg) and core antibody (HBcAb) were negative at all visits tested; defined as resolved if HBsAg-negative and HBcAb-positive at the current visit and HBsAg-positive and/or HBcAb-positive at two or more prior visits; and defined as chronic if HBsAg-positive. Hepatitis C status was defined as positive if there was a reactive HCV antibody or detectable HCV RNA level, and negative otherwise.

Measurement of HIV-related Characteristics

HIV viral suppression following ART initiation was defined as having at least one undetectable HIV RNA measurement within 12 months of starting ART; for this analysis, undetectable was consistently defined as < 50 copies/mL given the change from < 50 copies/mL (Ultra-sensitive 2nd Generation; Roche) to < 20 (COBAS TaqMan or COBAS; Roche) during the time period evaluated. Evidence of virological failure following viral suppression was defined as having any measure > 200 copies/mL within 2 years after starting ART (to allow for viral detection variation, i.e. 'blips').

CD4 cell count and HIV viral load were available in 36% of the sample from medical abstraction at the exact time of treatment initiation. The other subjects had these measures from the closest visit prior to initiation, which was within 6 months before initiation for 58%, within 12 months for 4% and within 18 months for 2%. Other covariates were taken from the closest visit prior to the estimated date of ART initiation where available (n = 178, 53% of men), which was a median of 3 months before ART initiation [interquartile range (IQR): 2.3–3.4 months], and was within 6 months before initiation for 49%, within 12 months for 2% and within 18 months for 1%; otherwise at the first visit following ART initiation (n = 160, 47% of men), which was a median of 4.4 months after initiation [IQR: 2.8–8.3 months] and was within 6 months before initiation for 30%, within 12 months for 17% and within 18 months for 1%.

Statistical Analysis

Characteristics of each group were described with percentages for categorical variables and median (IQR) for continuous variables. HIV-related characteristics were compared across time periods of ART initiation using a test of proportions or a test of medians as appropriate. Prevalence of HIV suppression, therapy failure, sexual behaviour, substance use, comorbidity, depression and quality of life were explored among those with data for each characteristic. Prevalence was compared in recent and earlier initiators using χ 2 or Fisher's exact test for categorical variables, as appropriate. HIV-related characteristics were compared across time periods of ART initiation using Wilcoxon two-sample tests to compare medians for continuous variables. Odds ratios were calculated using logistic regression. The analysis was conducted using SAS v.9.4. All data were collected under institutional review board-approved MACS protocols at all study sites.