CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

454 PATTERNS OF EFAVIRENZ USE AS FIRST-LINE THERAPY IN THE UNITED STATES: 1999–2015

Angela Bengtson 1 , Brian W. Pence 1 , Ellen F. Eaton 2 , Jessie K. Edwards 1 , Joseph J. Eron 1 , W. C. Mathews 3 , Katie Mollan 1 , Richard D. Moore 4 , Connall O’Cleirigh 5 , Michael J. Mugavero 2 1 Univ of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 2 Univ of Alabama at Birmingham, Birmingham, AL, USA, 3 Univ of California San Diego, San Diego, CA, USA, 4 Johns Hopkins, Baltimore, MD, USA, 5 Massachusetts General Hosp, Boston, MA, USA Background: Between 1999 and 2014 efavirenz was recommended as first-line therapy for HIV-infected adults in the US, and continues to be recommended globally by the WHO. However, efavirenz has been linked to suicidal behavior and may not be appropriate for patients with mental illness. Methods: We examined the patterns of initiating efavirenz-containing first line combination antiretroviral therapy (cART) overall and by mental illness status using data from CNICS, a cohort of 31,000 HIV-infected adults in care at 8 sites in the US. Participants were included if they initiated cART between 1999 and 2015 and were followed from cART initiation until initial cART regimen or cART discontinuation, death, loss to care (>12 months with no HIV appointment), or administrative censoring (Oct 2014-Sept 2015, depending on site), whichever date came first. We used multivariable log binomial models to examine factors associated with initiating and discontinuing efavirenz-containing cART. Results: We included 9,775 new cART users. Of those, 4,239 (43%) initiated efavirenz-containing cART; and 772 of those participants (18%) discontinued efavirenz. At cART initiation, 2,492 (25%) of participants had a history of a mental illness associated with suicidal behavior, including depression, psychosis, post-traumatic stress disorder, or obsessive compulsive disorder. Efavirenz initiation peaked in 2007 and declined rapidly thereafter (Figure). Over time, persons with a history of a mental illness were modestly less likely to initiate efavirenz-containing cART, compared to those with no history of mental illness. In a multivariable analysis adjusted for site and year of cART initiation, factors associated with initiating efavirenz-containing cART were: prior mono or dual therapy use (prevalence ratio (PR) 0.74, 95% CI 0.66, 0.83), being female (PR 0.80, 95% CI 0.74, 0.86), intravenous drug use (PR 0.83, 95% CI 0.77, 0.90), history of mental illness (PR 0.79, 95% CI 0.75, 0.84) and CD4 cell count >350 (PR 0.91, 95% CI 0.86, 0.95). History of mental illness (PR 1.18, 95% CI 1.02, 1.37) and CD4 cell count >350 (PR 0.86, 95% CI 0.73, 0.98) were associated with discontinuing efavirenz. Conclusion: Until recently, efavirenz was widely used as first line therapy for HIV-infected adults in the US, including among individuals with a history of mental illness. Given the widespread use of efavirenz globally, greater clarity about the implications of efavirenz use among persons with mental illness is needed.

Poster and Themed Discussion Abstracts

455 ART INITIATION, REGARDLESS OF COMPOSITION, IMPROVES PATIENT-REPORTED DEPRESSION

Ellen Eaton , Ashutosh Tamhane, Michael J. Mugavero Univ of Alabama at Birmingham, Birmingham, AL, USA

Background: More than 50% of persons living with HIV (PLWH) have depression. Co-morbid HIV and depression yield reduced rates of antiretroviral therapy (ART) adherence and higher rates of HIV viral progression and mortality. We hypothesized that patient and ART-specific factors are associated with depression 12 months after ART initiation. Methods: This retrospective study included treatment-naïve PLWH initiating ART between January 2007 and December 2012 at the HIV clinic of the University of Alabama at Birmingham (Alabama, USA). In addition to baseline (at ART initiation) sociodemographic and clinical characteristics, PHQ9 scores (self-reported depression) were obtained both at baseline and while receiving ART at 12-month. Patients were categorized as having depression (PHQ score ≥10) or mild/no depression (score ≤9) and receiving Efavirenz or Non- Efavirenz-based regimens. Associations were examined using univariate and multivariable logistic regression analyses. Results: 291 patients were included: 61 (21%) had depression at 12 months, 83%were male, 60%were Black/African American, and mean age was 35 years. Odds of depression were higher at baseline than at 12 months (30% vs 21%, p=0.003), more so in the Non-Efavirenz regimen (39% vs 24%, p<0.001) than in Efavirenz-based regimen (20% vs 17%, p=0.48). Differential baseline depression between the regimens (20% vs 39%, p<0.001) was indicative of channeling bias. Depression at 12 months was higher among uninsured patients (30/142=27%) than in public (4/22=18%) or privately insured (18/127=14%). Overall, baseline median PHQ score decreased from 6 to 3 at 12 months (Efavirenz: 4 vs 2, Non-efavirenz: 7.5. vs 4.0). In univariate analysis, recipients of the Efavirenz-based regimen were at lower odds of having 12-month depression than Non-efavirenz regimens (OR=- 0.7, 95% CI 0.4-1.2) (Table 1); after adjusting for baseline depression (channeling bias), the difference in 12 month depression between the regimens decreased (OR=0.9; 95% CI: 0.5-1.8). In multivariable analysis, baseline depression (ORadjusted=7.6; 95% CI: 3.7 – 15.5) and lack of insurance (ORadjusted=2.8; 95% CI: 1.3 – 6.0) were significantly associated with greater odds of 12 month depression; no difference was observed between ART regimens (Table 1). Conclusion: In addition to aggressively treating existing depression, initiating ART and enrolling PLWH in insurance has the potential to improve depression and HIV health outcomes.

CROI 2017 190