CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

Viremia outcomes were grouped as undetectable (≤ 20 copies/ml), low level suppressed (21-200 copies/ml), or not suppressed (>200 copies/ml). Results: The average age of included participants was 34 years with 38.5% African American and 37.2% Hispanic/Latino. The average years since HIV diagnosis was 8 years. The prevalence of substance use across 825 study visits was 73%, with methamphetamine (MA) use most prevalent (50%). After adjusting for unstable housing and ART adherence, MA use, either alone (adjusted OR=1.87; 95% CI 1.03-3.40) or with other substances (adjusted OR=1.82; 95% CI 1.12-2.95), was associated with higher odds of increasing viremia categories (low level suppressed 21-200 copies/ml; not suppressed >200 copies/ml) compared to non-substance users. Other substance use excluding MA did not show a similar association (adjusted OR=1.29; 95% CI 0.80-2.09). These findings suggest that among MA users, nearly half the instances of viremia would be reduced if MA was discontinued (attributable fraction=46%; 95% CI 3-71%). Conclusion: MA use, either alone or in combination with other drugs, is associated with failure of viral suppression among HIV-positive MSM on ART independent of adherence and sociodemographic factors. This accounts for nearly half of the observed instances of unsuppressed viremia. In contrast, other substance use does not impose the same risk. This study underscores the importance of MA use on clinical outcomes among people living with HIV. 885 METHAMPHETAMINE DOSE, CLINICAL OUTCOMES, AND HIV STATUS LINKS AMONG DIVERSE MSM IN LA Steven Shoptaw 1 , Marjan Javanbakht 1 , Amy Ragsdale 1 , Risa Flynn 2 , Robert Bolan 2 , Raul Mandler 3 , Pamina M. Gorbach 1 1 University of California Los Angeles, Los Angeles, CA, USA, 2 Los Angeles LGBT Center, Los Angeles, CA, USA, 3 National Institute on Drug Abuse, Rockville, MD, USA Background: Men who have sex with men (MSM) who use methamphetamine (meth) report patterns of use that accumulate to varying amounts of exposures each month. Reliable patterns are: users who dose daily, users for several days once or twice monthly (“weekend warriors”) and users who rarely use over several months. Our hypothesis posits cumulative reported meth exposure over 6 month periods links significantly and in dose-dependent fashion with poorer physical and mental health outcomes and with confirmed HIV serostatus. Methods: Data were baseline and follow-up visits (1,798 visits) from 529 mSTUDY participants (HIV+ 266; HIV- 263) in mSTUDY, a prospective cohort of diverse MSM in Los Angeles. Analyses tested links between pattern of self- reported past meth use (past 6 months) with other substance use and select physical and mental health status variables. Reported meth use data at each visit were: (none=1,116 visits; ≥ weekly=330 visits; ≤monthly=352 visits). Results: Univariate analyses supported our hypothesis of an ordered dose- response association over 6-month periods between outcomes of cumulative dose of meth use and other drug use, HIV-risk behaviors, HIV status, STIs and clinical conditions. By contrast, any reported meth use significantly correlated with HIV-seropositive status (see Table). In multvariable analyses, visits where MSM reported ≥ weekly meth use (compared to nonusers) showed significantly higher likelihood of clinical hepatic (AOR 2.4, 95%CI 2.2, 5.4), neurologic (AOR 2.1, 95%CI 1.2, 3.4), psychologic (AOR 1.6, 95% CI 1.2, 2.3) and renal abnormalities (AOR 2.2, 95% CI 1.1, 4.3). In visits where MSM reported ≤ monthly meth use, lower, but significantly higher odds than non-users were observed for all conditions, though less than visits where ≥ weekly meth use was reported (excepting renal conditions). Conclusion: Findings show a robust and ordered signal between reported cumulative meth use and physical and mental health outcomes. By contrast, any reported meth use linked with HIV-positive serostatus. Findings also show a strong signal between level of reported meth use over 6 month periods and likelihood of hepatic, neurologic, psychologic and renal abnormalities, suggesting a dose-response link between cumulative dose of meth and negative health impacts in MSM in urban areas similar to Los Angeles.

Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA Background: Overdose mortality has been increasing in the United States for the past decade. People who inject drugs (PWID) with HIV infection may have heightened overdose risk due to a higher burden of age-related comorbidities. Moreover, the context of drug use has changed across the US. We characterize trends in fatal and non-fatal overdose and associations with HIV infection among a community-based cohort of PWID in a city with a long-standing opioid epidemic. Methods: The AIDS Linked to the IntraVenous Experience (ALIVE) cohort has followed PWID in Baltimore since 1988. We characterize the incidence of fatal and non-fatal overdose from 2014-2018 among all PWID enrolled in ALIVE who were alive in 2014. Mortality through 2016 was ascertained via linkage to the National Death Index with death certificate confirmation. Deaths classified as overdose/drug-related were examined using survival methods with censoring (administrative, other causes of death). Non-fatal overdose was ascertained via self-report among PWID actively using drugs with >1 study follow-up visit on or after 2014. Poisson regression was used to evaluate time trends and covariate associations. Results: Of 3,156 PWID enrolled in ALIVE who were living at the start of 2014, the median age was 54, 28%were female, 79%were African-American. 51 died of a drug-related cause between 2014-2016 (mortality rate: 6.2 per 1,000 person-years). PWID with HIV were at significantly higher risk of mortality from a drug-related cause (aHR = 2.40, 95% CI: 1.33-4.31) compared to HIV-negative PWID, adjusting for age, sex, and race. Overall, between 2014-2018 among 1104 in follow-up, 194 PWID experienced 530 non-fatal overdoses (rate: 28.5 per 100 person-years). Moreover, rates increased significantly between 2014 and 2018 (Figure) from 8.0 to 44.5 per 100 person-years (p-value for trend<0.0001). HIV- infection was associated with decreased risk of non-fatal overdose controlling for age, race and sex (adjusted incidence rate ratio (aIRR) =0.64, 95% CI: 0.42- 0.98). There were no differences in non-fatal overdose by viral suppression for HIV-positive PWID. Conclusion: While HIV-infected PWID in this cohort appear less likely to experience drug overdose, they may be at higher risk for drug-related mortality. Concurrently, overall rates of overdose are increasing among all PWID. Additional efforts are needed to mitigate the impact of non-fatal overdose among all PWID and fatal overdose among HIV-positive PWID.

Poster Abstracts

884 ATTRIBUTABLE RISK OF METHAMPHETAMINE USE ON VIRAL SUPPRESSION AMONG MSM ON ART

Jennifer A. Fulcher 1 , Marjan Javanbakht 1 , Chelsea L. Shover 2 , Amy Ragsdale 1 , Ron Brookmeyer 1 , Steven Shoptaw 1 , Pamina Gorbach 1 1 University of California Los Angeles, Los Angeles, CA, USA, 2 Stanford University, Stanford, CA, USA Background: Viral suppression improves clinical outcomes, while virologic failure (HIV RNA >200 copies/ml) increases risk of HIV morbidity/mortality. Substance use decreases the likelihood of achieving undetectable viremia; however, the comparative effects by substance have not been described. In this study, we examine the effect of different drugs on levels of viremia in a cohort of HIV+men who have sex with men (MSM) on antiretroviral therapy (ART). Methods: Participants (N=230) were selected from an ongoing cohort (The mSTUDY) of diverse young MSM enrolled from August 2014 to May 2018. Only participants who were HIV+ currently on ART as reported in review of concurrent medications and self-report were included. Plasma HIV RNA copies were measured at semi-annual visits and substance use over the preceding six months assessed by computer-assisted self-interview. Substance use and sociodemographic factors associated with viremia outcomes were assessed using ordinal regression analysis with generalized estimating equations.

CROI 2019 344