CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

1015 Impact of Unplanned Care Interruption on Immune Recovery After ART Initiation in Nigeria Aimalohi A. Ahonkhai 1 ; Juliet Adeola 2 ; Bolanle Banigbe 2 ; Ifeyinwa Onwuatuelo 2 ; IngridV. Bassett 1 ; Elena Losina 3 ; Kenneth A. Freedberg 1 ; Prosper Okonkwo 2 ; Susan Regan 1 1 Massachusetts General Hospital, Harvard Medical School, Boston, MA, US; 2 AIDS Prevention Initiative Nigeria, Jabi District, Nigeria; 3 Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, US Background: Unplanned care interruption (UCI) from HIV treatment is common in most settings. The clinical impact of UCI has not been well studied in resource-limited settings. Our objective was to determine the immunologic consequence of UCI in the first year on antiretroviral therapy (ART) in Nigeria. Methods: We examined data on adults ( ≥ 15 years) who enrolled in HIV care and started ART at a university-affiliated HIV clinic between 1/2009 and 12/2011. Follow-up was through 12/2012. In this retrospective cohort analysis, UCI was defined as ≥ 90 days with no clinician, laboratory, or pharmacy visits, but later return to care. We categorized patients into 3 groups: 0, 1, or ≥ 2 UCI in the first year on ART. We used multivariate repeated measures linear regression with patient as a random effect. We modeled the change in CD4 count early (1-6 months) and later (7-12 months) after ART initiation in each group, and determined the impact of UCI on predicted CD4. The model was adjusted for CD4 count and tuberculosis diagnosis at ART initiation in addition to patient age and sex. Results: Among the 2,029 patients in our cohort 69%were female, and median age was 32 years [IQR 27, 39], and 4% had tuberculosis co-infection at enrollment. Fifty-four percent of patients had 0, 37% had 1, and 8% had ≥ 2 UCI. Follow-up CD4 was not available for 380 patients within one year of observation. Of the remaining 1,649 patients, mean baseline CD4 cell counts for those with 0, 1, and ≥ 2 UCI were 228/uL [SD 176], 354/uL [SD 228], and 392/ul [SD 241]. Overall, mean CD4 increase was 11 cells/uL/month [95%CI 8-14] in months 1-6, and 4 cells/uL/month [95%CI 2-5] in months 7-12. Patients with 1 UCI gained an average of 52 cells/ul [95% CI 35-68] fewer than those with 0 UCI. Those with ≥ 2 UCI lost 172 cells/uL at one year, ending with substantially lower counts (228/uL, 95%CI 184-271) than 0 UCI (327/uL, 95%CI 320-334), despite starting with much higher counts. Conclusions: UCI was extremely common and occurred in almost half of the patients in a cohort initiating ART in Nigeria. UCI was associated with blunted CD4 cell responses in the first year on ART. Despite initiating ART with the highest counts, CD4 losses were greatest in patients with ≥ 2 UCI, negating the potential benefit of earlier care. Interventions to prevent interruptions in HIV care are critical to ensure the maximal benefits of ART. 1016 Linkage to HIV Care Among MenWho Have Sex With Men and Drug Users in India: Getting to 90 Sunil S. Solomon 1 ; Allison M. McFall 2 ; Aylur K. Srikrishnan 3 ; Gregory M. Lucas 1 ; Canjeeveram K.Vasudevan 3 ; David D. Celentano 2 ; Muniratnam S. Kumar 3 ; Suniti Solomon 3 ; Shruti H. Mehta 2 1 Johns Hopkins University School of Medicine, Baltimore, MD, US; 2 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, US; 3 YR Gaitonde Centre for AIDS Research and Education, Chennai, India Background: UNAIDS has set an ambitious target of 90/90/90 by 2020. While the first 90 assumes 90% of those infected will be aware of their status, the second 90 assumes 90% of those diagnosed are linked to care and initiated on ART. This can be particularly challenging among key populations (men who have sex with men [MSM] and people who inject drugs [PWID]) in resource-limited settings (RLS). Identifying modifiable factors associated with linkage to care in these groups will be critical to achieving this target. Methods: 26,503 individuals (MSM=12,022 and PWID=14,481) were recruited across 27 sites in India (~1000/site) using respondent-driven sampling. Participants had to be ≥ 18 years and self-identify as male and report sex with a man in the prior year (MSM) or injection drug use in the prior 2 years (PWID). 1726 of the 4051 (41%) HIV positive-persons were aware of their status and included. Linkage was defined as ever having visited a health care professional for HIV after diagnosis. We explored whether there were modifiable factors around the time of diagnosis that discriminated between those linked and not linked using multi-level logistic regression and receiver operating characteristic curves (AUC). Results: Median age was 35; 64% of PWID were male. Overall, 80%were linked to care. Among those not linked, 59% had been diagnosed in the past year, 20% 1-2 years ago and 21%>2 years ago. The primary reasons for not seeking care were not being ready/interested (51%) and not knowing where to go (10%). Modifiable factors that best discriminated those who were and were not linked included receiving tangible help with a medical referral at the time of diagnosis such as an appointment/transportation (odds ratio [OR]: 9.2, 95% confidence interval [CI]): 5.2-16.5) and disclosure of HIV status to ≥ 1 person at diagnosis (OR: 2.7; 95% CI: 1.5, 4.8). The AUC for these two factors was 0.85 (Figure 1), which was significantly higher than the AUC for demographics (0.79) or other combinations of modifiable factors. Overall, 91% of those who responded positively to both of these questions were linked vs. only 41% of those who responded no to both.

Poster Abstracts

Conclusions: We identified two simple modifiable factors that could substantially impact linkage to care among MSM and PWID in RLS. Promoting tangible assistance with referral and disclosure at diagnosis are simple easy to implement strategies even in the context of community or home-based testing that could help achieve UNAIDS targets.

597

CROI 2015