CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

age and black race were also associated with poor retention and concurrent viremia. Conclusion: In one of the first multi-site, clinic-based studies of HIV stigma in the US, we found that higher stigma had a modest but statistically significant independent effect on the likelihood of poor retention and concurrent viremia. These findings suggest that ameliorating stigma will be necessary to optimize HIV outcomes in pursuit of “getting to zero.” Future research will examine prospective associations between stigma and HIV outcomes and the role of potential mediators such as depression and antiretroviral adherence.

to ensure viremic patients are provided linkages to additional adherence and support services and that stable patients are correctly identified.

Poster Abstracts

1129 HIGH VIREMIA AMONG ‘STABLE’ PATIENTS RECEIVING ANTIRETROVIRAL THERAPY IN ZAMBIA Mpande Mukumbwa-Mwenechanya 1 , Emilie Efronson 1 , Paul Somwe 1 , Jillian L. Kadota 1 , Monika Roy 2 , Samuel Bosomprah 1 , Kelvin Kaputula 1 , Jake Pry 1 , Nancy Padian 3 , Elvin Geng 2 , Charles B. Holmes 4 , Izukanji Sikazwe 1 , Anjali Sharma 1 , Carolyn Bolton Moore 1 1 Centre for Infectious Disease Research in Zambia, Lusaka, Zambia, 2 University of California San Francisco, San Francisco, CA, USA, 3 University of California Berkeley, Berkeley, CA, USA, 4 Georgetown University, Washington, DC, USA Background: Despite widespread availability of ART in Zambia, only 59.8% HIV-positive adults 15-59 years old on ART were found to have viral load (VL) suppression (<1000 copies of viral RNA/ml) in a recent community based study (1). This high level of silent non-suppression threatens to undermine the country’s ability to attain the UNAIDS 2020 treatment target of 90-90-90. 1.Zambia Population Based HIV Impact Assessment, 2015-2016 Methods: A systematic sample of eligible stable patients (HIV+, on ART > 6 months, not acutely ill, CD4 >200/µl (if available)) in two purposively selected urban and heavy clinic patient load (>4500) sites were selected to assess the FastTrack model (accelerated pharmacy pick-ups to receive ART every 3 months (FastTrack model)). We collected dry blood spots (DBS) to test for baseline VL and extracted current socio-demographic and, clinical and pharmacy data retrospective to 12 months from study enrolment. Bivariate and backward stepwise logistic regression analyses were conducted to assess viral suppression rates and factors associated with odds of unsuppressed HIV VL at enrollment. Results: Of the 407 enrolled, 308 patients had complete clinical data; 73% were female (n=225), median age was 34.7 years (IQR: 29.7-39.9) and median time on ART was 2144 days (IQR: 1087-2947). 43.5% (n=134) had no CD4 test result within the last 6 months and were enrolled based on clinical judgement of ‘stability.’ At enrollment into the FastTrack model, 40.3% (n=124) had >1000 copies of viral RNA/Ml. Analyses showed no significant association of unsuppressed HIV VL with gender, age, WHO staging, time on ART and missed pharmacy appointments. Odds of unsuppressed VL was greater for those in WHO clinical stage 2 (OR: 1.25 (95% CI: 0.65, 2.42) and stage ≥3 (OR: 1.47 (95% CI: 0.87, 2.50) compared to those in stage 1, although effect sizes were statistically insignificant. Conclusion: Rates of VL non-suppression among patients identified as stable were high and did not appear to have a specific age/ gender distribution. Clinical judgment combined with CD4 did not identify a large proportion of at risk patients. Strengthening access to routine VL testing is essential in order

1130 COMPLIANCE TO GUIDELINES FOR ROUTINE VIRAL LOAD TESTING IN RESOURCE LIMITED SETTINGS Barbara Castelnuovo 1 , Frank Mubiru 1 , Mark Nsumba 1 , Isaac Lwanga 1 , Ivan Lumu 1 , Agnes Kiragga 1 , Steven J. Reynolds 2 1 Infectious Disease Institute, Kampala, Uganda, 2 NIH, Bethesda, MD, USA Background: To achieve the final 90 of the UNAIDS 90-90-90 targets, it is critical that timely viral load (VL) testing is offered to all patients on ART. The WHO guidelines recommend VL every 12 months; patients with VL≥1000 copies/ml should repeat VL after 6 months and be promptly switched to 2nd line ART if treatment failure is confirmed. We aimed to evaluate the compliance to the WHO guidelines in implementing VL testing and managing patients according to guidelines in a large urban clinic in Uganda. Methods: This analysis included all patients at the Infectious Diseases Institute in Kampala on 1st line ART after December 2014 (date of VL monitoring implementation). Patients not yet due for a repeat VL at database closure were excluded. We describe the “cascade” of VL management by reporting the proportion who received a VL test. Among those with VL15 months), and never. For those with VL≥1,000 copies/ml we evaluated the proportion with a VL repeated within 7 months (6 months per guidelines+ 1-month window for patients on 1-month prescription), late (>7 months), and never. For those with 2 VL≥1000 copies/ml we evaluated the proportion promptly switched to 2nd line defined as within 3 months, late (>3 months) and not switched. Results: Among 9599 registered patients, 6,893 were eligible; 61.0%were female, median age at ART start was 36 years (IQR: 30-42), CD4 count was 166 (IQR: 68-293) cell/µL and median(IQR) duration on ART was 41(IQR: 17-87) months 6,357 (92.6%) had at least 1 VL, of which 512 (8.1%) was ≥1000 copies/ ml. Figure 1 shows compliance to WHO guidelines. Among patients with VL<1000 copies/ml, VL was repeated ≤ 15 months in 73.2% and totally in 89%. In patients with VL≥1000 copies/ml VL was repeated in 56.1% (287/512) within 7 months, totally in 86.3%, and was confirmed ≥1000 copies/ml in 249 (56.3%). Of the latter only 165 (66.3 %) were switched to 2nd line within 3 months with a total of being switched of 86.4%. Details for all the other patients are shown in the figure. Conclusion: We found a high rate of viral suppression (92%) in a population where the majority of the patients had been on long term ART with no VL monitoring. Compliance to timing of repeated VL in any category (VL < and ≥1000 copies/ml) was sub-optimal, however the proportion of patients for which no action was taken at the different steps was low.

CROI 2018 436