CROI 2024 Abstract eBook

Abstract eBook

Poster Abstracts

905

ART Adherence And Elevated Viral Load in Pregnant & Postpartum Women Initiating DTG Versus EFV Thokozile R Malaba 1 , Catherine Orrell 2 , Laura Else 3 , Duolao Wang 4 , Catriona Waitt 5 , Angela Colbers 6 , Helen Reynolds 4 , Nengjie He 7 , Lucy Read 4 , Mohammed Lamorde 8 , Saye Khoo 3 , Landon Myer 1 , for the DolPHIN-2 Study Group 1 University of Cape Town, Cape Town, South Africa, 2 Desmond Tutu HIV Foundation, Cape Town, South Africa, 3 University of Liverpool, Liverpool, United Kingdom, 4 Liverpool School of Tropical Medicine, Liverpool, United Kingdom, 5 Infectious Diseases Institute, Kampala, Uganda, 6 Radboud University Medical Center, Nijmegen, Netherlands, 7 Liverpool School of Tropical Medicine, Liverpool, UK, 8 Infectious Disease Institute, Kampala, Uganda Background: There is extensive evidence of non-adherence in pregnant and postpartum women living with HIV (PHIV). But despite the expansion of dolutegravir (DTG) replacing efavirenz (EFV) in first-line ART, there are few data on objective adherence to DTG vs EFV and how non-adherence is associated with elevated viral load (VL) in this population. Methods: The DolPHIN-2 trial (NCT03249181) randomised pregnant women initiating ART from 28w gestation to DTG vs EFV with tenofovir (TDF) and lamivudine/emtricitabine. Within the trial cohort we conducted a nested case-control study to examine adherence to DTG vs EFV using random plasma tenofovir (TFV) levels as an objective adherence measure in both arms (≥35.5ng/mL indicating effective adherence [EA]). Eligible participants had an initial VL>1000 at enrolment and achieved viral suppression (VS <20) during follow-up through 18m postpartum. Case specimens had ≥1 VL>20 after initial VS; control specimens were incidence density sampled from PHIV with persistent VS, matched on ART duration and trial arm (DTG vs EFV). Additional specimens were included from the suppressed visit preceding the first VL>20 (cases) or a time-matched visit (controls) and all visits after viraemia for cases (subsequent visits). Logistic regression, with conditional models for matched data, was used to examine associations between EA and elevated VL. Results: Overall 172 case and 338 control specimens were included from 88 PHIV (mean age, 28y). At preceding visits with VS, EA was higher in DTG compared to EFV (58% vs 42%). Self-reported missed doses (4%) and ARV related side effects (2%) were low and similar by regimen. At the time of VL>20, cases had a mean VL2.95 log 10 copies/mL; EA was observed in 37% of cases compared with 74% of controls with VL<20 at a matched ART duration. Differences were consistent between DTG (OR=6.3; 95% CI=2.3–17.2) and EFV (OR=3.8; 95% CI=1.3-10.8). Among cases, at the preceding visit 46% had EA compared with 37% at viraemic visit (conditional OR=1.4 95% CI 0.7-2.6) with no difference by regimen. At subsequent visits, 86% with detectable TFV achieved VS again with no differences by regimen. Conclusion: The association between objectively measured adherence and viraemia was similar with DTG versus EFV. EA was higher in DTG at visits with VL<20. Taken together, these data suggest that DTG may be associated with better ART adherence compared to EFV but is not more forgiving of the short term non-adherence that occurs commonly during the postpartum period. A Novel Risk Calculator to Predict Sub-Optimal Outcomes Among Pregnant and Postpartum Women With HV Karen Hampanda 1 , Kevin Owuor 2 , Laura K. Beres 3 , Emmah Ouma 2 , Maricianah A. Onono 4 , Anna Helova 5 , Mercelline Onyando 2 , Jeff Szychowski 5 , l L. Abuogi 1 , Janet M. Turan 5 1 University of Colorado Anschutz Medical Campus, Aurora, CO, USA, 2 Kenya Medical Research Institute, Kilifi, Kenya, 3 The Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA, 4 Kenya Medical Research Institute, Nairobi, Kenya, 5 University of Alabama at Birmingham, Birmingham, AL, USA Background: Pregnant and postpartum women living with HIV (PPWH) experience differential treatment success. No tool currently exists to assess a PPWH's cumulative risk of disengagement from care or treatment failure with a quantifiable score during the peripartum period. To identify PPWH at risk and intervene before negative outcomes occur, this study sought to develop and validate a parsimonious risk calculator capable of predicting disengagement from care and treatment failure. Methods: We used a derivation dataset with data from 1331 PPWH enrolled in the Mother-Infant Visit Adherence and Treatment Engagement trial in southwestern Kenya. Least absolute shrinkage and selection operator (LASSO) logistic regression procedures selected the most predictive variables from a list of 16 candidate factors based on prior research, including psychosocial, demographic, and clinical factors. We applied the Minimum Extended Bayes information criterion (EBIC) and 10-fold cross validation methods to find the regularization parameter lambda to give the minimum mean cross-validated

Results: A greater proportion of perinatal transmitters (7 out of 15, 47%) had neutralization breadth compared to non-transmitters (15 out of 47, 32%); these responses mapped to the V2/V3 glycan region for 4 out of 6 (67%) transmitters with neutralization breadth tested so far. Maternal transmitters also trended towards higher ADCC against WITO compared to non-transmitters. Finally, infants seemed to frequently a develop a bNAb response between ages 1 - 3 irrespective of maternal bNAb status, as infants of both transmitters with bNAb activity (4 of 5, 80%), and without ( 7out of 7, 100%), exhibited bNAb activity. Conclusion: Our findings support the existence of increased bNAb activity in transmitters that is frequently specific to a single epitope, which could lead to emergence of bNAb-resistant viral variants that can be transmitted perinatally to the infant, which in turn may contribute to endogenous bNAb responses. Thus, bNAb-based pediatric HIV prevention and treatments that are synergistic with ART will likely need to be multi-specific to effectively eliminate and cure pediatric HIV. Long-term Outcomes After Loss to Follow-Up From PMTCT Services for Women and Children in Kenya John Humphrey 1 , Bett Kipchumba 2 , Edwin Sang 3 , Marsha Alera 3 , Beverly Musick 1 , Lindah Muli 3 , Justin Kipsang 3 , Julia Songok 4 , Constantin Yiannoutsos 1 , Kara Wools-Kaloustian 5 1 Indiana University, Indianapolis, IN, USA, 2 Moi Teaching and Referral Hospital, Eldoret, Kenya, 3 Academic Model Providing Access to Healthcare, Eldoret, Kenya, 4 Moi University, Eldoret, Kenya, 5 Indiana University, Bloomington, IN, USA Background: Many prevention of mother-to-child HIV transmission (PMTCT) studies assess outcomes within a year post-delivery and exclude patients who became lost to follow-up (LTFU) or transferred out, biasing outcomes toward those retained in care at the facility where they first enrolled in PMTCT services. Methods: We recruited women living with HIV (WLH) ≥18 years that enrolled in antenatal clinic (ANC) at five public facilities in western Kenya. WLH retained in care (RW) were recruited during the 3rd trimester of pregnancy and followed with their children through 6 months post-delivery; WLH who became LTFU (LW, last visit >90 days) after ANC enrollment and before 6 months postpartum were recruited through community tracing. Re-contact at 3 years post-delivery was attempted for all WLH, using community tracing for WLH LTFU (>60 days since last missed scheduled visit before 36 months) and transferred. Primary outcomes of retention in care and child HIV-free survival were determined at 6 months and 3 years post-delivery. Results: 333 WLH were recruited from 2018-2019. At 6 months postpartum, 222 WLH were classified as RW and 111 as LW (79 disengaged from care, 32 silently transferred/retained elsewhere). More LW compared to RW were newly diagnosed with HIV at ANC enrollment (50% vs. 24%), not virally suppressed at study enrollment (40% vs. 8%), and miscarried (12% vs. 1%) (p<0.01 for all). HIV-free survival at 6 months was lower for children of LW vs. RW (88% vs. 99%, p<0.01). At 3 years, 230 WLH were retained at the study facility (81% of RW, 46% of LW), 30 officially transferred out (28 retained at a new facility, 2 unknown), 70 LTFU (8 silently transferred/retained elsewhere, 19 disengaged, 43 unknown), and 3 deceased. Child HIV-free survival at 3 years was 82% (59% for children of LW, 92% for RW), 3.7% were living with HIV (11% LW, 0.4% RW), 3.7% were deceased (7% LW, 2% RW), and 11% had unknown HIV/vital status (23% LW, 5% RW). Conclusion: HIV-free survival was lower for children of LW compared to RW at 6 months and 3 years post-delivery, emphasizing the need for interventions targeting early loss to follow-up from PMTCT services. Although most LW had re-engaged in care by 3 years, many remained LTFU and tracing-ascertained engagement in care was lower for WLH silently vs. officially transferred. Community tracing of patients who become LTFU can inform PMTCT outcome estimates and service delivery priorities for this population.

904

Poster Abstracts

906

CROI 2024 283