Decline in CD4 T lymphocytes with monotherapy bridging strategy for non-adherent adolescents living with HIV infection: Results of the IMPAACT P1094 randomized trial

Introduction: Management of persistently non-adherent youth living with HIV (YLHIV) with virologic failure (VF) on combination antiretroviral therapy (cART) remains challenging. One strategy has been using 3TC/FTC monotherapy (3TC/FTC), which in the presence of the M184V resistance mutation, does not suppress viral replication nor select for additional drug resistance mutations, and reduces viral fitness with limited side effects. P1094 compared the immunologic outcome of continuing failing cART vs. switching to 3TC/FTC as a “bridging strategy” to subsequent suppressive cART for non-adherent YLHIV with pre-existing M184V resistance. Materials & methods: Participants with documented nonadherence, M184V mutation, CD4+ T cell count ≥100 cells/mm3 and VF (HIV-1 plasma RNA ≥400 copies/mL (2.6 log10 HIV-1 RNA) were enrolled and randomized to continue failing cART vs. switch to 3TC/FTC. The primary endpoint (time to ≥30% CD4+ T cell decline or development of CDC class C events) at 28-weeks were assessed by Kaplan-Meier (K-M) curves in an intent-to-treat analysis. Results: Thirty-three perinatally acquired YLHIV participants (16 continuing cART and 17 3TC/FTC) enrolled in the study. The median age, entry CD4+ T cell count, and viral load were 15 years (Inter-quartile range (IQR) 14-20), 472 cells/mm3 (IQR 384-651), and 4.0 log10HIV-1 RNA copies/ml (IQR 3.2-4.5), respectively. Five participants, all in the 3TC/FTC arm, reached the primary endpoint for absolute CD4+ T cell decline (p = 0.02, exact log-rank test comparing monotherapy to cART). The Kaplan-Meier estimate of probability of primary endpoint on 3TC/FTC at 28 weeks was 0.41 (standard error 0.14). There were no CDC class C events or deaths and no statistically significant difference in frequencies of adverse events between the arms. Conclusions: Non-adherent participants randomized to 3TC/FTC were more likely than those maintained on failing cART to experience a confirmed decline in CD4+ count of ≥30%. Although this study suffers from limitations of small sample size and premature discontinuation, the randomized comparison to continuing failing cART indicates that 3TC/FTC provides inferior protection from immunologic deterioration for non-adherent youth with M184V resistance. Better alternatives to 3TC/FTC such as ART with higher barriers to resistance and novel adherence and treatment strategies for nonadherent youth are urgently needed. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. International Maternal Pediatric Adolescent AIDS Clinical Trials (IMPAACT) P1094 was a multi-center (domestic and international), open-label, Phase IV randomized, controlled, comparative trial to evaluate continuation of non-suppressive cART (Arm A) vs. 3TC/FTC monotherapy (Arm B) for 28 weeks (in both arms) ({“type”:”clinical-trial”,”attrs”:{“text”:”NCT01338025″,”term_id”:”NCT01338025″}}NCT01338025).[23] All of the participating sites are part of the IMPAACT network of clinical research sites. Participants were (perinatally or non perinatally) HIV-infected children, adolescents, and young adults ≥8 to <25 years old, on a non-NNRTI (non-nucleoside reverse transcriptase inhibitor) regimen for at least 6 months, with documented nonadherence (e.g., self-report or pharmacy report, despite documented interventions to improve nonadherence), M184V resistance mutation at or prior to screening on standard, clinically available genotyping assays, absolute CD4+ T cell count ≥100 cells/mm3, and VF (defined as HIV-1 plasma RNA ≥400 copies/mL on at least two occasions including screening at least 2 months after initiating the regimen and within 6 months of study entry). When slow enrolment was observed, a site survey was conducted and in response, the following changes were made to the eligibility criteria (March 7 2012): absolute CD4+ T cell count was lowered from 250 to 100 cells/mm3 and the time demonstrating failure from 6 months to 2 months. Main exclusion criteria included hepatitis B infection, active malignancy within the prior two years, active opportunistic or other infection; ≥ 1 CDC class C event within 12 months of entry, or viral load >250,000 copies/mL at screening. Participants were enrolled between May 2011 and January 2013 and randomized 1:1 to the two treatment groups using a dynamic permuted block system with institutional balancing (maximum institutional imbalance of 2), stratified by CD4+ T cell count (<400 cells/mm3 versus ≥400 cells/mm3). Participants were enrolled online and randomized at time of enrollment via computer algorithm at the Data Management Center. For participants assigned to Arm A, the protocol required that the participant continue their current ART regimen, which was prescribed and monitored by the participant’s provider. Participants assigned to the monotherapy arm (Arm B) received either 3TC or FTC (the choice of 3TC or FTC was left to the provider) and discontinued other ART. Sites were expected to provide and document adherence support measures (per their standard of care) during the study. The target sample size of 344 was chosen to provide 80% power to detect a difference of 0.15 between study arms in proportions of participants with immunologic deterioration using a log-rank test with a 2-sided Type I error rate of 0.05. Due to poor accrual, the final sample size was 32 eligible participants. The antiretroviral agents were not provided by the study. The study was approved by the Institutional review boards (IRBs) of all participating clinical research sites (CRS) (Bronx-Lebanon CRS, New York, United States; Chiang Mai University Pediatrics-Obstetrics CRS, Thailand; Children's National Med. Ctr. Washington DC NICHD CRS, United States; Duke University Medical Center CRS, Durham, North Carolina, United States; Hosp. General de Agudos Buenos Aires Argentina NICHD CRS; Hosp. Geral De Nova Igaucu Brazil NICHD CRS; Inst de Infectologia Emilio Ribas, Sao Paulo Brazil NICHD CRS; and Johns Hopkins Univ. Baltimore NICHD CRS, United States. All participating sites received ethics approval prior to study initiation at their site. Approvals were received from April 13, 2011 to March 7, 2012. Participants and their parent/guardian, where relevant, signed written informed consent, which was documented in the participant’s chart and/or study log as directed by their local IRB requirements. A data safety monitoring board (DSMB) monitored the study conduct throughout the study. Study measurements included lymphocyte subsets at each study visit (0, 4, 12, 20, 28 weeks), HIV RNA testing (0, 4, 12, 28 weeks), and safety monitoring labs including hematology (0, 4, 12, 20, 28 weeks and chemistries (0, 12, 28 weeks). Plasma and peripheral blood mononuclear cells (PBMCs) for inflammatory and immune activation markers were collected at 0, 12, and 28 weeks. Self-reported adherence was also assessed at each study visit by collection of 3-day recall and evaluation of interventions employed to enhance adherence. Batched resistance testing was planned as part of the protocol, however, this was not performed when the study ended early. Plasma and Ficoll-Hypaque density gradient isolated peripheral blood mononuclear cells (PBMCs) were cryopreserved within 8 hours of blood draw. Plasma markers of inflammation and immune activation were determined with ELISA kits, using duplicate samples on the same plate per subject, for D-Dimer (Imuclone/Sekisui Diagnostics, Lexington, MA), sVCAM-1 (Millipore, Billerica, MA), and soluble CD4 (sCD14), interleukin-6 (IL-6), and high sensitivity C-reactive protein (hsCRP) (R&D Systems, Minneapolis, MN). These markers were selected based on correlation with morbidity and mortality in studies of HIV-1 infected adults[23–27]. Multi-parameter flow cytometric analysis (LSRII, Becton Dickinson, Franklin Lakes, NJ; FlowJo Version 9.6.1, Treestar, Ashland, OR) of cryopreserved PBMC stained with CD3-FITC and CD4-Qdot605 (Invitrogen/Life technologies, Grand Island, NY), CD8-AF700 and HLA-DR-APC (BD Biosciences, San Jose CA), CD38-PE, CD95-Pacific Blue, CD31-PE-CY7, CD27-APC-CY7, CD45RO-PE-CY5 (Biolegend, San Diego, CA), and LIVE/DEAD Fixable Aqua Dead Cell Stain (Invitrogen/Life technologies, Grand Island, NY) was performed. The percent of naive (CD27+ CD45RO- CD95-) and activated (CD38+ HLA-DR+) cells within total CD3+/CD4+ and CD3+/CD8+ viable lymphocytes was determined. In addition, the percentage of recent thymic emigrants (RTE) (CD31+ CD27+ CD45RO- CD95-) among CD3+/CD4+ lymphocytes was determined[28]. The primary endpoint was time to time to immunologic deterioration, defined as first ≥30% decline in absolute CD4+ T cell count from baseline or development of CDC class C events. Additional analyses were done on rates of adverse events or deaths, change in viral load, changed in biomarkers of inflammation and immune activation, and adherence at the end of follow-up; as well as baseline immune characteristics on the 3TC/FTC monotherapy arm. Kaplan-Meier curves and log-rank tests were used to compare the treatment arms with respect to time to immunologic failure.[29] The probability of avoiding immunologic deterioration by 28 weeks was estimated via the method of Kaplan and Meier, and the 95% confidence intervals (CIs) for this probability were computed for each arm. Linear mixed effects models were used to estimate, in an exploratory manner, rates of change in CD4+ T cell counts and viral load levels over time for the two arms.[30] Three-day recall of adherence was categorized as missed doses in the past 3 days (yes/no) at entry and 28 weeks. The proportion of participants at week 28 that were in the different categories of adherence, as measured by 3-day recall, was compared between the two study arms using Fisher’s exact test. Lastly, as an exploratory objective, changes in immune activation, inflammatory biomarkers, and LDL and HDL cholesterol were assessed for participants in the 3TC/FTC arm followed at domestic sites only, and compared between groups formed by presence or absence of immunologic deterioration by week 28 using Kruskal-Wallis tests. All analyses were intent-to-treat and included all eligible participants unless otherwise specified. All tests were done using SAS 9.4, StatXact.