Secondary Logo

Journal Logo

Rapid Communication

Increased Persistence of Initial Treatment for HIV Infection With Modern Antiretroviral Therapy

Davy-Mendez, Thibaut MSPH; Eron, Joseph J. MD; Zakharova, Oksana MS; Wohl, David A. MD; Napravnik, Sonia PhD

Author Information
JAIDS Journal of Acquired Immune Deficiency Syndromes: October 1, 2017 - Volume 76 - Issue 2 - p 111-115
doi: 10.1097/QAI.0000000000001481
  • Free



Initiating antiretroviral therapy (ART) early in HIV infection improves clinical outcomes and prevents transmission.1,2 US treatment guidelines have changed with the availability of newer ART agents and currently recommend starting ART with a combination of 2 NRTIs, and an integrase strand transfer inhibitor (INSTI), or a boosted protease inhibitor (bPI).3 The effectiveness of INSTI-containing regimens compared with previously available regimens has not been well characterized. In this study, we compared response to initial ART in the clinical setting including continuation of the initial regimen, known as persistence or durability, and virologic response.


All patients initiating ART in the University of North Carolina (UNC) Center for AIDS Research HIV Clinical Cohort, 1996–2014, were included. This prospective cohort of all primary HIV care patients at the UNC Hospitals is representative of HIV-infected patients in care in North Carolina.4 Patients were followed from the first ART initiation (baseline) until the first of outcome, loss to follow-up, death, or November 2015. The 2 primary outcomes evaluated were ART discontinuation, defined as change in anchor agent class or stopping ART for longer than 2 weeks, and virologic failure, defined as the first HIV RNA level ≥400 copies/mL after 24 weeks of therapy in an intention-to-treat approach, where patients lost to follow-up with HIV RNA <400 copies/mL were censored and changes in therapy were ignored. ART regimen categories were based on anchor agent: INSTI (any INSTI), bPI (ritonavir-boosted atazanavir, darunavir, or lopinavir), non–nucleoside reverse transcriptase inhibitor (NNRTI, efavirenz, or rilpivirine), other (including unboosted and other bPI), and NRTI (regimens including only NRTIs). Patients provided written informed consent to participate in the clinical cohort, and the UNC Institutional Review Board approved both the cohort study and this secondary data analysis.

Separate time-to-event analyses were performed for each outcome of interest, including Kaplan–Meier survival curves and Cox proportional hazards models adjusted for baseline age, sex, race, CD4 cell count, HIV RNA level, and calendar year. We excluded patients missing baseline CD4 or HIV RNA measurements. In a sensitivity analysis, we included patients missing baseline measurements and used multiple imputations with Markov Chain Monte Carlo and 50 imputations based on age, sex, race, men who have sex with men status, injection drug use, year of ART initiation, and ART regimen. We included the calendar year using disjoint indicator variables for the periods 1996–2000, 2001–2005, 2006–2010, and 2011–2014. Because ART agent availability changed over time, we also fit separate models for INSTI, bPI, other, and NRTI, each in comparison with NNRTI, restricting to calendar years where both ART regimen types were available, including calendar years as a continuous variable. To account for shorter follow-up of INSTI-initiating patients, we also conducted sensitivity analyses, where we restricted follow-up time for all patients to 3 years after ART initiation. For all estimates, 95% confidence intervals (95% CIs) were calculated and P values were 2 sided; P values <0.05 were considered statistically significant. Analyses were performed in SAS software, version 9.4 (SAS Institute, Cary, NC).


A total of 1624 patients who initiated ART between 1996 and 2014 were 28% women, 60% African American, 28% white, and a median of 37 years old at baseline (Table 1). Eleven percent initiated INSTI, 33% NNRTI, 20% bPI, 27% other, and 9% NRTI only regimens. The most common NRTI backbone combinations were emtricitabine/tenofovir disoproxil fumarate (FTC/TDF) among patients on INSTI (92%); FTC/TDF and zidovudine/lamivudine (3TC) among patients on NNRTI (59% and 22%, respectively); and TDF/FTC, zidovudine/3TC, and abacavir/3TC among patients on bPI (57%, 18%, and 9%, respectively). Patients initiating different ART regimens differed significantly on most baseline characteristics including sex, race, and year of starting ART. Notably, CD4 cell counts were statistically significantly different by ART regimen with a median of 403, 279, 230, 237, and 341 cells/mm3 for patients starting INSTI, NNRTI, bPI, other, and NRTI regimens, respectively.

Baseline Patient Characteristics and Time to Discontinuation and Virologic Failure by Initial ART

Median times to discontinuation and virologic failure for NNRTI patients were 3.5 and 5.1 years, respectively (Fig. 1A–B), compared with >7.4 years for INSTI patients for each of these events. Among 1111 patients who discontinued ART, 23% and 37% were because of virologic failure and ART interruption, respectively, and this varied with ART regimen. In unadjusted analyses, the estimated hazard ratio (HR) among those initiating an INSTI, as compared to an NNRTI, was 0.48 (95% CI: 0.35 to 0.67) for time to discontinuation and 0.46 (95% CI: 0.31 to 0.67) for time to virologic failure (Table 1). After adjustment, the estimated HR for patients initiating an INSTI versus an NNRTI was 0.49 (95% CI: 0.35 to 0.69) for time to discontinuation and 0.70 (95% CI: 0.46 to 1.06) for time to virologic failure. When comparing only patients initiating INSTI and NNRTI regimens between 2007 and 2014, INSTI regimens still fared better (Fig. 1C–D). In adjusted models restricted to years 2007–2014 and comparing only INSTI regimens to NNRTI, the HR was 0.52 (0.34 to 0.80) for time to discontinuation and 0.59 (0.34 to 1.03) for time to virologic failure. In this comparison, >90% of both INSTI and NNRTI regimens had TDF/FTC as the NRTI backbone. In a sensitivity analysis including patients missing CD4 and HIV RNA at baseline and using multiple imputation, in adjusted models restricted to 2007–2014 and comparing INSTI regimens (n = 180) with NNRTI (n = 295), the HR was 0.52 (95% CI: 0.34 to 0.79) for time to discontinuation and 0.57 (0.33 to 0.98) for time to virologic failure. Results were similar when follow-up was limited to 3 years after ART initiation, and when NRTI backbone was included as an adjustment variable (results not shown). Patients initiating bPI, other, or NRTI regimens fared worse in time-to-discontinuation and virologic failure analyses than patients initiating an NNRTI.

Primary endpoints. Shown are unadjusted Kaplan–Meier estimates of time to discontinuation of first ART (A), time to virologic failure of initial ART (B), and by ART regimen type. C and D show unadjusted Kaplan–Meier estimates of time to discontinuation and time to virologic failure, respectively, restricting to patients initiating INSTI and NNRTI regimens 2007 and later.


To our knowledge, this study demonstrates for the first time the dramatic increase in initial therapy virologic response and persistence with INSTI-containing regimens. Studies on initial ART duration before the availability of INSTI agents estimated median times on first therapy ranging from 1 to 2.9 years.5–7 More recent reports have observed increases in durability with median times from 2.9 to 4.6 years, with patients initiated on NNRTIs remaining on their initial ART longer than those started on bPIs and other agents.8,9 In this study, patients started on non-INSTI regimens had first ART durability comparable with these previous reports with similar associations observed between NNRTI and bPI regimens. Notably, we found that patients initiating INSTI-containing regimens remained on their initial regimen at least twice as long on average than patients initiating other types of ART. Our study is the first to examine specifically INSTI-containing regimens and to focus on changes of anchor class as regimen discontinuation. This definition likely reflects more relevant regimen modifications not related to simplification or to backbone agent changes.

The greater INSTI regimen persistence likely captures the contributions of favorable safety, efficacy, and tolerability profiles of INSTI regimens. Randomized clinical trials of INSTI agents have reported rare drug discontinuations because of adverse events, which were generally mild or moderate and lower than for NNRTI-based regimens and high virologic response rates and low failure rates that were comparable or better than NNRTI.10–12 Our findings of INSTI effectiveness from routine clinical care therefore extend these previous results and suggest that more tolerable regimens with high efficacy result in increased initial ART persistence in a real-world setting and with longer observation.

Our findings are limited by smaller sample size and person-time available for INSTI initiators and possible residual confounding because of differences in baseline clinical covariates. In addition, specific reasons for discontinuation were not examined and may have varied across initial ART regimens. This study was also conducted at a single clinical site and, therefore, the generalizability of our results should be considered carefully. The demographic make-up of our clinical site is, however, very similar to the demographics of the HIV epidemic in Southeastern United States.4 Expanding these analyses to national and international cohort collaborations and including future years of INSTI clinical care experience will be important to confirm these initial findings and inform clinical care practice.


The authors especially thank all participating patients who made this study possible.


1. Lundren JD, Babiker AG, Gordin F, et al. Initiation of antiretroviral therapy in early asymptomatic HIV infection. N Engl J Med. 2015;373:795–807.
2. Cohen MS, Chen YQ, McCauley M, et al. Antiretroviral therapy for the prevention of HIV-1 transmission. N Engl J Med. 2016;375:830–839.
3. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. Department of Health and Human Services. Available at: Accessed January 23, 2017.
4. Napravnik S, Eron JJ Jr, McKaig RG, et al. Factors associated with fewer visits for HIV primary care at a tertiary care center in the Southeastern US. AIDS Care. 2006;18:S45–S50.
5. Palella FJ Jr, Chmiel JS, Moorman AC, et al. HIV Outpatient Study Investigators. Durability and predictors of success of highly active antiretroviral therapy for ambulatory HIV-infected patients. AIDS. 2002;16:1617–1626.
6. Chen RY, Westfall AO, Mugavero MJ, et al. Duration of highly active antiretroviral therapy regimens. Clin Infect Dis. 2003;37:714–722.
7. Willig JH, Abroms S, Westfall AO, et al. Increased regimen durability in the era of once-daily fixed-dose combination antiretroviral therapy. AIDS. 2008;22:1951–1960.
8. Abgrall S, Ingle SM, May MT, et al. Durability of first ART regimen and risk factors for modification, interruption or death in HIV-positive patients starting ART in Europe and North America 2002-2009. AIDS. 2013;27:803–813.
9. Sheth AN, Ofotokun I, Buchacz K, et al. Antiretroviral regimen durability and success in treatment-naïve and treatment-experienced patients by year of treatment initiation, United States, 1996-2011. J Acqui Immune Defic Syndr. 2016;71:47–56.
10. Sax PE, Wohl D, Yin MT, et al. Tenofovir alafenamide versus tenofovir disoproxil fumarate, coformulated with elvitegravir, cobicistat, and emtricitabine, for initial treatment of HIV-1 infection: two randomised, double-blind, phase 3, non-inferiority trials. Lancet. 2015;385:2606–2615.
11. Lennox JL, DeJesus E, Berger DS, et al. Raltegravir versus efavirenz regimens in treatment-naïve HIV-1-infected patients: 96-week efficacy, durability, subgroup, safety, and metabolic analyses. J Acquir Immune Defic Syndr. 2010;55:39–48.
12. Walmsley SL, Antela A, Clumeck N, et al. Dolutegravir plus abacavir-lamivudine for the treatment of HIV-1 infection. N Engl J Med. 2013;369:1807–1818.

HIV; antiretroviral therapy; integrase inhibitors; prospective studies; United States

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.