ACTG A5353: A Pilot Study of Dolutegravir Plus Lamivudine for Initial Treatment of Human Immunodeficiency Virus-1 (HIV-1)–infected Participants With HIV-1 RNA <500000 Copies/mL

Dolutegravir plus lamivudine was virologically effective and safe in treatment-naive individuals with human immunodeficiency virus type 1 RNA up to 500 000 copies/mL. A participant who experienced failure developed the M184V lamivudine and the R263R/K integrase resistance mutations.

Abstract

Background

Limited data exist on initial human immunodeficiency virus type 1 (HIV-1) treatment with dolutegravir plus lamivudine.

Methods

A5353 is a phase 2, single-arm, pilot study of once-daily dolutegravir (50 mg) plus lamivudine (300 mg) in treatment-naive participants with HIV-1 RNA ≥1000 and <500000 copies/mL. Exclusion criteria included active hepatitis B or major protease, reverse transcriptase, or integrase resistance. The primary efficacy measure was the proportion with HIV-1 RNA <50 copies/mL (FDA [US Food and Drug Administration] Snapshot) at week 24. Virologic failure (VF) was confirmed HIV-1 RNA >400 copies/mL at week 16/20 or >200 copies/mL at or after week 24. Dolutegravir levels and drug resistance testing were performed at VF.

Results

One hundred and twenty participants (87% male, median age 30 years, 37 (31%) HIV-1 RNA >100000 copies/mL) initiated study treatment. Median entry HIV-1 RNA and CD4 count were 4.61 log₁₀ copies/mL and 387 cells/mm³. Virologic efficacy at week 24 was 108/120 (90%, confidence interval [83%, 95%]), with comparable results in the >100000 copies/mL and ≤100000 copies/mL strata, that is, 89% (75%, 97%) and 90% (82%, 96%), respectively. Three participants with VF, had undetected plasma dolutegravir at ≥1 time points; the M184V and R263R/K mutations developed in 1 participant. Two participants experienced grade 3 possible/probable treatment-related adverse events; none discontinued treatment due to adverse events.

Conclusions

Dolutegravir plus lamivudine demonstrated efficacy in individuals with pretreatment HIV-1 RNA up to 500000 copies/mL in this pilot trial, but a participant developed resistance mutations.

Clinical Trials Registration

NCT02582684.

((See Brief Report by Taiwo et al on pages 1794–97.)

Advances in the virologic and safety profiles of antiretroviral drugs have increased options for 3-drug antiretroviral therapy (ART), which is the standard of care globally [1–4]. These advances have also created opportunities to investigate novel 2-drug combinations. Potential advantages of 2-drug regimens include lower cost and improved long-term safety. Dolutegravir, an integrase strand transfer inhibitor, is attractive for 2-drug ART given its antiviral potency [5] and high resistance barrier [6–9]. Maintenance therapy with dolutegravir plus rilpivirine was comparable to standard 3-drug therapy in large, randomized trials [10].

Dolutegravir plus lamivudine is an investigational 2-drug regimen with few drug–drug interactions, potential for coformulation into a once-daily tablet, and potential for lower ART cost [11]. In the PADDLE study of treatment-naive individuals, 18/20 participants had HIV-1 RNA <50 copies/mL through week 96 [12]. Although these results were promising, the study was small and limited participation to those with screening HIV-1 RNA ≤100000 copies/mL and CD4 count >200 copies/mm³.

The AIDS Clinical Trials Group (ACTG) Study A5353 was designed to investigate dolutegravir and lamivudine in a broader population of treatment-naive HIV-1–infected participants, including those with pretreatment HIV-1 RNA 100000 copies/mL. Here, we present the week 24 primary results of the study.

METHODS

Study Design and Participants

A5353 is a phase 2, single-arm, open-label, 52-week pilot study of dolutegravir plus lamivudine conducted at 26 ACTG research sites in the United States. Antiretroviral treatment-naive HIV-1–infected adults (aged ≥18 years) with plasma HIV-1 RNA ≥1000 copies/mL and <500000 copies/mL were included. We prespecified enrollment of at least 25% of participants with screening HIV-1 RNA >100000 copies/mL. There was no CD4 count restriction. Exclusion criteria included any major protease (PR), reverse transcriptase (RT), or integrase mutation as defined by the 2014 International AIDS Society (US drug resistance mutations were excluded [13]); active hepatitis B virus infection; anticipated hepatitis C treatment during the study period; unstable or severe hepatic impairment; and pregnancy. The ethics committees at each study site approved the study, and written informed consent was obtained from each participant.

Procedures

At the screening visit, participants underwent a clinical assessment and laboratory testing to evaluate study eligibility. Pretreatment resistance was evaluated using PR/RT genotyping conducted through routine care at any time prior to study entry plus study-provided integrase genotyping performed at Quest Diagnostics, Baltimore, Maryland.

Within 45 days of the screening visit, participants initiated the study regimen of open-label oral dolutegravir 50 mg and lamivudine 300 mg once daily taken as 2 separate tablets. Post-entry evaluations at weeks 2, 4, 8, 12, 16, 20, 24, 32, 40, and 48 included a clinical assessment, adherence assessment using a self-completed questionnaire [14], and plasma HIV-1 RNA measurement using the Abbott RealTime HIV-1 assay (lower detection limit of 40 copies/mL) at Quest Diagnostics, Baltimore. Complete blood count, liver function, and chemistries were monitored at weeks 4, 12, 24, 32, 40, and 48. Genomic DNA from whole blood was Taqman genotyped for the UGT1A1 rs887829 decreased conjugation variant that, when homozygous T/T, confers approximately 30% greater peak plasma dolutegravir concentrations [15]. Signs, symptoms, and laboratory events were graded by site investigators according to the Division of AIDS, National Institutes of Allergy and Infectious Diseases Table for Grading the Severity of Adult and Pediatric Adverse Events, version 2.0 [16]. All symptoms ≥ grade 3, rash ≥ grade 2, or any grade that led to treatment discontinuation were reported. For laboratory tests, all creatinine and fasting lipid values and other laboratory results ≥ grade 3 or that led to a change in study treatment were reported. CD4 count was determined at weeks 4, 12, 24, and 48. The final study visit at the completion of 52 weeks of study treatment was primarily to confirm virologic failure in participants with suspected virologic failure at week 48.

Defining virologic failure as confirmed HIV-1 RNA >400 copies/mL at weeks 16 or 20 or confirmed HIV-1 RNA >200 copies/mL at or after week 24, participants with suspected virologic failure and those with HIV-1 RNA 50–200 copies/mL at week 24 underwent confirmatory HIV-1 RNA measurement within 7–28 days of the initial sample. Participants with confirmed virologic failure had real-time PR/RT/integrase genotyping performed on plasma collected at the confirmation time point. Plasma was stored at each study visit to evaluate the impact of minority variants and dolutegravir plasma levels on observed virologic responses. Dolutegravir and lamivudine were dispensed at study entry and all subsequent visits except weeks 2 and 52. A pregnancy test was performed any time pregnancy was suspected.

Objectives

The primary objective was to estimate virologic success, defined as HIV-1 RNA <50 copies/mL on dolutegravir plus lamivudine (FDA [US Food and Drug Administration] Snapshot) at week 24. Participants with HIV-1 RNA ≥50 copies/mL at week 24 or who discontinued or changed study treatment earlier and either had HIV-1 RNA ≥50 copies/mL or did not have HIV-1 RNA data were counted as virologic nonsuccesses. The secondary objectives included evaluating safety and tolerability of the study regimen, efficacy in those with pretreatment HIV-1 RNA levels ≤100000 vs >100000 copies/mL, emergent integrase or RT resistance, changes in CD4 counts, and associations between plasma dolutegravir exposure and observed outcomes.

Statistical Analyses

The primary efficacy measure was virologic success at week 24 by FDA Snapshot definition. The analysis window was week 22–28. The proportion of participants achieved virologic success was estimated using a 2-sided exact Clopper-Pearson 95% confidence interval (CI) based on binomial distribution. With a sample size of 120 participants, if the observed virologic success rate is between 85% and 90%, the 95% CI would have a width of ±5.75% to ±6.75%. For subgroup analysis of participants with higher viral load, assuming N = 30 and the observed virologic success rate between 85% and 90%, the 95% CI would have a width of (−12.2%, +10%) to ±14.1%. The proportions of participants with HIV-1 RNA <50 copies/mL at scheduled study weeks over time were estimated using modified intent-to-treat (ITT; missing/off-treatment was considered a failure) approach. Self-reported adherence was classified as perfect (zero reported missed doses) or imperfect (any missed doses of dolutegravir or lamivudine, failure to answer the 4-day recall question, or missed study visit). Comparisons between baseline HIV-1 RNA strata were carried out using Fisher exact test. All P values and CIs presented were 2-sided and nominal, unadjusted for interim analysis and multiple comparisons. Analyses were done using SAS, version 9.4 (SAS Institute, Cary, North Carolina).

RESULTS

Between 8 November 2015 and 13 September 2016, 165 participants were screened and 122 were enrolled from 26 sites in the United States (Figure 1). Two participants who enrolled with the exclusionary K103N mutation were terminated from the study and excluded from the analysis. The 120 eligible participants included in the primary analysis had a median age of 30 years and were racially and ethnically diverse, with a median study entry HIV-1 RNA level of 4.61 log₁₀ copies/mL and CD4 count of 387 cells/mm³ (Table 1). Approximately one-third of the participants had study entry HIV-1 RNA >100000 copies/mL. For 117 participants successfully genotyped for UGT1A1 rs887829, 53 (45%) were C/C, 46 (39%) were C/T, and 18 (15%) were T/T. Four participants discontinued study follow-up prior to week 24 for the following reasons: moved (n = 2), incarceration (n = 1), and lost to follow-up (n = 1). In addition, 3 participants discontinued dolutegravir plus lamivudine prior to week 24 for the following reasons: pregnancy (n = 1), noncompliance (n = 1), and unable to attend clinic (n = 1) but continued study follow-up.

Figure 1.

Consort diagram. *Number of screening instances and screening failure instances because of rescreening these numbers may not reflect people screened.

Table 1.

Study Entry Demographics

Characteristic		Baseline HIV-1 RNA Category		Total (N = 120)
		>100000 Copies/mL (N = 37)	≤100000 Copies/mL (N = 83)
Age (y)	Median (Q1, Q3)	30 (25, 40)	30 (24, 42)	30 (24, 41)
	Min, Max	18, 67	18, 60	18, 67
Sex (%)	Male	33 (89)	71 (86)	104 (87)
	Female	4 (11)	12 (14)	16 (13)
Race/Ethnicity (%)	White non-Hispanic	10 (27)	24 (29)	34 (28)
	Black non-Hispanic	12 (32)	36 (43)	48 (40)
	Hispanic (regardless of race)	13 (35)	19 (23)	32 (27)
	Asian, Pacific Islander	1 (3)	1 (1)	2 (2)
CD4 (cells/mm3)	Median (Q1, Q3)	350 (173, 458)	413 (328, 671)	387 (288, 596)
	<200	11 (30%)	5 (6%)	16 (14%)
HIV RNA (copies/mL)	Median, log10 (Q1, Q3)	5.23 (5.09, 5.46)	4.23 (3.82, 4.65)	4.61 (3.94, 5.05)
	<10000	0%	35 (42%)	35 (29%)
	10000–100000	0%	48 (58%)	48 (40%)
	>100000–200000	23 (62%)	0%	23 (19%)
	>200000	14 (38%)	0%	14 (12%)

Abbreviation: HIV, human immunodeficiency virus.

Efficacy

At week 24, 108 (90%) of 120 participants achieved virologic success per FDA Snapshot definition (plasma HIV-1 RNA <50 copies/mL on study treatment) with 95% CI of 83%, 95% (Table 2). Five (4%) participants had virologic nonsuccess. Seven (6%) participants had no virologic data in the analysis window, 6 of whom discontinued the study prior to week 24 with an HIV-1 RNA <50 copies/mL, and 1 had HIV-1 RNA <50 copies/mL at weeks 22 and 32 but a missed visit at week 24. The proportion of participants who achieved virologic success was similar between the study entry HIV-1 RNA strata: 33/37 (89% [75%, 97%]) in HIV-1 RNA >100000 copies/mL and 75/83 (90% [82%, 96%]) in HIV-1 RNA ≤100000 copies/mL (P = 1.00; Table 2). The proportion of participants with HIV-1 RNA <50 copies/mL over time was also estimated using an ITT missing/off treatment = failure approach (Figure 2), where HIV-1 RNA was <50 copies/mL in 108/120 (90% [83%, 95%]) of the participants at week 24, 33/37 (89% [75%, 97%]) in participants with HIV-1 RNA >100000 copies/mL, and 75/83 (90% [82%, 96%] in participants with HIV-1 RNA ≤100000 copies/mL.

Table 2.

Virologic Outcome by US Food and Drug Administration Snapshot at Week 24

	Baseline HIV-1 RNA				Total (N = 120)
	>100000 Copies/mL (N = 37)		≤100000 Copies/mL (N = 83)
Virologic success, HIV-1 RNA <50 copies/mL, N (%)	33	(89)	75	(90)	108	(90)
[95% confidence interval]		(75, 97)		(82,96)		(83,95)
Virologic nonsuccess	3	(8%)	2	(2%)	5	(4%)
HIV-1 RNA ≥50 copies/mL	3		0		3
Discontinued study treatment for other reasonsa while HIV-1 RNA ≥50 copies/mL	0		2		2
No virologic data in window	1	(3%)	6	(7%)	7	(6%)
Discontinued study treatment for other reasonsb	1		5		6
On study but missing data in window	0		1		1

Abbreviation: HIV, human immunodeficiency virus.

^aPoor adherence.

^bLost to follow-up, pregnancy

Figure 2.

Proportion (95% confidence interval) of participants with human immunodeficiency virus type 1 RNA levels <50 copies/mL by week (intention-to-treat/ missing/off treatment = ignored).

The 5 participants with virologic nonsuccess at week 24 are described in Figure 3. Two of these participants (participants 1 and 2) had study entry HIV-1 RNA >100 000 copies/mL and HIV-1 RNA levels between 50 and 200 copies/mL at week 24, indicating nonsuccess by the FDA Snapshot algorithm while not meeting protocol-defined virologic failure. Three participants experienced protocol-defined virologic failure prior to or at week 24: 1 with study entry HIV-1 RNA >100000 copies/mL (participant 3) and 2 with study entry HIV-1 RNA ≤100000 copies/mL (participants 4 and 5). Participant 3 had virologic failure on study treatment at week 24 with no evidence of emergent resistance mutations. Participant 4 experienced virologic failure on study treatment at week 24 with detection of the polymorphic V106I mutation, which was not detected pre-treatment. Participant 5 experienced viral rebound and prolonged viremia on study treatment, discontinuing study treatment due to noncompliance at week 18. This participant had confirmed virologic failure at week 24 (off ART) with the M184V mutation in RT detected (no mutation found pre-treatment). Population-based genotyping of samples obtained prior to protocol-defined virologic failure showed no mutations at week 8 (HIV-1 RNA 6579 copies/mL) and M184V plus the mixture R263R/K in integrase at week 14 (HIV-1 RNA 446 copies/mL). Dolutegravir concentrations assessed on stored samples from these 3 participants with virologic failure demonstrated dolutegravir levels below the limit of detection (<5 ng/mL) at 1 or more time points while on study treatment. For the 5 participants with virologic nonsuccess, UGT1A1 rs887829 genotypes were C/C (n = 4) and C/T (n = 1). All 3 with virologic failure were C/C.

Figure 3.

HIV-1 RNA, DTG plasma concentration, and genotyping results for virologic failures and Snapshot nonsuccesses. Open circles represent HIV-1 RNA results that are less than the lower limit of quantification. Open diamonds represent drug level below the in vitro inhibitory concentration for 90% inhibition (IC90) (64 ng/mL) [17]. Participants 3, 4, and 5 had at least 1 time point with undetectable plasma DTG. Resistance mutations are shown at the time points tested, with “None” representing no mutation detected. Vertical dashed line represents discontinuation of study treatment. Horizontal dashed line represents HIV-1 RNA 50 copies/mL. Abbreviations: DTG, dolutegravir; HIV-1, human immunodeficiency virus type 1.

The median (interquartile range) CD4 cell count change from study entry was +167 (86, 275) cells/mm³ at week 24: +212 (125, 299) cells/mm³ in the participants with study entry HIV-1 RNA >100000 copies/mL and +133 (60, 262) cells/mm³ in participants with study entry HIV-1 RNA ≤100000 copies/mL. Self-reported adherence was generally high throughout the study, with 90% of participants reporting perfect adherence based on 4-day recall at week 24.

Safety and Tolerability

Ten (8.3%) participants reported grade 3 or 4 adverse events, only 2 (grade 3 creatinine clearance reduction and grade 3 palpitation) of which were considered to be possibly or probably related to study treatment by the site investigator. There were also 2 cases of grade 2 allergic rash. No participant discontinued study treatment due to adverse events.

DISCUSSION

In this single-arm pilot study, 90% of HIV-1–infected antiretroviral-naive individuals with study entry HIV-1 RNA levels up to 500000 copies/mL who initiated the 2-drug regimen of dolutegravir and lamivudine were on study treatment and had HIV-1 RNA levels <50 copies/mL at week 24, using the FDA Snapshot algorithm. An intent-to-treat, missing = failure analysis showed similar results. Six percent of the participants were virologically suppressed at their last study visit but had no data in the week 24 window, primarily due to early discontinuation of study treatment for reasons unrelated to tolerability. Overall, this 2-drug regimen potently suppressed HIV-1 RNA regardless of study entry HIV-1 RNA stratum (≤100000 copies/mL vs >100000 copies/mL) and was safe and well tolerated.

A 3-drug combination is the current standard for initial HIV therapy globally. Although available 3-drug regimens are potent, convenient, and generally safe and well tolerated, issues of toxicity and cost remain over decades of exposure. A virologically potent 2-drug regimen, with the potential to reduce toxicity and cost, could address these issues and challenge standard-of-care therapy. Earlier large randomized studies of initial 2-drug antiretroviral regimens were disappointing, with suboptimal virologic responses or excess toxicity in all or some groups of participants [18, 19] or unacceptable rates of drug resistance during virologic failure [20]. The recent exception was the GARDEL study, which demonstrated noninferiority of lopinavir/ritonavir (LPV/r) plus lamivudine to LPV/r plus 2 nucleos(t)ide reverse transcriptase inhibitors [21]. However, LPV/r is associated with gastrointestinal and metabolic side effects and a high risk of drug–drug interactions, making it an undesirable initial ART. Our results suggest that dolutegravir plus lamivudine may be a potent, well-tolerated, and safe 2-drug regimen in individuals with HIV-1 RNA levels up to 500000 copies/mL. These findings extend the results of the PADDLE study, which restricted enrollment to individuals with screening HIV-1 RNA levels ≤100000 copies/mL and CD4 counts >200 cells/mm³.

Virologic failure was rare and associated with suboptimal adherence. In fact, all 3 participants who met protocol criteria for virologic failure had viral suppression to <50 copies/mL by week 4 (2 participants) or week 12 (1 participant) before viral rebound that appeared temporally linked to adherence lapse. All had no measurable dolutegravir (<5 ng/mL) at 1 or more time points while on study treatment. In 1 participant, population sequencing revealed emergence of the M184V mutation, which confers resistance to lamivudine and emtricitabine, plus 263K, which confers integrase resistance. Another participant developed the polymorphic 106I mutation, which is associated with nonnucleoside reverse-transcriptase inhibitor (NNRTI) resistance [13, 22]. Since this mutation was not detected pre-treatment and the participant had no known exposure to an NNRTI, its development is being investigated with more sensitive sequencing assays. Of note, the only participant with virologic failure in the PADDLE study had no RT mutations detected, integrase did not amplify, and he resuppressed on the same regimen with improved adherence. Because a frequent UGT1A1 variant is associated with altered dolutegravir pharmacokinetics, for thoroughness we characterized UGT1A1 genotypes in the present study. This genotype almost certainly did not contribute to virologic failure, given its modest effect on pharmacokinetics.

The 263K mutation was the most common mutation that developed during in vitro passage experiments with dolutegravir [23]. This mutation is associated with approximately 2-fold reduction in susceptibility to dolutegravir and a 6-fold reduction in susceptibility to elvitegravir, imparting no significant effect on raltegravir [23–25]. Further, 263K is known to substantially hamper viral replicative capacity and enzymatic activity [23, 26]. Our study is the first to report 263K mutation in an antiretroviral treatment-naive individual receiving dolutegravir. The mutation in our participant was detected as a mixture (impact on dolutegravir sensitivity unknown) and was not seen at a later time point after the participant had discontinued dolutegravir, possibly related to the known impact of 263K on viral fitness. A European cross-sectional study of 278 treatment-naive individuals found 1 individual with 263K [27], indicating this mutation can be transmitted. However, it was not detected by population sequencing in our participant’s pre-treatment plasma sample. Phenotypic and minority variant testing of pre-treatment and virologic failure samples are planned. Previous reports of the 263K mutation were in treatment-experienced individuals with virologic failure: 2 in the dolutegravir arm of the SAILING study [28] and 2 in a retrospective cohort study from British Columbia [29]. Since emergence of any drug resistance is rare during virologic failure of initial dolutegravir-based 3-drug ART [30], the emergence of resistance mutations in both RT (M184V) and integrase (R263R/K) in 1 participant in this pilot study underscores the need to investigate the resistance barrier of dolutegravir plus lamivudine further. Two ongoing, fully powered randomized trials, GEMINI 1 (NCT02831673) and GEMINI 2 (NCT02831764), will provide data to better address this.

In this study, the 2-drug regimen of dolutegravir and lamivudine was generally safe and well tolerated, consistent with the known favorable toxicity profiles of these 2 agents. No participant discontinued the regimen because of adverse events. Notably, the combination of dolutegravir and lamivudine has the potential to be coformulated into a single daily pill without dietary restrictions. Antiretroviral regimens must be well tolerated and convenient to support long-term adherence.

The cost of recommended 3-drug antiretroviral regimens exceeds $30000 a year in developed countries; a 2-drug regimen of dolutegravir and lamivudine could cost less. Girouard and colleagues estimated that if the 2-drug regimen of dolutegravir and lamivudine demonstrated high rates of virologic suppression, the regimen would be cost-effective and could save more than $500 million over 5 years [11] using current US drug pricing and projecting a lower cost due to use of fewer agents and 1 generic drug. Projected cost savings exceeded $3 billion over 5 years when the model included switching a quarter of currently suppressed individuals to dolutegravir plus lamivudine.

This study has strengths and limitations. The study was conducted at 26 sites across the United States and enrolled a diverse patient population (40% black and 27% Latino) with an average age of 30 years. Study follow-up was excellent, with only 6% discontinuing the study regimen prematurely. Drug concentration and virologic assays helped elucidate the causes and consequences of virologic failure. On the other hand, the study population consisted of only 13% women and excluded those with pre-treatment HIV-1 RNA levels >500000 copies/mL. Further, there was no comparator arm to conclusively assess efficacy or safety/tolerability vs the standard of care. The GEMINI 1 and 2 studies of dolutegravir plus lamivudine have enrolled approximately 1400 treatment-naive participants, and results are anticipated. The regimen is contraindicated with chronic hepatitis B infection.

In summary, the week 24 results of the ACTG A5353 study showed that the 2-drug antiretroviral regimen of dolutegravir and lamivudine was virologically potent and generally safe and well tolerated in individuals with pre-treatment HIV-1 RNA levels up to 500000 copies/mL. Virologic failure was uncommon and associated with suboptimal dolutegravir concentrations, likely indicative of suboptimal adherence. One participant selected for a mutation associated with resistance to lamivudine and a partial substitution at codon 263 of the integrase enzyme that may confer reduced sensitivity to dolutegravir. The ongoing fully powered phase 3 studies will provide needed data on dolutegravir plus lamivudine compared to dolutegravir-based 3-drug therapy in treatment-naive individuals.

Notes

Author contributions. B. O. T. contributed to the study design, data collection and interpretation, and manuscript drafting and revision. L. Z. contributed to the study design, data collection and interpretation, and manuscript drafting and revision. A. S. contributed to the data collection and interpretation and manuscript revision. A. N. N. contributed to participant recruitment, data collection and interpretation, and manuscript revision. B. B. contributed to the study design, data collection, and manuscript drafting and revision. C. L. W. contributed to the study design, virological resistance studies, data interpretation, and manuscript revision. C. G. contributed to the study design, data interpretation, and manuscript revision. P. S. contributed to the study design, data interpretation, and manuscript revision. E. A. performed the plasma dolutegravir assays and contributed to manuscript revision. D. H. contributed to the study design, data interpretation, and manuscript revision. K. Y. S. contributed to the study design, data interpretation, and liaison with industry support. B. S. contributed to participant recruitment and manuscript revision. C. V. D. contributed to participant recruitment and manuscript revision. R. M. G. contributed to the study design, data collection and interpretation, and manuscript drafting and revision.

Acknowledgments. We acknowledge the support of the Pharmacology Specialty Laboratory of the ACTG, University of Alabama (grant UM1 AI106701-05) and the Statistical Data Management Center (grant 2UM1 AI068634-08). We thank the study participants and the other members of the A5353 team, including Elizabeth Hawkins, Johnstone Kumwenda, Angel Hernandez, Belinda Ha, Bernadette Jarocki, Gerald Tegha, and Tanisha Sullivan. We acknowledge the contributions of the staff at the sites and grants supporting their work, including Mark Mall, RN, and Antoinette Lewis, MPA at Rush University Medical Center Clinical Research Site (CRS) (site 2702; grant U01 AI06947; Cornelius N. Van Dam and Kimberly Epperson, RN, at Greensboro CRS (site 3203; grant 5UM1A069423-11); Johnny Perez and Nina Lambert at Northwestern University CRS (site 2701; grant 2UM1 AI069471); Roberto C. Arduino and Aristoteles Villamil at Houston AIDS Research Team (site 31473; grants 5UM1 AI069503-11, 2UM1 AI068636-11); Cathi Basler and Christine Griesmer at University of Colorado Hospital CRS (site 6101; grants 2UM1AI069432, UL1 TR001082); Margaret A. Fischl, MD, and Hector Bolivar, MD, at University of Miami AIDS Clinical Research Unit (site 901; grants AI069477, 5UM1AI069477); Roger Bedimo, MD, and Lauren Rogers at Trinity Health and Wellness Center (site 31443; grant U01 AI069471); Becky Straub, BSN, MPH, and Chris Evans, MSN, at Chapel Hill CRS (site 3201; grants UM1 AI069423, Clinical and Translational Science Award [CTSA] 1UL1TR001111, Center for AIDS Research [CFAR] P30 AI50410; Beverly Woodward, MSNRN, and Michael Leonard at Vanderbilt Therapeutics (site 3652; grants AI069439, TR002243;) Susan Koletar, MD, and Leah Kofmehl at Ohio State University (site 2301; grant UM1AI06949); Teresa Spitz and Lisa Kessels at Washington University Therapeutics CRS (site 2101; grant 5UM1AI069439-10); Helen Patterson, LPN, and Karen Tashima, MD, at the Miriam Hospital (site 2951; grant AI069412); Todd Stroberg and Tiina Ilmet at Weill Cornell Chelsea CRS (site 7804; grants UM1AI069419, UL1TR000457 [Clinical and Translational Science Center]); Phillip Dube, MD, and Frances Canchola, RN, at University of Southern California CRS (site 1201; grant 2UM1AI069432); Jorge L Santana, MD, FIDSA, and Sigrid Perez, MD, at Puerto Rico AIDS Clinical Trails Unit CRS (site 5401; grant 5 UM1 AI069415-12); Paul Sax, MD, and Cheryl Keenan, RN, BC, at Brigham and Women’s Hospital (site 107; grant 5UM1AI068636); Eva Whitehead, RN, and Carl J. Fichtenbaum, MD, at Cincinnati CRS (site 2401; grant AI69439); Michael T. Yin, MD, MS, and Jolene Noel-Connor, RN, at Columbia Physicians and Surgeons CRS (site 30329; ACTG Clinical Trials Unit [CTU], grants 2UM1-AI069470, CTSA UL1RR024156); Mary Adams, RN, MPH, and Elizabeth Keller, RN at University of Rochester (site 31787; grant UM1 AI069511); Carlos Del Rio, MD, and Ericka R Patrick, RN, MSN, at Emory-Centers for Disease Control and Prevention Ponce de Leon (site 5802; grant 1U01AI069418-01 and Emory University CFAR P30AI050409); Eric Daar and Ruben Lopez at Harbor–University of California–Los Angeles (UCLA) Medical Center (site 603; grants A1069424, UL1TR000124); Pablo Tebas, MD, and Yan Jiang, MSN, at Philadelphia HIV Therapeutics and Prevention CTU (site 6201; grants UM1AI068636, UM1AI069534); Dee Dee Pacheco and Constance Benson, MD, at University of California–San Diego (site 701; grant AI069432); Valery Hughes and Brian Mangano at Weill Cornell Uptown CRS (site 7803; grants U01 AI69419, UL1 TR000457); Teri Flynn, ANP, and Amy Sbrolla, RN, at Massachusetts General Hospital (site 101; grant AI069412); Raphael Landovitz, MD, and Sana Majid at UCLA Care Center CRS (site 601; grants AI069424, UCLA CTSI UL-1TR000124, CFAR P30-AI028697).

Disclaimer. This paper was written by C. G. in her capacity as a National Institutes of Health (NIH) employee. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. ViiV Healthcare provided the study drugs; this publication does not represent the official views of ViiV Healthcare.

Financial support. This work was supported by the Institute of Allergy and Infectious Diseases of the National Institutes of Health (grants UM1 AI068634, UM1 AI068636, and UM1 AI106701).

Potential conflicts of interest. B. O. T. served as a paid consultant to GlaxoSmithKline/ViiV, Gilead, and Janssen and received research funding through Northwestern University from GlaxoSmithKline/ViiV. E. A. reports nonfinancial support from ViiV. P. E. S. has served as a paid consultant to Gilead, Abbvie, Merck, Janssen, Glaxo-SmithKline/ViiV, and Bristol-Myers Squibb. K. Y. S. is employed by ViiV Healthcare. C. L. W. received honoraria from Abbott Molecular and Mylan. C. V. D. served as a paid consultant/speaker to ViiV Healthcare, Gilead, and Janssen. All remaining authors: No reported conflicts. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.