7 A n n a ls

O rig in a l R esearch

o f In te r n a l M e d ic in e

Efficacy and Tolerability of 3 Nonnucleoside Reverse Transcriptase In h ib ito r-S p arin g A ntiretroviral Regim ens fo r Treatm ent-N aive Volunteers Infected W ith HIV-1 A Randomized, Controlled Equivalence Trial J e ffrey L. Lennox, M D ; R aphael J. L an d o vitz, M D , M Sc; H e a th e r J. R ibaudo, PhD; Ig h o v w e rh a O fo to k u n , M D , M S c; L um in e H. N a, M S ; C a th e rin e G o d fre y , M D ; D a n ie l R. K uritzkes, M D ; M a n is h Sagar, M D ; T o d d T. B row n , M D , PhD; Susan E. C ohn, M D , M P H ; G race A . M c C o m s e y , M D ; Francesca A w e e k a , Ph arm D ; Carl J. F ich te n b a u m , M D ; Rachel M . Presti, M D , P hD ; Susan L. K oletar, M D ; D a v id W . H aas, M D ; K ristine B. P atterson, M D ; C o nstance A . Benson, M D ; Bryan P. B augh, M D ; Randi Y . L ea v itt, M D , PhD; Jam es F. R o oney, M D ; D a n ie l Seekins, M D ; and Ju d ith S. C urrier, M D , M S c, fo r th e A C T G A 5 2 5 7 T e a m *

Background: Nonnucleoside reverse transcriptase inhibitor-based

Results: A m on g 1809 participants, all pairwise comparisons o f in­

antiretroviral therapy is n o t suitable fo r all treatm ent-naive HIVinfected persons.

cidence o f virologic failure over 96 weeks showed equivalence

Objective: To evaluate 3

reverse transcriptase

gravir and ritonavir-boosted darunavir were equivalent fo r tolera­ bility, whereas ritonavir-boosted atazanavir resulted in a 12.7% and

inhibitor-sparing initial antiretroviral regimens to show equivalence fo r virologic efficacy and tolerability.

9.2 % higher incidence o f tolerability discontinuation than raltegravir

nonnucleoside

w ith in a margin o f equivalence defined as - 1 0 % to 10% . Ralte­

and ritonavir-boosted darunavir, respectively, primarily because o f hyperbilirubinemia.

Design: A phase 3, open-label study randomized in a 1:1:1 ratio w ith fo llo w -u p NC T00811954)

fo r

at

least

96

weeks.

(ClinicalTrials.gov:

For com bined

virologic

efficacy

and

toler­

ability, ritonavir-boosted darunavir was superior to ritonavir-boosted atazanavir, and raltegravir was superior to both protease inhibitors. Antiretroviral resistance at the tim e o f virologic failure was rare but more frequent w ith raltegravir.

Setting: 57 sites in the United States and Puerto Rico. Patients: Treatm ent-naive persons aged 18 years o r older w ith

Limitation: The trial was open-label, and

ritonavir was

not

provided.

HIV-1 RNA levels greater than 1000 copies/m L w ith o u t resistance to nucleoside reverse transcriptase inhibitors or protease inhibitors.

Conclusion: O ver 2 years, all 3 regimens attained high and equiv­

Intervention: Atazanavir, 300 m g/d, w ith ritonavir, 100 m g/d;

alent rates o f virologic control. Tolerability o f regimens containing

raltegravir, 40 0 m g tw ice daily; or darunavir, 800 m g /d , w ith

raltegravir or ritonavir-boosted darunavir was superior to th a t o f the ritonavir-boosted atazanavir regimen.

ritonavir, 100 m g /d , plus com bination emtricitabine, 20 0 m g/d, and te n o fo vir disoproxil fum arate, 300 m g/d.

Measurements: Virologic failure, defined as a confirm ed HIV-1

Primary Funding Source: National Institute o f Allergy and Infec­ tious Diseases.

RNA level greater than 1000 copies/m L a t or after 16 weeks and before 24 weeks or greater than 20 0 copies/m L at o r after 24

Ann Intern Med. 2 0 1 4 ;1 6 1 :4 6 1 -4 7 1 . d o i:1 0 .7 3 2 6 /M 1 4 -1 0 8 4

www.annals.org

weeks, and tolerability failure, defined as discontinuation o f atazanavir, raltegravir, or darunavir fo r toxicity. A secondary end

For a u th o r affiliations, see end o f text.

p o in t was a com bination o f virologic efficacy and tolerability.

w w w .a n n a ls .o rg ).

he 2014 U.S. Department of Health and Human Ser­ vices antiretroviral therapy guidelines recommend a combination of 2 reverse transcriptase inhibitors plus a nonnucleoside reverse transcriptase inhibitor (NNRTI), a ritonavir-boosted protease inhibitor (PI), or an integrase inhibitor for the initial treatment of H IV -1-infected adults and adolescents (1). The recommended NNRTI is efavirenz that, when coformulated with emtricitabine and tenofovir disoproxil fumarate (TDF), allows 1-pill, oncedaily dosing. Efavirenz-based combinations are recom­ mended as first-line therapy by the World Health Organi­ zation (2). However, women who are contemplating becoming pregnant, patients with preexisting NNRTI re­ sistance, and those with severe psychiatric disorders are not considered good candidates for efavirenz-based therapy when other options are available.

Ritonavir-boosted Pi-containing therapy may be lim­ ited by hepatic, gastrointestinal, and metabolic adverse ef­ fects and may increase cardiovascular and cerebrovascular morbidity (3-5). Integrase inhibitors are virologically po­ tent first-line agents with a favorable toxicity profile but have more limited long-term safety data and are less widely available in resource-constrained settings. To better under­ stand the long-term efficacy and tolerability of alternatives to efavirenz, we did a randomized study of combination TDF plus emtricitabine with ritonavir-boosted atazanavir, raltegravir, or ritonavir-boosted darunavir.

T

* For m em bers o f th e AC TG A 5 2 5 7 Team , see Appendix 1 (available at

S e e a ls o :

S u m m a ry f o r P a tie n ts ..........................................................|-2 2

© 2014 American College of Physicians 461

O

r ig in a l

R

esear c h

Context

NNRTI-Sparing Antiretroviral Regimens for Treatment-Naive Volunteers Infected W ith HIV-1

___

Antiretroviral regimens that include efavirenz are widely used for initial therapy fo r HIV infection. However, efa­ virenz is not recommended fo r women who are consider­ ing becoming pregnant or patients w ith a history o f severe psychiatric disease.

Contribution

____

___

In an open-label, randomized, controlled trial, 3 antiretro­ viral regimens that did not contain efavirenz were com­ pared fo r long-term efficacy and safety fo r initial treat­ ment o f HIV infection. All 3 regimens resulted in excellent HIV suppression w ith some differences in tolerability and adverse effect profiles.

Caution

___

_

To reduce pill burden, the trial was not blinded.

Implication

__

_

__

Several reasonable options fo r initial HIV antiretroviral therapy that do not contain efavirenz are available.

— The Editors M

etho ds

Study Patients

The ACTG (AIDS Clinical Trials Group) A5257 study included adults infected with HIV-1 receiving care in the United States and Puerto Rico with plasma HIV-1 RNA levels greater than 1000 copies/mL who had received 10 or fewer days of antiretroviral therapy. Participants had documented absence of genotypic resistance to nucleoside reverse transcriptase inhibitors and Pis; integrase genotyping was not required because transmitted integrase resis­ tance is rare (6, 7). The C D 4+ cell count at entry was not limited. This study was approved by the institutional review boards and/or ethics committees at each site, and all participants gave written informed consent before enrollment. Study Design

O ur study was a phase 3, randomized, open-label trial. Participants were followed, regardless of meeting an end point, for 96 weeks after enrollment of the final volunteer. They were randomly assigned in a 1:1:1 ratio to receive 1 of 3 regimens: 300 mg of atazanavir (Reyataz, BristolMyers Squibb) with 100 mg of ritonavir (Norvir, Abbott Laboratories), both once daily (ritonavir-boosted atazana­ vir); 800 mg of darunavir (Prezista, Janssen Therapeutics) with 100 mg of ritonavir, both once daily (ritonavirboosted darunavir); or 400 mg of raltegravir (Isentress, Merck) twice daily, each with a fixed-dose combination of 300 mg of TDF plus 200 mg of emtricitabine (Truvada, Gilead Sciences) daily. Randomization used permuted blocks stratified ac­ cording to the HIV-1 RNA level (>100 000 vs. < 100 000 copies/mL) with balancing by institution. To ensure treat4 6 2 7 October 2014 Annals oflnternal Medicine Volume 161 • Number 7

ment balance by cardiovascular risk for an embedded car­ diovascular substudy (8), randomization was stratified by intent to participate in the substudy and 10-year Framing­ ham risk for myocardial infarction or coronary death (6% ). Screening HIV-1 RNA levels were mea­ sured at Clinical Laboratory Improvement Amendmentsadherent laboratories; subsequent levels were measured using the RealTime HIV-1 Qualitative assay (Abbott Lab­ oratories) at Johns Hopkins University, Baltimore, Mary­ land. Study evaluations were completed before entry; at entry; at weeks 4, 8, 16, 24, and 32; and every 16 weeks thereafter. At the time of protocol-defined virologic failure, the HIV-1 reverse transcriptase and protease regions were genotyped at Brigham and W omen’s Hospital, Boston, Massachusetts, and the University of Alabama, Birming­ ham, Birmingham, Alabama; samples obtained at study entry were also assayed concurrently. The HIV-1 integrase region was genotyped at the end of the study at Brigham and W omen’s Hospital for participants with virologic fail­ ure receiving raltegravir and for a few randomly selected participants from each Pi-containing group. In the event of treatment changes for virologic or tolerability failure, within-class substitutions for participants receiving PIcontaining regimens were recommended but not man­ dated. Adverse events were graded using the 2004 Division of AIDS Table for Grading the Severity of Adult and Pe­ diatric Adverse Events (9). Statistical Analysis

The primary objective was to evaluate regimen equiv­ alence regarding virologic efficacy and tolerability over 96 weeks. Virologic failure was defined as a confirmed HIV-1 RNA level greater than 1000 copies/mL at or after 16 weeks and before 24 weeks from randomization or less than 200 copies/mL at or after 24 weeks (10). The primary tolerability end point was the time from randomization to discontinuation of the randomized regimen component for toxicity (per-site attribution); treatment discontinuations for other reasons were considered competing events. Sub­ stitution of any component of the fixed-dose combination of TDF plus emtricitabine was not considered tolerability failure. A preplanned composite end point was defined as the earlier occurrence of virologic or tolerability failure. We also analyzed the composite U.S. Food and Drug Admin­ istration (FDA) end point of time to loss of virologic re­ sponse (TLOVR), with an HIV-1 RNA failure threshold of 200 copies/mL (11). Furthermore, we did the FDA snapshot analysis of the proportion of participants receiv­ ing randomized treatment with an HIV-1 RNA level less than 30 copies/mL at 96 weeks (12). The target sample size of 600 participants per group would provide 90% power to show equivalence in pairwise regimen comparisons, assum­ ing rates of virologic failure, tolerability failure, and loss to follow-up of 25%, 10%, and 12%, respectively. w w w .an n a ls.org

NNRTI-Sparing Antiretroviral Regimens for Treatment-Naive Volunteers Infected With HIV-1

Equivalence was predefined as a 2-sided 97.5% Cl on the pairwise difference in 96-week cumulative incidence of each individual or composite end point falling between —10% and 10%. For comparisons in which equivalence was not shown, superiority was defined as exclusion of 0 from the 97.5% Cl. The equivalence bound was deter­ mined by the protocol team consistent with FDA guide­ lines and contemporary trials. Inference was based on the 97.5% Cl to control the study-wide type I error rate at 5% with evaluation of 3 pairwise comparisons (13). Unless otherwise noted, analyses followed the intention-to-treat (ITT) principle. All randomly assigned participants who were not subsequently found to be ineli­ gible were included according to their randomized treat­ ment assignment; sensitivity as-treated analyses of virologic failure included treatment discontinuation as a competing event. Event cumulative incidence was estimated over time, accounting for competing events as appropriate (for tolerability failure and as-treated virologic failure out­ comes). Inference regarding equivalence was based on the pairwise differences in the 96-week cumulative incidence of the event of interest with 97.5% CIs stratified by HIV-1 RNA level at screening. In each case, the inverse of the stratum-specific variance was used for the stratum weights (14). Modification of treatment group differences across prespecified subgroups of screening HIV-1 RNA levels, sex, and race or ethnicity was evaluated by Wald chi-square tests of interaction constructed from the weighted sum of the squared differences of the subgroup-specific treatment effects from a pooled treatment effect. Weighting was by the inverse of the variances of the subgroup-specific effects. Pairwise comparisons of the distributions of changes in continuous outcomes by treatment group used t tests. The key toxicity secondary end point was defined as the time from initiation of treatment to the first grade 2, 3, or 4 sign or symptom (grade 3 or 4 if after week 48) or any grade 3 or 4 laboratory abnormality while the patient was receiving the randomized treatment (as-treated). A pre­ specified sensitivity analysis excluded hyperbilirubinemia and elevated creatine phosphokinase levels from the end point definition. Review of the data raised a concern about underestimation of rates of adverse events due to treatment switching, so a further sensitivity analysis was done that included all qualifying adverse events regardless of status on randomized treatment (ITT analysis). An independent data safety monitoring board re­ viewed the data on 3 occasions according to prespecified monitoring guidelines; no concerns with the conduct, safety, or efficacy of the study were noted. Reported P values were 2-sided. Analyses were done using SAS soft­ ware, version 9.2 (SAS Institute). R ole o f th e F u n d in g Source

The study was funded by the Division of AIDS of the National Institute of Allergy and Infectious Diseases, www.annals.org

O

r ig in a l

R

esea rch

which provided input into protocol design, conduct, and manuscript development. Bristol-Myers Squibb, Merck, Janssen Therapeutics, and Gilead Sciences provided the study medications (atazanavir, raltegravir, darunavir, and combination TDF plus emtricitabine, respectively) and had input into protocol development and review of the manuscript. Ritonavir was not provided by the study; how­ ever, copayments were reimbursed for participants ran­ domly assigned to ritonavir-containing regimens where al­ lowed by law. The ACTG was solely responsible for final review and approval of the manuscript.

R esults

Between 22 May 2009 and 9 June 2011, a total of 1814 participants from 57 sites enrolled in the study. After being randomly assigned, 5 participants were found to be ineligible because of an acute illness, an exclusionary resis­ tance mutation, or previous use of antiretroviral therapy; per protocol, these participants were excluded from all analyses. The final analysis therefore included 1809 partic­ ipants (Figure 1). Demographic characteristics of the population were well-balanced among the 3 groups (Table 1). Women composed 24.0% of the study population, 34.0% of the participants were non-Hispanic white, the median (first to third quartile) C D 4+ count was 0.308 (0.170 to 0.425) X 10; cells/L, the median (first to third quartile) HIV-1 RNA level at entry was 4.62 (4.14 to 5.12) log10 copies/mL, and 69.4% of the population had a baseline HIV-1 RNA level less than 100 000 copies/mL. A total of 262 participants (14.5%) discontinued follow-up before completion of the study in June 2013. Among these, 162 (61.8%) discontinued after reaching a virologic failure end point and thus contributed events to the primary virologic failure analysis. The cumulative probability of study discontinuation before virologic failure over time did not differ among the 3 groups (P = 0.134; log-rank test). Ten persons died in the ritonavir-boosted atazanavir group, 6 in the raltegravir group, and 13 in the ritonavir-boosted darunavir group; the most common rea­ sons were sudden death of unknown cause and cancer (Ap­ pendix Table 1 , available at www.annals.org). A total of 295 participants (16.3%) experienced con­ firmed virologic failure, which occurred in 54 participants (18.3%) before 24 weeks, 67 (22.7%) between weeks 24 and 48, and 174 (59.0%) after week 48. The cumulative incidence of virologic failure by treatment group and dif­ ference by 96 weeks is shown in Figure 2 (.A). The cumu­ lative probability of virologic failure by 96 weeks was 12.6% in the ritonavir-boosted atazanavir group, 9.0% in the raltegravir group, and 14.9% in the ritonavir-boosted darunavir group. For all pairwise treatment comparisons, the 97.5% CIs were within the prespecified equivalence bound of ± 10%, meeting the definition of equivalence of the 3 regimens regarding this end point; results for as7 October 2014 Annals of Internal Medicine Volume 161 • Number 7 I463

O

r ig in a l

Figure 1.

R

esea rch

NNRTI-Sparing Antiretroviral Regimens for Treatment-Naive Volunteers Infected With HIV-1

S tu d y f l o w d ia g r a m .

Completed 96 w k (n = 556 [9 1 .9 % ]) Participants discontinued before study closure (n = 89 [1 4 .7 % ])ll Died: 10 Lost to fo llo w -u p : 29 Unable to travel to clinic: 28 Nonadherent or other reasons: 22

Completed 96 w k (n = 560 [92.9% ]) Participants discontinued before study closure (n = 72 [11.9% ])ll Died: 6 Lost to follow -up: 23 Unable to travel to clinic: 35 Nonadherent or other reasons: 8

Completed 96 w k (n = 546 [90.8% ]) Participants discontinued before study closure (n = 101 [16 .8% ])ll Died: 13 Lost to follow -up: 34 Unable to travel to clinic: 30 Nonadherent or other reasons: 24

Analyzed (n = 605)

Analyzed (n = 603)

Analyzed (n = 601)

A total of 114 participants discontinued randomized treatment before virologic failure: 28 in the ATV/r group, 35 in the RAL group, and 51 in the DRV/r group. ART = antiretroviral therapy; ATV/r = atazanavir plus ritonavir; DRV/r = darunavir plus ritonavir; RAL = raltegravir. * 10 participants were excluded for >1 reason. t Per protocol, participants found to be ineligible were excluded from analysis, f 5 participants never started ART. § 4 participants never started ART. || Study closure occurred when the last participants completed 96 wk of follow-up.

treated analyses were consistent (Figure 2, B). For all com­ parisons, differential treatment effects by screening HIV-1 RNA level {P for interaction > 0.32), race or ethnicity (P for interaction > 0.150), and sex {P for interaction > 0.170) were not apparent (Appendix Figure 1, available at www.annals.org). The primary tolerability end point of toxicityassociated discontinuation of the randomized treatment was equivalent between the raltegravir and ritonavirboosted darunavir groups at 96 weeks, whereas ritonavirboosted atazanavir resulted in a 12.7% (97.5% Cl, 9.4% 4 6 4 7 October 2014 Annals of Internal Medicine Volume 161 • Number 7

to 16.1%) higher incidence of tolerability discontinuation than raltegravir and a 9.2% (97.5% Cl, 5.5% to 12.9%) higher incidence of discontinuation than ritonavir-boosted darunavir (Figure 2, Q . Ninety-five participants discontin­ ued ritonavir-boosted atazanavir because of toxicity, 46 (48%) because of jaundice or an elevated serum bilirubin level, and 25 (26%) because of nausea or other gastrointes­ tinal toxicities (Appendix Table 2, available at www.annals .org). In the ritonavir-boosted darunavir group, 14 of the 32 (44%) tolerability end points were attributed to gastrointestinal symptoms; in the raltegravir group, w w w . a n n a ls .o r g

N N RTI-Sparing Antiretroviral Regimens for Treatment-Naive Volunteers Infected W ith HIV-1

T a ble 1.

O r ig in a l

R esearch

Baseline Characteristics of the Study Population, by Treatment Group

Characteristic

Treatment Group Total (n = 1809)

Women, n (% )

435 (24.0)

Median age, y

37

A TV /r (n = 605)

RAL (n = 603)

DRV/r (n = 601)

144 (23.8)

148 (24.5)

143 (23.8)

37

36

37

Race/ethnicity, n (% )

White non-Hispanic Black non-Hispanic Hispanic* Asian or Pacific Islander American Indian or Alaskan Native >1 race Mean baseline C D 4+ count, x 109 cells/L Baseline C D 4+ count, n (% ) 0.500 x 109 cells/L Median baseline HIV-1 RNA level, /og10 copies/mL

615 757 390 30 6 7

(34.0) (41.8) (21.6) (1.7) « 1 .0 ) « 1 .0)

0.308 216(11.9) 320 (17.7) 528 (29.2) 504 (27.9) 241 (13.3) 4.62

212 252 125 11 2 2

(35.0) (41.6) (20.7) (1.8) (

i i■ i i i i -2 0 -1 0

i

D ifference in 9 6 -w k cum ulative incidence (9 7 .5 % Cl)

ATV/r (13.9%) vs. RAL (0.9%) 12 .7 % (9 .4 % to 1 6.1% )

DRV/r (4.7%) vs. RAL (0.9%) 3 .6 % (1.4 % to 5 .8 % )

! Favors D R V /r

ATV/r (13.9%) vs. DRV/r (4.7%) 0

10

9 .2% (5.5 % to 12 .9 % )

20

0 .5 0 -

£

0 .2 5

a

0 .0 0 -

j5 3

E

3

U

— I— 24

48

-- 1----- 1--64

80

— i-------- 1---------r

96

112

128

144

Study W eek Participants in th e risk set, n

Participants in th e risk set, n

317

A T V /r

605

548

522

467

34 9

A T V /r

60 5

536

494

427

RAL

60 3

585

562

53 4

411

RAL

60 3

57 4

54 5

511

30 7

D R V /r

601

57 6

55 3

51 7

39 2

D R V /r

601

559

520

470

35 8

------- A T V /r -------- RAL ......... D R V /r

ATV/r = atazanavir plus ritonavir; DRV/r = darunavir plus ritonavir; RAL — raltegravir. * “As-treated” signifies that a participant’s time to virologic failure was censored at the time of the competing event of discontinuation of randomized treatment.

only 2 of 8 (25%) tolerability end points were for this reason. Some evidence of differential treatment effects for tol­ erability by screening HIV-1 RNA level was apparent for ritonavir-boosted atazanavir versus raltegravir (P = 0.036) and by sex for ritonavir-boosted darunavir versus raltegra­ vir (P = 0.047). A greater tolerability benefit of raltegravir compared with ritonavir-boosted atazanavir was observed among participants with a baseline HIV-1 RNA level less 4 6 6 7 October 2014 Annals of Internal Medicine Volume 161 • Number 7

than 100 000 copies/mL; similarly, a greater tolerability benefit of raltegravir over ritonavir-boosted darunavir was observed in women (Appendix Figure 2, available at www .annals.org). No other differential treatment effects were apparent (P > 0.128). In pairwise comparisons of the cumulative incidence of a prespecified secondaiy composite end point of time to the first episode of virologic or tolerability failure, ritonavir-boosted atazanavir was inferior to raltegravir by www.annals.org

NNRTI-Sparing Antiretroviral Regimens for Treatment-Naive Volunteers Infected With HIV-1

Figure 3.

P ro p o r tio n o f p a r tic ip a n ts w it h a n H IV - 1 R N A le v e l

< 5 0 c o p ie s /m L o v e r tim e , w it h 9 5 %

C Is.

H I V - 1 R N A L e v e l < 5 0 C o p ie s /m L , R e g a rd le s s o f A R T C h a n g e ( IT T A n a ly s is )

A T V /r

605

563

553

515

394

RAL

603

566

555

526

410

D R V /r

601

564

542

518

387

H I V - 1 R N A L e v e l < 5 0 C o p ie s /m L W h i l e R e c e iv in g R a n d o m iz e d A R T (A s -T re a te d )

A T V /r

601

509

482

425

361

RAL

599

555

534

501

386

D R V /r

597

546

514

481

363

O

r ig in a l

R esea r c h

14.9% (97.5% Cl, 10.2% to 19.6%) and to ritonavirboosted darunavir by 7.5% (97.5% Cl, 2.3% to 12.7%) (Figure 2, D). Ritonavir-boosted darunavir was inferior to raltegravir by 7.5% (97.5% Cl, 3.2% to 11.8%). No dif­ ferential treatment effects by HIV-1 RNA level, race or ethnicity, or sex were apparent (P > 0.090) (Appendix Figure 3, available at www.annals.org). A TLOVR end point analysis using a virologic failure threshold greater than 200 copies/mL was consistent with the ITT compos­ ite end point results. The proportion of participants with an HIV-1 RNA level of 50 copies/mL or less at 96 weeks by ITT analysis (regardless of treatment status) was 88.3% for ritonavirboosted atazanavir, 93.9% for raltegravir, and 89.4% for ritonavir-boosted darunavir (Figure 3, top). As-treated analyses yielded similar proportions of virologic suppres­ sion as those observed by ITT analyses (Figure 3, middle). The proportion of participants with an HIV-1 RNA level of 50 copies/mL or less who were receiving randomized treatment at week 96 according to the FDA snapshot def­ inition was 62.6% for ritonavir-boosted atazanavir, 79.8% for raltegravir, and 72.7% for ritonavir-boosted darunavir (Figure 3, bottom). Figure 4 shows the corresponding out­ comes for participants not receiving randomized treatment at week 96 according to the FDA snapshot definition. The mean change in C D 4+ count from baseline to week 96 was 0.280 X 109 cells/L for ritonavir-boosted atazanavir, 0.288 X 109 cells/L for raltegravir, and 0.256 X 109 cells/L for ritonavir-boosted darunavir. The difference in the mean change in C D 4+ count from base­ line was greater for raltegravir (0.032 X 109 cells/L [95% Cl, 6 to 57 X 109 cells/L]; P = 0.005) and ritonavirboosted atazanavir (0.028 X 109 cells/L [95% Cl, 3 to 54 X 109 cells/L]; P = 0.011) than for ritonavir-boosted darunavir. O f 295 participants who met criteria for virologic fail­ ure, 294 had HIV-1 sequence data available at baseline and

H I V - 1 R N A L e v e l < 5 0 C o p ie s /m L a n d R e c e iv in g R a n d o m iz e d A R T ( IT T a n d S n a p s h o t A n a ly s e s )

A T V /r

605

605

605

605

471

RAL

603

603

603

603

483

D R V /r

601

601

601

601

468

www.annals.org

E rror bars for all panels indicate pointw ise 95% CIs at prespecified key study weeks. A R T = antiretroviral therapy; A T V /r = atazanavir plus ritonavir; D R V /r = darunavir plus ritonavir; IT T = intention-to-treat; RAL = raltegravir. T o p . T h e n um ber o f participants contributing data at each tim e p o in t (the denom inator for the proportion) represents the nu m b er o f participants w ith an H IV -1 R N A level available for th at study week, regardless o f status o f random ized treatm ent. T h e n um erator rep­ resents the num ber o f participants w ith an H IV -1 RN A level < 5 0 copies/mL; missing evaluations are assum ed to be missing com pletely at random , and loss to follow -up is assum ed to be noninform ative. M iddle. T h e n um ber o f participants contributing data at each tim e point (the d enom inator for the proportion) represents the n um ber o f participants continuing random ized treatm ent w ith an H IV -1 R N A level available for th at study week. T h e num erator represents the num ber o f participants w ith an HIV-1 R N A level ^ 5 0 copies/m L; m issing evaluations are as­ sum ed to be missing com pletely at random , and loss to follow -up is assum ed to be noninform ative. B ottom . T h e n um ber o f participants contributing data at each tim e p o in t (the denom inator for the p rop o r­ tion) represents the num ber o f participants w ith the potential for follow -up to th at week based on date o f random ization. T h e num erator represents the n um ber o f participants w ith an H IV -1 R N A level < 5 0 copies/m L still receiving random ized treatm ent. 7 October 2014 Annals of Internal Medicine Volume 161 • Number 7 467

O r ig in a l R esearc h

NNRTI-Sparing Antiretroviral Regimens for Treatment-Naive Volunteers Infected W ith H IV -1

F igure 4. O u tc o m e s a t w e e k 9 6 , a c c o rd in g to th e F D A s n a p s h o t d e fin itio n .

greater increases in these lipid levels than the raltegravir group ( P < 0.001 for all). A total of 7, 4, and 12 partici­ pants treated with ritonavir-boosted atazanavir, raltegravir, and ritonavir-boosted darunavir, respectively, had a grade 3 or 4 elevation in serum creatinine level. A total of 20, 9, and 23 participants in the ritonavir-boosted atazanavir, raltegravir, and ritonavir-boosted darunavir groups, respec­ tively, substituted any component of the fixed-dose combination of TDF plus emtricitabine while receiving randomized therapy.

D

ATV/r

RAL

DRV/r

Treatment

ART = antiretroviral therapy; ATV/r = atazanavir plus ritonavir; DRV/r = darunavir plus ritonavir; FDA = U.S. Food and Drug Ad­ ministration; RAL = raltegravir.

239 (81.0%) had available data at the time of virologic failure; sequencing failure due to low-level viremia was the primary reason for lack of sequence data (Table 2). Over­ all, virologic failure with resistance occurred in 3.0% of study participants randomly assigned to raltegravir (2 of whom developed intermediate-level resistance to dolutegravir) and in 1.5% or fewer of those in either boosted PI group. Twenty-seven participants randomly assigned to a ritonavir-boosted PI regimen who experienced virologic failure had integrase genotyping. Two participants had ev­ idence of treatment-emergent raltegravir resistance despite the absence of known exposure to an integrase inhibitor. The 96-week cumulative incidence of the first clinical or laboratory adverse event was 80.8% in the ritonavirboosted atazanavir group, 59.5% in the raltegravir group, and 64.9% in the ritonavir-boosted darunavir group; when elevated bilirubin or creatine phosphokinase levels were ex­ cluded, these cumulative incidences were 62.3%, 59.3%, and 64.9%, respectively. Table 3 lists clinical adverse events and laboratory abnormalities (grade 2 or higher) occurring in at least 5% of the participants in any group. The most frequently reported abnormality was elevated bil­ irubin level; 286 of the 295 participants with this finding were in the ritonavir-boosted atazanavir group. In prespecified safety analyses excluding participants with elevated bilirubin levels, all 3 study regimens were equivalent regarding the cumulative incidence of a first adverse event over 96 weeks; ITT analyses yielded consis­ tent findings. Fasting low-density lipoprotein cholesterol and triglyceride levels increased in the ritonavir-boosted Pi-containing groups to a similar degree, and each had 468

7 October 2014 Annals o f Internal Medicine Volume 161 • Number 7

is c u s s io n

In this prospective, randomized, open-label study of the initial treatment of HIV infection, all 3 study regimens yielded equivalent rates of virologic suppression over 96 weeks. The racial and sexual diversity of our study popu­ lation allowed us to examine the effect of these factors on treatment success. Rates of virologic failure remained equivalent among the 3 groups when the analysis was strat­ ified by baseline viral load, race/ethnicity, and sex. The ritonavir-boosted atazanavir-containing regimen was less well-tolerated than regimens containing ritonavir-boosted darunavir or raltegravir. A composite assessment of viro­ logic efficacy and tolerability found that the raltegravirbased regimen was superior to both Pi-containing regi­ mens and that ritonavir-boosted darunavir was superior to ritonavir-boosted atazanavir. The tolerability result was caused primarily by participant-driven regimen change for jaundice in the ritonavir-boosted atazanavir group and by nonbiliary gas­ trointestinal toxicity for both Pi-containing regimens compared with the raltegravir regimen. Although ritonavirboosted atazanavir was less well-tolerated than ritonavirboosted darunavir and raltegravir across all subgroups, the relative benefit of raltegravir over ritonavir-boosted daruna­ vir was greater in women than in men. This result is con­ sistent with findings of the GRACE (Gender, Race, and Clinical Experience) study, in which more women discon­ tinued ritonavir-boosted darunavir than men (15). Crea­ tine phosphokinase levels were elevated in fewer than 1% of raltegravir-treated participants, in contrast to other re­ ports (16). The secondary composite end point aggregates 2 ma­ jor reasons for regimen change in clinical practice: failure to achieve and maintain virologic suppression and lack of tolerability. In this analysis, subtle differences in rates of virologic failure among groups, combined with small dif­ ferences in tolerability between raltegravir and ritonavirboosted darunavir, yielded the observed superiority of the raltegravir regimen over both Pi-based comparators. Jaun­ dice and other gastrointestinal toxicity leading to discon­ tinuation of randomized therapy drove the composite superiority of ritonavir-boosted darunavir over ritonavirboosted atazanavir. w w w .a n n a ls .o rg

NNRTI-Sparing Antiretroviral Regimens for Treatment-Naive Volunteers Infected With HIV-1

Participants receiving ritonavir-boosted atazanavir and raltegravir had greater increases in mean C D 4+ cell counts over 96 weeks of therapy than those receiving ritonavirboosted darunavir, although the magnitude of the differ­ ences may be of limited clinical significance. Increases in triglyceride and fasting low-density lipoprotein cholesterol levels observed in both Pi-based regimens did not differ between ritonavir-boosted atazanavir and ritonavir-boosted darunavir, consistent with 48-week results from a prior study, and were greater than those observed for raltegravir (17). Despite the inclusion of TDF in each group, only 1.27% of participants had a grade 3 or 4 elevation of the serum creatinine level, possibly because of the careful prestudy screening and close on-study monitoring. To our knowledge, Dejesus and colleagues (18) re­ ported the first study showing that an integrase inhibitorbased regimen (combined TDF, emtricitabine, cobicistat, and elvitegravir) was noninferior to a Pi-based regimen (TDF plus emtricitabine with ritonavir-boosted atazanavir) for the initial treatment of HIV-1 infection. In contrast to our study, the rate of ocular icterus was 14% and only 5.1% of participants discontinued ritonavir-boosted atazanavir for toxicity compared with 13.9% in our study. Moreover, total or low-density lipoprotein cholesterol lev­ els did not significantly differ among groups, probably be­ cause of the known effects of cobicistat on lipid metabo­ lism (19, 20). Important differences from our study are that a change in study treatment before week 48 was clas­ sified as treatment failure and the rates of medication dis­ continuation did not differ among groups, perhaps because of the placebo-controlled nature of Dejesus and colleagues’ study.

O riginal Research

Table 2. Genotypic Analysis for Resistance to Randomized Medications at Virologic Failure, by Treatm ent Group

Variable

Treatment Group, n ATV/r

RAL

DRV/r

V iro lo g ic fa ilu re

95

85

115

G e n o ty p e te s tin g c o m p le te

75

65

99

A n y resistance dete cte d

9

18

4

PI resistance d ete c te d

0

0

0

N R T I-o n ly resistance d ete c te d

8

7

3

E m tric ita b in e * TDFt TD F a nd e m tric ita b in e

5

7

3

2

0

0 0

1

0

IN I-o n ly resistance d e te c te d *

1

1

1

NRTI a nd INI resistance d ete c te d

0

10

0

E m tric ita b in e and RAL

0

7

0

E m tricita bin e , TD F, and RAL

0

3

0

ATV/r - atazanavir plus ritonavir; DRV/r = darunavir plus ritonavir; INI = integrase inhibitor; NRTI = nucleoside reverse transcriptase inhibitor; PI = pro­ tease inhibitor; RAL = raltegravir; TDF — tenofovir disoproxil fumarate. * Emtricitabine resistance mutations: M184V (21 participants), M184I (4 partic­ ipants), or M184V/I (1 participant). t TDF resistance mutations: K65R (3 participants), K70N (2 participants), T215A (1 participant), or T215I (1 participant). 4- 15 of 75 participants in the ATV/r group and 12 of 99 participants in the DRV/r group were tested for INI resistance. RAL resistance: N155H (6 partici­ pants) plus >1 of E92Q (4 participants), E138K (2 participants), V151I (3 par­ ticipants), E157Q (2 participants), and G163R (1 participant) or single mutations at E92Q (1 participant), T97A (1 participant receiving ATV/r), N155H (2 par­ ticipants), E157Q (2 participants), and G163K (1 participant).

Clotet and associates (21) reported a randomized, open-label study showing superiority of dolutegravir over ritonavir-boosted darunavir for initial treatment of H IV -1 infection, driven by tolerability differences and higher rates of virologic suppression for dolutegravir among the subset

Table 3. Grade 2 or Higher Adverse Effects Occurring in > 5 % of Participants, by Treatm ent Group Adverse Effect

Treatment Group ATV/r Grade, n* 2

3

RAL Total, n (%)

4

Grade, n* 2

DRV/r Total, n (%)

Grade, n*

Total, n (%)

3

4

D ia rrh ea

35

11

0

4 6 (7 .6 )

26

10

0

N ausea

3 6 (6 .0 )

36

8

46

1

6

0

52 (8 .6 )

4 5 (7 .4 )

21

12

0

33 (5 .5 )

29

12

0

41 (6 .8 )

2

3

4

V o m itin g

22

7

1

3 0 (5 .0 )

15

9

0

A b d o m in a l pain

2 4 (4 .0 )

21

11

13

0

17

32 (5 .3 )

1

31 (5 .1 )

6

10

H eadache

1

17 (2 .8 )

13

14

23

10

2

2 9 (4 .8 )

2

35 (5 .8 )

35

7

4 4 (7 .3 )

0

Pain in e x tre m ity

42 (7 .0 )

30

27

12

2

14

1

4 2 (6 .9 )

31

14

A rth ra lg ia

0

4 5 (7 .5 )

17

18

13

1

8

32 (5 .3 )

0

25 (4 .1 )

17

4

1

Back pain

22 (3 .6 )

13

14

14

1

4

2 8 (4 .7 )

0

18 (3 .0 )

21

10

F atigue

0

31 (5 .1 )

9

32

12

0

6

21 (3 .5 )

1

3 9 (6 .4 )

26

5

0

C ou g h

31 (5 .1 )

26

7

33

0

9

0

33 (5 .5 )

42 (6 .9 )

32

8

0

4 0 (6 .6 )

31

5

0

3 6 (6 .0 )

0

2 8 (4 .6 )

8

14

1

23 (3 8)

35 (5 .8 )

18

7

2

2 7 (4 .5 )

0

4

0

D yspnea

16

9

1

2 6 (4 .3 )

16

12

Pyrexia

16

9

1

2 6 (4 .3 )

Elevated b lo o d b iliru b in le v e l*

22

217

47

3

30

11

15

Decreased b lo o d p ho sp h oru s le v e l* E levated b lo o d g lucose level§

25

10

0

2 8 6 (4 7 .3 )

0

5

0

1

3 4 (5 .6 )

4

24

1

2 9 (4 .8 )

2

35

0

0

3 7 (6 .2 )

2 6 (4 .3 )

15

9

2

2 6 (4 .3 )

15

11

1

2 7 (4 .5 )

5 ( < 1 .0)

4 ( < 1 .0)

ATV/r = atazanavir plus ritonavir; RAL = raltegravir; DRV/r = darunavir plus ritonavir. * Grade 2 = moderate, grade 3 = severe, and grade 4 = potentially life-threatening, t Grade 2 = 1.6-2.5 X upper limit of normal, grade 3 = 2.6-5.0 X upper limit of normal, and grade 4 > 5.0 X upper limit of normal. t Grade 2 = 0.65-0.80 mmol/L, grade 3 = 0.32-0.64 mmol/L, and grade 4 = 27.75 mmol/L (>500 mg/dL).

www.annals.org

7 October 2014 Annals oflnternal Medicine IVolume 161 • Number 7 I469

O riginal R esearch

NNRTI-Sparing Antiretroviral Regimens for Treatment-Naive Volunteers Infected W ith HIV-1

of participants with baseline HIV-1 RNA levels greater than 100 000 copies/mL. Important differences in the study population (only 15% were women, 28% were non­ white, and the nucleoside backbone was not uniform) compared with our study may have contributed to differ­ ences in subgroup results. The observed development of antiretroviral resistance at virologic failure is consistent with other studies, with minimal resistance to Pis and supporting nucleoside back­ bones in both Pi-containing regimens. Patients randomly assigned to raltegravir frequently developed integrase mu­ tations and nucleoside backbone resistance when virologic failure occurred. Two of the raltegravir-treated participants also developed intermediate-level dolutegravir resistance, potentially limiting the use of this medication. The appar­ ent emergence of integrase-inhibitor resistance in 2 of 27 participants receiving Pis with no known exposure to inte­ grase inhibitors is unexplained. Limitations of our study include its open-label design. Although our study reduced pill burden compared with double-blind dummy designs, participants and investiga­ tors may have been more prone to switch ritonavir-boosted atazanavir treatment for signs or symptoms of elevated bil­ irubin levels. We did not provide ritonavir, but copay re­ imbursements limited withdrawal due to difficulty obtain­ ing this agent. Analyses of the primary study end points assumed that censoring of participants who prematurely discontinued therapy was noninformative. Although statistical methods exist for alternative approaches to such cases— such as in­ verse probability of censoring weighting— these analyses rely heavily on modeling assumptions that would be ques­ tionable in this setting. Further, they would evaluate the relative benefit of these regimens in a setting where treat­ ment changes (adverse events that were associated or oth­ erwise) do not occur, a situation that is not clinically rele­ vant. These approaches could have been used to adjust for censoring due to loss to follow-up, but this adjustment was not believed to be warranted given the low rate of such withdrawals in the study (4% for virologic failure and 4% for the composite outcome) and the consistency of the results of our composite outcome with those of the FDA end point of TLOVR that treats such outcomes as failures. O ur study allowed participants who were intolerant to randomized treatment to change to study-provided options that were convenient, well-tolerated, and potent. Partici­ pants’ threshold for discontinuing less tolerable treatments may have been lower than in prior studies in which dis­ continuation required participants to leave the study or receive suboptimal alternatives. To our knowledge, our trial is the first head-to-head comparison of these initial treatment regimens recom­ mended by the U.S. Department of Health and Human Services and provides useful information to guide clinicians about choosing among them. All 3 regimens achieved equivalent and high rates of virologic suppression, whereas

1

4 7 0 7 October 2014 Annals of Internal Medicine Volume 161 • Number 7

raltegravir had a more favorable tolerability profile and caused less elevation in lipid levels. Among Pi-containing regimens, ritonavir-boosted darunavir was better tolerated than ritonavir-boosted atazanavir, primarily because of dif­ ferences in hyperbilirubinemia. When tolerability and virologic response are consid­ ered together, raltegravir-based therapy was superior overall to both Pi-based therapies and ritonavir-boosted darunavir was superior to ritonavir-boosted atazanavir. An advantage of the ritonavir-boosted Pis over the raltegravir-based reg­ imen is the reduced likelihood of drug resistance if viro­ logic failure occurs. From Grady Memorial Hospital and Emory University School of Med­ icine, Atlanta, Georgia; Center for Clinical AIDS Research and Educa­ tion and University of California, Los Angeles, Los Angeles, California; Harvard School of Public Health, Brigham and Women’s Hospital, Har­ vard Medical School, and Boston University School of Medicine, Bos­ ton, Massachusetts; National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, Maryland; Johns Hopkins Uni­ versity, Baltimore, Maryland; Northwestern University Feinberg School of Medicine, Chicago, Illinois; Case Western Reserve University, Cleve­ land, Ohio; University of California, San Francisco, School of Pharmacy, San Francisco, California; University of Cincinnati College of Medicine, Cincinnati, Ohio; Washington University School of Medicine, St. Louis, Missouri; The Ohio State University College of Medicine, Columbus, Ohio; Vanderbilt University School of Medicine, Nashville, Tennessee; University of North Carolina School of Medicine, Chapel Hill, North Carolina; University of California, San Diego, School of Medicine, San Diego, California; Janssen Scientific Affairs, Titusville, New Jersey; Merck, Whitehouse Station, New Jersey; Gilead Sciences, Foster City, California; and Bristol-Myers Squibb, New York, New York. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institute of Allergy and Infectious Diseases or the National Institutes of Health. D is c la im e r :

By the National Institute of Allergy and Infectious Diseases (award UM1AI068636), National Institute of Mental Health, and National Institute of Dental and Craniofacial Research. The proto­ col received financial support from the ACTG and the Site Data Man­ agement Center (grant UM1AI68634), the ACTG specialty laboratories listed in the article, and the 57 clinical research sites. The list of person­ nel and grant numbers from these clinical research sites are listed in Appendix 2 (available at www.annals.org). F in a n c ia l S u p p o r t:

D is c lo s u re s : Disclosures can be viewed at www.acponline.org/authors /icmje/ ConflictOflnterestForms.doimsN um= M 14-1084.

R e p r o d u c ib le R e s e a rc h S ta te m e n t: Study protocol, statistical code, and data set: Available through written agreement with the Division of AIDS of the National Institute of Allergy and Infectious Diseases.

R e p r in ts : Jeffrey L. Lennox, MD, Professor of Medicine, Division of Infectious Diseases, Emory University School of Medicine, 69 Jesse Hill Jr. Drive SE, Suite 202, Atlanta, GA 30303; e-mail, [email protected]. R e q u e s ts f o r S in g le

Current author addresses and author contributions are available at www.annals.org. w w w . a n n a ls .o r g

N N R T I-S p a rin g A ntiretroviral Regim ens for T reatm ent-N aive V olunteers Infected W ith H IV -1

R e fe re n c e s 1. Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the Use of Antiretroviral Agents in HIV-1-Infected Adults and Adolescents. Washington, DC: U.S. Department of Health and Human Services; 2014. Ac­ cessed at http://aidsinfo.nih.gov/ContentFiles/AdultandAdolescentGL.pdf on 3 June 2014. 2. W orld Health Organization. Consolidated Guidelines on the Use of Antiretroviral Dmgs for Treating and Preventing HIV Infection: Recommendations for a Public Health Approach. Geneva: World Health Organization; 2013. Accessed at www.who.int/hiv/pub/guidelines/arv2013/download/en/ on 3 June 2014. 3. Periard D, Telenti A, Sudre P, Cheseaux JJ, Halfon P, Reymond MJ, et al. Atherogenic dyslipidemia in HIV-infected individuals treated with protease in­ hibitors. The Swiss HIV Cohort Study. Circulation. 1999;100:700-5. [PMID: 10449690] 4. D ’Ascenzo F, Cerrato E, Biondi-Zoccai G, Moretti C, Omede P, Sciuto F, et al. Acute coronary syndromes in human immunodeficiency virus patients: a meta-analysis investigating adverse event rates and the role o f anuretroviral ther­ apy. Eur Heart J. 2012;33:875-80. [PMID: 22187508] doi:10.1093/eurheartj /ehr456

O r i g i n a l R esearch

/GuidanceComplianceRegulatoryInformation/Guidances/ucm070968.pdf on 1 July 2014. 12. U.S. Department o f Health and Hum an Services; U.S. Food and Drug Administration; Center for Drug Evaluation and Research. Guidance for Indus­ try: Human Immunodeficiency Virus-1 Infection— Developing Antiretroviral Drugs for Treatment. Draft Guidance. Silver Spring, MD: U.S. Food and Drug Administration; 2013. Accessed at www.fda.gov/downloads/Drugs /GuidanceComplianceRegulatoryInformation/Guidances/UCM355128.pdf on 1 July 2014. 13. Wiens BL, Iglewicz B. Design and analysis of three treatment equivalence trials. Control Clin Trials. 2000;21:127-37. [PMID: 10715510] 14. Prentice RL, Kalbfleisch JD. Hazard rate models with covariates. Biometrics. 1979;35:25-39. [PMID: 497336] 15. Currier J, Averitt Bridge D, Hagins D, Zorrilla CD, Feinberg J, Ryan R, et al; GRACE (Gender, Race, And Clinical Experience) Study Group. Sexbased outcomes of darunavir-ritonavir therapy: a single-group trial. Ann Intern Med. 2010;153:349-57. [PMID: 20855799] doi: 10.7326/0003-4819-153-6 -201009210-00002 16. Monteiro P, Perez I, Pich J, Gatell JM , Martinez E. Creatine kinase eleva­ tion in HIV-l-infected patients receiving raltegravir-containing antiretroviral therapy: a cohort study. J Antimicrob Chemother. 2013;68:404-8. [PMID: 23109185] doi:10.1093/jac/dks416 17. Aberg JA, Tebas P, Overton ET, Gupta SK, Sax PE, Landay A, et al. Metabolic effects of darunavir/ritonavir versus atazanavir/ritonavir in treatmentnaive, HIV type 1-infected subjects over 48 weeks. AIDS Res Hum Retroviruses. 2012;28:1184-95. [PMID: 22352336] 18. Dejesus E, Rockstroh JK, Henry K, Molina JM, Gathe J, Ramanathan S, et al; GS-236-0103 Study Team. Co-formulated elvitegravir, cobicistat, emtricitabine, and tenofovir disoproxil fumarate versus ritonavir-boosted atazanavir plus co-formulated emtricitabine and tenofovir disoproxil fumarate for initial treat­ ment of HIV-1 infection: a randomised, double-blind, phase 3, non-inferiority trial. Lancet. 2012;379:2429-38. [PMID: 22748590] doi:10.1016/S0140 -6736(12)60918-0

5. W orm SW, Kamara DA, Reiss P, Fontas E, De W it S, El-Sadr W , et al; D:A:D Study Group. Evaluation of HIV protease inhibitor use and the risk of sudden death or nonhemorrhagic stroke. J Infect Dis. 2012;205:535-9. [PMID: 22223855] doi: 10.1093/infdis/jir788 6. Boyd SD, Maldarelli F, Sereti I, Ouedraogo GL, Rehm CA, Boltz V, et al. Transmitted raltegravir resistance in an HIV-1 CRF_AG-infected patient. Antivir Ther. 2011;16:257-61. [PMID: 21447876] doi:10.3851/IMP1749 7. Young B, Fransen S, Greenberg KS, Thomas A, Martens S, St Clair M, et al. Transmission o f integrase strand-transfer inhibitor multidrug-resistant HIV-1: case report and response to raltegravir-containing antiretroviral therapy. Antivir Ther. 2011;16:253-6. [PMID: 21447875] doi:10.3851/IMP1748 8. Stein JH , Brown T T , Ribaudo HJ, Chen Y, Yan M, Lauer-Brodell E, et al. Ultrasonographic measures of cardiovascular disease risk in antiretroviral treatment-naive individuals with HIV infection. AIDS. 2013;27:929-37. [PMID: 23196938] doi: 10.1097/QAD.0b013e32835ce27e 9. National Institute of Allergy and Infectious Diseases. Division of AIDS Table for Grading the Severity of Adult and Pediatric Adverse Events. 28 December 2004. Accessed at www.niaid.nih.gov/LabsAndResources/resources /DAIDSClinRsrch/Documents/daidsaegradingtable.pdf on 28 March 2014. 10. Ribaudo H, Lennox J, Currier J, Kuritzkes D, Gulick R, Haubrich R, et al. Virologic failure endpoint definition in clinical trials: is using HIV-1 RNA threshold < 2 0 0 copies/mL better than < 5 0 copies/mL: an analysis of ACTG studies [Abstract]. Presented at 16th Conference on Retroviruses and Opportu­ nistic Infections, Montreal, Quebec, Canada, 8-11 February 2009. Abstract no 580.

19. Elion R, Cohen C, Gathe J, Shalit P, Hawkins T , Liu H C, et al; GS-US216-0105 Study Team. Phase 2 study of cobicistat versus ritonavir each with once-daily atazanavir and fixed-dose emtricitabine/tenofovir d f in the initial treat­ ment of H IV infection. AIDS. 2011;25:1881-6. [PMID: 21811136] doi: 10.1097/QAD.0b013e32834b4d48 20. Gallant JE, Koenig E, Andrade-Vdlanueva J, Chetchotisakd P, Dejesus E, Antunes F, et al. Cobicistat versus ritonavir as a pharmacoenhancer o f atazanavir plus emtricitabine/tenofovir disoproxil fumarate in treatment-naive HIV type 1-infected patients: week 48 results. J Infect Dis. 2013;208:32-9. [PMID: 23532097] doi: 10.1093/infdis/jitl22

11. Appendix B: Time to loss of virologic response algorithm. In: U.S. Depart­ ment o f Health and Human Services; U.S. Food and Drug Administration; Center for Drug Evaluation and Research. Guidance for Industry: Antiretroviral Dmgs Using Plasma HIV RNA Measurements—Clinical Considerations for Accelerated and Traditional Approval. Silver Spring, MD: U.S. Food and Drug Administration; 2002:20-1. Accessed at www.fda.gov/downloads/Drugs

21. Clotet B, Feinberg J, van Lunzen J, Khuong-Josses MA, Antinori A, Dum itru I, et al; ING114915 Study Team. Once-daily dolutegravir versus darunavir plus ritonavir in antiretroviral-naive adults with HIV-1 infection (FLAMINGO): 48 week results from the randomised open-label phase 3b study. Lancet. 2014;383:2222-31. [PMID: 24698485] doi:10.1016 /S0140-6736(14)60084-2

www.annals.org

7 October 2014 Annals of Internal Medicine Volume 161 • N u m b e r 7 |4 7 1

Copyright © American College of Physicians 2014.