Abstract
Background:
Randomized trials of long-acting injectable antiretroviral therapy (ART) in persons with HIV (PWH) who face challenges to oral medication adherence are lacking.
Methods:
We conducted a randomized, open-label trial among PWH with suboptimal ART adherence indicated by persistent HIV-1 RNA >200 copies/mL or loss to clinical follow-up. Enrolled participants received up to 24 weeks of adherence support and conditional economic incentives while on oral standard of care (SOC) ART (Step 1). Participants achieving HIV-1 RNA ≤200 copies/mL were randomized 1:1 to continue SOC or switch to monthly cabotegravir plus rilpivirine (CAB+RPV) long-acting injectable with or without oral lead-in (Step 2). The primary outcome was regimen failure, defined as virologic failure (confirmed HIV-1 RNA >200 copies/mL) or treatment discontinuation during Step 2.
Results:
Step 1 enrolled 453 participants; median age was 40 years, 63% were Black/African American, and 29% female sex at birth. There were 306 participants randomized into Step 2 (long-acting injectable n=152, SOC n=154). Step 2 randomization was stopped early based on superiority of long-acting injectable at secondary outcomes of virologic failure and treatment-related failure at a pre-planned review after a median follow-up of 48 weeks (interquartile range 20, 52). Cumulative probabilities of regimen failure by Week 48 were 22.8% in long-acting injectable and 41.2% in SOC (difference −18.4%, 98.4% CI, −32.4% to −4.3%; p=0.002). Adverse events reported were similar (cumulative probability difference 1.1%, 95% CI, −12.7%, 15.0%). Two participants in each arm with confirmed virologic failure acquired drug resistance-associated mutations.
Conclusions:
Monthly CAB+RPV long-acting injectable ART demonstrated superior efficacy in reducing regimen failure over SOC in PWH with adherence challenges. (Trial Registration Number: NCT03635788)
Antiretroviral therapy (ART) reduces mortality and disease progression in persons with HIV (PWH) and prevents HIV transmission.1 Although modern ART is mostly comprised of well-tolerated, fixed-dose, combination daily oral tablets, only 67% of PWH in the United States are estimated to be virally suppressed.2,3 Many PWH experience challenges to ART adherence such as social and structural barriers, side-effects, stigma, and competing priorities. Long-acting injectable (LAI) ART offers a less-frequent, directly-observed dosing alternative that could facilitate durable virologic suppression in this population.
The combination of LAI cabotegravir and rilpivirine (CAB/RPV) was approved by the United States (US) Food and Drug Administration in 2021 for PWH who are virologically suppressed on oral ART to be used as monthly (Q1M) or every 2-month (Q2M) injections.4-6 However, phase III studies generally excluded PWH with viremia and barriers to oral ART adherence. Emerging data from observational cohorts and case series suggest that durable viral suppression can be achieved with LAI in this population.7-9 These prompted recent changes to US-based treatment guidelines, which now recommend considering LAI CAB/RPV in PWH with suboptimal adherence to ART, advanced disease, and limited treatment options though it is a CIII recommendation (expert opinion).10,11 To date, no randomized clinical trials have evaluated this treatment strategy in this population.
We report the primary results of the Advancing Clinical Therapeutics Globally (ACTG) A5359 LATITUDE (Long-Acting Therapy to Improve Treatment SUccess in Daily LifE) trial, a phase III, prospective, randomized, multi-center, open-label trial comparing the safety and efficacy of monthly LAI CAB/RPV with daily oral standard of care (SOC) ART among PWH with a history of adherence challenges in the US.
METHODS
Trial Oversight and Study Participants
Eligible participants were ART-experienced, PWH ≥18 years of age without Hepatitis B co-infection, and no clinically relevant RPV or integrase strand-transfer inhibitor (INSTI) resistance-associated mutations with non-adherence, defined as poor virologic response while being prescribed ART, and/or loss to clinical follow-up with ART non-adherence, each for ≥6 consecutive months, and within 18 months prior to study entry. Full eligibility criteria, trial design and protocol are provided in the Supplementary Appendix at nejm.org.
The A5359 protocol was approved by both central and local institutional review boards. All participants provided written informed consent. The National Institute of Allergy and Infectious Diseases Division of AIDS provided regulatory sponsorship and held responsibility for clinical monitoring. ViiV Healthcare and Johnson & Johnson donated ART and contributed to trial design. All the authors vouch for the accuracy and completeness of the data and the fidelity to the protocol.
Study Design and Randomization
The trial was designed with two steps for the primary outcome followed by either an extension phase (Step 3) for participants continuing LAI or an observation phase (Step 4) for those discontinuing LAI (Figure 1A); this report focuses on the primary outcome. In Step 1, participants initiated SOC oral ART consisting of at least ≥3 drugs with ≥2 predicted to be fully active, including a boosted protease inhibitor and/or an INSTI for up to 24 weeks. Participants in Step 1 received adherence support and were eligible for conditional economic incentives at each Step 1 visit (maximum $675 total) for achieving benchmarks in visit completion and/or decreased viral load (Supplementary Appendix).12,13 Participants achieving virologic suppression during Step 1 were randomized 1:1 at Step 2 enrollment to continue SOC or switch to LAI for 52 weeks. Those randomized to LAI received an optional oral lead-in with daily CAB 30 mg and RPV 25 mg for approximately 4 weeks followed by loading injections of CAB 600 mg (3mL)/RPV 900 mg (3mL). Subsequent injections of CAB 400 mg (2mL)/RPV 600 mg (2 mL) were administered within 24 to 32 days after the previous injection for the second and third injections, and within 21 to 35 days thereafter. Step 2 participants received continued adherence support without conditional economic incentives. Randomization used permuted blocks with balancing by site. Three protocol versions were implemented prior to this primary analysis (Supplementary Appendix). Major changes among versions included: a) modification in the viral load threshold to advance into Step 2 from <50 copies/mL in version 1 to <200 copies/mL in version 2; b) shortening of the required time in Step 1 prior to randomization in Step 2 from 20 weeks in version 1 to 12 weeks in version 2 and 4 weeks in version 3; c) allowing enrollment of participants who were virally suppressed at screening but had an HIV VL >200 copies/mL within 12 months prior to study entry in version 3, and; d) optionality of optional oral lead-in in versions 2 and 3.
FIGURE 1. STUDY DESIGN AND CONSORT DIAGRAMS.
FIGURE 1A STUDY DESIGN
a Defined as positive Hepatitis B Virus surface antigen test or any detectable HBV DNA in participants with isolated HBV core antibody and HBV DNA.
b RPV rilpivirine
c INSTI integrase strand transfer inhibitor
d CEI, conditional economic incentives were distributed during Step 1 at week 2($75) for visit attendance and subsequently every 4 weeks ($75 weeks 4 and 8, $150 weeks 12, 16, 20) for HIV-1 RNA benchmark completion (see supplementary material) for a total $675. In protocol versions 2 and 3, participants could receive the full $675 once achieving HIV-1 RNA <200 copies/mL after weeks 12 and 4, respectively.
FIGURE 1B STEP 1 CONSORT DIAGRAM
FIGURE 1C STEP 2 CONSORT DIAGRAM
ART antiretroviral therapy; VL viral load; HBV Hepatitis B Virus; RPV rilpivirine; INSTI integrase stand transfer inhibitor; SOC standard of care; CEI conditional economic incentives; LAI Long-acting injectable cabotegravir+rilpivirine
Study procedures and outcomes
After enrollment, participants were asked to return every 4 weeks to assess conditional economic incentives eligibility in Step 1 or receive LAI injections in Step 2. Participants randomized to continuation of SOC in Step 2 had less frequent visits (every 8 weeks starting at week 12 and every 12 weeks at week 24) to more closely mimic US standard of care for individuals with adherence challenges. The primary outcome was regimen failure in Step 2, defined as the earliest occurrence of either confirmed virologic failure (two consecutive HIV-1 RNA >200 copies/mL post randomization) or permanent discontinuation of study treatment (including death and never initiated treatment). Secondary outcomes reported here included virologic failure, treatment-related failure (i.e., earliest occurrence of virologic failure or premature treatment discontinuation due to treatment-related AE), genotypic resistance with virologic failure, adverse events (AE), and plasma pharmacokinetics of CAB and RPV in participants with confirmed virologic failure. Additional participant reported outcomes are available in the Supplementary Appendix.
Statistical Analysis
A sample size of 320 participants in Step 2 was estimated to provide approximately 80% power to detect a 16-percentage point difference in the cumulative probability of regimen failure between the LAI and SOC arms, using a two-sided 5% type I error rate. The primary analysis was based on the intention-to-treat principle. Comparison of treatment arms was made using the difference in the Kaplan-Meier estimate for the week 48 cumulative probability of regimen failure. Analysis of secondary efficacy outcomes of virologic failure, treatment-related failure, and premature treatment discontinuation followed the same analysis approach. For analysis of virologic failure, participants who prematurely discontinued the study follow-up or died were censored. Treatment effects for the primary and secondary efficacy outcomes were evaluated in pre-specified subgroups. Details of other secondary outcome measures, supportive analyses and statistical analysis methods are available in the Supplementary Appendix.
Interim efficacy analyses were conducted using the Lan-DeMets spending function analog of the O’Brien-Fleming boundaries.14 At the second planned efficacy review on February 12, 2024 which included trial data through January 3, 2024, the Data and Safety Monitoring Board recommended halting randomization to Step 2 based on efficacy of secondary outcomes of virologic failure and treatment-related failure (Supplementary Appendix). The analyses in this manuscript include data collected up to the date of the efficacy review on February 12, 2024 with 76.6% information fraction, leading to a Type I error of 0.0161. Two-sided 98.4% CIs were provided for the primary and secondary efficacy outcomes. Other outcomes were summarized with point estimates and two-sided 95% CIs. The widths of the confidence intervals were not adjusted for multiplicity and should not be used to infer definitive treatment effects for secondary outcomes and subgroup analyses. All analyses were performed with the use of SAS software (version 9.4M7 for Linux).
RESULTS
Participants
Between March 28, 2019 and February 12, 2024, 820 adults were screened at 33 sites across the US and 453 eligible participants enrolled into Step 1 (Figure 1B). Median age for Step 1 participants was 40 (Q1, Q3: 31, 52) years, 29% were female sex at birth, 63% were Black, 17% were Hispanic, and 14% reported either ongoing or prior injection drug use (Table 1). Social determinants of health informing outcomes among PWH15,16 including household income and housing status (Table 1), as well as employment, insurance, and education level (Supplementary Appendix) were assessed. Alcohol use screening and urine drug testing were assessed at baseline and throughout the study (Table 1), in addition to assessments of depression and anxiety (Supplementary Table).
Table 1.
Step 1 and Step 2 demographics by Treatment arm and Total
| Characteristic | Step 1 | Step 2 | ||||
|---|---|---|---|---|---|---|
| LAI | SOC | Step 2 Overall | ||||
| (N=453) | (N=152) | (N=154) | (N=306) | |||
| Age, years | Median (Q1, Q3) | 40 (32, 51) | 41 (32, 53) | 42 (33, 52) | 42 (32, 52) | |
| 18-24 | 29 (6%) | 8 (5%) | 7 (5%) | 15 (5%) | ||
| 25-30 | 64 (14%) | 21 (14%) | 20 (13%) | 41 (13%) | ||
| 31-40 | 142 (31%) | 46 (30%) | 40 (26%) | 86 (28%) | ||
| 41-50 | 102 (23%) | 27 (18%) | 45 (29%) | 72 (24%) | ||
| 51+ | 116 (26%) | 50 (33%) | 42 (27%) | 92 (30%) | ||
| Sex at birth | Female | 133 (29%) | 43 (28%) | 41 (27%) | 84 (27%) | |
| Male | 320 (71%) | 109 (72%) | 113 (73%) | 222 (73%) | ||
| Race, n (%) | Black/African American | 286 (63%) | 94 (62%) | 103 (67%) | 197 (64%) | |
| White | 125 (28%) | 44 (29%) | 36 (23%) | 80 (26%) | ||
| Other/multiple/unknown | 42 (9%) | 14 (9%) | 15 (10%) | 29 (9%) | ||
| Ethnicity | Hispanic/Latinx | 79 (17%) | 29 (19%) | 26 (17%) | 55 (18%) | |
| Non-Adherence Enrollment Criteria | ||||||
| Lost to follow-up | 93 (21%) | 33 (22%) | 37 (24%) | 70 (23%) | ||
| Poor virologic response | 294 (65%) | 99 (65%) | 95 (62%) | 194 (63%) | ||
| Both | 66 (15%) | 20 (13%) | 22 (14%) | 42 (14%) | ||
| Duration of ARV use prior to study entry (years) | Median (Q1, Q3) | 7 (3, 13) | 9 (3, 15) | 7 (4,13) | 8 (3, 14) | |
| Baseline HIV-1 RNA, copies/mL, n (%) | ||||||
| ≤200 | 154 (34%) | 126 (84%) | 142 (93%) | 268 (89%) | ||
| 201-10,000 | 114 (25%) | 16 (11%) | 9 (6%) | 25 (8%) | ||
| >10,000 | 185 (41%) | 8 (5%) | 1 (1%) | 9 (3%) | ||
| Baseline CD4+ T cells/mm3 | Median (Q1, Q3) | 273 (119, 511) | 417 (198, 703) | 379 (202, 611) | 389 (199, 638) | |
| Time since HIV Diagnosis, years | Median (Q1, Q3) | 13 (7, 21) | 13 (7, 22) | 12 (7, 19) | 13 (7, 20) | |
| Baseline BMI, kg/m2 | Median (Q1, Q3) | 26 (22, 30) | 27 (23, 31) | 26 (23, 31) | 26 (23, 31) | |
| Annual Household Income | ||||||
| <$5000 | 128/439 (29%) | 42/149 (28%) | 47/151 (31%) | 89/300 (30%) | ||
| $5000-9999 | 59/439 (13%) | 24/149 (16%) | 22/151 (15%) | 46/300 (15%) | ||
| $10000-19999 | 99/439 (23%) | 32/149 (21%) | 31/151 (21%) | 63/300 (21%) | ||
| >$20,000 | 155/439 (34%) | 51/149 (33%) | 51/151 (33%) | 102/300 (33%) | ||
| Housing Status | ||||||
| In own house/apartment | 270/444 (61%) | 101/150 (67%) | 99/152 (65%) | 200/302 (66%) | ||
| Parents’ House | 64/444 (14%) | 15/150 (10%) | 21/152 (14%) | 36/302 (12%) | ||
| Someone Else’s Housing | 57/444 (13%) | 25/150 (17%) | 16/152 (11%) | 41/302 (14%) | ||
| Unstable/Homeless | 42/444 (9%) | 7/150 (5%) | 14/152 (9%) | 21/302 (7%) | ||
| Other | 11/444 (2%) | 2/150 (1%) | 2/152 (1%) | 4/302 (1%) | ||
| Alcohol Use AUDIT-C Score | ||||||
| Hazardous Drinking (≥4 males or ≥3 females), n (%) | 144/452 (32%) | 44/152 (29%) | 53/154 (34%) | 97/306 (32%) | ||
| Urine Drug Screen Positivity anytime during Step, n (%) | ||||||
| Cocaine | 98/405 (24%) | 41/148 (28%) | 47/147 (32%) | 88/295 (30%) | ||
| Amphetamine/Methamphetamine | 113/405 (28%) | 46/148 (31%) | 42/147 (29%) | 88/295 (30%) | ||
| Opiate/Methadone | 45/400 (11%) | 26/147(18%) | 13/147 (9%) | 39/294 (13%) | ||
| Marijuana | 239/405 (59%) | 98/147 (67%) | 86/147 (59%) | 184/294 (63%) | ||
| Benzodiazepine/Barbiturate/PCP | 45/405 (11%) | 19/147 (13%) | 16/147 (11%) | 35/294 (12%) | ||
| ≥2 drugs | 186/405 (41%) | 75/148 (49%) | 71/147 (46%) | 146/295 (48%) | ||
LAI Long-Acting Injectable; SOC Standard of Care; BMI Body Mass Index; AUDIT-C Alcohol Use Disorders Identification Test-C; PCP phencyclidine
Of the 453 participants enrolled in Step 1, 306 (68%) successfully completed Step 1 and enrolled to Step 2; 54 (12%) did not meet virologic criteria for randomization, 66 (15%) discontinued Step 1 prematurely, and 27 (6%) remained on Step 1 as of February 12, 2024. Of the 306 eligible participants enrolled into Step 2, 152 were randomized to LAI and 154 to SOC (Figure 1C). Three participants (one on LAI and two on SOC) did not initiate randomized treatment. Baseline characteristics were comparable between those who entered Step 2 and those who did not except for CD4+ T-cell count and HIV-1 RNA (Supplementary Appendix). Participants who did not enroll into Step 2 had lower CD4+ T-cells (median 206 vs. 300 cells/mm3) and higher HIV-1 RNA at enrollment (4.23 vs. 2.91 log10 copies/mL). While participants were required to have HIV-1 RNA ≤200 copies/mL at or after Week 4 of Step 1 to be eligible for randomization in Step 2, 24 (16%) LAI and 10 (7%) SOC participants had a HIV-1 RNA >200 copies/mL at the Step 2 randomization visit itself; eight participants (5%) on LAI had HIV-1 RNA >10,000 copies/mL.
The median duration of follow-up was 48 weeks (interquartile range: 20, 52). Twenty-five (16%) participants on LAI and 35 (23%) on SOC discontinued randomized treatment prematurely. Twenty participants (13%) on LAI and 24 (16%) on SOC prematurely discontinued study follow-up. Among the 152 eligible participants randomized to LAI, 11 did not initiate injections (6 completed oral lead-in, 2 prematurely discontinued oral lead-in, 2 remained on oral lead-in at data cut-off, and one never initiated oral lead-in). The remaining 141 participants received a total of 1359 injections on Step 2. Ninety-four percent of post-loading dose injections were administered on time. Fifteen (11%) participants had at least one missed injection. Four LAI participants received bridging with oral ART due to anticipated delay in injection administration.
Outcomes
The primary outcome of regimen failure was observed in 29 (19%) participants on LAI and 55 (36%) participants on SOC (Figure 2A). Among LAI regimen failures, 5 had virologic failure as the first event and 24 had permanent treatment discontinuation as the first event. Among SOC failures, 32 had virologic failure as the first event and 23 had permanent treatment discontinuation as the first event. The difference in the Week 48 cumulative probability of regimen failure (LAI - SOC) was −18.4% (98.4% CI: −32.4%, −4.3%, p=0.002) demonstrating superiority of the LAI regimen (Figures 2A/2B). Results of subgroup analyses and supportive analyses are shown in supplemental tables.
FIGURE 2. PRIMARY AND SECONDARY OUTCOMES.
Figure 2A Primary Outcome and Key secondary outcomes
* The widths of the confidence intervals were not adjusted for multiplicity and should not be used to infer definitive treatment effects for secondary outcomes.
Figure 2B Time to Event for 1) Cumulative Probability of Regimen Failure and 2) Virologic Failure in Step 2 by Treatment Arm
*The widths of the confidence intervals were not adjusted for multiplicity and should not be used to infer de nitive treatment e ects for secondary outcomes.
CI Confidence Interval; LAI Long-acting injectable cabotegravir+rilpivirine; SOC standard of care; VF virologic failure; TRT-DISC Treatment discontinuation; AE Adverse events
Six LAI and 34 SOC participants had virologic failure. The treatment difference in the Week 48 cumulative probability of virologic failure was −21.4% (98.4% CI: −33.5%, −9.3%). Nine LAI and 34 SOC participants had treatment-related failure. The difference in the Week 48 cumulative probability of treatment-related failure was −19.2% (98.4% CI: −31.6%, −6.9%). Twenty-six LAI and 37 SOC participants had permanent treatment discontinuation. The treatment difference in the Week 48 cumulative probability of permanent treatment discontinuation was −8.4% (98.4% CI: −21.3%, 4.5%).
A total of 60 participants discontinued treatment in Step 2 (25 LAI, 35 SOC). The most common reasons for treatment discontinuation on SOC were loss to follow-up/participant withdrawal (n=20, 57%) and non-compliance with study drug (n=6, 17%), whereas for LAI the reasons were more varied, including: loss to follow-up/participant withdrawal (n=9, 36%), AE (n=4, 16%), death not related to study drug (n=2, 8%), and incarceration (n=2, 8%). Amongst participants who experienced virologic failure and had resistance data available, 2 of 5 (40%) in the LAI arm acquired INSTI resistance-associated mutations (RAMs); 2 of 22 (9%) in the SOC arm acquired RAMs (Figure 3).
FIGURE 3.
Characteristics of Participants Experiencing Confirmation of Virologic Failure After Initiating Step 2a
CAB cabotegravir; RPV rilpivirine; INI integrase inhibitor ; PI protease inhibitor; RT reverse transcriptase ; VF virologic failure; BLQ below limit of quantification ; LLOQ lower limit of quantification; PA-IC90 Protein-adjusted concentration at which 90% inhibition of viral replication is achieved ; BMI Body Mass Index
aOne CVF participant was randomized but never received LAI and is not included in the figure
Virologic Failures on LAI
Six participants randomized to LAI experienced virologic failure. One of these initiated oral lead-in but discontinued treatment prior to initiating injections (meeting primary outcome), and was followed in the study with subsequent virologic failure. Of the remaining 5 participants, none had a detectable HIV-1 RNA at LAI initiation, nor any missed injections (Figure 3). Three had delayed injections at least once, though all were delayed by <5 days. While guidelines suggest those with BMI >30 kg/m2 should receive injections with 2” needle lengths,10 the 4 virologic failure participants with BMI >30 kg/m2 received injections with needle lengths of ≤1.6”. All except one virologic failure occurred after week 24 on Step 2. All 4 virologic failure s with successful sub-typing of virus were clade B. All participants with virologic failure subsequently suppressed on oral ART except 1 participant who was lost to follow-up (no RAMs were identified at their virologic failure confirmation visit).
Plasma concentrations of CAB and RPV were all above the lowest quartile at the time of initial and confirmed virologic failure, consistent with previously reported adequate exposure (Figure 3).17,18 However, each participant demonstrated trends in trough concentrations below the lowest quartile for at least one of the agents early after initiating LAI, some of which were also accompanied by a single detectable viral load that re-suppressed by the subsequent visit.
Safety
Fifty-two (34%) LAI and 46 (30%) SOC participants had ≥1 AE during Step 2 (Table 2). The difference in the week 48 cumulative probability of AE was 1.1% (95% CI: −12.7%, 15.0%). Twenty-one (14%) LAI and 16 (10%) SOC participants had ≥1 serious AE reported during Step 2 with the difference in the week 48 cumulative probability of 4.8% (95% CI: −5.6%, 15.2%). Four LAI participants discontinued treatment due to AE, two due to low-grade injection site pain or pain/nodule, one due to seizure, and one due to cerebrovascular accident/hemiparesis requiring anti-coagulation that was deemed by the site investigator as not related to study treatment. Only 1 SOC participant discontinued treatment due to AE (ischemic cerebral infarction/encephalopathy). Two deaths (both on the LAI arm, 1 due to fentanyl overdose at week 43, 1 due to coronary artery disease at week 44) were reported on Step 2, and both were determined by the site investigators to be not related to study treatment.
Table 2.
Step 2 Adverse Events, including Injection Site Reactions
| All Adverse Events | Adverse Events Investigator Assessed as Related to Trial Regimen ≥Grade 3 or Led to treatment discontinuationa |
||||
|---|---|---|---|---|---|
| n (%) | LAI | SOC | LAI | SOC | |
| (N=152) | (N=154) | (N=152) | (N=154) | ||
| Any event | 52 (34%) | 46 (30%) | 8 | 1 | |
| Any event, excluding injection site-reactions | 50 (33%) | 46 (30%) | 3 | 1 | |
| Grade 3 or 4 events | 45 (30%) | 35 (23%) | 6 | 1 | |
| Grade 3 or 4 events, excluding injection-site reactions | 43 (28%) | 35 (23%) | 3 | 1 | |
| Events leading to withdrawal | 4 (3%) | 1 (<1%) | 3 | 1 | |
| Any serious adverse events | 21 (14%) | 16 (10%) | 0 | 0 | |
| Fatal serious adverse events | 2 (1%) | 0 | 0 | 0 | |
| Any injection-site reaction (n=141)b | 84 (60%) | ||||
| Any injection site pain | 5 | 5 | n/a | ||
| Grade 3 injection-site pain | 3 | 3 | n/a | ||
| Any injection-site nodule | 2 | 2 | n/a | ||
| Any injection-site swelling | 2 | 2 | n/a | ||
| Injection-site pain or nodule leading to withdrawala | 2 | 2 | n/a | ||
| Injection-site reaction duration, days median (Q1, Q3) | 3 (2, 6) | - | n/a | ||
| Events, excluding injection site-reactions reported in ≥ 5 % of participants in either group | |||||
| COVID-19 | 4 (3%) | 9 (6%) | - | - | |
| Creatine clearance decreased | 3 (2%) | 8 (5%) | - | - | |
Injection site reactions leading to withdrawal were Grade 1 pain and nodule (n=1) and Grade 2 pain (n=1)
Of those randomized to LAI, 11 participants did not initiate injections at time of data analysis. Among 11 participants did not initiate injection: 1 never initiated oral lead-in (OLI), 2 prematurely discontinued OLI, 2 still on OLI as of Feb 12, 2024, 6 completed OLI but never initiated LAI.
Injection Site Reactions
Among the 141 participants who initiated LAI, 84 (60%) had at least one injection site reaction, with the most frequently reported injection site reactions being pain, swelling, tenderness, or nodule (Table 2). Three participants had ≥1 Grade 3 injection site reaction (pain), and only 2 participants discontinued treatment due to injection site reaction (Grade 1 and Grade 2).
Pregnancy
Four pregnancies occurred on study, with 2 pregnancies in each Step. Both Step 2 pregnancies were in the LAI arm, one of whom experienced virologic failure and developed RAMs (Participant A, Figure 3). All 4 pregnancies resulted in live births, with no congenital abnormalities reported.
Discussion
The LATITUDE trial established the superiority of monthly LAI CAB/RPV over oral ART to reduce regimen failure in PWH who face adherence barriers. Results appeared similar within subgroups in a vulnerable population that is frequently excluded from clinical trials. While a growing body of observational data reporting benefits of this approach have recently accrued, this study provides randomized clinical trial data indicating efficacy of LAI ART as a treatment strategy to reduce regimen failure in PWH for whom oral ART has not proven effective.
This trial was launched when HIV treatment guidelines limited LAI to PWH with sustained viral suppression on oral ART. Thus, to foster viral suppression in the viremic population enrolled, the trial provided comprehensive adherence support and financial incentives contingent on viral load reduction during Step 1. Adherence support to individuals in this study reflected routine clinical practice and provision of financial incentives tied to virologic benchmarks in a viremic population has implementation precedence.19,20 However, the utility of alternative strategies to best initiate individuals with viremia on LAI regimens will benefit from further systematic research. To date, “direct-to-inject” approaches show promise in clinical cohorts7-9 , and our study demonstrated no virologic failure or treatment discontinuation for 23 of the 24 participants who initiated LAI without viral suppression. An ongoing study is further evaluating the use of LAI administered every other month versus oral SOC in viremic PWH (NCT06694805).
The main risk factors present in the five LAI participants experiencing virologic failure were higher BMI and lower CAB and RPV concentrations earlier in treatment, which have been identified in other studies17,18. Additionally needle lengths <2” were used in these participants. This inverse relationship between BMI and trough concentrations is known21, and thus longer needle lengths of 2” are typically recommended to ensure adequate drug exposure. While comparative thresholds to historical PK data were used in our analysis,18,22 a threshold clearly associated with virologic breakthrough has not been identified. Notably none of the participants with virologic failure had missed injections, and three had short delays ≤ 5 days in LAI administration at a maximum of 2 time points.
Our study is strengthened by the broad study population of individuals in whom viral suppression has not been achieved due to mental health comorbidities, substance and alcohol use, and unstable housing, among other causes, representative of the out-of-care population in the United States23 (Supplementary Appendix). This is reflected in the finding of almost 40% of those randomized to SOC had regimen failure during the study. This emphasizes the potential public health contributions of LAI to ending the HIV epidemic. Protocol modifications were made to address slow accrual, lower than expected transition from Step 1 into Step 2, and to overcome barriers imposed by the COVID-19 public health emergency. While these factors could have introduced heterogeneity in the study population, the results from the subgroup and stratified analyses by protocol version indicated no substantial variability. Similarly, while the use of conditional economic incentives could raise concerns about the generalizability of this strategy, the financial support was time-limited and was not included in Step 2, thus allowing for a direct comparison between LAI and oral SOC.
Our study results showed that LAI has superior efficacy over SOC in reducing regimen failure in PWH who face adherence barriers. This strategy provides a key intervention for both averting disease progression and preventing HIV transmission. Wide implementation of this treatment strategy in the people who need it most will require a multidisciplinary approach to maximize its benefit and reach.
Supplementary Material
Acknowledgements:
The study team would like to thank all the participants who enrolled in this study, the Advancing Clinical Therapeutics Globally Network (formerly the AIDS Clinical Trials Group), the Statistical Data and Management Center, Specialty Laboratories, and members of the NIAID Division of AIDS Therapeutic & Prevention Data Safety Monitoring Board. We also thank ViiV and Johnson & Johnson for providing the oral ART and LAI products for the study. We thank Dr. Susan Cohn and Dr. Thomas Campbell for serving as independent adjudicators on this protocol. Dr. Jose Castillo-Mancilla served as study co-chair until February 2023. Pablo F Belaunzaran-Zamudio is a contractor for DAIDS and the current Medical Officer for the study; Karin Klingman, Lawrence Fox, and Tia Morton were NIAID employees during the conduction of the study.
Funding.
This work was supported by the National Institute of Allergy and Infectious Diseases (NIAID) of the National Institutes of Health under Award Number [UM1 AI068634, UM1 AI068636 and UM1 AI106701], with support from the National Institute of Mental Health (NIMH), the National Institute on Drug Abuse (NIDA), and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The study is registered with clinicaltrials.gov (NCT03635788).
References
- 1.Cohen MS, Chen YQ, McCauley M, et al. Antiretroviral Therapy for the Prevention of HIV-1 Transmission. N Engl J Med 2016;375(9):830–9. DOI: 10.1056/NEJMoa1600693. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.CDC. Centers for Disease Control and Prevention: Monitoring Selected National HIV Prevention and Care Objectives by Using HIV Surveillance Data United States and 6 Territories and Freely Associated States, 2023. 2025. [Google Scholar]
- 3.Fauci AS, Redfield RR, Sigounas G, Weahkee MD, Giroir BP. Ending the HIV Epidemic: A Plan for the United States. JAMA 2019. DOI: 10.1001/jama.2019.1343. [DOI] [Google Scholar]
- 4.Orkin C, Arasteh K, Górgolas Hernández-Mora M, et al. Long-Acting Cabotegravir and Rilpivirine after Oral Induction for HIV-1 Infection. New England Journal of Medicine 2020;382(12):1124–1135. DOI: 10.1056/NEJMoa1909512. [DOI] [PubMed] [Google Scholar]
- 5.Swindells S, Andrade-Villanueva JF, Richmond GJ, et al. Long-Acting Cabotegravir and Rilpivirine for Maintenance of HIV-1 Suppression. N Engl J Med 2020;382(12):1112–1123. DOI: 10.1056/NEJMoa1904398. [DOI] [PubMed] [Google Scholar]
- 6.Overton ET, Richmond G, Rizzardini G, et al. Long-acting cabotegravir and rilpivirine dosed every 2 months in adults with HIV-1 infection (ATLAS-2M), 48-week results: a randomised, multicentre, open-label, phase 3b, non-inferiority study. Lancet 2021;396(10267):1994–2005. DOI: 10.1016/S0140-6736(20)32666-0. [DOI] [PubMed] [Google Scholar]
- 7.Brock JB, Herrington P, Hickman M, Hickman A. Long-acting Injectable Cabotegravir/Rilpivirine Effective in a Small Patient Cohort With Virologic Failure on Oral Antiretroviral Therapy. Clin Infect Dis 2023. DOI: 10.1093/cid/ciad511. [DOI] [Google Scholar]
- 8.Hastie E, Hill L, Bamford L, Karim A, Martin TCS. Long-acting injectable HIV therapy outcomes among persons with HIV who have adherence challenges to oral antiretroviral therapy. Clin Infect Dis 2025. DOI: 10.1093/cid/ciaf120. [DOI] [Google Scholar]
- 9.Spinelli MA, Heise MJ, Gistand N, et al. HIV Viral Suppression With Use of Long-Acting Antiretroviral Therapy in People With and Without Initial Viremia. JAMA 2025;333(16):1451–1453. DOI: 10.1001/jama.2025.0109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Panel on Antiretroviral Guidelines for Adults and Adolescents. Guidelines for the Use of Antiretroviral Agents in Adults and Adolescents with HIV. Department of Health and Human Services. Year. Available at https://clinicalinfo.hiv.gov/en/guidelines/adult-and-adolescent-arv. Accessed on October 4, 2024. [Google Scholar]
- 11.Sax PE, Thompson MA, Saag MS. Updated treatment recommendation on use of cabotegravir and rilpivirine for people with HIV from the IAS-USA guidelines panel. JAMA 2024. [Google Scholar]
- 12.Galarraga O, Sosa-Rubi SG. Conditional economic incentives to improve HIV prevention and treatment in low-income and middle-income countries. Lancet HIV 2019;6(10):e705–e714. DOI: 10.1016/S2352-3018(19)30233-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Silverman K, Holtyn AF, Rodewald AM, et al. Incentives for Viral Suppression in People Living with HIV: A Randomized Clinical Trial. AIDS Behav 2019;23(9):2337–2346. DOI: 10.1007/s10461-019-02592-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Jennison C, Turnbull BW. Group sequential methods with applications to clinical trials. Boca Raton: Chapman & Hall/CRC, 2000. [Google Scholar]
- 15.Gant ZM, Morales J, Dailey A, et al. Social Determinants of Health among Adults with Diagnosed HIV Infection in the United States and Puerto Rico, 2020. CDC HIV surveillance report Supplemental report; ; v 28, no 2 2023. (https://stacks.cdc.gov/view/cdc/130299). [Google Scholar]
- 16.Menza TW, Hixson LK, Lipira L, Drach L. Social Determinants of Health and Care Outcomes Among People With HIV in the United States. Open Forum Infect Dis 2021;8(7):ofab330. DOI: 10.1093/ofid/ofab330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Orkin C, Schapiro JM, Perno CF, et al. Expanded Multivariable Models to Assist Patient Selection for Long-Acting Cabotegravir + Rilpivirine Treatment: Clinical Utility of a Combination of Patient, Drug Concentration, and Viral Factors Associated With Virologic Failure. Clin Infect Dis 2023;77(10):1423–1431. (In eng). DOI: 10.1093/cid/ciad370. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Cutrell AG, Schapiro JM, Perno CF, et al. Exploring predictors of HIV-1 virologic failure to long-acting cabotegravir and rilpivirine: a multivariable analysis. AIDS 2021;35(9):1333–1342. DOI: 10.1097/QAD.0000000000002883. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Dombrowski JC, Ramchandani MS, Golden MR. Implementation of Low-Barrier Human Immunodeficiency Virus Care: Lessons Learned From the Max Clinic in Seattle. Clin Infect Dis 2023;77(2):252–257. DOI: 10.1093/cid/ciad202. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Gambone GF, Feldman MB, Thomas-Ferraioli AY, Shubert V, Ghose T. Integrating Financial Incentives for Viral Load Suppression into HIV Care Coordination Programs: Considerations for Development and Implementation. J Public Health Manag Pract 2020;26(5):471–480. DOI: 10.1097/PHH.0000000000001020. [DOI] [PubMed] [Google Scholar]
- 21.Han K, D’Amico RD, Spreen WR, Ford SL. Population pharmacokinetics of cabotegravir following intramuscular thigh injections in adults with and without HIV. Antimicrob Agents Chemother 2024;68(12):e0088024. DOI: 10.1128/aac.00880-24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Thoueille P, Cavassini M, Guidi M, et al. Guidance for the Interpretation of Long-Acting Cabotegravir and Rilpivirine Concentrations Based on Real-World Therapeutic Drug Monitoring Data and Documented Failures. Open Forum Infect Dis 2024;11(2):ofae023. DOI: 10.1093/ofid/ofae023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Dawson LKJ What Do We Know About People with HIV Who Are Not Engaged In Regular HIV Care? 2023(August 26, 2025) (https://www.kff.org/hiv-aids/what-do-we-know-about-people-with-hiv-who-are-not-engaged-in-regular-hiv-care/).
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.