| |
Long-Acting Cabotegravir and Rilpivirine after Oral Induction for HIV-1 Infection FLAIR
|
| |
| |
Download the PDF here
Download the PDF here
NEJM March 19 2020 - C. Orkin, K. Arasteh, M.G. Hern®¢ndez‑Mora, V. Pokrovsky, E.T. Overton, P.-M. Girard, S. Oka, S. Walmsley, C. Bettacchi, C. Brinson, P. Philibert,J. Lombaard, M. St. Clair, H. Crauwels, S.L. Ford, P. Patel, V. Chounta,
R. D’Amico, S. Vanveggel, D. Dorey, A. Cutrell, S. Griffith, D.A. Margolis, P.E. Williams, W. Parys, K.Y. Smith, and W.R. Spreen
LONG-ACTING CABOTEGRAVIR + RILPIVIRINE FOR HIV TREATMENT: FLAIR WEEK 96 RESULTS - (03/11/20)
Abstract
Background
Long-acting injectable regimens may simplify therapy for patients with human immunodeficiency virus type 1 (HIV-1) infection.
Methods
We conducted a phase 3, randomized, open-label trial in which adults with HIV-1 infection who had not previously received antiretroviral therapy were given 20 weeks of daily oral induction therapy with dolutegravir-abacavir-lamivudine. Participants who had an HIV-1 RNA level of less than 50 copies per milliliter after 16 weeks were randomly assigned (1:1) to continue the current oral therapy or switch to oral cabotegravir plus rilpivirine for 1 month followed by monthly injections of long-acting cabotegravir plus rilpivirine. The primary end point was the percentage of participants who had an HIV-1 RNA level of 50 copies per milliliter or higher at week 48 (Food and Drug Administration snapshot algorithm).
Results
At week 48, an HIV-1 RNA level of 50 copies per milliliter or higher was found in 6 of 283 participants (2.1%) who received long-acting therapy and in 7 of 283 (2.5%) who received oral therapy (adjusted difference, -0.4 percentage points; 95% confidence interval [CI], -2.8 to 2.1), a result that met the criterion for noninferiority for the primary end point (margin, 6 percentage points). An HIV-1 RNA level of less than 50 copies per milliliter at week 48 was found in 93.6% who received long-acting therapy and in 93.3% who received oral therapy (adjusted difference, 0.4 percentage points; 95% CI, -3.7 to 4.5), a result that met the criterion for noninferiority for this end point (margin, -10 percentage points). Of the participants who received long-acting therapy, 86% reported injection-site reactions (median duration, 3 days; mild or moderate severity, 99% of cases); 4 participants withdrew from the trial for injection-related reasons. Grade 3 or higher adverse events and events that met liver-related stopping criteria occurred in 11% and 2%, respectively, who received long-acting therapy and in 4% and 1% who received oral therapy. Treatment satisfaction increased after participants switched to long-acting therapy; 91% preferred long-acting therapy at week 48.
Conclusions
Therapy with long-acting cabotegravir plus rilpivirine was noninferior to oral therapy with dolutegravir-abacavir-lamivudine with regard to maintaining HIV-1 suppression. Injection-site reactions were common. (Funded by ViiV Healthcare and Janssen; FLAIR ClinicalTrials.gov number, NCT02938520. opens in new tab.)
Introduction
Combination antiretroviral therapy can significantly prolong the life expectancy of people living with human immunodeficiency virus (HIV), but HIV infection remains a chronic condition that requires lifelong daily oral treatment.1,2 Current antiretroviral regimens are highly effective, and one focus of ongoing drug development is improvement of the side-effect profile and convenience to reduce disengagement from care. Prolonged daily regimens can engender dissatisfaction, contribute to stigma, and increase the risk of nonadherence to treatment and treatment failure.3,4 Surveys indicate that many patients would prefer therapeutic alternatives.5,6 Two-drug regimens have been developed as an option7,8; long-acting injectable regimens are another alternative that can free patients from the burden of daily regimens and potentially provide a more acceptable therapeutic approach.
Long-acting injectable formulations are being developed for cabotegravir, which is an integrase strand-transfer inhibitor (INSTI), and rilpivirine, which is a nonnucleoside reverse-transcriptase inhibitor (NNRTI) with an approved oral formulation.9-11 In the Long-Acting Antiretroviral Treatment Enabling Trial 2 (LATTE-2), participants who had not previously received antiretroviral therapy were given oral cabotegravir-based treatment; those who had viral suppression were randomly assigned to continue oral treatment or switch to monthly intramuscular injections of long-acting cabotegravir plus rilpivirine. Viral suppression was maintained through week 96 in 87% of the participants who switched to monthly long-acting therapy, as compared with 84% of the participants who continued oral therapy12; viral suppression was maintained through week 160 in 83% of the recipients of long-acting therapy.13
The First Long-Acting Injectable Regimen (FLAIR) trial evaluated a potential pathway to long-acting injectable therapy for patients who had not previously received antiretroviral therapy. We assessed whether switching to monthly injections of long-acting cabotegravir plus rilpivirine would be noninferior to continuing oral therapy in patients with HIV type 1 (HIV-1) who had viral suppression in response to oral induction therapy.
Results
Participants
A total of 809 adults were screened at 108 sites in 11 countries beginning on October 27, 2016 (Figure 1B); the last participant completed week 48 on August 30, 2018. Oral induction therapy was initiated in 629 participants; 63 of those participants withdrew from the trial before randomization, primarily because of a lack of efficacy, and the remaining 566 were randomly assigned to treatment in the maintenance phase (283 to each treatment group). During the maintenance phase, 25 participants (9%) in the long-acting-therapy group and 22 participants (8%) in the oral-therapy group withdrew from the trial; withdrawals were most frequently due to adverse events in the long-acting-therapy group (in 9 participants) and participant decision to withdraw in the oral-therapy group (in 7). In the long-acting-therapy group, 98% of the 3577 expected injection visits (12 per participant by week 48, with additional visits beyond week 48) occurred within a window of 21 to 35 days after the previous injection; four of five missed injections were covered with oral bridging therapy (Fig. S1). In the oral-therapy group, adherence to treatment was more than 90% on the basis of patient-reported treatment interruptions of 3 or more consecutive days.
Across the two treatment groups, participants were a median of 34 years of age, 22% were female, and 74% were white; approximately 20% had an HIV-1 RNA level of 100,000 copies per milliliter or higher at baseline (Table 1). Before the induction phase, 69% of the participants had a CD4+ lymphocyte count of 350 per microliter or higher, and this percentage increased to 90% by the start of maintenance therapy.
Efficacy
At week 48, an HIV-1 RNA level of 50 copies per milliliter or higher was found in 6 participants (2.1%) who received long-acting therapy and in 7 participants (2.5%) who received oral therapy (adjusted difference, -0.4 percentage points; 95% confidence interval [CI], -2.8 to 2.1), a result that met the prespecified noninferiority criterion for the primary end point (Table 2). For this end point, there was no meaningful estimated difference between treatments across subgroups (Fig. S2). Similarly, long-acting therapy was noninferior to oral therapy with regard to the key secondary end point of the percentage of participants with an HIV-1 RNA level of less than 50 copies per milliliter at week 48 (93.6% and 93.3%, respectively; adjusted difference, 0.4 percentage points; 95% CI, -3.7 to 4.5). On tests for evidence against homogeneity in the differences between treatments across randomization strata, results were not significant. All these results were similar in the per-protocol population (Table 2).
In the long-acting-therapy group, 4 participants had confirmed virologic failure. In 1 of those participants (who had HIV-1 subtype AG), oral lead-in therapy was suspended owing to a false-positive pregnancy test; on reinitiation of oral therapy, criteria for confirmed virologic failure were met and the participant was withdrawn from the trial before long-acting therapy was initiated, with no resistance mutations detected. The other 3 participants had NNRTI and INSTI resistance mutations (Table S1) that developed during long-acting therapy; these mutations reduced susceptibility to rilpivirine in 2 participants by a factor of more than 2 and reduced susceptibility to cabotegravir in all 3 participants by a factor of more than 5. These 3 participants had HIV-1 subtype A1 with the L74I integrase polymorphism at baseline. However, 51 of the 54 participants in the long-acting-therapy group who had HIV-1 with the L74I integrase polymorphism at baseline did not have virologic failure (Table S2). In subgroup analyses of the primary end point, no significant difference between treatments was observed in subgroups defined according to the presence or absence of the L74I integrase polymorphism or according to HIV-1 subtype (Fig. S2). In the oral-therapy group, 3 participants had confirmed virologic failure without the development of resistance mutations or phenotypic changes during treatment. In the primary analysis, 2 participants who received long-acting therapy and 4 participants who received oral therapy had an HIV-1 RNA level of 50 copies per milliliter or higher without confirmed virologic failure; neither of the participants who received long-acting therapy had HIV-1 resistance mutations.
Safety and Side Effects
During the maintenance phase, the most common adverse events in the long-acting-therapy group, excluding injection-site reactions, were nasopharyngitis, headache, upper respiratory tract infection, and diarrhea (Table 3 and Table S3). Serious adverse events occurred in 18 participants (6%) who received long-acting therapy and in 12 participants (4%) who received oral therapy (Table S4); each type of event occurred in 1 participant (except hepatitis A, which occurred in 3 participants who received long-acting therapy), with no deaths. Adverse events that led to withdrawal from the trial occurred in 9 participants (3%) who received long-acting therapy and in 4 participants (1%) who received oral therapy. In the long-acting-therapy group, the only events that led to withdrawal in more than 1 participant were viral hepatitis and injection-site pain (in 5 and 2 participants, respectively) (Table S5). During the maintenance phase, events that met liver-related stopping criteria occurred in 7 participants (2%) who received long-acting therapy and in 2 participants (1%) who received oral therapy, including eight cases of acute viral hepatitis and one case of inorganic solvent toxicity. One participant in the oral-therapy group was pregnant and had a healthy baby. At week 48, the median weight gain from baseline was 1.3 kg (interquartile range, -1.0 to 5.0) in the long-acting-therapy group and 1.5 kg (interquartile range, -1.0 to 3.9) in the oral-therapy group.
Of the participants who received long-acting therapy, 86% had at least one injection-site reaction. The most common injection-site reaction was pain, which was reported by 227 of the 278 participants (82%) who received at least one injection (Table S3). Most of the 1879 pain events were of mild or moderate severity (86% and 13%, respectively); 12 events (<1%) in 11 participants were severe (grade 3), and there were no grade 4 events. The incidence of injection-site reactions was highest (71%) after the initial 3-ml injections at week 4 and subsequently decreased to 20% at week 48 (Fig. S3). The median duration of injection-site reactions was 3 days; 88% of cases resolved within 7 days. Injection-site reactions led to withdrawal from the trial in 2 participants, and another 2 participants withdrew consent for other injection-related reasons. Other adverse events that were considered by investigators to be drug-related were more common with long-acting therapy than with oral therapy (occurring in 28% vs. 10% of the participants). There was no discernible pattern to these events; other than injection-site reactions, the most common drug-related events (occurring in ≥5% of the participants) in the long-acting-therapy group were headache and pyrexia.
Pharmacokinetics
Plasma concentrations of cabotegravir and rilpivirine during long-acting therapy (Figure 2) were similar to the concentrations reported during oral therapy.19,20 The geometric mean cabotegravir trough concentration was 1.56 μg per milliliter at week 8 and 3.13 μg per milliliter at week 48; these concentrations are 9.4 times and 18.9 times, respectively, as high as the in vitro protein-adjusted 90% inhibitory concentration (PA-IC90) for the drug. The geometric mean rilpivirine trough concentration was 41.2 ng per milliliter at week 8 and 82.4 ng per milliliter at week 48; these concentrations are 3.4 times and 6.9 times as high as the PA-IC90. The three participants who received long-acting therapy and had confirmed virologic failure had cabotegravir and rilpivirine concentrations in the lowest quartile, with one having drug concentrations below the fifth percentile (Fig. S4); all three had a body-mass index (the weight in kilograms divided by the square of the height in meters) of more than 30 at baseline, but none missed an injection or received injections outside the permitted window.
Patient-Reported Outcomes
At week 48, the HIVTSQc total score for satisfaction with current treatment as compared with induction treatment was higher in the long-acting-therapy group than in the oral-therapy group (adjusted mean difference, 4.1 points; 95% CI, 2.8 to 5.5) (Table S7). An exploratory analysis of a single-item question regarding therapy preference at week 48 indicated that 257 of 283 participants (91%) who received long-acting therapy in the intention-to-treat exposed population and 257 of 259 participants (99%) who responded to the survey preferred the long-acting regimen over the previous oral therapy.
Discussion
The results of this trial show a pathway for patients who have not previously received antiretroviral therapy to reach and maintain HIV-1 suppression with oral induction therapy and a subsequent transition to monthly injectable therapy. After viral suppression was achieved with a standard oral INSTI-based regimen, the simplified injectable regimen, which consisted of long-acting formulations of cabotegravir and rilpivirine, was noninferior to continued oral therapy with regard to maintaining suppression through week 48.
Baseline viral load had no significant effect on the results for the primary end point; this finding suggests that if the viral load is suppressed without evidence of cabotegravir or rilpivirine resistance mutations, a transition to long-acting therapy is feasible. At baseline, HIV-1 subtype A1 was present in 8 participants in the long-acting-therapy group, 5 of whom maintained viral suppression and 3 of whom had virologic failure. All 3 of the participants who had virologic failure during long-acting therapy also had a body-mass index of more than 30, plasma drug levels in the lowest quartile, and HIV-1 with the L74I integrase polymorphism. However, the L74I integrase polymorphism does not confer resistance to cabotegravir by itself: 51 of 54 participants in the long-acting-therapy group who had the L74I integrase polymorphism did not have virologic failure at week 48.21,22 No dosing complications that may have contributed to virologic failure were reported. With virologic failure occurring in only 4 participants in the long-acting-therapy group (and occurring before the initiation of long-acting therapy in 1 of those participants), the potential contributions of virologic, pharmacokinetic, demographic, and other factors to virologic outcomes remain uncertain; future analyses of pooled study data may provide clarification.
The three-drug combination of dolutegravir, abacavir, and lamivudine that was used for induction therapy and as the control regimen during the maintenance phase is a recommended first-line regimen for patients who have not previously received antiretroviral therapy.23 Efficacy results with long-acting therapy in this trial were similar to those seen in the ATLAS trial, which had a similar switch design but enrolled participants who had longer-term viral suppression, which had been achieved with the use of other standard three-drug oral therapies.18 Together, the FLAIR and ATLAS trials show that the long-acting regimen effectively maintained viral suppression that had initially been achieved with the use of oral regimens, both in adults new to treatment and in those who had received prolonged previous treatment.
Other than injection-site reactions, no patterns of adverse events with the long-acting regimen were evident, a finding consistent with the reported safety profiles of oral cabotegravir and rilpivirine.12,20 Injection-site reactions, primarily pain, were common, but the incidence decreased from 71% to 20% during the trial. Injection-site reactions were generally of mild or moderate severity and transient; 4 of the 283 participants in the long-acting-therapy group withdrew from the trial owing to injection-site reactions or other injection-related reasons. The incidences of some adverse events other than injection-site reactions were higher in the long-acting-therapy group than in the oral-therapy group; however, effects associated with starting a new treatment (as opposed to continuing the same treatment) may have contributed to the observed differences - a possibility consistent with observations in previous switch studies.8,24
The long-acting regimen was preferred over the previous oral therapy by 91% of recipients, even after 12 monthly injections. Conclusions derived from this finding are limited to patients who are willing to consider injectable therapy, reflecting the enrolled trial population. Similarly, in clinical practice, the long-acting regimen is a therapeutic option that patients can select according to their preference. For patients who choose long-acting therapy, the data regarding treatment preference suggest that their expectations of regimen benefits will be met.
The potential clinical role of the long-acting regimen remains to be fully defined for the spectrum of patients with HIV-1 infection, particularly those who have adherence challenges, in different practice settings. In this regard, additional randomized efficacy and safety trials of the long-acting regimen include the ATLAS trial,18 which enrolled participants who had previously received antiretroviral therapy, and the LATITUDE trial (ClinicalTrials.gov number, NCT03635788. opens in new tab), which enrolled participants who had adherence difficulties. Ongoing follow-up of the FLAIR and ATLAS trials will assess outcomes of long-acting therapy extended beyond 48 weeks. Several additional ongoing or planned studies (e.g., the Cabotegravir plus Rilpivirine in the U.S. to Optimize and Measure Implementation and Experience [CUSTOMIZE] trial; NCT04001803. opens in new tab) are focused on implementation of the long-acting regimen in various settings, including university hospitals and private and public health clinics.
In conclusion, monthly two-drug long-acting therapy was noninferior to standard three-drug oral therapy with regard to maintaining viral suppression for 48 weeks in adults with HIV-1 infection who had not previously received antiretroviral therapy, with greater reported treatment satisfaction.
|
|
| |
| |
|
|
|
