Efficacy, Safety, and Pharmacokinetics by Body Mass Index Category in Phase 3/3b Long-Acting Cabotegravir Plus Rilpivirine Trials

Abstract Background Cabotegravir plus rilpivirine (CAB + RPV) is a guideline-recommended long-acting (LA) injectable regimen for the maintenance of human immunodeficiency virus-1 (HIV-1) virologic suppression. This post hoc analysis summarizes CAB + RPV LA results by baseline body mass index (BMI) category among phase 3/3b trial participants. Methods Data from CAB + RPV-naive participants receiving every 4 or 8 week dosing in FLAIR, ATLAS, and ATLAS-2M were pooled through week 48. Data beyond week 48 were summarized by study (FLAIR through week 96 and ATLAS-2M through week 152). HIV-1 RNA <50 and ≥50 copies/mL, confirmed virologic failure (CVF; 2 consecutive HIV-1 RNA ≥200 copies/mL), safety and tolerability, and plasma CAB and RPV trough concentrations were evaluated by baseline BMI (<30 kg/m2, lower; ≥30 kg/m2, higher). Results Among 1245 CAB + RPV LA participants, 213 (17%) had a baseline BMI ≥30 kg/m2. At week 48, 92% versus 93% of participants with lower versus higher BMI had HIV-1 RNA <50 copies/mL, respectively. Including data beyond week 48, 18 participants had CVF; those in the higher BMI group (n = 8) all had at least 1 other baseline factor associated with CVF (archived RPV resistance-associated mutations or HIV-1 subtype A6/A1). Safety and pharmacokinetic profiles were comparable between BMI categories. Conclusions CAB + RPV LA was efficacious and well tolerated, regardless of baseline BMI category. Clinical Trials Registration NCT02938520, NCT02951052, and NCT03299049.

Antiretroviral therapy (ART) consists of a combination of ≥2 agents from at least 2 drug classes [1][2][3].Cabotegravir (CAB), an integrase strand transfer inhibitor (INSTI), plus rilpivirine (RPV), a nonnucleoside reverse transcriptase inhibitor (NNRTI), is the first complete long-acting (LA) ART regimen, administered monthly (Q1M) or every 2 months (Q2M) via intramuscular injection, recommended by treatment guidelines for the maintenance of human immunodeficiency virus-1 (HIV-1) virologic suppression [1][2][3].LA injectable regimens have the potential to address some of the psychosocial stressors associated with daily oral ART that are experienced by some people with HIV (PWH) [4].Treatment guidelines recognize this potential, including the US Department of Health and Human Services guidelines, which state that PWH who are interested in nonoral ART options due to concerns surrounding privacy, stigma, or convenience have demonstrated greater satisfaction with CAB + RPV LA than continued oral therapy [1].
The global prevalence of obesity (BMI ≥30 kg/m 2 ) has been rising and is associated with many comorbidities [19].Obesity has the potential to affect drug absorption of medications administered by the intramuscular route, as well as the incidence of inadvertent administration into subcutaneous tissues due to inadequate needle length [19,20].
To examine the potential impact of obesity on treatment outcomes for those receiving CAB + RPV LA, we conducted a post hoc analysis examining the efficacy, safety, and pharmacokinetics (PK) of CAB + RPV LA using pooled data through week 48 from participants in the FLAIR, ATLAS, and ATLAS-2M studies, stratified by baseline BMI category (<30 kg/m 2 , lower BMI group; ≥30 kg/m 2 , higher BMI group).Data beyond week 48 for FLAIR (week 96) and ATLAS-2M (week 152) are also described.

Study Population
Data from participants without prior exposure to CAB + RPV who received CAB + RPV LA every 4 weeks (Q4W) or every 8 weeks (Q8W) dosing within the phase 3/3b FLAIR (NCT02938520), ATLAS (NCT02951052), and ATLAS-2M (NCT03299049) trials through week 48 were pooled in a post hoc analysis.Additionally, data beyond week 48 were summarized separately for FLAIR and ATLAS-2M participants (data beyond week 48 for the ATLAS study were not summarized as most participants transitioned to ATLAS-2M after week 48 during the extension phase of ATLAS).
FLAIR, ATLAS, and ATLAS-2M are randomized, multicenter, parallel-group, open-label, noninferiority phase 3/3b studies evaluating CAB + RPV LA dosed Q4W versus continuing daily oral therapy (FLAIR and ATLAS), or CAB + RPV LA dosed Q8W versus Q4W (ATLAS-2M).The full study designs and eligibility criteria have been published previously [6,7,9].All studies included a screening phase, a maintenance phase, and an extension phase.Enrolled participants were 18 years of age or older and virologically suppressed with a plasma HIV-1 RNA <50 copies/mL prior to randomization without evidence of any major INSTI or NNRTI RAMs (except K103N).Baseline characteristics were generally similar across the 3 studies and across arms [6,7,9].FLAIR participants were ART naive at study entry and underwent an induction phase with daily oral dolutegravir/ abacavir/lamivudine for 20 weeks (participants who were HLA-B*5701-positive received a non-abacavir regimen) to achieve virologic suppression, while ATLAS and ATLAS-2M participants were virologically suppressed (HIV-1 RNA <50 copies/mL) on their current oral regimen at study entry.Most participants who received CAB + RPV LA in the ATLAS study rolled over to the ATLAS-2M study after week 48 (n = 391).ATLAS-2M data from participants who had rolled over from ATLAS with prior exposure to CAB + RPV were excluded to align duration of exposure and PK parameters in the 48-week pooled study population.Beyond week 48, data were summarized separately for FLAIR (week 96) and ATLAS-2M (week 152) as different time points were being evaluated; for ATLAS-2M, all participants (including those who entered the study with prior CAB + RPV exposure) were included in the summary.ATLAS data beyond week 48 were not summarized as most participants had transitioned to ATLAS-2M after week 48 [7,10].
All 3 studies were conducted in accordance with the Declaration of Helsinki [21].All participants provided written informed consent, and the study protocols, any amendments, informed consent, and other information that required preapproval were reviewed and approved by a national, regional, or investigational center ethics committee or institutional review board.

End Points and Assessment
End points assessed by baseline BMI category were the proportion of participants with a plasma HIV-1 RNA <50 copies/mL and ≥50 copies/mL by the Food and Drug Administration (FDA) Snapshot algorithm (weeks 48, 96, and 152), the incidence of CVF (2 consecutive measurements of plasma HIV-1 RNA ≥200 copies/mL; baseline through week 152), adverse events (AEs; baseline through week 152) including injection site reactions (ISRs; week 4 through week 152), and CAB and RPV plasma trough concentrations (C trough ; week 4 through week 96; including analysis of CAB concentrations by needle length for participants with a BMI ≥30 kg/m 2 , week 4 through week 48).Analysis of RPV concentration by needle length was not conducted as previous analyses have indicated that BMI has no impact on RPV LA absorption [22,23].PK plasma samples were analyzed for CAB and RPV concentrations using liquid chromatography with tandem mass spectrometry methods.

Statistical Analysis
Participants were stratified by baseline BMI category (<30 kg/ m 2 , lower BMI group; ≥30 kg/m 2 , higher BMI group) and CAB + RPV LA dosing regimen (Q8W and Q4W).All data presented in this post hoc analysis are descriptive.

Participants
In total, 1245 randomized participants (Q8W, n = 327; Q4W, n = 918) with a median baseline BMI of 25.1 kg/m 2 (range, 15.30-54.02kg/m 2 ) were included in the week 48 analysis; 1032 (83%) had a baseline BMI <30 kg/m 2 and 213 (17%) had a baseline BMI ≥30 kg/m 2 .Overall, 22% (n = 47 of 213) of participants in the higher BMI group had a baseline BMI of 35-<40 kg/m 2 , and 12% (n = 25 of 213) had a baseline BMI of ≥40 kg/m 2 .Median age was comparable across BMI categories and dosing regimens; however, there was a lower proportion of female (sex at birth) participants and black or African American participants in the lower BMI group versus the higher BMI group (Table 1).For data beyond week 48, 283 participants (86%, n = 243 of 283, in the lower BMI group; 14%, n = 40 of 283, in the higher BMI group) from FLAIR and 1045 participants (80%, n = 834 of 1045, in the lower BMI group; 20%, n = 211 of 1045, in the higher BMI group) from ATLAS-2M were included in the analysis.

Efficacy
Viral suppression was high and comparable across BMI categories at week 48 (Table 2).Within the lower BMI group, 94% and 93% of participants in the Q8W and Q4W dosing regimen had HIV-1 RNA <50 copies/mL at week 48, respectively.In the higher BMI group, 92% of participants across both dosing regimens had HIV-1 RNA <50 copies/mL at week 48.Overall, 10 (1%) versus 11 (5%) participants had HIV-1 RNA ≥50 copies/mL in the lower and higher BMI groups, respectively.A higher proportion of participants had no virologic data at week 48 in the lower BMI group versus the higher BMI group, driven by a higher number of discontinuations related to AEs and other reasons.

Confirmed Virologic Failure
Through 48 weeks of CAB + RPV LA therapy, 1% (n = 15 of 1246) of participants met the CVF criterion (Supplementary Table 1).Of these participants, 1 had oral CAB + RPV dosing interrupted due to a false-positive pregnancy test and, upon reinitiating oral therapy, had suspected virologic failure that was confirmed; this participant did not receive a CAB + RPV LA injection.Of the 14 participants who met the CVF criterion and received injections, 6 (<1%) and 8 (4%) occurred in the lower and higher BMI groups, respectively.Within the higher BMI group, all 8 participants with CVF had at least 1 other baseline factor previously associated with increased risk of CVF: RPV RAMs (n = 3), HIV-1 subtype A6/A1 (n = 4), or both (n = 1).Among the 153 participants with BMI ≥30 kg/m 2 as their only baseline factor, none met the CVF criterion.Within the lower BMI group, 3 participants had no factors, 2 had HIV-1 subtype A6/A1, and 1 had both HIV-1 subtype A6/A1 and RPV RAMs.

Safety
AEs Excluding ISRs.The frequency of AEs was broadly comparable across BMI categories and dosing regimens, with 69%-86% of participants reporting an AE.Drug-related AEs were reported in 14%-29% of participants (Table 3).Drug-related grade ≥3 AEs were uncommon, occurring in 0%-2% of participants across both BMI categories and dosing regimens.Overall, 3% (n = 32 of 1032) and <1% (n = 1 of 213) of participants in the lower and higher BMI groups experienced AEs leading to withdrawal, respectively.Serious drug-related AEs occurred in 2 participants (<1%, both of whom were in the lower BMI group): right knee monoarthritis (n = 1, FLAIR) and hypersensitivity due to suspected (partial) intravenous administration of RPV (n = 1, ATLAS-2M).No fatal AEs occurred.

Pharmacokinetics
CAB and RPV C troughs remained above the respective protein-adjusted 90% inhibitory concentrations for both drugs (0.166 µg/mL for CAB and 12 ng/mL for RPV) regardless of baseline BMI category throughout the dosing period (Figure 2).

Efficacy
High levels of virologic suppression were observed at week 96 (FLAIR) and week 152 (ATLAS-2M) and were comparable across BMI categories.At week 96 in FLAIR, 87% (n = 211 of 243) of participants in the lower BMI group and 85% (n = 34 of 40) in the higher BMI group maintained HIV-1 RNA <50 copies/mL (Supplementary Table 3).At week 152 in ATLAS-2M, virologic suppression was maintained in 87% (n = 725 of 834) and 85% (n = 180 of 211) of participants in the lower and higher BMI groups, respectively.

Confirmed Virologic Failure
Of the 18 participants with CVF (10 in the lower BMI group; 8 in the higher BMI group) through week 96 in FLAIR and week 152 in ATLAS-2M combined (excluding ATLAS; including those who entered ATLAS-2M with prior CAB + RPV exposure; Supplementary Table 4), 3 met the criterion after week 48; all 3 participants were in the lower BMI group.Of these 3 participants with CVF after week 48, 2 had at least 1 baseline factor associated with increased risk of CVF (1 participant had RPV RAMs, and the other participant had HIV-1 subtype A6/A1).

Safety
AEs Excluding ISRs.A summary of AEs (excluding ISRs) occurring between week 48 and week 96 in FLAIR and between week 48 and week 152 in ATLAS-2M is shown in Supplementary Table 5.In FLAIR, of those participants with no drug-related AEs in the week 0-48 period, 9% (n = 16 of 177) and 4% (n = 1 of 28) in the lower BMI and the higher BMI groups, respectively, had drug-related AEs (excluding ISRs) by week 96.In total, 4 (2%) participants in the lower BMI group had AEs leading to withdrawal after week 48; no AEs leading to withdrawal were reported for participants in the higher BMI group.There were no drug-related grade ≥3 or drug-related serious AEs (SAEs) reported beyond week 48 in FLAIR (Supplementary Table 5).In ATLAS-2M, of those participants with no drug-related AEs in the week 0-48 period, 10% (n = 63 of 636) and 7% (n = 12 of 175) in the lower and higher BMI groups, respectively, had drug-related AEs by week 152.Fourteen (2%) and 5 participants (2%) in the lower and higher BMI groups, respectively, withdrew due to AEs between week 48 and week 152.Since the week 48 analysis, 3 (<1%) participants in the lower BMI group had drug-related SAEs.
ISR AEs Over Time.The proportion of participants reporting ISRs generally decreased over time regardless of BMI group, with a numerical trend toward fewer ISRs in the higher BMI group (Supplementary Figure 1).

DISCUSSION
The influence of BMI on efficacy and safety outcomes with ART has not been widely studied.Given that CAB + RPV LA is an injectable regimen, BMI is of particular interest, with previous PK analyses indicating that BMI values ≥30 kg/m 2 are associated with initially slower CAB LA absorption [20,24], but with no impact on RPV LA absorption [22].Notably, baseline BMI (per unit increase) was associated with increased CVF risk in PWH receiving CAB + RPV LA in multivariable logistical regression analyses (adjusted incidence rate ratio, 1.09) [17,18].To evaluate the effect of BMI on virologic and safety outcomes with CAB + RPV LA, we pooled data through week 48 from 3 phase 3/3b trials, totaling 1245 PWH.Data beyond week 48 were also summarized to examine longer-term outcomes.
The efficacy results demonstrate that CAB + RPV LA Q4W and Q8W maintained high virologic suppression rates at week 48 (FDA Snapshot algorithm) regardless of baseline BMI category.High virologic suppression was also maintained beyond week 48, through week 96 in FLAIR, and through week 152 in ATLAS-2M.Rates of suppression in participants with a BMI <30 kg/m 2 and ≥30 kg/m 2 were consistent with overall virologic suppression rates observed in the primary analyses of the individual studies [6,7,9].
Through week 48, CVF was more frequent in participants with a BMI ≥30 kg/m 2 (4% vs 1%, BMI <30 kg/m 2 ) when combined with at least 1 other baseline factor; no participant with a BMI ≥30 kg/m 2 as the only baseline factor met the CVF criterion through week 48.Additionally, none of the 3 participants with CVF after week 48 in FLAIR and ATLAS-2M had a BMI ≥30 kg/m 2 .These results are consistent with previous observations from expanded population multivariable analyses of studies with CAB + RPV, in which the incidence of CVF in participants with a BMI ≥30 kg/m 2 as their sole baseline factor (0.5%) was similar to that of the overall population with no baseline factor through to week 152 (0.4%) [17].
The safety profiles were generally comparable between BMI categories through week 48 and up to week 152, with no new safety signals observed.Overall, the incidence of AEs leading to withdrawal was low for both BMI groups throughout week 152, consistent with the overall phase 3/3b population.Injections were well tolerated across both BMI categories, with most ISRs classified as mild to moderate in severity, the ISRs decreasing in incidence over time, and few ISRs leading to study withdrawal.
PK observations were consistent with population PK modeling data for both CAB and RPV [22,23].RPV concentrations were similar regardless of BMI category.Median CAB C troughs tended to be lower in the first 16 weeks of therapy in participants with a baseline BMI ≥30 kg/m 2 compared with the lower BMI group; this trend disappeared with drug accumulation after week 16.Importantly, higher median CAB C trough were observed in the first 16 weeks when the longerlength needles (≥2 inches) were available and used for administration in participants with a BMI ≥30 kg/m 2 .This supports the product label that recommends longer-length needles to accommodate individual body habitus in participants with a BMI ≥30 kg/m 2 and ensures appropriate administration into gluteal muscle [25].When evaluating PK further in the higher BMI subgroup (BMI ≥40 kg/m 2 ), C troughs were also comparable.
This analysis had several limitations.Owing to the difference in dosing frequency, the Q4W arm received more frequent safety assessments than the Q8W arm, which may have increased the overall number of AEs reported in the Q4W arm for both BMI groups.
There were more participants in the lower BMI group versus the higher BMI group; therefore, results should be interpreted with caution in view of the different group sizes.Furthermore, there were differing proportions of men versus women and race/ethnicities between the 2 BMI categories.The generalizability of the data is also limited for those with very high BMI values, given the small number of participants included who had a BMI ≥40 kg/m 2 (<1%).Finally, no formal statistical analyses were performed; therefore, the data are purely descriptive.

e34•
JID 2024:230 (15 July) • Elliot et al The Journal of Infectious Diseases M A J O R A R T I C L E

Table 1 . Baseline Characteristics for Pooled CAB + RPV LA Participants Across the ATLAS, FLAIR, and ATLAS-2M Clinical Trials Through Week 48
ISRs in the lower BMI group (Supplementary Table2).No grade 4 or 5 ISRs occurred.Overall, 3% (n = 34 of 1011) and 1% (n = 3 of 212) of participants in the lower and higher BMI groups had ISRs leading to withdrawal, respectively.
Abbreviations: BMI, body mass index; CAB, cabotegravir; ITT-E, intention-to-treat exposed; LA, long-acting; NA, not applicable; Q4W, every 4 weeks; Q8W, every 8 weeks; RPV, rilpivirine.e36 • JID 2024:230 (15 July) • Elliot et al the lower and higher BMI groups, respectively.The ISR profiles were generally similar between dosing regimens, with slightly fewer pain events in the higher BMI group and fewer grade 3 CAB C troughs were initially slightly lower in participants in the higher BMI group versus those in the lower BMI group (Q8W, 1.14 vs 1.95 µg/mL; Q4W, 1.00 vs 1.70 µg/ mL) at week 8, although by week 16 (1.48vs 1.52 µg/mL) for Q8W dosing, and by week 28 (2.17 vs 2.48 µg/mL) for Q4W dosing, concentrations were similar between BMI groups through week 48.In the higher BMI group, the use of longer needle lengths (≥2 vs <2 inches [≥5.08 vs <5.08 cm]) resulted

Table 2 . Summary of Pooled Study Outcomes at Week 48: Snapshot Analysis (ITT-E Population)
1.48-5.21)and 2.58 µg/mL Summary of ISRs through week 48.AE grade is the maximum grade reported by the participant at each visit.Few ISRs were classified as grade 3 (approximately 1% of ISR events), consistent across both BMI categories and dosing regimens.There were no grade 4 or 5 ISR events.Abbreviations: AE, adverse event; BMI, body mass index; CAB, cabotegravir; ISR, injection site reaction; LA, long-acting; Q4W, every 4 weeks; Q8W, every 8 weeks; RPV, rilpivirine; W, week.