The tail phase of CAB-LA was assessed among participants in the Study to Evaluate the Safety Tolerability and Acceptability of Long Acting Injections of the Human Immunodeficiency Virus (HIV) Integrase Inhibitor, GSK1265744, in HIV Uninfected Men (ECLAIR), a phase 2 randomized, placebo-controlled trial that enrolled HIV-uninfected male volunteers and provided CAB-LA 800 mg injections every 12 weeks. Overall, 17% of participants had detectable plasma cabotegravir concentrations 52 weeks after the last injection. Notably, because the dose of CAB-LA used in ECLAIR did not meet PK targets derived from non-human primate studies, a different dosing regimen was studied in the phase 3 CAB-LA PrEP trials (600 mg given 4 weeks apart, followed by 600 mg every 8 weeks) [
48].
Tail phase PK was also assessed among healthy HIV-uninfected men and women in the HPTN 077 trial. HPTN 077 was a double-blind, randomized, placebo-controlled phase IIa trial conducted at 8 global sites that included 600 mg and 800 mg dosing regimens. Median time from last CAB-LA injection to cabotegravir concentrations decreasing below the lower-limit of quantitation (LLOQ) was 43.7 weeks (IQR 31.1–66.6; range 20.4–152.5) for male participants compared to 67.3 weeks (IQR 29.1–89.6; range 17.7–225.5) for female participants. The time to LLOQ was approximately 30% longer among women than men. Moreover, the apparent terminal-phase half-life of CAB was approximately 31% longer among participants with a higher body mass index (BMI). Of note, however, less than 10% of PK variability was explained by sex and BMI. The investigators note that additional factors, such as muscle size, muscle-fat content, host genetics, and potential delivery of injections into subcutaneous tissues or intravenously rather than intramuscularly should also be explored as potential contributors [
23••].
Knowledge Gaps on CAB-LA as PrEP
Despite concerns about the potential for INSTI resistance during the CAB-LA PK tail, little is currently known about the risk of resistance after stopping CAB-LA injections. Incident HIV infections reported from the CAB-LA PrEP trials have not occurred during the PK tail, potentially because these infections occurred during active product use and because the trial protocols included the use of daily oral TDF/FTC for 48 weeks after the last CAB-LA injection to protect against HIV acquisition during waning cabotegravir concentrations. Additional data are anticipated on the risk of resistance during the CAB-LA PK tail with longer follow-up of trial participants during the open-label extensions of HPTN-083 and 084. Moreover, with use of CAB-LA PrEP in routine clinical settings, where injections may be stopped or missed without switching to oral PrEP or oral CAB to cover the PK tail, further data on the risk of resistance during the tail phase may be forthcoming.
Special Considerations: PK Variability in Different Populations
As CAB-LA PrEP rolls out in routine clinical settings, PK variability in diverse populations could result in higher or lower drug concentrations and influence both protection from HIV acquisition and the risk of ARV resistance. CAB-LA PK during pregnancy is of high interest given that women of childbearing age, particularly in Sub-Saharan Africa, account for a disproportionate burden of new HIV infections globally [
50]. Moreover, the risk of HIV acquisition may be elevated during pregnancy and pregnant women are thus a priority population for HIV prevention [
51]. Plasma concentrations of several ARVs are altered during pregnancy, particularly during the third trimester [
52]. Inadequate exposure to an ARV for prevention during pregnancy could lead to HIV acquisition and perinatal transmission to the infant. Notably, for dolutegravir use for HIV treatment, the area under the curve (AUC) may be decreased during the third trimester of pregnancy, but no dose adjustment has been recommended during pregnancy [
52].
Data are currently limited on cabotegravir PK among pregnant persons. Data have been reported among a small number of women who became pregnant while receiving CAB/RPV (oral or LA) for HIV treatment; CAB/RPV was discontinued upon identification of pregnancy and quarterly PK sampling was performed for 52 weeks following the last injection. Plasma cabotegravir concentrations were reported to be within the range of expected concentrations for non-pregnant women [
53]. The HPTN 084 CAB-LA PrEP trial required participants to use long-acting contraception, although pregnancies did occur during the phase III trial. After pregnancy was diagnosed, participants were unblinded and no further CAB-LA was received. Participants were offered daily oral TDF/FTC for prevention (similar to non-pregnant participants who discontinued CAB-LA injection). An analysis of participants who received CAB-LA until pregnancy diagnosis in HPTN 084 found that cabotegravir concentrations were similar to those in non-pregnant women in HPTN 077 [
23••]. Age, weight, race, and pregnancy status were not associated with changes in the apparent terminal phase half-life (
t1/2app) of CAB-LA. Higher BMI (> 27.2 kg/m
2) was associated with longer CAB-LA
t1/2app (fold-change 1.49,
p = 0.069) [
54]. Notably, in the results reported to date, CAB-LA was only given in very early pregnancy, before pregnancy was detected by frequent testing in the trial. Therefore, CAB-LA PK with continued injections during the second and third trimesters remains unknown. The open-label extension (OLE) of HPTN 084 does not require long-acting contraception and participants will be allowed to continue active dosing of CAB-LA during pregnancy and will be followed to assess pregnancy and infant outcomes. Therefore, additional data on CAB-LA PK during pregnancy and the postpartum period, as well as cabotegravir exposure to infants in utero and during breastfeeding, are anticipated from the HPTN 084 OLE and other studies.
Potential drug interactions between CAB-LA and hormone therapy are another important consideration for CAB-LA PrEP use among diverse populations. A study of repeat dosing of cabotegravir 30 mg daily found no significant impact on the PK of oral contraceptives (OC) containing levonorgestrel and ethinyl estradiol, and the FDA label states that no dose adjustments are needed with co-administration [
55]. In a secondary analysis of ciswomen in the phase II HPTN 077 trial, the use of oral hormonal contraceptives was associated with significantly lower peak cabotegravir concentrations compared to women not on hormonal contraception [
56•]. No differences in other CAB-LA PK parameters were observed, nor were differences seen with injectable, implantable, or other contraceptives. The phase III HPTN 084 CAB-LA PrEP trial required the use of long-acting reversible contraception (LARC) and did not find differences in HIV incidence across LARC methods, although there were only four HIV infections in the CAB-LA arm, thus limiting comparisons.
Transgender individuals are another priority population for HIV prevention, given the higher burden of HIV infection among trans persons in the USA and globally [
57]. Several studies have investigated the impact of tenofovir-based PrEP on gender-affirming hormone therapy (GAHT) concentrations among transgender individuals, and the impact of various GAHT regimens on tenofovir-based PrEP [
57‐
59]. In the HPTN 083 phase 3 trial, 12.5% of participants were transwomen who have sex with men. The point estimate for reduction in HIV incidence was similar among transwomen to MSM in the trial, but was not statistically significant due to smaller numbers in this subgroup (HR 0.34, 95% CI 0.08–1.56) [
17••]. A preliminary analysis of PK data from HPTN 083 found that plasma concentrations of CAB were similar among transwomen receiving and not receiving GAHT [
60].