Abstract
Background:
Long-acting injectable antiretroviral therapy (LAI-ART) with cabotegravir and rilpivirine offers a promising alternative to daily oral ART, particularly for young people living with HIV (YLHIV). However, there are limited data on the feasibility, acceptability and efficacy of LAI-ART delivery in lower- and middle-income countries.
Methods:
We conducted a 48-week, community-based study of cabotegravir and rilpivirine LAI-ART among YLHIV aged 12–24 years in Cape Town, South Africa. Participants (virally suppressed, struggling with adherence or ART naïve) were required to be virally suppressed (<50 copies/ml) on first-line ART at the end of the screening phase before switching to cabotegravir and rilpivirine LAI-ART every 8 weeks for 48 weeks. Feasibility, retention, adherence to injection schedules, viral suppression and safety were assessed. Virological failure was defined as two consecutive viral loads ≥200 copies/ml.
Results:
Between December 2022 and December 2023, 172 participants were screened and 134 (77·9%) switched to LAI-ART. The median age was 20 years and 62·7% were assigned female at birth. Nearly one quarter (23·1%) were under 18 years of age. Overall, 47% had prior NNRTI exposure. Retention at week 48 was 98·5% with 97·5% of all injections given on schedule. Viral suppression, as indicated by VL <50 copies/ml, at week 48 was achieved by 95·5% with no confirmed virological failures. All participants who completed week 48 opted to remain on LAI-ART.
Interpretation:
LAI-ART was safe, highly acceptable and effective among YLHIV. These findings support the feasibility of implementing LAI-ART through community-based models in sub-Saharan Africa.
Funding:
The AFINAty study (PACTR202402700796695) was funded by Johnson & Johnson (Award number TMC278LAHTX3004).
Keywords: Long-acting injectable ART, Cabotegravir, Rilpivirine, Young people living with HIV
INTRODUCTION:
South Africa has the largest population of people living with HIV (PLWH) globally, with an estimated 8.15 million people affected, equating to a prevalence of 12·9%.1 Among young people aged 15–24 years, HIV prevalence is 6·84%, and although this has declined over the past decade, younger people continue to experience sub-optimal retention in care and lower rates of viral suppression.2–4 HIV viral load data from the National Health Laboratory Service (NHLS) from July 2019 to June 2020 showed that only 57% of young people living with HIV (YLHIV) aged 15–19 years on antiretroviral therapy (ART) had viral load testing done. Among those tested, only 50·71% were virally suppressed (viral load <50 copies/mL), 21·23% had low level viraemia (50- <1000 copies/mL) and 28·06% had a viral load ≥1000 copies/mL5, underscoring how far this group remains from achieving the UNAIDS 95–95-95 targets.6 Consistent with these gaps in treatment outcomes, mortality rates amongst South African adolescents living with HIV are significantly higher compared to their HIV-negative peers, particularly in those struggling with ART adherence.7
In this context, alternative ART delivery approaches that better address adherence challenges are urgently needed. South Africa’s ART delivery currently relies exclusively on oral ART.8 Long-acting injectable antiretroviral therapy (LAI-ART), specifically intramuscular dosing of cabotegravir in combination with rilpivirine administered every 8 weeks, represents a novel ART delivery approach that has demonstrated non-inferiority to oral ART in international clinical trials.9 In African cohorts, the CARES and IMPALA studies have similarly shown non-inferiority to oral ART among adults who were either virologically suppressed on ART or had evidence of prior sub-optimal adherence.10,11 To date, the MOCHA and LATA studies provide the only evidence evaluating LAI-ART using adult dosing among virologically suppressed adolescents. Week 48 and subsequent Week 96 analyses of MOCHA confirmed durable virological suppression with 94.4% of adolescents maintaining a viral load below 50 copies/mL as well as sustained safety and tolerability in this cohort.12–15 More recently, the LATA study demonstrated that cabotegravir and rilpivirine was non-inferior to oral ART.16 These findings provide further evidence supporting the long-term effectiveness of LAI-ART in young people.
Beyond efficacy, multiple factors influence the preference for LAI-ART among YLHIV. These include inadvertent disclosure from being seen with ARV tablets, HIV-related stigma and struggles with pill burden.17 The MOCHA study demonstrated high acceptability and tolerability of LAI-ART in YLHIV,12 and early qualitative data from our study, the Acceptability and Feasibility of long-acting INjectable ART in Adolescents and Young Adults (AFINAty), corroborates these findings.18 The convenience and discretion of LAI-ART make it a potentially highly acceptable option for certain groups of PLWH in sub-Saharan Africa, particularly those facing stigma and discrimination due to their HIV status.19
Although LAI-ART is a promising option, several concerns remain for implementation in resource constrained settings, including cold-chain requirements, drug costs, hepatitis B coinfection and the efficacy of rilpivirine in populations with a high prevalence of non-nucleoside reverse-transcriptase inhibitor (NNRTI) resistance.20 To address these, the AFINAty trial was designed to evaluate the acceptability and tolerability of LAI-ART among YLWH aged 12–24 years across the HIV care continuum: ART-naive, virally suppressed or struggling with adherence; and to assess feasibility of delivering this regimen in a community-based research site. This study will provide critical evidence for future scale up of LAI-ART in settings facing similar constraints.
METHODS:
Study design
The AFINAty trial was a non-randomised, longitudinal, uncontrolled, interventional, three-cohort, switch study among 3 cohorts of YLHIV on standard of care first-line ART, exploring the acceptability and feasibility of the LAI-ART delivery. These 3 cohorts included participants who were virally suppressed, struggling with adherence or ART naïve commencing ART during the screening phase of the study.
Study setting
The trial was conducted at a community-based clinical research site in Cape Town, South Africa. Participants were either recruited from a large public ART community health centre (CHC) just outside of Cape Town, which provides free ART services to over 8500 PLWH; or from mobile clinics providing HIV testing and prevention services operating in the same health district as the CHC. The research site is housed within a densely populated peri-urban area, with a population of around 100 000 people, predominantly black African and Xhosa-speaking. The area has a mixture of formal and informal dwellings, high rates of unemployment and a high HIV prevalence rate.21
Participants and procedures:
Screening phase
In the screening phase participants aged 12–24 years, living with HIV, and weighing at least 35 kg, were enrolled into one of three cohorts. Cohort 1 included participants who were taking first-line oral ART and were virally suppressed for 1-year preceding enrolment with no evidence of recent poor adherence. Cohort 2 included participants on first-line ART with sub-optimal adherence in the preceding year as evidenced by their most recent viral load (≥ 50 copies/mL), missed clinic visits or missed pharmacy collection. Cohort 3 included participants who were ART naïve and about to commence first-line ART. We excluded participants who had evidence of hepatitis B infection, were pregnant, breastfeeding or were using any contraindicated medication e.g. Rifampicin. Additionally, we excluded those with known resistance to NNRTIs or integrase strand transcriptase inhibitors (INSTIs) based on clinical history and folder review. Baseline HIV resistance testing was not conducted.
To be eligible to initiate LAI-ART, all participants needed to be virally suppressed (<50 copies/mL copies) at the end of the screening phase. Cohort 1 participants were eligible to switch to LAI-ART within 4 weeks of confirmation of baseline viral suppression. Cohort 2 participants remained on their current ART regimen and Cohort 3 participants were commenced on first-line ART, both received enhanced adherence support for up to 24 weeks prior to switching. Enhanced adherence support consisted of at least 3 to 4 counselling sessions with a trained ART counsellor. Viral load testing was repeated at 12 weeks and 24 weeks during the screening phase. If their viral load was <50 copies/mL at either of these time points, they were switched to LAI-ART within 7 days.
Injection phase
All participants had the option of choosing oral lead-in (OLI) consisting of daily oral cabotegravir 30mg and rilpivirine 25mg for 28 days prior to switching to LAI-ART. Participants who opted out of OLI received their first doses immediately, as intragluteal injections of long-acting cabotegravir 600mg and long-acting rilpivirine 900mg. The second dose was administered 4 weeks later, after which injections were administered 8-weekly. The optimal dosing window for injections was within 7 days before or after the target date. Oral-bridging of daily oral cabotegravir 30mg and rilpivirine 25mg was given for up to 60 days to participants who would be unable to attend their next injection visit within the optimal dosing window.
We clinically assessed participants who presented more than 7 days after their target injection date and if continuation of LAI-ART was found to be appropriate, the injection schedule was resumed. If participants were more than 4 weeks late for their injection (i.e. >12 weeks since last dose), the loading dose was repeated (i.e. same day injection with another dose after 4 weeks). Thereafter the 8-weekly injection schedule was resumed. Participants with a BMI >30kg/m2 were injected with a longer 2-inch needle. Participants received a dose of cabotegravir and rilpivirine at 6 visits: week 4, 8, 16, 24, 32 and 40 after which they were offered 2 years of post-trial access to LAI-ART or return to standard of care oral ART.
At the screening visit all participants had a CD4 count, viral load and full safety blood tests, which included a full blood count, sodium, potassium, creatinine, glucose, alanine aminotransferase (ALT), alkaline phosphatase, bilirubin, hepatitis B surface antigen and Hepatitis C antibodies. Limited safety blood tests (ALT only) were done at weeks 4, 8, 16, 24 and 48. Viral load was measured at weeks 4, 8 (cohorts 2 and 3 only), 24 and 48 with plasma stored to allow for genotyping of any samples with viral load ≥50 copies/mL. All viral loads ≥50 copies/mL were repeated within 4 weeks. We defined virological failure as two consecutive viral loads ≥200 copies/mL. Serum pregnancy tests were done for all women of childbearing potential at screening, with urine pregnancy tests at all subsequent visits. All women of child-bearing potential were required to use contraception throughout study participation as the study drugs have not yet been approved for use in pregnancy.
We assessed adverse events, injection site reactions and concomitant medications at all study visits. Adverse events were graded using the DAIDS adverse events grading system (version 2·1).22 Injection site reactions were assessed using a standard set of healthcare worker administered questions at every visit from week 8 onwards. We evaluated adherence based on the timing of injections or pill counts for participants receiving standard of care ART, OLI or oral bridging.
Outcomes:
The primary outcomes were acceptability and tolerability of the LAI-ART delivery determined by incidence and severity of drug-related adverse events. Feasibility of delivering injectable ART in a community setting was determined by adherence to injection schedule. Secondary objectives included efficacy of the long-acting injection regimen in each cohort. We conducted in-depth interviews with a subset of participants. Qualitative outcomes will be reported elsewhere.
Statistical analysis:
There was no sample size calculation done as all statistical analyses were exploratory with no tests of hypothesis. All baseline characteristics are summarized using descriptive statistics with counts and percentages for binary or categorical variables and means, medians, standard deviations and interquartile ranges for continuous variables. All participants who met the eligibility criteria and were switched to LAI-ART were included in the viral load analysis.
Ethics approval
This trial was approved by the South African Health Products Regulatory Authority (SAHPRA, 20220801) as well as the University of Cape Town Human Research Ethics Committee (UCT HREC, 359/2022). A Data and Safety Monitoring Board (DSMB) consisting of independent expert advisors was established and reviewed the study at regular intervals. Participants ≥16 years old provided written informed consent, while participants <16 years old provided informed assent with a parent/guardian providing informed consent. Parental consent was waived for participants aged 16–17 years by UCT HREC. All participants were reimbursed for time, inconvenience and travel expenses at each study visit in compliance with SAHPRA recommendations. The trial is registered on the South African National Clinical Trials Registry (DOH-27–102022-7051) and the Pan African Clinical Trials Registry (PACTR202402700796695).
RESULTS:
Recruitment took place between December 2022 and December 2023. We recruited 172 participants into the screening phase of the study (Figure 1). Of these, 134 (77·9%) entered the injection phase and were switched to LAI-ART: 59 of 66 (89·4%) participants screened in cohort 1, 35 of 52 (67·3%) in cohort 2, and 40 of 54 (74·1%) in cohort 3. There were 38 participants that did not enter the injection phase. Twenty participants did not meet the eligibility criteria: eight were not virally suppressed at the end of screening, two had a positive pregnancy test at screening, two had symptoms of depression and suicidal ideation, two cohort 3 participants were not truly ART naïve, two participants had a raised ALT, one was on a contraindicated medication, one did not meet the weight cutoff, one cohort 1 participant was virally suppressed for less than a year and one had a history of seizures. The other reasons for not entering the injection phase included being lost to follow-up (n=14), voluntary withdrawals (n=2), transferring out the area (n=1) and investigator discretion (n=1). There were no cases of hepatitis B in the screening phase, and no participants were excluded on the basis of breastfeeding.
Figure 1: Flow diagram describing participant progress throughout the study.
*The screening phase for cohort 1 was up to 4 weeks, for cohort 2 and 3 it was 12–24 weeks.
Table 1 describes the baseline demographic characteristics of the cohort in the injection phase. Eighty-four (62·7%) participants were assigned female at birth. The median (IQR) age was 20 (18–22) years old, with almost a quarter of participants being under the age of 18 (n=31, 23·1%). The majority of participants in cohorts 1 (n=42, 71·2%) and 2 (n=31, 88·6%) had vertically acquired HIV, while in cohort 3, almost all participants (n=39, 97·5%) had horizontally acquired HIV. At baseline, 59·7% (n=80) had WHO Clinical Stage 1 HIV disease, and 79·1% (n=106) had a CD4 count >350 cells/μl. For cohorts 1 and 2, the mean (±SD) time on ART at study start was 10·6 (6·1) years and 13·4 (5·4) years respectively, with 63 participants (47%) having had prior exposure to an NNRTI-containing regimen. None of the participants that had prior NNRTI exposure were known to have failed an NNRTI-containing regimen.
Table 1:
Baseline demographics
| Cohort 1 virally suppressed, no adherence challenges |
Cohort 2 struggling with adherence |
Cohort 3 ART-naive |
Total | |
|---|---|---|---|---|
|
| ||||
| Number initiated on LAI-ART | 59 | 35 | 40 | 134 |
|
| ||||
| Sex at birth: n(%) | ||||
| Male | 21 (35·6) | 19 (54·3) | 10 (25·0) | 50 (37·3) |
| Female | 38 (64·4) | 16 (45·7) | 30 (75·0) | 84 (62·7) |
|
| ||||
| Gender | ||||
| Male | 20 (33·9) | 19 (54·3) | 8 (20·0) | 47 (35·1) |
| Female | 38 (64·4) | 16 (45·7) | 30 (75·0) | 84 (62·7) |
| Transgender | 1 (1·7) | 0 (0) | 2 (5·0) | 3 (2·2) |
|
| ||||
| Age (years): median (IQR) | 20 (16·5–22) | 20 (17·5–21) | 20.5 (19–22) | 20 (18–22) |
|
| ||||
| Number under 18: n (%) | 18 (30·5) | 9 (25·7) | 4 (10) | 31 (23·1) |
|
| ||||
| Median age <18 years: median (IQR) | 15 (14–16) | 16 (15–17) | 16·5 (16–17) | 16 (14–16.5) |
|
| ||||
| Race: % | ||||
| Black African | 100 | 100 | 100 | 100 |
|
| ||||
| BMI: Median (IQR) | 22·5 (20.5–26.2) | 21·8 (20.3–27.2) | 24·1 (21.6–30.1) | 23·1 (20.6–27.1) |
|
| ||||
| Mode of transmission: n (%) | ||||
| Vertical | 42 (71·2) | 31 (88·6) | 1 (2·5) | 74 (55·2) |
| Horizontal | 17 (28·8) | 4 (11·4) | 39 (97·5) | 60 (44·8) |
|
| ||||
| WHO Clinical stage: n(%) | ||||
| Stage 1 | 34 (57·6) | 8(22·9) | 38 (95·0) | 80 (59·7) |
| Stage 2 | 4 (6·8) | 2 (5·7) | 2 (5·0) | 8 (6·0) |
| Stage 3 | 20 (33·9) | 13 (37·1) | 0 (0) | 33 (24·6) |
| Stage 4 | 1 (1·7) | 12 (34·3) | 0 (0) | 13 (9·7) |
|
| ||||
| CD4 count at baseline: n (%) | ||||
| <100 cells/μl | 0 (0) | 1 (2·9) | 0 (0) | 1 (0·7) |
| 100–<200 cells/μl | 0 (0) | 1 (2·9) | 3 (7·5) | 4 (3·0) |
| 200–<350 cells/μl | 2 (3·4) | 7 (20) | 14 (35) | 23 (17·2) |
| 350–<500 cells/μl | 9 (15·3) | 7(20) | 11 (27·5) | 27 (20·1) |
| ≥500 cells/μl | 48 (81·4) | 19 (54·3) | 12 (30·0) | 79 (59·0) |
|
| ||||
| Unsuppressed viral load at study entry: n(%) | 0 (0) | 14 (40.0) | 40 (100) | 54 (40.3) |
| VL 50–1000 copies/ml | 0 (0) | 3 (8·6) | 0 (0) | 3 (2·2) |
| VL >1000 copies/ml | 0 (0) | 11 (31·4) | 40 (100) | 51 (38·1) |
|
| ||||
| Prior NNRTI exposure: n (%) | 39 (66·1) | 24 (68·6) | - | 63 (47·0) |
|
| ||||
| Time on ART before study entry (years): mean (sd) * | 10·6 (6·1) | 13·4 (5·4) | - | 11·61 (6·0) |
Cohort 3 excluded from total calculation due to being ART naïve at study entry.
Abbreviations: LAI-ART = Long-acting injectable antiretroviral therapy; sd = standard deviation; BMI = body mass index; WHO = World Health Organisation; VL = viral load; NNRTI = Non-nucleoside reverse transcriptase inhibitor
Eleven (8·2%) participants that switched to LAI-ART chose to receive OLI. After initiation of LAI-ART there were 794 follow-up doses of LAI-ART given, 774 (97·5%) of these were given within the target window (Figure 2). Four doses (0·5%) were given between 8 and 14 days early and 16 doses (2·0%) were given late. Only 3 of these participants required a re-loading dose (>28 days late).
Figure 2: Distribution of injection visits relative to the visit window (7 days before or after the scheduled target date).
Of the 134 participants who initiated LAI-ART, 132 (98·5%) attended the week 48 visit. One participant (cohort 1) transferred out of the study area after the week 24 visit. One participant (cohort 3) was lost to follow-up after the week 8 visit. During the study period five participants required oral bridging.
A total of 149 injection site reactions were reported by 100 participants. These included pain, tenderness, swelling, erythema and subcutaneous nodules. Injection site reactions were all grade 1 and grade 2 except for one participant that had a secondary infection at the injection site (grade 3). The participant was treated with oral antibiotics and the infection resolved before her next injection visit. The injection site reactions with the longest duration were subcutaneous nodules. There were four post-injection reactions, with the most common symptoms being sweating, dizziness, headache, shortness of breath and the feeling of a tight chest. These reactions were of sudden onset after administration of the cabotegravir or rilpivirine injection, lasted a few minutes, resolved spontaneously, did not require any intervention and all 4 participants opted to continue with cabotegravir and rilpivirine. Other drug-related adverse events included nausea, headache, fatigue, insomnia, hypersomnia, myalgia, pruritis and a high ALT. These adverse events were all grade 1 or 2 and none led to injection discontinuation. There were no anaphylactic reactions. There were two serious adverse events reported, neither of which were related to the study drug. This included a participant who attempted suicide by overdose requiring an overnight stay in an emergency facility. Another participant developed a Bartholin’s abscess requiring in-patient hospitalisation and treatment.
One participant (cohort 1) was diagnosed with drug-sensitive pulmonary tuberculosis (TB) after the first set of cabotegravir and rilpivirine injections. She was initiated on TB treatment, and her ART was switched from the study drug to oral ART due to drug-drug interactions between Rifampicin and the LAI-ART. After completion of her TB treatment, she was reinitiated on LAI-ART. She was successfully treated for TB and remained virally suppressed throughout. There were no other contra-indicated medications reported.
Due to receipt of an incorrectly labelled result, one participant was initiated on LAI-ART while viraemic (52976 copies/mL). The error was found after the participant received the loading dose of cabotegravir and rilpivirine. The decision was made to allow the participant to continue in the study but to exclude her from viral outcome analyses. This participant was found to be virally suppressed to <50 copies/ml within four weeks of initiation of LAI-ART and remained virally suppressed to week 48.
There were nine unsuppressed viral loads (VL>50 copies/mL) in the study period up to week 48: three at week 8 (one from cohort 2, two from cohort 3), two at week 24 (one from cohort 1, one from cohort 2), and four at week 48 (two from cohort 1, one from cohort 2, one from cohort 3). All of these were low level viraemia (50–1000 copies/mL) apart from one at week 8. All participants resuppressed prior to the next injection being due. At week 48, there were no cases of confirmed viral failure and the viral suppression rate to <50 copies/ml at week 48 for all participants was 95·5% (94·9% for cohort 1, 97·1% for cohort 2 and 95% for cohort 3) (Table 2).
Table 2:
Week 48 Viral load outcomes
| Suppressed (VL<50 copies/mL) | Low level viraemia (VL 50–1000 copies/mL) | No virologic data in window | Confirmed virological failure (2 consecutive viral loads ≥200 copies/mL) | |
|---|---|---|---|---|
| Cohort 1 (n=59) | 56 (94·9%) | 2 (3·4%) | 1(1·7%)* | 0 (0%) |
| Cohort 2 (n=34)** | 33 (97·1%) | 1 (2·9%) | 0 (0%) | 0 (0%) |
| Cohort 3 (n=40) | 38 (95·0%) | 1 (2·5%) | 1 (2·5%)† | 0 (0%) |
| Total (n=133) | 127 (95·5%) | 4 (3·0%) | 2 (1·5%) | 0 (0%) |
Participant transferred out
One participant excluded due to being initiated on LAI-ART while viraemic
One participant was lost to follow-up.
Abbreviations: VL = viral load; mL = millilitre
Of the 132 participants who attended the Week 48 visit, all of them opted to continue LAI-ART in the long-term extension phase.
DISCUSSION
We switched 134 young people (aged 12–24 years) living with HIV onto LAI-ART in a community setting in Cape Town South Africa. Despite concerns about the feasibility of delivering LAI-ART in lower- and middle-income countries, our findings demonstrate that this delivery model is both feasible and effective in a real-world community context. Retention at week 48 was 98·5%, no participants experienced virological failure, and more than 85% of participants received all injections within the designated window. Although injection site reactions were assessed using the same questionnaire consistently until week 48, there was a significant decrease in reported injection site reactions after the week 24 visit. This likely demonstrates improving tolerance to cabotegravir and rilpivirine over time. Importantly, all participants chose to continue LAI-ART after the 48-week study period, reflecting sustained acceptability among this age group. Acceptability and tolerability were further demonstrated by the majority of drug-related adverse events being assessed as mild or moderate, with only one grade 3 and no grade 4 adverse events.
Our results add to a growing body of evidence showing that LAI-ART can be implemented successfully across diverse populations, including young people who had either vertically or horizontally acquired HIV and were either ART-naïve or treatment experienced, with or without a history of sustained viral suppression or prior adherence challenges. These results are in keeping with findings of high efficacy and participant satisfaction in other African studies in adults, and international studies in young people.10–12 The MOCHA and LATA studies have also demonstrated that long-acting cabotegravir and rilpivirine can successfully maintain viral suppression, has a favourable safety profile and was strongly preferred by their adolescent cohorts including South Africa.14–16 Our study extends these findings by demonstrating similar effectiveness and acceptability in a more heterogenous cohort of young people in a community-based African setting and supports the use of LAI-ART beyond clinical trial settings.
These results are particularly important for YLHIV, who continue to have the poorest ART outcomes.3,4 YLHIV face multiple structural and psychosocial barriers to daily oral therapy, such as stigma, pill fatigue, busy schedules and unstable living situations, all of which contribute to suboptimal adherence and lower rates of viral suppression.23 The high levels of acceptability and viral suppression in this study demonstrate that LAI-ART can overcome some of these barriers in this population.17
In addition, these results highlight the feasibility of delivering LAI-ART in a community setting, maintaining continuity of care close to participants’ homes and using existing health infrastructure to maintain cold chain storage at 2 to 8°C. Our findings are reassuring in a population with high prior NNRTI exposure and in the absence of baseline genotyping. We also found no cases of hepatitis B during screening for this study. Safety was not compromised by less frequent viral load monitoring than has been seen in phase III registrational clinical trials. These results provide valuable insights for future scale-up of long-acting HIV treatment in a Southern African setting.
This study has several limitations. As cabotegravir and rilpivirine are not registered in South Africa at the time of the study launch, this study was implemented as a clinical trial with research infrastructure and participant compensation. The participants were supported by a clinical trials team, with increased capacity to call and trace should visits be missed. These factors may limit generalisability in the broader South African context; however, the strong acceptability and clinical outcomes suggest that, with appropriate systems support, similar results could be achievable in programmatic settings. Baseline genotyping was not performed, although stored samples will allow for retrospective analysis of resistance patterns.
Delivering this method of ART needs to be assessed in a community clinic with an attempt to integrate it into their routine clinic services to assess true feasibility of LAI-ART.
CONCLUSIONS
In conclusion, this study demonstrates that LAI-ART can be feasibly and effectively delivered to young people living with HIV in a community-based setting in South Africa. The exceptionally high rates of viral suppression and retention highlight the potential of LAI-ART to transform HIV treatment outcomes among YLHIV, who have historically struggled to remain engaged in care. These findings underscore the importance of continued research to find effective and scalable long-acting HIV treatment modalities that can be delivered in youth-friendly settings and the need for ongoing advocacy for access to LAI-ART worldwide. Further follow-up from this cohort, including outcomes at week 96 and beyond, will provide important evidence on the long-term durability, safety and acceptability of LAI-ART in this population.
Supplementary Material
Table 3:
Self-reported injection site reactions (number of reactions as a percentage of total participants receiving injections at each time point)
| Week 4 | Week 8 | Week 16 | Week 24 | Week 32 | Week 40 | Week 48 | |
|---|---|---|---|---|---|---|---|
| Pain or tenderness | 91(67·9) | 22(16·4) | 6(4·5) | 6(4·5) | 1(0·8) | 0(0) | 0(0) |
| Erythema or redness | 9(6·7) | 2(1·5) | 1(0·8) | 0(0) | 0(0) | 0(0) | 0(0) |
| Induration or swelling | 17(12·7) | 5(3·7) | 3(2·3) | 0(0) | 0(0) | 0(0) | 1(0·8) |
| Pruritis | 17(12·7) | 6(4·5) | 5(3·8) | 3(2·3) | 0(0) | 1(0·8) | 0(0) |
| Rash | 0(0) | 0(0) | 0(0) | 1(0·8) | 0(0) | 0(0) | 0(0) |
| Bruising | 10(7·5) | 1(0·7) | 0(0) | 0(0) | 0(0) | 0(0) | 0(0) |
| Nodule | 5(3·7) | 2(1·5) | 1(0·8) | 1(0·8) | 1(0·8) | 0(0) | 0(0) |
| Total * | 149 | 38 | 16 | 11 | 2 | 1 | 1 |
Percentage not reported as some participants reported more than one type of injection site reaction per study visit
Acknowledgements:
We would like to sincerely thank all our participants that took part in this study. We would also like to acknowledge our site staff that contributed towards the implementation of all study-related activities. Finally, we acknowledge the Western Cape Department of Health for their role in ensuring this study could be run successfully.
Conflicts of Interest and Source of Funding:
LGB received honoraria for advisories from ViiV Healthcare, Merck PTY LTD and Gilead Sciences. CO has received honoraria from MSD for advisory boards. LJ has honoraria from ViiV Healthcare. CS, NM, RM and MA have no conflicts to declare. The AFINAty study was funded by Johnson & Johnson (Award number TMC278LAHTX3004).
Data availability statement:
The data that support the findings of this study are not publicly available due to privacy and ethical restrictions. However, the de-identified participant data are available from the corresponding author on reasonable request.
REFERENCES
- 1.Statistics South Africa. Mid-Year Population Estimates 2025. 2025. www.statssa.gov.za
- 2.Azia IN, Mukumbang FC, Van Wyk B. Barriers to adherence to antiretroviral treatment in a regional hospital in Vredenburg, Western Cape, South Africa. South Afr J HIV Med. 2016;17(1). doi: 10.4102/sajhivmed.v17i1.476 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Zanoni BC, Archary M, Buchan S, Katz IT, Haberer JE. Systematic review and meta-analysis of the adolescent HIV continuum of care in South Africa: The Cresting Wave. BMJ Glob Health. BMJ Publishing Group Ltd. 2016;1(3):1. doi: 10.1136/bmjgh-2015-000004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Naidoo I, Takatshana S, Sewpaul R, et al. Past and current status of adolescents living with HIV in South Africa, 2005–2017. BMC Res Notes. BMC Res Notes. 2022;15(1). doi: 10.1186/s13104-022-06006-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mazanderani AH, Sherman GG. PAEDIATRIC & ADOLESCENT HIV VIRAL LOAD MONITORING , 2014–2020. Vol 18. 2020. [Google Scholar]
- 6.Global Task Team on 2030 Targets. Global HIV Target Setting for 2030. 2025. [Google Scholar]
- 7.Zhou S, Toska E, Gwampi B, et al. Age-specific all-cause mortality rates among adolescents and youth living with and without HIV: Evidence from a cohort study in South Africa. J Int AIDS Soc. 2025;28(6):1–9. doi: 10.1002/jia2.26522 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.South African National Department of Health. 2023 ART Clinical Guidelines. 2023;(June). https://knowledgehub.health.gov.za/elibrary/2023-art-clinical-guidelines-management-hiv-adults-pregnancy-and-breastfeeding-adolescents
- 9.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. The Lancet. 2020;396(10267):1994–2005. doi: 10.1016/S0140-6736(20)32666-0 [DOI] [PubMed] [Google Scholar]
- 10.Kityo C, Mambule IK, Musaazi J, et al. Switch to long-acting cabotegravir and rilpivirine in virologically suppressed adults with HIV in Africa (CARES): week 48 results from a randomised, multicentre, open-label, non-inferiority trial. Lancet Infect Dis. 2024;24(10):1083–1092. doi: 10.1016/S1473-3099(24)00289-5 [DOI] [PubMed] [Google Scholar]
- 11.Cresswell FV, Kafeero P, Norcross C, et al. Long-acting cabotegravir and rilpivirine in adults with sub-optimal HIV control in sub-Saharan Africa: the IMPALA trial 48-week results. IAS. Oral late-breaker abstract OAS0105LBP. [Google Scholar]
- 12.Lowenthal ED, Chapman J, Ohrenschall R, et al. Acceptability and tolerability of long-acting injectable cabotegravir or rilpivirine in the first cohort of virologically suppressed adolescents living with HIV (IMPAACT 2017/MOCHA): a secondary analysis of a phase 1/2, multicentre, open-label, non-compar. Lancet HIV. 2024;11(4):e222–e232. doi: 10.1016/S2352-3018(23)00301-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Gaur A, Baltrusaitis K, Capparelli EV, et al. Long-Acting Cabotegravir+Rilpivirine in Adolescents: IMPAACT 2017 Week 96 & End-of-Study Results [CROI Abstract 155]. Top Antivir Med. 2026;34(1):33. [Google Scholar]
- 14.Gaur AH, Baltrusaitis K, Capparelli EV., et al. Safety, antiviral activity, and pharmacokinetics of long-acting injectable cabotegravir–rilpivirine in virologically suppressed adolescents living with HIV-1 (IMPAACT 2017/MOCHA): 48-week results of a multinational, phase 1/2, single-arm study. Lancet HIV. 2026;13(2):e85–e94. doi: 10.1016/S2352-3018(25)00242-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Lowenthal ED, Chapman J, Baltrusaitis K, et al. Acceptability and tolerability of long-acting injectable cabotegravir–rilpivirine in adolescents with HIV-1 (IMPAACT 2017/MOCHA): 48-week results of a multicentre, open-label, non-comparative phase 1/2 trial. Lancet HIV. 2026;13(2):e95–e103. doi: 10.1016/S2352-3018(25)00241-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Bwakura Dangarembizi M, Chappell E, Kekitiinwa AR, et al. Every-8-weeks injectable cabotegravir and rilpivirine is superior to daily oral tenofovir disoproxil fumarate/lamivudine/dolutegravir in adolescents living with HIV in sub-Saharan Africa: LATA 96-week results. In: 26th International AIDS Conference. [Google Scholar]
- 17.Toska E, Zhou S, Chen-Charles J, Gittings L, Operario D, Cluver L. Factors Associated with Preferences for Long-Acting Injectable Antiretroviral Therapy Among Adolescents and Young People Living with HIV in South Africa. AIDS Behav. 2023;27(7):2163–2175. doi: 10.1007/s10461-022-03949-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Atujuna M, Jennings L, Rousseau E, et al. Experiences and acceptability of Long-acting Injectable antiretroviral therapy (ART) among adolescents enrolled in the Acceptability and Feasibility of long-acting INjectable ART in Adolescents and Young-Adults (AFINAty) study in South Africa. Preprint posted online 2024:OAD3705. [Google Scholar]
- 19.Gashema P, Iradukunda PG, Saramba E, et al. Bridging the gap: identifying barriers and strategies for widespread implementation of long-acting injectable antiretroviral therapy in Sub-Saharan Africa: a scoping review. BMC Infect Dis. 2025;25(1). doi: 10.1186/s12879-025-10871-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.da Silva J, Siedner M, McCluskey S, Chandiwana N, Venter F, Raizes E. Drug resistance and use of long-acting ART. Lancet HIV. 2022;9(6):e374–e375. doi: 10.1016/S2352-3018(22)00059-5 [DOI] [PubMed] [Google Scholar]
- 21.Statistics South Africa. Gugulethu Census 2011. 2011. Accessed August 14, 2026. https://www.statssa.gov.za/?id=320&page_id=4286
- 22.Division of AIDS (DAIDS) Table for Grading the Severity of Adult and Pediatric Adverse Events, Corrected Version 2.1. Division of AIDS, National Institute of Allergy and Infectious Diseases, National Institutes of Health, US Department of Health and Huma Services. [Google Scholar]
- 23.Mabasa RA, Madiba S, Mothiba TM. Structural, Familial, and Psychosocial Factors Affecting Long-Term Antiretroviral Treatment Adherence amongst Adolescents Living with Perinatally Acquired HIV in Limpopo, South Africa. Int J Environ Res Public Health. 2023;20(22). doi: 10.3390/IJERPH20227074 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are not publicly available due to privacy and ethical restrictions. However, the de-identified participant data are available from the corresponding author on reasonable request.