Viral Suppression Rates at 48 Weeks in People With Human Immunodeficiency Virus Starting Long-Acting Cabotegravir/Rilpivirine With Initial Viremia

Abstract

Background

We previously demonstrated at the Ward 86 human immunodeficiency virus (HIV) clinic in San Francisco that long-acting (LA) cabotegravir (CAB)/rilpivirine (RPV) (LA-CAB/RPV) can rapidly lead to viral suppression in people with HIV (PWH) with viremia due to adherence challenges. We now evaluate the durability of viral suppression in this population.

Methods

We conducted a retrospective cohort study of PWH who started LA-CAB/RPV with viremia (HIV RNA viral load ≥50 copies/mL) before December 2022. Our primary outcome was viral suppression (viral load <50 copies/mL) with LA-CAB/RPV persistence (not discontinued or late by >14 days) at 48 weeks, using the viral load closest to 48 ± 8 weeks. We also describe viral failure, defined as a <2-log decline in viral load at 4 weeks or a viral load ≥200 copies/mL after initial viral suppression with emergent CAB- or RPV-associated resistance mutations; overall 48-week viral suppression including those switched to alternative antiretroviral therapy (ART).

Results

Fifty-nine PWH initiated LA-CAB/RPV with viremia and were included in the analysis; 49% had a CD4 cell count <200/µL, and the median baseline viral load was 42 900 copies/mL (quarter 1–quarter 3, 5272–139 038). At 48 weeks, 47 PWH met the primary outcome of viral suppression with LA-CAB/RPV persistence (80% [95% confidence interval, 67%–89%]). Five had viral failure with resistance (3 with RPV-associated and 2 with CAB- and RPV-associated mutations), and 1 was lost to follow-up. At week 48, 2 of those with viral failure were suppressed on alternative regimens (lenacapavir + bictegravir/tenofovir alafenamide/emtricitabine and CAB + lenacapavir). The overall viral suppression at week 48 with either LA-CAB/RPV or alternative ART was 92% (54 of 59).

Conclusions

In PWH initiating LA-CAB/RPV with initial viremia, 48-week viral suppression (<50 copies/mL) was seen in 92%. LA ART can be an important tool for improving viral suppression among patients who face adherence challenges to oral ART.

Graphical Abstract

This graphical abstract is also available at Tidbit: https://tidbitapp.io/tidbits/48-week-viral-suppression-rates-in-people-with-hiv-starting-long-acting-cab-rpv-with-initial-viremia-e96411a0-d008-4ac6-bf5d-8e30dcaec8b0

HIV/AIDS, long-acting antiretroviral therapy, injectable cabotegravir and rilpivirine, viremia, viral suppression

Long-acting (LA) injectable cabotegravir (CAB)/rilpivirine (RPV; LA-CAB/RPV) is approved by the US Food and Drug Administration for people with human immunodeficiency virus (HIV; PWH) on oral antiretroviral therapy (ART) with stable viral suppression [1–3]. However, LA-CAB/RPV may benefit PWH who are unable to attain viral suppression with oral ART [4], particularly those with psychosocial factors that affect oral ART adherence, such as housing instability, mental illness, substance use, and HIV-associated stigma [5–8]. Clinical cohort data suggest that most PWH with baseline viremia can achieve HIV viral suppression when starting LA-CAB/RPV, though there are no published data on sustained suppression [9–14]. In February 2024, ACTG LATITUDE A5359, a randomized trial of LA-CAB/RPV versus oral ART among PWH with adherence challenges who first attained viral suppression on oral ART [15], was stopped early due to the superior efficacy of LA-CAB/RPV in the interim analysis [15]. Both the International AIDS Society–USA and the US Department of Health and Human Services recently changed guidelines to support use of LA-CAB/RPV for PWH with viremia who are unable to take oral ART consistently despite adherence support [16, 17].

Our group previously reported outcomes from the LA-CAB/RPV program at Ward 86, demonstrating that ≥95% of PWH who initiated LA-CAB/RPV with initial viremia attained viral suppression within the first 8 months of therapy (median follow-up, 33 weeks; range, 10–83 weeks) [9, 10]. In the current analysis, we sought to evaluate longer-term virologic outcomes after initiation of LA-CAB/RPV among PWH with HIV RNA levels ≥50 copies/mL at baseline.

METHODS

Study Setting

Our study took place at Ward 86, a publicly funded academic HIV clinic in San Francisco that serves approximately 2400 PWH with high levels of psychosocial and economic vulnerability. Ward 86 provides HIV treatment and prevention services for people with public insurance (eg, Medicaid and Medicare) or who are uninsured. The clinic includes the POP-UP program, which provides low-barrier HIV primary care for people experiencing homelessness who are viremic and have difficulty engaging in scheduled appointments [18].

Ward 86 LA Injectable ART Program

We have previously described the Ward 86 Special Programs on Long-Acting Antiretrovirals to Stop HIV (SPLASH) program, reported early clinical outcomes, and characterized the multilevel implementation strategies used by SPLASH [9, 10, 19]. From program inception, we allowed LA-CAB/RPV initiation for PWH who are unable to attain HIV viral suppression on oral ART due to adherence challenges. Our clinic protocol [20] (supplement) recommends every-4-week dosing with optional transition to every-8-week dosing after viral suppression is achieved and maintained for 3–6 months [21]. Key strategies include clinic pharmacist/pharmacy technician medication procurement, injection appointment adherence monitoring, and text/phone reminders. Incentives were offered if the provider/pharmacy team determined that the patient would need an extra incentive ($10 grocery vouchers per on-time injection, $20 if on every-8-week dosing). Patients in the POP-UP low-barrier HIV primary care program for people experiencing homelessness [18] received additional outreach by a social worker in the event of a missed injection appointment and had access to additional incentives (maximum $50/month for clinic visits, laboratory measurements, and viral suppression) [22].

Study Design and Population

We conducted a retrospective cohort study of PWH aged ≥18 years who initiated LA-CAB/RPV at Ward 86, were viremic at the time of their first injection (HIV plasma RNA ≥50 copies/mL), and initiated LA-CAB/RPV prior to 5 December 2022 to allow sufficient duration of follow-up for 48-week outcome ascertainment before database closure. We excluded PWH who initiated other LA ART agents alongside LA-CAB/RPV at baseline (eg, lenacapavir [LEN]).

Data Collection and Analysis

We collected baseline demographic and clinical data, LA-CAB/RPV injection dates, and HIV RNA measurements through record abstraction. All LA-CAB/RPV administrations and laboratory measurements were conducted through routine clinical care. Clinicians documented baseline HIV genotype, hepatitis B status, housing status, and substance use (stimulants or opioids) in a standardized note template at the time of referral for LA-CAB/RPV initiation.

Our primary outcome was HIV viral suppression (defined as <50 copies/mL) with persistence 48 weeks after LA-CAB/RPV initiation. We defined viral suppression using the HIV viral load measured closest to 48 weeks with a ±8-week window, considering missing values to be unsuppressed. We defined LA-CAB/RPV persistence as not discontinued or late by >14 days at 48 weeks after initiation. Secondary outcomes included viral suppression to <50 copies/mL at 48 weeks regardless of ART regimen. We also report virologic failure, defined as <2-log viral load decline at 4 weeks after LA-CAB/RPV initiation or viral load ≥200 copies/mL after initial viral suppression with emergent CAB- or RPV-associated mutations (Supplementary Table 1) [20]. We also report the number and proportion of patients with late LA-CAB/RPV injections (>7 and >14 days after 30- or 60-day manufacturer-specified dosing interval) [20]. We calculated 95% confidence intervals (CIs) using the bias-ascertained Wilson method. The University of California, San Francisco Institutional Review Board approved data abstraction and analysis for this study.

RESULTS

As of 31 January 2024, 286 PWH had received ≥1 dose of LA ART at Ward 86, 101 of whom had baseline viremia (HIV viral load ≥50 copies/mL). We included 59 PWH who started LA-CAB/RPV with baseline viremia by 5 December 2022. Approximately half (53%) were experiencing homelessness or unstable housing, and 63% were using stimulants at baseline (Table 1). One-third were receiving care through the POP-UP program [18]. Half (49%) had a CD4 cell count <200/µL. The median baseline viral load was 42 900 copies/mL (quarter 1–quarter 3, 5272–139 038 copies/mL), and 69% had a baseline viral load ≥10 000 copies/mL. No patients had active hepatitis B infection at baseline; 73% had surface antibody positivity (n = 26 vaccination; n = 17 prior cleared infection), and those without hepatitis B immunity were offered vaccination.

Table 1.

Baseline Patient Characteristics (N = 59)

Characteristics	Category	Patients, No. (%)
Gender	Female	5 (8.5)
	Male	53 (89.8)
	Gender minority	1 (1.7)
Age, y	18–29	2 (3.4)
	30–49	29 (49.2)
	≥50	28 (47.5)
Race/ethnicity	White	24 (40.7)
	Black/AA	14 (23.7)
	Latino	17 (28.8)
	Other	4 (6.8)
Housing status	Stable	28 (47.5)
	Unstable	26 (44.1)
	Homeless	5 (8.5)
Substance use	Methamphetamine/cocaine	36 (61.0)
	Opioids	6 (10.2)
CD4 count, cells/µL	<50	9 (15.3)
	50–199	20 (33.9)
	200–349	13 (22.0)
	350–499	7 (11.9)
	≥500	10 (16.9)
HIV VL, copies/mL	50 to <200	3 (5.1)
	200 to <1000	5 (8.5)
	1000 to <10 000	10 (16.9)
	10 000 to <100 000	22 (37.3)
	≥100 000	19 (32.2)
Baseline resistance mutationsa
NNRTI	K103N	9 (15)
	N348I	2 (3)
	V108I	1 (2)
INSTI	T97A	2 (3)
	E157Q	1 (2)
	S230R	1 (2)

Abbreviations: AA, African American; HIV, human immunodeficiency virus; INSTI, integrase strand transfer inhibitor; NNRTI, nonnucleoside reverse-transcriptase inhibitor; VL, viral load.

^aIncluding major or accessory INSTI or NNRTI mutations listed in the Stanford HIV Drug Resistance Database (hivdb.stanford.edu/hivdb).

Viral Suppression Outcomes

At 48 weeks, 80% (47 of 59 patients [95% CI, 67%–89%]) remained on LA-CAB/RPV and were virally suppressed (Table 2). The overall viral suppression at week 48 irrespective of ART regimen was 92% (54 of 59 [95% CI, 81%–97%]).

Table 2.

Patient Status at Week 48

Status	Patients, No. (%)a
	VL <50 Copies/mL (n = 54)	VL ≥50 Copies/mL (n = 5)	Overall (N = 59)
Remained on LA-CAB/RPV with no virologic failure	47 (87)	1 (20)b	48 (81)
Discontinued LA-CAB/RPV and resumed oral ARTc	5 (9)	0	5 (8)
Failure with resistance			5 (8)
On-time injections	2 (4)	0
Late injections	0	1 (20)
Lost to follow-up, off oral ART, later determined to have resistanced	0	2 (40)
Lost to follow-up and off oral ART without resistance testingd	0	1 (20)	1 (2)

Abbreviations: ART, antiretroviral therapy; LA-CAB/RPV, long-acting cabotegravir/rilpivirine; VL, viral load;

^aThree VLs were missing at week 48, all with VL ≥50 copies/mL before or after the window and off ART.

^bIntensified to LA-CAB/RPV + lenacapavir for low-level viremia (week 48 VL <200 copies/mL).

^cThe reasons for discontinuation and switching to oral ART included adverse effects (n = 3), provider-initiated switch due to viremia associated with incorrect needle length in patient with body mass index ≥30 (calculated as weight in kilograms divided by height in meters squared) (no resistance; n = 1), and transfer to another clinic that did not have LA-CAB/RPV available (n = 1).

^dReasons for discontinuation/loss to follow-up and not taking oral ART included a fixed belief that the patient was cured of human immunodeficiency virus, psychosis, and depression (each n = 1).

LA-CAB/RPV Discontinuation and Virologic Failure

Among the 12 patients who did not meet the primary outcome of LA-CAB/RPV persistence with viral suppression, 5 discontinued and remained virally suppressed on oral ART at week 48, 5 had virologic failure (3 with RPV-associated mutations and 2 with both CAB- and RPV-associated resistance mutations), 1 was lost to follow-up and off oral ART without resistance testing available, and 1 was intensified to LEN + LA-CAB/RPV due to persistent low-level viremia without emergent CAB/RPV-associated resistance mutations (range, 98–614 copies/mL) (Table 2). As such, 5 patients (5 of 59 [8%]) had direct evidence of emergent viral resistance, though up to 6 may have had viral failure with resistance if those lost to follow-up are included (6 of 59; [10% maximum estimate]). All 6 of the patients who either developed emergent resistance (n = 5) or were lost to follow-up without resuming oral ART (n = 1) had baseline viral loads >10 000 copies/mL, and 5 had baseline CD4 count cells <200/µL, indicating advanced HIV disease. We provide details on these patients below and in Table 3.

Table 3.

Details in 7 Patients With Adverse Virologic Outcomes

Patient	Baseline	Follow-up	Status at wk 48
Patient 1 (early failure despite on-time injections)	VL 137 000 copies/mL; CD4 cell count: 15/µL; mutations: T97A (minor INSTI)	VL 4400 copies/mL at wk 4 (<2 log decline); genotyping with E138 K (NNRTI) and R263 K (INSTI); disengaged from care but later returned and was resuppressed on BIC/TAF/FTC + LEN	Suppressed on BIC/TAF/FTC + LEN
Patient 2 (early failure despite on-time injections)	VL 215 000 copies/mL; CD4 cell count 71/µL; mutations: V179I and N348I (minor NNRTI)	VL 29 000 copies/mL at wk 4 (<2-log decline); genotyping showed new L100I and Y181I mutations (NNRTI); disengaged from care but later returned and was resuppressed on CAB + LEN	Suppressed on CAB + LEN
Patient 3 (failure after late injection)	VL 363 789 copies/mL; CD4 cell count 306/µL; mutations: none	VL 29 000 copies/mL when 10 d late to 10th injection (1st injection after switch to every-8-wk dosing); genotyping showed a new K101E mutation. VL was 79 copies/mL 6 d later, and initial viremia was thought to be a laboratory error; oral ART was offered, but patient was unwilling to switch and was continued on LA-CAB/RPV. Patient subsequently received 3 on-time every-8-wk injections but had a VL of 4500 copies/mL with a new Q148R mutation at the 13th injection (wk 68); he was unwilling to take oral ART but was resuppressed with addition of subcutaneous LEN (CAB/RPV + LEN)	Emergent resistance after late injection; suppressed on LA-CAB/RPV + LEN
Patient 4 (failure after discontinuation)	VL 67 000 copies/mL; CD4 cell count 20/µL; mutations: none	Received 10 on-time injections; VL <200 copies/mL by wk 4 and <50 copies/mL by wk 24; self-discontinued due to severe depression; genotyping 18 wk after discontinuation showed new extensive NNRTI mutations: K101E, E138 K, Y181F/I/N, and M230L	Emergent resistance after discontinuation; presumed viremic
Patient 5 (failure after discontinuation)	VL 801 000 copies/mL; CD4 cell count: 27/µL; mutations: none	Received 6 on-time injections then discontinued all ART due to the patient's belief they were cured of HIV; declined all subsequent laboratory testing or care engagement until presenting 1 y later with Shigella diarrhea; VL at that time was 226 000 copies/mL with genotyping showing reverse-transcriptase mutations E138A/K/T and Y181C; patient resumed ART with CAB + LEN (only willing to take LAI ART); VL <30 copies/mL after 1 m	Emergent resistance after discontinuation; presumed viremic
Patient 6 (discontinued; lost to follow-up)	VL 19 000 copies/mL; CD4 cell count 20/µL; mutations: none	Received 2 injections, suppressed by wk 4; discontinued LA-CAB/RPV due to psychosis; at wk 36, VL from emergency department was 32 000 copies/mL (no genotyping)	Lost to follow-up, presumed viremic
Patient 7 (on LA-CAB/RPV + LEN; VL ≥50 copies/mL)	VL 284 000 copies/mL; CD4 cell count 118/µL; mutations: none	VL 418 copies/mL after 8 wk; persistent low-level viremia (range, 98–614 copies/mL); VL of 614 copies/mL at 6th injection prompted addition of LEN; subsequent continued low-level viremia (VL range, 94–185 copies/mL; 126 copies/mL at wk 48)	Low-level viremia on LA-CAB/RPV + LEN

Abbreviations: ART, antiretroviral therapy; BIC, bictegravir; CAB, cabotegravir; FTC, emtricitabine; HIV, human immunodeficiency virus; INSTI, integrase strand transfer inhibitor; LA, long-acting; LAI, long-acting injectable; LEN, lenacapavir; NNRTI, nonnucleoside reverse-transcriptase inhibitor; RPV, rilpivirine; TAF, tenofovir alafenamide; VL, viral load.

We previously described the 2 patients with early virologic failure; since our last report [10], both attained viral suppression on alternative partially or fully injectable ART and were suppressed to <50 copies/mL at week 48. Patient 1 had virologic failure at week 4 with a new E138K (nonnucleoside reverse-transcriptase inhibitor) and R263K (integrase strand transfer inhibitor) mutation. This patient was switched back to oral ART (darunavir/cobicistat [DRV/c]/tenofovir alafenamide [TAF]/emtricitabine [FTC] single-tablet regimen) but had ongoing adherence challenges due to pill size and nausea, which were unrelieved by cutting the pill and using antiemetics. After multiple repeated genotypes did not reveal additional resistance mutations, Patient 1 was switched to LEN and twice-daily dolutegravir plus TAF/FTC (subsequently simplified to LEN and bictegravir/TAF/FTC once viral suppression was attained and repeated genotype testing confirmed no additional integrase mutations). Patient 1 remained virally suppressed on this regimen at week 48, and out to 100 weeks at the time of analysis.

The second patient (patient 2) had virologic failure at week 4 with genotyping revealing a new L100I RPV-associated mutation. Patient 2 was switched to oral ART (DRV/c/TAF/FTC), though subsequent viral loads ranged from 35 000 to 300 000 copies/mL due to persistent challenges adhering to oral ART. Subsequent genotyping revealed additional RPV-associated resistance (Y181I) but no emergent integrase mutations. Considering ongoing oral adherence challenges, patient 2 was restarted on LA-ART with monthly intramuscular CAB and every-6-month LEN. Patient 2's viral load declined from 35 000 to <50 copies/mL within 1 month, and this patient remained virally suppressed at week 48, and out to 88 weeks at the time of analysis.

Patient 3 was on time for his first 9 every-4-week injections, and at the ninth injection he transitioned to every-8-week dosing. At the 10th LA-CAB/RPV injection, he presented 10 days late and had an HIV viral load of 29 000 copies/mL. He was prescribed DRV/c/TAF/FTC but never took it. A repeated viral load 6 days later was 79 copies/mL (week 46). After clinic team review, high-level viremia was thought due to laboratory error, and the patient was continued on LA-CAB/RPV given his inability to adhere to oral ART. Genotyping later returned with a new K101E mutation. The patient subsequently received 3 on-time every-8-week injections but had a viral load of 4500 copies/mL with a new Q148R mutation at the time of his 13th injection (week 68). He was strongly encouraged to switch to DRV/c/TAF/FTC at this time, though he was completely unable to take oral ART. Thus, LEN was added to LA-CAB/RPV; the patient has remained suppressed on LA-CAB/RPV + LEN out to week 85 at the time of analysis.

Patient 4 discontinued LA-CAB/RPV after the 10th administration due to severe depression and did not resume oral ART; the viral load was missing at week 48, though the subsequent viral load was unsuppressed, with genotyping showing extensive nonnucleoside reverse-transcriptase inhibitor resistance (K101E, E138K, Y181F/I/N, and M230L). The patient subsequently died 81 weeks after LA-CAB/RPV initiation from complications of AIDS.

Patient 5 discontinued ART after 6 on-time injections due to their belief that they were cured of HIV and had no further need for testing or treatment. One year after discontinuation, this patient returned to care due to diarrhea caused by Shigella. At that time, the CD4 cell count was 24/µL (1%), the HIV RNA level was 226 000 copies/mL, and genotyping showed reverse-transcriptase mutations at E138A/K/T and Y181C. The patient wanted to restart ART but was only willing to take injectable medications. CAB + LEN was started, and the HIV RNA level was <30 copies/mL after 1 month.

Patient 6 received 2 injections and attained viral suppression by week 4; they subsequently self-discontinued LA-CAB/RPV due to severe psychosis and were lost to follow-up. Their viral load was 32 000 copies/mL at week 36 (32 weeks after last LA-CAB/RPV injection) during an emergency department visit, but no genotype data were available.

Dosing and Alternate Injection Locations

By 48 weeks, 19 patients (32%) with initial viremia had transitioned from every-4-week to every-8-week LA-CAB/RPV administration. Over the first 48 weeks following LA-CAB/RPV initiation, the median viral load frequency across patients was 8.1 weeks (quarter 1–quarter 3, 7.1–9.1 weeks). Nine patients received ≥1 injection outside Ward 86: 6 in healthcare settings outside the clinic (inpatient hospitalization, emergency department, skilled nursing facility, or another clinic while traveling) and 5 in settings outside a healthcare facility (public health street medicine nursing, supportive housing nursing, or home health), with some patients receiving out-of-clinic injections at >1 site (Supplementary Table 2). No patients receiving out-of-clinic injections experienced virologic failure.

Late Injections

Among 530 postbaseline injections given, 499 (94%) were administered on-time, 28 (5.3%) were administered >7 days late (0–5 late injections per patient), and 3 (0.6%) were missed due to loss to follow-up. Among all patients, 25% (15 of 59) received ≥1 injection >7 days late and 8% (5 of 59) ≥14 days late. Of those ≥14 days late (range, 14–19 days), all received reinitiation of LA-CAB/RPV 600/900 mg (induction dose) and remained virally suppressed at reinitiation and through 48 weeks. Seven patients in the POP-UP clinic received in-person outreach by a social worker: 4 for missed injection appointments (2 returned and received on-time injections, 1 returned and received a late injection, and 1 remained lost to follow-up), 1 following virologic failure to encourage return to clinic for alternative ART initiation, and 2 following LA-CAB/RPV discontinuation to ensure ongoing adherence to oral ART. Social worker outreach in the community often required multiple attempts, with an average of 3 outreach visits per person (range, 1–5).

DISCUSSION

In PWH who are viremic due to adherence challenges, our data support that LA-CAB/RPV can attain and maintain high levels of viral suppression for ≥48 weeks. Among PWH initiating LA-CAB/RPV with viremia, we observed that 80% remained on LA-CAB/RPV and were virally suppressed; overall, 92% had 48-week viral suppression to <50 copies/mL, including suppression on alternative ART regimens. In our cohort, half had advanced immunosuppression, highlighting the importance of attaining and maintaining viral suppression in this population to avoid opportunistic infections and death. Our findings lend additional support for recent International AIDS Society–USA and Department of Health and Human Services guideline changes to recommend LA-CAB/RPV for PWH who are unable to take oral ART consistently despite adherence support and who are at high risk of HIV disease progression [16, 17].

Among PWH initiating LA-CAB/RPV with viremia, 10% had virologic failure or discontinued without resumption of alternate ART over the first year, similar to the 7.2% confirmed virologic failure rate reported over the first year in the ACTG LATITUTDE study [15]. We observed 2 distinct categories of virologic failure: (1) early failure despite on-time injections and (2) late failure due to late injections, personal choice not to resume oral ART, or loss to follow-up. It is unclear why the 2 patients with early virologic failure developed emergent resistance after only a single injection; the rapidity of failure raises the possibility of preexisting genotypic resistance that was not seen on prior genotypes. Low CAB or RPV concentrations are also a possibility and have been associated with virologic failure [23–25]. However, we did not have the ability to measure drug levels in routine practice.

These 2 patients had persistent oral ART adherence challenges after LA-CAB/RPV discontinuation but were both ultimately able to reattain viral suppression with a partially or fully injectable regimen. While failure after late injections was uncommon, 1 patient had a very high viral load with emergent RPV-associated resistance after presenting to the clinic only 10 days late for an injection. However, this patient steadfastly refused oral ART while remaining engaged in care. Rather than withhold ART altogether, his care team opted for intensification with LEN, and the patient has remained suppressed. This case highlights the complexity of long-acting injectable treatment, in which the unique circumstances of a patient's care trajectory require nuanced decision making.

Despite significant psychosocial challenges in our population, loss to follow-up was uncommon, though 3 patients were lost to follow-up due to mental health challenges. Since these patients all had advanced immunosuppression at the time of initiation, LA-CAB/RPV may have been the best option to avoid disease progression. Two of these patients were lost to follow-up after ≥6 on-time injections, highlighting the need for ongoing support to stay on schedule with LA-CAB/RPV therapy even after clinical routines appear to be established.

Although most PWH initiating LA-CAB/RPV without viral suppression remained on this regimen after 48 weeks, 20% switched back to oral ART, required a regimen change, or were lost to follow-up. In this small sample, all who discontinued and switched back to oral ART remained virally suppressed. A qualitative study of early adopters of LA-CAB/RPV has found that some PWH who initiate LA-CAB/RPV with viremia and subsequently discontinue due to side effects or injection site pain may experience renewed motivation for oral ART adherence [26].

We have previously described patient-, clinic-, and system-level implementation determinants for use of LA-CAB/RPV among PWH with viremia, along with strategies used at Ward 86 [19]. Key elements of our program include a dedicated pharmacy technician to facilitate medication procurement, drop-in access for injections, standardized reminder calls/text messages, and gift cards to incentivize injections. One-third of patients also received HIV primary care through the POP-UP program, providing drop-in access to integrated HIV primary care, substance use treatment, mental health treatment, social work, and case management support alongside drop-in injection access [18].

Within the Ward 86 program, some patients required additional support to maintain their injection schedule, but most did not. Over 48 weeks, one-quarter of PWH initiating LA-CAB/RPV with viremia had ≥1 late injection (>7 days late), triggering additional phone outreach and communication with case managers and other community partners. POP-UP patients also received additional support with in-person outreach by a social worker to facilitate return to clinic after a missed injection appointment. Nine patients (15%) received ≥1 out-of-clinic injection, which required coordination with other care teams, including physically bringing the medication to the hospital or emergency department when administration was required in those locations. Support for responding to late injections and facilitating offsite administration are important for retention but require additional resources and coordination.

The proportion of late or missed injections in our overall cohort (approximately 6% of all injections and 25% with ≥1 late injection) is comparable to that seen in the ACTG A5359 LATITUDE study [15], though higher than observed in cohorts of patients who initiated LA-CAB/RPV while virally suppressed [27–29]. While most patients did not have adverse virologic outcomes with injections administered up to 19 days late, 1 patient did have emergent resistance when 10 days late for his first every-8-week injection. Patients transitioning to every-8-week dosing may require more careful monitoring and may also require additional reminders to adhere to the new longer dosing interval.

Our study has several limitations, including small sample size, limited to a single clinic with support services that may not be available in all settings. Despite the small size of our study, we do provide clear evidence for high levels of viral suppression at 48 weeks after LA-CAB/RPV initiation among PWH who are not virally suppressed—even with imperfect adherence to injection schedules. All patients received LA-CAB/RPV, so we do not have the ability to compare viral suppression oral ART. However, it is likely that our observed viral suppression of 92% represents a substantial improvement over continued attempts to attain and maintain viral suppression with oral ART in this population.

In conclusion, initiation of LA-CAB/RPV in PWH with baseline viremia resulted in high levels of viral suppression out to 48 weeks. Longer-term follow-up and assessment in other settings is necessary, but our findings demonstrate that LA-CAB/RPV can be an important tool for improving viral suppression among PWH with challenges adhering to oral ART.

Supplementary Data

Supplementary materials are available at Clinical Infectious Diseases online. Consisting of data provided by the authors to benefit the reader, the posted materials are not copyedited and are the sole responsibility of the authors, so questions or comments should be addressed to the corresponding author.

Notes

Financial support. This work was supported by the National Institute of Mental Health, the National Institute of Allergy and Infectious Diseases (grants R01MH123396 and K24AI167805 [principal investigator, K. C.] and 5R37AI098472 [principal investigator, M. G.]), the National Heart, Lung, and Blood Institute (grant K23HL162578 to M. D. H.), and the City and County of San Francisco (support to Ward 86).

Author notes

Potential conflicts of interest. The authors: No reported conflicts of interest. All authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.