Assessing Real-world Weight Change in the Treatment and Prevention of HIV-1 with a Cabotegravir-containing Regimen

1. INTRODUCTION

Human immunodeficiency virus (HIV)-1 remains a serious and life-threatening global public health emergency. In 2022, an estimated 39 million people globally were living with HIV, with about 630,000 people dying from HIV-related causes [1]. With the introduction of modern antiretroviral therapy (ART), the life expectancy of a person living with HIV (PLWH) has been prolonged significantly, with a 51% reduction in HIV-related deaths since 2010 [1]. While safe and efficacious, current combination antiretroviral regimens require lifelong daily oral intake of medication, which can be burdensome for patients and potentially associated with increased risk of non-adherence, treatment failure, and the emergence of drug resistance [2].

Long-acting injectable (LAI) therapy was introduced to help improve patient satisfaction and overcome the challenges associated with other ART regimens [2, 3]. In a recent survey, 73% of patients reported interest in trying LAI therapy, noting less frequent dosing [4]. Cabotegravir, an integrase strand-transfer inhibitor (INSTI), and rilpivirine, a non-nucleoside reverse transcriptase inhibitor (NNRTI), were co-formulated into a long-acting parenteral nanosuspension which was approved by the FDA as the first LAI therapy for PLWH in January 2021 [5]. In December 2021, LAI cabotegravir alone was approved by the FDA for pre-exposure prophylaxis (PrEP) against HIV infection [6]. Both agents have a long duration of action, which allows for maintenance dosing of LAI therapy as infrequently as every eight weeks, with no oral background therapy required [5, 6].

One question surrounding the use of INSTIs in LAI therapy is the incidence of adverse events. Several oral INSTIs were found to be associated with weight gain in clinical trials [7-10]. A retrospective, observational study using data obtained from the North American AIDS Cohort Collaboration on Research and Design (NA-ACCORD) found that INSTI-based regimens, primarily dolutegravir- or raltegravir-based, were associated with greater weight gain compared to NNRTI-based regimens, and similar weight change compared to protease inhibitor (PI)-based regimens [7]. Another pooled analysis found similar results, with the magnitude of weight gain being larger in more recent studies with the use of newer ART regimens. Finally, women living with HIV who were switched to an INSTI-based ART regimen experienced a significantly greater increase in body weight, BMI, body circumferences, and percent body fat compared to women on non-INSTI-based ART regimens in a recent analysis [9]. In addition to the female sex, studies have revealed baseline factors of lower CD4 count, higher HIV-1 RNA, and no injection drug use, and Black and Hispanic race were associated with weight gain. However, cabotegravir was not included in these trials [8, 10].

Little data exists on whether treatment with cabotegravir is associated with weight gain. The HIV Prevention Trials Network Study 077 (HPTN 077) analyzed weight gain in participants receiving at least one injection of cabotegravir or placebo (134 vs 43 participants) for HIV prevention in low-risk, uninfected individuals. This phase 2b study revealed no significant differences in weight change between the two arms over a 41-week period, prompting further investigation [10, 11]. The ATLAS and FLAIR trials were two phase 3, randomized, open-label trials assessing the efficacy of cabotegravir-rilpivirine in maintaining HIV-1 viral suppression [2, 12]. A 48-week pooled analysis of these two trials reported a weight change of 2.34 kg in the LAI arm and 1.17 kg in the current ART arm from baseline to week 48 [13]. However, the ATLAS and FLAIR trials included predominantly white males (73%), which is not representative of the patient population we see at our institution’s infectious disease clinics [2, 12]. The pathogenesis of weight gain in PLWH is poorly understood and may lead to serious comorbidities, including diabetes mellitus and cardiovascular disease, making early identification of these effects a priority [10, 14]. Hence, the primary objective of this study is to assess the real-world weight change in patients who switched to LAI cabotegravir or cabotegravir-rilpivirine from another anti-retroviral regimen.

2. MATERIALS AND METHODS

3. RESULTS

4. DISCUSSION

Treatment with LAI therapy has been a focus of research in recent years to improve patient satisfaction and adherence and decrease the negative stigma associated with oral ART therapy. The results of this study suggest that LAI cabotegravir-based regimens can be effective at maintaining HIV-1 viral suppression without significant weight change. Comparing mean weight change from baseline to 3, 6, and 12 months, there was no significant weight change at any time point for the primary and secondary outcomes. However, given the small sample size of this study, there were 3 patients who lost a significant amount of weight due to other causes that may have influenced our results. One patient experienced unexplained weight loss of ~10 kg. During this time, they had an abnormal TSH and were referred for evaluation of their thyroid. Another patient had a prolonged hospital stay >1 month, leading to a weight loss of ~23 kg. A third patient lost ~16 kg after the health decline of a loved one. They stopped eating and began heavily drinking alcohol. While these values can be considered outliners, the primary endpoint was not close to reaching significance at any time point, and we do not believe they affected the outcome of the analysis.

Adverse events were similar between patients on LAI cabotegravir-rilpivirine versus cabotegravir (60% vs. 50%). Injection site reactions, primarily pain or nodules at the injection site, were common but subsided within a few days after receiving the injection and infrequently led to discontinuation of therapy. One patient experienced significant numbness and tingling of the lower extremities and muscle spasms, which worsened at night, affecting their ability to sleep. This was not believed to be related to cabotegravir-rilpivirine, as symptoms remained despite the eventual discontinuation of therapy. They were later referred to neurology for evaluation for restless leg syndrome and peripheral neuropathy. Another patient reported experiencing headaches on cabotegravir-rilpivirine and discontinued therapy as they believed it was making them sick. Around the same time, the patient had significant uncontrolled hypertension, hypokalemia, and stopped drinking alcohol after previously drinking 2-3 glasses of liquor per day. Therefore, this adverse event may also be unrelated to cabotegravir-rilpivirine.

There were two confirmed cases of virologic failure in patients on cabotegravir-rilpivirine. Studies have shown that mutations associated with rilpivirine (RPV) resistance, HIV-1 subtype A1/A6, BMI > 30 kg/m², and missed doses or receiving doses outside the treatment window have been associated with increased risk of confirmed virologic failure [3, 15, 16]. In our study, all injections in those with confirmed virologic failure were received within the recommended treatment window. In both cases, the patients were HIV-1 subtype B and obese (BMI > 30 kg/m²) at baseline. The first patient was found to have V90I, K101E, and Y181C mutations on DNA sequencing, resulting in resistance to rilpivirine and other NNRTIs. The second patient was found to have a Q148K mutation, resulting in resistance to cabotegravir and other INSTIs. Both patients were transitioned back to oral ART.

Our population was distinct from the patients in previous trials since our patients were primarily minorities. Pre-specified subgroup analysis showed no difference between race and gender in relation to weight change at 3, 6, or 12 months. This non-significant weight change was generally consistent across other pre-specified subgroups except for patients who were overweight prior to initiation of LAI therapy, were on concurrent medications known to cause weight gain, and in patients who achieved clinically significant weight loss in the year prior to initiation of LAI therapy. Interestingly, the significant weight change in these subgroups was only seen at the 6-month time point, not at 3- or 12-months. This warrants further investigation in future studies.

This study has several limitations. The small sample size and high rate of discontinuation prevent generalizability to other institutions. Given the retrospective design, weights were obtained from the patient’s chart. With injections being given bimonthly, patients often did not have an appointment in the clinic or recorded weight in the chart at each time point. Therefore, the weight collected was defined as the first weight found on chart search recorded immediately after the 3-, 6-, or 12-month endpoint was reached. This allowed us to better capture data for as many patients as possible. If a weight was not recorded at a clinic visit and no other weight was found on the chart search, the weight was recorded as not available, and the patient was not included in the analysis for that specific time point. In this regard, studies conducted with real-time weights may produce additional useful results.

CONCLUSION

In this study, bimonthly LAI cabotegravir and cabotegravir-rilpivirine were not associated with significant weight change from baseline to 12 months. Adverse events were common, but most patients were able to maintain viral suppression or remained HIV-negative throughout 12 months of therapy. This represents real-world data that may be used to influence provider clinical decision-making when selecting a regimen for treatment or prevention of HIV-1.

AUTHORS’ CONTRIBUTION

J.B.S and S.J.J.: Study conception and design; N.A.T: Data collection; N.A.T. and J.M.: Analysis and interpretation of results; N.A.T.: Manuscript was written.

REFERENCES