4.1. Summary of Available Evidence
The evidence for this review was drawn from three RCTs — LIPO-010 (N = 412), CTR-1011 (N = 404), and Stanley et al. 2014 (N = 54) — each of which compared tesamorelin 2 mg/day (subcutaneous injection) with placebo. LIPO-010 and CTR-1011 consisted of a 26-week main phase, followed by a 26-week extension phase. The study by Stanley et al. 2014 was six months in duration. The primary efficacy outcome for LIPO-010 and CTR-1011 was the per cent change in VAT at week 26, while the co-primary end points in Stanley et al. 2014 were the changes in VAT and hepatic fat at six months. Other relevant efficacy outcomes from LIPO-010 and CTR-1011 included body image, specifically belly size evaluation, belly appearance distress, and belly profile evaluation, waist circumference, and QoL. Investigators of CTR-1011 also administered the EQ-5D instrument. Relevant harms outcomes were mortality, AEs, SAEs, WDAEs, and notable harms, including injection-site reactions, myalgia, arthralgia, fluid retention or edema, diabetes, and malignancies. In all, the type of evidence from the three trials was congruent with patient expectations (0).
All three trials enrolled participants who were aged 18 to 65 years, HIV-positive, on stable ART regimen, and had objective evidence of abdominal fat accumulation as follows: waist circumference ≥ 95 cm and waist-to-hip ratio ≥ 0.94 for males, and waist circumference ≥ 94 cm and waist-to-hip ratio ≥ 0.88 for females. Some limitations were noted for the trials. First, discussions with the clinical expert consulted by CDR for the purpose of this review highlighted that the populations in all three trials were substantially different than those seen in routine clinical practice in Canada today. In particular, the studies appear to have been conducted at a time when HIV patients were commonly receiving ART regimens that were associated with accumulation of visceral fat. The expert indicated that today’s HIV patients are much more likely to receive ART regimens consisting of backbones other than PIs, including INSTIs and NNRTIs. To this end, of the six regimens recommended by the US DHHS to manage ART-naive patients, five are INSTI-based and one is PI/r-based. In LIPO-010, the number of participants receiving ART regimens consisting of integrase inhibitors was not reported, whereas fewer than 5% of participants in CTR-1011 were receiving INSTI-based therapies. The number of participants on integrase inhibitors was unclear, as the authors reported that seven individuals in the tesamorelin group (25.0%) and six in the placebo group (27.3%) were receiving entry inhibitors and integrase inhibitors. Further, as a result of the large number of HIV treatments available today, single or multiple substitutions of the components of an ART regimen offer optimal virologic suppression with fewer AEs, such as the accumulation of visceral fat. In addition, the psychometric properties of the instruments used to evaluate body image and QoL in LIPO-010 and CTR-1011 are not clear, which leaves uncertain the degree to which the results are valid and clinically meaningful.
4.2. Interpretation of Results
4.2.1. Efficacy
Across all studies, tesamorelin was associated with a statistically significantly greater reduction in VAT versus placebo at week 26 (LIPO-010 and CTR-1011) or six months (Stanley et al. 2014). Both the FDA and Health Canada accepted that an 8% decrease in VAT was an appropriate measure of clinical benefit, although Health Canada noted that this cut-off is “arbitrary.”21 While tesamorelin consistently exceeded this magnitude of effect versus placebo in the main phases of the included trials, the degree to which these results are clinically meaningful is unclear. The FDA, for instance, concluded that tesamorelin is “clearly effective on modifying a biomarker but what the clinical benefit is with this degree of VAT reduction is uncertain.”15 In particular, while there is evidence suggesting that increased visceral fat is associated with greater cardiometabolic risk,18 the FDA indicated that the “link between VAT reduction and improved cardiovascular risk has not been established” in patients with HIV-associated lipodystrophy.15 Health Canada also raised a question about the extent to which tesamorelin provides added benefit to a structured diet and exercise regimen, specifically noting that a “substantial” percentage of placebo-treated participants achieved a decrease of ≥ 8% of VAT at week 26 in the two manufacturer-conducted studies — 25.0% in LIPO-010 and 34.1% in CRR-1011 — even though neither trial included a specified diet and exercise regimen.21 Nevertheless, a post-hoc pooled analysis of LIPO-010 and CTR-1011 suggested that tesamorelin responders experienced statistically significantly improved endocrine and metabolic changes, including triglyceride levels, adiponectin levels, and preservation of glucose homeostasis, as well as statistically significantly greater decreases in waist circumference and VAT over 52 weeks compared with non-responders.25 However, this study is associated with several important limitations, including the post-hoc nature of the analyses and a failure to adjust for multiplicity, which necessitate caution when interpreting the results. Furthermore, the relevance of the aforementioned endocrine changes to hard clinical end points, such as cardiovascular events and death, is unknown.
Further, in LIPO-010 and CTR-1011, with respect to change in VAT at week 26, neither the ART regimen nor the treatment-by-ART regimen interaction were statistically significant in the analytical model, thus suggesting that type of ART regimen did not statistically significantly affect the results. However, these results may be limited by the fact that individual drugs within a class were grouped and analyzed together. Moreover, the clinical expert indicated that VAT (as measured by CT scan) is not routinely used to gauge treatment response, thus limiting the application of the VAT results to support clinical decision-making. Setting aside the utility of VAT, results from the 26-week extension phases of LIPO-010 and CTR-1011 demonstrated that participants who discontinued treatment with tesamorelin experienced increases in VAT to levels that were similar to those measured at baseline or even higher (0, Figure 2). Thus, in order to maintain any reductions in VAT that might appear in the short term, patients would have to require chronic treatment with tesamorelin.
Per the FDA, the manufacturer acknowledged the assessment of VAT through routine CT scans, and proposed monitoring treatment response based on changes in waist circumference.15 In LIPO-010 and CTR-1011, tesamorelin was associated with a statistically significantly greater reduction in waist circumference versus placebo at week 26. Health Canada noted that a 1 cm reduction in waist circumference “should be considered as the minimal acceptable decrease for a satisfactory response” following six months of treatment with tesamorelin.26 Across both studies, the magnitude of the treatment effect exceeded the threshold of 1 cm, specifically an absolute difference of 1.8 cm in LIPO-010 and 1.3 cm in CTR-1011. However, discussions with the clinical expert suggested that these differences are unlikely to be clinically meaningful, especially when considering the baseline waist circumferences, as well as how relatively little they decreased in each treatment group. The expert elaborated that the effects on waist circumference will “likely change little in terms of appearances or how people feel about themselves.”
The clinical expert also emphasized that changes to body image and QoL are the primary factors in guiding the management of patients with HIV-associated lipohypertrophy; feedback from patients appears to indicate that they, too, prioritize improvements to these parameters over other clinical benefits (0). Both LIPO-010 and CTR-1011 collected several PROs related to body image, including belly size evaluation, belly appearance distress, and belly profile evaluation. The FDA considered belly appearance distress a “consequential component” of the PRO evaluation, and required the manufacturer to show a positive effect on this parameter to support the primary efficacy outcome.
Across both trials, the effects of tesamorelin versus placebo on change in belly appearance distress at week 26 were inconsistent: specifically, per the parametric ANCOVA models, there were no statistically significant differences between treatment groups; conversely, per the ranked (non-parametric) ANCOVA models, tesamorelin was associated with a statistically significantly greater reduction in belly appearance distress at week 26 versus placebo. Health Canada considered the non-parametric ANCOVA model to be the “most appropriate robust method” given that the data were not normally distributed.20 Across both trials, there were no statistically significant differences between treatment groups with respect to change in belly size evaluation at week 26. Further, tesamorelin was associated with a statistically significantly greater reduction in belly dysmorphia (belly profile evaluation) versus placebo at week 26 in LIPO-010 but not in CTR-1011. However, the degree to which the observed results are clinically meaningful is uncertain due to the relatively unknown psychometric properties of the instrument used to evaluate changes to body image (0). In particular, the FDA’s SEALD team concluded that the instrument used in the two trials had “questionable” validity, and it raised concerns regarding the instrument’s ability to measure a clinically important treatment effect.24 Still, the FDA voted unanimously in favour of approving tesamorelin, primarily citing the “powerful” patient testimonies,15 which “indicated that whether or not the statistical data holds up, patients subjectively believe treatment with the drug had a beneficial impact in their psychological well-being.”18
4.2.2. Harms
Across all three studies, at least 70% of study participants in each trial experienced a treatment-emergent AE. A greater percentage of participants in Stanley et al. 2014 experienced an AE (tesamorelin: 89.3%; placebo: 95.5%) than those in the main phase of LIPO-010, followed by those in CTR-1011. Further, across the main phases of LIPO-010 and CTR-1011, approximately 5% more participants receiving tesamorelin (LIPO-010: 82.8%; CTR-1011: 74.1%) experienced an AE than those on placebo (LIPO-010: 75.9%; CTR-1011: 69.8%). Results from the 26-week extension phases of LIPO-010 and CTR-1011 were inconsistent with respect to the relative occurrence of AEs between participants who continued tesamorelin from the main phase versus those who switched to placebo (0).
Across all three studies, a greater percentage of participants receiving tesamorelin reported any injection-site condition, myalgia, or any fluid retention or edema than those in the placebo group. According to the clinical expert consulted by CDR for the purpose of this review, these types of AEs are consistent with the administration method and biological properties of tesamorelin. Another notable harm associated with tesamorelin use is glucose intolerance, and, while pre-existing diabetes was an exclusionary criterion, at least for LIPO-010, it is important to highlight the development of diabetes among those participants enrolled in the trial. In LIPO-010, one participant (0.4%) receiving tesamorelin (versus none receiving placebo) developed diabetes mellitus; this was recorded as a treatment-emergent AE, and the manner in which diabetes mellitus was defined was unclear. However, both the FDA and Health Canada note that the odds of developing diabetes (as defined by hemoglobin A1C ≥ 6.5%) was approximately three-fold greater (statistically significant) among participants receiving tesamorelin versus placebo.18,26 Tesamorelin also increases serum IGF-1, which, according to the clinical expert, may be associated with the development or worsening of some cancers. In LIPO-010, a greater percentage of participants receiving tesamorelin versus placebo (2.9% versus 1.5%) developed a malignancy, whereas fewer participants receiving tesamorelin in CTR-1011 versus those receiving placebo (0.4% versus 3.2%) developed a malignancy. Last, there were no deaths in the main phase of LIPO-010 and Stanley et al. 2014. Two participants died in CTR-1011, one in each treatment group. Two participants died in the extension phase of LIPO-010: one who continued tesamorelin from the main phase and another individual who switched from placebo to tesamorelin (0).
The durations of all trials included in this review were insufficient to adequately capture some important safety outcomes, including the risk of cardiovascular diseases, as well as the occurrence of diabetes and cancer, thus leaving uncertain the long-term safety profile of tesamorelin. To that end, Health Canada notes that treatment with tesamorelin “should be discontinued in patients with IGF-1 standard deviation scores greater than 2 after 26 weeks.”26 Given the increase in IGF-1 levels, the tesamorelin product monograph also indicates that individuals who are receiving ongoing treatment with tesamorelin should be monitored for development or worsening of retinopathy.26 To provide evidence for the long-term safety of tesamorelin, the manufacturer is conducting a 10-year prospective cohort study in which HIV-infected patients with abdominal lipohypertrophy and receiving tesamorelin will be compared with a similar group of individuals who are not receiving the treatment.27 The primary outcome of this study is the time to development of malignancies, while secondary outcomes are the time to development or worsening of type 2 diabetes mellitus, diabetic retinopathy, hypersensitivity reactions, hepatic and renal function, AEs and major adverse cardiovascular events. However, the FDA required more conclusive evidence on the risk of developing diabetic retinopathy with tesamorelin,15 so the manufacturer is also conducting a randomized, placebo-controlled, DB non-inferiority trial in patients with HIV-associated lipodystrophy and type 2 diabetes mellitus.28
Overall, Health Canada and the FDA concluded that the data from the manufacturer-submitted trials are inadequate to assess the long-term safety of tesamorelin.15,19 Nevertheless, the FDA noted that the safety concerns “are not immediately life-threatening or severe to reasonably argue against the availability of Egrifta to the HIV population with lipodystrophy who have few to no options.”15
4.3. Potential Place in Therapy
The information in this section is based on information provided in draft form by the clinical expert consulted by CDR reviewers for the purpose of this review.
The prevalence of lipodystrophy in HIV-infected persons is uncertain, particularly as regard the prevalence among patients treated with more recently approved antiviral therapies.
The changes associated with lipodystrophy can be subdivided into the following categories: metabolic changes (elevated low-density lipoprotein cholesterol, reduced high-density lipoprotein cholesterol, elevated triglycerides, glucose intolerance); lipoatrophy (fat thinning, typically involving the face, arms, legs and buttocks); and lipohypertrophy (fat deposition, typically involving the abdomen, the base of the neck posteriorly, and the anterior neck). Metabolic changes, which are mostly associated with PI use, usually respond to a change in antiviral therapy or to therapies like HMG CoA reductase inhibitors (“statins”), whereas lipoatrophy, typically associated with NRTI therapy (especially stavudine, didanosine, and zidovudine), is at best slowly and incompletely reversible with treatment change or discontinuation, although these remain the best treatment option.29–31
Abdominal lipohypertrophy is the result of fat deposition around the viscera within the abdomen, not underneath the skin. Although associated with a number of risk factors, it appears to be associated with the use of older PIs, such as ritonavir and indinavir. Abdominal lipohypertrophy may be cosmetically displeasing32 or associated with glucose intolerance and dyslipidemia.33–36 It is uncommon to see lipohypertrophy develop in treated HIV-infected patients in the present era, as the antiviral therapies that have been implicated in developing lipohypertrophy are no longer in common use in clinical practice. For those patients who have lipohypertrophy, changing or discontinuing antiretroviral therapy to reduce or resolve fat deposition is not always successful, and there are therefore a small number of patients who remain with this problem.
Even though lipohypertrophy is an increasingly uncommon problem, it remains a clinically and psychologically important issue in a small percentage of HIV-infected persons32 and there are few effective treatment options. Treatment options for abdominal lipohypertrophy, at present, are limited. Visceral fat accumulation is not amenable to surgical removal or liposuction. As noted above, lipohypertrophy does not consistently resolve or improve with a change in HIV antiviral therapy. Dietary and lifestyle modifications may be beneficial,37 as may be treatment with metformin,38,39 but these are inconsistent and modest in effectiveness at best. There are no therapies to offer besides tesamorelin. The primary goal of treatment with tesamorelin would be to improve quality of life through improving self-image, and therefore would be patient-guided. Although there is evidence of a reduction in triglyceride and glucose levels among responders to tesamorelin,25 there is no evidence that treatment of abdominal lipohypertrophy with tesamorelin is associated with reduced cardiovascular risk, and therefore it is unlikely that tesamorelin would be used to treat the metabolic changes associated with HIV lipodystrophy.
Barriers to treatment with tesamorelin include its formulation as an injected therapy administered daily, the potential for local and systemic side effects, and the need for VAT assessment using a CT scan, which is not standard medical care in Canada.