Reimbursement Review

Metreleptin (Myalepta)

Sponsor: Medison Pharma Canada Inc.

Therapeutic area: Leptin deficiency in lipodystrophy

This multi-part report includes:

Clinical Review

Pharmacoeconomic Review

Ethics Review

Clinical Review

Executive Summary

An overview of the submission details for the drug under review is provided in Table 1.

Table 1: Background Information of Application Submitted for Review

LD = lipodystrophy; NOC = Notice of Compliance; PL = partial lipodystrophy.

Introduction

Lipodystrophy is a rare, progressive, chronic, and life-threatening disease characterized by selective absence of adipose tissue. Generalized lipodystrophy (GL) and partial lipodystrophy (PL) encompass a heterogeneous group of disorders featuring complete or partial loss of adipose tissue; these disorders may be congenital (congenital GL [CGL] or familial PL [FPLD]) or acquired (acquired GL [AGL] or acquired PL [APL]).¹ The lack of adipose tissue is also associated with leptin deficiency, which results in the early development of serious metabolic disorders such as severe insulin-resistant diabetes and hypertriglyceridemia.^2,3 Complications of lipodystrophy also frequently include multiorgan damage that may become irreversible, affecting organs such as the liver, kidneys, and pancreas.^4-11

In addition to the clinical burden, lipodystrophy also has a major detrimental emotional, psychological, and physical burden on patients, reducing life expectancy and health-related quality of life (HRQoL), and compromising the ability to carry out even basic daily activities.^12-18 People with lipodystrophy often experience insatiable hunger and hyperphagia, which causes distress to them and caregivers, including those who need to ensure that children with lipodystrophy do not eat inedible objects.^1,19 The impact of lipodystrophy also leads to a high direct and indirect economic burden.^11,15,18

The lack of precise diagnostic criteria for lipodystrophy makes it hard to firmly establish the diagnosis of lipodystrophy; overestimation or underestimation of disease prevalence is likely.^20,21 The prevalence of GL has been estimated to be 0.23 cases to 0.96 cases per million and the prevalence of PL has been estimated to be 1.67 cases to 2.84 cases per million.¹⁰ There are no epidemiological studies of lipodystrophy in Canada; however, it is estimated that there are fewer than 30 GL cases and fewer than 200 PL cases in Canada.

Metreleptin mimics the physiological effects of leptin by binding to and activating the human leptin receptor, which belongs to the Class I cytokine family of receptors that signal through the Janus kinase (JAK)/signal transducer and activator of transcription (STAT) transduction pathway.²² Metreleptin is indicated as an adjunct to diet as a replacement therapy to treat the complications of leptin deficiency experienced by patients with lipodystrophy:

• with confirmed CGL (Berardinelli-Seip syndrome) or AGL (Lawrence syndrome) in adults and children aged 2 years and older

• with confirmed FPLD or APL (Barraquer-Simons syndrome), in adults and children aged 12 years and older with persistent significant metabolic disease for whom standard treatments have failed to achieve adequate metabolic control.

The sponsor reimbursement request is consistent with the approved Health Canada indication. Metreleptin is administered once daily as a subcutaneous injection. The recommended daily dose is based on body weight. Based on clinical response (e.g., inadequate metabolic control) or other considerations (e.g., tolerability issues, excessive weight loss, especially in pediatric patients), the dose may be adjusted.²²

Stakeholder Perspectives

The information in this section is a summary of input provided by the patient and clinician groups who responded to CADTH’s call for input and from clinical experts consulted by CADTH for the purpose of this review.

Patient Input

One patient group, Lipodystrophy Canada, responded to CADTH’s call for input for the current review of metreleptin as an adjunct to diet as a replacement therapy to treat the complications of leptin deficiency in patients with lipodystrophy. Information for this input was gathered from 2 patients with PL, 1 in Canada (Patient 1) and 1 in the UK (Patient 2).

According to both patients, lipodystrophy tremendously affects their physical and mental health and every aspect of their life. Patients experience hormonal imbalance, insulin resistance, diabetes, uncontrolled hunger, hypertriglyceridemia, hypertension, body image issues, low self-esteem, and fatigue.

According to the patient input, symptoms associated with the disease affect school life and social relationships and contribute to bullying based on their masculine appearance, which increases their symptoms of depression. Patients noted that disease symptoms and management affect their everyday activities and HRQoL.

Both patients manage their disease by addressing comorbid conditions, and they agreed that the currently available treatments are not ideal, with no treatment available that directly targets lipodystrophy. The 2 patients with prior metreleptin experience reported significant improvements in their disease symptoms and quality of life.

Clinician Input

Input From Clinical Experts Consulted by CADTH

The information in this section is based on input received from a panel of 4 clinical specialists consulted by CADTH for the purpose of this review.

The clinical experts explained that there is an unmet need for effective therapies that control metabolic parameters for patients with GL and an unmet need for effective therapies that control metabolic parameters for patients with PL whose metabolic parameters are not controlled with current standard-of-care therapies. The experts expect metreleptin to become first-line therapy for patients with GL and to be used to treat PL in those patients with metabolic parameters that are not controlled with current standard-of-care therapies. The clinical experts noted that while genetic testing can be helpful to confirm a diagnosis of familial GL, often there is not a perfect correlation between a true positive in terms of genetic testing and the clinical presentation of GL. As such, the clinical experts did not consider that a confirmed genetic test result should be required before initiating therapy for this patient population. To identify patients with PL that would be suitable for treatment with metreleptin, the clinical experts suggested that elevated hemoglobin A1C and triglyceride levels are an adequate substitute given the impracticalities of measuring leptin levels directly. The clinical experts noted that the levels used in the submitted pivotal trial to define severe PL (baseline hemoglobin A1C ≥ 6.5% and/or triglycerides ≥ 5.65 mmol/L) would be appropriate criteria for identifying patients with PL who have uncontrolled disease while receiving standard-of-care therapies. The clinical experts noted that to assess response to metreleptin for patients with lipodystrophy, hemoglobin A1C and triglyceride levels would be monitored to determine whether metabolic control has been improved. The experts suggested that determining a clinically meaningful response would be context-dependent on a number of factors including the baseline hemoglobin A1C and triglyceride levels, as well as the background therapies that the patient was receiving at the time of metreleptin treatment initiation. The clinical experts suggested that the prescribing of metreleptin should be done by an endocrinologist or a pediatric endocrinologist.

Clinician Group Input

One clinician group responded to CADTH’s call for input by a group of endocrinologists, medical geneticists, lipidologists, and internal medicine specialists. Information for this input was gathered mainly through the clinical registries of patients in Canada with various forms of lipodystrophies.

The clinician group indicated that the current treatment paradigm for lipodystrophy, which does not target the underlying pathophysiology, consists of supportive care for comorbid conditions or complications. This includes diet, exercise, and antidyslipidemic and antihyperglycemic medications.

The clinician group stated that there are significant unmet therapeutic needs for patients living with lipodystrophy, as there is no cure for this disease, and available treatments address the associated metabolic complications. Conventional therapies are considered inadequate due to the severity of metabolic abnormalities experienced by patients with GL or with the more severe forms of PL, increasing their risk of end-organ damage and early death. Therefore, there is a need for a therapy that aims at correcting the underlying pathophysiology of leptin deficiency.

The clinician group noted that metreleptin can ameliorate hyperphagia and improve hepatic and peripheral insulin sensitivity, and has an established benefit versus risk profile. According to the clinician group, metreleptin is the primary first-line therapy for people with GL, including children, and for people with PL with more severe metabolic diseases and/or who do not respond well to standard treatment approaches.

The clinician group indicated that the outcomes of interest in assessing clinical response are changes in metabolic control. If clinical response is not seen after 6 months of treatment and the patient adheres to the administration technique, is receiving the correct dose, and adheres to diet, a dose increase should be considered before stopping treatment.

Drug Program Input

The drug programs identified the following jurisdictional implementation issues: relevant comparators, considerations for initiation of therapy, considerations for continuation or renewal of therapy, considerations for prescribing of therapy, generalizability, care provision issues, and system and economic issues. Refer to Table 6 for more details.

Clinical Evidence

Systematic Review

Description of Studies

The National Institutes of Health (NIH) 991265/20010769 study was a phase II/III, open-label, single-arm, single-centre, investigator-sponsored study. Study 991265 was a pilot, dose-escalation study; its objectives were to determine if metreleptin can be safely administered to a group of participants with clinically significant lipodystrophy and to determine if metreleptin treatment would be effective in lowering plasma glucose and lipid abnormalities experienced by people with clinically significant lipodystrophy. Study 20010769 was a long-term study conducted to determine the long-term safety and efficacy of metreleptin treatment for patients with lipodystrophy. Participant enrolment occurred between July 24, 2000, and March 26, 2014; the data cut-off date was in December 2014. Study 20010769 allowed for the rollover of participants from the pilot study, as well as for direct enrolment of new participants. A total of 107 participants were enrolled in the studies, which were conducted at the NIH, but also enrolled participants from countries outside the US, including Canada. Nine participants enrolled in the pilot Study 991265; of these, 8 rolled over to receive metreleptin in Study 20010769, into which 98 participants enrolled directly. A total of 66 of the 107 participants had GL and 41 had PL. There were 31 participants in a specified PL subgroup, that is, those patients with PL with baseline hemoglobin A1C greater than or equal to 6.5% and/or triglycerides greater than or equal to 5.65 mmol/L.

Actual change from baseline in hemoglobin A1C and percent change from baseline in fasting triglyceride levels to month 12 were the co-primary efficacy end points. The clinical experts considered month 12 an appropriate time point for analysis as the effects of metreleptin could be expected to be seen by this time. The sponsor noted that this 12-month period would allow for individual dose titrations to achieve maximum effects in given patients; 12 months was also an acceptable length of time over which to assess the clinical impact of the treatment. To account for patients who may have discontinued treatment before month 12, last observation carried forward (LOCF) methods were used to determine changes from baseline to month 12. Specifically, hemoglobin A1C and triglyceride samples obtained on or after day 180 were used in the analysis for patients without samples obtained within the month 12 window (day 365 ± 65 days).

Efficacy Results

Change From Baseline in Hemoglobin A1C at 12 Months

In the GL cohort, mean hemoglobin A1C was 8.6% (standard deviation [SD] = 2.33) at baseline and 6.4% (SD = 1.68) at month 12, for a mean change from baseline of –2.2% (95% confidence interval [CI], –2.7% to –1.6%). In the overall PL cohort, mean hemoglobin A1C was 7.9% (SD = ||||%) at baseline and 7.4% (SD = ||||%) at month 12, for a mean change from baseline of –0.6% (95% CI, |||| || ||||). In the specified PL subgroup, mean hemoglobin A1C was 8.7% (SD = ||||%) at baseline and 7.9% (SD = ||||%) at month 12, for a mean change from baseline of –0.9% (95% CI, |||| || ||||).

Change From Baseline in Fasting Triglycerides at 12 Months

In the GL cohort, mean triglyceride level was 14.7 mmol/L (SD = 25.66 mmol/L) at baseline and 4.5 mmol/L (SD = 6.10 mmol/L) at month 12, for a relative mean change from baseline of –32.1% (95% CI, –51.0% to –13.2%). In the overall PL cohort, mean triglyceride level were |||| |||||| ||||||| at baseline and ||| |||||| at month 12, for a relative mean change from baseline of ||||| |||| ||| ||||| || |||||. In the specified PL subgroup, mean triglyceride level was |||| |||||| ||||||| and at month 12 were ||| |||||| |||||||, for a mean change from baseline of |||| |||| ||| ||||| || |||||.

The sponsor, |||||| ||| |||| ||| || || ||| ||||||| ||||||||, conducted an ad hoc sensitivity analysis, removing 1 patient in the PL cohort who was recorded as a patient with noncompliance behaviour. The results of this ad hoc analysis showed a mean change from baseline in triglyceride levels of –20.8% (95% CI, ||||| || ||||) in the PL cohort and –37.4% (95% CI, ||||| || |||||) in the specified PL subgroup.

Change From Baseline in Fasting Glucose at 12 Months

In the GL cohort, mean glucose level was 10.2 mmol/L (SD = 5.05 mmol/L) at baseline and 7.0 mmol/L (SD = 3.40 mmol/L) at month 12, for a relative mean change from baseline of –19.7% (95% CI, –29.4% to –10.0%). In the overall PL cohort, mean glucose level was 8.8 mmol/L (SD = 4.39 mmol/L) at baseline and 7.5 mmol/L (SD = 3.28 mmol/L) at month 12, for a relative mean change from baseline of –6.1% (95% CI, –16.0% to 3.8%). In the specified PL subgroup, mean glucose level was 10.0 mmol/L (SD = |||| mmol/L) at baseline and 8.1 mmol/L (SD = |||| mmol/L) at month 12, for a relative mean change from baseline of –13.2% (95% CI, –24.4% to –1.9%).

Change From Baseline in Liver Volume at 12 Months

In the GL cohort (N = 21), mean baseline liver volume was 3,357.7 mL (SD = 1,121.74 mL); the relative mean change from baseline was –33.8% (SD = 14.78%). In the overall PL cohort (N = 9), mean baseline liver volume was 2,624.6 mL (SD = 936.21 mL); the relative mean change from baseline was –13.4% (SD = ||||). In the specified PL subgroup (N = 8), mean baseline liver volume was 2,411.7 mL (SD = 731.91 mL), the relative mean change from baseline was –12.4% (SD = ||||).

Harms Results

Treatment-emergent adverse events (TEAEs) occurred in 89.4% of patients in the GL cohort of the safety analysis set and 85.4% of patients in the PL cohort of the safety analysis set. The most common adverse events (AEs) in the GL cohort were weight decrease (25.8%), abdominal pain (16.7%), and hypoglycemia (15.2%). The most common AEs in the PL cohort were hypoglycemia (17.1%), abdominal pain (14.6%), and nausea (14.6%).

Serious adverse events (SAEs) occurred in 34.8% of patients in the GL cohort, |||| ||||||||| |||| |||||| ||| |||||||||||| |||||| ||| |||| ||||||. SAEs occurred in 24.4% of the PL cohort, |||| ||||||||| |||| ||||||| |||||||||||| ||||||| ||| |||||||||| |||||| ||| |||| ||||||.

TEAEs that resulted in treatment discontinuation occurred among 5 patients (7.6%) in the GL cohort and 1 patient (2.4%) in the PL cohort.

Death occurred in 4.5% of the patients in the GL cohort, including due to renal failure, cardiac arrest, and chronic hepatic failure. Death occurred in 2.4% of the patients in the PL cohort, including due to hypoxic-ischemic encephalopathy.

Critical Appraisal

The major limitations associated with the NIH 991265/20010769 study include the single-arm, open-label design of the study. The lack of comparative data is a key limitation to the interpretation of the results from a single-arm trial, as it is difficult to distinguish between the effect of the intervention relative to that of a placebo effect, or the effect of natural history. It is acknowledged that there may be practical limitations to conducting a randomized controlled trial (RCT) with patients with lipodystrophy due to the rarity of the condition. The open-label nature of the trial also potentially increases the risk of bias, however the end points included are objective laboratory values and therefore are unlikely to have been influenced by this bias. Harms outcomes however may be impacted by the open-label design of the study.

The NIH 991265/20010769 study had a large number of drop-outs and missing data at the 12-month primary analysis because of the challenges in conducting a clinical study that included international participants at the NIH. LOCF methods were used to carry forward the results from 6 months onward. Patients who did not have an observation after 6 months from baseline were considered to have missing data and were not included in the analysis. Excluding patients with final observations before 6 months violates intent-to-treat principles as not all randomized patients were included in the primary analysis. In addition, this imputation may underestimate the variance in the results, potentially resulting in narrower CIs. Also, interim analyses were conducted without adjusting for multiplicity to account for the increased risk of type I error. The co-primary end points did not require multiplicity adjustment because both end points needed to achieve statistical significance to be considered a positive result. However, statistical significance was only achieved for the PL cohort on removal from the analysis of a patient with noncompliance behaviour.

The NIH 991265/20010769 study began to enrol patients in July 2000. As this was 23 years before the writing of this report, the clinical experts suggested that standards of therapy and patient support may have evolved since then. However, the clinical benefit of metreleptin is anticipated to be consistent with that observed in the NIH 991265/20010769 study. Lipodystrophy is a chronic disease and patients would be expected to receive treatment for life. The generalizability of the results beyond the maximum 14-year follow-up of the NIH 991265/20010769 study is unknown, although the clinical experts consulted by CADTH did not expect the efficacy of metreleptin to change beyond the time horizon of the NIH 991265/20010769 study. The clinical experts consulted considered the patient characteristics from the NIH 991265/20010769 study to be broadly generalizable to that of the expected patient population in Canada.

GRADE Summary of Findings and Certainty of the Evidence

The selection of outcomes for Grading of Recommendations Assessment, Development and Evaluation (GRADE) assessment was based on the sponsor’s Summary of Clinical Evidence, consultation with clinical experts, and input received from patient and clinician groups and public drug plans. The following list of outcomes was finalized in consultation with expert committee members:

• change from baseline to month 12 in hemoglobin A1C

• change from baseline to month 12 in fasting triglycerides

• change from baseline to month 12 in fasting glucose

• change from baseline to month 12 in liver volume.

Table 2: Summary of Findings for Metreleptin Treatment for Patients With Leptin Deficiency in GL (NIH 991265/20010769 Study)

CFB = change from baseline; CI = confidence interval; GL = generalized lipodystrophy; NIH = National Institutes of Health; SAE = serious adverse event; SD = standard deviation.

Note: Study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias were considered when assessing the certainty of the evidence. All serious concerns in these domains that led to the rating down of the level of certainty are documented in the table footnotes. For single-arm trials, all serious concerns with study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias are documented in the table footnotes.

^aIn the absence of a comparator group, conclusions about efficacy relative to any comparator cannot be drawn and the certainty of evidence starts at very low and cannot be rated up.

^bRated down 2 levels for very serious imprecision due to the small sample size. Rated down 1 level for a high amount of missing data requiring imputation.

^cRated down 1 level for serious risk of bias due to potential for bias in favour of metreleptin arising from the open-label nature of the study.

Table 3: Summary of Findings for Metreleptin Treatment of Patients With Leptin Deficiency in PL (NIH 991265/20010769 Study)

CFB = change from baseline; CI = confidence interval; NIH = National Institutes of Health; PL = partial lipodystrophy; SAE = serious adverse event; SD = standard deviation.

Note: Study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias were considered when assessing the certainty of the evidence. All serious concerns in these domains that led to the rating down of the level of certainty are documented in the table footnotes. For single-arm trials, all serious concerns with study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias are documented in the table footnotes.

^aIn the absence of a comparator group, conclusions about efficacy relative to any comparator cannot be drawn and the certainty of evidence starts at very low and cannot be rated up.

^bRated down 2 levels for very serious imprecision due to the small sample size. Rated down 1 level for a high amount of missing data requiring imputation.

^cRated down 1 level for serious risk of bias due to potential for bias in favour of metreleptin arising from the open-label nature of the study.

Indirect Comparisons

Description of Studies

• One unpublished supportive analysis was conducted to estimate the comparative treatment effect of metreleptin with or without supportive care compared to supportive care alone using an inverse probability weighting (IPW) and multivariate regression methods to adjust for differences between patients from the NIH follow-up study and an observational study of patients with GL and PL receiving supportive care alone.²³

• One published supportive analysis by Cook et al. (2021)⁴⁴ estimated the treatment effect of metreleptin on mortality of patients with GL or PL using a Cox proportional hazards model to control for differences between patients who have received metreleptin treatment and the historical cohort of patients with no experience with metreleptin treatment.⁶

Efficacy Results

In the unpublished supportive analysis, after IPW, the mean difference between metreleptin with or without supportive care compared to supportive care alone in hemoglobin A1C levels was |||||| |||| ||| ||||| || |||||, the mean difference in triglyceride levels was ||||| |||||| |||| ||| |||||| || ||||| and the hazard ratio (HR) for all-cause mortality was |||| |||| ||| |||| || |||||.

In the Cook et al. (2021) supportive analysis, the Cox proportional hazards model–predicted mortality HR for the overall metreleptin-treated cohort versus the matched metreleptin-naive cohort was 0.35 (95% CI, 0.13 to 0.90). Statistically significant differences in mortality risk between patients who received metreleptin treatment and patients with no experience of metreleptin in the GL subgroup were not detected from the Cox proportional hazards model (HR = 0.455; 95% CI, 0.150 to 1.387).

Critical Appraisal

The unpublished supportive analysis was associated with major limitations relating to the use of retrospective chart reviews and missing data, inability to adjust for important prognostic covariates and small sample sizes resulting in imprecise and wide 95% CIs.

The published Cook et al. (2021) historical control arm analysis utilized a more robust methodology for adjusting the patient populations on important prognostic factors (though still not capturing all important factors). However, mortality was the only end point assessed, and the few events captured resulted in imprecise and wide 95% CIs. The Cook et al. (2021) analysis also had missing data because details of the standard-of-care therapies used in the historical control arm were not available.

Studies Addressing Gaps in the Evidence From the Systematic Review

Study FHA101 was a single-arm, multicentre, open-label, expanded-access study conducted at multiple treatment centres in the US with patients with lipodystrophy. The primary objective was to provide metreleptin, an investigational medication, under a treatment protocol to patients with lipodystrophy that is associated with diabetes mellitus and/or hypertriglyceridemia. A secondary objective was to assess the long-term efficacy, safety, and tolerability of metreleptin among people with diabetes mellitus and/or hypertriglyceridemia. Patient enrolment occurred between March 30, 2009, and January 23, 2016. A total of 41 patients were enrolled across 6 centres in the US.

Efficacy Results

Study FHA101 found that treatment with metreleptin led to sustained improvements in glycemic control and hypertriglyceridemia in this small group of patients, both with GL and in the PL subgroup. Among the 9 patients with GL included in the full analysis set (FAS), mean hemoglobin A1C was reduced from 7.7% at baseline (n = 9) to 6.2% at month 12 using LOCF (n = 5), a mean change of –1.2%. Results were similar for the 7 patients in the PL subgroup included in the FAS; treatment with metreleptin led to reductions in hemoglobin A1C from 7.8% at baseline (n = 7) to 7.0% at month 12 using LOCF (n = 7), a mean change of –0.8%.

Mean fasting glucose levels were reduced from 11.4 mmol/L at baseline (n = 9) to 10.2 mmol/L at month 12 using LOCF (n = 6) in the GL group, a mean change of –1.5 mmol/L, representing a 7.3% decrease in fasting glucose levels. For the PL subgroup, mean fasting glucose levels were reduced from 8.0 mmol/L at baseline (n = 7) to 6.9 mmol/L at month 12 using LOCF (n = 7), a mean change of –1.1 mmol/L, representing a 9% decrease from baseline.

Mean fasting triglyceride concentrations were reduced from 19.9 mmol/L at baseline (n = 8) to 7.6 mmol/L at month 12 using LOCF (n = 6) in the GL group, corresponding to a mean percent change of –26.9%. In the PL subgroup, mean fasting triglyceride concentrations, which were lower in this group of patients compared to those with GL, were reduced from 4.0 mmol/L (n = 7) at baseline to 3.6 mmol/L at month 12 using LOCF (n = 7), a mean change of –8.5%.

Harms Results

Treatment with metreleptin was safe and generally well tolerated by patients with GL and by patients in the PL subgroup. The most common TEAEs among patients in the GL group were hypoglycemia, infections, abdominal pain and increased liver function tests. Most TEAEs were mild to moderate in severity. The AE profile of patients in the PL subgroup was generally similar to that of patients with GL. The most common TEAEs among patients in the PL subgroup were hypoglycemia, urinary tract infection, upper respiratory tract infection, anxiety, nausea, and sinusitis. Two patients were reported to have neoplasms, but the investigator considered that these were unrelated to metreleptin. A total of 3 patients, including 1 with GL, 1 in the PL subgroup, and 1 with PL (not in the subgroup), developed neutralizing antibodies.

During the 5-year study, 2 deaths were reported, neither of which was assessed as drug-related.

Critical Appraisal

The open-label design of Study FHA101 is considered a limitation that could bias the results parameters. The lack of a control arm is considered a key constraint that limits the interpretation of study outcomes. A small number of patients with GL and PL were evaluated; therefore, observed results should be interpreted with caution.

External Validity

As there were no study sites in Canada, there may be limitations in generalizing these findings to the Canadian context.

Conclusions

Lipodystrophy is a rare disease and no currently available therapies directly target the underlying disease pathology. There is an unmet need for effective therapies that control metabolic parameters for patients with GL and an unmet need for effective therapies that control metabolic parameters for patients with PL who are unable to achieve metabolic control with current standard-of-care therapies. Metreleptin is a first-in-class treatment that replaces leptin and directly targets the underlying leptin deficiency experienced by patients with lipodystrophy. The NIH 991265/20010769 study demonstrated improvements from baseline in hemoglobin A1C and triglyceride levels although the evidence is very uncertain regarding the effects of metreleptin on metabolic parameters, mainly due to the lack of a comparator arm. Evidence gaps from the NIH 991265/20010769 study include the effect of metreleptin treatment on outcomes identified by patients as important, such as impact on hunger, HRQoL, and fertility; as such, the impact of metreleptin on these outcomes is uncertain. Treatment with metreleptin was well tolerated over the study period and the safety profile was as expected, according to clinical experts.

There were important technical limitations in the conduct of the historical control arm comparisons, including not adjusting for important prognostic covariates, missing data, and unclear standard of therapies used for the historical control arm. As such, the historical control arm analyses were inconclusive and imprecise given the small sample sizes and wide 95% CIs.

While the results of the included studies aligned with clinical experts’ expectation that metreleptin will address the unmet needs in this patient population, there were important limitations in the included studies leading to uncertainty in the evidence due to the single-arm design of the trial and the high amount of missing data at the 12-month time point.

Introduction

The objective of this report is to review and critically appraise the evidence submitted by the sponsor on the beneficial and harmful effects of metreleptin (3 mg, 5.8 mg, and 11.3 mg, powder for solution, subcutaneous injection) for the treatment of the complications of leptin deficiency experienced by patients with lipodystrophy.

Disease Background

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the CADTH review team.

Lipodystrophy is a rare, progressive, chronic, and life-threatening disease characterized by selective absence of adipose tissue. Lipodystrophy encompasses a heterogeneous group of disorders featuring complete or partial loss of adipose tissue; these disorders may be congenital (CGL or FPLD) or acquired (AGL or APL).¹ The absence of adipose tissue in people with lipodystrophy leads to reduced storage capacity for lipids, which are therefore accumulated ectopically in other organs. The lack of adipose tissue is also associated with leptin deficiency, which results in the early development of serious metabolic disorders such as severe insulin-resistant diabetes and hypertriglyceridemia.^2,3 Complications of lipodystrophy also frequently include multiorgan damage that may become irreversible, affecting organs such as the liver, kidneys, and pancreas.^4-11 Data from a study that assessed the natural history of non–HIV-related GL and PL noted Kaplan-Meier estimates of mean time to first organ abnormality of 7.7 years (standard error [SE] = 0.9 years) in GL and 16.1 years (SE = 1.5 years) in PL. Mean time to diabetes or insulin resistance was 12.7 years (SE = 1.2 years) among patients with GL and 19.1 years (SE = 1.7 years) among patients with PL.⁷ These comorbidities of lipodystrophy ultimately lead to premature death.^1,7,24 The same systematic review of patients with non–HIV-related lipodystrophy indicated that the mean time to death was 51.2 years (SE = 3.5 years) for patients with GL and 66.6 years (SE = 1.0 years) for patients with PL.⁷ However, the mean age at mortality varied between studies. In another systematic review of people with non–HIV-related lipodystrophy, the mean age at mortality was 12.5 years and 32.2 years for CGL and AGL, respectively, and 27.8 years and 22.7 years for FPLD and APL, respectively.²⁵

In addition to the clinical burden, lipodystrophy has a major detrimental emotional, psychological, and physical burden on patients, reducing life expectancy and HRQoL, and compromising the ability to carry out even basic daily activities.^12-18 Caregivers (and family members) often experience many of the same complications as patients with lipodystrophy, from deterioration of mental health to a reduced quality of day-to-day life.^15,26 The emotional, psychological, and physical (body image) burden faced by younger pediatric or adolescent patients as well as their caregivers is considerable on top of the earlier onset of the complications.^1,15-18 People with lipodystrophy often experience insatiable hunger and hyperphagia, which causes distress to them and their caregivers, not least those who must ensure that children in their care do not eat inedible objects.^1,19 Lipodystrophy also results in a high direct economic burden due health care resource utilization (physician visits, hospitalizations, medications, and therapy for mental health) as well as an indirect economic burden from lost opportunities, reduced household income, time burden, and social stigma.^11,15-18

Estimated Disease Prevalence

Awareness of GL and PL is low due to its rarity, making it difficult to accurately estimate its prevalence or incidence.¹⁰ The overlap of clinical presentation with more common diseases may further complicate the diagnosis of lipodystrophy. Moreover, the lack of precise diagnostic criteria for lipodystrophy makes it hard to firmly establish the diagnosis; overestimation or underestimation of disease prevalence is likely.^20,21 The prevalence of GL has been estimated to be 0.23 cases to 0.96 cases per million and the prevalence of PL has been estimated to be 1.67 cases to 2.84 cases per million.¹⁰ There are no epidemiological studies of lipodystrophy in Canada; however, it is estimated that there are fewer than 30 cases of GL and fewer than 200 cases of PL in Canada.

Limited published international and European estimates of the 4 major subtypes of lipodystrophy indicate that prevalence ranges from 0.1 cases to 90 cases per million, depending on the information source and the methodology used for computation of the estimate.¹⁰ Chiquette et al. attempted to quantitatively estimate prevalence of lipodystrophy by conducting electronic medical record and literature searches to identify cases.¹⁰ The prevalence range of all lipodystrophies across electronic medical record databases was reported to be 1.3 cases to 4.7 cases per million. For the quintiles search, the estimated prevalence of diagnosed lipodystrophy was 0.23 cases per million for GL and 2.84 cases per million for PL. For all literature searches, the prevalence of all lipodystrophy in Europe was 2.6 cases per million (0.96 cases per million for GL and 1.67 cases per million for PL). Other studies have reported higher prevalence estimates of lipodystrophy at 3.23 cases to 4.7 cases per 100,000.^27,28

New cases of lipodystrophy are identified mostly by diabetes specialists, endocrinologists, and lipid specialists. However, due to the rarity of this disease, many clinicians are unfamiliar with diagnosis and management, and diagnosis can take many years. In a chart review, Akinci et al. reported that the average time from first symptoms to diagnosis was 3.1 years for patients with GL and 9.0 years for patients with PL.⁷

Although experienced lipodystrophy specialists can easily recognize and diagnose lipodystrophy, no firm objective diagnostic criteria have been established for less experienced clinicians.¹ Multiple societies, including a 17-member committee of nominees from worldwide endocrine societies, have attempted to develop consensus recommendations for the detection of lipodystrophy.^1,29-31 In these multisociety practice guidelines, published in 2016, Brown et al. recommended that diagnosis be initially based on clinical history, physical examination, body composition, and metabolic status.¹ Confirmatory genetic testing is helpful in suspected familial lipodystrophy and should be considered in at-risk family members.¹ For patients in whom lipodystrophy is suspected, Brown et al. recommended screening for comorbidities associated with the disease including diabetes, dyslipidemia, nonalcoholic fatty liver disease (NAFLD), nonalcoholic steatohepatitis (NASH), and cardiovascular and reproductive dysfunction.¹ These recommendations are similar to those of other guidelines that state that clinical characteristics and comorbid conditions should be the basis for referral to a lipodystrophy specialist.²⁹

Methods to identify suspected lipodystrophy are widely available and conducted across Canada; these include clinical history, physical examination, body composition, and metabolic status. Similarly, screening for comorbidities associated with lipodystrophy, including diabetes, dyslipidemia, NAFLD, NASH, and cardiovascular and reproductive dysfunction, are also widely available and conducted across Canada. Confirmatory genetic testing for suspected familial lipodystrophy may not be readily available across the country.

Standards of Therapy

Contents within this section have been informed by materials submitted by the sponsor and clinical expert input. The following has been summarized and validated by the CADTH review team.

There are no approved drugs for the treatment of lipodystrophy in Canada;³² patients currently receive supportive care for comorbid conditions or complications of lipodystrophy (i.e., diet and exercise, and antihyperglycemic and lipid-lowering medications).^29,31 However, these supportive care treatments do not address the underlying mechanism of the disease, i.e., the lack of adipose tissue and resulting leptin deficiency, and patients require numerous medications to manage comorbid conditions of lipodystrophy. The American Association of Clinical Endocrinologists²⁹ and the Japan Endocrine Society³¹ both suggest diet and exercise as options for the metabolic management of lipodystrophy alongside antihyperglycemic and lipid-lowering medications. Metformin, sulfonylureas, thiazolidinediones, and insulin can be used to manage hyperglycemia, while fibrates and statins can be used to manage hypertriglyceridemia. They acknowledge, however, that when the complications associated with lipodystrophy are severe, the aforementioned supportive care treatments (alone or in combination) are likely to be inadequate at establishing metabolic control as they do not address the underlying leptin deficiency.

The clinical experts consulted agreed that standard of care includes a variety of supportive therapies aimed at achieving metabolic control to reduce comorbidities. The clinical experts suggested that standard-of-care therapies can have success in some patients with PL, but that patients with GL do not experience the same level of metabolic control while receiving standard-of-care therapies. Of note, the Ethics Review Report uses the term “metabolic management” in place of “metabolic control” in alignment with recommendations from patient groups regarding the use of language in the context of treating people with metabolic conditions.

Drug Under Review

Metreleptin mimics the physiological effects of leptin by binding to and activating the human leptin receptor, which belongs to the Class I cytokine family of receptors that signal through the JAK/STAT transduction pathway.²² Metreleptin is indicated as an adjunct to diet as a replacement therapy to treat the complications of leptin deficiency experienced by patients with lipodystrophy:

• with confirmed CGL (Berardinelli-Seip syndrome) or AGL (Lawrence syndrome) in adults and children aged 2 years and older

• with confirmed FPLD or APL (Barraquer-Simons syndrome), in adults and children aged 12 years and older with persistent significant metabolic disease for whom standard treatments have failed to achieve adequate metabolic control.

Metreleptin is contraindicated for patients with general obesity not associated with confirmed generalized leptin deficiency or confirmed PL. Metreleptin is also contraindicated for patients with HIV-related lipodystrophy.

The sponsor reimbursement request is consistent with the approved Health Canada indication. Metreleptin is administered once daily as a subcutaneous injection. The recommended daily dose is based on body weight (refer to Table 4).²² Based on clinical response (e.g., inadequate metabolic control) or other considerations (e.g., tolerability issues, excessive weight loss, especially in pediatric patients), the dose may be adjusted.²²

Table 4: Recommended Dose of Metreleptin

Source: Metreleptin draft product monograph.²²

Adipocytes store lipids to meet the fuel requirements of nonadipose tissues during fasting. For patients with lipodystrophy, the deficiency of adipose tissue leads to hypertriglyceridemia and ectopic deposition of fat in nonadipose tissues such as liver and muscle, contributing to metabolic abnormalities including insulin resistance. Native leptin is a hormone predominantly secreted by adipose tissues that informs the central nervous system of the status of energy stores in the body. In people with lipodystrophy, leptin deficiency, resulting from the loss of adipose tissue, contributes to excess caloric intake, which exacerbates the metabolic abnormalities, though leptin likely has effects on insulin sensitivity independent of food intake.³³ Metreleptin mimics the physiological effects of leptin by binding to and activating the human leptin receptor, which belongs to the Class I cytokine family of receptors that signals through the JAK/STAT transduction pathway.²²

In the absence of treatments that directly address leptin deficiency experienced by patients with lipodystrophy, patients often receive supportive care to manage comorbid conditions or complications of lipodystrophy. Although supportive care treatments are not indicated for lipodystrophy, they are indicated for comorbid conditions that are commonly diagnosed in patients with lipodystrophy (i.e., diabetes, hypertriglyceridemia, hypertension); as such, they are included in the basket of medications currently used to manage complications of lipodystrophy. Key characteristics of metreleptin are summarized in Table 5.

Table 5: Key Characteristics of Metreleptin

APL = acquired partial lipodystrophy; CGL = congenital generalized lipodystrophy; FPLD = familial partial lipodystrophy; JAK = Janus kinase; STAT = signal transducer and activator of transcription.

Source: Metreleptin Product Monograph.²²

Stakeholder Perspectives

Patient Group Input

This section was prepared by the CADTH review team based on the input provided by patient groups. The full original patient input received by CADTH has been included in the Stakeholder section of this report.

One patient group, the Lipodystrophy Canada, responded to CADTH’s call for input for the current review of metreleptin as an adjunct to diet as a replacement therapy to treat the complications of leptin deficiency experienced by patients with lipodystrophy. The Lipodystrophy Canada is a not-for-profit foundation with the core mission of providing support and resources to patients with lipodystrophy and their caregivers. Information for this input was gathered from 2 patients from Canada (patient 1) and the UK (patient 2) with PL.

According to both patients, lipodystrophy affects their physical and mental health and every other aspect of their life tremendously. They experience hormonal imbalance, insulin resistance, diabetes, uncontrolled hunger, hypertriglyceridemia, hypertension, body image issues, low self-esteem, and fatigue. In addition, patient 1, whose sibling died after severe insulin resistance damaged their kidneys, experienced hirsutism, acanthosis nigricans, myocardial infarction, and heart arrhythmias that necessitated surgery to implant a cardiac device.

According to the patient input, symptoms associated with the disease affect school life and social relationships and contribute to bullying due to their masculine appearance, which increases their symptoms of depression. The patients noted that disease symptoms and constant medical appointments, medication intake, very high doses of insulin, and check-ups affect their everyday activities and HRQoL.

Both patients manage their disease by addressing comorbid conditions such as hyperglycemia, insulin resistance, hypertriglyceridemia, and hypertension. Both patients agreed that the treatments that are currently available are not ideal, as conventional treatments keep all health parameters at suboptimal levels and no available treatment directly targets lipodystrophy.

The 2 patients with prior metreleptin experience reported significant improvements in their disease symptoms and quality of life. They stated that the drug improved fatty liver symptoms, triglycerides, and hypertension symptoms; regulated their satiety; and improved their overall health and HRQoL. Metreleptin allowed patient 1 to cease insulin therapy and patient 2 to reduce insulin intake.

Clinician Input

Input From Clinical Experts Consulted by CADTH

All CADTH review teams include at least 1 clinical specialist with expertise on the diagnosis and management of the condition for which the drug is indicated. Clinical experts are a critical part of the review team and are involved in all phases of the review process (e.g., providing guidance on the development of the review protocol, assisting in the critical appraisal of clinical evidence, interpreting the clinical relevance of the results, and providing guidance on the potential place in therapy of the drug). In addition, as part of the review of metreleptin, a panel of 4 clinical experts from across Canada was convened to characterize unmet therapeutic needs, assist in identifying and communicating situations where there are gaps in the evidence that could be addressed through the collection of additional data, promote the early identification of potential implementation challenges, gain further insight into the clinical management of a condition, and explore the potential place in therapy of the drug (e.g., potential reimbursement conditions). A summary of this panel discussion follows.

Unmet Needs

The clinical panel noted that for patients with GL, standard-of-care therapies show little response. As a result of the disease, patients experience metabolic abnormalities such as elevated hemoglobin A1C and triglyceride levels putting them at risk for complications such as pancreatitis, damage to the liver and other organs, or cardiovascular abnormalities. The lack of circulating leptin results in uncontrolled constant increased appetite that can be debilitating. The clinical experts noted that caregivers of children with lipodystrophy must restrict access to food in the house to keep their child from overeating. A lack of circulating leptin also results in delayed onset of puberty and a markedly changed physical appearance that can lead to adolescent patients being bullied at school. The clinical experts agreed that current treatment options do not address these unmet needs for patients with GL. The clinical experts noted that an unmet need exists for patients with PL who are unable to have their metabolic parameters adequately controlled with standard-of-care therapies.

Place in Therapy

The clinical panel noted that the place in therapy for metreleptin would be different depending on the type of lipodystrophy the patient has (i.e., GL or PL). For patients with GL, metreleptin would be used as first-line therapy given the lack of efficacy provided by current standard-of-care therapies and in keeping with the treatment philosophy for a rare proven genetic disorder that matches the pathophysiology as closely as possible. For patients with PL, a portion are able to have their metabolic parameters controlled with standard-of-care therapies without the use of metreleptin, often with low to moderate doses of insulin. However, for a subset of patients, PL cannot be managed even with high doses of insulin; it is for these patients with uncontrolled disease and very low leptin levels, that the addition of metreleptin would be considered. The clinical expert panel also noted that many patients with PL receive treatment for underlying metabolic abnormalities before they receive a diagnosis of PL. If there is documented inability to control metabolic abnormalities with standard-of-care therapies before the diagnosis of PL, then metreleptin could be considered immediately.

Patient Population

The clinical expert panel explained that lipodystrophy is generally split into 2 subtypes, GL and PL. GL is a very rare genetic disorder, often identified in a younger patient population. The clinical experts noted that while genetic testing can help confirm a diagnosis of familial GL, often there is no perfect correlation between a true positive in terms of genetic testing and the clinical presentation of GL. As such, the clinical experts did not believe a confirmed genetic test result should be required for initiation of therapy for this patient population. Patients with GL have very low levels of circulating leptin, which is the underlying cause of the multitude of comorbidities that affect this population.

The clinical expert panel noted that patients with PL are a more heterogenous population, often diagnosed later in life than patients with GL. When determining which patients with PL would be most suited for treatment with metreleptin, the clinical experts suggested those patients who have uncontrolled metabolic parameters and low levels of circulating leptin. However, the clinical experts described testing for leptin levels as an uncommon and difficult-to-access laboratory test. To identify patients with PL for whom treatment with metreleptin could be suitable, the clinical experts suggested that elevated hemoglobin A1C and triglyceride levels were an adequate substitute given the impracticalities of measuring leptin levels directly. The experts noted that the levels used in the submitted pivotal trial to define severe PL (baseline hemoglobin A1C ≥ 6.5% and/or triglycerides ≥ 5.65 mmol/L) would be an appropriate criterion for identifying patients with PL who have uncontrolled disease on standard-of-care therapies.

Assessing the Response Treatment

The clinical expert panel noted that to assess response to metreleptin for patients with lipodystrophy, hemoglobin A1C and triglyceride levels would be monitored to determine whether metabolic control has improved. The experts suggested that determining a clinically meaningful response would be context-dependent on a number of factors including the baseline hemoglobin A1C and triglyceride levels, as well as the background therapies that the patient was receiving at the time of metreleptin initiation. Other clinical measures such as change in liver volume and change in plasma glucose levels can be informative but are not as heavily relied upon as hemoglobin A1C and triglyceride levels. The clinical experts suggested that it is important for patients with lipodystrophy to have their metabolic parameters reduced to a level where they are at a reduced risk for complications such as pancreatitis.

Other parameters the clinical experts noted as important included the ability to reduce the amount of insulin required by the patient as well as to reduce other background therapies that the patient may have been receiving before metreleptin initiation. The clinical experts also noted that for patients with child-bearing potential experiencing amenorrhea, the resumption of normal menstrual cycles and fertility is a very important outcome to monitor given how meaningful it is to patients’ quality of life.

Discontinuing Treatment

The expert panel agreed that treatment with metreleptin is life-long for patients with lipodystrophy provided the patient is still experiencing metabolic control while on treatment. Patients would be monitored for continued efficacy based on hemoglobin A1C and a lipid panel. Treatment would be discontinued if the patient experienced a worsening of metabolic parameters (potentially due to neutralizing antibodies) or if there were intolerable AEs attributable to metreleptin, that is, immunological side effects. However, the clinical expert panel noted that intolerable AEs appear to be uncommon.

Prescribing Considerations

The clinical panel suggested that metreleptin should be prescribing by an endocrinologist or a pediatric endocrinologist. The panel noted that specialist involvement is particularly important because initiating metreleptin treatment can impact insulin sensitivity experienced by patients with lipodystrophy, potentially leading to hypoglycemia if the insulin dose is not prophylactically adjusted. The experts noted that with the rise of telemedicine and virtual appointments, patients with PL who live in remote areas and may not have had access to an endocrinologist can have virtual consultations. Prescribing for PL can also be virtual, in coordination with the local primary care team. The panel noted that patients with GL should continue to be seen in person in hospital as treatment often involves the coordination of multiple specialties.

Clinician Group Input

This section was prepared by the CADTH review team based on the input provided by clinician groups. The full original clinician group input received by CADTH has been included in the Stakeholder section of this report.

One clinician group responded to CADTH’s call for input from a group of endocrinologists, medical geneticists, lipidologists, and internal medicine specialists from across Canada linked by their common interest in the care of patients with rare lipodystrophies, a serious group of disorders for which there is no cure, and which can lead to severe life-threatening complications. Information for this input was gathered mainly through the clinical registries of patients in Canada with various forms of lipodystrophies.

According to the clinician group input, lipodystrophy is characterized by complete or partial loss, or absence of, subcutaneous adipose tissue, known as GL and PL, leading to accumulation of lipids in the liver, skeletal muscle, heart, and pancreas, triggering hyperphagia, and resulting in insulin resistance, diabetes, hepatic steatosis, and severe hypertriglyceridemia. At an early age, this condition can cause potentially life-threatening acute pancreatitis.

The clinician group indicated that the current treatment paradigm for lipodystrophy, which does not target the underlying pathophysiology, consists of supportive care for comorbid conditions or complications. This includes diet and exercise, statins, fibrates, and omega-3 fatty acids to control dyslipidemia, and antihyperglycemic medications such as metformin, sulfonylureas, thiazolidinediones, glucagon-like peptide-1 receptor agonists, and sodium-glucose cotransporter-2 inhibitors.

The clinician group stated that there are significant unmet therapeutic needs for patients living with lipodystrophy, as there is no cure for this disease, and available treatments address the associated metabolic complications. Conventional therapies are inadequate because of the severity of metabolic abnormalities experienced by patients with GL and more severe forms of PL, increasing their risk of end-organ damage and early death. Therefore, there is a need for a therapy that aims at correcting the underlying pathophysiology of leptin deficiency.

The clinician group noted that metreleptin, which mimics the physiological effects of leptin, can ameliorate hyperphagia and improve hepatic and peripheral insulin sensitivity, and has an established benefit versus risk profile. According to the clinician group, metreleptin is the primary first-line therapy for patients with GL, including children, for whom starting early treatment with this drug is recommended to prevent serious complications. Also, patients with PL and with more severe metabolic diseases and who do not respond well to standard treatment approaches could benefit from metreleptin.

The clinician group indicated that the outcomes of interest in assessing clinical response are changes in metabolic control, such as 0.5% hemoglobin A1C reduction and 15% reduction in serum triglycerides. If clinical response is not seen after 6 months of treatment and the patient adheres to the administration technique, is receiving the correct dose, and adheres to diet, a dose increase should be considered before stopping treatment.

Drug Program Input

The drug programs provide input on each drug being reviewed through CADTH’s Reimbursement Review processes by identifying issues that may impact their ability to implement a recommendation. The implementation questions and corresponding responses from the clinical experts consulted by CADTH are summarized in Table 6.

Table 6: Summary of Drug Plan Input and Clinical Expert Response

CDEC = Canadian Drug Expert Committee; DPP = dipeptidyl peptidase; GL = generalized lipodystrophy; GLP-1 = glucagon-like peptide-1 receptor; LD = lipodystrophy; PL = partial lipodystrophy; SGLT = sodium-glucose cotransporter.

Clinical Evidence

The objective of CADTH’s Clinical Review Report is to review and critically appraise the clinical evidence submitted by the sponsor on the beneficial and harmful effects of metreleptin (3 mg, 5.8 mg, and 11.3 mg, powder for solution, subcutaneous injection) for the treatment of the complications of leptin deficiency experienced by patients with lipodystrophy. The focus is on comparing metreleptin to relevant comparators and identifying gaps in the current evidence.

A summary of the clinical evidence provided by the sponsor in the review of metreleptin is presented in 2 sections with CADTH’s critical appraisal of the evidence included at the end of each section. The first section, the systematic review, includes pivotal studies and RCTs that were selected according to the sponsor’s systematic review protocol. CADTH’s assessment of the certainty of the evidence in this first section using the GRADE approach follows the critical appraisal of the evidence. The second section includes additional studies that were considered by the sponsor to address important gaps in the systematic review evidence.

Included Studies

Clinical evidence from the following are included in the CADTH review and appraised in this document:

• 1 pivotal clinical trial identified in a systematic review

• 1 supportive clinical trial

• 2 historical control arm analyses addressing gaps in evidence.

Systematic Review

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the CADTH review team.

Description of Studies

Characteristics of the included studies are summarized in Table 7.

Table 7: Details of Studies Included in the Systematic Review

ACTH = adrenocorticotropic hormone; ALT = alanine aminotransferase; AST = aspartate aminotransferase; FFA = free fatty acid; HDL-C = high-density lipoprotein cholesterol; IVGTT = IV glucose tolerance test; LDL-C = low-density lipoprotein cholesterol; NIH = National Institutes of Health; OGTT = oral glucose tolerance test.

Source: Study 991265/20010769 Clinical Study Report.³⁹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

The NIH 991265/20010769 was a phase II/III, open-label, single-arm, single-centre, investigator-sponsored study. Study 991265 was a pilot, dose-escalation study, with the objectives to determine if metreleptin can be safely administered to a group of patients with clinically significant lipodystrophy and to determine if metreleptin treatment will be effective in lowering plasma glucose and lipid abnormalities experienced by patients with clinically significant lipodystrophy. Additional objectives of this study were to determine if treatment with metreleptin could ameliorate lipid deposition in liver and muscle or improve the hypogonadotropic hypogonadism observed in some patients. Study 20010769 was a long-term study conducted to determine the long-term safety and efficacy of metreleptin treatment for patients with lipodystrophy.

Patient enrolment occurred between July 24, 2000, and March 26, 2014; the data cut-off date was in December 2014. Study 20010769 allowed for the rollover of patients from the pilot study, as well as for direct enrolment of new patients. A total of 107 patients were enrolled in the studies, which were conducted at the NIH. Although these studies were conducted at the NIH, patients were also enrolled from countries outside the US, including Canada. Nine of the 107 patients were enrolled in the pilot Study 991265; of these, 8 rolled over to receive metreleptin in Study 20010769, and 98 patients enrolled directly into Study 20010769. A total of 66 of the 107 patients had GL and 41 had PL, including 31 patients in the PL subgroup, that is, those patients with PL with baseline hemoglobin A1C greater than or equal to 6.5% and/or triglyceride levels greater than or equal to 5.65 mmol/L.

Patients were assessed 7 days before initiation of metreleptin for evaluation of study eligibility and to undergo baseline assessments, including a complete medical history and physical examination, including vital signs. At the end of the baseline evaluation, patients began subcutaneous metreleptin injections and were observed on therapy as inpatients for at least 48 hours before discharge. Patients received detailed education on home glucose-monitoring and self-injection techniques.

Populations

Inclusion and Exclusion Criteria

Eligible patients were aged older than 5 years in Study 991265 and 6 months or older in Study 20010769 with clinically significant lipodystrophy. To be eligible to participate in Study 991265, participants’ circulating leptin levels had to be less than or equal to 8.0 ng/mL in females and less than or equal to 6.0 ng/mL in males to be eligible. To be eligible to participate in Study 20010769, circulating leptin levels had to be less than 12.0 ng/mL in females and less than 8.0 ng/mL in males as measured using a Linco assay on a specimen obtained after an overnight fast; in children aged 6 months to 5 years, a circulating leptin level of less than 6 ng/mL was used. In both studies, the presence of at least 1 of the following metabolic abnormalities was an inclusion criterion: diabetes as defined by American Diabetes Association criteria; fasting insulin greater than 30 µU/mL; fasting hypertriglyceridemia, defined as triglyceride levels greater than 2.26 mmol/L in Study 991265 and greater than 2.26 mmol/L or postprandially elevated to greater than 5.65 mmol/L when fasting was clinically not indicated (e.g., in infants) in Study 20010769. General exclusion criteria included “pregnant women, women in their reproductive years who did not use an effective method of birth control, and women who were nursing or who were lactating within 6 weeks of having completed nursing.”³⁹

Interventions

All patients received metreleptin subcutaneously, either administered by themselves or by caregivers. Dosing was administered on a body weight basis and varied by age and sex. Injections were to be administered at a single site at a maximum allowable volume of 2.0 mL. Administering the study drug at the same time each day was recommended; dosing was to alternate between sites on the abdomen and limbs.

Because of variability in individual metabolic profiles at baseline and differences in response to metreleptin (likely due in large part to sex and lipodystrophy subtype), metreleptin doses for each patient were adjusted according to individual response, for example, increased to achieve better efficacy or decreased (based on AEs or effects such as excessive weight loss). Thus, an individualized approach was utilized to determine each patient’s dose.

All patients were admitted to the NIH Clinical Center for screening and baseline assessments. These were scheduled to occur over 7 days. Patients received detailed education on self-injection and home glucose-monitoring techniques.

Patients were to be discontinued from treatment immediately if there was a systemic life-threatening allergic reaction to metreleptin, AEs that were not readily explained by other causes unrelated to current protocol, development of any medical problem listed in the exclusion criteria, and noncompliance with self-monitoring requirements and clinical centre visits, among other reasons.

Study 991265

At the time of initiation of the pilot Study 991265, the dose of metreleptin required to achieve a normal leptin concentration was proposed to be 0.03 mg/kg of lean body weight for females aged between 14 and 18 years, 0.04 mg/kg of lean body weight for adult females, and 0.02 mg/kg of lean body weight for males, regardless of age. Patients were initially started at 50% of this predicted dose with subsequent dose increases (to 100% and 200%) based on safety and effectiveness. A conservative, empirical approach to dosing was employed: the dose was calculated based on body weight (mg per kg) and gradually titrated from a low starting dose that was escalated over several months, with the total daily dose administered twice a day (a.m. and p.m.). Approximately twice the dose was used for females than that for males based upon the observation that females in general have higher concentrations of leptin even after adjustment for differences in body compositions. The target dose of metreleptin for each patient was achieved via a 2-step dose escalation and the dosing regimen in this pilot study was as follows:

• month 1: 0.02 mg/kg/day for females aged 18 years or older, 0.015 mg/kg/day for females aged younger than 18 years, and 0.01 mg/kg for males (considered “50%” target dose)

• month 2: 0.04 mg/kg/day for females aged 18 years or older, 0.03 mg/kg/day for females younger than 18 years, and 0.02 mg/kg for males (considered “100%” target dose)

• month 3 to 4: 0.08 mg/kg/day for females aged 18 years or older, 0.06 mg/kg/day for females younger than 18 years, and 0.04 mg/kg for males (considered “200%” target dose).

Any individual patient could participate in Study 991265 for up to 8 months, after which they could roll over into the long-term Study 20010769.

Study 20010769

The dosing regimen in Study 20010769 was initially the same as in Study 991265. However, dosing of metreleptin evolved to initiating at more typical efficacious doses with minimal titration, and dosing frequency was modified from twice daily to once daily. In females aged 5 years and older, the modified starting dose was 0.08 mg/kg/day to 0.10 mg/kg/day; in females younger than 5 years and all males, the starting dose was 0.06 mg/kg/day. The metreleptin target dose for each patient was initially achieved via a 2-step dose escalation. As information about metreleptin dosing grew, patients who initiated later started at higher doses and required minimal to no dose escalation.

The dose of metreleptin could be increased after the 6-month follow-up. All dose escalations were performed in increments of 0.02 mg/kg/day for females aged 10 years and older and 0.01 mg/kg/day in all other patients. Only 1 dose increase could be done per week to evaluate the effect on body weight and on injection sites. In individuals aged older than 5 years, the dose of metreleptin was not to be increased beyond 0.12 mg/kg/day unless there was a clear decline in metabolic status without alternative explanations for the metabolic change (such as an infection, noncompliance, or dietary indiscretion). Dose escalations without prior approval were capped at 0.24 mg/kg/day for any patient. If patients could not tolerate a higher dose level, they could continue treatment at the next lowest tolerated dose. The dosing frequency was also modified during the course of the study. The total daily dose was originally administered twice a day in 2 equal doses approximately 12 hours apart. With an amendment to the protocol, patients could switch to once daily dosing while keeping the total daily dose constant after the 1-year follow-up. If, according to the principal investigator or the coordinating nurse practitioner, there was a decline in clinical benefit with once daily dosing, the patients were to go back to twice daily dosing. Dosing was later initiated at once a day.

Outcomes

A list of efficacy end points assessed in this Clinical Review Report is provided in Table 8, and descriptions of the outcome measures follow. Summarized end points are based on outcomes included in the sponsor’s Summary of Clinical Evidence as well as any outcomes identified as important to this review according to the clinical expert(s) consulted by CADTH and stakeholder input from patient and clinician groups and public drug plans. Using the same considerations, the CADTH review team selected end points considered most relevant to informing CADTH’s expert committee deliberations and finalized this list of end points in consultation with members of the expert committee. All summarized efficacy end points were assessed using GRADE. Select notable harms outcomes considered important for informing CADTH’s expert committee deliberations were also assessed using GRADE.

Table 8: Outcomes Summarized From the Studies Included in the Systematic Review

ACTH = adrenocorticotropic hormone; ALT = alanine aminotransferase; AST = aspartate aminotransferase; FFA = free fatty acid; HDL-C = high-density lipoprotein cholesterol; IVGTT = IV glucose tolerance test; LDL-C = low-density lipoprotein cholesterol; NIH = National Institutes of Health; OGTT = oral glucose tolerance test.

^aNo adjustment for analysis of 2 primary end points is required since the end points of change from baseline in hemoglobin A1C and fasting triglyceride levels are regarded as co-primary, that is, significance on both end points is required for an overall significant result.

^bIn the pilot study, postbaseline visits were initially scheduled at months 1, 2, and 4; visits at months 6, 8, and 12 were added by protocol amendment. Note that patients in this study who elected to continue metreleptin treatment were transferred to the long-term study at approximately month 8 of treatment. The original protocol for Study 20010769 included postbaseline visits at months 1, 2, 6, 4, 8, and 12; Amendment 1 removed the month 6 assessment and added assessments every 6 months after month 12. As more experience was gained with the safety and effectiveness profile of metreleptin, and as many of the patients travelled a great distance to the NIH for treatment, the visit schedule was reduced or modified with Amendment F to include visits at months 3, 6, and 12 and every 6 months thereafter.

Source: Study 991265/20010769 Clinical Study Report.³⁹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

According to the clinical experts and clinician group input, change in hemoglobin A1C and triglyceride levels are the common metabolic parameters that are measured to assess disease control in the population of patients with lipodystrophy. The thresholds that determine a clinically meaningful response depend on the individual patient, the starting level of metabolic parameters, and the background therapies received. Other important end points included were change in plasma glucose and change in liver volume as this can impact long-term complications.

Primary Efficacy Outcomes

Actual change from baseline in hemoglobin A1C and percent change from baseline in fasting triglyceride levels to month 12 were the co-primary efficacy end points. The clinical experts considered month 12 an appropriate time point for analysis as the effects of metreleptin could be expected to be seen by this time. The sponsor noted that this 12-month period would allow for individual dose titrations to achieve maximum effects in given patients; 12 months was also an acceptable length of time over which to assess the clinical impact of the treatment. To account for patients who discontinued treatment before month 12, LOCF methods were used to determine changes from baseline to month 12. Specifically, hemoglobin A1C and triglyceride samples obtained on or after day 180 were used in the analysis for patients without samples obtained within the month 12 window (day 365 ± 65 days).

Secondary Efficacy Outcomes

Serum chemistry parameters, including fasting glucose, were tested using laboratory assessments at screening or baseline, then at months 1, 2, 4, 6, 8, and 12, and then every 6 months. The actual and percent change from baseline in fasting plasma glucose levels at month 12 was assessed as a key secondary efficacy end point. The actual and percent change from baseline in fasting glucose at each postbaseline visit was also assessed as a secondary efficacy end point.

Patients aged 5 years and older underwent MRI of the liver to evaluate its size and estimate fat deposition. IV gadolinium was administered to accurately calculate fat volume in the liver. Assessments were conducted at screening or baseline, at months 4, 8, and 12, and then every 6 months. Actual change from baseline in liver volume at each postbaseline visit through month 12 was assessed as a secondary efficacy end point.

Statistical Analysis

Sample Size and Power Calculation

The NIH investigators specified a sample size justification in both study protocols as follows:

Based on preliminary data in a cross-sectional study, the mean ± SD HbA1c data for 8 patients with GL was 9.1 ± 2.2%. Based on assumption of a 1.5% (for protocol 991265 and 1.0% for 20010769) actual decrease in HbA1c levels over a period of 4 months (for protocol 991265 (and 12 months for 20010769)) as clinically meaningful, 10 patients would be required for 80% study power and an alpha error of 5%. Also, based on previous cross-sectional data, the mean ± SD fasting TG levels for 8 patients with GL was 2200 ± 900 mg/dL. Based on assumption of 660 mg/dL (or 30% decrease from the mean baseline) decrease as clinically meaningful, 10 patients with hypertriglyceridemia would be required for 80% study power and 5% alpha error.

Upon validation of the sample size calculation, it was found that, based on these assumptions, 32 patients would be required to participate in the sample to be able to detect a 1% actual decrease in hemoglobin A1C and 15 patients would be required to detect a 1.5% actual decrease in hemoglobin A1C with 80% power and 5% 1-sided alpha error. For triglycerides, a sample size of 13 would be required to detect a reduction of 660 mg/dL with 80% power and 5% 1-sided alpha error. The final sample size across the 2 protocols was 107 patients. As it became clear that treatment with metreleptin was improving the metabolic abnormalities associated with lipodystrophy and could be safely administered long-term with benefits to patients, the protocol was amended to expand the eligibility criteria and to increase the sample size.

Statistical Testing

Tabulations were produced for appropriate demographic, baseline, efficacy, pharmacokinetic, and safety parameters. For categorical variables, summary tabulations of the number and percentage of patients within each category (with a category for missing data, where appropriate) of the parameter were presented. For continuous variables, the number of patients, mean, median, SD, minimum, and maximum values were presented. Geometric mean and SE of the geometric mean were presented as appropriate.

For analysis of change from baseline, the total number of patients by time point reflected the actual number of patients with data for that specific parameter at baseline and the specified time point. The total number of patients at a given time point depended on whether data for that time point were available and whether the study visit fell within the specified analysis visit window.

Formal statistical hypothesis testing was performed on the primary efficacy end point with all tests conducted at the 2-sided, 0.05 level of significance. Summary statistics were presented, as well as CIs on selected parameters.

Interim safety and efficacy data were examined at several data cuts over the course of the study. As such, multiple interim reports were produced for regulatory submissions. The analyses presented in this report are the final analyses.

Co-Primary End Points

The co-primary efficacy analyses were performed using the FAS. Actual change from baseline in hemoglobin A1C and actual and percent changes from baseline in fasting triglyceride levels were summarized using descriptive statistics and 95% CIs by type (GL or PL) and PL subgroup, by sex and overall for GL and PL separately. P values were computed using paired t tests to determine if the change from baseline to month 12 was significantly different from 0, at a 1-sided alpha-level of 0.025. The LOCF method was used to impute any missing month 12 hemoglobin A1C and fasting triglyceride results. The imputation only took into account results that were at least 6 months (180 days) postbaseline. Thus, analysis of primary efficacy end points included all patients who had baseline and at least month 6 measurements and were presented by type and PL subgroup, by sex and overall for GL and PL separately.

Mean actual and percent changes over time through month 12 in hemoglobin A1C and fasting triglycerides were plotted with SEs by type and PL subgroup, by sex and overall for GL and PL separately. Waterfall plots of by-patient changes from baseline to month 12 in hemoglobin A1C and fasting triglyceride levels were also presented by peak titre and overall.

Scatter plots showing the correlated by-patient changes in hemoglobin A1C and fasting triglyceride levels were also presented by type and PL subgroup, by sex and overall for GL and PL separately.

No adjustment for analysis of 2 primary end points was required since the end points of change from baseline in hemoglobin A1C and fasting triglyceride levels were regarded as co-primary, that is, significance on both end points was required for an overall significant result. No formal statistical analyses on the primary end point that adjust for possible covariate effects were planned.

A sensitivity analysis on the co-primary efficacy end points was performed using the efficacy-evaluable analysis set, controlled concomitant medication full analysis set (CFAS), and controlled concomitant medication efficacy-evaluable analysis set (CEEAS). For tables using the CFAS or CEEAS, a patient only contributed to the change in parameter for which the concomitant medication was controlled. In other words, a patient could only contribute to the change in hemoglobin A1C if there was no increase in the first 12 months in antidiabetic medications and could only contribute to the change in fasting triglycerides if there was no increase in the first 12 months in lipid-lowering therapies. Analyses using the CFAS also used the LOCF imputation method. An additional sensitivity analysis using the FAS on the co-primary end points used the worst observation carried forward (WOCF) imputation method. An analysis using the baseline observation carried forward (BOCF) imputation was also conducted to comply with a specific request from Health Canada.

Secondary End Points

Actual and percent change from baseline in liver volume at each postbaseline analysis visit through month 12 was summarized using descriptive statistics by type and PL subgroup, by sex and overall for GL and PL separately. It was also presented by age subgroups (< 18 years and ≥ 18 years).

Table 9: Statistical Analysis of Efficacy End Points

ALT = alanine aminotransferase; AST = aspartate aminotransferase; BOCF = baseline observation carried forward; CEEAS = Controlled Concomitant Medication Efficacy-Evaluable Analysis Set; CFAS = controlled concomitant medication full analysis set; CI = confidence interval; EEAS = efficacy-evaluable analysis set; FAS = full analysis set; FFA = free fatty acid; GL = generalized lipodystrophy; HDL-C = high-density lipoprotein cholesterol; LDL-C = low-density lipoprotein cholesterol; LOCF = last observation carried forward; MMRM = mixed-effect model for repeated measures; NA = not applicable; PL = partial lipodystrophy; WOCF = worst observation carried forward.

^aAll descriptive statistical analyses were performed using SAS statistical software version 9.3, unless otherwise noted.

Source: Study 991265/20010769 Clinical Study Report.³⁹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Analysis Populations

Table 10 summarizes the analysis populations in the NIH 991265/20010769 study.

Table 10: Analysis Populations of the NIH 991265/20010769 Study

CEEAS = controlled concomitant medication efficacy-evaluable analysis set; CFAS = controlled concomitant medication full analysis set; EEAS = efficacy-evaluable analysis set; FAS = full analysis set; NIH = National Institutes of Health; SAS = safety analysis set.

^aData for all antidiabetic or lipid-lowering therapies, including type, dose, regimen, and route of administration, underwent medical review. Patients who had these types of medications added or doses increased, which may have had an impact on the efficacy end points, were excluded from the CFAS and CEEAS. Patients were excluded separately based on the type of medication that was added or increased, e.g., patients with potentially confounding antidiabetes medications were excluded from the analyses of hemoglobin A1C and those with potentially confounding lipid-lowering therapies were excluded from the analyses of triglycerides.

Source: Study 991265/20010769 Clinical Study Report.³⁹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Results

Patient Disposition

Table 11 summarizes the patient disposition from the NIH 991265/20010769 study. A total of 66 patients with GL and 41 patients with PL received treatment. Of the patients in the GL cohort, 34.8% discontinued treatment prematurely, with the most common reason cited being transfer to another program. In the PL cohort, 36.6% of patients discontinued treatment prematurely, with the most common reason cited being noncompliance.

Table 11: Summary of Patient Disposition From Studies Included in the Systematic Review

GL = generalized lipodystrophy; NIH = National Institutes of Health; PL = partial lipodystrophy.

^aThe PL subgroup included patients with baseline hemoglobin A1C ≥ 6.5% and/or triglycerides ≥ 5.65 mmol/L.

^bOther reasons included diagnosis of bipolar disorder, health issues, and gastric bypass surgery.

^cPatients who were receiving treatment at the time when metreleptin was approved in the US were contacted to determine if and how they were able to continue therapy.

^dAll enrolled patients who received at least 1 dose of the study drug.

^eAll patients in the safety analysis set who had either primary efficacy parameter measured at baseline and at ≥ 1 postbaseline visit.

^fAll patients in the full analysis set who had controlled concomitant medication use before month 12.

^gAll patients in the full analysis set who had either efficacy parameter of interest measured at month 12 and no major protocol violations before month 12.

^hAll patients in the controlled concomitant medication full analysis set who had either efficacy parameter of interest measured at month 12 and no major protocol violations before month 12.

^IAll patients in the safety analysis set who have antibody status assessed ≥ 1 while enrolled in the study.

Source: Study 991265/20010769 Clinical Study Report.³⁹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Baseline Characteristics

A summary of baseline characteristics in the NIH 991265/20010769 study is shown in Table 12. In the GL cohort, most of the patients were female (77.3%) and the mean age was 17.8 years (SD = ||||||| years). Mean baseline hemoglobin A1C was 8.6% (SD = 2.32%) and mean fasting triglyceride levels were 14.5 mmol/L (SD = |||||||| mmol/L).

Most of the patients in the overall PL cohort were also female (97.6%) and the mean age was 34.1 years (SD = ||||||| years). Mean baseline hemoglobin A1C was 8.0% (SD = 2.15%) and mean fasting triglyceride levels were |||| |||||| |||||||, while the specified PL subgroup had baseline hemoglobin A1C of 8.8% (SD = 1.88%) and mean fasting triglyceride levels of |||| |||||| |||||||.

The baseline characteristics outlined in Table 12 are limited to those that are most relevant to this review or were considered as having the greatest effect on the outcomes or interpretation of the study results.

Table 12: Summary of Baseline Characteristics From Studies Included in the Systematic Review

ALT = alanine aminotransferase; AST = aspartate aminotransferase; BMI = body mass index; GL = generalized lipodystrophy; max = maximum; min = minimum; NIH = National Institutes of Health; PL = partial lipodystrophy; SD = standard deviation; TG = triglyceride; ULN = upper limit of normal.

^aThe PL subgroup includes patients with baseline hemoglobin A1C ≥ 6.5% and/or triglycerides ≥ 5.65 mmol/L.

Source: Study 991265/20010769 Clinical Study Report.³⁹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Exposure to Study Treatments

A summary of metreleptin exposure in the NIH 991265/20010769 study is shown in Table 13. The mean overall duration (from first to last dose, including any drug interruptions), in the GL cohort was 62.5 months (SD = 45.66 months), in the overall PL cohort was 48.1 months (SD = 44.14 months), and in the specified PL subgroup was 47.5 months (SD = 47.11 months). The average daily dose was 5.0 mg/day (SD = 3.01 mg/day) in the GL cohort, 8.4 mg/day (2.45 mg/day) in the overall PL cohort, and 8.4 mg/day (2.40 mg/day) in the specified PL subgroup.

Table 13: Summary of Patient Exposure From Studies Included in the Systematic Review

GL = generalized lipodystrophy; max = maximum; min = minimum; NIH = National Institutes of Health; PL = partial lipodystrophy; SD = standard deviation.

^aThe PL subgroup includes patients with baseline hemoglobin A1C ≥ 6.5% and/or triglycerides ≥ 5.65 mmol/L.

^bOverall duration is from first to last dose; actual duration excludes drug interruptions.

^cDose intensity is the cumulative amount of drug exposure divided by actual treatment duration.

^dAverage daily dose is the cumulative amount of drug exposure divided by overall number of planned dosing days.

^eWeighted average dose over the study period is the sum of daily doses (in mg/kg or mg) over the overall duration at that dose/regimen divided by the total duration of all dose regimens.

Source: Study 991265/20010769 Clinical Study Report.³⁹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

A summary of baseline medication use in the NIH 991265/20010769 study is shown in Table 14. Antidiabetic medication use was reported for 80.3% of patients in the GL cohort, 90.2% in the overall PL cohort, and 96.8% in the specified PL subgroup. Lipid-lowering medication use was reported by 51.5% of patients in the GL cohort, 82.9% in the overall PL cohort, and 83.9% in the specified PL subgroup.

Table 14: Baseline Medication Use in NIH 991265/20010769 Study (Safety Analysis Set)

ATC = anatomic therapeutic class; GL = generalized lipodystrophy; HMG-CoA = 3-hydroxy-3-methyl-glutaryl-coenzyme A; NIH = National Institutes of Health; PL = partial lipodystrophy; WHODD = WHO Drug Dictionary.

^aThe PL subgroup includes patients with baseline hemoglobin A1C ≥ 6.5% and/or triglycerides ≥ 5.65 mmol/L.

^bTerms across multiple WHODD ATC classes are grouped to provide overall incidence.

^cIndividual WHODD preferred terms reported with incidence > 10% in GL or PL groups.

Source: Study 991265/20010769 Clinical Study Report.³⁹

Efficacy

A summary of key efficacy outcomes from the NIH 991265/20010769 study is shown in Table 15.

Change From Baseline in Hemoglobin A1C at 12 Months

Change from baseline in hemoglobin A1C at 12 months for the FAS population is shown in Table 15. In the GL cohort, mean hemoglobin A1C was 8.6% (SD = 2.33%) at baseline and 6.4% (SD = 1.68%) at month 12, for a mean change from baseline of –2.2% (95% CI, –2.7% to –1.6%). Included in the 12-month analysis were 59 of 62 patients in the FAS, with || patients imputed using LOCF. In the overall PL cohort, mean hemoglobin A1C was 7.9% (SD = ||||) at baseline and 7.4% (SD = ||||) at month 12, for a mean change from baseline of –0.6% (95% CI, |||| || ||||). Included in the 12-month analysis were 37 of 40 patients in the FAS, with | patients imputed using LOCF. In the specified PL subgroup, mean hemoglobin A1C was 8.7% (SD = ||||) at baseline and 7.9% (SD = ||||) at month 12, for a mean change from baseline of –0.9% (95% CI, |||| || ||||). Included in the 12-month analysis were 28 of 30 patients in the PL subgroup, with | patients imputed using LOCF.

Sensitivity analyses were conducted using the CFAS population. In the GL cohort (N = 54), the mean change from baseline was –1.9% (95% CI, –2.6% to –1.2%). In the overall PL cohort (N = 31), the mean change from baseline was –0.4% (95% CI, –0.9% to 0.0%). In the specified PL subgroup (N = 23), the mean change from baseline was –0.7% (95% CI, –1.2% to –0.2%).

Table 15: Summary of Key Efficacy Results From Studies Included in the Systematic Review

CI = confidence interval; FAS = full analysis set; GL = generalized lipodystrophy; max = maximum; min = minimum; LOCF = last observation carried forward; PL = partial lipodystrophy; SD = standard deviation; TG = triglyceride.

Note: The LOCF imputation method only includes results that are at least 6 months (180 days) postbaseline.

^aThe PL subgroup included patients with baseline hemoglobin A1C ≥ 6.5% and/or triglycerides ≥ 5.65 mmol/L.

^bP values computed using paired t tests.

Source: Study 991265/20010769 Clinical Study Report.³⁹

Change From Baseline in Fasting Triglycerides at 12 Months

Change from baseline in fasting triglyceride levels at 12 months for the FAS population is shown in Table 15. In the GL cohort, mean triglyceride level was 14.7 mmol/L (SD = 25.66 mmol/L) at baseline and 4.5 mmol/L (SD = 6.10 mmol/L) at month 12, for a relative mean change from baseline of –32.1% (95% CI, −51.0% to −13.2%). Included in the 12-month analysis were 57 of 62 patients in the FAS, with || patients imputed using LOCF. In the overall PL cohort, mean triglyceride level was |||| |||||| ||||||| at baseline and ||| |||||| at month 12, for a relative mean change from baseline of ||||| |||| ||| ||||| || ||||). Included in the 12-month analysis were 37 of 40 patients in the FAS, with | patients imputed using LOCF. In the specified PL subgroup, mean triglyceride level was |||| |||||| ||||||| at baseline and ||| |||||| ||||||| at month 12, for a mean change from baseline of |||| |||| ||| ||||| || ||||). Included in the 12-month analysis were 28 of 30 patients in the PL subgroup, with ||| patients imputed using LOCF.

Sensitivity analyses were conducted using the CFAS population. In the GL cohort (N = 54), the mean change from baseline was –26.5% (95% CI, –49.7% to –3.3%). In the overall PL cohort (N = 31), the mean change from baseline was –19.2% (95% CI, –42.7% to 4.3%). In the specified PL subgroup (N = 23), the mean change from baseline was –34.0% (95% CI, –49.6% to 4.3%). The sponsor, |||||| ||| |||| ||| || || ||| ||||||| ||||||||, conducted an ad hoc sensitivity analysis, removing 1 patient in the PL cohort who was recorded as patient with noncompliance behaviour. The results of this ad hoc analysis showed a mean change from baseline in triglyceride levels of –20.8% (95% CI, ||||| || ||||) in the PL cohort and –37.4% (95% CI, ||||| || |||||) in the specified PL subgroup.

Change From Baseline in Fasting Glucose at 12 Months

Change from baseline in fasting glucose at 12 months for the FAS population is shown in Table 15. In the GL cohort, mean glucose level was 10.2 mmol/L (SD = 5.05 mmol/L) at baseline and 7.0 mmol/L (SD = 3.40 mmol/L) at month 12, for a relative mean change from baseline of –19.7% (95% CI, –29.4% to –10.0%). Included in the 12-month analysis were 59 of 62 patients in the FAS, with || patients imputed using LOCF. In the overall PL cohort, mean glucose level was 8.8 mmol/L (SD = 4.39 mmol/L) at baseline and 7.5 mmol/L (SD = 3.28 mmol/L) at month 12, for a relative mean change from baseline of –6.1% (95% CI, –16.0% to 3.8%). Included in the 12-month analysis were 37 of 40 patients in the FAS, with |||| patients imputed using LOCF. In the specified PL subgroup, mean glucose level was 10.0 mmol/L (SD = |||| mmol/L) at baseline and 8.1 mmol/L (SD = |||| mmol/L) at month 12, for a relative mean change from baseline of –13.2% (95% CI, –24.4% to –1.9%). Included in the 12-month analysis were 28 of 30 patients in PL subgroup, with | patients imputed using LOCF.

Sensitivity analyses were conducted using the CFAS population. In the GL cohort (N = 54), the mean change from baseline was |||||% (95% CI, ||||| || ||||). In the overall PL cohort (N = 31), the mean change from baseline was ||||| |||| ||| ||||| || ||||). In the specified PL subgroup (N = 23), the mean change from baseline was ||||| |||| ||| ||||| || ||||).

Change From Baseline in Liver Volume at 12 Months

Change from baseline in liver volume at 12 months for the FAS population is shown in Table 15. In the GL cohort (N = 21), mean baseline liver volume was 3,357.7 mL (SD = 1,121.74 mL); the relative mean change from baseline was –33.8% (SD = |||||). In the overall PL cohort (N = 9), mean baseline liver volume was 2,624.6 mL (SD = 936.21 mL); the relative mean change from baseline was –13.4% (SD = ||||). In the specified PL subgroup (N = 8), mean baseline liver volume was 2,411.7 mL (SD = 731.91 mL); the relative mean change from baseline was –12.4% (SD = ||||).

Sensitivity analyses were not conducted using the CFAS population for the liver volume end point.

Harms

Refer to Table 16 for harms data.

Adverse Events

Of the patients in the GL cohort, 59 (89.4%) experienced TEAEs. Of the patients in the PL cohort, 35 (85.4%) experienced TEAEs. The most common AEs in the GL cohort were weight decrease (25.8%), abdominal pain (16.7%), and hypoglycemia (15.2%). The most common AEs in the PL cohort were hypoglycemia (17.1%), abdominal pain (14.6%), and nausea (14.6%).

Serious Adverse Events

In the GL cohort, 23 (34.8%) patients experienced SAEs, with ||||||||| |||| |||||| ||| |||||||||||| |||||| ||| |||| ||||||. In the PL cohort, 10 (24.4%) patients SAEs, with ||||||||| |||| ||||||| |||||||||||| ||||||| ||| |||||||||| |||||| ||| |||| ||||||.

Withdrawals Due to Adverse Events

Five (7.6%) of the patients in the GL cohort and 1 (2.4%) of the patients in the PL cohort experienced TEAEs that resulted in treatment discontinuation.

Mortality

Three (4.5%) of the patients in the GL cohort died, due to renal failure, cardiac arrest, and chronic hepatic failure. One (2.4%) of the patients in the PL cohort died, due to hypoxic-ischemic encephalopathy.

Notable Harms

No notable harms were identified by the clinical experts consulted for this review.

Table 16: Summary of Harms Results From Studies Included in the Systematic Review

GL = generalized lipodystrophy; NIH = National Institutes of Health; PL = partial lipodystrophy; SAE = serious adverse event; TEAE = treatment-emergent adverse event.

Source: Study 991265/20010769 Clinical Study Report.³⁹ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Pubertal Status

Pubertal status was assessed as a safety evaluation in the NIH 991265/20010769 study. There were 27 patients with GL younger than 18 years and 6 patients with PL (33 patients total) younger than 18 years with pubertal status assessed at baseline. Of these 33 patients, 26 had complete, near complete, or likely complete puberty. Of the remaining 7 patients (all with GL), 4 had delayed puberty before metreleptin treatment and 3 had precocious puberty. Follow-up was available for 3 of the patients with delayed puberty, 2 of whom experienced normal development while receiving metreleptin while 1 continued to have delayed puberty.

Endocrine Abnormalities

For 10 female patients in the NIH 991265/20010769 study, luteinizing hormone responses to luteinizing hormone–releasing hormone were noted as being more robust with metreleptin therapy, particularly for the 3 youngest patients. Of the 10 female patients, 8 had amenorrhea before therapy and all 8 developed normal menses after metreleptin therapy initiation.⁴⁰

Critical Appraisal

Internal Validity

The NIH 991265/20010769 study, the pivotal trial submitted by the sponsor, was a phase II/III, single-arm, open-label clinical trial. The lack of comparative data is a key limitation to the interpretation of the results from a single-arm trial, as it is difficult to distinguish between the effect of the intervention relative to a placebo effect, or the effect of natural history. It is acknowledged that there may be practical limitations to conducting an RCT of patients with lipodystrophy due to the rarity of the condition. However, while this provides context to the decision to conduct a single-arm trial, the limitations to the interpretation of the results remain. The open-label nature of the trial also increases the risk of bias. However, the study end points are objective laboratory values and therefore unlikely to have been influenced by this bias. Harms outcomes, however, may be impacted by the open-label design of the study.

When assessing the efficacy end points in the NIH 991265/20010769 study, the large number of drop-outs and missing data, due to the challenges of conducting a clinical study at the NIH that included international participants, must be taken into account. The co-primary end points of change from baseline in hemoglobin A1C and triglyceride levels were to be assessed at month 12. The statistical plan allowed for patients with missing data at month 12 to carry forward, according to LOCF methodology, if data from month 6 onwards were available. The clinical experts consulted by CADTH considered this to be a reasonable method to account for missing data given that most responses should be apparent in the laboratory findings by month 6. Despite this methodology to account for missing data, only 59 of 66 patients with GL and 37 of 41 patients with PL who received metreleptin were included in the change from baseline in hemoglobin A1C results and 57 of 66 patients with GL and 37 of 41 patients with PL who received metreleptin were included in the change from baseline in triglyceride results. The number of patients available for analysis at the 12-month time point was reduced because no measurements were available from 6 months onward. This analysis did not follow the intention-to-treat philosophy and likely overestimated the treatment effect. An additional impact of imputation methodology allowing results from 6 months onwards is the likely reduction in imprecision. It is likely that results after 12 months would be more variable than results after 6 months given the additional time available for treatment to improve laboratory measurements or wane in efficacy. In particular, data for ||% of the patients with GL included in the hemoglobin A1C end point were imputed as were data for ||% of the patients in the PL subgroup. As such, the 95% CIs presented are likely artificially narrow. Most notably, in the change from baseline in liver volume results, only 12 patients with GL and 8 patients with PL were included in the results because no imputation methodology was used for this end point. Therefore, any conclusions drawn from the change from baseline results are likely to be uninformative.

Statistical hypothesis testing was performed on the primary efficacy end point with all tests conducted at the 2-sided, 0.05 level of significance. It was noted that for the trial to be considered positive, both end points had to reach statistical significance, removing the need to adjust for multiplicity. However, multiple interim analyses were conducted with no reports of methodology changes that adjusted for the increased risk of type I error associated with interim analyses. In addition, it must be acknowledged that a statistically significant change from baseline in triglyceride levels in the PL cohort occurred only on the removal from the analysis of patient with noncompliance behaviour .

Given the possibility for concomitant medications to have an impact on end points, particularly in this disease area, the NIH 991265/20010769 study included a sensitivity analysis using the concomitant therapy–controlled population. In this analysis, only those patients who did not increase concomitant therapy use for a given end point (i.e., increase in antidiabetic therapies for the hemoglobin A1C end point) were included in the analysis. This removed proportionally more patients from the PL cohort than from the GL cohort, with results generally consistent with the primary analysis. BOCF and WOCF sensitivity analyses were also presented by the sponsor (data not reported in this report), in addition to the primary analysis of LOCF, to assess the impact of missing data on the end points of interest. The results of these sensitivity analyses were generally consistent with the primary analysis. It is noted that alternative assessments of the impact of missing data (such as using multiple imputation, assessments for data not missing at random) were not performed.

External Validity

The NIH 991265/20010769 study began to enrol patients in July 2000. As this was 23 years before the writing of this report, the clinical experts suggested that standards of therapy and patient support may have evolved since then. However, the clinical benefit of metreleptin is anticipated to be consistent with that observed in the NIH 991265/20010769 study. Lipodystrophy is a chronic disease and patients would be expected to receive treatment for life if experiencing benefit. The generalizability of the results beyond the 14-yrear follow-up of the NIH 991265/20010769 study is unknown, although the clinical experts consulted by CADTH did not expect the efficacy of metreleptin to change beyond the time horizon of the NIH 991265/20010769 study. The clinical experts consulted considered the patient characteristics from the NIH 991265/20010769 study to be broadly generalizable to that of the expected patient population in Canada.

GRADE Summary of Findings and Certainty of the Evidence

Methods for Assessing the Certainty of the Evidence

For pivotal studies and RCTs identified in the sponsor’s systematic review, GRADE was used to assess the certainty of the evidence for outcomes considered most relevant to informing CADTH’s expert committee deliberations, and a final certainty rating was determined as outlined by the GRADE Working Group:^41,42

• High certainty: We are very confident that the true effect lies close to that of the estimate of the effect.

• Moderate certainty: We are moderately confident in the effect estimate – the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. We use the word “likely” for evidence of moderate certainty (e.g., “X intervention likely results in Y outcome”).

• Low certainty: Our confidence in the effect estimate is limited – the true effect may be substantially different from the estimate of the effect. We use the word “may” for evidence of low certainty (e.g., “X intervention may result in Y outcome”).

• Very low certainty: We have very little confidence in the effect estimate – the true effect is likely to be substantially different from the estimate of effect. We describe evidence of very low certainty as “very uncertain.”

For single-arm trials: Although GRADE guidance is not available for noncomparative studies, the CADTH review team assessed pivotal single-arm trials for study limitations (which refers to internal validity or risk of bias), inconsistency across studies, indirectness, imprecision of effects, and publication bias to present these important considerations. Because the lack of a comparator arm does not allow for a conclusion to be drawn on the effect of the intervention versus any comparator, the certainty of evidence for single-arm trials starts at very low and there is no opportunity for rating up.

When possible, certainty was rated in the context of the presence of an important (nontrivial) treatment effect; if this was not possible, certainty was rated in the context of the presence of any treatment effect (i.e., the clinical importance is unclear). In all cases, the target of the certainty of evidence assessment was based on the point estimate and where it was located relative to the threshold for a clinically important effect (when a threshold was available) or to the null.

Results of GRADE Assessments

Table 2 shows the GRADE summary of findings for metreleptin.

Indirect Evidence

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the CADTH review team.

Objectives for the Summary of Supportive Analyses

As the pivotal NIH 991265/20010769 study was a single-arm study, no head-to-head efficacy data are available for metreleptin compared to supportive care for patients with lipodystrophy. To provide further context for the relative efficacy of metreleptin to supportive care, and to support the pharmacoeconomic model for metreleptin, the sponsor identified 2 supportive analyses estimating the relative effects of metreleptin compared to supportive care for patients with lipodystrophy.^6,23

Description of Supportive Analyses

• One unpublished supportive analysis was conducted to estimate the comparative treatment effect of metreleptin with or without supportive care compared to supportive care alone using an IPW and multivariate regression methods to adjust for differences between patients from the NIH follow-up study and an observational study of patients with GL and PL on supportive care alone.²³

• One published supportive analysis by Cook et al. (2021) estimated the effect of metreleptin treatment on mortality among patients with GL or PL using a Cox proportional hazards model to control for differences between patients receiving metreleptin treatment and those who have no experience of metreleptin.⁶

The study selection criteria for each supportive analysis are shown in Table 17.

Table 17: Study Selection Criteria and Methods for Supportive Analyses Submitted by the Sponsor

DPP = dipeptidyl peptidase; EASD = European Association for the Study of Diabetes; ECE = European Conference of Endocrinology; ESPE = European Society for Pediatric Oncology; PES = Pediatric Endocrine Society; GL = generalized lipodystrophy; GLP-1 = glucagon-like peptide-1 receptor; IPD = individual patient-level data; PL = partial lipodystrophy; NIH = National Institutes of Health; SGLT = sodium-glucose cotransporter.

Sources: Sponsor-submitted Indirect Treatment Comparison Technical Report;⁴³ Cook et al. (2021).⁴⁴ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Supportive Analysis Design

Study Selection Methods

Unpublished Supportive Analysis

The NIH follow-up study extended the NIH 991265/20010769 study by undertaking a chart review to collect long-term data and additional outcomes for patients with lipodystrophy who received metreleptin therapy at the NIH. The study is based on the participants in the original NIH 991265/20010769 study. These long-term data from the NIH follow-up study were used for the metreleptin data in the supportive analysis where individual patient-level data were available.

The systematic literature review identified 36 observational studies with participants who were not receiving metreleptin treatment and were receiving supportive care. The key relevant study identified, the GL and PL natural history study, was an observational chart review study of participants who had not received metreleptin treatment. As individual patient-level data were available for both studies, this supported the use of methods selecting observables to minimize bias to average treatment effect (ATE) of metreleptin with or without supportive care compared to supportive care alone.

Cook et al. (2021) Study

No systematic review to inform study selection was reported in published Cook et al. (2021) study. The study included a retrospective chart review of participants enrolled in the NIH 991265/20010769 study; 7 participants who met the eligibility criteria for the NIH 991265/20010769 study enrolled in a parallel nonrandomized study evaluating metreleptin.³³ A retrospective chart review of 230 patients with metreleptin-naive lipodystrophy from Brazil, Turkey, and the US was used as the comparator population.⁷

Supportive Analyses Methods

Unpublished Supportive Analysis

The IPW uses the propensity score function, which is a function of a set of observed covariates. Each participant’s weighting is equal to the inverse of the probability of receiving metreleptin, given the person’s certain characteristics. As the IPW uses the inverse of the probability of treatment assignment to weight outcomes, the ATE corresponds to the difference in these weighted means. Due to the relatively small sample size in the hemoglobin A1C outcome of the supportive care alone arm compared to the metreleptin with or without supportive care arm, stabilized IPW was used to avoid excessively high weights in the supportive care arm.

An ATE was calculated using linear models for continuous outcomes (change in hemoglobin A1C, triglycerides, alanine aminotransferase [ALT], and aspartate aminotransferase [AST] from baseline to month 12), generalized linear models for categorical outcomes (incidence of pancreatitis), and Cox proportional hazards models for time-to-event outcomes (all-cause mortality), using the propensity score weights as a link function. For continuous outcomes, the ATE was estimated by the weighted mean difference between the 2 groups, where the inverse of the probability of treatment assignment was used to weight outcomes. For categorical outcomes, the ATE was estimated by the odds ratio using the exponential of the coefficient of treatment assignment. For time-to-event outcomes, the ATE was estimated by the HR using the exponential of the coefficient of treatment assignment. Robust SEs were calculated using a robust sandwich estimator to take into account that the IPW uses weighted data.

Potential sensitivity analyses were considered, including the addition of extra covariates (history of baseline elevated hemoglobin A1C and elevated triglycerides, baseline leptin levels, and baseline pancreatitis) and subgroups (patients with GL and PL for whom supportive care failed). These sensitivity analyses were not deemed feasible due to the extent of the missing data in the GL and PL natural history study, alongside the limited number of mortality and pancreatitis events across the studies.

The covariates in the model, which were validated by clinical expert opinion, included sex, age at baseline, and lipodystrophy type. A variety of additional covariates were considered for inclusion in the propensity score model, but they were not deemed feasible due to the extent of missing data in the GL and PL natural history study.

Cook et al. (2021) Study

To adjust for patient characteristic differences between the participants with no experience with metreleptin treatment and those who had received metreleptin treatment, balanced risk set–matching was used to match participants in both groups in terms of similarity in age, sex, lipodystrophy diagnosis type, number of organs (heart, kidney, or liver) with an observed abnormality, and presence of elevated hemoglobin A1C levels (≥ 6.5%) at treatment initiation. Pancreatitis was not used as a matching criterion due to the large imbalance between the cohort of patients with no experience with metreleptin treatment and of patients who received metreleptin treatment.

A Mahalanobis distance calculation was conducted to determine the optimal index date for the metreleptin-naive population. The Mahalanobis distance was calculated for all possible monthly index dates for the metreleptin-naive population with a lower Mahalanobis distance indicating a closer match across selected patient characteristics. Participants were followed until the date of data abstraction or until loss to follow-up or death. A participant with no experience of metreleptin could be matched to only 1 participant who received metreleptin treatment to reduce the possibility of 1 person having an outsized impact on the analyses. Participants with no experience of metreleptin were also required to have at least 6 months of follow-up from the selected index date.

Primary analysis was assessed in the combined populations of GL and PL for patients with no experience of metreleptin treatment and those who received metreleptin treatment, with a subgroup analysis conducted in the GL population alone. The ATE of metreleptin on mortality was estimated by comparing the outcome of a metreleptin-treated cohort with that of matched metreleptin-naive cohort. Kaplan-Meier survival analysis was used to estimate mean time-to-mortality from treatment initiation for patients who received metreleptin treatment and from the index observation date for matched patients with no experience with metreleptin treatment. A log-rank test was conducted to compare time-to-mortality between the 2 cohorts.

A Cox proportional hazards model that controlled for treatment status, lipodystrophy diagnosis (GL or PL), age, triglyceride levels, hemoglobin A1C levels at 6.5% or greater, 1 or more episodes of pancreatitis, and abnormalities observed in the heart or kidneys at the time of treatment initiation or index date was used. A comparison of analysis methods is shown in Table 18.

Table 18: Supportive Analyses Methods

ALT = alanine aminotransferase; AST = aspartate aminotransferase; GL = generalized lipodystrophy; NA = not applicable.

Sources: Sponsor-submitted Indirect Treatment Comparison Technical Report;⁴³ Cook et al. (2021).⁴⁴ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Results of Supportive Analyses

Summary of Included Studies

Unpublished Supportive Analysis

An assessment of homogeneity for the sponsor-submitted unpublished supportive analysis is shown in Table 19.

Cook et al. (2021) Study

An assessment of homogeneity for the Cook et al. (2021) supportive analysis is shown in Table 20.

Results

Unpublished Supportive Analysis

A description of the covariates before and after IPW for each analyzed outcome is shown in Table 21. After weighting, covariates were broadly balanced. The biggest difference after weighting was in the mortality outcome; death occurred among 24.7% of the male patients in the supportive care arm compared to 19.4% of the male patients in the metreleptin with or without supportive care arm.

Table 19: Assessment of Homogeneity for Unpublished Supportive Analysis

ATE = average treatment effect; GL = generalized lipodystrophy; NIH = National Institutes of Health; PL = partial lipodystrophy.

Source: Sponsor-submitted Indirect Treatment Comparison Technical Report.⁴³ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 20: Assessment of Homogeneity for Cook et al. (2021)

GL = generalized lipodystrophy; NIH = National Institutes of Health; PL = partial lipodystrophy.

Sources: Cook et al. (2021).⁴⁴ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 21: Covariates Used in Comparison of Effects of Metreleptin With or Without SC Versus SC Alone Before and After IPW

IPW = inverse probability weighting; LD = lipodystrophy; NA = not available; SD = standard deviation; SC = supportive care.

^aWhile reported as N values, the weighted sample sizes are not true sample sizes and should be interpreted with caution. Weighted sample sizes: SC, N = 125; metreleptin with or without SC, N = 121.

^bWhile reported as N values, the weighted sample sizes are not true sample sizes and should be interpreted with caution. Weighted sample sizes: SC, N = 150; metreleptin with or without SC, N = 145.

^cWhile reported as N values, the weighted sample sizes are not true sample sizes and should be interpreted with caution. Weighted sample sizes: SC, N = 335; metreleptin with or without SC, N = 322.

Sources: Sponsor-submitted Indirect Treatment Comparison Technical Report.⁴³ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Table 22: IPW Efficacy Results for Metreleptin With or Without SC Versus SC Alone

ATE = average treatment effect; CI = confidence interval; IPW = inverse probability weighting; SC = supportive care; vs. = versus.

^aDenotes significance at the P < 0.05 level.

Source: Sponsor-submitted Indirect Treatment Comparison Technical Report.⁴³ Details included in the table are from the sponsor’s Summary of Clinical Evidence.

Results of the IPW analysis from the sponsor-submitted unpublished supportive analysis are shown in Table 22. The mean change in hemoglobin A1C was |||||| |||| ||| ||||| || |||||. An alternative stabilized IPW analysis was conducted to account for large weights being allocated to a small number of patients in the supportive care alone arm. This analysis yielded a mean change in hemoglobin A1C of –1.52% (95% CI, –2.28 to –0.77; P < 0.001).

Cook et al. (2021) Study

A summary of the characteristics of patients in the metreleptin-treated cohort and those in the metreleptin-naive cohort prematching and postmatching is shown in Table 23.

In the metreleptin-treated cohort, 11 patients with GL and 1 patient with PL (the patient had FPLD) died. In the matched metreleptin-naive cohort, 9 patients with GL and 3 patients with PL died. Figure 1 shows the Kaplan-Meier curve for time-to-mortality in the 2 populations (log-rank test P value = 0.2). The Kaplan-Meier curve from the Cox proportional hazards model, adjusted for selected covariates, is shown in Figure 2. The HR was 0.35 (95% CI, 0.134 to 0.900; P = 0.029).

Mortality risk of patients receiving metreleptin treatment and patients with no experience with metreleptin treatment in the GL subgroup were described using Kaplan-Meier analysis (log-rank test P = 0.6). The Cox proportional hazards model resulted in an HR of 0.455 (95% CI, 0.150 to 1.387; P = 0.166).

Critical Appraisal of Unpublished Supportive Analysis

The unpublished supportive analysis submitted by the sponsor was designed as a comparison of a prospective single-arm (NIH 991265/20010769) study compared against retrospective chart review. Data from retrospective chart reviews can have a number of confounding factors both in reported characteristics and unreported characteristics that may bias the results. Retrospective chart review analysis also tends to have a high amount of missing or unreliable data. In the case of the sponsor-provided unpublished supportive analysis, missing data resulted in a low sample size for outcomes such as change from baseline in hemoglobin A1C. This resulted in a small number of patients in the supportive care population being heavily weighted and in an imprecise 95% CI. The sponsor conducted a stabilized IPW to account for this, resulting in a significant benefit in favour of metreleptin in this end point; however, the ad hoc nature introduces bias and likely overestimates the true effect for this analysis since the ability to account for patient differences is more limited compared to the primary analysis.

Table 23: Patient Demographic and Clinical Characteristics Prematching and Postmatching (Overall Cohort)

AGL = acquired generalized lipodystrophy; CGL = congenital generalized lipodystrophy; FPLD = familial partial lipodystrophy; GL = generalized lipodystrophy; NA = not applicable; SD = standard deviation.

^aP < 0.01 compared with metreleptin-treated cohort.

^bP < 0.05 compared with metreleptin-treated cohort.

^cGL and PL subtypes, triglyceride levels, and pancreatitis were not used as matching parameters for the metreleptin-naive cohort.

^dPatients with mutations in AGPAT2 were the most common (n = 26 in metreleptin-treated cohort; n = 22 in matched metreleptin-naive cohort), followed by those with mutations in BSCL2 (n = 15 in metreleptin-treated cohort; n = 15 in matched metreleptin-naive cohort). The metreleptin-treated cohort also included 2 patients with CGL with other mutations, and 1 patient with CGL whose genetic testing data were either missing or a mutation could not be confirmed. The matched metreleptin-naive cohort also included 2 patients with PTRF4 mutations, 2 patients with CGL with other mutations, and 16 patients with CGL whose genetic testing data were either missing or a mutation could not be confirmed.

^ePatients with mutations in LMNA were the most common (n = 20 in metreleptin-treated cohort; n = 25 in matched metreleptin-naive cohort), followed by those with mutations in PPARG (n = 8 in metreleptin-treated cohort; n = 3 in matched metreleptin-naive cohort). The metreleptin-treated cohort also had 1 patient with FPLD with a mutation in PCYT1A and 9 patients with FPLD whose genetic testing data were either missing or a mutation could not be confirmed. The matched metreleptin-naive cohort also included 3 patients with Köbberling type FPLD, 5 patients with FPLD who had other mutations, and 2 patients with FPLD whose genetic testing data were either missing or a mutation could not be confirmed.

^fIndex observation date for the prematching metreleptin-naive cohort was defined as the time at which patients with no experience with metreleptin treatment achieved the mean age at the start of treatment of patients receiving metreleptin treatment (24.7 years) or the date of their last available observation, whichever came first.

^gCounts only include patients who have laboratory measurements taken on or after their index observation date.

Source: Cook et al. (2021). Copyright Cook et al. (2021). Reprinted in accordance with Creative Commons Attribution-NonCommercial-NoDerivs licence (CC BY-NC-ND 4.0 DEED): https://creativecommons.org/licenses/by-nc-nd/4.0/.⁶

The high amount of missing data in the retrospective chart review impacted the ability to control for important covariates in the IPW. Only 3 covariates were included: age, sex, and lipodystrophy type. Important variables such as baseline level of hemoglobin A1C and triglycerides could not be included, further reducing the reliability of the results. Clinical experts consulted by CADTH did not consider the list of included covariates to be sufficient for capturing the prognostic factors and treatment-effect modifiers important for patients with lipodystrophy. It is noted that these 2 variables were included as preplanned sensitivity analyses but deemed infeasible due to missing data. Further to the issue of missing data, population characteristics of these important variables were not presented for any patients included in the historical control arm of the analysis. The details of the standard-of-care therapies received by the historical control arm are also unknown, further increasing the uncertainty of the presented results, particularly when considering the retrospective chart review began in the year 1960.

The variables that were included in the IPW were mostly well balanced, although the low starting sample sizes for the supportive care population (N = 46 in unweighted population for the triglyceride level outcome) suggests that a few patients were very heavily weighted in the analysis.

Figure 1: Mortality Risk for Patients Receiving Metreleptin Treatment Versus Matched Patients With No Experience With Metreleptin Treatment (Overall Cohort)

Note: Vertical bars denote censoring events.

Source: Cook et al. (2021). Copyright Cook et al. (2021). Reprinted in accordance with Creative Commons Attribution-NonCommercial-NoDerivs licence (CC BY-NC-ND 4.0 DEED): https://creativecommons.org/licenses/by-nc-nd/4.0/.⁶

Figure 2: Cox Proportional Hazards Model–Predicted Mortality for Patients Receiving Metreleptin Treatment Versus Matched Patients With No Experience With Metreleptin Treatment (Overall Cohort)

GL = generalized lipodystrophy; PL = partial lipodystrophy.

Note: Modelled results are after adjusting for the following covariates: lipodystrophy diagnosis (GL or PL), birth year, triglyceride levels, elevated hemoglobin A1C, history of ≥ 1 episodes of pancreatitis, and the presence of observed abnormalities in the heart or kidneys.

Source: Cook et al. (2021). Copyright Cook et al. (2021). Reprinted in accordance with Creative Commons Attribution-NonCommercial-NoDerivs licence (CC BY-NC-ND 4.0 DEED): https://creativecommons.org/licenses/by-nc-nd/4.0/.⁶

Critical Appraisal of Cook et al. (2021)

Similar to the sponsor-submitted unpublished supportive analysis, the supportive analysis from Cook et al. (2021) is based on a comparison of a prospective single-arm trial (NIH 991265/20010769) against retrospective chart review. Data from retrospective chart reviews can have a number of confounding factors both in reported characteristics and unknown unreported characteristics that may bias the results.

The Cook et al. (2021) supportive analysis used a matching approach to identify patients from the retrospective chart review that most closely match patients from the NIH 991265/20010769 study. Matching was based on selected variables of age, sex, lipodystrophy diagnosis type, number of organs (heart, kidney or liver) with an observed abnormality, and presence of elevated hemoglobin A1C levels (≥ 6.5%) at treatment initiation. In the postmatching population, there were some residual imbalances in important prognostic variables such as presence of elevated hemoglobin A1C (78.6% in the metreleptin-treated versus 60.2% in the metreleptin-naive population) and the number of organs with observed abnormalities. The patients receiving metreleptin treatment appear to have a more severe presentation of lipodystrophy, potentially increasing the likelihood of seeing a larger change from baseline in hemoglobin A1C and triglyceride levels and therefore biasing the results in favour of metreleptin.

The end points included in the Cook et al. (2021) analyses were focused on risk of mortality. Given the small number of mortality events recorded in both populations, the results were imprecise and uncertain. While mortality is a very important end point in this disease context, the extended survival horizons of patients with lipodystrophy make any mortality analysis difficult to assess due to the low number of events. As such, metabolic parameter end points such as hemoglobin A1C and triglyceride levels offer more immediate and reliable insight into the effectiveness of therapies in treating lipodystrophy. Metabolic end points were not considered in the Cook et al. (2021) analysis. In addition, the exact standard of therapies used in the retrospective control arm was not described, making it difficult to assess the relevance of comparing metreleptin to standard of care.

The Cox proportional hazards model adjusted further for treatment status, lipodystrophy diagnosis (GL or PL), age, triglyceride levels, hemoglobin A1C levels at 6.5% or greater, 1 or more episodes of pancreatitis, and abnormalities observed in the heart or kidneys at the time of treatment initiation or index date. The clinical experts consulted by CADTH considered this list of adjustment factors to be a more comprehensive list than the one that was included in the unpublished analysis; however, it is still unlikely to capture all prognostic factors and effect modifiers. After the Cox proportional hazards model adjustment, the results showed a benefit in risk of mortality for patients receiving metreleptin treatment compared with those with no experience with metreleptin treatment, but these results cannot be considered conclusive due to these noted limitations of the analysis.

Summary

Both the sponsor-submitted historical control arm analyses compared retrospective chart review data of patients with lipodystrophy receiving standard of care with data from a cohort of patients with lipodystrophy, mostly from the NIH 991265/20010769 study, who received metreleptin. There was little overlap between the outcomes assessed in each analysis. As such, determining the level of consistency between the analyses is difficult. The Cook et al. (2021) supportive analysis assessed the time-to-mortality and risk of mortality in the combined GL and PL cohort. The matched analysis found no difference between the metreleptin-treated and metreleptin-naive cohorts. A Cox proportional hazards analysis controlling for selected variables found a benefit in time-to-mortality in the combined GL and PL cohort; no difference was found in the GL only cohort. However, the major limitations of this analysis relating to the small number of events preclude drawing any conclusions from this analysis.

The sponsor-submitted unpublished supportive analysis assessed risk of mortality in the combined GL and PL cohorts, finding no difference between the metreleptin with or without supportive care cohort and the supportive care alone cohort. Given the small number of deaths reported in the follow-up time available, the results in this outcome are very uncertain.

The sponsor-submitted supportive analysis also assessed end points of interest including change in hemoglobin A1C and change in fasting triglycerides. The major limitations relating to the small number of variables included for adjustment preclude drawing any conclusions from this analysis.

Studies Addressing Gaps in the Systematic Review Evidence

Contents within this section have been informed by materials submitted by the sponsor. The following has been summarized and validated by the CADTH review team.

Description of Study

Study FHA101⁴⁵ was a single-arm, multicentre, open-label, expanded-access study conducted at multiple treatment centres in the US. The primary objective was to provide metreleptin, an investigational medication, under a treatment protocol to patients with lipodystrophy associated with diabetes mellitus and/or hypertriglyceridemia. A secondary objective was to assess the long-term efficacy, safety, and tolerability of metreleptin among people with diabetes mellitus and/or hypertriglyceridemia. Participant enrolment took place between March 30, 2009, and January 23, 2016. A total of 41 participants were enrolled across 6 centres in the US.

Eligibility Criteria

Patients were eligible to enrol in Study FHA101 if they were aged 5 years or older and had physician-confirmed GL or PL with diabetes mellitus and/or hypertriglyceridemia. In addition, participants were required to adhere with the contraception and reproduction restrictions of the study. Patients were excluded from the study if they had been diagnosed with HIV infection, had acquired lipodystrophy and clinically significant hematologic abnormalities, had known infectious liver disease, or had known allergies to Escherichia coli–derived proteins or hypersensitivity to any component of the study treatment.

Interventions

All patients participating in Study FHA101 received metreleptin subcutaneously. Dosing was initially determined based on body weight and varied by age and sex.

Investigators were aware that the doses of patients’ concomitant medications (including antidiabetic medications such as insulin, or antihyperlipidemic medications) might have required adjustment upon initiation of metreleptin treatment. Patients were carefully monitored during the period of adjustments to their concomitant medication(s).

Patients who receiving insulin therapy were monitored closely during the first several weeks of treatment as insulin doses may have needed to be adjusted downward as frequently as weekly (especially for those on high doses of insulin) to avoid hypoglycemia. Consideration was to be given to an empiric decrease in insulin doses of approximately 25% upon initiation of metreleptin treatment. Patients who were receiving oral insulin sensitizers alone had less risk for hypoglycemia; investigators evaluated the need to adjust the dosages of these medications at each follow-up visit during metreleptin treatment. Similarly, investigators evaluated the need to adjust the dosages of lipid medications at each follow-up visit during metreleptin treatment for patients who were receiving lipid medications for hypertriglyceridemia. If metreleptin treatment was discontinued for any reason, further adjustment of concomitant medications could have been warranted.

Outcomes

The main efficacy outcomes assessed in Study FHA101 were hemoglobin A1C and fasting serum triglyceride levels. Other outcomes measured included levels of fasting plasma glucose and hepatic enzymes such as ALT and AST.

Hemoglobin A1C and Triglyceride Levels

Primary efficacy outcomes for Study FHA101 were the actual change from baseline in hemoglobin A1C at month 12 and percent change from baseline in fasting triglyceride levels at month 12.

Key secondary efficacy outcomes included the proportion of patients achieving target actual decreases of 1% or greater in hemoglobin A1C; or a 30% or greater decrease in fasting triglycerides at month 12, a 1.5% or greater decrease in hemoglobin A1C at month 12, or a 35% or greater decrease in fasting triglycerides at month 12; a 2% or greater actual decrease in hemoglobin A1C, or a 40% or greater decrease in fasting triglycerides at month 12.

Target decreases in both hemoglobin A1C (≥ 1%, ≥ 1.5%, and ≥ 2%) and fasting serum triglycerides (≥ 30%, ≥ 35%, and ≥ 40%) at each postbaseline visit through month 12 were assessed as an exploratory efficacy outcome.

Regarding hemoglobin A1C levels alone, the following were assessed as secondary or exploratory efficacy outcomes:

• Other secondary: Actual change from baseline in hemoglobin A1C at each postbaseline visit

• Other secondary: Proportion of patients with baseline hemoglobin A1C greater than 7% who achieve:

  • a target decrease to less than or equal to 7% postbaseline through month 12

  • a target decrease of greater than or equal to 1% or who decrease to less than or equal to 7% postbaseline through month 12

• Other secondary: Proportion of patients who achieve target actual decreases (≥ 1%, ≥ 1.5%, and ≥ 2%) in hemoglobin A1C at 2 consecutive postbaseline visits on or after month 3 through month 12

• Exploratory: Proportion of patients who achieve target actual decreases (≥ 1%, ≥ 1.5%, and ≥ 2%) hemoglobin A1C at month 3.

Regarding triglyceride levels alone, the following were assessed as secondary (or exploratory) efficacy outcomes:

• Other secondary: Percent and actual change from baseline in fasting serum triglycerides at each postbaseline visit

• Other secondary: Proportion of patients with baseline triglycerides greater than or equal to 2.26 mmol/L (≥ 200 mg/dL) who achieve:

  • a target decrease to less than 2.26 mmol/L (< 200 mg/dL) postbaseline through month 12

  • a target decrease of greater than or equal to 30% or who decrease to less than 2.26 mmol/L (< 200 mg/dL) postbaseline through month 12

• Other secondary: Proportion of patients with baseline triglycerides greater than 5.65 mmol/L (> 500 mg/dL) at baseline who achieve:

  • a target decrease to less than 5.65 mmol/L (< 500 mg/dL) postbaseline through month 12

  • a target decrease of greater than or equal to 30% or who decrease to less than 5.65 mmol/L (< 500 mg/dL) postbaseline through month 12

• Other secondary: Proportion of patients who achieve target percent decreases (≥ 30%, ≥ 35%, and ≥ 40%) in fasting serum triglycerides at 2 consecutive postbaseline visits on or after month 3 through month 12

• Exploratory: Proportion of patients who achieve target percent decreases (≥ 30%, ≥ 35%, and ≥ 40%) in fasting serum triglycerides at month 3.

Fasting Plasma Glucose

The actual and percent change from baseline for fasting glucose levels at month 12 was assessed as a key secondary efficacy outcome. The percent and actual change from baseline in fasting plasma glucose at each postbaseline visit was assessed as a secondary efficacy end point.

Hepatic Enzymes

The actual change from baseline in ALT and AST at each postbaseline visit through month 12 was assessed as a secondary efficacy outcome.

Safety was assessed by examining SAEs as well as the AEs the investigator considered to be clinically significant and/or related. Safety evaluations also included reviewing any new concomitant medications and changes in dose or frequency of existing concomitant medications, clinical laboratory evaluations, measuring body weight and vital signs (systolic and diastolic blood pressure and heart rate), and collecting blood samples for potential assessment of antileptin antibodies.

Statistical Analysis

There was no formal hypothesis and no formal statistical testing in Study FHA101. Safety and efficacy end points were summarized descriptively. Subgroup analyses were planned for the primary and key secondary efficacy end points based on the presence of metabolic abnormalities at baseline (hemoglobin A1C [< 6.5% and ≥ 6.5%], ≥ 7%, and ≥ 8% and fasting triglycerides [< 2.26 mmol/L and ≥ 2.26 mmol/L or < 200 mg/dL and ≥ 200 mg/dL; ≥ 5.65 mmol/L or ≥ 500 mg/dL]).

For analysis of change from baseline, the total number of patients by time point reflected the actual number of patients with data for that specific parameter at baseline and the specified time point. The total number of patients at a given time point depended on whether data for that time point were available and whether the study visit fell within the specified analysis visit window.

Summary statistics were presented, as well as CIs on selected parameters.

The FAS was the primary population for the analyses of primary and secondary efficacy end points, with analyses performed for the primary and key secondary end points on the efficacy-evaluable analysis set, CFAS, and CEEAS as supportive or sensitivity analyses. In general, no substitutions were made to accommodate missing data points. All data recorded in the electronic case report form were included in data listings that accompanied the Clinical Study Report. For the primary efficacy end point analysis at month 12, a LOCF method of missing data imputation was used; in addition, a WOCF method was used as a supportive analysis.

Baseline Characteristics

A summary of demographic and key clinical characteristics of patients in Study FHA101 is presented in Table 24. Of the 9 patients with GL, most were female (8 patients; 89%) and white (8 patients; 89%); the majority (6 patients, 67%) were aged 18 years and older at the time of enrolment and median age was 25 years. Of the 9 patients with GL, 2 (22%) had the congenital form of the disease and 6 (67%) had the acquired form; the subtype was not reported for 1 patient. Of the 3 patients in the GL group who were enrolled at the University of Michigan and whose baseline fasting levels were available, baseline leptin levels were ||| ||||| in all 3 patients. Median body mass index was 21.3; 8 (89%) of the ||| || patients with GL weighed more than 40 kg at study entry.

Overall, 31 (97%) of the 32 patients with PL were female, as were all the patients in the PL subgroup. The majority of the 32 patients with PL were white (69%), as were 5 (71%) of the 7 patients in the PL subgroup. At the time of enrolment, the patients with PL were older than those with GL, and they were all aged 18 years and older. Most (29 of 32 patients; 91%) had the familial form of the disease, including 6 of 7 patients in the PL subgroup. Of the 24 patients with PL who were enrolled at the University of Michigan, baseline leptin levels were available for ||||| patients with a median of |||| ||||| and ||| ||||| in the ||| || patients in the PL subgroup. Median body mass index was 30.3 kg/m² in the overall PL group and 27.6 kg/m² in the PL subgroup.

Table 24: Summary of Baseline Characteristics of Study FHA101 (Safety Analysis Set)

AGL = acquired generalized lipodystrophy; ALT = alanine aminotransferase; APL = acquired partial lipodystrophy; AST = aspartate aminotransferase; BMI = body mass index; CGL = congenital generalized lipodystrophy; FPLD = familial partial lipodystrophy; GL = generalized lipodystrophy; max = maximum; min = minimum; PL = partial lipodystrophy; SD = standard deviation; TG = triglyceride; ULN = upper limit of normal.

^aThe PL subgroup includes patients with baseline leptin < 12 ng/mL and hemoglobin A1C ≥ 6.5% and/or triglycerides ≥ 5.65 mmol/L.

Sources: Sponsor’s Summary of Clinical Evidence; Study FHA101 Clinical Study Report.⁴⁵

Patient Disposition

Patient disposition in Study FHA101 is summarized in Table 25. Overall, 24 (59%) of the 41 patients participating in the study were reported to have prematurely discontinued the study treatment, including 4 (44%) of the 9 patients with GL and 20 (63%) of the 32 patients with PL. The most common reason for early discontinuation from the study was withdrawal by the patient, reported for 1 (11%) of the patients with GL and 9 (28%) of the patients with PL. Death was reported as the cause for withdrawal for 2 patients. AEs led to the withdrawal of 3 patients in the overall PL group.

Table 25: Summary of Patient Disposition – Study FHA101

GL = generalized lipodystrophy; PL = partial lipodystrophy.

^aThe PL subgroup includes patients with baseline leptin < 12 ng/mL and hemoglobin A1C ≥ 6.5% and/or triglycerides ≥ 5.65 mmol/L.

^bAll enrolled patients who received at least 1 dose of study drug.

^cAll patients in the safety analysis set who had either primary efficacy parameter measured at baseline and at ≥ 1 postbaseline visit.

^dAll patients in the full analysis set who had controlled concomitant medication use before month 12.

^eAll patients in the full analysis set who had either efficacy parameter of interest measured at month 12 and no major protocol violations before month 12.

^fAll patients in the controlled concomitant medication full analysis set who had either efficacy parameter of interest measured at month 12 and no major protocol violations before month 12.

^gAll patients in the safety analysis set who provided at least 1 blood sample for pharmacokinetic evaluation of metreleptin.

^hAll patients in the safety analysis set who had antibody status assessed at least once while participating in the study.

Sources: Sponsor’s Summary of Clinical Evidence; Study FHA101 Clinical Study Report.⁴⁵

Exposure to Study Treatments

Patient exposure to study treatment is summarized in Table 26. In Study FHA101, the median overall duration of treatment was 21.3 months for patients in the GL group and ranged from ||| || |||| months. Median actual duration of treatment (excluding dose interruptions) was |||| months. The median average daily dose received by patients with GL was ||| mg/day. The median maximum daily dose over the study period was 5.0 mg; the maximum daily weight-based dose administered in the study was ||||| mg/kg. Because of the dose titration specified by protocol through Amendment 4 and subsequent dose adjustments based on individual clinical response, as well as evolution of the dosing regimen over time, a weighted average dose was calculated. The median weighted average daily dose received by patients with GL was 3.7 mg or 0.057 mg/kg over the study period.

In the PL subgroup, median overall duration of treatment with metreleptin was 53.1 months and ranged from |||| || |||| months; median actual duration of treatment (excluding dose interruptions) was similar at |||| months. Median average daily dose in the PL subgroup was ||| mg/day and median maximum daily dose was 10.0 mg, both higher than in the GL group. The median maximum daily dose administered in the PL subgroup was ||||| mg/kg. The median weighted average daily dose received by patients in the PL subgroup over the study period was 9.0 mg or 0.110 mg/kg.

Table 26: Summary of Patient Exposure – Study FHA101 (Safety Analysis Set)

GL = generalized lipodystrophy; max = maximum; min = minimum; PL = partial lipodystrophy; SD = standard deviation.

^aThe PL subgroup includes patients with baseline leptin < 12 ng/mL and hemoglobin A1C ≥ 6.5% and/or triglycerides ≥ 5.65 mmol/L.

^bOverall duration is from first to last dose; actual duration excludes drug interruptions.

^cDose intensity is the cumulative amount of drug exposure divide by actual treatment duration.

^dAverage daily dose is the cumulative amount of drug exposure divided by overall number of planned dosing days.

^eWeighted average dose over the study period is the sum of daily doses (in mg/kg or mg) over the overall duration at that dose or regimen divided by the total duration of all dose regimens.

Sources: Sponsor’s Summary of Clinical Evidence; Study FHA101 Clinical Study Report.⁴⁵

Concomitant Medication and Co-intervention

Concomitant medications were defined as medications used on or after the first date of study medication administration. Medications were coded using the WHO Drug Dictionary version 2016Q1. Key medications patients were receiving at the study baseline are summarized in Table 27.

Antidiabetic medications were administered to 2 (22%) of the patients with GL, 19 (59%) of the patients with PL, and 6 (86%) of the patients in the PL subgroup. Lipid-lowering therapies were administered to 19 (59%) of the patients with PL, 2 (22%) of those with GL, and 6 (86%) of those in the PL subgroup. Other commonly administered medications included lisinopril and hydrochlorothiazide for hypertension, spironolactone for edema, acetylsalicylic acid for cardiac prophylaxis, and vitamins.

Table 27: Baseline Medication Use – Study FHA101 (Safety Analysis Set)

ATC = anatomic therapeutic class; GL = generalized lipodystrophy; HMG-CoA = 3-hydroxy-3-methyl-glutaryl-coenzyme A; PL = partial lipodystrophy; WHODD = WHO Drug Dictionary.

^aThe PL subgroup includes patients with baseline leptin < 12 ng/mL and hemoglobin A1C ≥ 6.5% and/or triglycerides ≥ 5.65 mmol/L.

^bTerms across multiple WHODD ATC classes are grouped to provide overall incidence.

^cIndividual WHODD preferred terms reported with incidence > 15% in the overall GL group or PL subgroup.

Sources: Sponsor’s Summary of Clinical Evidence; Study FHA101 Clinical Study Report.⁴⁵

Efficacy

All primary evaluations of the efficacy of metreleptin were based on the FAS. Results for the primary and key secondary efficacy end points were analyzed for all 4 analysis sets. Results of the sensitivity analyses for the primary and secondary end points were consistent with those in the primary analysis; therefore, these results were not presented in this report.

Co-Primary End Point: Change From Baseline in Hemoglobin A1C and Triglycerides at Month 12

The co-primary efficacy end points were actual change from baseline in hemoglobin A1C at month 12 and percent change from baseline in fasting triglyceride levels at month 12. Results are presented in Table 28.

Among the patients with GL, mean hemoglobin A1C was reduced from 7.7% at baseline to 6.2% at month 12 (LOCF), a mean change of –1.2%, and mean fasting triglyceride concentrations were reduced from 19.9 mmol/L at baseline to 7.6 mmol/L at month 12 (LOCF), corresponding to a mean percent change of –26.9%.

Among the 29 patients in the overall PL group, mean hemoglobin A1C was reduced from 8.1% at baseline to 7.8% at month 12 (LOCF), a mean change of –0.4%, and mean fasting triglyceride concentrations were reduced from 8.5 mmol/L at baseline to 6.4 mmol/L at month 12 (LOCF), corresponding to a mean percent change of 8.7%.

Table 28: Change From Baseline to Month 12 in Hemoglobin A1C and Fasting Triglycerides Using LOCF in Study FHA101 (FAS Population)