CADTH Reimbursement Recommendation

Luspatercept (Reblozyl)

Indication: For the treatment of adult patients with red blood cell transfusion-dependent anemia associated with beta-thalassemia

Sponsor: Celgene Inc., a Bristol Myers Squibb Company

Final Recommendation: Reimburse with conditions

Recommendation

Summary

Key Messages

What is the CADTH reimbursement recommendation for Reblozyl?

CADTH recommends that Reblozyl should be reimbursed by public drug plans for the treatment of transfusion-dependent anemia associated with beta-thalassemia if certain conditions are met.

What are the conditions for reimbursement?

Reblozyl should only be reimbursed if it is prescribed by a hematologist and if the price is reduced by at least 85%.

Which patients are eligible for coverage?

Reblozyl should only be covered for adult patients with beta-thalassemia who require blood transfusions on a regular basis.

Why did CADTH make this recommendation?

Evidence from 1 clinical trial demonstrated that Reblozyl reduced transfusion burden by at least a 33% over a 12-week period. At the submitted price, Reblozyl is not cost-effective and will cost $33,415,422 over 3 years.

What is transfusion-dependent anemia associated with beta-thalassemia?

Beta-thalassemia is a blood disorder that is present from birth and reduces the amount of hemoglobin produced. Patients with beta-thalassemia may eventually develop organ disease, pulmonary hypertension, and often need blood transfusions throughout their whole life.

What is Reblozyl?

Reblozyl (luspatercept) is a recombinant fusion protein that promotes red blood cell growth. It is approved by Health Canada for the treatment of adult patients with red blood cell transfusion-dependent anemia associated with beta-thalassemia. Reblozyl is an injection that must be taken once every 3 weeks.

How much does Reblozyl cost?

Treatment with Reblozyl is expected to cost between $113,828 and $151,771 per patient annually depending on the dosage.

What other treatments are available for transfusion-dependent anemia associated with beta-thalassemia?

The only other treatment available is blood transfusion, which is used with iron chelators (deferasirox, deferiprone, and deferoxamine mesylate) to manage iron overload that happens because of the transfusion. Patients taking Reblozyl may still need these treatments.

Unmet needs in transfusion-dependent anemia associated with beta-thalassemia

There is a need for more tolerable treatments that reduce the treatment burden for patients with beta-thalassemia.

How much do other treatments cost?

The cost of iron chelation therapy ranges from $4,524 to $87,272 per patient annually.

Recommendation

The CADTH Canadian Drug Expert Committee (CDEC) recommends that luspatercept should be reimbursed for the treatment of adult patients with red blood cell (RBC) transfusion-dependent anemia associated with beta-thalassemia, only if the conditions listed in Table 1 are met.

Rationale for the Recommendation

One phase III, randomized, double-blind, placebo-controlled study (BELIEVE, N = 336) in adult patients with transfusion-dependent anemia associated with beta-thalassemia demonstrated that treatment with luspatercept in addition to best supportive care (BSC) was associated with a statistically significant and clinically meaningful reduction in transfusion burden compared with placebo. In the BELIEVE study, 21.4% of patients in the luspatercept group achieved an erythroid response, defined as at least a 33% reduction from baseline in RBC transfusion burden, with a reduction of at least 2 RBC units during the protocol-defined period of 12 consecutive weeks (week 13 to week 24), compared with 4.5% of the patients in placebo group (P < 0.0001). Similarly, the proportion of patients who achieved an erythroid response of greater than or equal to a 50% reduction from baseline in transfusion burden in the same period was 7.6% in the luspatercept treatment group versus 1.8% in placebo group (P < 0.05). Patients expressed to CADTH the need for a treatment that reduces the burden and adverse effects of transfusion and improves health-related quality of life (HRQoL). Based on the evidence reviewed, luspatercept meets the need for reduction in transfusion burden, but no differences were observed on measures of HRQoL, and symptom reduction was not assessed.

The sponsor’s submitted price of luspatercept is $2,189 per 25 mg and $6,567 per 75 mg. The recommended dose of luspatercept depends on treatment response; therefore, the average daily treatment ranges from $312.71 to $416.95, while the average annual cost of treatment is between $113,828 and $151,771 per patient. CADTH estimated the incremental cost-effectiveness ratio (ICER) of luspatercept compared with BSC to be $659,395 per QALY, with a 0% probability of being cost-effective at a $50,000 per QALY willingness-to-pay (WTP) threshold at the price submitted by the sponsor. A price reduction of 85% would be required for luspatercept to be cost-effective at this threshold. Scenario analyses were conducted to assess the impact of clinical uncertainty with luspatercept (examining the predictive value of serum ferritin, mortality benefit of luspatercept, and treatment durability of luspatercept). The scenario analyses resulted in ICERs greater than $1 million per QALY.

Table 1: Reimbursement Conditions and Reasons

Reimbursement condition

Reason

Initiation

1. Adults with RBC transfusion-dependent anemia associated with beta-thalassemia.

The results of the BELIEVE study demonstrated that luspatercept is superior to placebo in reducing transfusion burden in adult patients with RBC transfusion-dependent anemia associated with beta-thalassemia.

2. Patients must be receiving regular transfusions, defined as:

  • 6 to 20 RBC units in the 24 weeks prior to initiating treatment with luspatercept, and

  • No transfusion-free period greater than 35 days in the 24 weeks prior to initiating treatment with luspatercept.

These conditions are based on the inclusion criteria of the BELIEVE study.

Renewal

1. Patients should be assessed for a response to luspatercept every 6 months.

  • Initial response is defined as a ≥ 33% reduction in transfusion burden (RBC units/time) and should be compared to the pre-treatment baseline RBC transfusion burden, measured over 24 weeks before initiating treatment with luspatercept.

  • At each subsequent assessment, a reduction in transfusion burden of ≥ 33% compared to the pre-luspatercept transfusion burden must be maintained.

The duration of the BELIEVE study was not adequate to determine how long the treatment effect of luspatercept would be maintained.

The definition of response to treatment as a ≥ 33% reduction in transfusion burden versus the pre-treatment baseline aligns with clinical expert opinion and the Health Canada product monograph. This definition is also consistent with the primary outcome used in the BELIEVE study.

Discontinuation

1. Luspatercept should be discontinued if a patient does not respond after 9 weeks of treatment (3 doses) at the maximum dose. A response is defined as per renewal condition 1.1.

The discontinuation of luspatercept after 9 weeks is aligned with the Health Canada product monograph.

Prescribing

1. The patient should be under the care of a hematologist with experience in managing patients with beta-thalassemia.

Accurate diagnosis by a clinician with experience and expertise in treating and managing beta-thalassemia is important to ensure that luspatercept is prescribed only for appropriate patients and to ensure that patients receive optimal care and monitoring.

2. The maximum dose of luspatercept should not exceed 1.25 mg/kg (or 120 mg total dose) per administration.

In the BELIEVE study, the dose levels were titrated stepwise up to a maximum of 1.25 mg/kg, and the maximum total dose per administration was not to exceed 120 mg.

Pricing

1. Price reduction

Based on CADTH reanalyses, the ICER of luspatercept plus BSC compared with BSC for patients with beta-thalassemia was $659,395 per QALY, with a 0% chance of being cost-effective at a WTP threshold of $50,000 per QALY. A price reduction of at least 85% is needed to meet the $50,000 WTP threshold.

ICER = incremental cost-effectiveness ratio; QALY = quality-adjusted life-year; RBC = red blood cell; WTP = willingness to pay.

Implementation Guidance

  1. Regular access to a hematologist for luspatercept administration may be limited for some patients. There is potential for luspatercept to be administered by a health care professional in other settings, such as a community pharmacy.

Discussion Points

Background

Luspatercept has a Health Canada indication for treatment of adult patients with RBC transfusion-dependent anemia associated with beta-thalassemia. Luspatercept is a recombinant fusion protein consisting of 2 identical chains, each consisting of a modified form of the extracellular domain of human activin receptor type IIB (ActRIIB) linked to the human immunoglobulin G1 (IgG1) Fc domain. Luspatercept is available as lyophilized powder for solution for subcutaneous injection in 2 strengths: 25 mg/vial and 75 mg/vial. The Health Canada recommended starting dose of luspatercept is 1 mg/kg up to a maximum of 1.25 mg/kg administered by a subcutaneous injection every 3 weeks.

Summary of Evidence

To make their recommendation, CDEC considered the following information:

Summary of Patient Input

Two patient groups, TFC and CORD, conducted focus groups and an online survey to solicit patient input. The following is a summary of key input from the perspective of the patient groups:

Clinical Trials

The CADTH systematic review included 1 ongoing phase III, multi-centre, double-blind randomized placebo-controlled study (BELIEVE, N = 336) evaluating the efficacy and safety of luspatercept in adult patients with transfusion-dependent anemia associated with beta-thalassemia. The BELIEVE trial had a 12-week screening/run-in period during which the patient’s prior 24-week transfusion history was documented. Following the 12-week screening/run-in phase, eligible patients were randomized (2:1) to receive either luspatercept or placebo along with BSC for 48 weeks in a double-blind manner. BSC included RBC transfusions, ICTs, antibiotic, antiviral, and antifungal therapies; and/or nutritional support as needed.

Patients received a starting dose of 1 mg/kg of the study drug administered by a subcutaneous injection every 3 weeks for 48 weeks. During this period, the dose levels were titrated (increased) stepwise up to a maximum of 1.25 mg/kg or reduced based on a clinical response. The maximum total dose per administration was not to exceed 120 mg.

The baseline characteristics of the patients enrolled in the BELIEVE study were overall well balanced. The mean (SD) age of the patients was 32.2 (10.67) and 31.9 (9.89) in the luspatercept and placebo groups, respectively, and 58.9% of the patients in the luspatercept treatment group and 56.3% of the patients in the placebo group of the study were females. The percentage of patients who discontinued from the study was 12.5% in the placebo treatment group and 11.6% in the luspatercept treatment group.

The main limitations were potential for unblinding due to lack of efficacy or adverse events, missing data for some end points (i.e., HRQoL), and inadequate duration of the trial to demonstrate maintenance of the treatment effect of luspatercept in the long-term.

Outcomes

Outcomes were defined a priori in CADTH’s systematic review protocol. Of these, CDEC discussed the following: hematologic response, HRQoL, and iron accumulation.

Hematologic Response

Hematologic response was assessed by RBC transfusion burden, transfusion frequency, and RBC units.

The primary outcome of BELIEVE was based on an erythroid response of greater than or equal to 33% reduction from baseline in transfusion burden (RBC units/time) with a reduction of at least 2 units, in the fixed 12-week period from week 13 to week 24. The secondary outcomes assessed an erythroid response greater than or equal to 33% reduction from baseline with a reduction of at least 2 units in the fixed 12-week period from week 37 to week 48, greater than or equal to 50% reduction with a reduction of at least 2 units in the fixed 12-week period from week 13 to week 24 and from week 37 to week 48, and the mean change from baseline in RBC transfusion burden to the fixed week 13 to week 24 interval.

Health-Related Quality of Life

HRQoL was assessed as mean change from baseline at week 24 and 48 using the Short Form (36) Health Survey (SF-36) and TranQoL instruments.

The TranQoL Questionnaire is a disease-specific questionnaire for adults and children with thalassemia major that focuses on quality of life issues related to transfusion burden. The TranQoL assesses the following 4 domains: physical health, emotional health, family functioning, school and career functioning, and physical health. The total score and domain scores range from 0 (worst) to 100 (best). Internal consistency, test-retest reliability, and reliability of individual TranQoL domains were acceptable. The ability for the TranQoL to detect a meaningful change in quality of life was determined as patients who rated their quality of life as better had a 4.0-point (SD 9.0) improvement in TranQoL scores, from baseline of 67.1 to 71.1 points 1 week later (P = 0.008). A MID for the TranQoL in patients with transfusion-dependent thalassemia was not identified in the literature.

The SF-36 is a 36-item, generic, self-reported questionnaire that is scored from 0 to 100 and has been used extensively in clinical trials in many disease areas. The SF-36 consists of 8 domains: physical functioning, role physical, bodily pain, general health, vitality, social functioning, role emotional, and mental health. For each of the 8 domains, a subscale score can be calculated. The SF-36 also yields 2 summary measures of physical health (the Physical Component Score measure) and mental health (the Mental Component Score measure) derived from scale aggregates. Higher global scores are associated with better quality of life. The scores can also be standardized to the general US population, where an average score is 50. Validity, reliability, and responsiveness for patients with transfusion-dependent thalassemia was not identified in the literature for the English version of the scale.

Iron Accumulation

Iron accumulation was assessed using liver iron concentration (LIC) and myocardial iron concentration (MIC) using T2* MRI and were included in the CADTH review. The LIC was measured at 3 time periods: baseline, 24-weeks, and 48-weeks. MIC was measured at 2 time points: baseline and 48-weeks.

Efficacy

In the BELIEVE study, the primary efficacy outcome demonstrated that a significantly greater proportion of patients treated with luspatercept exhibited at least a 33% reduction from baseline in transfusion burden during the fixed 12-week period from weeks 13 to 24 than in the placebo group. In the luspatercept treatment group 21.4% of the patients achieved the primary end point compared with 4.5% of the patients in placebo group (difference in proportions [95% CI], 17.0 [10.4 to 23.6]; P < 0.0001).

In a fixed 12-week period from week 37 to week 48, 19.6% of the patients in the luspatercept treatment group and 3.6% of the patients in placebo group achieved an erythroid response of greater than or equal to 33% reduction from baseline in transfusion burden (units RBCs/time) with a reduction of at least 2 units (difference in proportions [95% CI], 16.1 [9.8 to 22.3]; P < 0.0001). The proportion of patients who achieved an erythroid response of greater than or equal to 50% reduction from baseline in transfusion burden (units RBCs/time) with a reduction of at least 2 units during the fixed 12-week period from week 13 to week 24 was 7.6% in the luspatercept treatment group, and 1.8% in placebo group (difference in proportions [95% CI], 5.8 [1.6 to 10.1]; P = 0.0303). In the fixed 12-week period from week 37 to week 48 the proportion of patients who achieved an erythroid response of greater than or equal to 50% reduction from baseline in transfusion burden (units RBCs/time) with a reduction of at least 2 units, was 10.3% in the luspatercept group and 0.9% in the placebo group (difference in proportions [95% CI], 9.4 [5.0 to 13.7]; P = 0.0017). The clinical experts consulted by CADTH suggested that the 33% or greater and 50% or greater reduction in transfusion burden were clinically meaningful outcomes.

HRQoL as measured by the TranQoL and SF-36 were outside of the statistical testing hierarchy and were therefore considered as supportive evidence. Overall, luspatercept did not show benefit in terms of HRQoL when evaluated against the placebo treatment group.

Iron accumulation was measured through LIC, myocardial concentration, and serum ferritin. These outcomes were outside of the statistical testing hierarchy and hence were considered as supportive evidence. The iron accumulation was measured as a change from baseline. The clinical experts consulted by CADTH suggested that LIC and MIC were more reliable indicators of iron overload as compared to serum ferritin levels, as there are frequently large fluctuations with this measurement.

Harms (Safety)

In BELIEVE, 96.0% and 92.7% of the patients in the luspatercept and placebo treatment group, reported at least 1 adverse event (AE) respectively. In the luspatercept treatment group 4% of the patients and 0.9% of the patients in the placebo group reported experiencing at least 1 thromboembolic event. The most commonly occurring AEs in luspatercept and placebo treatment group, respectively, were back pain (27.4% and 29.4%), upper respiratory tract infection (26.5% and 33.0%), headache (26.0% and 23.9%), and bone pain (19.7% and 8.3%).

In BELIEVE, serious adverse events (SAEs) were reported by 15.2% of patients in the luspatercept treatment group and 5.5% of patients in the placebo group. The most commonly reported SAE was infections and infestations, reported by 5.8% of patients in the luspatercept group and 2.8% of patients in the placebo group. The proportion of patients who stopped treatment due to an AE was 5.4% and 0.9% in the luspatercept and placebo treatment group, respectively.

One patient died in each treatment group.

Cost and Cost-Effectiveness

Luspatercept is available at a cost of $2,189 per 25 mg and $6,567 per 75 mg powder. The recommended dose is between 1.0 and 1.25 mg/kg every 3 weeks, leading to an average daily cost of $312.71 to $416.95 per patient (or $113,828 to $151,771 annually).

The sponsor submitted a cost-utility analysis comparing luspatercept and BSC, consisting of RBC transfusions and ICT) to BSC alone, for the treatment of adult patients with transfusion-dependent anemia associated with beta-thalassemia. The modelled population aligned with the Health Canada indication and reimbursement request. A semi-Markov model was submitted that included a decision tree of up to 48 weeks followed by a Markov model. The decision tree informed whether patients would be considered responders to luspatercept. The Markov model included mutually exclusive health states based on transfusion burden (complication-free), as well as states pertaining to cardiac, liver, and endocrine complications within these states, and a death state. Transitions into the complication states were based on hazard ratios associated with serum ferritin levels and complication states were associated with an increased risk of death. The model was run over the patient’s lifetime (70 years).

The following key limitations were identified:

CADTH undertook reanalyses to address several key limitations of the sponsor’s model, including more conservative assumptions about luspatercept efficacy as well as transfusion burden for those receiving BSC after 48 weeks, alternative utility estimates, use of the full intention-to-treat population, increase to the average dose intensity, use of a fixed response criteria, and use of the January 2019 data cut of BELIEVE.

In the CADTH base case, the ICER was $659,395 per QALY compared with BSC. Based on CADTH reanalyses, the probability of luspatercept being cost-effective at a WTP threshold of $50,000 per QALY was 0%. A price reduction of at least 85% would be required for luspatercept to be cost-effective at this threshold.

A key driver of the analysis are the assumptions around the long-term use and effects of luspatercept. When exploring the impact of a loss of treatment efficacy after 5 years, luspatercept was dominated by BSC (luspatercept is associated with greater costs and less QALYs); and, after 10 years, the ICER for luspatercept compared with BSC was $1,352,159 per QALY. When the predictive nature of SF levels was removed, the resulting ICER was $1,398,609 per QALY. And, when the mortality benefit of luspatercept was removed, the resulting ICER was $1,198,773 per QALY. This highlights the impact of clinical uncertainty on the cost-effectiveness of luspatercept.

Members of the Canadian Drug Expert Committee

Dr. James Silvius (Chair), Dr. Ahmed Bayoumi, Dr. Sally Bean, Dr. Bruce Carleton, Dr. Alun Edwards, Mr. Bob Gagne, Dr. Ran Goldman, Dr. Allan Grill, Mr. Allen Lefebvre, Dr. Kerry Mansell, Ms. Heather Neville, Dr. Danyaal Raza, Dr. Emily Reynen, Dr. Yvonne Shevchuk, and Dr. Adil Virani.

Meeting date: April 21, 2021

Regrets: Three expert committee members did not attend.

Conflicts of interest: None