Authors' objectives: The Academic Technology Assessment Group (ATAG) reviewed a report by the manufacturer (Bristol- Myers Squibb K.K.) on the additional benefits and cost-effectiveness of luspatercept for low-risk myelodysplastic syndromes (MDS) with anemia. This report summarizes the ATAG review and reanalysis. The target population was divided into four groups: (a) erythropoiesis-stimulating agent (ESA)-naive patients with ring sideroblast (RS), (b) ESA-naive patients without RS, (c) patients with RS resistant to, intolerant of, or ineligible for ESA, (d) patients without RS resistant to, intolerant of, or ineligible for ESA. The comparator was darbepoetin alfa, best supportive care (BSC) with red blood cell (RBC) transfusions for populations (a) and (b), and BSC with RBC transfusions for (c) and (d). The intervention was luspatercept, BSC with RBC transfusions. To assess additional benefits for populations (a) and (b), the manufacturer cited the luspatercept clinical trial (COMMANDS). They concluded that luspatercept had additional benefits over the comparator in population (a) because it produced a superior response regarding RBC transfusion independence. In contrast, they determined that no additional benefit was demonstrated in population (b). For population (c), they evaluated that luspatercept demonstrated additional benefits based on another trial (MEDALIST). For population (d), their systematic review (SR) focusing solely on randomized controlled trials (RCT) reported that it was impossible to assess additional benefits because no relevant trials were identified. ATAG conducted an SR independently. Because ATAG's SR results for populations (a), (b), and (c) were generally consistent with the manufacturer's results, ATAG concluded that the manufacturer's assertion regarding additional benefits in these populations was acceptable. For population (d), ATAG considered the manufacturer's SR insufficient and expanded it to non-RCTs and found two single-arm trials (PACE-MDS and MDS-003) partially including population (d). However, the small sample size and heterogeneity of study designs made them unsuitable for assessing additional benefits. Furthermore, ATAG examined clinical trials registered in databases, including ongoing ones, but none were identified. Thus, ATAG concluded that it was not feasible to evaluate additional benefits of luspatercept in population (d). ATAG considered it appropriate to conduct a cost-effectiveness analysis for populations (a) and (c), and a cost-minimization analysis for population (b). In the cost-effectiveness evaluation, the manufacturer used a Markov model with five health states: "transfusion dependent," "transfusion independent," "high-risk MDS," "acute myeloid leukemia," and "death," with quality-adjusted life year (QALY) as the outcome. For population (b), ATAG noted that the manufacturer used treatment continuation rates from the overall population in COMMANDS and applied them separately to each arm. ATAG considered it more appropriate to use data specific to population (b) and apply pooled continuation rates combining two arms. Accordingly, ATAG reanalyzed populations (a) and (b), using the updated price of darbepoetin alfa. The reanalysis showed that, compared to darbepoetin alfa, BSC with RBC transfusions, luspatercept had an incremental cost of JPY 23,280,337 and 0.85 QALYs gained in population (a), with an incremental cost-effectiveness ratio (ICER) of JPY 27,268,507/QALY; and an incremental cost of JPY 14,089,839 in population (b). The reanalysis also showed that, compared to BSC with RBC transfusions, luspatercept had an incremental cost of JPY 17,299,887 and 0.42 QALYs gained in population (c), with an ICER of JPY 41,138,889/QALY. For population (d), ATAG determined that analysis was infeasible. In conclusion, these results suggest that the ICER for luspatercept compared to each comparator is likely to belong to the more than JPY 10 million/ QALY interval for population (a), the equivalent (or inferior) in effectiveness and expensive interval for population (b), and the more than JPY 10 million/QALY interval for population (c) from the perspective of public healthcare payers in Japan.

Project Status: Completed

URL for project: https://c2h.niph.go.jp/en/

Year Published: 2025

URL for published report: https://c2h.niph.go.jp/en/results/C2H2401.html

English language abstract: An English language summary is available

Publication Type: Not Assigned

Country: Japan

Organisation Name: Center for Outcomes Research and Economic Evaluation for Health

Contact Address: 2-3-6 Minami, Wako-shi, Saitama 351-0197 Japan

Contact Name: Takeru Shiroiwa

Contact Email: t.shiroiwa@gmail.com