Safety and Efficacy of Luspatercept in Low-Risk Myelodysplastic Neoplasms with Anaemia

Background: Therapy of anaemia in people with low-risk Myelodysplastic Neoplasms (MDNs) is challenging. Luspatercept targets ineffective erythropoiesis.

Objective: Define the safety and efficacy of luspatercept in people with low-risk MDNs and anaemia.

Methods: A meta-analysis of prospective clinical trials. Relevant studies were identified through PubMed, Embase, Web of Science and the Cochrane Library up to January 31, 2024. Response rates was analyzed using fixed- or random-effects models. Analysis were done to identify the co-variates correlated with efficay. Adverse events were grade using National Cancer Institute Common Terminology Criteria for Adverse Events. Response was graded by whether achieving RBC transfusion independence for a duration of ≥ 8 weeks.

Results: The meta-analysis retrieved 935 references and ultimately included 4 studies： the PACE-MDNs-base, PACE-MDNs-extension, MEDALIST, and COMMANDS Overall, 497 patients who received Luspatercept treatment were included. The median age of the patients was 72.5 years (range, 27-95 years), and 62% of the patients were men. The mean time since the original diagnosis of MDNs was 17.1 months (range, 0-421 months). With regard to the IPSS-R categories, 8%, 65%, and 23% of the patients had an MDNs defined as very low-risk, low-risk, and intermediate-risk, respectively. Baseline serum EPO levels were less than 200 U/L, 200 to 500 U/L, and > 500 U/L in 63%, 24% and 12% of the patients, respectively. The baseline transfusion burden was ≤ 4 units per 8 weeks in 42% of the patients and > 4 units per 8 weeks in 51%; 7% of the patients did not have a baseline burden. RS-positive status and SF3B1 mutations were detected in 78% and 67% of the patients, respectively.

In this analysis, achieving RBC-TI for a duration of ≥ 8 weeks (equivalent to 56 consecutive days on-study) was considered indicative of therapeutic efficacy. There was significant heterogeneity in the response results among the four studies (Tau2=0.034, Chi2=50.22, p <0.01, I2=94.03%). Thus, random-effects models were applied, and the pooled estimates of the RBC-TI ≥ 8 weeks rate was 52.3% (95% confidence interval [CI]: 0.335-0.707) for all patients.

Usng the Chi-squared test, the impact of baseline characteristics on the efficacy of Luspatercept was assessed. Our findings revealed no statistically significant correlation between age, gender, RS status, SF3B1 status, IPSS-R classification, prior ESA treatment, and the effectiveness of Luspatercept. However, both baseline RBC transfusion burden and serum erythropoietin (EPO) concentration exhibited a significant relationship with Luspatercept efficacy (p < 0.001 and p < 0.001, respectively). Additionally, we conducted Chi-squared tests within each clinical trial to examine the relationship between the RS status or SF3B1 status and Luspatercept efficacy. Our analysis revealed no statistically significant correlation between the RS status and response to Luspatercept in either case. Only one clinical trial (COMMANDS) revealed a correlation between SF3B1 mutation status and response to Luspatercept(p=0.033), whereas the two remaining clinical trials did not show a significant correlation.

The commonly reported AEs across the four clinical trials involving Luspatercept were as follows: fatigue (16%), diarrhea (14%), nausea (12%), bone pain (12%), peripheral edema (11%), and headache (10%). Serious adverse events included hypertension (4%) and fatigue (2%). These AEs were relieved with symptomatic treatment. Additionally, the rates of progression to AML and high-risk MDNs were 1.7% (8/464) and 1.0% (3/286), respectively.

Conclusions: Luspatercept was safe and effectiveive in people with low-risk MDNs with anaemia. Most adverse events were low-grade and mangeable. Low transfusion frequency and low serum erythropoietin concentration but ring sideroblasts nor SF3B1 mutation correlated with efficay.

PROSPERO CRD registration number: CRD42024513235

Keywords: Myelodysplastic Neoplasms, Anemia, Luspatercept, Transfusion, Meta-analysis

Key Points：

1. Luspatercept was safe and effectiveive in people with low-risk MDNswith anaemia.

2. Low transfusion frequency and low serum erythropoietin concentration but ring sideroblasts nor SF3B1mutation correlated with efficay.

No relevant conflicts of interest to declare.