Efficacy and Safety of Cyclosporine Monotherapy for Patients with Non-Severe Aplastic Anemia Not Requiring Transfusions: Results from a Prospective Multicenter Phase II Study

Introduction

Patients with non-severe aplastic anemia (NSAA) have traditionally been observed without therapy if they do not require transfusions. However, nearly two-thirds of cases show gradual progression of cytopenia and eventually become transfusion-dependent over a long period of time. Patients with long-standing NSAA tend to respond poorly to immunosuppressive therapy (IST) when their NSAA progresses to a severe form. Therefore, initiating IST for mild AA is a reasonable option. Cyclosporine (CsA) is a key drug in the treatment of AA and is known to be safe when administered at lower doses. CsA monotherapy may be more successful in the treatment of NSAA when guided by the presence or absence of immune markers, such as paroxysmal nocturnal hemoglobinuria (PNH)-type cells and HLA class I allele-lacking leukocytes (HLA-LLs), which have been shown to be associated with a good response to IST. However, the efficacy of CsA in patients with mild AA and the role of immune markers in predicting response to CsA have not been evaluated in prospective studies. Although the presence of somatic mutations, which are detected in approximately 20% of patients with NSAA, may adversely affect the response to CsA monotherapy, their influence has also not been investigated. To address these issues, we conducted a prospective clinical trial in which patients with NSAA who did not require transfusion were treated with low-dose CsA alone.

Methods

Patients with NSAA who were ≥16 years old and met all of the criteria (1) through (3) were enrolled: (1) platelet <100×10⁹/L and reticulocyte count <60×10⁹/L, with either or both hemoglobin (Hb) concentration <10.0 g/dL and neutrophil count <1.5×10⁹/L; (2) bone marrow hypocellularity as demonstrated by a biopsy or normocellularity with a decrease in the number of megakaryocytes; and (3) no dysplastic cells represented >10% of each lineage. CsA 3.5 mg/kg/day was administered orally, and the dose was adjusted to achieve a blood concentration >600 ng/mL at 2 h after CsA administration. The primary endpoint was treatment response rate at 8 weeks after CsA initiation, and the secondary endpoints were treatment response rate at 16 and 52 weeks and the incidence of grade ≥3 adverse events (AEs) related to CsA therapy. Panel sequencing of 81 genes associated with myeloid malignancies was performed on peripheral blood samples collected at diagnosis.

Results

Between January 2018 and June 2021, 33 patients from 21 institutions were enrolled; 32 cases were evaluable. The median age was 63.5 (range: 16-83) years old, and the median interval from the diagnosis to treatment was 20.5 days. The median neutrophil count was 1.20 (range: 0.59-2.68) ×10⁹/L, the Hb level was 9.8 (range: 5.8-13.5) g/dL, and the platelet count was 40.5 (range: 9-104) ×10⁹/L. A total of 20 patients completed the study. At 8 weeks, hematologic improvement, defined by increases in Hb ≥1.5 g/dL (HI-E) and a platelet count ≥30×10⁹/L (HI-P), occurred in 0 (0%) of 25 and 6 (19%) of 32 evaluable cases, respectively. HI-E and HI-P occurred in 1 (4%) of 25 and 10 (31%) of 32 patients, respectively, at 16 weeks and in 5 (20%) of 25 and 16 (50%) of 32 patients, respectively, at 52 weeks. Two of the 15 responders at 52 weeks received eltrombopag at 26 and 30 weeks and continued to the end of the study. Nine grade 3 AEs were reported in six patients, but there were no grade ≥4 AEs. PNH-type cells and HLA-LLs were detected at enrollment in 19 (61.3%) of the 31 patients and 4 (13.9%) of the 29 evaluable patients, respectively. High plasma thrombopoietin levels (≥9.02 fmol/mL) were observed in 26 (81.3%) of 32 patients. None of these markers correlated with the response to CsA. Somatic gene mutations, such as BCOR/BCORL in 4, DNMT3A in 3, and PIGA in 2 patients, were detected in 12 (40%) of 30 evaluable patients and did not affect the response to CsA.

Conclusions

This prospective study showed that low-dose CsA had limited efficacy in transfusion-naïve NSAA patients treated for 8 weeks, but moderate efficacy when the treatment duration was extended beyond 16 weeks, regardless of the presence of immune markers or somatic mutations. These findings are consistent with previous research emphasizing the importance of prolonged therapy over dose escalation in CsA therapy and also support the strategy of adding eltrombopag for patients who do not respond to CsA after 16 weeks.