Pure red cell aplasia (PRCA) is characterized by normocytic normochromic anemia, reticulocytopenia, and absence of erythroid precursors in the bone marrow [1]. It can be classified into congenital or acquired PRCA (aPRCA) according to the pathogenesis and etiologies. Acquired PRCA can be further stratified into primary and secondary ones [2]. aPRCA can be secondary to a variety of diseases including thymoma, large granular lymphocytic leukemia, parvovirus B19 infection, drug uses, hematological or immune diseases, solid tumors, etc [2, 3]. Primary aPRCA refers to those with no detectable underlying diseases [2, 4].

Currently, the first-line agents used for the treatment of aPRCA include cyclosporine A (CsA) and corticosteroid. However, corticosteroid alone does not have durable efficacy and may result in severe side effects [3]. CsA, as a first-line treatment, is widely used to treat aPRCA patients with a response rate of ~65–85% [2,3,4]. However, about 20–30% of patients treated with CsA relapse later when CsA is tapered or withdrawn [2]. In addition, some patients cannot tolerate CsA either due to severe side effects, advanced age, or renal dysfunction, and have to rely on blood transfusion.

Sirolimus (rapamycin), a lipophilic macrolide antibiotic synthesized by streptomyces hygroscopicus, has strong immunosuppressive and anti-cell proliferation activities and has been used for antitumor therapy and post-transplantation immunosuppression [5, 6]. We recently demonstrated that sirolimus is an effective treatment for patients with refractory /relapsed/intolerant aPRCA [7]. Even patients who are refractory to CsA may respond to sirolimus, indicating that sirolimus may have a different mechanism of action from that of CsA. So far, no study has been conducted on the use of sirolimus to treat newly diagnosed aPRCA.

In this prospective cohort study (NCT04470804), we evaluated the safety and efficacy of sirolimus versus CsA as the frontline therapy for patients diagnosed with primary aPRCA. A total of 68 patients were screened for eligibility. Among these, 11 patients were excluded for not meeting the inclusion criteria. The remaining 57 patients were ultimately enrolled: 27 were assigned to the sirolimus group and the other 30 were assigned to the CsA group. One patient in the sirolimus group was lost after 3-month treatment and was excluded from the final analysis. The remaining 56 patients all finished the designed treatment and were followed up for at least 1 year (Fig. 1A). The details of patient selection, therapy regimens, laboratory tests, evaluation of response and monitoring of toxicity, sample size calculation and statistical analyses in this clinical trial were included in Supplementary Methods.