Results from the first clinical trial of the selective GlyT1 inhibitor bitopertin for steroid-refractory Diamond-Blackfan Anemia (NCT05828108) Free

Background: Diamond-Blackfan Anemia (DBA) is an inherited macrocytic anemia. Despite an initial 80% steroid response rate, approximately 50% of patients require chronic transfusions due to relapsed/refractory disease or steroid intolerance. The primary genetic defect in DBA is ribosomal protein haploinsufficiency, which reduces globin production and causes unbalanced heme accumulation during erythrocyte hemoglobinization. Bitopertin selectively inhibits GlyT1, the erythroid-specific transporter necessary for glycine uptake in the initial, rate-limiting step of heme synthesis. We hypothesize that reducing heme by inhibiting glycine can restore erythropoiesis in DBA, supported by improved erythropoiesis in human cells in vitro, and in a haploinsufficient Rpl11 murine model (Doty et al, Blood Red Cells & Iron, in press).

Methods: This was an open label, intrapatient dose escalation of single-agent bitopertin. The primary endpoint was response (improved pre-transfusion hemoglobin or transfusion frequency) through 8 months of treatment. Secondary endpoints were safety (treatment-related serious adverse events, TRSAE) and tolerability (maximum tolerated dose).

Adult patients underwent medical, laboratory and marrow evaluations at the NIH. Bitopertin was escalated monthly (5 up to 60 mg daily), followed by 3-months maintenance. Patients underwent monthly laboratory assessments. Dose adjustments were allowed for toxicity. Eight months from initiation, patients underwent repeat evaluations at the NIH.

Results: We enrolled 15 patients from July 2023-December 2024. Follow-up completed July 2025. Six were female. One was Asian, one Black, the remainder white. Variants were identified in RPS19 (n=7), RPL5 (n=2), and RPL11, RPS10 and RPS26 (n=1 each). No variant was identified in 3 patients. RPL-mutated patients had greater absolute baseline reticulocyte counts (36 vs 9.7x10³/µL, P=0.003) and mean corpuscular volume (92 vs 84fL, P=0.005) compared with RPS-mutated patients. Sex and age distributions, ferritin, leukocyte count, and the likelihood of completing an 8-month course were comparable between mutation types (P≥0.2). All patients received chelation. Average ferritin was 1946 ng/mL (427-5478). Even accounting for erythroid hypoplasia, 14/15 baseline bone marrow specimens were hypocellular for age.

Seven serious adverse events were reported, all unlikely/unrelated to bitopertin. Three patients did not reach the 8-month endpoint (1 lost-to-follow-up, 2 for non-TRSAE). There were no early discontinuations for TRSAE. All patients completing the study reached the goal dose of 60 mg once daily. Two required subsequent reduction to 40 mg (1 grade 2 lightheadedness, 1 concern for loss of early effect). There were no responses.

Plasma drug levels were consistent with prior studies in patients and healthy volunteers. Median plasma levels at 10 and 60 mg daily were 79 and 489 ng/mL (expected: 56 and 442 ng/mL), respectively. EC50 levels (26 ng/mL) were observed for 86% of patients at 5 mg and all patients at 10 mg daily. Three-quarters of patients achieved ≥100 ng/mL at 20 mg daily. All patients at 40 mg daily achieved at least twice that level (≥200 ng/mL).

There were time-dependent and dose-dependent effects upon reticulocyte hemoglobin content. There was a 6.9 pg per µg/mL bitopertin reduction from baseline by 8 months. All patients (14) reaching 20 mg daily experienced an 11% median reduction in reticulocyte hemoglobin content from baseline (range 0-36%).

Conclusions: Bitopertin is safe and well-tolerated in steroid-refractory DBA. Bioavailability, pharmacokinetics and effects upon hemoglobin are consistent with prior experience in patients initially treated with bitopertin in large clinical trials for schizophrenia. Although bitopertin did not yield transfusion-independence in these patients, the baseline iron overload, marrow hypocellularity and prior failure of steroids reflect a cohort with long-standing disease and minimal erythropoietic marrow reserve. Ongoing work with model systems underscores the importance of this reserve while continuing to suggest efficacy in DBA, especially in steroid-dependent patients (i.e., responsive to medical intervention). Future studies may consider bitopertin as a steroid-sparing agent for steroid-responsive DBA patients.

Funding: This study was supported by the NHLBI Division of Intramural Research and a Cooperative Research and Development Agreement (CRADA) with Disc Medicine, Inc. (Watertown, MA).