To the editor:
In a paper by Abkowitz et al,1 metoclopramide was identified as a potential new therapeutic approach for Diamond-Blackfan anemia (DBA). In order to further assess the drug efficacy, we performed a prospective study in patients included in the French DBA registry. Patients older than 2 years of age and dependent on regular transfusions were randomized at inclusion to be treated either at day 1 (arm B) or 6 weeks later (arm A); the comparison of responders in these 2 groups was planned to provide a more accurate determination of the delay of response. All patients were dependent on regular transfusion. All patients received metoclopramide for at least 16 weeks: 10 mg × 3/d in adults and 0.2 mg/kg × 3/d in children (weight < 50 kg). Packed red blood cell transfusions were prescribed for hemoglobin (Hb) values lower than 80 g/L. The definition of response relied on Hb levels, reticulocyte counts, and the intervals between transfusions that were recorded for the year before inclusion. Ten adults and 23 children have been included (Table 1). The treatment had to be stopped in 2 patients because of side effects: depression (8th week of treatment) and anaphylaxis (14th week of treatment). Other side effects were mild: most patients complained of asthenia and drowsiness, which both decreased after the first week of treatment; amenorrhea (2 patients) and bulimia (3 patients) were also observed. No increase in transfusion need occurred. None of the patients experienced a complete response. Two patients exhibited a partial response with an increase in the mean transfusion interval: from 3 to 8 weeks (patient 4) and from 4 to 5 weeks (patient 28). In these 2 patients, total duration of treatment with metoclopramide was 12 and 27 months, respectively. Last, it has to be emphasized that the 2 French responder adult patients previously included in the pilot study are still on treatment and transfusion independent more than 5 years after the initiation of metoclopramide. To date, only one other responder patient has been reported.2 Obviously, only very rare transfusion-dependent DBA patients will respond to metoclopramide and, until now, we have no clue how to identify such patients. It is of note, however, that patients 1, 2, and 3 from the Abkowitz et al1 study were not strictly resistant to steroids. Actually, patient 1, who is followed by one of us, does need the association of metoclopramide with low-dose prednisone (10 mg) to maintain the response. Lastly, our 2 partial responders are known to have previously experienced a response to steroids (Table 1). If the sensitivity to metoclopramide is actually associated with the sensitivity to steroids, the low response rate observed among transfusion-dependent patients may be related to the proportion of patients who may be classified as strictly resistant to steroids (20/33 in our cohort). Metoclopramide, which is cheap and nontoxic, may be tested in all patients with regular transfusion need, as a slight decrease in transfusion requirement may be of clinical value in patients with severe hemochromatosis. A synergy in patients with high steroid threshold remains to be prospectively tested.
Correspondence: Thierry M. Leblanc, Pediatric Hematology, Hôpital Saint-Louis, 1 avenue Claude Vellefaux, 75475 Paris, France; e-mail: email@example.com.
We thank our colleagues from the SHIP (Société d'Hématologie et d'Immunologie Pédiatrique) for inclusion of their patients in this study, Assistance-Publique Hôpitaux de Paris for promotion of the study, and Maria Daniela Arturi and DBA foundations for financial support.
Conflict-of-interest disclosure: The authors declare no competing financial interests.