rHuEpo for the treatment of anemia in myelofibrosis with myeloid metaplasia. Experience in 6 patients and meta-analytical approach

Abstract

BACKGROUND AND OBJECTIVE: Experience with recombinant human erythropoietin (rHuEPO) in the treatment of the anemia secondary to myelofibrosis with myeloid metaplasia (MMM) is slight up to now. We present our results of the treatment of 6 patients and a review of the literature in search of possible parameters predicting response to this treatment. DESIGN AND METHODS: From January 1994 to June 1996 all transfusion-dependent patients with MMM diagnosed in our hospital were included in this study. We established a minimum period of 4 weeks of treatment and a maximum of 12 if no response was observed. Initial dosages used were 100 U/kg s.c. 3 times weekly, increasing by 50 U/kg every 4 weeks where no response was observed. Response was defined as a reduction > or = 30% of the previous transfusional needs. The review of the literature was made using a MEDLINE search (January 1990-December 1996) on the keywords erythropoietin, myelofibrosis, and agnogenic myeloid metaplasia. A statistical study was made in search of possible parameters to predict response. The parameters studied include age, sex, hemoglobin, serum erythropoietin (sEPO) levels, transfusional dependency, transfusional requirements per month prior to treatment, maximum dosages used and dosage at which response was obtained. RESULTS: Only 2 of our 6 patients responded, both at a dosage of 600 U/kg/week (200 U/kg 3 times weekly s.c.). In addition to our 6 patients we have found only 28 other patients in the literature. For statistical calculation 2 of our patients were not considered as they did not complete the period of study. The overall rate of response was 17/32 (53.1%). In the univariate analysis comparing responders and non-responders we found a tendency to significance with respect to sex (p = 0.07), sEPO (p = 0.07) and transfusional needs in units of packed red blood cells per month (PRBC/m) (p = 0.13). In this way patients with low sEPO, females and those with low transfusional needs (< 3 PRBC/m) respond better. This better response in females could be explained by the fact that their disease situation was more stable (with both lower sEPO levels and transfusional dependency). The best cut-off point in the sEPO to predict response was 123 mU/mL. No important side-effects have been observed except three cases of aggravation of splenomegaly. In two cases this condition improved when the rHuEPO was discontinued. The association of rHuEPO with hydroxyurea or interferon does not seem to affect the response. INTERPRETATION AND CONCLUSIONS: Though the number of patients is low, our data suggest that some MMM patients, in particular females and individuals with low sEPO levels and with low transfusional needs, might benefit from rHuEPO in terms of elevation of hemoglobin levels. Unfortunately, transfusion dependent-patients, i.e. those in whom a beneficial effect of rHuEPO would be most welcome, are unlikely to respond, and more generally, treatment is not cost effective in medically responsive patients.