The clinical value of erythropoietin in patients with cancer.

Erythropoietin has been successfully used in the treatment of cancer-related anaemia. About two-thirds of patients with the 'anaemia of chronic disorders', anaemia due to neoplastic bone marrow infiltration or therapy-related anaemia, are expected to respond to high doses of erythropoietin with a haemoglobin increase of at least 2 g/dl. In the myelodysplastic syndromes, about one-third of patients will show a response when very high doses of erythropoietin are combined with granulocyte colony-stimulating factor. The response to erythropoietin is slow, requiring several months to develop. Various factors have been reported to predict a response, but the prediction models proposed are contradictory and have not been prospectively validated. Therefore, the most common strategy to determine the responsiveness of cancer-related anaemia to erythropoietin is to subject the patient to a treatment trial of several months' duration. Treatment with erythropoietin needs to be compared with the transfusion of red blood cells, which has similar effects on the patient's haemoglobin level. Erythropoietin is a generally well tolerated drug, but it is slow to exert an effect and ineffective in a substantial proportion of patients. Red blood cell transfusion is associated with a small risk of infectious, allergic or toxic complications, but it leads to a rapid haemoglobin increase in virtually all patients treated. Cost and cost-benefit analyses from several countries indicate that, in patients with cancer-related anaemia, treatment with erythropoietin is considerably more expensive than the transfusion of allogeneic red blood cells. Thus, the choice between the two treatment options will be influenced by the financial resources of the respective healthcare systems.