The prevalence of anaemia in patients with cancer lies between 10 and 40%, depending on the type of tumor and chemotherapy. Anaemia has a significant impact on the quality of life, along with pain or disease progression. There are multiple causes but the physiopathology resembles that of inflammatory anaemia. The following mechanisms can be distinguished: a resistance of the erythroid precursor cells (BFU-e, CFU-e) to erythropoietin, an inappropriately decreased renal erythropoietin secretion for a given haemoglobin value and alterations of the iron metabolism leading to a functional iron deficiency. Recombinant human erythropoietin (r-hu-EPO) is safe and efficient in the treatment of anaemia of chronic renal failure and rheumatoid arthritis. In oncology different phase I and II studies have demonstrated an efficacy (increase of haemoglobin, decrease of transfusion requirements) in about 50% of all adult patients. A response to a subcutaneous r-hu-EPO treatment with a relatively high posology of 150 U/kg three times a week can be expected after one to two months. No single reliable parameter will predict a response to the r-hu-EPO treatment. Several phase III studies confirm that anaemia in cancer patients undergoing chemotherapy (notably with cisplatin) can be corrected in 40 to 60% of all cases and that the haemoglobin increase improves the quality of life. Finally, recent clinical trials suggest that an early r-hu-EPO treatment might prevent the occurrence of anaemia secondary to chemotherapy. Several parameters will have to be specified such as the precise definition of the groups at risk, the appropriate haemoglobin level to initiate a r-hu-EPO treatment, its optimal posology, as well as the role of the iron substitution and its route of administration. The impact of the r-hu-EPO treatment on the quality of life of cancer patients constitutes a priority for future studies, which will have define the exact role of r-hu-EPO in oncology management.