This site is intended for health professionals only
Anaemia is one of the most frequent side-effects of cancer treatment. Anaemia is usually defined when haemoglobin (Hb) levels fall below 12g/dl but WHO first grade is defined as Hb levels ≤10g/dl. Recommendations have been published proposing symptomatic treatment for anaemia based on red blood cell transfusions and deficiencies monitoring. Erythropoietin (EPO) constitutes the only preventive and curative treatment. Kidney cells release EPO when oxygen levels are low in the kidneys. EPO acts on bone marrow erythrocyte progenitors. According to international consensus guidelines,(1) a target Hb level of 12g/dl (or a gain of about 2g/dl) should be reached after four weeks of treatment, with treatment being started when Hb levels are ≤10.5g/dl.
Nevertheless, no recommendations are currently available for paediatric indications, radiotherapy and haematological cancers treated without chemotherapy. In 2004, recommendations were elaborated by a medical and pharmaceutical committee at Saint-Louis Hospital, based on the international guidelines for EPO prescription in oncology inpatients.
The objectives of this prospective study, carried out from January to June 2005, were twofold. First, current practices in oncology and haematology were listed and compared with national and international guidelines. Simultaneously, Hb levels of patients treated with EPO were retrospectively analysed in order to measure efficacy. Second, the potential costs associated with misuse were estimated.
Materials and methods
Five cancer units collaborated in this audit. No control existed for EPO before January 2005 but since then and following agreement from a therapeutic committee, a prefilled prescription has been implemented, allowing injections of epoietin beta (30,000UI) or darbepoietin alfa (150mg) in chemotherapy-induced anaemia on a weekly schedule when Hb levels were <11g/dl. Doctors had to give details on the patient’s characteristics (first name, age, weight and unit), and clinical and bibliographical proofs were required when anaemia was not chemotherapy-induced. Prescriptions were analysed by a pharmacist who verified the written information and asked for more details when required.
Hb levels at days –1, 7 (potential premature efficacy) and 30 were recorded by a pharmacist on specific software. Patients who suffered from renal failure were not included. Transfusions needs and chemotherapy schedules (especially with platinum) were recorded for each patient receiving EPO.
Costs were first evaluated from the French healthcare system perspective, based on the number of syringes used for nonvalidated prescriptions and valued on local prices, and then compared with global EPO costs at Saint-Louis Hospital.
Results and discussion
One hundred and seventy-seven prescriptions were analysed for 95 patients (46.2% of which were female; average age 60.7 years [age range: 19–95]) receiving EPO (40% of them receiving it twice or more). A total of 10.5% of the patients were treated with platinum-based chemotherapy.
Conformity analyses revealed two types of misuse: clinical (wrong indication) and biological (Hb levels) – 15.8% of prescriptions were made for the wrong indication, 89.2% of which were made for haematological pathologies such as multiple myeloma, myelodysplastic syndrome, chronic lymphoid leukaemia, myelofibrosis or myeloproliferative syndrome. Although efficacy was assessed through the monitoring of Hb levels, data were lacking for days –1, 7 and 30 (23.0%).
Comparisons of Hb levels between day –1 and day 30 showed that EPO was correctly prescribed (ie, Hb level <11g/dl) in 76.8% of cases; however, they also showed that it is difficult to reach the target Hb level of >12g/dl at day 30 (only 2.7%). It is important to note that 10.9% of the prescriptions had been made for Hb levels <8g/dl, which is an indication for transfusion. This, added to misuse related to prescription when Hb levels are >11g/dl, amounts for 28% of misuses identified. In addition, 63.2% of prescriptions were made for patients with Hb levels between 8 and 10.5g/dl, for which EPO treatment is not the optimal one according to a recent meta-analysis.(2) The results of our audit confirm this study, as a large percentage of patients with Hb levels in this range present with a decrease in their Hb level at day 30. However, none of these patients had a double dose of EPO after four weeks.
EPO treatment led, at day 30, to an increase of 1–2g/dl in 20.8% of patients, with 48.5% of patients having a decrease in their Hb level and 9.9% having a gain >2g/dl. The target was a 2g/dl increase in four weeks but a more modest gain appears to be a reasonable target in case of severe anaemia.
Finally, monitoring Hb levels at day 7 is not relevant because of the lack of data and the very low percentage of patients presenting with raised Hb levels at that stage.
In the six-month study, the total cost of EPO treatment was estimated at €58,973. If we consider the 38.3% of patients in which EPO treatment was not required, the total costs that could be avoided reached €22,587 (if validated haematological indications such as multiple myeloma or myelodysplastic syndrome are not included, the amount reaches €20,384). Savings could be made by coprescribing red blood cell transfusions.
This audit allowed us to identify that 38.3% of cases of EPO are wrongly prescribed, mostly due to different practices and use of guidelines, even at local level. Sixty-eight syringes were administrated outside of the clinical and haematological recommendations range (this result has to be contrasted with the 5,500 EPO syringes sold in the Saint-Louis Hospital out‑patients unit).
Following this audit, medical and pharmaceutical staff at the hospital will be drafting new recommendations, to try and reach a consensus regarding indications in haematology and the coprescription of transfusions in some cases.