Patterns of recruitment into acute myeloid leukaemia (AML) 15 and outcome for young patients with AML at a single referral centre.

This study assessed the recruitment to an acute myeloid leukaemia (AML) trial (AML15) in a single centre, evaluated whether outcome was influenced by trial entry and whether the trial population could be considered representative of all AML patients by retrospective comparison of patient characteristics, trial entry and outcome for 81 consecutive patients (<60 years). All patients were considered for trial entry, however the trial was not offered to 12 (15%) patients. These patients had a worse outcome than the 69 (85%) patients that were invited to participate (P = 0.04). Sixteen patients (23%) invited to participate in the trial declined and were treated on equivalent protocols. These patients had a similar outcome to those who accepted entry into the trial (P = 0.2). These results suggested that physicians exert a selection bias when evaluating patients for trial entry. Thus the overall survival estimates generated from large phase III trials may indicate that the outcome for patients with AML is better than the outcome experienced in the 'real' world. Furthermore, patients who are considered appropriate for randomization into a trial, but decline entry, experience a similar outcome to those treated on trial when treated in an equivalent manner.