Analysis of factors influencing inclusion of 102 patients with stage III/IV Hodgkin's lymphoma in a randomized trial for first-line chemotherapy

Background: Data on factors influencing inclusion of Hodgkin's lymphoma (HL) patients in randomized clinical trials (RCT) are limited and, for the present study they were analyzed in a RCT for III/IV HL. Patients and methods: All patients with stage III/IV HL referred to the Saint-Louis Hospital between January 2003 and May 2007 were studied. A Groupe d'Etudes des Lymphomes de l'Adulte/European Organisation for Research and Treatment of Cancer RCT, to compare ABVD (doxorubicin, bleomycin, vinblastine, dacarbazine) with increased-dose BEACOPP (bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, prednisone), was open for recruitment. Noninclusion criteria and physician's reasons for non-recruitment were prospectively recorded. The reasons for patient's refusal were collected retrospectively. Logistic regression analyses were carried out in order to identify factors predicting inclusion. Results: A total of 102 patients were diagnosed, among whom 51% were included. Seven patients were ineligible, 22 refused to participate, and 21 were not enrolled due to the physician's decision. Main reasons for patients' refusal were standard treatment preference and concerns about experimental arm toxicity, mainly infertility risk. Conditions that could hamper accurate follow-up and toxicity concerns accounted for most of the physicians' reasons. Adverse prognostic factors [B symptoms (odds ratio, OR = 5.35) and international prognostic score >= 3 (OR = 2.69)] were independently associated with inclusion. Conclusion: Despite an attractive protocol, only 51% of patients were included. It highlights concerns about selection of patients and the difficulty to obtain informed consent with better prognostic profile patients.