BACKGROUND AND OBJECTIVE(#br)One of the consequences of the enormous improvement in survival rates of patients treated for Hodgkin's disease (HD) is the emergence in the long term of treatment-related complications, particularly secondary cancers. This study was undertaken to observe the occurrence of non-Hodgkin's lymphoma (NHL) in patients treated for HD and to identify the etiological role of various risk factors, especially spleen irradiation, in the pathogenesis of this illness.(#br)DESIGN AND METHODS(#br)From 1972 to 1996, the Department of Radiation Oncology and the Hematology Section of "La Sapienza" University of Rome observed and analyzed the occurrence of NHL in 1,391 patients treated for HD. The average follow-up period was 84 months. For a more accurate calculation of the... risk of the occurrence of NHL, the patients were first divided into 3 groups according to their initial treatment and also according to the total treatment they had received. Then, in order to establish the possible connection between NHL and splenic treatment the patients were also divided into 3 subgroups according to whether they had undergone splenectomy, splenic irradiation or neither of these. Two different methods of statistical analysis were used: (a) the cumulative risk (confidence interval) was evaluated in relation to treatment (initial and at the time of salvage) and (b) the Cox model was applied to identify the variables which play a role in the appearance of NHL. The cumulative risk of developing NHL was assessed using the Kaplan and Meier method. A multivariate analysis was performed using the Cox Proportional Hazard Model.(#br)RESULTS(#br)A total of 20 cases of NHL were observed, appearing between 17 and 206 months after initial treatment. The cumulative risk was 0.8%, 1.8%, 2.6% and 3.5% at 5, 10, 15 and 20 years respectively. According to the multivariate analysis, significant risk factors were splenic irradiation and age (> 40 years). Splenic irradiation (vs no splenectomy/no splenic irradiation) showed a relative risk of 5.69, p = 0.0280, while age over 40 showed a relative risk of 3.05, p = 0.0152.(#br)INTERPRETATION AND CONCLUSIONS(#br)From the results of this study, if appears that there is a possibility that splenic irradiation and age over 40 increase the risk of NHL in HD patients. Further studies are needed to investigate in greater depth the role of spleen irradiation in the occurrence of this illness.