Hodgkin's disease: Study of treatment intensities and incidences of second malignancies

Background: Advances in radiotherapy and chemotherapy have gradually increased cure rates for patients with Hodgkin's disease. With improved long-term survivals, increases in observed second malignancies over those of the general population have been reported as early as 1972. Recently, a number of investigators have suggested that the relative importance of recognized risk factors contributing to the development of acute myelogenous leukemia (AML), non-Hodgkin's lymphomas, and solid tumors may be different. Our study is concerned with the influence of various risk factors on patients who have been treated with modern radiotherapy and combination chemotherapy between 1966 and 1987. Patients and methods: We reviewed the records of 1,022 patients with Hodgkin's disease of whom 1,013 had sufficient data for analysis. Kaplan-Meier methodology was used to calculate overall and determinate survivals and occurrences of acute myelogenous leukemia, non-Hodgkin's lymphoma, and solid tumors. The observed to expected incidences, calculated from the SEER incidence and population files for 1976, were compared. Using Cox's proportional hazards model, the following were analyzed singly for risk significance for the entire population: age, stage, splenectomy, treatment modality, treatment intensity, and number of treated relapses. Separate analyses were performed to determine the relative risks for subsets of the population. These included pelvic radiotherapy for those with stage III disease and specific alkylating agents for patients who were treated with chemotherapy only. Results: Sixty-six instances of second malignancy were documented as follows: AML 14, non-Hodgkin's lymphoma 14, and solid tumors 38. The overall incidence of second malignancy was significantly greater than the expected incidence of 21.75 (p = 0.0001) and it was also significant for AML, non-Hodgkin's lymphoma and solid tumors. Analyses for risk of second malignancy demonstrated that age ≥40 years, stage III or stage IV disease, and treatment with chemotherapy only were all associated with a significantly higher risk of second malignancy than any of the other factors. However, only treatment with regimens containing nitrogen mustard had a significantly higher risk for second malignancy. Treatment intensity and number of treated relapses had no specific effect on risk. Joint modeling of age, stage, and treatment showed that the combination of age and stage was the most significant risk factor for AML and non-Hodgkin's lymphoma (p = < 0.003). However, only age was important for solid tumors. Conclusions: Our analysis suggests that the most critical host factor for developing a second malignancy was age. The fact that patients with stages III and IV disease had an increased risk of second malignancy regardless of age suggests that biologic factors related to the tumor also may have been significant. However, it is possible that the effect of treatment was hidden by stage.