Increased risk of second cancers in managing Hodgkin's disease: the 20-year Leiden experience

Publisher Website
Google Scholar
Abstract
Between January 1969 and December 1988, 482 patients were treated for Hodgkin's disease at the Leiden University Hospital. All cases were routinely recorded in the Hospital Information System, which has an active annual follow-up. Of all patients, 57% remained relapse free. According to the kinds of treatment they received, the following major categories were established: radiotherapy only (28.2%), chemotherapy only (20.1%), only initial combination of radiotherapy and chemotherapy (34.2%), all other combinations of radio- and chemotherapy (15.4%), or not registered (2.1%). Twenty-seven second cancers were observed; six leukemias, five non-Hodgkin lymphomas, and 16 solid tumors. Of all solid tumors only nine occurred in relapse-free patients. The overall relative risk of second cancers increased with the duration of follow-up. Using general population incidence rates to calculate expected numbers, the risk for developing leukemia, non-Hodgkin lymphoma, and solid tumors was increased 36-fold, 31-fold, and 2.4-fold, respectively. The cumulative risk of developing a second cancer 10 years after diagnosis of Hodgkin's disease was 7% for both the radiotherapy-only and the initial combination of radio- and chemotherapy group. It was 16% and 17% for the chemotherapy-only and the other combinations of radio- and chemotherapy group, respectively. Multivariate analysis (using the Cox regression model) show an increased risk of second cancers (RR=0.7) when a relapse of Hodgkin's disease resulting in increasing cumulative therapy occurred. Age at diagnosis of Hodgkin's disease was an important determinant for the risk of non-Hodgkin lymphoma and solid tumors. Cumulative chemotherapy intensity was an important factor in increasing leukemic risk in a dose-response fashion. Apart from this, the stage of Hodgkin's disease, although closely related to the kind of therapy, seemed to have an independent effect on leukemic risk.