Secondary malignancies after treatment of aggressive non-Hodgkin lymphoma: A GISL cohort study on 1259 patients

BACKGROUND Secondary malignancies have been associated with Non Hodgkin Lymphoma (NHL) treatment. Nevertheless few analyses have addressed this issue focusing on aggressive lymphoma. Aims of this study were to determine the incidence and the risk factors for developing secondary cancer during long term follow up of patients treated for aggressive lymphoma. METHODS For the purpose of this study we identified in the GISL database, 1259 naïve patients with histologically confirmed diagnosis of aggressive NHL. Observed cancer were classified by site. The incidence numbers of second neoplasia was compared to the incidence of malignancies in the Italian population. The standardized incidence ratio was calculated from the ratio between observed and expected number of cancers. Absolute excess risk was calculated by subtracting the expected from the observed cases and dividing by the person-years at risk. The Time Free 2nd Tumour (TF2T) was measured from the end of the first treatment to last follow-up or date of diagnosis. Cumulative incidences were estimated either with a Kaplan-Meier estimate or by Gooley’s method. Effects of potential risk factors on second cancer rates were examined in a Cox proportional-hazard model. RESULTS The cohort consisted of 1259 patients enrolled in GISL trials in the period 1988–2003, accounting for 6180 person-year at risk of second tumor. Median age at diagnosis was 58 years. All patients were treated with chemotherapy either alone or in combination with radiotherapy. During follow up, 44 patients (3.5%) developed a second cancer. Twelve out of 44 patients developed hematological malignancies and 32 solid tumors, including 7 lung cancer, 6 colorectal cancer, 4 prostate cancer and 15 other type of cancers. The median time for developing second tumors was 42 months. The risk of secondary malignancy overall was not increased. The analysis of risk by cohort of age at diagnosis of second cancer showed an excess of risk for the cohort age 20–39 and 40–59 years. Cumulative incidence of second malignancy, estimated by Kaplan Meier and by Gooley’s method at 5, 10 and 15 years was 3.2%, 6.9% and 13.8%, and 2.7%, 5.2% and 9.6%, respectively. By univariate analysis we observed a significant negative impact on TF2T for age at first treatment and only marginally significant for elevated LDH level. Further, we performed a Cox regression analysis with gender, age at 1st treatment and LDH >UNL that showed the prognostic ability of these variables in predicting the risk of second tumour. We divided the log(HR) predicted from multivariate analysis, at the 33° and the 66° percentiles to obtain a score system. We observed three groups with significant difference (p< .0001) in the risk of developing second cancers. CONCLUSIONS Our results showed that the risk of second malignancy overall was not increased in patients treated for aggressive lymphoma. Cancer risk was age-related, as demonstrated by the excess of risk observed in the cohort age 20–39 and 40–59 years. Further, utilizing age, male gender and LDH in a Cox regression analysis, we demonstrated the prognostic value of these variables and we produced a score system able to identify groups with different risk of developing