BACKGROUND AND OBJECTIVES: The 5 factors of the International Prognostic Index (IPI) for aggressive non Hodgkin's lymphoma (NHL) age, disease stage, serum lactate dehydrogenase (LDH), performance status, number of extranodal sites) are validated predictors of a patient's survival. Given the need for economic evaluations, we analyzed whether the IPI and the presence of B-symptoms (night sweats, fever, weight loss) can identify subgroups of patients with favorable or unfavorable cost profiles. DESIGN AND METHODS: Chart data for 374 patients with newly diagnosed stage II-IV aggressive NHL treated between 1993-2001 with CHOP chemotherapy were used. Costs were calculated up to two years from the start of treatment. The cost of granulocyte colony-stimulating factor (G-CSF) was not included, as some patients received this due to trial participation. Regression analyses and non-parametric bootstrap tests were performed to determine the significance of prognostic factors. RESULTS: Mean first-line treatment costs (excluding G-CSF) were pound sterling 10047 (<60 years) and pound sterling 12232 (>60 years). Two-year follow-up costs averaged pound sterling 14039 and pound sterling 9026 for the two age groups, respectively. The 5 IPI variables, the 2 IPI risk group variables (resulting from the 5 factors) and B-symptoms all showed significant univariate associations with first-line treatment costs. They were also associated with higher 2-year costs, except for age, LDH, and standard risk group index. Lower predictability of total 2-year costs was due to wide variations in second-line treatments. INTERPRETATION AND CONCLUSIONS: The IPI factors and B-symptoms are predictive of treatment costs. The detailed information presented in this paper is of value for those who need to make cost-effectiveness estimations in NHL, which is a relevant topic, given new treatment modalities that are emerging.
BACKGROUND AND OBJECTIVES: The 5 factors of the International Prognostic Index (IPI) for aggressive non Hodgkin's lymphoma (NHL) age, disease stage, serum lactate dehydrogenase (LDH), performance status, number of extranodal sites) are validated predictors of a patient's survival. Given the need for economic evaluations, we analyzed whether the IPI and the presence of B-symptoms (night sweats, fever, weight loss) can identify subgroups of patients with favorable or unfavorable cost profiles. DESIGN AND METHODS: Chart data for 374 patients with newly diagnosed stage II-IV aggressive NHL treated between 1993-2001 with CHOP chemotherapy were used. Costs were calculated up to two years from the start of treatment. The cost of granulocyte colony-stimulating factor (G-CSF) was not included, as some patients received this due to trial participation. Regression analyses and non-parametric bootstrap tests were performed to determine the significance of prognostic factors. RESULTS: Mean first-line treatment costs (excluding G-CSF) were pound sterling 10047 (<60 years) and pound sterling 12232 (>60 years). Two-year follow-up costs averaged pound sterling 14039 and pound sterling 9026 for the two age groups, respectively. The 5 IPI variables, the 2 IPI risk group variables (resulting from the 5 factors) and B-symptoms all showed significant univariate associations with first-line treatment costs. They were also associated with higher 2-year costs, except for age, LDH, and standard risk group index. Lower predictability of total 2-year costs was due to wide variations in second-line treatments. INTERPRETATION AND CONCLUSIONS: The IPI factors and B-symptoms are predictive of treatment costs. The detailed information presented in this paper is of value for those who need to make cost-effectiveness estimations in NHL, which is a relevant topic, given new treatment modalities that are emerging.