Cost-effectiveness analysis of the addition of rituximab to CHOP in young patients with good-prognosis diffuse large-B-cell lymphoma.

BACKGROUND AND OBJECTIVE: Diffuse large-B-cell lymphoma (DLBCL) is an aggressive form of lymphoma. It accounts for 30-40% of all new cases of non-Hodgkin's lymphoma and is the subtype with the highest overall incidence. Before the development of monoclonal antibodies, the standard treatment of newly diagnosed DLBCL was based on combination therapy with cyclophosphamide, doxorubicin, vincristine and prednisone (CHOP). Current treatment consists of the combination of CHOP and the monoclonal antibody rituximab (R-CHOP regimen), with recovery rates of 70%. The aim of this study was to compare the efficacy (survival) and direct medical costs of two chemotherapy regimens, R-CHOP and CHOP, in the treatment of DLBCL in young patients with a good prognosis.

METHODS: A decision-analysis model with tree structure was used to compare R-CHOP and CHOP in the treatment of young patients with DLBCL from the perspective of the Italian National Health Service. Patients entered at the root of the tree and followed one of the six possible therapeutic pathways. After receiving one of the two chemotherapy treatments (R-CHOP or CHOP) for 5 months, patients could have a complete response or not. In the presence of no response, patients underwent rescue therapy. In the case of relapse after 3 years' follow-up (following an initial complete response at 5 months), patients were given rescue therapy. The model provided an estimate of mean survival (life-years gained [LYG]) and mean costs (direct medical costs) over a period of 3 years. Both survival and costs were discounted at a rate of 3%. Costs were in euro, year 2007 values. Several sensitivity analyses were carried out with varying clinical parameters.

RESULTS: The LYG with the R-CHOP regimen was higher (2.697 LYG per patient) than with the CHOP regimen (2.517 LYG per patient). When taking into account the cost of rescue therapy, the overall mean treatment cost per patient was lower with the R-CHOP regimen (22,113.44 euro) than with the CHOP regimen (22,831.17 euro). Sensitivity analyses showed that the incremental cost-effectiveness ratios per LYG for complete response at 5 months (16,816.00 euro) and for relapse-free survival at 3 years (11,967.12 euro) were below the internationally accepted threshold (50,000 euro). Furthermore, for survival at 3 years, R-CHOP was confirmed as the dominant therapy (lower expected mean costs, higher number of LYG).

CONCLUSIONS: Our study demonstrated the clinical and economic benefits of adding rituximab to a CHOP chemotherapy regimen in young patients who present with DLBCL with good prognosis. The higher costs associated with rituximab were offset by the significantly lower rescue therapy costs. Further studies that include patients with unfavourable prognosis are needed to confirm these findings.