Cost effectiveness of amphotericin B plus G-CSF compared with amphotericinB monotherapy - Treatment of presumed deep-seated fungal infection in neutropenic patients in the UK

Objective: To assess the economic impact of adding granulocyte colony-stimu lating factor (G-CSF) to amphotericin B to treat a presumed deep-seated fun gal infection in neutropenic patients. This study was conducted from the pe rspective of the National Health Service (NHS) hospital sector. Design: We used our previously reported trial as the clinical basis for the analysis (see Participants and interventions). These data were combined wi th resource utilisation data, enabling us to construct a decision tree mode l of the treatment paths attributable to managing patients in each arm of t he trial. The model was used to calculate the cost effectiveness of using a mphotericin B plus G-CSF compared to amphotericin B monotherapy in neutrope nic patients with a presumed deep-seated fungal infection. Setting: An adult leukaemia/bone marrow transplant (BMT) unit in a large UK teaching hospital. Participants: Patients with a neutrophil count of <0.5 x 10(9)/L and having a presumed deep-seated fungal infection after either chemotherapy or stem cell/bone marrow transplantation for haematological malignancy. Interventions: 29 patients received intravenous amphotericin B (1 mg/kg dai ly) plus subcutaneous G-CSF (3 to 5 mu g/kg daily) and 30 patients received intravenous amphotericin B (1 mg/kg daily) monotherapy. The clinical trial showed that 62% of patients responded to antifungal treatment with amphote ricin B plus G-CSF compared to 33% of patients who responded to amphoterici n B monotherapy (p = 0.027). Nonresponders went on to receive a lipid formu lation of amphotericin B. Main outcome measure and results: The mean cost per patient treated with am photericin B plus G-CSF was pound 11 247 and the corresponding cost for amp hotericin B monotherapy was pound 14 317 (1996/1997 values) - a cost reduct ion of pound 3070 per patient. Sensitivity analyses demonstrated that the a ddition of G-CSF to conventional amphotericin B in the treatment of a presu med deep-seated fungal infection offers not only clinical benefits, but cos t benefits which are robust to changes in clinical and economic parameters. Conclusion: From a UK hospital perspective, amphotericin a plus G-CSF is co st effective compared with amphotericin B monotherapy in managing a presume d deep-seated fungal infection in neutropenic patients. This result should provide strong arguments to clinicians and policy-matters for the adoption of this treatment strategy in such patients.