Incidence of neutropenic fever in patients treated with standard-dose combination chemotherapy for small-cell lung cancer and the cost impact of treatment with granulocyte colony-stimulating factor.

PURPOSE: We sought to determine the incidence of neutropenic fever associated with the use of standard-dose combination chemotherapy for small-cell lung cancer (SCLC) and to use these data as a template to analyze the costs and benefits of the routine use of granulocyte colony-stimulating factor (G-CSF). PATIENTS AND METHODS: We retrospectively reviewed records of 137 consecutive, unselected patients with SCLC treated with combination chemotherapy from January 1987 to March 1992. Admission criteria for neutropenic fever were temperature > or = 38.5 degrees C and an absolute neutrophil count < or = 500/microL. Neutropenic fevers were managed with a 25% dose reduction of the myelosuppressive drugs in subsequent cycles. Charge estimates for hospitalization ($1,244 per day) and G-CSF use ($2,027 per course) were estimated by reviewing charges to patients at Indiana University hospitalized for neutropenic fever or treated with outpatient G-CSF. We imposed assumptions from the Neupogen (filgrastim; Amgen Inc, Thousand Oaks, CA) licensing trial regarding the effectiveness of G-CSF and the Indiana University charge estimates on three models of G-CSF use: (1) preemptive--with all courses of chemotherapy, (2) reactive--with all cycles of chemotherapy following a neutropenic fever, and (3) dose reduction only (no G-CSF)--to derive charge estimates for G-CSF use.
RESULTS: Records of 137 patients with SCLC were identified and reviewed. The incidence of neutropenic fever was 12% in the first cycle of chemotherapy, and 18% overall, compared with the placebo- and G-CSF-treated arms of the Neupogen licensing trial, in which the incidence of neutropenic fever was 77% and 40%, respectively. Other therapeutic outcomes, such as neutropenic septic deaths, response rates, and survival, were comparable. We derived the following charge estimates for the three models of G-CSF: (1) preemptive--total charges = $1,287,481; (2) reactive--total charges = $276,154; and (3) dose reduction only--total charges = $192,820.
CONCLUSION: The incidence of neutropenic fever with standard-dose chemotherapy for SCLC was 18%. Routine use of G-CSF in SCLC patients treated with standard-dose chemotherapy appears to be expensive and is not associated with an obvious therapeutic benefit or cost savings. We suggest that careful analysis of the incidence of infectious complications, rather than granulocyte nadir and duration, be performed, and that clinical guidelines for the use of these effective, but expensive, products be developed.