Cost-Benefit Analysis of Primary Prophylaxis with Granulocyte Colony Stimulating Factor (GCSF) in Reducing Hospital Admissions with Febrile Neutropenia in Patients Undergoing High Dose Chemotherapy for Non-Hodgkin Lymphoma (NHL).

Abstract

Abstract 4764 IntroductionHigh dose chemotherapy for lymphoma is known to be associated with significant risk of febrile neutropenia (FN), especially in elderly patients. FN is a major dose-limiting toxicity of high dose chemotherapy and can lead to delay in treatment. In addition, FN also causes significant patient morbidity and has an adverse impact on quality of life. Although a meta-analysis of 12 randomised trials with Hodgkin disease(HD) and non-Hodgkin lymphoma (NHL) patients has shown that prophylactic GCSF significantly reduces the relative risk of FN (Bohlius J, et al: Granulopoiesis-stimulating factors to prevent adverse effects in the treatment of malignant lymphoma. Cochrane Database of Systematic Reviews CD003189, 2004), it is not known whether primary prophylaxis is also cost-effective. Since December 2007, we implemented guidelines in our haematology department for the use of primary GCSF prophylaxis in patients age >65 years undergoing high dose chemotherapy treatment for NHL. We studied the impact on the incidence of FN as well as cost-effectiveness of this approach. Patients and MethodsA cohort of 62 patients who underwent high dose chemotherapy for NHL from December 2006 to May 2009 was identified. Patient demographics, type of lymphoma, chemotherapy regimen, details of GCSF prophylaxis, hospital admissions due to FN, and outcome of treatment at last follow up were recorded. Cost-benefit analysis was carried out using the actual cost of hospital admissions and numbers needed to treat (NNT). Chi-square test was used to calculate p-value. ResultsGCSF was given in a dose of 300 mcg subcutaneously daily, for an average of 7 days. The median age of the cohort was 63 years (IQR 54, 70). Fifty three patients received CHOP based regimen. Forty one patients had diffuse large B cell lymphoma. Overall, patients who received primary GCSF therapy had significantly less FN compared to those who did not [8% vs 56%, p<0.0001, absolute risk reduction (ARR) 48% (95% confidence interval(C.I.) 25% to 66%) with NNT of 2 (95% C.I. 2 to 4)]. In subgroup analysis, even patients aged '65 years appeared to receive benefit from primary GCSF prophylaxis in reduction of FN [4.8% vs 66.7%, p=0.005, ARR 62%(95% C.I. 34%-90%), NNT 2(95% C.I. 1-3)]. In patients who underwent CHOP-based chemotherapy, the benefit of primary GCSF prophylaxis appeared to be greater compared to the cohort [11% vs 68%, p<0.0001, ARR 58%(C.I. 30%-76%) with NNT of 2 (95% C.I. 1-3) – table 1. There was no difference in treatment outcomes such as achieving at least PR or CR, and no difference in mortality, either from disease or all cause combined, during the study period. The average cost of hospital admission with FN was calculated to be NZD$6753.31. Using the NNT value obtained, primary GCSF prophylaxis appeared to be cost effective, leading to saving of $1991.22 per patient. ConclusionOur study shows that primary GCSF prophylaxis is effective in preventing FN in patients undergoing high dose chemotherapy and that it is highly cost-effective; hence primary GCSF prophylaxis is recommended in this patient group.All patients (n=62)GCSF prophylaxis (n=37)No GCSF prophylaxis (n=25)P-valueAge median years (IQR)63 (54,70)62 (53,70)64 (54,70)0.99Febrile neutropenia (FN) – all patientsOne or more episodes17314<0.0001None453411Subgroup analyses (patients >65 years)One or more episodes (FN)8260.04None21147Subgroup analyses (patients'65 years)One or more episodes (FN)9180.005None24204Subgroup analyses (CHOP-based chemotherapy only)One or more episodes (FN)15411<0.0001None38335Subgroup analyses (patients >65 years and CHOP-based chemotherapy)One or more episodes (FN)6240.02None18153 Disclosures:No relevant conflicts of interest to declare.