Weycker, D., Li, X., Figueredo, J., Barron, R., Tzivelekis, S., & Hagiwara, M. (2016). Risk of chemotherapy-induced febrile neutropenia in cancer patients receiving pegfilgrastim prophylaxis: Does timing of administration matter? Supportive Care in Cancer, 24, 2309–2316. 

To evaluate the timing of pegfilgrastim administration

Data from two large claim systems repositories obtained from 2003 to 2011 were pooled for analysis. Patients included were adults receiving at least one course of chemotherapy for a solid tumor or non-Hodgkin lymphoma. Patients were identified as receiving intermediate or high-risk regimens, and pegfilgrastim use was determined using Healthcare Common Procedure Coding System (HCPCS) codes on claims with service dates up to three days after the last chemotherapy administration of a cycle. Patients were grouped based on whether pegfilgrastim was given on the same day as the last chemotherapy dose or on days 2–4 after chemotherapy. Chemotherapy cycles that used different colony-stimulating factors (CSFs) were excluded from the analysis. The analysis was done for all cases in the first cycle and for all febrile neutropenia (FN) episodes.

• N = 45, 592 patients–179, 152 chemotherapy cycles   
• MEAN AGE = 55 years
• MALES: 10.9%, FEMALES: 89.1%
• CURRENT TREATMENT: Chemotherapy
• KEY DISEASE CHARACTERISTICS: Seventy-eight percent had metastatic breast cancer.
• OTHER KEY SAMPLE CHARACTERISTICS: Nine percent had cardiovascular comorbidities, and 11% had diabetes.

• SITE: Multi-site   
• SETTING TYPE: Other    
• LOCATION: United States

• PHASE OF CARE: Active antitumor treatment

• Retrospective cohort comparison

• FN episodes requiring inpatient care were determined based on admissions with a diagnosis of neutropenia, fever, or infection.
• FN episodes in the outpatient setting were identified by any ambulatory encounter with a diagnosis of neutropenia, fever, or infection and a HCPCS code on the same day for the IV administration of an antibiotic. 
• A narrower definition including only diagnoses of neutropenia and antibiotic administration was also evaluated.

The study was shown to be sufficiently powered. Overall, FN occurred in 3.9% of those receiving same-day prophylaxis in the first cycle compared to 2.8% of those receiving pegfilgrastim on days 2–4 after chemotherapy (p < 0.001). Across all cycles, FN occurred in 2.5% of the same-day patients versus 1.9% of the prophylaxis patients on days 2–4 (p < 0.001). Overall, pegfilgrastim was given on the same day as the last chemotherapy dose in 12% of the cases.

The findings suggest that the timing of pegfilgrastim administration does matter, and that administration according to prescribing information is more effective than administration on the last day of the chemotherapy cycle.

• Risk of bias (no random assignment)
• Measurement validity/reliability questionable
• Reliance on claims data is known to be prone to errors
• The definition of FN is somewhat questionable, as no specific diagnosis code exists for FN. The use of the combination of documented fever, use of antibiotics, etc., may not accurately represent clinical FN.

The findings suggest that the administration of pegfilgrastim according to the timing outlined in prescription information is more effective than administration on the last day of the chemotherapy cycle. While an additional patient visit for this injection may not be convenient, it is worth the reduction in risk for FN and attendant infectious complications. The findings point to the need for the development of practical alternatives to facilitate adherence to essential schedules, such as simple devices for patients to self-administer CSFs as needed, as is done with medications like insulin.