Weycker, D., Bensink, M., Wu, H., Doroff, R., & Chandler, D. (2017). Risk of chemotherapy-induced febrile neutropenia with early discontinuation of pegfilgrastim prophylaxis based on real-world data from 2010 to 2015. Current Medical Research and Opinion, 33, 2115–2120.

To estimate the odds of FN, beginning with second chemotherapy cycle, among patients who received prophylactic pegfilgrastim in that cycle and all previous cycles versus those who received prophylactic pegfilgrastim in all previous cycles only.

This is a descriptive study using a retrospective matched-cohort design utilizing pooled data from two healthcare claims repositories, the MarketScan Database and LifeLink Database, spanning the period from January 1, 2010 through September 30, 2015. The study is not prospective, but rather completes a secondary analysis of existing data.

• N: 8,450 matched pairs of patients undergoing chemotherapy: comparison group received pegfilgrastim following first cycle only, source group received pegfilgrastim following each cycle
• AGE: Mean in comparison = 53.6 years, in pegfilgrastim = 53.9 years 
• MALES: Not identified, missing data 
• FEMALES: Not identified; study reports 89% of matched patients had breast cancer and 2% ovarian 
• CURRENT TREATMENT: Chemotherapy
• KEY DISEASE CHARACTERISTICS: 89% of matched had breast cancer, 7% had NHL, 2% lung, 2% ovarian, less than 1% colorectal
• OTHER KEY SAMPLE CHARACTERISTICS: 52% of the 89% of those with breast cancer received doxyrubicin and cyclophosphamide. The majority  of participants had surgery within the preceding 90 days

• SITE: Multi-site
• SETTING TYPE: Not specified; data obtained from larger number of participating private U.S. health plans. Appears to be primarily outpatients receiving chemotherapy with some hospital admissions    
• LOCATION: Varied throughout the United States

PHASE OF CARE: Active anti-tumor treatment

Retrospective matched-cohort design

Comparison of the incidence and odds ratios of FN were determined between comparison and pegfilgrastim patients. Data was evaluated using generalized estimating equation regression models. No standardized measurement or instrument used; data obtained from charts.

Second-cycle incidence of FN was greater is the comparison patients with an odds ratio of 1.7 and p < 0.001 (broad definition). Second-cycle incidence proportions for FN was greater in the comparison group (narrow definition), with an odds ratio of 4.3 and p < 0.001. All cycles from the third through course completions, the odds ratio for FN was 1.6 with p < 0.001 (broad definition).

In this retrospective, matched-cohort study, those who did not continue pegfilgrastim prophylaxis subsequent to the first cycle of chemotherapy had a higher risk for FN compared to those who continue pegfilgrastim prophylaxis after each cycle of chemotherapy. Therefore, the administration of pegfilgrastim following each chemotherapy cycle (as opposed to abbreviated use) appears to have a protective effective in reducing the incidence of FN.

• Risk of bias (no blinding)
• Risk of bias (no random assignment)
• Risk of bias (no appropriate attentional control condition)
• Unintended interventions or applicable interventions not described that would influence results
• Selective outcomes reporting
• Key sample group differences that could influence results
• Measurement/methods not well described
• Measurement validity/reliability questionable 
• Findings not generalizable
• Other limitations/explanation: Cannot be generalized to the larger population due to the lack of rigor in the design. The non-experimental retrospective design cannot establish causal relationships, only show correlations. In addition, those whose who are publicly insured and older adults were poorly represented. Lastly, the study was funded by maker of pegfilgrastim and the authors are stockholders in Amgen.

This retrospective, matched-cohort study indicates that premature discontinuation of pegfilgrastim prophylaxis following chemotherapy is commonplace in U.S. practice and is associated with additional febrile neutropenia events. The decision to prematurely discontinue pegfilgrastim prophylaxis following chemotherapy should be carefully weighed against the associated risk of FN. Although this study was a retrospective design, the findings continue to support the importance of adhering to current standards of nursing care practice with pegfilgrastim prophylaxis for the prevention of FN. Additional independent research is necessary, particularly evaluating the risk of FN in the older adult and publicly insured populations, which are under-represented in this study.