Trends in use of primary prophylactic colony stimulating factors and neutropenia-related hospitalization in commercially insured patients receiving myelosuppressive chemotherapy in the United States: 2005-2017.

PURPOSE: Describe temporal changes in use of myelosuppressive chemotherapy, primary prophylactic colony-stimulating factor, and neutropenia-related hospitalization, in commercially insured patients. METHODS: Using a large commercial administrative database, we identified annual cohorts of adult patients diagnosed with breast or lung cancer, or non-Hodgkin lymphoma and initiating myelosuppressive chemotherapy during 2005-2017. We described yearly changes in proportions of myelosuppressive chemotherapy by febrile neutropenia risk category (high, intermediate, unclassified) and proportion of prophylactic colony-stimulating factor use and unadjusted incidence of neutropenia-related hospitalization in the first cycle of myelosuppressive chemotherapy. RESULTS: Annual cohorts included 4383-5888 eligible patients during 2005-2017. The proportion of eligible patients aged ≥ 65 years increased from 26.0% in 2005 to 58.2% in 2017. Myelosuppressive chemotherapy use with regimens with high risk for febrile neutropenia increased from 15.1% in 2005 to 31.0% in 2017; and regimens with intermediate risk for febrile neutropenia decreased from 63.7% to 48.1% in 2017. Prophylactic colony-stimulating factor use increased from 41.6% in 2005 to 54.3% in 2017. Crude incidence of neutropenia-related hospitalization for all cancers increased from 2.0% to 3.1%, with a substantial increase in neutropenia-related hospitalization observed among non-Hodgkin lymphoma patients (2.8% to 8.5%) during 2005-2017. CONCLUSION: Among adult patients with breast and lung cancer, and non-Hodgkin lymphoma receiving myelosuppressive chemotherapy, use of regimens with high risk for febrile neutropenia increased, as did the use of prophylactic colony-stimulating factors after 2005. Incidence of neutropenia-related hospitalization increased slightly, particularly among non-Hodgkin lymphoma patients. Further studies are required to understand this increasing trend of neutropenia-related hospitalization, changing patient-level risk factors, and febrile neutropenia management.

Initial response to chemotherapy, not delay in diagnosis, predicts overall survival in inflammatory breast cancer cases.

OBJECTIVES: To determine whether chemotherapy response and diagnostic delay affect overall survival (OS) of classic inflammatory breast cancer (IBC) cases receiving chemotherapy as initial treatment and to determine whether OS differs between classic and "atypical" IBC cases. METHODS: This is a prospective cohort study of 155 patients enrolled in the IBC Registry. "Classic" IBC cases met AJCC or SEER case definitions. "Atypical" IBC cases exhibited classic features but involved <1/2 breast without documented dermal lymphatic invasion. Variables included OS (years from initial chemotherapy treatment until death or last contact), chemotherapy response (complete, partial, or none), diagnostic delay (days from first medical contact for signs/symptoms of abnormal breast to definitive pathologic IBC diagnosis), age at diagnosis (y), and triple-negative status (yes or no). OS curves stratified by individual predictors were estimated and compared using Kaplan-Meier methods and log-rank tests. Associations between OS and predictors were examined collectively using Cox proportional hazards regression. RESULTS: Classic IBC cases with complete, partial, or no response had respective median (95% confidence interval [CI]) OS times of 10.30 (6.78, +), 6.27 (4.42, +), and 2.86 (1.11, 11.42) years (P=0.0072). Chemotherapy response was significantly associated with OS after controlling for covariates (P=0.003). Women not responding to chemotherapy had a significantly higher hazard of death compared with women with complete (hazard ratio [HR]=5.76; 95% CI, 2.09-15.84) or partial (HR=3.40; 95% CI, 1.27-9.10) response. Diagnostic delay was not significantly associated with OS (HR=1.003; 95% CI, 0.999-1.007). OS did not differ significantly between classic and "atypical" IBC cases (P=0.60). CONCLUSIONS: Response to standard IBC chemotherapy is a dominant prognostic factor in determining patient outcomes. In our study, with limited statistical power, delay in diagnosis defined as >60 days from the time of first physician contact did not seem to affect patient outcomes. Data support similarities between classic and "atypical" IBC.