Multivariable regression analysis of febrile neutropenia occurrence in early breast cancer patients receiving chemotherapy assessing patient-related, chemotherapy-related and genetic risk factors

Abstract Background Febrile neutropenia (FN) is common in breast cancer patients undergoing chemotherapy. Risk factors for FN have been reported, but risk models that include genetic variability have yet to be described. This study aimed to evaluate the predictive value of patient-related, chemotherapy-related, and genetic risk factors. Methods Data from consecutive breast cancer patients receiving chemotherapy with 4–6 cycles of fluorouracil, epirubicin, and cyclophosphamide (FEC) or three cycles of FEC and docetaxel were retrospectively recorded. Multivariable logistic regression was carried out to assess risk of FN during FEC chemotherapy cycles. Results Overall, 166 (16.7%) out of 994 patients developed FN. Significant risk factors for FN in any cycle and the first cycle were lower platelet count (OR = 0.78 [0.65; 0.93]) and haemoglobin (OR = 0.81 [0.67; 0.98]) and homozygous carriers of the rs4148350 variant T-allele (OR = 6.7 [1.04; 43.17]) in MRP1 . Other significant factors for FN in any cycle were higher alanine aminotransferase (OR = 1.02 [1.01; 1.03]), carriers of the rs246221 variant C-allele (OR = 2.0 [1.03; 3.86]) in MRP1 and the rs351855 variant C-allele (OR = 2.48 [1.13; 5.44]) in FGFR4 . Lower height (OR = 0.62 [0.41; 0.92]) increased risk of FN in the first cycle. Conclusions Both established clinical risk factors and genetic factors predicted FN in breast cancer patients. Prediction was improved by adding genetic information but overall remained limited. Internal validity was satisfactory. Further independent validation is required to confirm these findings.