Febrile neutropenia in adjuvant and neoadjuvant chemotherapy for breast cancer: a retrospective study in routine clinical practice from a single institution

BACKGROUND

Febrile neutropenia (FN) is one of the most common and most critical adverse effects of chemotherapy. Despite many existing guidelines based on the use of granulocyte-colony stimulating factor (G-CSF), FN continues to impair the quality of life and interfere with the treatment of many patients. The purpose of this study was to assess the incidence and management of FN associated with chemotherapy for early breast cancer in routine clinical practice.

METHODS

All patients with early-stage breast cancer (ESBC) treated by chemotherapy at Institut Curie, Hôpital René Huguenin, in 2014 were retrospectively included. The incidence and management of FN were reported. Risk factors associated with FN were studied by robust-error-variance Poisson regression.

RESULTS

A total of 524 patients received either neoadjuvant (N = 130) or adjuvant chemotherapy (N = 394). Most patients (80%) were treated with a combination of 5-fluorouracil, epirubicin, and cyclophosphamide (FEC100; 3 cycles) followed by docetaxel 100 mg/m² (D; 3 cycles). The overall incidence of FN was 17%. Eighteen percent of patients received primary prophylaxis (PP) for FN with G-CSF, using pegfilgrastim in 64% of cases and 74% of patients over the age of 70 received PP. Less than 5% of patients who received PP experienced FN. Recurrent FN after secondary prophylaxis was observed in 9% of patients. Forty-seven percent of cases of FN occurred after the first cycle and 30% occurred after the fourth cycle, corresponding to D ± trastuzumab (T). The FEC100 regimen was associated with a relative risk of FN of 1.98 (p = 0.09). Autoimmune (AI) and inflammatory diseases were associated with a higher risk of FN (RR 3.08; p < 0.01). No significant difference in the incidence of FN was observed between adjuvant and neoadjuvant chemotherapy. FN was managed on an outpatient basis in 72% of cases. Outpatients with FN were mainly treated by a combination of amoxicillin-clavulanic acid and ciprofloxacin. Dose reduction or chemotherapy regimen modification were necessary in 25% of patients after FN. No toxic death was reported.

CONCLUSION

The incidence of FN induced by adjuvant/neoadjuvant chemotherapy in ESBC is higher in routine clinical practice than in clinical trials. AI or inflammatory diseases were significant independent risk factors for FN. Primary prophylaxis in patients at risk (elderly, comorbid patients), especially treated with the FEC regimen, is the keystone of management of this adverse effect. Prevention and management of FN to ensure the patient's safety and quality of life are a major issue for both medical oncologists and supportive care physicians.

[Mitomycin-vinorelbine combination as second-line chemotherapy in metastatic cancer of the breast]

Fourty-six patients (41 evaluable) were treated in second line chemotherapy of metastatic breast cancer (MBC) by an association of mitomycin (M), vinorelbine (V) (M 8 mg/m2 D1, V 25 mg/m2 D1 and DI 8 every 4 weeks). Median age was 58 years (36-78), median performance status 1 (0-3). Thirty-seven per cent of the tumors were estrogen receptors positive and 17% progesterone receptors positive. Seventeen patients received an adjuvant chemotherapy and 39 a first line chemotherapy with anthracycline (A). The median number of metastatic sites was 2 (1-4) and 27 patients (67%) had visceral metastases. Twelve patients were refractory to anthracyclines and 5 resistant. No toxic death nor hemolytic uremic syndrome were observed. Seven (3.7%) febrile neutropenias happened responsible for 4 hospitalizations. A grade 3 or 4 neutropenia was noted in 34% of the cycles but no other clinic toxicity nor grade 3 or 4 thrombopenia. The rate of objective response (OR) was 37.5% with 2 complete responses (CR) and 13 partial responses (PR). Seven patients had stable disease and 18 progressed. The rate of hepatic OR was 31%. Five (40%) A-refractory patients responded but no resistant patient. Median OR time was 10 weeks (8-12) and median OR duration was 5 months (3-6). Median survival was 11.5 months. MV association is well tolerated and effective in second line chemotherapy for MBC even with hepatic metastasis and in patients refractory to anthracyclines.