Background: There is an urgent need for the identification of commonly assessable predictive factors in the treatment of patients with metastatic breast cancer.
Methods: During the course of a treatment including low dose metronomic oral cyclophosphamide and capecitabine plus i.v. bevacizumab (plus erlotinib in one third of the patients) for metastatic breast cancer, we observed that a relevant number of patients developed repeatedly elevated levels of mean corpuscular volume (MCV) of red blood cells without a significant fall in hemoglobin levels. We conducted a retrospective analysis on these 69 patients to evaluate if the increase in MCV could be associated to tumor response.
Results: During the course of treatment 42 out of 69 patients (61%) developed macrocytosis. Using Cox proportional hazards modeling that incorporated macrocytosis (MCV ≥ 100 fl) as a time-dependent covariate, macrocytosis resulted in a halved risk of disease progression (HR 0.45; 95% CI, 0.22-0.92, p-value 0.028). In a landmark analysis limited to patients with no sign of progression after 24 weeks of treatment, median time to progression was 72 weeks (48 weeks after landmark) in patients who had developed macrocytosis, and 43 weeks (19 weeks after landmark) in patients who had not (p=0.023).
Conclusion: Macrocytosis inversely related to risk of disease progression in patients treated with metronomic capecitabine plus cyclophosphamide and bevacizumab for metastatic breast cancer. This finding may be explained through thymidylate synthase inhibition by capecitabine. Whether bevacizumab has a role in determining macrocytosis, similarly to what happens with sunitinib, has to be further investigated. If other studies will confirm our findings, macrocytosis might be used as an early marker of response during metronomic treatment with capecitabine and cyclophosphamide with or without bevacizumab.
Copyright © 2012 Elsevier Ltd. All rights reserved.