Open Access Research article

An open cohort study of bone metastasis incidence following surgery in breast cancer patients

Mitsuru Koizumi12*, Masataka Yoshimoto34, Fujio Kasumi35 and Takuji Iwase3

Author Affiliations

1 Diagnostic Imaging Group, Molecular Imaging Center, National Institutes of Radiological Sciences, 4-9-1 Anagawa, Inage-ku, Chiba,263-8555 Japan

2 Department of Nuclear Medicine, Cancer Institute Hospital, 3-8-31, Ariake, Koto-ku, Tokyo 135-8550, Japan

3 Department of Breast Surgery, Cancer Institute Hospital, 3-8-31, Ariake, Koto-ku, Tokyo 135-8550, Japan

4 Breast Center, Mita Hospital, International University of Health and Welfare, 1-4-3 Mita, Minato-ku, Tokyo, 108-8329, Japan

5 Breast Center, Juntendo University, 3-1-3 Hongo, Bunkyo-ku, Tokyo 113-8431, Japan

For all author emails, please log on.

BMC Cancer 2010, 10:381  doi:10.1186/1471-2407-10-381

Published: 21 July 2010



To help design clinical trials of adjuvant bisphosphonate therapy for breast cancer, the temporal incidence of bone metastasis was investigated in a cohort of patients. We have tried to draw the criteria to use adjuvant bisphosphonate.


Consecutive breast cancer patients undergoing surgery between 1988 and 1998 (5459 patients) were followed up regarding bone metastasis until December 2006. Patients' characteristics at the time of surgery were analyzed by Cox's method, with bone metastasis as events. Patient groups were assigned according to Cox's analysis, and were judged either to require the adjuvant bisphosphonate or not, using the tentative criteria: high risk (>3% person-year), medium risk (1-3%), and low risk (<1%).


Bone metastasis incidence was constant between 1.0 and 2.8% per person-year more than 10 years. Non-invasive cancer was associated with a very low incidence of bone metastasis (1/436). Multivariate Cox's analysis indicated important factors for bone metastasis were tumor grade (T), nodal grade (pN), and histology. Because T and pN were important factors for bone metastasis prediction, subgroups were made by pTNM stage. Patients at stages IIIA, IIIB and IV had an incidence of >3% per person-year, patients with stage I <1% per person-year, and those with stages II were between 1 and 3%. Further analysis with histology in stage II patients showed that stage IIB with high risk histology also had a high incidence (3% person year), whereas stage IIA with medium risk histology were <1%.


Bone metastasis incidence remained constant for many years. Using pN, T, and histopathology, patients could be classified into high, medium, and low risk groups.