Metastases from breast cancer appear to have a predilection for the skeleton. In addition to focal skeletal disease, there is evidence that breast cancer is associated with generalized disturbances in skeletal metabolism.Increased resorption and accelerated turnover of bone have been shown in biopsies taken at sites distant from skeletal metastases (Taube et al, 1994), perhaps mediated by the systemic secretion of parathyroid hormone-related protein (PTHrP) which is expressed in breast cancer tissue (Orloff et al, 1989;Powell et al, 1991).In addition to the expression of PTHrP, both chemotherapy and a chemotherapy-induced menopause are likely to increase the risk of osteoporosis (Rivkees and Crawford, 1988;Saarto et al, 1997). Whereas tamoxifen may protect against bone loss in postmenopausal women, recent evidence suggests that bone losses may be accelerated in women with normal ovarian function Saarto et al, 1997). For these reasons, we wished to determine whether the risk of osteoporotic fractures was higher in women with non-metastatic breast cancer than in the healthy population.
PATIENTS AND METHODSWe studied two populations of women with breast cancer and a control population of women randomly drawn from the female population. Eighty-two women had histologically proven breast cancer and recurrent disease diagnosed according to standard criteria (Breast Cancer Task Force, 1977;Hayward et al, 1978), but without skeletal metastases as judged by skeletal radiographs and scintigraphy. They comprised a cohort of women in whom the effects of clodronate on the incidence of skeletal metastases and associated morbidity were studied. The findings of this trial have been published elsewhere .After initial assessment, patients were randomized to receive clodronate by mouth, 1600 mg daily as four capsules, or an identical placebo supplied by Leiras Oy, Helsinki, Finland. Patients were instructed to take four capsules daily in the morning 1 h before meals and away from calcium-containing liquids. Bone scintigraphy and skeletal radiographs (hands, pelvis, skull, lateral lumbar and thoracic spine) were obtained at 6-monthly intervals.The second population of women with breast cancer comprised 352 women recruited at the time of the first diagnosis to study the effects of clodronate, 1600 mg daily by mouth, on the incidence of skeletal metastases. None of the women had radiographic or scintigraphic evidence of skeletal disease. No selection was made on the basis of histological grade or the involvement of axillary nodes. After diagnosis, 36% of women received tamoxifen, 18% chemotherapy and 43% both chemotherapy and tamoxifen. After randomization to clodronate or placebo treatment, patients were assessed at 6-monthly intervals and spinal radiographs taken yearly. The study is continuing and remains double blind, so that the data are presented irrespective of treatment allocation.A control population comprised 776 women aged 45Ð69 years randomly drawn from a catchment population in London . At initial assessment, 193 women (2...