br Patients and methods br

Patients and methods
Discussion In this nitric oxide secondary endpoint analysis of the OPTION trial, we have shown that the addition of goserelin to neo(adjuvant) chemotherapy temporarily increased bone turnover during treatment and that this reverses over the next 6–12 months after completion of treatment. This is probably mediated by the effects on ovarian function, with a rapid and profound fall in oestradiol levels but, once the drug is stopped, ovarian function resumes to a level that is sufficient to maintain normal bone turnover, even if this is not manifested by a demonstrable difference in the frequency of chemotherapy induced amenorrhoea. We found no acute direct effect of chemotherapy on bone turnover markers, but there was a delayed effect on bone with increased bone turnover starting at 6 months after completion of chemotherapy and continuing throughout the 36 months of observation. This was predominantly in women aged >40 years and in those women who developed chemotherapy induced amenorrhoea, suggesting there may not be a direct toxic effect of chemotherapy on bone but that the BMD loss is driven predominantly by POI. Our study however has limitations that include a lack of ovarian function information beyond 12 months, which limits the interpretation of the association between bone markers and POI after this time point. In addition the number of serum/urine samples available for the bone marker analysis decreased during follow up thus limiting the interpretation of data at later time points. We found that BALP remained persistently raised following completion of adjuvant therapy in women who experienced POI at 12 months, an effect not seen in menstruating women. Cameron et al. showed that BALP and serum CTX increased during chemotherapy in women who continue to menstruate (n=16) but normalised after cessation of treatment, whereas women who are amenorrheic (n=25) at the end of chemotherapy show a continuing rise in BALP [15], an observation that supports our findings. This altered bone turnover appears to persist and leads to evidence of bone loss at 5 years post chemotherapy that is more marked in women who develop amenorrhoea (lumbar spine BMD; −11.3%±0.9%) than those who maintain menses (lumbar spine BMD; −6.4%±1.2% respectively) [16]. In support of this, BMD loss following 6 months of CMF chemotherapy treatment in premenopausal breast cancer patients has been demonstrated at both the lumbar spine (−5.9%) and the femoral neck (−2%) at 2 years, with the largest changes in those patients who developed POI. [17]. Similarly, osteoporosis-free survival at 10 years following CMF chemotherapy is higher in menstruating women (100%) than those with irregular menses (69%) or amenorrhoea (67%) [18]. Therefore, patients with POI are at long term risk of bone loss. This study has focused on the bone turnover markers NTX and BALP. In previous studies elevated NTX and BALP have been shown to correlate with risk of skeletal complications in patients with bone metastases from solid tumours and myeloma [19]. Moreover, clinical trials of anti-bone resorptive therapies i.e. bisphosphonates have demonstrated translation lowering of bone turnover marker values to premenopausal levels reduces fracture risk [20–22]. Thus the persistent changes in bone turnover markers seen in our study have potential implications for subsequent fracture risk. The effects of permanent and temporary ovarian suppression on bone have been evaluated in prospective breast cancer trials comparing chemotherapy to GnRH analogues in premenopausal women. Evaluation of BMD was undertaken in the bone sub-study of the ZEBRA trial (Zoladex in Early Breast Cancer Research Association) which compared 6x CMF with 2 years of goserelin; at 2 years BMD losses at lumbar spine and femoral neck compared to baseline were −10.5% and −6.4% with goserelin, and −6.5% and −4.5% with CMF [23]. However at 3 years (12 months after completion of goserelin) BMD at both anatomical sites partially recovered in the goserelin cohort compared to persistent loss in the CMF cohort. Amenorrhoea was observed in 100% of patients treated with goserelin and 64% treated with CMF at 2 years, however, there was resumption of menses in 73% of the goserelin group at 3 years but continued amenorrhoea in 76.5% CMF patients.