Background: Urinary deoxypyridinoline (DPD)-crosslinks have been shown to be a highly specific parameter for type I collagen metabolism.
Materials and methods: In a prospective breast cancer study, urine samples were collected in after-care patients and in patients with bone metastases. DPD-crosslinks were measured every three weeks using a fully automated chemiluminescence immunoassay. Bone metastases were confirmed by bone scan and/or x-ray, and were followed-up over six months. To validate the test, a receiver operating characteristics (ROC)-curve was set up to find the DPD cut-off concentration which separates patients with no evidence of disease (NED) from patients with bone metastases.
Results: 73 breast cancer patients (41 with NED, 32 with bone metastases) were included into the ROC analysis. At a DPD cut-off value of 8 nmol/mmol creatinine, we found the best sensitivity (84.4%) for the detection of bone metastases with a specificity of 70.7%. Patients with stable bone disease under intravenous pamidronate treatment (90 mg q3w) and specific therapy had a significant (p = 0.007) fall of the DPD-crosslinks in comparison to the progressive subset with 72.7% falling below 8 nmol/mmol.
Conclusions: We conclude that the net bone turnover is not increased at a DPD-crosslinks elimination < 8 nmol/mmol.