e21039 Background: Bone metastasis of breast cancer significantly compromises patient morbidity and mortality. Currently, no reliable methods detect or predict patients at increased risk for developing bone metastasis. We utilized 3 independent cohorts of breast cancer patients to validate a highly discriminatory plasma-based proteomic profile that identifies breast cancer bone metastasis. The identity of the most discriminatory protein component identified was a parathyroid hormone-related protein fragment, PTHrP(12-48). Methods: Plasma samples collected from 21 breast cancer patients with clinical evidence of a bone metastasis and 21 patients with no evidence of bone metastasis from time of diagnosis to clinical outcome were evaluated. A novel mass spectrometry-based assay using human serum spiked with synthetic PTHrP(12-48) was used to measure PTHrP(12-48) concentrations (pg/μl). Statistical significance was assessed by one-way ANOVA. ROC curves evaluated the diagnostic potential of PTHrP(12-48) and a simple logistic regression derived from the combined measurement of PTHrP(12-48) and NTx. Results: PTHrP(12-48) concentrations ranged between 11.6 and 92.1 pg/μl in bone metastasis patients and between 4.5 and 34.2 pg/μl in patients without bone metastases. PTHrP(12-48) was significantly increased in bone metastasis plasma (p < 0.05). No significant correlation was identified between PTHrP(12-48) and NTx. ROC analysis of PTHrP(12-48), threshold 18 pg/μl, classified the two groups with high accuracy. Class prediction by the PTHrP(12-48)/NTx logistic regression model increased diagnostic specificity. Conclusions: The measurement of PTHrP(12-48) in patient plasma has potential as a viable clinical measure of bone metastasis. In combination with serum NTx, PTHrP(12-48) may assist in identifying bone metastases in patients presenting with low to normal bone turnover markers.
Parathyroid hormone-related protein expression, in combination with nodal status, predicts bone metastasis and prognosis of breast cancer patients
