Clinical Research Of The Value Of BSP Expression In Primary Breast Cancer Tissue For Early Predicting Bone Metastasis

Objective: Bone metastasis, which resulted in skeletal-related eventsgreatly affects the quality of life and prognosis of patients with breast cancer.A number of studies have confirmed that prophylactic use of bisphosphonatesin postoperative breast cancer can reduce the incidence of bone metastases,improve quality of life and prognosis as well.Human bone sialoprotein (hBSP)involves in tumor cell adhesion, vascular generation, anti-apoptosis, bonedestruction.Our studies have confirmed the role of BSP in predicting bonemetastasis and high expression BSP in breast cancer shows high sensitivityand specificity in predicting bone metastasis.The different expression of BSP in breast cancer tissues and clinicalpathological features such as tumor size,lymph nodes status,ER,PR,HER-2,Ki-67between bone metastasis and non-bone metastasis arms were analyzed in ourstudy to validated the role of BSP in predicting bone metastasis in breastcancer,and to investigate whether these factors had statistical impacts on thebone metastases free survival and which were independent risk factors for bonemetastasis. As a result, it provided basis for screening accuratly high-riskpopulation who are prone to bone metastases in early phasea and implementprophylactic use of bisphosphonates for them.Methods: A total of171women patients with breast cancer fromDepartment of Medical Oncology, the Fourth Hospital of Hebei MedicalUniversity from January1st2005to December31st2007were included.Inclusion criteria were resection of primary tumor in this hospital andcomplete clinical and follow-up data.According to ECT and X-ray or CT scanto diagnose bone metastases,92patients who were diagnosed with bonemetastasis in3years after surgery were assigned to bone metastasis(BM)groupand the rest79patients were non-BM group(NBM).The follw-up started on the surgery day and ended on Decembe31st2011by outpatient,telephone andmail etc. The end point was decided as the death of the patient, lost in contactor the first time of bone metastasis diagnosed.All clinical data recorded in thisstudy included age, surgery day, pathologic type, tumor size, lymph nodesstatus, ER,PR, HER-2, Ki-67, the date on which bone metastases wasdiagnosed for the first time and KPS score at final follow-up.Immunohistochemistry method was used for evaluation of the expressionof BSP,Ki-67,HER-2,ER and PR. Semi-quantitative scoring, based on thepercentage of positive tumour cells, was used to assess the immunereaction(for protein BSP: negative≤25%;positive:>25%;for Ki-67:negative≤50%;positive>50%;for protein HER-2:low expression-～++;highexpression:+++).All data were analyzed with SPSS13.0statistical software.The differenceof all the pathological features including BSP among BM group and non-BMgroup was stimated by chi-square test；Univariate survival analysis was basedon the Kaplan–Meier product limit estimate.Differences between survivalcurves were compared with the use of log-rank test. The parameters for thepredominant impact on the BM-free survival which P<0.10in the univariateanalysis were estimated by use of the Cox proportional hazards regressionmodel. All statistical tests were two-tailed, and P<0.05was consideredstatistically significant.Results:1.The expression of BSP in breast cancer1.1BSP protein expression was stronger in patients with BM (67.4%vs44.3%,P=0.002), and the positive expression rate of BSP in all patients was57.2%.1.2HER-2expression was weaker in patients with BM(25%VS43%,P=0.015).The correlation was not significant(P>0.05) for other clinicalparameters such as tumor size, hormone receptor, lymph nodes, pathologicaltype, TNM stage,age,ki-67,KPS score,etc.2. The correlation among BSP expression and other clinical characteristicsThere was not significant for BSP with clinical characteristics(P>0.05). 3Univariate analysis of3year BM-free survival analysis3.1The3year BM-free survival analysis showed significantly differentbetween BSP-negetive arm and BSP-positive arm(68.92%and47.42%,respectively,P=0.001);3.2The3year BM-free survival rate was also significantly lower in patientswith tumor size>3cm(62.50%vs49.15%for patients with≤3cm and>3cmrespectively, P=0.014).3.3BM-free survival was also correlated with ER status (P=0.027),the survivalrate was higher in patients with ER positive(48.05%) compared with ERnegative(37.49%).However,other clinical characteristics such as pathologictypes,PR,Ki-67, lymph nodes,age(P>0.10)as well as HER-2(P=0.077) were notsignificantly associated with3year bone metastasis free survival.4Multivariate Cox regression analysisCox proportional hazards regression model demonstrated that BSPexpression,tumor size were independent risk factors for BM,while ER andHER-2status were independent protective factors for BM.4.1BSP expressed was an independent prognostic risk factor forBM(HR=2.179,95%CI1:429-3.322,P<0.001);4.2Tumor size was another independent prognostic risk factor forBM(HR=1.631,95%CI:1.095-2.37,P=0.011);4.3ER was a protective factor for BM in breast cancer(HR=0.464,95%CI:0.289-0.735,P=0.001);4.4HER-2was another statisticly protective factor for BM (HR=0.54,95%CI:0.337-0.864,P=0.01).Conclusion:1The expression of BSP was stronger in BM group compared with non-BMgroup(P=0.002); the3year BM-free survival analysis showed a significantpoorer prognosis (P=0.001) in patients with stronger BSP protein expressionand it was an independent risk factor for BM(HR=2.179, P<0.001),which canbe used to early screen bone metastasis of breast cancer.2The3year BM-free survival analysis showed a significant poorer prognosis in patients with tumor size>3cm(P=0.014). Tumor size was also anindependent risk factor for BM(HR=1.580, P=0.011);3The3year BM-free survival analysis showed a significant better prognosisin patients with ER expression(P=0.024); ER was an independent protectivefactor for BM(HR=0.464, P=0.001);4HER-2showed weaker staining in BM arm compared with that inNBMarm(P=0.015), it was also an independent protective factor for BM(HR=0.540,P=0.010).