Abstract 50P
Background
Bone metastases affect 80% of patients with advanced breast cancer(BC), resulting in several skeletal-related events, which results in significantly reduced overall survival and poorer quality of life for patients. This study retrospectively analyzed the clinical features of breast cancer bone metastasis(BCBM) patients and the risk factors related to the incidence and prognosis of bone metastasis(BM), and established and validated the prediction model of risk and prognosis of BM.
Methods
Clinical characteristics of 2617 patients with BC treated in Liaoning Cancer Hospital from January 2018 to December 2020 were collected, including 629 BCBM patients and 1988 patients without BM. All patients were randomly divided into a training group and a verification group, including 1755 in the training set and 862 in the verification set. A stepwise LR regression analysis method was used to screen out the predictors of BCBM prediction model and nomogram model was constructed.
Results
The results of multivariate logistic regression analysis showed that T stage, N stage, tissue type, menstrual status, lymphocyte, hemoglobin, alkaline phosphatase and CA153 were independent risk factors for BCBM (P<0.05). The AUC (95%CI) in the training set and validation set of the nomogram constructed by the selected risk factors were 0.930 and 0.938 respectively. Through Hosmer-Lemeshow test on the prediction model, Bootstrap sampling conducted 1000 internal calibration on the nomogram, which showed that the calibration curves of the training set and the verification set were close to the ideal 45°reference line. The analysis of decision curve of training set and validation set in this study shows that it has strong clinical practicability in a certain threshold probability range.
Conclusions
The model is a practical tool for predicting BM in BC patients, which can provide an effective decision-making basis for clinicians and a relatively reliable clinical preventive treatment plan to a certain extent.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.