Oops, you're using an old version of your browser so some of the features on this page may not be displaying properly.

MINIMAL Requirements: Google Chrome 24+Mozilla Firefox 20+Internet Explorer 11Opera 15–18Apple Safari 7SeaMonkey 2.15-2.23

Poster Display session

350P - Development and Validation of a machine learning-based early warning model for initial diagnosis of prostate cancer bone metastasis

Date

07 Dec 2024

Session

Poster Display session

Presenters

Leibo Wang

Citation

Annals of Oncology (2024) 35 (suppl_4): S1531-S1543. 10.1016/annonc/annonc1690

Authors

L. Wang1, W. He2, C. Zhao3

Author affiliations

  • 1 Urology, Beijing Jishuitan Hospital Guizhou Hospital(Guizhou Orthopaedic Hospital), 550000 - Guiyang/CN
  • 2 Urology, Beijing Jishuitan Hospital Guizhou Hospital, 550000 - Guiyang/CN
  • 3 Urology, The Affiliated Hospital of Zunyi Medical University, 563003 - Zunyi/CN

Resources

This content is available to ESMO members and event participants.

Abstract 350P

Background

The purpose of this study was to analyze the factors related to the occurrence of bone metastases in prostate cancer and to establish a nomogram for predicting the occurrence of bone metastases. This tool is intended to aid urologists in making early diagnosis and treatment decisions for patients at risk of bone metastases in prostate cancer.

Methods

In this two-center retrospective study, we included 311 newly diagnosed prostate cancer patients from the urology Departments of Zunyi Medical University Affiliated Hospital and Beijing Jishuitan Hospital. The clinical data of 260 patients from Zunyi Medical University Affiliated Hospital were randomly divided into a training cohort (n=184) and an internal validation cohort (Zunyi internal validation cohort, n=76) using a 7:3 ratio with computer-generated random numbers. Univariate analysis, multivariate logistic regression analysis, and Lasso regression analysis were used with the training cohort to determine risk factors for bone metastasis. Based on these factors, a nomogram prediction model was established. The model was validated using an internal validation cohort and an external validation cohort (Beijing external cohort, n=51) to assess its prognostic accuracy. All statistical analyses were performed using R4.2.2 software.

Results

A total of 311 patients were included in the study. Four independent risk factors for bone metastasis in prostate cancer were identified: fPSA (OR=7.34, P=0.0252), ALP (OR=40.34, P=0.0008), N (OR=5.82, P=0.0006), and tPSA. A nomogram was constructed based on these risk factors. The AUC of the nomogram was 0.911. After internal validation, the AUC was 0.885, and external validation showed an AUC of 0.839, indicating good discriminatory ability. The calibration curve demonstrated good consistency, and the decision curve analysis indicated high clinical applicability.

Conclusions

fPSA, ALP, N, and tPSA are independent risk factors for the development of bone metastasis in prostate cancer. The nomogram model based on these factors may have significant value for urologists in the early diagnosis and treatment of patients with bone metastasis in prostate cancer.

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.

This site uses cookies. Some of these cookies are essential, while others help us improve your experience by providing insights into how the site is being used.

For more detailed information on the cookies we use, please check our Privacy Policy.

Customise settings
  • Necessary cookies enable core functionality. The website cannot function properly without these cookies, and you can only disable them by changing your browser preferences.