Conditional survival analysis and real-time prognosis prediction for prostate cancer patients

Abstract To assess shifts in long-term survival outcomes for prostate cancer patients, the present study conducted conditional survival (CS) analysis and developed a personalized CS-nomogram to provide dynamic prognostic predictions. Data from 301,441 prostate cancer patients in the Surveillance, ep...

Full description

Saved in:
Bibliographic Details
Main Authors: Li Xu, Xudong Mao, Yiming Ding, Zhenwei Zhou, Yuanlei Chen, Mingchao Wang, Gonghui Li, Zeyi Lu
Format: Article
Language:English
Published: Nature Portfolio 2025-05-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-00420-9
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Abstract To assess shifts in long-term survival outcomes for prostate cancer patients, the present study conducted conditional survival (CS) analysis and developed a personalized CS-nomogram to provide dynamic prognostic predictions. Data from 301,441 prostate cancer patients in the Surveillance, epidemiology, and end results (SEER) database were analyzed, with cancer-specific survival (CSS) estimated using the Aalen Johansen estimator. Key predictive factors, including tumor size, lymph node involvement, distant metastasis, and histological grade, were identified via LASSO and multivariable Cox regression analyses to construct the nomogram. Results showed an increase in 12-year CSS from 94.3% at diagnosis to 99.4% for those surviving more than 5 years, with prostate cancer-specific mortality falling below 5% after 5–6 years of survival. The CS-nomogram demonstrated high predictive accuracy, achieving a concordance index (C-index) of 0.869 and area under the curve (AUC) values between 0.847 and 0.904 over a 12-year follow-up period, which provides dynamic, individualized prognostic estimates, supporting the long-term management of prostate cancer patients.
ISSN:2045-2322