Prostate Cancer Risk Prediction Model
Author Information
Author(s): Ji Wen-Tong, Wang Yong-Kun, Han Zhan-Yang, Wang Si-Qi, Wang Yao
Primary Institution: China-Japan Union Hospital of Jilin University
Hypothesis
To construct a prediction model for clinically significant prostate cancer (csPCa) based on traditional diagnostic methods.
Conclusion
The study developed a cost-effective and highly accurate model for predicting clinically significant prostate cancer before biopsy.
Supporting Evidence
- The model showed an area under the ROC curve of 0.890 in the training set.
- The model had a negative predictive value of 89.8% and a positive predictive value of 68.0%.
- The calibration curves indicated good calibration of the model.
- The decision curve analysis demonstrated good clinical utility of the model.
Takeaway
This study created a tool to help doctors predict if someone has serious prostate cancer using simple tests, which can save money and avoid unnecessary procedures.
Methodology
Retrospective analysis of 1196 patients using logistic regression to establish a csPCa risk prediction model.
Potential Biases
Potential selection bias due to the majority of patients being from underdeveloped areas.
Limitations
The study was based on a single-center retrospective design, which may limit generalizability.
Participant Demographics
Asian patients who underwent transrectal ultrasound-guided biopsy.
Statistical Information
P-Value
<0.001
Confidence Interval
95%CI: 0.865-0.816
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website