Abstract: FR-PO650
Development of a Radiomic-Clinical Nomogram to Predict the Treatment Resistance of Myeloperoxidase ANCA-Associated Vasculitis
Session Information
- Glomerular Diseases: Clinical, Outcomes, Trials - II
November 04, 2022 | Location: Exhibit Hall, Orange County Convention Center‚ West Building
Abstract Time: 10:00 AM - 12:00 PM
Category: Glomerular Diseases
- 1303 Glomerular Diseases: Clinical‚ Outcomes‚ and Trials
Author
- Yong, Zhong, Xiangya Hospital Central South University, Changsha, Hunan, China
Background
We aimed to develop a radiomics nomogram to predict the treatment resistance of MPO-AAV patients.
Methods
MPO-AAV patients were randomly assigned to a training cohort and a test cohort. Treatment-resistance related features were selected using multivariate logistic regression to obtained the radiomic score (Rad-score). Two models were built. The clinical utility of the good models was assessed by decision curve analysis (DCA). The calibration of combined radiomics nomogram was evaluated.
Results
Serum creatinine and 9 radiomics features were selected, which were related to the treatment resistance of MPO-AAV. The accuracy of Model 2 (radiomics nomogram) for the prediction of treatment resistance was 0.948 and 0.913 in the training and test cohorts, which was higher than Model 1 (radiomics) with the AUC of 0.824 (p=0.039), and 0.898 (p=0.043) respectively. The DCA curve demonstrated that Model 2 had a higher net clinical benefit than that of Models 1. The calibration curves of Model 2 closely aligned with the true treatment resistance rate in the training (p =0.28) and validation sets (p =0.70). Furthermore, the validation cohort gained a good AUC of 0.929 in the Model 2.
Conclusion
The radiomics nomogram is a useful for predicting the treatment resistance of MPO-AAV.