Updating Predictive Model for Cisplatin-Induced Acute Kidney Injury: Incorporating Concomitant Medications and the...

Area of Science:

• Nephrology
• Oncology
• Clinical Prediction Models

Background:

• Cisplatin-induced acute kidney injury (C-AKI) complicates cancer treatment.
• Existing C-AKI prediction models do not incorporate concomitant medications.
• This study aimed to improve C-AKI prediction by including medication data.

Purpose of the Study:

• To enhance existing cisplatin-induced acute kidney injury (C-AKI) prediction models.
• To investigate the impact of concomitant medications on C-AKI prediction.
• To evaluate the clinical utility of an updated C-AKI prediction model.

Main Methods:

• Retrospective analysis of 1785 patients receiving cisplatin.
• Assessed existing model performance and explored new predictors (medications).
• Utilized C-statistic, net reclassification improvement (NRI), and decision curve analysis (DCA) for model evaluation.

Main Results:

• The existing model showed poor discrimination (C-statistic: 0.621).
• Incorporating magnesium supplementation improved the C-statistic to 0.695.
• The updated model demonstrated superior NRI (0.143) and better net benefits in DCA.

Conclusions:

• The existing prediction model had unsatisfactory performance in the studied cohort.
• Magnesium supplementation improved prediction but is limited as a novel predictor due to standard care status.
• Developing C-AKI prediction models requires consideration of uniform supportive care regimens.