Risk Prediction Model for Acute Kidney Injury After First Course of Cisplatin


Key Points

  • A risk model was developed based on patient age, cisplatin dose, hypertension, and serum albumin level.
  • The c-statistic was 0.72 and 0.70 in the development and validation cohorts, respectively.

As reported in the Journal of Clinical Oncology, Motwani et al have developed a predictive model for acute kidney injury following a first course of cisplatin that includes patient age, cisplatin dose, hypertension, and serum albumin level.

Study Details

The study involved data from 2,118 patients in a development cohort (Massachusetts General Hospital) and 2,363 patients in a validation cohort (Dana-Farber Cancer Institute, Brigham and Women’s Hospital) who received cisplatin between 2000 and 2016. Cisplatin-associated acute kidney injury was defined as a 0.3 mg/dL increase in serum creatinine within 14 days of first receiving cisplatin. Multivariable logistic regression models were used to derive a scoring model from the development cohort that was tested in the validation cohort.

Predictive Model

Cisplatin-associated acute kidney injury was observed in 13.6% of the development cohort and 11.6% of the validation cohort. Factors significantly associated with acute kidney injury (with score assigned in predictive model) included: age 61 to 70 years (odds ratio [OR] = 1.64, P = .001, score = 1.5) and 71 to 90 years (OR = 2.97, P < .001, score = 2.5) vs ≤ 60 years (score = 0); cisplatin dose of 101 to 150 mg (OR = 1.58, P = .007; score = 1) and > 150 mg (OR = 3.73, P < .001, score = 3) vs < 100 mg (score = 0); history (score = 2) vs no history (score = 0) of hypertension (OR = 2.10, P < .001); and serum albumin of 2.0 to 3.5 vs > 3.5 g/dL (OR = 2.21, P < .001, score = 2 for 1.3–3.5 g/dL and 0 for > 3.5 g/dL). Baseline estimated glomerular filtration rate was not significantly associated with the risk of acute kidney injury.

The c-statistics of the score-based model in the development and validation cohorts were 0.72 and 0.70, respectively. Scores of 0, 3.5, and 8.5 were associated with a probability of acute kidney injury of 0.03, 0.12, and 0.51.  

The investigators concluded, “A score-based model created by using the patient’s age, cisplatin dose, hypertension, and serum albumin is predictive of [cisplatin-associated acute kidney injury].”

Shveta Motwani, MD, of Dana-Farber Cancer Institute, is the corresponding author for the Journal of Clinical Oncology article.

The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.