Validation of clinical risk assessment scores for venous thromboembolism in patients with cancer: a population-based cohort study.

Abstract

Background: Guidelines recommend using risk assessment tools to identify ambulatory patients with cancer at high risk of venous thromboembolism (VTE).

Objective: We aimed to validate a new cancer-associated thrombosis (CAT) risk score in a population-based healthcare setting.

Methods: We used healthcare registry data and electronic medical records from the Central Denmark Region to calculate the new CAT risk score and the guideline-recommended Khorana score in patients with a first-time cancer diagnosis who initiated systemic cancer therapy. Patients were followed for six months after initiation of therapy. The outcome was any VTE identified through hospital discharge diagnoses. Discrimination was assessed using c-statistics.

Results: We included 12,471 patients from 2012 through 2020. Of these, 416 (3.3%) developed VTE. The c-statistic was 0.71 (95% confidence interval [CI]: 0.68-0.74) for the new CAT score and 0.66 (95% CI: 0.63-0.70) for the Khorana score. The six-month cumulative VTE incidence was 5.0% in 6,175 patients classified as high risk by the new CAT score, compared with 1.7% in 6,296 patients classified as low risk. The six-month cumulative VTE incidence was 5.2% in 4,263 patients classified as high risk by the Khorana score, compared with 2.4% in 8,208 patients classified as low risk.

Conclusion: The new CAT score had a discriminatory ability similar to that reported in the derivation study. The c-statistic was numerically higher than that of the Khorana score. Our findings support implementation of the new CAT score to identify ambulatory patients with cancer who are at high risk of VTE.