Cardio-oncology & otherDiagnostic or prognostic study
A Biomarker-Based Diagnostic Model for Cardiac Dysfunction in Childhood Cancer Survivors
- Diagnostic accuracy study
- multicentre
- cross-sectional
- internal validation by bootstrapping
Authors' conclusion
The biomarker-based diagnostic model proves effective in ruling out LV dysfunction, offering the potential to minimize unnecessary surveillance echocardiography in childhood cancer survivors.
Study context
- Population
- Childhood cancer survivors (median age 34.2 years) and siblings (median age 36.8 years)1,334 survivors
- Comparison
- Biomarker-based diagnostic model vs clinical characteristics alone
- Outcome
- Ruling in or ruling out left ventricular (LV) dysfunction (LVEF <50%; LVEF <45%)
Performance
As reported in the abstract
Diagnostic performance for LV dysfunction
| Test or model | C statistic, LV ejection fraction (LVEF) <50% | C statistic, LVEF <45% | Ruled out, LVEF <50% | Sensitivity, LVEF <50% | NPV, LVEF <50% | Ruled out, LVEF <45% | Sensitivity, LVEF <45% | NPV, LVEF <45% |
|---|---|---|---|---|---|---|---|---|
| Biomarker model | 0.73 | 0.86 | 16.9% | 95.4% (95% CI: 90.4%-99.3%) | 97.5% (95% CI: 94.6%-99.7%) | 53.0% | 91.1% (95% CI: 79.2%-100%) | 99.4% (95% CI: 98.7%-100%) |
| Clinical characteristics | 0.69 | 0.80 |
Further findings
| Outcome | Values |
|---|---|
| Abnormal NT-proBNP | Survivors 22.1% · Siblings 5.4% |
| hs-cTnT >10 ng/L | Survivors 5.9% · Siblings 5.0% |
Original publication
A Biomarker-Based Diagnostic Model for Cardiac Dysfunction in Childhood Cancer Survivors. JACC CardioOncol. Published 16 April 2024. doi:10.1016/j.jaccao.2024.02.008
Summarised from the published abstract; every number is checked against it