Predictive Performance of Cardiovascular Risk Scores in Cancer Survivors From the UK Biobank
Authors' conclusion
Existing cardiovascular risk scores have significantly worse predictive accuracy in cancer survivors compared with noncancer comparators, leading to an underestimation of risk in this cohort.
Study context
- Population
- UK Biobank participants with and without a history of cancermatched cohort
- Participants
- 31,534 vs 126,136
- Comparison
- Cancer survivors vs matched noncancer controls
- Cancer survivorsn = 31,534
- Matched controlsn = 126,136reference
- Outcome
- Predictive performance (AUC, balanced accuracy, sensitivity) of 7 CV risk scores (QRISK3, SCORE2/Systematic Coronary Risk Evaluation for Older Persons (SCORE-OP), Framingham Risk Score, PCP-HF, CHARGE-AF, QStroke, CHA2DS2-VASc) for incident CV events
- Follow-up
- 10 years
Methodological detailspropensity-matched cohort, UK Biobank, propensity-score matched, 10-year follow-up via health record linkage
- Propensity-matched cohort, UK Biobank
- Propensity-score matched
- 10-year follow-up via health record linkage
Performance
As reported in the abstract
- Performance metrics significantly worse for all risk scores in cancer cases than in matched controls
- Risk score distributions near identical in cases and controls
- Matched cohort
Further findings
Incidence of CV outcomes (any cancer)
- Significantly higher incidence of all CV outcomes in any cancer than matched controls
Hematologic malignancies (subgroup)
- Significantly higher outcome rates and poorer risk score performance than matched controls; most notable differences
Stroke prediction, brain/CNS cancer (subgroup)
- Very poor performance; predictive accuracy lower than noncancer controls
- more than 30%
Original publication
Predictive Performance of Cardiovascular Risk Scores in Cancer Survivors From the UK Biobank. JACC CardioOncol. Published 23 July 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →