PB0919 - Polygenic risk scores for prediction of cancer-associated venous thromboembolism: a UK Biobank analysis

Monday, July 11, 2022

6:30 PM – 7:30 PM

Location: Exhibition

Presentation Format: In-Person

Presenting Author(s)

NG

Noori A.M. Guman, MD

PhD Student
Dept. of Vascular Medicine, Amsterdam UMC location AMC, University of Amsterdam, Amsterdam, The Netherlands
Amsterdam, Noord-Holland, Netherlands

Co-Author(s)

FM

Frits I. Mulder

PhD student
Dept. of Vascular Medicine, Amsterdam UMC location AMC, University of Amsterdam, Amsterdam, The Netherlands
Amsterdam, Noord-Holland, Netherlands
BF

Bart Ferwerda

PhD
Amsterdam University Medical Centers, University of Amsterdam
Amsterdam, Noord-Holland, Netherlands
AZ

Aeilko H. Zwinderman

Professor of Biostatistics
Amsterdam University Medical Centers, University of Amsterdam
Amsterdam, Noord-Holland, Netherlands
HB

Harry R. Büller

MD
Department of Vascular Medicine, Amsterdam UMC, University of Amsterdam, Amsterdam, The Netherlands
Amsterdam, Noord-Holland, Netherlands
Nick van Es, MD PhD

MD
Department of Vascular Medicine, Amsterdam University Medical Center, Amsterdam, The Netherlands
Amsterdam, Noord-Holland, Netherlands
PK

Pieter W. Kamphuisen

MD
Tergooi MC and Amsterdam University Medical Centers
Hilversum, Noord-Holland, Netherlands

Background: Guidelines recommend primary thromboprophylaxis for selected cancer patients at high risk of venous thromboembolism (VTE). Improvement of traditional clinical risk assessment models for VTE in cancer patients is needed.

Aims: To assess the predictive performance of three polygenic VTE risk scores derived in the general population (De Haan et al. 2012 [5-SNP score], Klarin et al. 2019 [297-SNP score], and Lindström et al. 2019 [37-SNP score]) in cancer patients enrolled in the UK Biobank, a large-scale prospective, observational cohort.

Methods: Participants with an incident cancer diagnosis after UK Biobank enrollment were included, using data from cancer registries or inpatient diagnosis data. Those with an anticoagulant therapy prescription at enrollment or history of VTE were excluded. The primary outcome was pulmonary embolism or deep-vein thrombosis. Patients were followed for 12 months after cancer diagnosis. Performance of the continuous scores was assessed by time-dependent c-indices and subdistribution hazard ratios (SHR) in univariable and multivariable analyses adjusted for sex and age. The risk in patients in the upper quartile of the polygenic scores was compared to that in the lower three quartiles.

Results: 36,479 cancer patients (median age 66; 49% females) were included of whom 874 (2.4%) developed VTE during 12-month follow-up. All continuous scores were significantly associated with VTE in univariable and multivariable analyses (SHRs 1.3 to 1.5). The 297-SNP score showed the best discrimination (c-index 0.59, 95%-CI: 0.58-0.61). For this score, the cumulative VTE incidence was 3.6% (95%CI: 3.2-4.0) in the upper quartile versus 2.0% (95%CI: 1.8-2.2) in the lower quartiles (SHR 1.8, 95%CI: 1.6-2.1). Performance of all scores is summarized in the Table and Figure.

Conclusion(s): Polygenic VTE risk scores developed for the general population can identify cancer patients with about a 2-fold increased VTE risk, and may potentially improve VTE risk assessment on top of clinical risk scores.