DOI: 10.1177/10760296261503703 ISSN: 1076-0296
Integrating
SERPINE1
Genotype With Clinical Indicators Improves Venous Thromboembolism Risk Prediction in Cancer Patients: A Retrospective Cohort Study From Xinjiang
Fan Guo, Yiliminuer Keremu, Gang Zhao, Huiting Hu, Yufei Wu, Xiaolu Yu, Chongxiao Hao, Weina Kong, Xiumin Ma
Objective
To develop and validate a venous thromboembolism (VTE) risk prediction model integrating genetic polymorphisms and routine clinical indicators in a cohort of cancer patients from Xinjiang, China.
Methods
A retrospective study was conducted on 1135 cancer patients (179 with VTE, 956 without) at Affiliated Cancer Hospital of Xinjiang Medical University between March 2024 and May 2025. Six gene polymorphisms (
SERPINE1
,
SERPINC1
,
PROC
,
MTHFR
,
F5
,
F2
) and clinical predictors were assessed. Chi-square tests and logistic regression identified independent predictors for model construction. Model performance was evaluated using ROC analysis, and three clinically optimized cut-off values were defined.
Results
The
SERPINE1
4G allele was identified as a strong genetic risk factor for VTE (4G/4G genotype: OR=2.330, 95%CI: 1.378-3.942; 4G/5G genotype: OR=2.399, 95%CI: 1.411-4.078). No significant associations were found for the other five genes. Age (OR=1.022, 95%CI: 1.008-1.037), D-dimer level (OR=1.288, 95%CI: 1.186-1.399), and hemoglobin level (OR=0.987, 95%CI: 0.979-0.995) were independent clinical predictors (
P
<0.01). The combined model achieved an area under the ROC curve (AUC) of 0.769 (95%CI: 0.731-0.807). Three optimized cut-off values were derived for different clinical contexts: high-sensitivity (90.5%), balanced (Youden index 0.434), and high-specificity (90.1%).
Conclusion
The
SERPINE1
4G allele independently predicts VTE risk in this cohort of cancer patients from Xinjiang. A prediction model integrating this genetic marker with age, D-dimer, and hemoglobin demonstrates good discriminatory performance. The three derived cut-off values may facilitate risk-stratified thromboprophylaxis in clinical practice; however, these findings are based on a retrospective, single-center cohort and internally validated only by ROC analysis.