Protein C Gene Variant Tied to Higher Venous Thrombosis Risk in Taiwanese Cohort

A common inherited variant in the protein C gene, PROC rs146922325, is associated with an increased risk of venous thrombosis in Taiwanese patients, according to a retrospective case-control study drawing on genotyping data from the Taiwan Precision Medicine Initiative. Researchers at Taichung Veterans General Hospital identified 805 adult carriers of the variant and compared them with 8,050 age- and sex-matched non-carriers. Carriers showed markedly lower protein C activity and a higher prevalence of venous thrombotic events, and the association persisted after statistical adjustment for a broad panel of clinical covariates.

Protein C is a vitamin K-dependent protein produced in hepatocytes that acts as a natural brake on blood coagulation. Working with its cofactor protein S, activated protein C degrades two procoagulant cofactors, Va and VIIIa, thereby restraining clot formation. The link between low plasma protein C levels and venous thrombosis was first described in a 1981 family study, and hereditary protein C deficiency caused by pathogenic PROC variants has since been recognized as a risk factor for venous thrombosis. More than 300 variants of the gene, located on chromosome 2q14.3, have been reported, and heterozygous deficiency is estimated to occur in roughly 1 in 200 to 1 in 500 births, while severe homozygous or compound heterozygous forms are far rarer.

The rs146922325 variant, also known as c.565C>T or p.Arg189Trp, has previously been identified as a common PROC defect and thrombotic risk factor in Taiwanese and other East Asian populations, accounting for up to 50 percent of protein C deficiency cases reported in Chinese patients. Earlier studies, however, were generally limited by small sample sizes, family-based designs, or incomplete adjustment for clinical covariates, leaving the precise magnitude of the thrombotic risk in a large hospital-based population uncertain. The new analysis set out to provide population-based risk estimates adjusted for potential confounders, with secondary aims of examining sex differences and the age at which venous thrombosis first appears.

The study drew on 58,091 patients aged 18 or older who visited 28 surgical and medical outpatient clinics at Taichung Veterans General Hospital between July 2019 and November 2022. Genotyping was performed with the Taiwan Biobank version 2 array, designed for Taiwan’s Han Chinese population and containing more than 714,000 single-nucleotide polymorphisms. After extensive quality control, including exclusion of samples with low call rates, sex inconsistencies, and related individuals, 591,048 high-quality SNPs were retained. The rs146922325 variant was included on the array because of its prior annotation as a pathogenic variant in ClinVar and its established association with hereditary protein C deficiency in East Asian populations.

Of 809 individuals carrying the variant, four were under 18 and were excluded, leaving 805 adult carriers, of whom 804 were heterozygous and one was homozygous. Because only a single homozygote was available, the researchers combined heterozygous and homozygous carriers under a dominant genetic model and compared them with 8,050 wild-type non-carriers selected at a 1:10 ratio matched for age and sex. The mean age of both groups was 58.46 years, and 52.05 percent of participants were female. Baseline comorbidities were broadly similar between the two groups.

The key biochemical finding was a substantial difference in protein C activity: carriers averaged 71.28 percent activity compared with 114.58 percent in non-carriers, a difference that was highly significant. Protein S, antithrombin, D-dimer, and activated partial thromboplastin time did not differ significantly between the groups. Venous thrombotic events were more prevalent among carriers, affecting 4.10 percent versus 2.48 percent of non-carriers, while arterial thrombotic events occurred at nearly identical rates of 11.80 percent and 11.42 percent. Overall thrombotic events, combining arterial and venous categories, did not differ significantly between the groups, underscoring that the genetic effect was specific to the venous side of the circulation.

In univariable logistic regression, carrier status was associated with an increased risk of venous thrombosis, with an odds ratio of 1.69 and a 95 percent confidence interval of 1.16 to 2.47, but showed no association with arterial thrombosis. An allelic analysis focusing on the variant T allele produced a similar result, with an odds ratio of 1.74. After multivariable adjustment for age, sex, smoking status, hypertension, diabetes, hyperlipidemia, atrial fibrillation, heart failure, chronic liver disease, chronic obstructive pulmonary disease, chronic kidney disease, antiphospholipid syndrome, inflammatory bowel disease, obesity, lower limb fracture, pregnancy, cancer, and hormone therapy, carrier status remained independently associated with venous thrombosis, with an adjusted odds ratio of 1.61. No clear dose-response pattern could be established because the cohort contained only one homozygous individual.

Sex-stratified analyses suggested a similar direction of association in both sexes, reaching statistical significance in females but not in males, while the formal sex-by-genotype interaction test was not significant, indicating no evidence of a true sex-specific effect. The median age at onset of venous thrombosis did not differ significantly between carriers and non-carriers, and subgroup analyses by sex likewise showed no significant differences, although the authors note that the onset-age analysis was constrained by the small number of venous thrombosis events among carriers, roughly 33, which limits statistical power.

Two prespecified sensitivity analyses addressed potential biases related to anticoagulant exposure and disease timing. Because warfarin suppresses vitamin K-dependent coagulation factors including protein C, laboratory comparisons were restricted to warfarin-naive participants, among whom carriers still showed significantly lower protein C activity than non-carriers. In this restricted analysis, the association between carrier status and venous thrombosis was no longer statistically significant in the multivariable model, with an adjusted odds ratio of 1.26. A second sensitivity analysis restricted to incident venous thrombosis occurring after the date of genotyping substantially strengthened the association, yielding an adjusted odds ratio of 2.31 for carrier status and 2.47 per T allele, with consistent results in both sexes. The authors interpret this enhancement as evidence that including preexisting thrombotic events in the primary analysis may have introduced reverse causation bias, and they regard the incident-case analysis as a more temporally rigorous estimate.

The effect size observed in this Taiwanese cohort was notably smaller than risks reported in some earlier studies, a discrepancy the researchers attribute to differences in study design and population context. Previous work was often conducted in clinically referred populations enriched for thrombotic events, which can inflate genetic risk estimates through referral and ascertainment bias, whereas the present analysis used an unselected biobank-based cohort that yields more conservative, population-representative estimates. Regional differences also matter: Factor V Leiden and the prothrombin G20210A variant, the two most common inherited thrombophilia variants in Western populations, are virtually absent in East Asians, which simplifies interpretation of the independent effect of PROC variants in this setting. The authors caution that the tertiary referral nature of the hospital means the T allele frequency and the prevalence of venous thrombosis among non-carriers may exceed general-population values, so absolute risk estimates may not be fully generalizable to the entire Taiwanese population. They suggest that identifying carriers could eventually support targeted risk assessment, particularly in individuals exposed to additional acquired risk factors such as pregnancy, surgery, prolonged immobilization, or hormone therapy, while emphasizing that the cost-effectiveness and optimal implementation of genotype-guided prevention require further validation, and that independent confirmation in prospective population-based cohorts is needed.

Subject of Research: Association of the PROC rs146922325 variant with venous thrombosis risk in a Taiwanese case-control cohort

Article Title: Association of the PROC rs146922325 variant with venous thrombosis in a Taiwanese population

Article References: Chen, C.-Y., Chen, I.-C., Lin, G.-C., Hsiao, T.-H., & Wang, J.-D. (2026). Association of the PROC rs146922325 variant with venous thrombosis in a Taiwanese population. Blood Research, 61(1), Article 48. https://doi.org/10.1007/s44313-026-00168-9

Image Credits: AI Generated

DOI: 10.1007/s44313-026-00168-9

Keywords: PROC rs146922325, protein C deficiency, venous thrombosis, Taiwan Precision Medicine Initiative, case-control study, thrombophilia, genetic risk stratification, protein C activity, East Asian genetics, logistic regression, sensitivity analysis, Taiwanese population