Predicting venous thromboembolism risk from endogenous thrombin potential-based normalized activated protein C sensitivity ratio assay: toward early assessment of combined oral contraceptives

Figures

• 

  Figure 1. Thrombin generation curves in the absence (continuous line) and presence of activated protein C (APC; dotted lines) of healthy pooled plasma (HPP; blue) and an APC-resistant phenotype plasma (pink), along with the normalized APC sensitivity ratio (nAPCsr) scale. The reference plasma used to calculate the nAPCsr values is derived from pooled plasma of healthy individuals (men and women in a 1:1 ratio) who are not using hormonal contraception and are not carriers of factor V Leiden or prothrombin G20210A mutations. In the presence of APC, the endogenous thrombin potential (ETP; area under the curve [AUC]) decreases by 90%, corresponding to a nAPCsr value of 1. In contrast, the APC-resistant phenotype plasma, typically seen in women using ethinylestradiol-containing products, shows increased thrombin generation both with and without APC compared with the reference plasma. On the nAPCsr scale, this results in a value above the upper limit of the normal range.

• 

  Figure 2. Box and whisker plot of the normalized activated protein C sensitivity ratio (nAPCsr) across study subgroups (A) and identified clusters (B). The box represents the central 50% of the data, with the lower edge indicating the first quartile and the upper edge the third quartile. The line inside the box represents the median, while the cross indicates the mean. The whiskers extend to the 2.5th and 97.5th percentiles. Differences between clusters were assessed using anova and an unpaired t-test with Welch’s correction. The significance threshold was set at .05. E2, estradiol; E4, estetrol; EE, ethinylestradiol; CPA, cyproterone acetate; DNG, dienogest; DRSP, drospirenone; DSG, desogestrel; LNG, levonorgestrel; NOMAC, nomegestrol acetate.

• 

  Figure 3. In silico modeling based on normalized activated protein C sensitivity ratio (nAPCsr) data and venous thromboembolism (VTE) relative risk (RR) estimates, as reported in the meta-analysis of de Bastos et al. [1]. COC, combined oral contraceptive; CPA, cyproterone acetate; DSG, desogestrel; EE, ethinylestradiol; LNG, levonorgestrel; Rs, Spearman’s rank correlation coefficient.

• 

  Figure 4. Predicted vs actual relative risk (RR) of venous thromboembolism based on normalized activated protein C sensitivity ratio data. Actual RR values are taken from the meta-analysis of de Bastos et al. [1]. The dashed line represents the identity line, indicating perfect agreement between predicted and observed values. CPA, cyproterone acetate; DSG, desogestrel; EE, ethinylestradiol; LNG, levonorgestrel; RMSE, root mean square error.

• 

  Figure 5. Venous thromboembolism (VTE) relative risk (RR) estimates for 4 combined oral contraceptives based on our in silico model: estetrol (E4) 15 mg with drospirenone (DRSP) 3 mg, estradiol (E2) 1.5 mg with nomegestrol acetate (NOMAC) 2.5 mg, ethinylestradiol (EE) 20 μg with DRSP 3 mg, and EE 30 μg with dienogest (DNG) 2 mg. COC, combined oral contraceptive; CPA, cyproterone acetate; DSG, desogestrel; LNG, levonorgestrel; nAPCsr, normalized activated protein C sensitivity ratio; Rs, Spearman’s rank correlation coefficient.

• 

  Supplementary Figure. Sensitivity analysis - model robustness after exclusion of the reference group Abbreviations: CI, confidence interval; CPA, cyproterone acetate; DRSP, drospirenone; DSG, desogestrel; EE, ethinylestradiol; LNG, levonorgestrel; nAPCsr, normalized activated protein C sensitivity ratio; RR, relative risk; SD, standard deviation; VTE, venous thromboembolism