NextDose: A web-based Bayesian dose forecasting tool

Warfarin

Target Concentration

Warfarin dose individualization is typically based on a target international normalized ratio (INR) of 2.5 with an acceptable range of 2 – 3. An INR of 2 is expected when synthesis of the coagulation factors in the prothrombin complex has been inhibited by 50%. Warfarin is administered as a racemic mixture of S- and R-enantiomers. Both influence the synthesis of coagulation factors by modifying the activity of vitamin K epoxide reductase (VKOR). S-warfarin is an inhibitor of VKOR while R-warfarin appears to act as a competitive inhibitor of S-warfarin (Xue, Holford et al. 2017).

The relationship between the time course of warfarin (S- and R-) concentration and changes in INR has been described using a turnover model for prothrombin complex (Xue, Holford et al. 2017). This model can be used to predict the steady state combination of S- and R- warfarin that will achieve a target steady state INR and the warfarin dose predicted from these target concentrations.

The use of NextDose has been compared to standard of care in a randomized controlled trial. The demonstrated that NextDose achieved improved time within acceptable range of INR and reduced clinically relevant bleeding events(Xue, Ma et al. 2024).

NextDose Model

NextDose uses the pharmacokinetic model for S- and R- warfarin developed to describe the change in INR in a population of patients who had undergone cardiac surgery (Xue, Holford et al. 2017). INR observations are used to update Bayesian estimates of both PK and PD parameters which are then used to predict the warfarin dose required to achieve the target iNR (Figure 1).

Figure 1 Example of INR change after started warfarin and predicted warfarin dose to achieve a target INR of 2.

If S- or R- warfarin concentration observations are available they can also be used to provide information about the PK component of the warfarin model and help with dose individualization (Figure 2).

Figure 2 Observations that may be used with warfarin.

Both amiodarone and fluconazole are known to inhibit metabolism of warfarin. The start and stop dates of amiodarone and fluconazone can be entered as observations to indicate treatment with these medications (Figure 2).

Vitamin K1 may be given to overcome the anticoagulant effect of warfarin. This effect of vitamin K1 can be included by specifying the date that vitamin K1 is started. A finish date one day after the start date should be entered to allow resumption of warfarin effect (Figure 2).

Xue, L., N. Holford, X. L. Ding, Z. Y. Shen, C. R. Huang, H. Zhang, J. J. Zhang, Z. N. Guo, C. Xie, L. Zhou, Z. Y. Chen, L. S. Liu and L. Y. Miao (2017). "Theory-based pharmacokinetics and pharmacodynamics of S- and R-warfarin and effects on international normalized ratio: influence of body size, composition and genotype in cardiac surgery patients." Br J Clin Pharmacol 83(4): 823-835.

Xue, L., G. Ma, N. Holford, Q. Qin, Y. Ding, J. A. Hannam, X. Ding, H. Fan, Z. Ji, B. Yang, H. Shen, Z. Shen and L. Miao (2024). "A Randomized Trial Comparing Standard of Care to Bayesian Warfarin Dose Individualization." Clinical Pharmacology & Therapeutics 115(6): 1316-1325.