Physiological based pharmacokinetic modeling and simulation for the distribution of tranexamic acid in human

Goal

The goal of this Physiological based pharmacokinetic (PBPK) modeling and simulation exercise is to understand the factor that impact tranexamic acid (TXA) distribution in human. In particular, the distribution of TXA distribuition in skin after oral intake of the drug. I would like to know more under the dosing regimen of Transino II (by Daiichi Sankyo Healthcare, which aims to treat melasma), how TXA would distribute and the exposure comparison between different organs

Note this is purelu a modeling and simulation (M&S) exercise. The result should be taken with a healthy dose of skeptism, and should not be used for medical advice.

Set up

The simulation is run with R 4.1 and the following packages (version)

• tidyverse (1.3.1)
• mrgsolve (0.11.1)
• ggplot2 (3.3.5)
• gridExtra (2.3)
• PKPDmisc (3.0.0)
• Rcpp (1.0.7)

Model Development

The PBPK model is borrowed from Utsey et al., 2020 with an additional compartment added to model TXA absoption with first-order kinetics and pancreas being lumped into the rest of body. Additional drug property related to TXA is obtained from its PubChem page.

The base model is built with following assumptions:

• All drug in the plasma is unbound
• blood:plasma ratio = 3
• TXA adsoption rate in the gut lumen = 0.4 h-1

To validate the model, we use the observed data from Pilbrant et al., 1981 with both IV and oral observed data from 3 healthy male wolunteers. I conclude the (parameters) provide a reasonable fit, and the pharmacokinetics is overall inline with what is described in Dunn 1999 and I will be using them for the following simulation.

Oral dosing from Transino II

The recommended oral dosing of Transino II roughlt translated into 1.5g of TXA per 12 hour. The simulation result is provided as follows. The blue line on the left indicate the Kd of TXA binding to plasminogen Dunn 1999, the blue line on the right indicates effective TXA concentration to inhibit fibrinolysis Picetti et al., 2019. The simulation predicts that the TXA concentation in the skin will be somewhat effective in inhibiting pigment forming for at least 12 hours, but the concentration in plasma is unlikely to inhibit the breakdown of blood clot.

5 dose of oral TXA for every 12 hours over 3 days is simulated. Multiple dose leads to slightly higher Cmax of TXA human, but with almost no impact on Tmax, or the time of exposure.

In addition, after TXA dosing stopped, all the TXA is cleared from plasma and skin in 24 hours.

When I look at the exposure in different organs, plasma exposure is unsurprisingly, the highest. In contrast, skin is one of the organs have the least exposure. This indicates that oral take of TXA results in most of the TXA not being delivered to skin.

Sensitivity Analysis

Blood:plasma ratio

The main reason for conducting blood:plasma ratio is due to the parameter is assumed, and my fitted result is different from Kane et al., 2021. The local sensitivity analysis show that blood:plasma ratio has a major impact on Tmax, Cmax, and the exposure. Higher blood:plasma ratio leads to smaller Tmax, lower Cmax, and lower exposure.

Absorption rate

I want to carry out the senseitivity analysis on TXA absoprtion rate not only because this parameter is assumed in the current model, but also Dunn 1999 reported a slightly different bioavailability in the with/ without food situation. As expected, higher oral absoprtion rate leads to earlier and higher peak of skin TXA concentration.

Body weight

The interest in body weight is partially based on the validation data are from heavier males, while small females may also take this drug. A 10% increase in the body weight barely has any impact on Tmax, Cmax, or the exposure in skin.

However, it is reasonable to caution that this weight adjustment can address gender-related issue for skin exposure, as women has the same percentage of cardiac output goes into skin (Ref). However, this may not be the case for exposure in other organ (e.g. adipose).

Renal Impairment

Since TXA is mostly cleared through renal pathway, thus it is reasonable to look at how renal impairment would impact the pharmacokinetics of the drug in both skin and in plasma. As expected, the more sereve the renal impairment is (base on FDA classification), the higher TXA exposure a person would has, in eitehr skin or in plasma.

Conclusion and Discussion

Overall, the plasma concentraion following the Transino II recommended dosing scheme would lead to safe plasma TXA concentration with no apparent risk of clotting. The concentration of TXA after oral intake peaks in 3 hours after the drug in the skin, and the concentration stays somewhat effective in preventing pigmentation for the next 12 hours.

(talk about uncertainty and weaknesses).

(talk about tissue volume adjustment)

We acknowledge the science behind our work isn't new, as Kane et al., 2021 published a similar study on TXA with more extensive discussion on different route of administration. But I concede this repo provides an open-source implementation is is beneficial for the science.

Content of the folder

• TranexamicAcidPBPK.cpp (model file)

• Transamin.Rmd (main script)

folders:

• img (where images that are used in this readme file stored)

• doc (where reference papers are stored)