In nlmixr2/rxode2: Facilities for Simulating from ODE-Based Models

options(cli.unicode=FALSE, crayon.enabled=FALSE);
options(knitr.table.format = "html")
knitr::opts_chunk$set( comment = "#>")
options(width=80, cli.width=80)
Sys.setenv(RSTUDIO_CONSOLE_WIDTH=80)

This is an example model for weight based dosing of daptomycin. Daptomycin is a cyclic lipopeptide antibiotic from fermented Streptomyces roseosporus.

There are 3 stages for weight-based dosing simulations: - Create rxode2 model - Simulate Covariates - Create event table with weight-based dosing (merged back to covariates)

Creating a 2-compartment model in rxode2

library(rxode2)

## Note the time covariate is not included in the simulation
m1 <- function() {
  model({
    CL ~ (1-0.2*SEX)*(0.807+0.00514*(CRCL-91.2))*exp(eta.cl)
    V1 ~ 4.8*exp(eta.v1)
    Q ~ (3.46+0.0593*(WT-75.1))*exp(eta.q);
    V2 ~ 1.93*(3.13+0.0458*(WT-75.1))*exp(eta.v2)
    A1 ~ centr;
    A2 ~ peri;
    d/dt(centr) <- - A1*(CL/V1 + Q/V1) + A2*Q/V2;
    d/dt(peri) <- A1*Q/V1 - A2*Q/V2;
    DV = centr / V1 * (1 + prop.err)
  })
}

Simulating Covariates

This simulation correlates age, sex, and weight. Since we will be using weight based dosing, this needs to be simulated first

set.seed(42)
rxSetSeed(42)
library(dplyr)
nsub=30
# Simulate Weight based on age and gender
AGE<-round(runif(nsub,min=18,max=70))
SEX<-round(runif(nsub,min=0,max=1))
HTm<-round(rnorm(nsub,176.3,0.17*sqrt(4482)),digits=1)
HTf<-round(rnorm(nsub,162.2,0.16*sqrt(4857)),digits=1)
WTm<-round(exp(3.28+1.92*log(HTm/100))*exp(rnorm(nsub,0,0.14)),digits=1)
WTf<-round(exp(3.49+1.45*log(HTf/100))*exp(rnorm(nsub,0,0.17)),digits=1)
WT<-ifelse(SEX==1,WTf,WTm)
CRCL<-round(runif(nsub,30,140))
## id is in lower case to match the event table
cov.df <- tibble(id=seq_along(AGE), AGE=AGE, SEX=SEX, WT=WT, CRCL=CRCL)
print(cov.df)

Creating weight based event table

s<-c(0,0.25,0.5,0.75,1,1.5,seq(2,24,by=1))
s <- lapply(s, function(x){.x <- 0.1 * x; c(x - .x, x + .x)})

e <- et() %>%
    ## Specify the id and weight based dosing from covariate data.frame
    ## This requires rxode2 XXX 
    et(id=cov.df$id, amt=6*cov.df$WT, rate=6 * cov.df$WT) %>%
    ## Sampling is added for each ID
    et(s) %>%
    as.data.frame %>%
    ## Merge the event table with the covarite information
    merge(cov.df, by="id") %>%
    as_tibble


e

Solving Daptomycin simulation

data <- rxSolve(m1, e,
             ## Lotri uses lower-triangular matrix rep. for named matrix
             omega=lotri(eta.cl ~ .306, 
                         eta.q ~0.0652,
                         eta.v1 ~.567,
                         eta.v2 ~ .191),
             sigma=lotri(prop.err ~ 0.15),
             addDosing = TRUE, addCov = TRUE)

print(data)

plot(data, log="y")

Daptomycin Reference

This weight-based simulation is adapted from the Daptomycin article below:

Dvorchik B, Arbeit RD, Chung J, Liu S, Knebel W, Kastrissios H. Population pharmacokinetics of daptomycin. Antimicrob Agents Che mother 2004; 48: 2799-2807. doi:(10.1128/AAC.48.8.2799-2807.2004)[https://dx.doi.org/10.1128%2FAAC.48.8.2799-2807.2004]

This simulation example was made available from the work of Sherwin Sy with modifications by Matthew Fidler

nlmixr2/rxode2 documentation built on Jan. 11, 2025, 8:48 a.m.