Implement Residua Unexplained Variability in a Population PK Model

Phoenix Certara Forums Modelling and Simulation
1

I am conducting a model-based simulation using PK parameters from a publication. The publication reports(table):

Table
Table921×159 7.2 KB

I attempted to include StudyA as a covariate, but not sure if I am doing right. I would greatly appreciate any guidance on how to properly implement this. Here are my settings

Capture 1
Capture 1570×290 38.1 KB

:
Residual error model: “Add+Mult”
tvCMultStdev: 0.14 (Convert residual error (38.3%) to variance)
dcCMultStdevStudyA: -0.455

Thanks

3

Here is one way to implement. Note that in this implementation the STUDYCEps1 and 2 parameters would be the individual CEps estimates by study, not a shift from reference CEps. You could implement a shift by unfreezing CEps and then modifying SigmaSTUDY to e.g., return 1 for STUDY==1 and STUDYCEps2 for STUDY!=1. If you share the paper I can give a more tailored example.

test(){
deriv(Aa = - Ka * Aa)
deriv(A1 = Ka * Aa - Cl * C)
dosepoint(Aa)
C = A1 / V
error(CEps(freeze) = 1)
SigmaSTUDY = ((STUDY==1) ? STUDYCEps1 : STUDYCEps2)
observe(CObs = C * (1 + CEps*SigmaSTUDY))
stparm(Ka = tvKa * exp(nKa))
stparm(V = tvV * exp(nV))
stparm(Cl = tvCl * exp(nCl))
stparm(STUDYCEps1 = tvSTUDYCEps1)
stparm(STUDYCEps2 = tvSTUDYCEps2)
fcovariate(STUDY())
fixef(tvKa = c(, 1.27, ))
fixef(tvV = c(, 93.4, ))
fixef(tvCl = c(, 1.88, ))
fixef(tvSTUDYCEps1 = c(, .04, ))
fixef(tvSTUDYCEps2 = c(, .25, ))
ranef(diag(nV, nCl, nKa) = c(0.1, 0.1, 0.1))
}

2 Likes
4

Thanks for your kindly help. I will review your comments and feedback. Thanks again!
DV