Dose-scale analysis to estimate relative bioavailability

[hrishikesh1985]

As per the USFDA guidance of Orlistate,( [https://www.accessdata.fda.gov/drugsatfda_docs/psg/PSG_020766.pdf]
Data from PD study on fecal fat excretion should be analyzed based on dose-scale analysis to
estimate relative bioavailability.

Model is y= E0 + Emax* Dose* Fi/ED50 + (Dose*Fi)
Where y = Response, Dose = Administered dose, E0 = Baseline response in the absence
of the drug, Emax = Fitted maximum drug effect, ED50 = Dose required to produce 50%
the fitted maximum effect, and i = Treatment indicator (0 = Ref, 1 = Test), with the
understanding that F0 = 1 and that F1 is the relative potency used to evaluate
bioequivalence.

The same approach has to be used for albuterol Bronchoprovocation study. [https://www.accessdata.fda.gov/drugsatfda_docs/psg/PSG_020503.pdf]
I am struggling to prepare the dataset and model (snapshot attached) in which have four treatment i,e, placebo, test, reference(2 doses) and from which F- relative bioavailability needs to be calculated

Data_al2 <- function() {
  ini({
    temax <- -20
    ted50 <- 20
    te0 <- 1
    eta.emax ~ pdDiag(1)
    eta.ed50 ~1
    eta.e0 ~1
    add.err <- 1
  })
  
  model({
    emax <- (temax *exp(eta.emax))
    ed50 <- (ted50)*exp(eta.ed50)
    e0 <- (te0 * exp(eta.e0))
    eff =   e0 + (((emax)* DOSE)* Frel/((ed50)+(DOSE*Frel)))
    eff ~ add(add.err)
  })
}

img.docx

[mattfidler]

I would suggest changing the independent variable DOSE to time in your dataset and in your model. Perhaps in the future you can declare a different idependent variable in nlmixr but right now it isn't possible.

[mattfidler]

There is some discussion on a similar issue in #517

[hrishikesh1985]

I am not able to run below code using nlmixr2 package.

Data <- function() {
ini({
temax <- -9.91
ted50 <- 11.9
eta.emax ~ pdDiag(2.9)
add.err <-12.7
})

model({
emax <- (temax + eta.emax)
ed50 <- (ted50)
eff = (((emax)* time)/((ed50)+time))
eff ~ add(add.err)
})
}

Data_fit <- nlmixr(Data, data_import,est = "focei", foceiControl(derivMethod = "central" ))

Error i got: lotri syntax error:

:001: temax <- -9.91
:002: ted50 <- 11.9
lotri error:
bad matrix expression: 'eta.emax ~ pdDiag(2.9)'
matrix expression should be 'name ~ c(lower-tri)'
:003: eta.emax ~ pdDiag(2.9)
:004: add.err <- 12.7

Error: lotri syntax errors above