Bioequivalence and Bioavailability Forum

Hi Darya,

❝ I want to compare the data from WinNonLin with data from RStudio.

❝ ncappc(obsFile=system.file("extdata","concentrationb2.csv", package = "ncappc"),


❝        concUnit = "ng/mL",
❝        timeUnit = "h",
❝        doseUnit = "g",
❝        adminType = "extravascular",
❝        doseType = "ns",
❝        method = "linear",
❝        printOut = FALSE,
❝        psnOut=FALSE,
❝        noPlot=TRUE,
❝        LambdaExclude = c(0,0.33,0.66,1,1.33,1.66,2,2.5,3,3.5,4)
❝       )

I used this code in the R-console

library(ncappc)
ncappc(obsFile="concentrationb2.csv", concUnit="ng/mL",
       timeUnit="h", doseUnit="g", adminType="extravascular",
       doseType="ns", method="linear", printOut=TRUE,
       psnOut=FALSE, noPlot=TRUE,
       LambdaExclude=c(0,0.33,0.66,1,1.33,1.66,2,2.5,3,3.5,4))

and got matching results in Phoenix/WinNonlin 7.0 with its automatic method of selecting the time range for λ_z (i.e., maximizing R²_adj):
                        ncappc    PHX/WNL
AUClast (h*ng/mL)     7645.4250  7645.425
Cmax (ng/mL)          1114.0000  1114
Tmax (h)                 3.0000     3
AUCINF_obs (h*ng/mL)  8315.9994  8315.9994
HL_Lambda_z (h)          8.9386     8.9386
Hence, seemingly the argument LambdaExclude did not work as expected. When I tried LambdaTimeRange=c(5,24) instead, I got:
Error in match(ntime[ntime >= Lambda_z_lower]) :
  argument "table" is missing, with no default

❝ I need calculate elimination constant with time points from 5h to 24h.

Why do you want to do that?
You would go from a good and reasonable fit (distribution/elimination)



to a completely wrong one:

❝ After all RStudio calculate pharmacokinetic parameters BUT the elimination constant is incorrectly calculated.

In this case it is good that ncappc did not observe your attempts. However, according to the documentation it should be possible to specify a time-range. Contact the authors. Mats Karlsson is a nice guy.

BTW, I strongly recommend to abandon the lousy linear trapezoidal method (positively biased AUC for decreasing concentrations). Better to use the argument method="linear-log" in the future.