Bioequivalence and Bioavailability Forum

Dear Ravi!

❝ We have profiles from 1000 mg single dose of a drug and want to predict 500mg profiles of the same drug given as Bid. at an interval of 12 hrs using data from 1000mg single dose profile.

❝

❝ Now my question is Can we predict 500 mg bid. profile using non parametric superposition option of the winNonlin 5.2 version.

Yes, but be aware of the assumptions and limitations (see the User’s Guide or the online help > The WinNonlin Toolbox > Nonparametric superposition > Data and assumptions):

Nonparametric Superposition assumes that each dose of a drug acts indepen­dently of every other dose; that the rate and extent of absorption and average systemic clearance are the same for each dosing interval; and that linear phar­macokinetics apply, so that a change in dose during the multiple dosing regi­men can be accommodated.
In order to predict the drug concentration resulting from multiple doses, one must have a complete characterization of the concentration-time profile after a single dose. That is, it is necessary to know C(t_i) at sufficient time points t_i, (i=1,2,…,n), to characterize the drug absorption and elimination process. Two assumptions about the data are required: independence of each dose effect, and linearity of the underlying pharmacokinetics. The former assumes that the effect of each dose can be separated from the effects of other doses. The latter, linear pharmacokinetics, assumes that changes in drug concentration will vary linearly with dose amount.
The required input data are the time, dosing, and drug concentration. The drug concentration at any particular time during multiple dosing is then predicted by simply adding the concentration values as shown below.

❝ If yes Please tell the procedure

In WinNonlin open Exp1.pwo in the program’s Examples folder (profile of one subject).
Run NCA first to check the terminal phase (you will need it later on).

With the Exp1.pwo workbook active, create a new column (Tools > Transform > Transform: x*n > Column x: Conc (ng/ml) > n: 0.5 > Destination: NewColumn > OK).
Tools > Nonparametric Superposition... > drag “Time” to “Time for first dose” and “NewColumn” to “Conc. for first dose” > (increase the default “Number of output data points” from 100 to 500 in order to catch C_max/C_min better) > Method for computations: Log > Define terminal phase 12 to 24 (see your results from the NCA method) > change to the “Variable Dosing tab” > first row: Time 0, Dose 500 > ☑ Repeat every 12 time units > Output time range 0 to 72 > Calculate…