1 reply to this topic

[niujing23@gmail.com]

Dear Certara Forum users,

I am modeling the PK data of a long-acting formulation. But I can't fit my absorption phase properly. A zero order+tlag absorption model could capture the overall shape, but will ignore the gradual increase at the initial time (when time <=tlag), a transit compartment model could capture the gradual increase at the beginning, but can't capture the shape of the peak. blow are my questions:

1. is that OK I use zero order+transit compartment model? I haven't read a paper with this model yet

2. if so, how can I code it in PML? I have attached my code, which could run successful but the parameter estimates are not plausible

3  meanwhile, IPRED+DV vs time with k0+tlag and ka+transit compartment model was provided as attachments, any other absorption model could I try?

Thanks,

Jing

 

test(){

transit(Aa, mtt, ntr, max = 50, out = -Aa * Ktr )

dosepoint(Aa)

deriv(A0 = (Aa * Ktr)- K0)

#deriv(Aa = - (Aa * Ktr))

deriv(A1 = K0- (CL * C))

#urinecpt(A0 = (CL * C2))

mtt = (ntr+1)/Ktr

#C = A1 / V

C = A1 / V

K0=Aa/InfDur

error(CEps = 0.00625961)

observe(CObs = C + CEps * sqrt(1 + ©^2 * (CMultStdev/sigma())^2))

stparm(V = tvV * exp(nV))

stparm(Ktr = tvKtr * exp(nKtr))

stparm(ntr = tvntr * exp(nntr))

#stparm(V2 = tvV2 * exp(nV2))

stparm(InfDur = tvInfDur * exp(nInfDur))

stparm(CL = tvCL * exp(nCL))

stparm(CMultStdev = tvCMultStdev)

fixef(tvV = c(0, 40, ))

fixef(tvKtr = c(0, 0.009, ))

fixef(tvntr = c(0, 7, ))

#fixef(tvV2 = c(, 1, ))

fixef(tvInfDur = c(0, 900, ))

fixef(tvCL = c(0, 0.28, ))

fixef(tvCMultStdev = c(, 0.0542587, ))

ranef(diag(nV, nCL, nKtr, nInfDur,nntr) = c(0.1, 0.1, 0.1, 0.1,0.1))

}

[serge guzy]

 

Dear Certara Forum users,

I am modeling the PK data of a long-acting formulation. But I can't fit my absorption phase properly. A zero order+tlag absorption model could capture the overall shape, but will ignore the gradual increase at the initial time (when time <=tlag), a transit compartment model could capture the gradual increase at the beginning, but can't capture the shape of the peak. blow are my questions:

1. is that OK I use zero order+transit compartment model? I haven't read a paper with this model yet

2. if so, how can I code it in PML? I have attached my code, which could run successful but the parameter estimates are not plausible

3  meanwhile, IPRED+DV vs time with k0+tlag and ka+transit compartment model was provided as attachments, any other absorption model could I try?

Thanks,

Jing

 

test(){

transit(Aa, mtt, ntr, max = 50, out = -Aa * Ktr )

dosepoint(Aa)

deriv(A0 = (Aa * Ktr)- K0)

#deriv(Aa = - (Aa * Ktr))

deriv(A1 = K0- (CL * C))

#urinecpt(A0 = (CL * C2))

mtt = (ntr+1)/Ktr

#C = A1 / V

C = A1 / V

K0=Aa/InfDur

error(CEps = 0.00625961)

observe(CObs = C + CEps * sqrt(1 + ©^2 * (CMultStdev/sigma())^2))

stparm(V = tvV * exp(nV))

stparm(Ktr = tvKtr * exp(nKtr))

stparm(ntr = tvntr * exp(nntr))

#stparm(V2 = tvV2 * exp(nV2))

stparm(InfDur = tvInfDur * exp(nInfDur))

stparm(CL = tvCL * exp(nCL))

stparm(CMultStdev = tvCMultStdev)

fixef(tvV = c(0, 40, ))

fixef(tvKtr = c(0, 0.009, ))

fixef(tvntr = c(0, 7, ))

#fixef(tvV2 = c(, 1, ))

fixef(tvInfDur = c(0, 900, ))

fixef(tvCL = c(0, 0.28, ))

fixef(tvCMultStdev = c(, 0.0542587, ))

ranef(diag(nV, nCL, nKtr, nInfDur,nntr) = c(0.1, 0.1, 0.1, 0.1,0.1))

}

ka_transit.pngk0_tlag.png

 

Dear colleague

Yes you can and I am attaching a project that will help you. I generated it with the graphical interface and then with the transit statement. Like you have the "out" expression, you have also the "in" expression. The trick is to define the in expression as the righ side of the infusion input differential equation.

 

Let me know if you need further help.

Best Regards

Serge

Attached Files

•  niujing23.phxproj   163.67KB   449 downloads