1 reply to this topic

[starsnt@gmail.com]

Hello,

 

I am trying to fit TMDD model for a mAb with nonlinear PK and I used the sample TMDD model code provided by the PML school.

I only have the total ligand and total receptor data. The sample full TMDD model code fit the total ligand data very nicely. At the lower dose level, it can capture the bend down followed by a bend back up.

Following are my questions,

1. Upon my research, the entities in the full TMDD code are free ligand and free receptor. Why did model code still fit my total ligand data when it is intended for free ligand? After I added total receptor data to the model, the model is not able to fit the receptor data (because the code is expecting the free receptor data?).
2. With the available data being total ligand and total receptor, how do I modify the TMDD code to recognize and fit the total ligand data and total receptor data, and to predict free ligand ©, free receptor ® and RC?

 

test(){

    deriv(A1 = ((A1*ke- A1*k12+A2*k21) -(kon*C*R -koff*LR)*V))

    deriv(A2 = (A1 * k12- A2 * k21))

    deriv(R = kin -kout*R -kon*C*R + koff*LR)

    deriv(LR = kon*C*R - koff*LR - keRL*LR)

    C = A1/V

    C2 = A2/Vt

    kin = R0*kout 

    dosepoint(A1)

    sequence{R = R0}

    error(CEps = 1)

    observe(CObs = C + C*CEps)

    error(REps = 1)

    observe(RObs = R + R*REps)

    fixef(V = c(, 0.04, ), Vt = c(, 0.04, ), ke = c(, 0.07, ), k12 = c(, 0.5, ), k21 = c(, 0.6, ), R0 =  c(, 12, ))

    fixef(kon = c(, 0.02, ), koff = c(, 0.02, ), kout = c(, 0.15, ), keRL = c(, 0.05, ))

}

 

Any help is greatly appreciated.

 

 

 

[frrd@lundbeck.com]

Hello,

 

I am trying to fit TMDD model for a mAb with nonlinear PK and I used the sample TMDD model code provided by the PML school.

I only have the total ligand and total receptor data. The sample full TMDD model code fit the total ligand data very nicely. At the lower dose level, it can capture the bend down followed by a bend back up.

Following are my questions,

1. Upon my research, the entities in the full TMDD code are free ligand and free receptor. Why did model code still fit my total ligand data when it is intended for free ligand? After I added total receptor data to the model, the model is not able to fit the receptor data (because the code is expecting the free receptor data?).
2. With the available data being total ligand and total receptor, how do I modify the TMDD code to recognize and fit the total ligand data and total receptor data, and to predict free ligand ©, free receptor ® and RC?

 

test(){

    deriv(A1 = ((A1*ke- A1*k12+A2*k21) -(kon*C*R -koff*LR)*V)) #you might want to try to add -keRL* LR * PR (Percent recycled)  to this term to allow for recycling of antibody without the receptor

    deriv(A2 = (A1 * k12- A2 * k21))

    deriv(R = kin -kout*R -kon*C*R + koff*LR)

    deriv(LR = kon*C*R - koff*LR - keRL*LR) #you might want to add -ke*LR to this term to reflect that the complex bound antibody is also broken down via the normal antibody clearance route

    C = A1/V

    C2 = A2/Vt

    Ct=C+LR  #total mab

    Rt=R+LR   # total receptor

    kin = R0*kout 

    dosepoint(A1)

    sequence{R = R0}

    error(CEps = 1)

    observe(CObs = Ct + C*CEps) #changed from C to CT

    error(REps = 1)

    observe(RObs = Rt + R*REps) # Changed from R to Rt

    fixef(V = c(, 0.04, ), Vt = c(, 0.04, ), ke = c(, 0.07, ), k12 = c(, 0.5, ), k21 = c(, 0.6, ), R0 =  c(, 12, ))

    fixef(kon = c(, 0.02, ), koff = c(, 0.02, ), kout = c(, 0.15, ), keRL = c(, 0.05, ))

}

 

Any help is greatly appreciated.

I have added some changes to fix your problem and some suggestions to allow some of the antibody to be recycled from the complex.

Edited by frrd@lundbeck.com, 10 September 2021 - 12:43 PM.