higher AUC after bolus compared to infusion?

Dear all, I have trouble modeling some rat PK data obtained after intravenous dosing. I have dense data for two different doses, given either as bolus, half-hour or 3-hour infusion. We know that the compound has a large binding affinity to intracellular structures and therefore has a high volume of distribution (~100L/kg in rats). However, what we observed in the mentioned study is that the AUC after bolus dosing is ~3 times higher than after infusions (no difference between 0.5 and 3 h), and this was observed for both dose groups (effect a bit less pronounced for high dose, but still there). Noncompartmental analysis indicates that this is due to lower clearance and not a change in Vss. At the moment I have no idea what kind of model would capture an observation like this. Has anybody ever observed something like this, and even better, have some ideas on model coding? Any ideas would be welcome! Best regards Nele _________________________________________ Bayer Schering Pharma AG Development Pharmacokinetics Berlin, S109, 03, 306A Phone: +49 30 468-15146 Fax: +49 30 468-11527 E-mail: nele.plock Web: http://www.bayerscheringpharma.de Vorstand: Andreas Fibig, Vorsitzender | Werner Baumann, Andreas Busch, Ulrich Kstlin, Kemal Malik, Bernd Metzner, Gunnar Riemann Vorsitzender des Aufsichtsrats: Werner Wenning Sitz der Gesellschaft: Berlin | Eintragung: Amtsgericht Charlottenburg HRB 283 B

Dear all, I have trouble modeling some rat PK data obtained after intravenous dosing. I have dense data for two different doses, given either as bolus, half-hour or 3-hour infusion. We know that the compound has a large binding affinity to intracellular structures and therefore has a high volume of distribution (~100L/kg in rats). However, what we observed in the mentioned study is that the AUC after bolus dosing is ~3 times higher than after infusions (no difference between 0.5 and 3 h), and this was observed for both dose groups (effect a bit less pronounced for high dose, but still there). Noncompartmental analysis indicates that this is due to lower clearance and not a change in Vss. At the moment I have no idea what kind of model would capture an observation like this. Has anybody ever observed something like this, and even better, have some ideas on model coding? Any ideas would be welcome! Best regards Nele _________________________________________ Bayer Schering Pharma AG Development Pharmacokinetics Berlin, S109, 03, 306A Phone: +49 30 468-15146 Fax: +49 30 468-11527 E-mail: [EMAIL PROTECTED] Web: http://www.bayerscheringpharma.de Vorstand: Andreas Fibig, Vorsitzender | Werner Baumann, Andreas Busch, Ulrich Köstlin, Kemal Malik, Bernd Metzner, Gunnar Riemann Vorsitzender des Aufsichtsrats: Werner Wenning Sitz der Gesellschaft: Berlin | Eintragung: Amtsgericht Charlottenburg HRB 283 B

Volume and AUC are independent - AUC is defined by dose and clearance. You have stated that your estimate of Vss is not impacted so your experimental design is probably capturing enough of the PK profile to describe the underlying phenomenon. I think the best starting point is a michaelis-menten model with diminished clearance at high plasma concentrations. Regards, Jeff ______________________________________________________ Jeff Wald, PhD jeffrey.a.wald 'at' gsk.com Director, Clinical Pharmacology Modeling and Simulation RTP, NC P Please consider the ENVIRONMENT before printing this mail P nele.plock Sent by: owner-nmusers 04-Dec-2008 04:02 To nmusers cc Subject [NMusers] higher AUC after bolus compared to infusion? Dear all, I have trouble modeling some rat PK data obtained after intravenous dosing. I have dense data for two different doses, given either as bolus, half-hour or 3-hour infusion. We know that the compound has a large binding affinity to intracellular structures and therefore has a high volume of distribution (~100L/kg in rats). However, what we observed in the mentioned study is that the AUC after bolus dosing is ~3 times higher than after infusions (no difference between 0.5 and 3 h), and this was observed for both dose groups (effect a bit less pronounced for high dose, but still there). Noncompartmental analysis indicates that this is due to lower clearance and not a change in Vss. At the moment I have no idea what kind of model would capture an observation like this. Has anybody ever observed something like this, and even better, have some ideas on model coding? Any ideas would be welcome! Best regards Nele _________________________________________ Bayer Schering Pharma AG Development Pharmacokinetics Berlin, S109, 03, 306A Phone: +49 30 468-15146 Fax: +49 30 468-11527 E-mail: nele.plock Web: http://www.bayerscheringpharma.de Vorstand: Andreas Fibig, Vorsitzender | Werner Baumann, Andreas Busch, Ulrich Kstlin, Kemal Malik, Bernd Metzner, Gunnar Riemann Vorsitzender des Aufsichtsrats: Werner Wenning Sitz der Gesellschaft: Berlin | Eintragung: Amtsgericht Charlottenburg HRB 283 B

Nele It is possible to simulate AUC_iv_bolus = 3*AUC_3h_inf This was done using the code below, assuming a saturation in both elimination and distribution in the peripheral compartment (similar to the pharmacokinetics of taxol) The problem is that the AUC_0.5h_inf is intermediate and closer to the bolus AUC than to the 3h_inf AUC value A model with auto-induction could also increase the AUC_iv_bolus with an intermediate value for the 0.5h infusion AUC Hoping that this helps Regards Saik $MODEL COMP=(CP,DEFOBS,DEFDOSE) COMP=(PERI) COMP=(AUC) $PK VM=THETA(1)*EXP(ETA(1)) KM=THETA(2)*EXP(ETA(2)) V1=THETA(3)*EXP(ETA(3)) VM12=THETA(4) KM12=THETA(5) K21=THETA(6) $DES DADT(1)=-VM*A(1)/(A(1)+V1*KM)+K21*A(2)-VM12*A(1)/(A(1)+V1*KM12) DADT(2)=-K21*A(2)+VM12*A(1)/(A(1)+V1*KM12) DADT(3)= A(1) $ERROR Y=A(1)/V1*EXP(EPS(1)) AUC = A(3) $THETA (0,20) ; VM $THETA (0,1) ; KM $THETA (0,1) ; V1 $THETA (0,5) ; VM12 $THETA (0,.5) ; KM12 $THETA (0,.1);k21

Dear, This can most likely be described by a model of non-instantaneous and potentially irreversible binding kinetics to the intracellular proteins/structures. With bolus, you see the drug in plasma (hence higher AUC) before it binds to the internal proteins irreversably. With infusions, more of the drug has time to bind irreversably to the internal structures, which is a tissue clearance (higher apparent clearance with infusions leading to lower AUC). At higher doses, you may saturate the proteins/internal structures and you no longer have that tissue clearance. And after a certain dose level, you will likely see more linear kinetics. I've done some research on that some time ago (modeled the non-instantaneous kinetics and measured them experimentally, then simulated a clinical profile where these non-instantaneous kinetics of protein binding played a role in the kinetics of the drug), a small part of the work was published but the remainder has not been published yet (I'm about 7 years late on writing that paper). Malaz A AbuTarif, Ph.D., M.B.A. Schering-Plough Kenilworth, NJ

Volume and AUC are independent - AUC is defined by dose and clearance. You have stated that your estimate of Vss is not impacted so your experimental design is probably capturing enough of the PK profile to describe the underlying phenomenon. I think the best starting point is a michaelis-menten model with diminished clearance at high plasma concentrations. Regards, Jeff ______________________________________________________ Jeff Wald, PhD jeffrey.a.wald 'at' gsk.com Director, Clinical Pharmacology Modeling and Simulation RTP, NC P Please consider the ENVIRONMENT before printing this mail P [EMAIL PROTECTED] Sent by: [EMAIL PROTECTED] 04-Dec-2008 04:02 To [email protected], [EMAIL PROTECTED] cc Subject [NMusers] higher AUC after bolus compared to infusion? Dear all, I have trouble modeling some rat PK data obtained after intravenous dosing. I have dense data for two different doses, given either as bolus, half-hour or 3-hour infusion. We know that the compound has a large binding affinity to intracellular structures and therefore has a high volume of distribution (~100L/kg in rats). However, what we observed in the mentioned study is that the AUC after bolus dosing is ~3 times higher than after infusions (no difference between 0.5 and 3 h), and this was observed for both dose groups (effect a bit less pronounced for high dose, but still there). Noncompartmental analysis indicates that this is due to lower clearance and not a change in Vss. At the moment I have no idea what kind of model would capture an observation like this. Has anybody ever observed something like this, and even better, have some ideas on model coding? Any ideas would be welcome! Best regards Nele _________________________________________ Bayer Schering Pharma AG Development Pharmacokinetics Berlin, S109, 03, 306A Phone: +49 30 468-15146 Fax: +49 30 468-11527 E-mail: [EMAIL PROTECTED] Web: http://www.bayerscheringpharma.de Vorstand: Andreas Fibig, Vorsitzender | Werner Baumann, Andreas Busch, Ulrich Köstlin, Kemal Malik, Bernd Metzner, Gunnar Riemann Vorsitzender des Aufsichtsrats: Werner Wenning Sitz der Gesellschaft: Berlin | Eintragung: Amtsgericht Charlottenburg HRB 283 B