# PharmPK Discussion - NONMEM model help

PharmPK Discussion List Archive Index page
• On 10 Nov 2005 at 08:40:56, "Lipinski, Scott [JJCUS NON J&J]" (SLipinsk.at.CORUS.JNJ.com) sent the message
`The following message was posted to: PharmPKhello allI am using NONMEM to model the clearance of a drug in prematureneonates(first time using this) . Currently my line of program is CL=(THETA(2)*VANC)+THETA(1)*(GA+PNA/7)+ETA(1)where GA - gestational age in weeks  and PNA - post natal age in days .Is there a better way of modeling the effect of the age on clearance.On looking at this i think my equation is suggesting that asgestational age increases clearance increases linearly however i guessclearance could also increase exponentially or in a sigmoidal shape.How can i get NONMEN to model in these ways ??On a second point im told that clearance is usually proportional toweight to the power of 0.7 and that this is "a well know fact" Doesanyone know why clearance is related to weight in this way  or have anyreferences to support this ???Padraig Galbraith`
Back to the Top

• On 10 Nov 2005 at 15:33:48, "RIGBY-JONES Ann, Research Fellow" (ann.rigby-jones.-a-.phnt.swest.nhs.uk) sent the message
`The following message was posted to: PharmPKHiIf you go to the NONMEM users archivehttp://www.phor.com/nonmem/nm/and search for "allometric scaling" you'll find plenty of usefulthreads with  references explaining the biological reasoning behind therelationship between  clearance and body weight to the power 0.75P.S.  Have you tried an exponential ETA e.g CL=THETA(n)*EXP(ETA(n))rather than an additive one?  Often works well.Good luckAnnDr. Ann Rigby-JonesResearch Fellow, Anaesthesia Research Group, Peninsula Medical School`
Back to the Top

• On 10 Nov 2005 at 16:38:37, "MacDonald, Alexander" (alexander.macdonald.at.roche.com) sent the message
`The following message was posted to: PharmPKPadraig,Nick Holfords 1996 paper is good place to start for the clearance/bodyweight relationship argument.1. Holford, N. H. G. A size standard for pharmacokinetics. ClinicalPharmacokinetics 30, 329-332 (1996).Hope this helps.CheersAlexAlex MacDonald Ph.DProject LeaderDrug Metabolism and PharmacokineticsF.Hoffmann-La Roche LtdPharmaceuticals DivisionBldg. 070/132CH-4070 BaselSwitzerlandTel. +41 (0) 61 688 40 98Fax +41 (0) 61 688 20 66alexander.macdonald.-at-.roche.com`
Back to the Top

• On 11 Nov 2005 at 04:46:46, Nick Holford (n.holford.-a-.auckland.ac.nz) sent the message
`The following message was posted to: PharmPKPadraig,Some background about modelling PK in neonates that incorporate weightand age can be found here:Bouwmeester NJ, Anderson BJ, Tibboel D, Holford NH. Developmentalpharmacokinetics of morphine and its metabolites in neonates, infantsand young children. Br J Anaesth 2004;92(2):208-17Anderson BJ, van Lingen RA, Hansen TG, Lin YC, Holford NHG.Acetaminophen developmental pharmacokinetics in premature neonates andinfants: a pooled population analysis. Anesthesiology 2002;96(6):1336-45The allometric exponent used in these papers is 0.75. This fact ispredicted by theory and confirmed by experiment in many biologicalsettings.Nick`
Back to the Top

• On 10 Nov 2005 at 10:52:28, "Bhattaram, Atul" (BhattaramA.at.cder.fda.gov) sent the message
`The following message was posted to: PharmPKHello Padraig GalbraithThe following publication will be a very useful reference in your NONMEManalysis.Developmental pharmacokinetics of morphine and its metabolites inneonates,infants and young children.Br J Anaesth. 2004 Feb;92(2):208-17.Bouwmeester NJ, Anderson BJ, Tibboel D, Holford NH.Also you may want to look at the pharmacometrics review and the printedlabel for Betapace (Sotalol) on deriving dosing regimens in pediatrics.Betapace (Sotalol Hydrochloride). Center for Drug Evaluation andResearch, United States Food and Drug Administration. Available fromhttp://www.fda.gov/cder/foi/nda/2001/19-865s10_Betapace.htm.Venkatesh Atul BhattaramPharmacometricsUS Food and Drug Administration"The contents of this message are mine personally and do not necessarilyreflect any position of the Government or the Food and DrugAdministration."`
Back to the Top

• On 10 Nov 2005 at 18:04:12, andreanicole.edginton.-at-.bayertechnology.com sent the message
`Hello,While I am not a NONMEM modeller, I can answer your final comment re:clearance in children.When we discuss clearance in children, there are two considerationsthat account for their altered clearance; namely maturity andphysiology.  There are also two camps of thinking about clearance andchildren; namely allometric scaling (as you mentioned) and mechanisticscaling.  By scaling I mean from adults to children.  There are manydifferences between these two methods and both have strengths andweaknesses.Allometric scaling in children is commonly done based on body weight. You are correct that the exponent 0.7 is used for allometric scaling,but this value is commonly used for scaling clearance between species,not within a species.  In the case of intraspecies scaling for humans,the exponent 0.75 is used (for the procedure see Anderson & Meakin. 2002.  Scaling for size: some implications for paediatric anaesthesiadosing.  Paediatic Anaesthesia 12: 205-219).  This procedure is basedon the differences in body weight between children and adults. Theseauthors have found it to be useful (and easy) for scaling clearanceamong different ages of children. However, allometric scaling is onlybased on size and says nothing about the maturity of the processesinvolved in the clearance of the compound.  This is understood, whichis why the authors, when scaling doses based on clearanceconsiderations, had to further reduce the dose by half for immaturityconsiderations (see Table 2).  For your particular case (prematureneonates), scaling clearance using allometric scaling does not tell youmuch since it does not take into account maturity of the processes ofclearance.The other method of scaling clearance is the mechanistic method.  Thisconsiders each mechanism involved in the clearance process of the drugin question and individually scales each process separately.  It iswell known that different enzymes and/or the processes of glomerularfiltration and tubular secretion mature at a different rate.  Some arefully mature at birth thus making the ratio of child to adult weightnormalized clearances based only on physiological differences betweenthem (e.g. liver weight to body weight ratio, consideration ofeliminating organ blood flow...). When the process matures slowly (e.g.CYP1A2 metabolism), the ratio of weight-normalized clearances is due toboth maturity and physiological considerations.  There are some verygood papers available to explain this method (Alcorn & McNamara. Ontogeny of hepatic and renal systemic clearance pathways in infants:part I and part II.  Clinical Pharmacokinetics.  2002.  41 (12):959-998 and 41 (13): 1077-1094) and (Bjorkman.  Prediction of drugdisposition in infants and children by means of physiologically basedpharmacokinetic (PBPK) modelling: theophylline and midazolam as modeldrugs.  2004.  59(6): 691-704.).  Our group has also submitted a paperfor peer-review on this topic; working from the earlier work of theauthors stated above.  The disadvantage of this method is that it ismore difficult than allometric scaling; however, the advantage, Ibelieve, is that it is more relevant.For premature neonates, the situation is further exaggerated.  Theirlevel of immaturity is dependent on both their gestational andpostnatal age (as you have already indicated).  Knowing their size willnot be sufficient enough to explain their clearance.  The mechanism ofclearance of the drug you are interested in and the maturity of thatmechanism in premature neonates (from in vitro liver data if its thecommon hepatic clearance situation) is required. From what I understand, you want to correlate clearance to gestationaland postnatal age and to determine which is the most importantparameter for the determination of clearance.  While this is indeedimportant, the underlying processes, regarding for example, curveshape, will not be understood.  Empirical correlations are sufficientfor retrospective analysis, but any prospective analysis will lack amechanistic basis from which to draw conclusions.Of course, when we move from the arena of clearance determination todosing determination in children, which is the primary use ofdetermining clearance in children, there occurs another importantparameter; volume of distribution.  Only with accurate clearance anddistribution volume prediction can dosing be consistently and reliablydetermined for children; especially in your case with prematureneonates.  To do this we used the age-specific PBPK model and ourclearance scaling module that is within the software, PK-Sim.  I have aposter from a conference I attended on the clearance scaling method ifyou would like me to send it to you.Take care,Andrea Edginton  Bayer Technology Services GmbH  Process Technology, BiophysicsLeverkusen, Germany`
Back to the Top

• On 11 Nov 2005 at 10:13:41, "Rob ter Heine" (Aprth.at.slz.nl) sent the message
`The following message was posted to: PharmPKDear Padraig, > On a second point im told that clearance is usually proportional to > weight to the power of 0.7 and that this is "a well know fact" Does > anyone know why clearance is related to weight in this way  orhave any > references to support this ???What you are talking about here is allometric scaling.There is enough literature on this subject.West GB, Brown JH, Enquist BJ. Science 1997 A general model for theorigin of allometric scaling laws in biology.West GB, Brown JH, Enquist BJ. Science 1999 The fourth dimension oflife: fractal geometry and allometric scaling of organisms.Holford NHG. Clinical Pharmacokinetics 1996 A size standard forpharmacokinetics.A good allometric model for clearance indeed  is CL= CLstd x (Wt/Wtstd)**(3/4)    -----> (** means to the power). I usually use astandard weight of 70 kilograms, but of course this depends on yourpopulation. >looking at this i think my equation is suggesting that as >gestational age increases clearance increases linearly however i guess >clearance could also increase exponentially or in a sigmoidal shape. >How can i get NONMEN to model in these ways ??By expressing the equations differently (with a power or a formulawhich expresses a sigmoidal curve).An good source for all your NONMEM questions is the NONMEM usersnetarchive: http://www.cognigencorp.com/nonmem/nm/Yours sincerely,Rob ter Heine--Rob ter Heine, MSc, PharmDDepartment of Pharmacology, Slotervaart HospitalAmsterdam, The NetherlandsE: aprth.-at-.slz.nlT: 020-5124737`
Back to the Top

Want to post a follow-up message on this topic? If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "NONMEM model help" as the subject

Copyright 1995-2010 David W. A. Bourne (david@boomer.org)