Re: RE: Simulation settgin in the precence of Shrinkage in PK when doing PK-PD analysis

On 2/18/2013 3:23 PM, Nick Holford wrote: > Jakob, Leonid, > > I think it should be pointed out more clearly that PKPD analyses can > be performed most simply using the PPP&D method. This method has > better properties compared with IPP and is easier to set up and run > (see Zhang et al Part I). Sequential methods are preferred over > simultaneous methods whenever there is a possibility of > mis-specification of the model linking concentration to effect. This > is nearly always a real risk so the sequential method is rarely a > sensible choice (see Zhang et al part II). > > The IPP+SE method has properties similar to PPP&D but it technically > more challenging to use and cannot be used without a > variance-covariance matrix of the estimates. This can usually only be > obtained with oversimplified models (asymptotic $COV) or by time > consuming bootstraps. > > The bottom line is to use PPP&D. > > Leonid makes a good point about using VPC for model evaluation. The > VPC is capable of detecting structural model misspecification for > PKPD analyses but it is not foolproof. A misspecified random effects > model can compensate for a misspecified structural model. See > references below for an example. > > Best wishes, > > Nick > > Karlsson MO, Holford NHG 2008. A Tutorial on Visual Predictive > Checks. PAGE 17 (2008) Abstr 1434 [wwwpage-meetingorg/?abstract=1434] > (last accessed 11 February 2012). > > http://holford.fmhs.auckland.ac.nz/docs/vpc-tutorial-and-datatop.pdf > > On 19/02/2013 7:55 a.m., Leonid Gibiansky wrote: > > > Hi Matts, > > > > One method to investigate the problem would be to conduct VPC. If > > VPC with model-estimated variances provides good (not inflated) > > range of PK profiles then one can argue that the PK model provides > > good description of the data and can be used for simulations > > (including PK-PD). > > > > Another test could be to do VPCs for the PK-PD model: one with > > fixed PK parameters (as was used in the sequential PK-PD modeling > > procedure) and the other one with model-simulated ETAs for both PK > > and PD parts. Again, if both provide good coverage of observed > > PK-PD data then combination of PK and PD models can be trusted, and > > any of the approaches can be applied. If one of the VPCs is > > inadequate, than it should be noticeable in the too narrow or too > > wide prediction intervals. > > > > Leonid > > On 19/02/2013 5:20 a.m., Ribbing, Jakob wrote: > > > Resending, since my posting from this morning (below) has not yet > > appeared on nmusers. > > > > Apologies for any duplicate postings! > > > > *From:*Ribbing, Jakob *Sent:* 18 February 2013 09:59 *To:* > > "Kågedal, Matts"; [email protected] *Cc:* Ribbing, Jakob > > *Subject:* RE: Simulation settgin in the precence of Shrinkage in > > PK when doing PK-PD analysis > > > > Hi Matts, > > > > I think you are correct; the problem you describe has not had much > > (public) discussion. > > > > It is also correct like you say that this is mostly a problem in > > case of all of the below > > > > ·sequential PK-PD analysis is applied (IPP approach, Zang et al) > > > > ·non-ignorable degree of shrinkage in PK parameters _of relevance_ > > > > oOf relevance: the PK parameters effectively driving PD for the > > mechanism, e.g. CL/F if AUC is driving. In addition, if PD response > > develops over several weeks/months then shrinkage in IOV may be > > ignored even for relevant PK parameters > > > > ·I would also like to add that for this to be an issue individual > > PK parameters must explain a fair degree of the variability in PD, > > which is not always the case > > > > oIf driving PD with typical PK parameters (along with dose and > > other PD covariates) does not increase PD omegas, compared to IPP, > > then either PK shrinkage is already massive, or else it is not an > > issue for the IPP-PD model > > > > If only the IPP approach is possible/practical a simplistic > > approach to simulate PD data is as follows: > > > > ·sample (with replacement) the individual PK parameters along with > > any potential covariates (maintaining correlation between IIP and > > covariates, i.e. whole subject vectors for these entities, but > > generally not for dose since generally should only have only random > > association with IPP or PK/PD-covariates) > > > > ·then use the re-sampled datasets for simulating PD according to > > the PD model (driven by IPP, covariates, dose, etc). The degree of > > shrinkage is then the same for PD estimation and simulation. > > > > This approach may for example allow to simulate realistic PD > > response at multiple dosing, based on only single dose PD. When the > > MD data becomes available then one may find that variabilities > > shift between PK and PD due to different PK shrinkage, but I would > > argue the simulated PD responses still were realistic. This > > approach is useful for predictions into the same population > > (especially if sufficient number of subjects available for > > re-sampling), but may not allow extrapolation into other > > populations where PK is projected to be different. > > > > When possible the obvious solution is to apply one of the > > alternative approaches to simultaneous PK-PD fit; after you have > > arrived at a final-IPP model. > > > > If a simultaneous fit is obtainable/practical this is the best > > option, but notice that e.g. if you have rich PK data in healthy > > and no PK data in patients (plus PD data in both populations): You > > can estimate separate omegas for PD parameters in healthy vs. > > patients, but it may be difficult to tell whether patients higher > > PD variability is due to PK shrinkage, or due to the actual PD > > variability being higher in this population (or both). PD > > variability may be confounded by a number of other factors that are > > actually variability in PK (fu, active metabolites and bio phase > > distribution, just to mention a few where information may be absent > > on the individual level). Depending on the purpose of the modelling > > this often not an issue, however. > > > > As you suggest there may be rare situations with IPP where a more > > complicated approach is needed, with a) simulation and > > re-estimation of PK model, to obtain Empirical-Bayes Estimates > > based on simulated data, and then feed these into the subsequent PD > > model. I would see this as a last resort. There are pitfalls in > > that if PD parameters have been estimated under one degree of PK > > shrinkage, then applying these estimates to a simulated example > > with different PK shrinkage requires adjustment of PD variability. > > I am not sure anyone has had to go down that route before and if > > not I hope you do not have to either. Maybe others can advice on > > this? > > > > Best regards > > > > Jakob > > > > Two methodological references: > > > > Simultaneous vs. sequential analysis for population PK/PD data II: > > robustness of methods. > > > > Zhang L, Beal SL, Sheiner LB. > > > > J Pharmacokinet Pharmacodyn. 2003 Dec;30(6):405-16. > > > > Simultaneous vs. sequential analysis for population PK/PD data I: > > best-case performance. > > > > Zhang L, Beal SL, Sheiner LB. > > > > J Pharmacokinet Pharmacodyn. 2003 Dec;30(6):387-404. e