Hi All, I am using NLME for the first time. I would like to developing a model to quantify the influence of phytochemical treatment on PgP and CYP probe substrates in healthy volunteers. Can the Concentrations of substrates before and after treatment with phytochemical be placed into the same input file mentioning without/with treatment(0/1) and added as the categorical variable during covariate analysis. And can any body, if had develop the model, share the control stream for modeling drug-drug interactions. Thanks VIJAY
Dear Vijay What can you share with respect to your data set? NLME can do all that but the main issue here is the mechanism of action we need to define and a good understanding of the problem (I am not sure it is 100% clear to me what we have here) Let me give you some ideas about how to proceed. 1: As a first step you can put the response columns PgP and CYP probe substrates in 2 different columns if you want to do a simultaneous analysis of the 2 substrates or do 2 separate analyses (do you have 2 probe substrates?). You add a covariate you call treatment and is 1 if treatment and 0 if not. This column treatment will be a covariate you are going to investigate if correlated to one of the PK parameters of your model like Cl or V. 2: You start with the base model where you do not take into consideration the treatment effect and you get a -2LL value. You start with 1 compartment and then add another one in a second base model. If -2LL decreases by more than 10 then 2 compartment is the base model to sue for the subsequent covariate analysis 3: You add the covariate and use the covariate search option which will allow you to find the best covariate model. You can put as optional PK parameters linked to treatment Ka, V and Cl. You then sue the automatic covariate search to find the best covariate model. Suppose it is Cl for both substrates (may be you should do a separate analysis for the 2 substrates ) 4: You conclude that Clearance is affected by treatment for both substrates 5: Now if you have the dynamics of the treatment (can you measure with time these compounds in plants? ), then there is a way to dynamically correlate the treatment to the PK dynamics of the substrates. I will be out for one week now at ACOP but if you can share something from the data, I will be happy to assist you when I am in a break at ACOP. I can help you with Pharmacometric aspects of the analysis but not with the biology/mechanism of action. Best Regards; Serge
Dear Serg, I am extremely sorry to make you confuse. This study investigates herb-drug interactions with probe substrate. Probe substrate ( and it’s metabolite for CYP substrate) concentration data from healthy subjects who received a single dose of probe substrate with or without phytochemical pretreatment (for10 days) are to be analyzed using a population pharmacokinetic modeling approach. Each phytochemical was tested for a Pgp and/or CYP substrates. Hope I made it clear this time. Is -2LL similar to OFV(Objective Function Value) in NONMEM? Thanks a lot for your future support. Best Regards, VIJAY
Dear vijay I am on my way to florida. Please send me the project where you delibarately hiding the response data and I will prepare for youba template you can use. The obj in nlme is different than nonmem but the diff between 2 models will be the same if they both optimize the same way. Best Serge