Predictive ADME: guiding the lead development and optimization process

The prediction of absorption, distribution, metabolism, and excretion (ADME) properties has become increasingly important as failures late in the drug discovery process become more costly. Increasingly, stringent in vitro and in vivo requirements have been placed on the hit-to-lead and lead optimization stages of the drug discovery process. Although it is tempting to dismiss ADME modeling and simply conduct an in vitro or in vivo experiment to get “the correct answer”, this approach is not practical. A skilled, competent medicinal chemist working on a lead optimization program can easily conceive of far more compounds than can reasonably be synthesized during the time of a lead optimization effort. In vivo studies are expensive and time-consuming and may become the rate-limiting step for some projects, particularly for small pharmaceutical companies. Rather than providing “the correct answer”, modeling provides a means of “stacking the deck” in favor of the medicinal chemistry effort, increasing the likelihood that a given compound will show the desired effect in vitro or in vivo.

At our Predictive ADME session chaired by Anthony Klon running October 16 at Bryn Mawr recent developments in the predictive modeling of ADME properties will be presented and discussed. Speakers will present their research into modeling microsomal stability, drug-drug interactions, and membrane transport processes such as blood-brain barrier penetration, intestinal absorption, and skin penetration. One topic of the accompanying discussions will be the appropriateness of relevant biological endpoints for ADME/PK modeling.

The session will be followed in the evening by a Knowledge Café to discuss Collaboration Opportunities in Predictive ADME & Predictive Toxicology.

A description of the session with presentation abstracts follows.  Please add your comments, discussion or questions at the end of the post.

Predictive ADME

http://echeminfo.com/COMTY_confprog08adme

(Please follow continuation here to read abstracts.  Comments can be made at the end.)

Abstracts

Predictions of Metabolic Drug-Drug Interactions

Heidi Einolf, Novartis

Drug-drug interactions (DDI) involving cytochrome P450 (CYP) enzymes remain an important factor in pharmaceutical drug development. Increased understanding of the potential clinical drug interaction magnitude caused by compounds deemed as CYP inhibitors (reversible or irreversible) is imperative to avoid compounds with potential dangerous DDI with likely co-medications and to have a competitive safety profile. In later development, this information is important for the strategic design of clinical DDI trials (i.e. rank-ordering of specific CYP inhibition studies and anticipation of actual DDI risk). Many reported prediction approaches for reversible (and irreversible) CYP inhibition focus on predictions of mean changes of the affected drug (or substrate) exposure at steady-state. These mathematical prediction models, expressed with varying levels of complexity, incorporate the relationship of a single in vivo inhibitor concentration [I] and the potency of the CYP inhibition determined from in vitro data. The simplest of the prediction models, although generally over-predictive for CYP reversible inhibition, is the ‘[I]/Ki’ approach. The pragmatic use of this model, using Cmax, total as [I], is currently the recommended approach by the FDA to evaluate whether a clinical drug interaction study might be warranted. However, models that incorporate the fraction of the affected drug metabolized by the inhibited enzyme (fm,CYP), first-pass intestinal availability for CYP3A substrates, and protein binding, in more mechanistic prediction models (herein termed the ‘Mechanistic-Static Model’ or MSM), have proven to be more predictive of actual DDI magnitude than the ‘[I]/Ki’ approach. Both these approaches are, however, limiting as they only consider mean finite inhibitor concentrations in DDI assessment. More physiologically-based drug interaction prediction models account for the time-varying concentration of inhibitors and are currently implemented in specialized platforms such as the Simcyp Population-Based ADME Simulator (Simcyp Ltd, Sheffield UK), herein termed the ‘Mechanistic-Dynamic Model’ or MDM. These types of models have the capability of being more informative for drug interaction assessments as they predict not only the mean, but also a range and frequency distribution of clearance or drug interaction magnitude in a population. The models implemented within Simcyp consider variables such as CYP expression level and genetic polymorphisms, first-pass intestinal metabolism, physiological, and demographic information in the generation of the virtual populations using a Monte Carlo approach. The program can simulate drug concentrations-time profiles of substrates and inhibitors and, therefore, has the potential to be more predictive than less physiologically-based models. In this talk, a comparison of the three approaches (‘[I]/Ki’, MSM, and MDM) to predict actual clinical DDI magnitude will be presented, including specific data required for best predictions using these different approaches. In addition, the importance of predicting the range and frequency of DDI magnitude using the more physiologically-based drug interaction prediction models will be emphasized.

Novel MI-QSAR Descriptors for Use in Modeling Membrane Transport Processes Such as Skin Penetration Enhancement Anton Hopfinger, Distinguished Research Professor of Pharmacy, University of New Mexico Membrane-interaction (MI) QSAR analysis is a structure-based design methodology combining unique intermolecular descriptors computed from molecular simulations with classic intramolecular QSAR descriptors to model chemically and structurally diverse compounds interacting with cellular membranes. The most recently developed MI-QSAR descriptor, the difference in the integrated cylindrical distribution functions over a phospholipid monolayer model, in and out of the presence of a monolayer penetrator, ÄÓh(r), greatly reduces the size and complexity of the MI-QSAR models as compared to corresponding classic intramolecular QSAR models. By way of example of demonstrating the utility of MI-QSAR analysis, as well as its descriptors including ÄÓh(r), for studying transport-related ADMET endpoints, two skin penetration enhancer data sets of 61 and 42 compounds, respectively, were investigated. These two data sets involve skin penetration enhancement of hydrocortisone and hydrocortisone acetate and the enhancers are generally similar in structure to lipids and surfactants. MI-QSAR models were constructed and compared to QSAR models constructed using only classic intramolecular QSAR descriptors. The MI-QSAR models are quite simple and compact in form as compared to the classic QSAR models. Good penetration enhancers are seen from the MI-QSAR models to make bigger ‘holes’ in the monolayer and are less aqueous soluble, so as to preferentially enter the monolayer, than are poor penetration enhancers. The skin penetration enhancer thus alters the structure and organization of the monolayer. This space and time alteration in the structure and dynamics of the membrane monolayer is captured by ÄÓh(r) and is simplistically referred to as ‘holes’ in the monolayer. The MI-QSAR models explain 70-80% of the variance in skin penetration enhancement across each of the two training sets, and are stable predictive models using accepted diagnostic measures of robustness and predictivity. Comparison of Machine Learning Algorithms to Predict ADME Properties Using Diverse Chemical Descriptors and Molecular Fingerprints Anthony E. Klon and David J. Diller We have compared the performance of ten different machine learning algorithms available in Weka to create binary classification models for blood-brain barrier (BBB) penetration and human intestinal absorption (HIA). For each data set, two models were constructed for each binary classifier; one using chemical descriptors and one using molecular fingerprints based on atom pairs and topological torsions, resulting in a total of 20 models for BBB penetration and HIA prediction. We describe the selection of descriptors used to train the chemical descriptor models. For both BBB and HIA datasets, the performance of all ten chemical descriptor models was tested by randomly scrambling the descriptors. For both datasets, the performance of all twenty models, descriptor and fingerprint-base, was further assessed and by randomly assigning compounds to the BBB penetrant / non-penetrant or HIA well-absorbed / poorly absorbed classes. Automatic QSAR Modeling of Blood-Brain Barrier Penetration by Gaussian Processes Method Olga Obrezanova*, Joelle M.R. Gola, Edmund J. Champness, Matthew D. Segall, BioFocus DPI, Chesterford Park, Saffron Walden, CB10 1XL, UK Blood-brain barrier (BBB) penetration is often one of the key properties considered during ADME (Absorption, Distribution, Metabolism and Excretion) studies in drug discovery. There have been a large number of in silico approaches to modeling BBB penetration reported in the literature describing either continuous models predicting the logarithm of the brain-blood concentration ratio (logBB) or classification models predicting BBB+/-. In this presentation we will discuss the results of modeling blood-brain barrier penetration by employing an automatic model generation process based on Gaussian Processes, a computational, machine learning technique. The rapid design-test-redesign cycles of modern drug discovery and the demand for fast model (re)building whenever data becomes available have given rise to a trend to develop computational algorithms for automatic model generation. Automatic modeling processes allow computational scientists to explore large numbers of modeling approaches very efficiently and make QSAR/QSPR model building accessible to non-experts. Automatic model generation requires unsupervised, computational techniques that are not dependant on any input from a user, are able to deal with a large number of descriptors and are not prone to overtraining. The automatic model generation process which we will present is based on Gaussian Processes, a powerful non-linear probabilistic method. The Gaussian Processes technique is highly appropriate for automatic model generation; it does not require subjective determination of model parameters, it is able to handle a large pool of descriptors and select the important ones, it is inherently resistant to overtraining and offers a way of estimating the uncertainty in predictions. In our previous work [1] we developed new techniques for implementing the Gaussian Processes method for modeling continuous data and compared their performance with other modeling techniques. In this presentation we will demonstrate how we have extended our Gaussian Processes techniques to model categorical data. We will present an automatic model generation process for building QSAR models and describe the stages of the process that ensure models are built and validated within a rigorous framework; descriptor calculation, splitting data into training, validation and test sets, descriptor filtering, application of modeling techniques and selection of the best model. We will then demonstrate the application of the automatic model generation process to modeling two blood-brain barrier penetration data sets; a continuous logBB data set and a classified BBB+/- data set. We will present examples of automatically generating both continuous and classification models, compare the resulting models with ‘manually’ built models and finally demonstrate the results of rebuilding an existing model by including new data points. Reference: 1. Obrezanova, O.; Csányi, G.; Gola, J.M.R.; Segall, M.D. Gaussian Processes: A Method for Automatic QSAR Modeling of ADME Properties. J. Chem. Inf. Model. 2007, 47, 1847-1857. Application of Machine Learning Tools for in silico ADME Screening Yojiro Sakiyama, Pfizer Global Research and Development, Pfizer Inc., IPC654 Ramsgate Road, Sandwich, Kent CT13 9NJ, UK To deal with the vast quantity of data from large compound libraries, computational in silico ADME screening is required to maximize efficiency of the drug discovery process. On the other hand, various machine learning tools originating from engineering areas have gradually gained considerable interest in drug discovery research. Here we have derived a relationship between the chemical structure and its ADME properties for a data set of in-house compounds by means of various in silico machine learning tools such as random forest, support vector machine, logistic regression and recursive partitioning. For model building, proprietary compounds comprising two classes (stable/unstable) were used with molecular descriptors calculated by the Molecular Operating Environment. The results using test compounds have demonstrated satisfactory results. Above all, classification by random forest as well as support vector machine yielded satisfactory results in an independent validation set, suggesting that nonlinear/ensemble-based classification methods might prove useful in the area of in silico ADME modeling.

Contact Information:
Program: Dr. Barry Hardy, eCheminfo Community of Practice & Research, Douglas Connect. Tel: +41 61 851 0170. barry.hardy -[at]- douglasconnect.com

Registration Enquiries: Nicki Douglas, Douglas Connect, Baermeggenweg 14, 4314 Zeiningen, Switzerland. Tel: +41 61 851 0461. eCheminfo -[at]- douglasconnect.com

or please visit:

http://echeminfobm810.eventsbot.com/

Barry Hardy

Community of Practice & Research Manager

InnovationWell eCheminfo ADME ADMET bioinformatics cheminformatics chemoinformatics Computational Biology Computational Chemistry Translational Medicine Molecular Medicine biology pharmaceutical pharma meeting workshop training Critical Path toxicology Life Sciences Drug Discovery Research and Development Drug Development Healthcare Innovation Innovation Management events

Comments

[ABHAY]

I WANT FULL INFORMATION ABOUT QSAR USED IN DRUG DISCOVERY & DISIGN .PLZ SEND ME URGENTLY................

[Compounding Pharmacist]

Compound Pharmacy has gained much popularity in the field of medicine. The medications are equally effective and safe for sick patients who cannot take the actual medications due to their personal allergies.

[fussball]

Gute Arbeit hier! Gute Inhalte.