
|
Leading Edge Predictors for Drug Discovery |

|
Download a Pre-print |
|
CSHIA ...Introduction to Oral Absorption |
|
CSHIA Predictor Capabilities |
|
CSHIA Calculates %HIA (%Oral Absorption) Predictions: |
| • | Continous %HIA values form 0% to 100% |
| • | Predicts the passive absorption of a drug via either the transcellular or paracellular route |
|
The only input required from the user is a structure for each compound to be predicted. The structure may be submitted as either a MOL file or SD file and it is not necessary for the structure to be in a three dimensional format or to be optimized to a minimum energy confirmation. |
|
For CSHIA, it is very important that the following types of compounds be entered as charged. |
| • | Quaternary amines (>Np<) |
| • | Pyridinium groups (aasNp) |
| • | Tertiary Sulfur (>Sp-) |
| Definition of HIA Activity Data |
|
Introduction to Human Intestinal Absorption |
|
Definition of %HIA (%Oral Absorption) Absorption is defined as the process involved in getting a drug from its dosage form into the body (1). Percent Human Intestinal Absorption (%Oral Absorption) is defined as the percent dose of orally administered drug to reach the hepatic portal vein, which drains the gastric, superior mesenteric and inferior mesenteric veins of the GI tract into the liver. Oral absorption takes into consideration metabolism that occurs in the gut wall, but not first pass metabolism in the liver. The most straight forward and common method of assay is to use radiolabeled drug and measure either: ((concentration oral dose - concentration excreted in feces) / oral dose) * 100 - or ( 100 - (concentration oral dose - concentration eliminatd in urine)) / oral dose) * 100 Often, both measurements are taken and compared for reciprocal agreement. This method does not take into account the fraction of drug absorbed and then eliminated in the feces through billary excretion, but it is generally to be considered reliable within reasonable limits of accuracy. HLPC assay methods are generally considered to have greater sensitivity as far as their detection limit, but often require a more thorough knowledge of the metabolites of the administered drug, each of which must be assayed separately in order to establish the total absorption before metabolism has taken place. (1) W. O. Foye, Principles of Medicinal Chemistry, 3rd Edition, Lea and Febiger, Philadelphia (1998) 24. |
|
The ability to predict the percent oral absorption (%OA) is primary goal in the design, optimization, and selection of potential candidates in the development of oral drugs. Most drugs cross the cellular barrier of the intestinal tract by passive diffusion through the cell wall (transcellular transport). A smaller number of drug candidates pass thorough the intercellular tight junctions (paracellular transport). In some cases, active and carrier-mediated transport mechanisms are involved. Such is the case with some amino acids, beta-lactams, nucleosides, vitamins, carbohydretes and certain monocarboxylic acids. The reverse situation, removal of xenobiotics, via efflux transporters can complicate the problem of in silico prediction of %HIA. Careful screening of hundreds of drugs (>750) to construct well populated training and external validation test sets has resulted in the generation of a robust predictor for %HIA for compounds that under go passive transcellular or paracellular transport. |
|
Data Sources used for CSHIA |
|
Oral Absorption Data Activity Range: The successful generation of a generalized in silico predictor for continous %OA requires a large and diverse set of drugs covering the complete range of oral absorption while, at same time, eliminating active and facilitated transport. Six hundred and twelve (612) were found (see CSHIA Dataset Profile). Below is histogram giving the distribution of compounds binned at intervals of 20% in OA values. Since oral drugs, by definition, should be “good” absorbers, published data is skewed towards high OA values. In the 612 drug dataset, 55% of the drugs had HIA values greater than 80% and 45% had values less than or equal to 80%.. Only 17% of dataset had an HIA value less than or equal to 40%. The compounds in the training and external validation datasets had an average percent HIA values of 71% and 79% respectively from random selection on 612 drugs. |
|
Sources for Experimental Oral Absorption Values: The values for HIA (% Oral Absorption) used in the generation of the CSHIA predictor were taken from the following sources. There was sufficient agreement of values reported for the same compound to justify combining the data from these sources. (1) Zhao Y. H. et al, J. Pharm. Sci. 90, 749 (2001) (2) Zmuidinavicus D. et al, J. Pharm. Sci. 92, 621 (2003) (3) Klopman G. et al, Eur. J. Pharm. Sci., 17, 253 (2002) (4) PDF (Physician Desk Reference), Publ. Thompson (2003) (5) Dolly C. Ed., Therapeutic Drugs, 2nd Ed, Publ. Churchill Livingstone (1999) |
| search | |
| links | |
| user login | |
| contact us | |
|
To contact us: |
![]() |
Phone: 978-501-0633 Fax: 781-275-5197 Email: sales@chemsilico.com |
Copyright © 2003 ChemSilico LLC All Rights Reserved Terms and Conditions of Use | Privacy Policy ChemSilico is a registered trademark of ChemSilico LLC, Tewksbury, MA 01876 |