Artificial Neural Network Based in Vitro-in Vivo Correlations
In vitro drug dissolution test facilitates design, development, and quality assurance of solid (oral) products. During product development dissolution tests are used to characterize impact of formulation variables on drug release and to optimize formulations to achieve a target drug release profile (assuming the in vitro test is predictive of in vivo drug release). Every manufactured batch must meet pre-set specifications prior to being released for distribution. Certain “minor” post-approval changes in the manufacturing process and/or formulation may be justified by demonstrating equivalent in vitro dissolution profiles1. For certain “major” post-approval changes the FDA may grant requests for waiver of in vivo bioequivalence studies when a sponsor demonstrates acceptable predictive ability of the selected in vitro test (i.e., availability of an validated in vitro to in vivo correlation -IVIVC)2.
KeywordsArtificial Neural Network Drug Release Root Mean Square Processing Element Slow Release Formulation
Unable to display preview. Download preview PDF.
- 1.“Draft Guidance for Industry: Modified Release Solid Oral Dosage Forms; Scale-Up and Post Approval Changes: Chemistry, Manufacturing and Controls, In Vitro Dissolution Testing and In Vivo Bioequivalence Documentation”, U. S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research, July 1996.Google Scholar
- 2.“Draft Guidance for Industry: Extended Release Solid Oral Dosage Forms; Development, Evaluation and Application of In Vitro/In Vivo Correlations”. U. S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research, July 1996.Google Scholar
- 3.Hussain, A. S., et. al. Artificial neural network (ANN) approach for establishing a predictive relationship between in vitro drug release and in vitro serum drug levels. Pharm. Res. 10: S-489 (1996)Google Scholar
- 4.Piscitelli, D. A., et. al. Investigation of an Level A correlation for diltiazem. Pharm. Res. 10: S-475 (1996).Google Scholar
- 5.Dowell, J. A.., et. al. Neural network model development for in vitro — in vivo relationships. Pharm. Res. 10:S-450(1996).Google Scholar
- 6.Robert Hecht-Nielsen. Neurocomputing, Addison-Wesley Publishing Company, Reading, Massachusetts, 1989.Google Scholar
- 7.Neural Computing: A Technology Handbook for Professional II/Plus and NeuralWorks Explorer. Neural-Ware, Inc., Pittsburgh, PA. (1993).Google Scholar
- 9.Hussain, A. S., Yu, X., Johnson, R. D. Application of neural computing to pharmaceutical product development, Pharm. Res. 8, 1248–1252 (1991).Google Scholar
- 18.Hussain A. S.: Application of artificial neural network to population pharmacokinetic and pharmacodynamic data analysis. In: Pharmacokinetic/Pharmacodynamic Analysis: Accelerating Drug Discovery and Development. IBC Biomedical Library Series, Southborough, MA., J. Schlegel (Ed.), July 1996Google Scholar