Advertisement

Computer-Aided Selection of Chemicals for Biological Testing: Estimation of Biological Activity

  • G. W. A. Milne
  • L. Hodes
Conference paper
  • 160 Downloads

Abstract

The United States National Cancer Institute manages a program whose goal is the discovery and development of drugs to be used in the treatment of cancer. This Program, which began in 1955, has examined approximately 500,000 materials for anticancer activity and it continues to screen much smaller numbers of compounds per year. The large database that has been built as a result of this work was used between 1980 and 1985 as the basis of an experiment in which the potential biological activity of a chemical structure was estimated before the compound was acquired and tested. The accuracy of such estimates and the impact upon the overall program is described.

Keywords

Anticancer Activity Inactive Compound Potential Biological Activity Connection Table High Decile 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    The starting point for this effort is generally regarded as the Congressional authorization, in 1955, of $5 million for the establishment by NCI of a Drug Development Program. See DeVita, V. T.; Oliverio, V. T.; Muggia, F. M.; Wiernik, P. W.; Ziegler, J.; Goldin, A.; Rubin, D.; Henney, J.; Schepartz, S. “The Drug Development and Clinical Trials Programs of the Division of Cancer Treatment, National Cancer Institute”. Cancer Clin. Trials, 1979, 2, 195–216.Google Scholar
  2. 2.
    On 23 June, 1986, the CAS Registry contained 7,908,232 entries (STN International online “News”). This number is currently increasing by about 750,000 compounds per year.Google Scholar
  3. 3.
    Beginning in 1986, NCI began to develop an in-vitro screening method which makes no use of the older P388 screen, or of the second stage, which involved more refractory tumors, also in mice.Google Scholar
  4. 4.
    Murine lymphocytic leukemia, strain P388, is the preliminary screen against which all compounds have been tested since 1976. The number of compounds for which there are P388 data in the DIS is now in excess of 200,000. Full details of all screening procedures are published by NCI in “Instruction 14” (1985). Copies of this booklet may be obtained from the Information Technology Branch, DTP, DCT, NCI, Bethesda, MD, 20205.Google Scholar
  5. 5.
    Milne, G. W. A., and Miller, J. A. The NCI Drug Information System. I. System. Overview. J. Chem. Inf. & Comp. Sci., 26, 154–159, (1986); Milne, G. W. A., Feldman, Alfred, Miller, J. A., Daly, G. P. and Hammel, M. J.: The NCI Drug Information System. II. The DIS Pre-Registry. J. Chem. Inf. & Comp. Sci., 26, 159–168, (1986); Milne, G. W. A., Feldman, Alfred, Miller, J. A. and Daly, G. P. The NCI Drug Information System. III. The DIS Chemistry Module. J. Chem. Inf. & Comp. Sci., 26, 168–179,(1986); Milne, G. W. A., Miller, J. A. and Hoover, J. R.: The NCI Drug Information System. IV. Inventory and Shipping Modules. J. Chem. Inf. & Comp. Sci., 26, 179–185, (1986); Zehnacker, M. T., Brennan, R. H., Milne, G. W. A., and Miller, J. A.: The NCI Drug Information System. V. The DIS Biology Module. J. Chem. Inf. & Comp. Sci., 26, 186–193, (1986); Zehnacker, M. T., Brennan, R. H., Milne, G. W. A., Miller, J. A. and Hammel, M. J.: The NCI Drug Information System. VI. System Maintenance. J. Chem. Inf. & Comp. Sci., 26, 193–197, (1986);CrossRefGoogle Scholar
  6. 6.
    The acronym “NSC” stands for National Service Center, a short form of Cancer Chemotherapy National Service Center, the early name for the program. The “NSC Number” is used by NCI as a Registry Number. Other Registry Numbers, such as the CAS Registry Number are not useful for NCI because CAS Registry Numbers can not be assigned to the confidential structures which constitute about half of the NCI database.Google Scholar
  7. 7.
    Hodes, L. “Computer-Aided Selection of Novel Antitumor Drugs for Animal Screening”. ACS Symp. Ser. 1979, 112, 583–603. Hodes, L. “Selection of Molecular Fragments for Structure-Activity Studies in Antitumor Screening”. J. Chem. Inf. & Comp. Sci., 1981, 21, 132–136. Hodes, L. “A Two-Component Approach to Predicting Antitumor Activity from Chemical Structure in Large Scale Screening”. J. Med. Chem., 1986, 29, 2207–2212. Paul!, K., Hodes, L., and Simon, R. M. “Efficiency of Antitumor Screening Relative to Activity Criteria”. J. Natl. Cancer Inst., 1986, 76, 1137–1142.CrossRefGoogle Scholar
  8. 8.
    Much of the data discussed here may be found in: Hodes, L. J. “Computer-Aided Selection of Compounds for Antitumor Screening: Validation of a Statistical-Heuristic Method”. J. Chem. Inf. & Comp. Sci. 1981, 21, 128–132.Google Scholar
  9. 9.
    Five of the original 400 compounds were dropped from consideration when it was discovered that their structures were incompletely described. About 2% of all compounds tested have incompletely defined structures but are tested in spite of this for a variety of reasons.Google Scholar
  10. 10.
    To be “highly active” against P388, a compound must show a reproducible %T/C greater than 175. Compounds whose %T/C values fall below 120 are considered “inactive” and the remainder, with %T/Cs lying between 120 and 175 are regarded as “moderately active”.Google Scholar
  11. 12.
    Milne, G. W. A., Feldman, Alfred, Miller, J. A., Daly, G. P. and Hammel, M. J.: The NCI Drug Information System. II. The DIS Pre-Registry. J. Chem. Inf. & Comp. Sci., 26, 159–168, (1986).CrossRefGoogle Scholar
  12. 13.
    In an attempt to create a “National Screening Center” as a service, NCI policy has always tested chemicals at no charge and has also declined, as a general matter, to purchase samples. Most of the compounds tested are not available for sale and even if they were, the Program could not afford to buy them. Testing data are supplied immediately to the supplier and, if so requested, the NCI will afford confidentiality to all such data.Google Scholar
  13. 14.
    In 1986, the decision was made to shift from animal, or in-vivo testing in the primary screen to in-vitro testing against human tumor cell lines. This decision renders the Hodes screen invalid at a stroke and is a good example of the susceptibility of such methods to external events.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1988

Authors and Affiliations

  • G. W. A. Milne
    • 1
  • L. Hodes
    • 1
  1. 1.Developmental Therapeutics Program, Division of Cancer TreatmentNational Cancer InstituteBethesdaUSA

Personalised recommendations