Abstract |
Four statistical techniques are described for comparing the in vitro cytotoxic activity of an analogue and its parent compound in the human tumor cloning system. These techniques include: the Spearman rank correlation coefficient; a sequential pairwise analysis; the kappa statistic for measuring agreement; and the McNemar test of symmetry. These statistical procedures were applied to results from 54 human tumor specimens that were simultaneously tested against multiple concentrations of the Vinca alkaloid vinblastine and its semisynthetic analogue, vinzolidine. Based on the percentage of survival of tumor colony-forming units in the human tumor cloning system, vinzolidine was shown to have a 26% in vitro response rate with a spectrum of activity across major tumor types. However, all four statistical procedures indicated that the activity of vinzolidine was not superior to that of its parent compound. Of the four statistical procedures examined, the sequential pairwise design appears to be best suited for comparing in vitro activity of an analogue and its parent compound and could result in a savings of both time and resources required for such comparisons.
|
Authors | B Lathan, G M Clark, D D Von Hoff |
Journal | Cancer research
(Cancer Res)
Vol. 45
Issue 12 Pt 1
Pg. 6286-9
(Dec 1985)
ISSN: 0008-5472 [Print] United States |
PMID | 4063979
(Publication Type: Comparative Study, Journal Article, Research Support, Non-U.S. Gov't)
|
Chemical References |
- Vinca Alkaloids
- vinzolidine
- Vinblastine
|
Topics |
- Cell Survival
(drug effects)
- Cells, Cultured
- Humans
- Models, Biological
- Neoplasms, Experimental
(pathology)
- Neoplastic Stem Cells
(drug effects)
- Statistics as Topic
- Structure-Activity Relationship
- Vinblastine
(pharmacology)
- Vinca Alkaloids
(pharmacology)
|