Statistical Analysis of in Vivo Anticancer Experiments: Tumor Growth Inhibition
- PDF / 9,898,508 Bytes
- 11 Pages / 648 x 864 pts Page_size
- 5 Downloads / 149 Views
Statistical Analysis of in Vivo Anticancer Experiments: Tumor Growth Inhibition Ludwig A. Hothorn Drug Information Journal 2006 40: 229 DOI: 10.1177/009286150604000212 The online version of this article can be found at: http://dij.sagepub.com/content/40/2/229
Published by: http://www.sagepublications.com
On behalf of:
Drug Information Association
Additional services and information for Drug Information Journal can be found at: Email Alerts: http://dij.sagepub.com/cgi/alerts Subscriptions: http://dij.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://dij.sagepub.com/content/40/2/229.refs.html
>> Version of Record - Apr 1, 2006 What is This?
Downloaded from dij.sagepub.com at Queen Mary, University of London on June 10, 2014
STATISTICS
229
Statistical Analysis of In Vivo Anticancer Experiments: Tumor Growth Inhibition
Ludwig A. Hothorn, PhD Professor of Biostatistics, Unit Biostatistics, University of Hannover,
Key Words Treatment-to-control ratio; Tumor inhibition: Confidence intervals Correspondence Address Ludwig A. Hothorn, R D , Unit Biostatistics, University of Hannover, Herrenhaeuserstr. 2 , D-30419, Hannover, Germany (email: hothorn Qbiostat .uni-hannove~de).
Tumor growth inhibition data in in vivo anticancer experiments are commonly analyzed using the treatment-to-control ratio (TCR). Parametric and nonparametric confidence interval approaches for this ratio are introduced, enabling a quantitative statistical decision. The growth curves are characterized by the areaunder-the-curve technique, adjustedfor animalspecific survival. Simple simultaneous ap-
IN T R O D U C T l O N The effect of antineoplastic compounds can be tested in experiments in vivo by analyzing the appropriateness of an inhibition of tumor growth relative to a control. In some experiments, the tumor is characterized by repeated measures of tumor volume, and the corresponding number of survivors is recorded over time. The treatment-to-control ratio (TCR) of tumor volume at a selected measurement time is commonly used for characterizing the effectiveness of therapy. For example, the tumor growth rate of human neuroblastoma xenograft was inhibited to a TCR of 0.3 after 16 days of treatment with the synthetic fumagillin analogue TNP-470 in mice (1). Two specific statistical problems arise in testing: (a) accounting for missing values caused by mortality and (b) analyzing the ratio to control instead of the widespread used difference tocontrol. Inherently connected with the second problem is the question of biological relevance, that is, which TCR is meaningful as opposed to which has formal statistical significance (P < 0.05). Previous statistical approaches (2,3) have been based on a two-stage model of carcinogenesis, which estimates the number of tumors over time and tests group differences. Moreover, a latent variable model for susceptibility and multi-
proaches are proposed for complex designs, including several treatment
Data Loading...
