Exact Poly-K Test

• Activity Number: 615
• Type: Contributed
• Date/Time: Thursday, August 2, 2012 : 8:30 AM to 10:20 AM
• Sponsor: Biometrics Section
• Author(s): Matthew Jackson*+ and Mohammad Rahman and Karl K Lin and Ram C. Tiwari
• Companies: FDA and FDA and FDA and FDA/CDER/OTS/OB
• Keywords: Permutation test ; Poly-K test ; Exact test ; Carcinogenicity study

Abstract

For animal carcinogenicity studies with multiple dose groups, positive trend (dose response relationship) test and pairwise comparisons of treated groups with control are generally performed using the Cochran-Armitage, Peto test, or Poly-K test. These tests are asymptotically normal test. Exact versions of the Cochran-Armitage and Peto tests are available based on the permutation test assuming fixed column and row totals. For the Poly-K test, column totals depend on the mortality pattern of the animals and can not be kept fixed over the permutations of the animals. Therefore, application of permutation test assuming fixed total does not give correct exact p-values. In this work a modification of the permutation test is suggested that can be applied to exact Poly-K test. This modification adjusts the column totals depending on the regrouping of animals in every permutation, and thus provides the correct exact p-values.

Disclaimer: This presentation reflects the views of the authors and should not be construed to represent FDA's views or policies