The
Biostatistics Journal Club
will meet on
Tuesday, November 3, 2009
COPH 2280 12:00p.m. – 1:00p.m.
The topic of discussion will be
presented by
Ralph l. kodelL, PH.D.
Professor, Department of Biostatistics
University of Arkansas for Medical Sciences
Should we conduct
pre-clinical assessments of tumorigenicity
with the Peto or Poly-k test?
A comparison of
underlying assumptions,
statistical characteristics, and regulatory perspectives.
Abstract
The 1980
IARC guidance document on testing for tumorigenicity in long-term animal
experiments recommended a statistical test for dose-related trend that
stratifies animals dying with tumors of a specific type into fatal and
incidental groupings, depending on whether or not, in the judgment of a
pathologist, the tumor was the cause of death. The recommended test has become
known as Peto’s cause-of-death (COD) test, or simply, the Peto test, in
recognition of Richard Peto’s 1974 paper and his lead authorship of the IARC
document. The Peto test has been used extensively in the pharmaceutical industry
for the evaluation of occult tumors, especially in the United States and Europe.
After being favored unofficially by the US Food and Drug Administration’s Center
for Drug Evaluation and Research (FDA/CDER) for many years, the Peto test was
officially recommended by FDA/CDER in its 2001 draft guidance for industry. In
1988 John Bailer and Chris Portier developed the Poly-k test for tumorigenicity
which does not require a COD determination. Instead it relies on the assumption
that the distribution of time to tumor onset can be expressed as a function of
the kth power of time. Animals dying without tumors contribute to the test
statistic according to weights determined by this kth power. Given that k is
almost never known, Bailer and Portier recommended using k=3 based on an
empirical study. The Poly-3 test is the version most often used and it has been
gaining popularity over the years. The 2001 FDA/CDER draft guidance document
recommends the Poly-3 test when COD information is unavailable for implementing
the Peto test. The US National Toxicology Program (NTP) has adopted the Poly-3
test as the official method for evaluating its animal bioassays for
carcinogenesis. In this presentation, the underlying assumptions and structures
of the Peto and Poly-k tests are outlined. A limited comparison of their
statistical performance in terms of size and power is presented, along with a
view of their regulatory standing. Some modifications and generalizations of
both tests are mentioned.
