PPDC Workgroup on 21st Century Toxicology/New Integrated Testing Strategies - March 3, 2009 Teleconference/Meeting - Presentation

Retrospective Cancer Analysis Overview: Cancer Assessment Review Committee

Hypothesis
Data analysis/methods
Preliminary Results
Next Steps

Background

OPP requires two sexes, two species rodent carcinogen bioassay for registration of all food use pesticides (870.4200a, 870.4200b and 870.4300). Results of over 500 studies have been reviewed by OPP's Cancer Assessment Review Committee (CARC). Concerns have been raised during the last two decades that the large number of animals, resources and time required to complete the two species cancer bioassays may no longer be warranted. To determine whether we could use a reduced animal bioassay, a subset of the CARC initiated the comprehensive review of all pesticides determined to be carcinogenic. The intent of this review is to provide insight regarding the usefulness of standard carcinogen bioassays for OPP's pesticide program.

The reduced animal cancer bioassay concept is not new. Many investigators have considered this in previous journal publications. For example:

Haseman and Huff (1987) found that the use of male rats and female mice only would have detected 96% of chemicals tested in all four sex-species combinations

Von Wittenau (1983) AND Ashby and Morrod (1991) also questioned the need for the four sex-species combinations

Gold et al., (1989), Huff et al. (1991) and Lai (1992) evaluated the utility of rodent cancer bioassays using results from the National Toxicology Program and from the International Agency for Research on Cancer database and determined that more than 85% of all chemicals would have been captured as a potential carcinogen using the combination of male rats and male mice (91‐94%) or male rats and female mice (91-92%) (Lai et al., 1992).

Van Osterhout et al (1996) evaluated the results of studies on pharmaceuticals included in the European Regulatory Database and concluded that "it appears that long-term carcinogenicity studies in mice in addition to those in rats have not been necessary to detect unacceptable carcinogen risks posed by pharmaceuticals. Carcinogen risk assessment on the basis of a lifespan study in a single rodent species in combination with short-term genotoxicity tests and mechanistic information seems feasible".

Hypothesis

To determine whether reducing the two species testing to a single species (rat) would be sufficient for determining carcinogenic potential for pesticides and whether there are possible alternative approaches to the detection of potential carcinogenic pesticides.

Data Analysis/Method

A subset of CARC members (Vicki Dellarco, Robert Mitkus, Brenda May, Jess Rowland, PV Shah, and Mary Manibusan) reviewed the cancer database. Criteria for inclusion of pesticide carcinogenic classifications and CARC determinations:

Available CARC reports based on a weight of evidence approach and application of Guidance for Carcinogen Risk Assessment (1986-2005). Only pesticide carcinogens determined between 1988 to 2008 were considered
Pesticide carcinogens with acceptable studies in both rats and mice as determined by the CARC
If data were used to support a cancer classification for more than one chemical, this data was only included once
Tumor responses included in this analysis were those determined to be the basis of the cancer classification as determined by the CARC and those determined to be relevant to humans.
The total number of positive carcinogenicity bioassays included in this current (n=156) analysis was not predetermined but was arrived at using the criteria as listed.

Preliminary Results

N=156 Pesticides that meet criteria, having both rat and mouse study
Mouse and Rat Tumors for All Chemicals did not significantly differ across sex; female and male exhibited similar cancer site profiles within each species.
Mouse tumors by strain: Most mice data are submitted using CD-1. No differences across strain in the major types of tumors (target organ‐liver and lung) identified.
Rat tumors by strain: Most rat data are submitted using Sprague Dawley. No differences across strain in the major types of tumors (target organ‐liver, thyroid, repro) identified.
Rat only tumors indicated that liver, thyroid and reproductive organ tumors were most prevalent among pesticide carcinogens tested to date. Mouse only tumors (n=53) indicated that liver and lung tumors were most prevalent among pesticide carcinogens tested to date.

Next Steps

Society of Toxicology Poster #1495 Presentation (March 16-20th) Baltimore, MD.
Consideration of alternative approaches (e.g., male rat and female mice combinations?)
Weight of evidence approach to include additional data streams as they become available.
Possible Scientific Journal Submission for Publication - Summer/Fall 2009.

Publications | Glossary | A-Z Index | Jobs