-2

that secondary metabolites present in the gut, including some that are potentially hepatocarcinogenic, may reach the liver by absorption.

The subject of hepatocarcinogenicity is very complex and has been the subject of much debate and numerous reviews. It is not the purpose of this Monograph simply to review the whole complex situation yet again. On the contrary its aim is to discuss two questions of great practical importance not only for those industries who produce, handle or sell chemicals such as drugs, pesticides, manufacturing aids etc., but also for all human kind.

The two questions are :

How should the finding of an increased incidence of liver tumours following exposure of laboratory animals to a test chemical be interpreted in terms of possible cancer risk (in the liver or at other sites) for man ?

How may it be established whether a new chemical carries such a risk for humans?

In relation to the first question, a survey of the literature indicates two important things: i) whereas numerous chemicals have been shown to be hepatocarcinogenic in mice and/or rats, very few have been shown to be so in man; and ii) although the liver may be the target organ for the carcinogenicity of a particular compound in, say, the mouse, another site (eg. the urinary bladder) may be the target in another species (e.g.man). In developing an approach to establishing whether a new chemical is a carcinogen for the liver, the system of studies advocated is necessarily relevant to investigating whether the agent is a carcinogen for other sites as well.

In listing causes of liver cancer in man, Hepatitis B virus infection, leading to chronic hepatitis and cirrhosis, is the most important contributory factor (cf.chapter F). There are similarities between man and laboratory animals with respect to hepatocarcinogenesis by ionising radiation, certain sex hormones, vinyl chloride, and possibly aflatoxin B1, but not by most other chemicals that have been shown to be liver carcinogens in laboratory animals. Consideration must be given to whether interspecies differences in the spectrum of factors that can predispose to hepatocarcinogenesis are so great that animal experiments are unhelpful in assessing whether a hazard to humans exists, or whether the apparent species differences are explained solely by the insensitivity of epidemiological methods under conditions of low levels of human exposure.

Today it is widely held that agents which facilitate the development of cancer fall into two broad types: those that act by a genotoxic mechanism and those

-3

that act by a non-genotoxic mechanism (the term non-genotoxic is preferred to epigenetic). This distinction is seen as fundamentally important in virtually all the sections of this report. Although, in general, more concern is attached to genotoxic than to non-genotoxic mechanisms, the potential importance of the latter type of activity for man should certainly not be lightly dismissed. The lower level of concern in the case of a non-genotoxic mechanism might be justified by a discontinuous dose-response relationship and/or by the fact that positive effects have been achieved only in animals exposed for prolonged periods to high, and often toxic, doses.

This monograph was prepared on the understanding that :

No single laboratory test and no group of tests could definitely predict either hepatocarcinogenicity or non-hepatocarcinogenicity for man.

The most that could be achieved would be to recommend procedures that would provide information pertinent to an understanding of the mechanism involved. Such an approach would be relevant to a more rational assessment of the hazard to man.

Chapters B to E constitute "state of the art" reviews of various aspects of hepatocarcinogenesis and in chapters F and G, based on these reviews, an attempt is made to answer the questions posed above.

The Task Force acknowledges the valuable contributions of Professor R. Wright (Southampton General Hospital, UK) and Dr. V.M. Craddock (Medical Research Council, UK) to the discussion of the aetiology of hepatic neoplasia in man, and the pathogenesis of hepatic neoplasia in laboratory rodents, respectively.

(Balance of document is held in the subcommittee files.)

Hearings have been scheduled by your subcommittee for February 22-23 on a report "Regulatory Procedures and Public Health Issues in the Environmental Protection Agency's Office of Pesticide Programs," commonly known as the "Benbrook Report."

Enclosed is a review of the Report. We respectfully request that this review be made a part of the Hearing record.

494

Regulatory Procedures

and Public Health Issues in the EPA's Office of Pesticide Programs

Comments of:

Agricultural Division
CIBA-GEIGY Corporation
P. O. Box 18300
Greensboro, NC 27419

January 31, 1983

INTRODUCTION

CIBA-GEIGY regulatory staff and scientific personnel have evaluated the December 16, 1982 report entitled "Regulatory Procedures and Public Health Issues in the EPA's Office of Pesticide Programs," commonly called the Benbrook Report. Comments contained in this document reflect the position of both the CIBA-GEIGY scientific staff and our management. As the draft report and appendices are massive, we have chosen to generally limit our specific comments to the six major issues raised in the report and to the report conclusions contained in Chapter 9. The only exception to this is Chapter 6, Issue 4. This should not be interpreted as concurrence or agreement with the many specific points raised in the individual chapters. On the contrary, the draft report contains so-called "statements of fact" and/or opinions which are incorrect or are presented in a manner which is unjustly critical of EPA's administration of the complex federal pesticide regulatory program.

We are both surprised and pleased with the depth and thoroughness of the report. Clearly, the report accurately describes this complex regulatory program from historical and procedural perspectives. The report, we feel, will be an excellent future reference for anyone wishing an overview of the federal pesticide regulatory process. While we disagree with some of the conclusions reached in the report, we commend those individuals responsible for researching and drafting a report which captures the complexities of the pesticide regulatory process and the difficulty in making risk/benefits decisions.

Issue 1. Regulatory Activities

This issue is primarily concerned with "registration" actions other than those authorized by Sections 3 and 5 of FIFRA. These are emergency exemption and special local need registrations which are authorized by FIFRA Sections 18 and 24 (c), respectively. State regulatory agencies play a key role in both the Section 18 emergency exemption program and in SLN registrations issued pursuant to Section 24(c). The report appears to suggest that potentially serious problems could be expected to occur from pesticides used under these programs because such usage:

places increased responsibilities for reviews on state
pesticide regulatory programs, some of which may have
difficulty in handling them;

increase the chances of localized adverse environmental and wildlife impacts; and

- entail significantly less complete and rigorous data requirements than required to obtain federal registrations.