Shell Battles to Save Dieldrin and to Weaken Federal Controls on Cancer Producing Chemicals

The Shell Chemical Company is fighting back vigorously against the Environmental Protection Agency's order suspending production of the pesticides aldrin and dieldrin. On August 2, EPA Administrator Russell Train announced the suspension, stating that continued manufacture of the chemicals posed an "imminent hazard" to human health. Train cited studies showing that dieldrin residues are present in most of the food Americans eat, and that small amounts of dieldrin produced cancer in mice and rats when added to their diets. Dieldrin is believed to be most dangerous to breast-fed babies, who may absorb up to 50 times as much dieldrin per pound of body weight as the average adult.

Unter the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA), the EPA Administrator is empowered to initiate "cancellation" proceedings to remove from the market any pesticide which presents a serious risk to human health or to the environment. When a manufacturer contests a cancellation, the agency must hold hearings on the chemical's hazardousness before an administrative law judge. If waiting for a final decision in cancellation proceedings would present an "imminent hazard" to human health, however, the Administrator may order production or use of the chemical "suspended" until the decision is rendered. The law provides for a speeded-up hearing and decision if a manufacturer opposes a suspension order. The final agency decision in both cancellation and suspension proceedings may be appealed to the U.S. Courts of Appeals.

The Environmental Defense Fund first demanded action against aldrin and dieldrin (the chemicals are closely related, and are commonly referred to together as "dieldrin") in 1970. In 1972, William Ruckelshaus, then EPA Administrator, cancelled the pesticide, but refused to order production suspended. Hearings did not get under way until October 1973. In view of the evidence presented at the cancellation hearings, much of it based on Shell's own studies of mice and rats, Train ordered the chemicals suspended last month, despite heavy pressure from the pesticide industry and its congressional friends.

Shell's attack on the suspension order is two-fold. In the first place, according to a succession of scientists retained by the company, the mouse is not a suitable animal for cancer research. Second, they maintain that the evidence does not meet the "imminent hazard" standard which the law requires for suspension of a chemical. In addition, the Shell experts claim to have found numerous errors in their own published studies of the effects of dieldrin, all of which, they say, tend to show dieldrin more cancer-producing in laboratory animals then in fact it is.

Independent scientists regard the outcome of the case as critical for the future of federal regulation of cancer-producing chemicals. Under the 1958 "Delaney amendment" to the Food, Drug, and Cosmetic Act, it is illegal to use any chemical on crops, in animal feed, or otherwise if it results in residues in food and if it has been shown to produce cancer in animals or man. (The law provides an exception for pesticide chemicals used on raw agricultural products, provided that the level of the pesticide in food does not exceed limits set by the Administrator of EPA. Under present guidelines, food must be seized and destroyed when dieldrin levels exceed .3 parts per million.) For decades, the mouse has been the standard laboratory animal testing the effects of food additives. If Shell's experts are able to establish as a matter of law that the mouse is an unsuitable animal for cancer research, numerous chemicals now banned on the basis of experiments with mice may find their way into the American diet. Two bills are now pending in Congress which would weaken the 1958 amendments and allow carcinogenic chemicals to be added to food. Though the bills are thought unlikely of passage, a victory for Shell in the dieldrin case could accomplish the same result with less publicity.

Cancer researchers find the mouse an extremely useful test animal for several reasons. Decades of careful breeding have produced various strains of genetically pure mice, each with its own characteristics. Some strains show a very low "spontaneous" rate of cancers, while others naturally develop cancer with greater frequency. The short lifespan of the mouse makes it possible to examine the effects of a chemical over the animal's entire life. Since cancer most commonly appears late in life, often long after exposure to the cancer-producing substance, this type of study is especially important. In addition, the animal's small size makes it practical to use much greater numbers of test and control animals, and therefore obtain more reliable results, than would be the case if a larger animal were studied.

According to Shell's witnesses, the fact that cancers occur spontaneously in the mouse makes it impossible to know for sure whether cancer in a test animal has been caused by a particular chemical. They also maintain that the incidence of cancer may be affected by such factors as the type of bedding material used in the animals' cages.

[4 ELR 10165]

The Shell position ignores the purpose of controlled experiments, in which all factors except one — in this case, dieldrin in the diet — are kept the same for test and control animals. Thus if bedding of cedar shavings causes the rate of cancer to increase for test animals, it will have the same effect on the controls. Nor is it necessary to prove that a particular case of cancer in an individual animal was caused by the substance being tested, only that animals exposed to a chemical showed a statistically higher rate of cancer than animals raised under otherwise identical conditions. Curiously, one of the scientists now condemning use of the mouse for cancer research has for several years been on a National Cancer Institute panel which has authorized grants totaling several hundred thousand dollars for studies of mice.

Shell argues that an extremely high standard must be met before a chemical can be described as carcinogenic (cancer-producing). According to the company, the chemical must be shown to produce cancers in two animals other than the mouse; these cancers must spread through the bloodstream to other organs of diseased animals; and cancerous tissues taken from one animal must grow when transplanted into the bodies of healthy animals.

The Shell position is in sharp contrast to the definition of a chemical carcinogen that has long been standard among all nationally and internationally recognized experts on cancer: a chemical is considered cancer-producing if animals which have been exposed to it develop cancer in greater numbers than unexposed animals, or if cancers develop sooner, or if certain unusual kinds of cancer are found. According to Shell, this is "too black-and-white" a view.

Most specialists in the field of cancer research believe that "a carcinogen is a carcinogen," and that it is meaningless to differentiate between "rat carcinogens," "mouse carcinogens," and "human carcinogens." Shell argues, on the other hand, that even clear proof that a chemical causes cancer in animals other than mice is not of itself sufficient of support a ban on the chemical. Dr. Don Stevenson, director of the Shell research laboratories in Tunstall, England, and a participant in the company's policy decisions, told the EPA hearing that even if a chemical were shown to produce cancer in rats, the odds of its being carcinogenic in man were still only about 70 percent. Administrative Law Judge Herbert Perlman, who presides over the hearings, turned to the witness. "Let me ask you this," he said to Stevenson, "and this is something that bothers me greatly, isn't 70 percent enough (to support a ban)?" The Shell official replied, "I do not think so, no, because on this basis you rule out many potentially useful compounds." Stevenson did not say what odds he believed were sufficient to justify banning achemical.

Shell places greatest reliance on a study of workers at its dieldrin plant in Pernis, Holland. Of 69 workers who were exposed to dieldrin for more than ten years, not a single case of cancer was reported. Quoting Alexander Pope, a Shell witness urged that "the proper study of mankind is man," and that this evidence should be given greater weight than any data on dieldrin-induced cancer in animals.

According to independent scientists called to testify on behalf of EPA, the Pernis study proves little. In the first place, Shell's plant appears to have kept worker contamination to a minimum. The mean level of dieldrin in the blood of the Pernis workers was only eight times that of a sample of North Carolinians whose only exposure to dieldrin came from their food and from particles of the chemical in the air. Moreover, the sample is far too small for meaningful conclusions to be drawn. In Hiroshima, the rate of leukemia among persons exposed to the atomic bomb is 1 in 60, a proportion considered extremely high. In addition, the length of the latency period between exposure to dieldrin and development of cancer in humans is still unknown, but it is believed to be as much as 30 years. The chemical has only been in use for 20 years, so that it may be some years before patterns of dieldrin-caused cancer begin to emerge. In Hiroshima, on the other hand, the full latency period has elapsed.

Shell contends, on the other hand, that cancer frequently appears before the end of the normal latency period. One witness cited Sir Percival Pott's study of cancer of the scrotum among London chimneysweeps in 1775. Though the usual onset of the disease was at puberty, Pott found cancer in one eight-year-old boy. Thus, according to Shell, if any cancer were going to occur among its Pernis workers, it would have arisen by now.

The Shell experts pooh-pooh data on the high levels of dieldrin in human milk, and on studies showing that some chemicals will produce cancer in animals when administered just after birth though not when administered later in life. A Shell witness, discussing breast-feeding in the first three months of the human baby's life, commented that "the three-month period about which concern was expressed is only 1/284 of a lifetime." This fraction assumes, of course, a lifetime of 71 years, and not one cut short by cancer.

Ironically, the most convincing proof that dieldrin causes cancer comes from studies which Shell itself conducted, using mice and rats. Though the company concedes that dieldrin produces cancer in the liver of mice, it now claims that its own published report on the study erroneously showed too high a rate of non-liver cancer.

Shell's animal testing was performed by a husband-and-wife team of veterinarians named Ferrigan at the Tunstall laboratories. According to Dr. E. Thorpe, a Shell scientist, company officials were dissatisfied with the Ferrigans' findings regarding the test animals' livers, and Thorpe was assigned to the project. A paper describing the Ferrigans' results was withdrawn from the scientific journal to which it had been submitted. Later, an article on the study appeared under the authoriship of Thorpe, with no acknowledgment or mention of the Ferrigans' research. [4 ELR 10166] Such failure to cite the principal researchers on a major study is considered extremely unusual in the scientific community. The Ferrigans have left Shell, and the company claims not to know their present whereabouts.

Testifying under oath, Dr. Thorpe told the suspension proceeding in August that 60 healthy mice had unaccountably been left out of the computations. He apologized for the error, saying he had no knowledge of what had happened to the missing mice. Pressed for an explanation, he suggested that some of the animals might have got "caught in the food hopper" of their cages. Dr. Thorpe was asked if he knew for a fact that the sixty mice had all been healthy. "No," he answered, "but my contention would be that they were almost certainly the animals without tumors which for some reason or aother got omitted from the computation." Dr. Thorpe did not explain what difference between healthy and diseased mice made only the former susceptible to entrapment in their food dispensers.

Judge Perlman asked the witness, "Isn't it rather strange that all sixty animals had no tumors?" Thorpe replied, "I don't know what you mean by 'strange,' sir." The judge explained: "Doesn't it seem rather significant to you that after a published study showing certain results, we come to the hearing and for the first time we learn that there are 61 missing animals, 60 of whom were in the treatment groups, and in the 60 there were no tumors …. I am just saying that it does stretch the credibility a little bit."

On September 4, Dr. M. Adrian Gross, a cancer research with the Food and Drug Administration, testified that his experience as an investigator for FDA led him to believe that the missing mice theory should be checked out throughly before being accepted as fact. While indignantly denouncing any suggestion of misrepresentation, Shell's counsel, William D. Rogers of the Washington law firm Arnold & Porter, explained that in fact there had been no missing mice. The problem, according to Rogers, was simply an error in counting the number of test animals. There was no immediate explanation for the divergence of opinion between Thorpe, who had already returned to Shell's English laboratories, and the company's Washington lawyer.

Gross, who also testified for EPA during the cancellation hearings that preceded the suspension hearings, was the subject of controversy this spring when EPA attorneys sought to recall him to the witness stand for further questioning. Rogers objected, and telephoned and wrote to a superior of Gross at FDA, stating — inaccurately — that Gross would testify that almost half the cancers occurring in the U.S. each year were caused by dieldrin. Rogers urged that Gross' statements were likely to be taken as official FDA policy and requested "consideration of this matter." After attorneys for the Environmental Defense Fund charged Rogers with attempting to silence a witness, Judge Perlman directed Rogers to refrain from further out-of-court contacts of this sort. Rogers has since been nominated by President Ford to be Assistant Secretary of State for Latin American Affairs, but continues to direct the presentation of Shell's case.

The validity of other data submitted by Shell has also come into question. In 1971, a highly respected pathologist, Dr. W. A. D. Anderson examined sections of tissue from animals fed varying quantities of dieldrin and from controls. No unusual tissue development was found. In 1973, Dr. Melvin Rueben, a scientist working for EPA examined the same slides and found numerous cancers in the test animals. Dr. Anderson then reexamined the slides and reported results similar to the EPA consultant's, although they differed on whether some of the growths were cancerous or benign. There is as yet no explanation for Anderson's initial diagnostic errors.

The suspension hearings came to an end on September 12 and Judge Perlman's decision is expected around September 20. EPA Administrator has the power to reverse the Administrative Law Judge's ruling, although in practice, this rarely occurs. The final agency decision will almost certainly be appealed to the federal courts by Shell, if it loses, or by the Environmental Defense Fund, if the chemical company is successful.

Even if the Shell Chemical Company fails in its effort to nullify the Delaney amendment and loosen the restrictions on cancer-producing chemicals in Americans' food, the case had raised serious questions about regulation on behalf of public health in this country. Critics of chemical and pharmaceutical manufacturers have long charged that where the profits at stake are high enough, experts can be found willing to advance views of highly questionable validity. Under the federal law governing pesticides, the task of weighing the voluminous evidence, and of deciding which of two conflicting viewpoints is scientifically correct, is assigned to a single non-scientist. While Judge Perlman's firm grasp of the issues in dispute and of the massive evidence in the record has won high praise, the structure of the law creates the possibility that under other circumstances, an administrative law judge — and the public — might become the victims of concocted evidence. The price of such an error might well be paid in human lives.

The dieldrin hearings also raise again the issue of the proper limits of the lawyer's role in cases affecting the public health and safety. While no one would deny the right of a corporation to have its case presented effectively, the adversary process assumes that there will be no use of pressure to keep witnesses off the stand, and that attorneys, in arguing, will not distort their own witnesses' earlier testimony. It is known that within Arnold & Porter, which under Abe Fortas and the late Thurman Arnold achieved an excellent reputation for high standards in choosing and representing clients, Rogers' handling of the dieldrin case has caused some dismay.

In April, Rogers wrote to EPA urging that dieldrin not be suspended, and suggested that removing the chemical from the market might suggest to Americans that dieldrin [4 ELR 10167] was "really a more serious cancer threat than cigarettes." Several weeks ago, a witness for Shell in the suspension hearings again cited the example of cigarettes as a substance which, though carcinogenic, is not illegal. Judge Perlman reminded the witness that the individual adult can decide whether or not to smoke cigarettes, while the public has no choice, and usually no knowledge, as to the level of dieldrin in our food supply.

As of 1971, the last year for which data is available, the average American already had .29 parts per million of dieldrin in his or her fatty tissues, and the figure was on the increase. This is just below the .30 parts per million level at which federal officials must seize and destroy meat, fish, and poultry intended for human consumption. Statistics on dieldrin levels in human fat are of significance primarily for children of nursing mothers.

When EPA announced its suspension of dieldrin in August, a company spokesman declared the issue "small potatoes." Shell Chemical is a division of the Shell Oil Company, which enjoyed record profits last year because of the increase in oil prices. Nonetheless, the company is fighting bitterly, and incurring substantial legal costs in the process, to retain those "small potatoes." To the mother who hears the child at her breast likened to a person buying a pack of cigarettes, and the first three months of the biologically hypersensitive newborn's life dismissed as "only 1/284 of a lifetime," the priorities of the company and its counsel must be difficult indeed to comprehend.