Conclusion

Chapter 6: The AEC Assumes Responsibility for Radioisotope Distribution

At first, there appears to have been some confusion over the responsibility of the AEC for its own research program and for its program to distribute radioisotopes to private researchers. As discussed in chapter 1, two 1947 letters from AEC General Manager Carroll Wilson describe strong consent requirements. The April letter to Stafford Warren was expressly directed to the terms on which research conducted by AEC contractors (including universities) would be approved. The November letter was sent to Robert Stone. As we have discussed, those clear statements to contract researchers do not seem to have been made to those applying for radioisotopes. This confusion about the relationship between contract research and isotope distribution is discussed in a September 26, 1947, memorandum from J. C. Franklin, manager of Oak Ridge Operations, to Carroll Wilson.[22] Other correspondence also indicates confusion over whether the AEC's own labs (which were themselves often operated by contractors) were to follow the procedures for the radioisotope distribution program, which would have placed their human use requests before the Subcommittee on Human Applications.

Initially, requests for by-product materials from within the AEC used a form that did not specify whether the radioisotope was to be used on humans.[23] By August 1949, Shields Warren, director of the AEC's Division of Biology and Medicine, had directed that human use by AEC laboratories be subject to review by the Subcommittee on Human Applications.[24] "Since this procedure has not been uniformly followed in the past, we are writing to acquaint you with the appropriate details." Paul Aebersold, Chief, Isotopes Division, to Carroll Tyler, Manager, Los Alamos, 5 October 1949 ("Use of Radioisotopes in Human Subjects") (ACHRE No. DOE-021095-B-4), 1. An identical memo was also sent to the manager of the AEC's New York office regarding requirements for Brookhaven National Laboratory. Paul Aebersold, Chief, Isotopes Division, to W. E. Kelley, Manager, New York, 5 October 1949 ("Use of Radioisotopes in Human Subjects") (ACHRE No. DOE-012795B). However, when regulations governing radioisotope distribution were first promulgated, AEC-owned facilities were specifically exempted from all such regulations.[25] Warren's goal was achieved instead by a memorandum from Carroll Wilson in July 1950. This memorandum discontinued use of the earlier form and directed that all requests use the same form used by outside purchasers, which directed human use requests to the Subcommittee on Human Applications.[26]

The AEC Subcommittee on Human Applications

The subcommittee formally met only once a year to discuss general issues. By its second meeting, in March 1948, membership had grown to four. Dowdy was no longer on the subcommittee; Joseph Hamilton and A. H. Holland had been added. Hamilton was, as described in chapter 5, a physician-investigator with the University of California's Radiation Laboratory in Berkeley. Holland was a physician-investigator who became medical director at the AEC's Oak Ridge Operations in late 1947. (As we shall see in chapter 13, he played a central role in the question of the declassification of secret experiments.) As the subcommittee continued to "examine each case on its own merits" it began to generate principles for "general guidance." In doing so, it began to categorize experiments, apparently according to the degree of hazard posed.

One category was tracer studies in "normal adult humans" using beta and gamma emitters with half-lives of twenty days or fewer. Applications needed to include information on biodistribution and biological half-life of the radioisotope, based on either animal studies or references to the literature.[28]

A second category was studies in "normal children." In 1948 the subcommittee did not issue detailed guidelines, but instead simply stated that such applications "would be given special scrutiny by the Subcommittee on Human Applications."[29] In 1949 it issued more detailed guidelines, which indicate that the concern was with minimizing risk, not requiring or overseeing consent:

In general the use of radioisotopes in normal children should be discouraged. However, the Subcommittee will consider proposals for use in important researches, provided the problem cannot be studied properly by other methods and provided the radiation dosage level in any tissue is low enough to be considered harmless. It should be noted that in general the amount of radioactive material per kilogram of body weight must be smaller in children than that required for similar studies in adults.[30]

Coupled with the children's category in 1949 were studies on pregnant women: "The use of radioactive materials in all normal pregnancies should be directly discouraged where no therapeutic benefit is to be derived."[31] Although not specifically mentioned in the minutes, such a policy may, like research in "normal children," have been waived for "important researches" that could not otherwise be undertaken.

One recurring difficulty was the problem of deciding when an application could be considered "safe." There was no simple, mechanical process for making such a judgment. This can be seen in the subcommittee's detailed consideration of an application for phosphorus 32 to be used in a blood volume study of children. The amount of radioactivity proposed ranged from 1/4 to 1 microcurie per kilogram of body weight. Initially, three of the four members approved the application and the allocation was made. However, the fourth member, replying late, reopened the question. Following reconsideration by the entire subcommittee, three of the four members concluded the original application for use on children should be turned down and the investigator asked to revise the application to "state the importance of making the study in children" and to keep the amount of activity less than 1/2 microcurie per kilogram.[32] The reduction in allowable amount of activity illustrates both the diligence with which the subcommittee pursued its task and the inherent difficulties in making judgments about what constituted "safe" practices in a rapidly developing field of research.

The subcommittee's task was made a bit easier when considering applications with adults, where it could draw upon occupational guidelines. Requests for "long-lived radioisotopes" were placed in a third category, defined as those with a biological half-life greater than twenty days. In contrast with experiments on children, here the subcommittee was willing to set a general dose limit: "The dosage in the critical tissue should be such as to conform to the limitations stated by the National Committee on Radiation Protection."[33] (The NCRP, now the National Council on Radiation Protection and Measurements, is an independent organization that publishes occupational radiation protection guidelines based on expert reviews of contemporary scientific knowledge.) As with children, such applications "must be reviewed separately." The subcommittee did not wish this limit to be ironclad: "In special cases, however, the Subcommittee on Human Applications may permit the use of radioisotopes in higher dosages."[34] At this point the subcommittee appears to have been establishing general principles; no specific radioisotopes or particular research proposals are mentioned.

A final category was applications using radioisotopes with long half-lives in patients with short life expectancies. The term moribund was used in correspondence by Paul Aebersold prior to the second meeting of the subcommittee in March 1948. He wrote to the subcommittee members explaining that the item was on the agenda because requests for such work had been received. He referred to a written request from a physician at Massachusetts General Hospital to use calcium 45 and an oral request from a staff member at Presbyterian Hospital in Chicago to use testosterone labeled with carbon 14. Aebersold did not provide any details as to the purposes of the proposed research. The issue was what policy to adopt when the patients were predicted not to live long enough for long-term hazards to develop. Aebersold told the subcommittee that "this office feels that such requests should be allowed if a satisfactory mechanism for determining the 'moribundness' of the patients in question is established. We believe that this question should be decided by a group of doctors and written evidence signed by the group filed with the Isotopes Division prior to use of the material."[35]

The subcommittee had no objection to the basic principle of applying larger doses to patients with short life expectancies, but its language was more oblique than Aebersold's letter: "It is recognized that there may be instances in which the disease from which the patient is suffering permits the administration of larger doses for investigative purposes."[36] Safeguards were to be provided by reliance on the judgment of local physicians, not a precise definition of moribund. Indeed, the subcommittee did not even use the term. Applications would be approved providing:

1. Full responsibility for conduct of the work is assumed by a special committee of at least three competent physicians in the institutions in which the work is to be done. This will not necessarily be the local Radioisotope Committee.

2. The subject has given his consent to the procedure.

3. There is no reasonable likelihood of producing manifest injury by the radioisotope to be employed.[37]

No further explanation was given of how the second requirement, giving consent, would be fulfilled by a "moribund" patient, nor was additional guidance provided to clarify the third criterion.

One instance in which this policy was applied took place at the Walter E. Fernald State School in Massachusetts (see chapter 7). Correspondence between the researchers and the AEC indicates that the AEC allowed the administration of 50 microcuries of calcium 45 (fifty times the amount the AEC allowed the researchers to administer to other subjects in the study) to a ten-year-old patient with a life expectancy of a few months, suffering from Hurler-Hunter syndrome (a degenerative disease of the nervous system). In applying for the radioisotope, Dr. Clemens Benda, the researcher, noted that "permission for the use of higher doses administered to moribund patients has been granted by you to other investigators . . . ."[38] This subject was part of a study of calcium metabolism approved by the superintendent of the school. Students had been described as "voluntarily participating" in a letter sent earlier to the parents asking if they objected, but that did not mention the use of radioactive tracers. Lack of response from a parent was presumed to be approval.[39] The subject with Hurler-Hunter syndrome was found to have abnormal calcium metabolism, but died before the study could be completed.[40]

Even as it developed procedures for unusual cases, the subcommittee recognized that some existing uses were becoming routine and did not need to be continuously reviewed by the subcommittee itself. The subcommittee delegated the review of such requests to the Isotopes Division, setting out the criteria to be applied:

Such applications should be justified by:

a) A commensurate increase in patient load.

b) An expanded research program.

c) Provision of adequate storage and handling facilities.

d) Assurance that personnel protection and supervision are adequate for the larger amounts requested.[41]

An additional simplification occurred with the introduction in 1951 of "general authorizations," which delegated more authority to the local radioisotope committees of approved institutions.[42] These authorizations enabled research institutions to obtain some radioisotopes for approved purposes after filing a single application each year, therefore eliminating the need to file a separate application for each radioisotope order. As such, they also reduced the oversight of the AEC's Subcommittee on Human Applications, as each order was no longer reviewed individually. However, at first the general authorizations did not apply to human use, and when they were expanded to human use in 1952, they were limited to certain radioisotopes for clinical use and excluded radioisotopes in cancer research, therapy, and diagnosis.[43]

Both the AEC and the subcommittee reacted strongly when proper bureaucratic procedures were not followed. One example was a private industrial lab that used iodine 131 for a human study that had not been properly reviewed. Even though no one was harmed, the AEC threatened to suspend shipments of all radioisotopes, not just iodine 131; such action would have put the company out of business.[44] Aebersold, at the direction of the subcommittee, notified the company president that while the incident "did not lead to any unfortunate results from the standpoint of radiation hazard . . . a recurrence of this type of violation should result in cessation of all shipment of radioactive materials to Tracerlab, Inc."[45] For his part, the company president reacted by notifying employees that such action would be grounds for automatic dismissal.[46]

Thus, as it proceeded in its work of evaluating individual applications, the subcommittee developed more general principles such as categories of human uses based upon risk and updating of criteria based upon developing knowledge. The goal, as the AEC's director of research, K. S. Pitzer, stated in 1950, was "to make radioisotopes as nearly as possible ordinary items of commerce in the technical world."[47] For example, cancer researchers initially received radioisotopes at no charge.[48] This free program was changed to an 80 percent discount program in 1952[49]for Medical, Biological, or Other Research Uses") (ACHRE No. TEX-101294-A-4), 1. and ended in July 1961. [50]

AEC Regulations and Published Guidelines

This greater precision can be seen, for example, in the guidelines for terminal patients. By the time of the 1956 guide, the use of radioisotopes with half-lives greater than thirty days ordinarily would not be permitted without prior animal studies establishing metabolic properties, unless patients had a short life expectancy. The judgment of local physicians was now to be guided by a more exact definition: exceptions would be "limited to patients suffering from diseased conditions of such a nature (life expectancy of one year or less) that there is no reasonable probability of the radioactivity employed producing manifest injury." [53] However, while a more precise definition of terminal was now provided, there was no longer explicit mention of a specific requirement for consent from these patient subjects, as had been made earlier.

Consent was required, though, in the section of the 1956 guide on the "use of radioisotopes in normal subjects for experimental purposes." (Presumably, "normal" here means "healthy.") This section included the earlier provisions that the tracer dose not exceed the permissible body burden and that such experiments not normally be conducted on infants or pregnant women. It also, however, included a new provision that such experiments were to be limited to "volunteers to whom the intent of the study and the effects of radiation have been outlined." [54] The term volunteer would seem to imply a requirement that consent be obtained following a disclosure of information to potential subjects. The disclosure requirement does not include, however, all of the elements of information that today are included in duties to obtain informed consent.

This 1956 consent requirement now governed all radioisotope experiments in normal subjects, a substantial expansion of the earlier requirement of consent only from terminal patients receiving larger-than-usual doses. It also explicitly required that both the purpose and effects of radiation be explained. It is unclear whether the failure to mention consent in the section on terminal patients was an oversight in drafting or a deliberate distinction between patients and "normal" subjects. The Advisory Committee has not found documents revealing the history of this provision, nor any explanation of the choice to limit the broad consent requirement to "normal" subjects. [55]

This broad requirement continued over the next decade as part of AEC policy. In 1965, the AEC published the "Guide for the Preparation of Applications for the Medical Use of Radioisotopes." The guide described the application process and specific policies for the "Non-Routine Medical Uses of Byproduct Material." This policy statement reiterated the exclusion of pregnant women and required that subject characteristics and selection criteria be clearly delineated in the application. Another requirement stated that applications should include "confirmation that consent of human subjects, or their representatives, will be obtained to participate in the investigation except where this is not feasible or in the investigator's professional judgment, is contrary to the best interests of the subjects." [56]

During the 1960s, the entire system of oversight of radioisotope research began to change as the Food and Drug Administration began developing a more active role in supervising the development of radiopharmaceuticals.[57] The regulatory history of this shift in authority is complex and beyond the scope of this report. Suffice it to say that by the mid-1960s the regulation of radioisotope research was beginning to merge with the regulation of pharmaceutical research in general.