Medical Device & Diagnostic Industry Magazine
MDDI Article Index

An MD&DI January 1999 Column

ISO 10993

This final installment in MD&DI's series of articles on the international biocompatibility standards discusses when and how to conduct carcinogenicity testing. Last month's installment covered Sample Preparation and Reference Materials.

THE POTENTIAL of a device material that comes in contact with a patient to cause or incite the growth of malignant cells—that is, its carcinogenicity—is among the issues addressed in the set of biocompatibility standards developed by the International Organization for Standardization (ISO). Part 3 of the ISO 10993 standards, which covers genotoxicity, carcinogenicity, and reproductive toxicity, describes carcinogenicity testing as the means "to determine the tumorigenic potential of devices, materials, and/or extracts to either a single or multiple exposures over a period of the total life-span of the test animal." The circumstances under which such an investigation may be required are indicated in ISO 10993-1, Table 2: Guidance for Supplementary Evaluation Tests. Specifically, such testing should be considered for a device that will have permanent contact (longer than 30 days) with tissues, either as an implant or as an externally communicating device. Although this definition clearly covers all permanent implants, including those that are designed to be absorbed, and extracorporeal devices that will be needed for the remainder of a patient's life, the standard further indicates that "carcinogenicity tests should be conducted only if there are suggestive data from other sources." Thus, not every implant or extracorporeal device needs to be subjected to this time-consuming and expensive testing.

TEST METHODS

Most investigators and regulators can agree on the criteria for undertaking carcinogenicity testing, but there is almost no consensus on how to actually do it. Rodents are invariably chosen as the test species because their relatively short life spans makes it practical to carry out lifetime studies. Traditionally, the effects of industrial chemicals, pesticides, food additives, and pharmaceuticals are evaluated via lifetime feeding, inhalation, or dermal application studies using two rodent species. However, there are no comparable validated in vivo or in vitro models available for testing devices or biomaterials. In addition, debate continues within the scientific community about whether two rodent species are needed and, if not, which rodent strain is preferable.

ISO 10993-3 provides some guidance by referencing the Organization for Economic Cooperation and Development (OECD) protocols 451 (Carcinogenicity Studies) and 453 (Combined Chronic Toxicity/Carcinogenicity Studies). Although these protocols were written for lifetime studies in rodents to evaluate chemicals that may be introduced into the body by means other than implantation, which is the type of exposure for many devices, they can nevertheless be used to select the key elements of a study. Such elements may include the number of test animals, the kinds of observations needed, the extent of histopathological evaluations, the number of survivors required at the end of the study, and the type of statistical evaluations that would be most meaningful. Having a basic study design in mind, a manufacturer can then construct a technically sound framework for carcinogenicity testing of a device or biomaterial.

The latest committee draft of ISO 10993-3 also cites the American Society for Testing and Materials document ASTM F 1439-92: "Performance of Lifetime Bioassay for Tumorigenic Potential of Implanted Materials." This guidance document was written specifically to address device or biomaterial evaluations, but recognizes the limitations of such efforts, stating that "the recommendations given in this guide may not be appropriate for all applications or types of implant materials." The ASTM method requires a minimum of 60 male and 60 female rodents per treatment or control group and identifies a basic study design as consisting of at least one group exposed to the test material, another group exposed to a reference material, and likely a sham surgical or vehicle dose group. Thus, the total number of animals in the study would be 360. Devices designed for use solely in male or female patients may be tested in rodents of the same gender, thus halving the number of animals required. A study should last a minimum of 18 months for mice and 24 months for rats.

The standard requires that tests be "appropriate for the route and duration of exposure or contact," which raises the issue of how to expose the test animals to the test article. For an extracorporeal device that will be in direct or indirect contact with blood, the testing may be conducted by injecting extracts of the device. In such cases, it is imperative to determine what is in the extract, which extraction vehicle or vehicles best represent the expected human exposure, and what the dose should be and how often and by what route it should be given. The goals are to mimic human exposure and to exaggerate that exposure on a milligram per kilogram of body weight basis. The decisions made on these matters should be worked out in concert with the FDA personnel that will review the device prior to marketing.

Devices that will be implanted present all of the challenges indicated above, and more. Virtually all solid materials with an extensive surface area cause what are known as solid-state tumors in rodents when implanted for long periods of time. The tumorigenic effect is due to the size and shape of the implant, not to leachable chemicals. (This phenomenon is also referred to as the Oppenheimer effect.) Investigators designing long-term implant studies must find ways to work around this effect or ensure that the number of animals in the study is sufficient for pathologists to distinguish solid-state from chemically induced tumors. Whenever possible, of course, the device should be implanted in the rodent body in an anatomic location that simulates clinical use.

Because the purpose of carcinogenicity studies is to evaluate the tumorigenic effects of lifetime exposure to a device or its extracts, the condition of tissues as revealed by both gross and microscopic examination is the most important end point to evaluate. A full complement of tissues, up to 40 per animal, must be harvested and preserved for examination by a pathologist. Such evaluations are both time-consuming and expensive, however, and the extent of the histopathology necessary remains a debatable issue.

A WORD TO THE WISE

Obviously manufacturers must take very seriously the requirement to understand the carcinogenic potential of their medical devices. However, before beginning these costly, time-consuming studies, it is good practice to conduct preliminary investigations. One should explore the chemistry of the device materials with special attention to their extractables, including degradation products. Information can be gathered about the absorption, distribution, metabolism, and excretion of these chemicals. In most instances, if detailed data about the carcinogenicity of the specific chemicals in the device materials are already available in the technical literature, animal studies may not be required.

CONCLUSION

This article brings to an end this multipart series summarizing the ISO 10993 biocompatibility standards. Previous installments, which appeared in MD&DI throughout 1998, covered such topics as materials characterization; tests for hemocompatibility, cytotoxicity, sensitization and irritation, and systemic toxicity; and sample preparation. Readers are encouraged to consult those articles and to read the standards for complete details.

The standards are available in the United States from the Association for the Advancement of Medical Instrumentation, 3330 Washington Blvd., Ste. 400, Arlington, VA 22201-4598; phone 703/525-4890, fax 703/276-0793; http://www.aami.org.

Paul J. Upman, PhD, is a senior scientist and Richard F. Wallin, DVM, PhD, is president of NAMSA (Northwood, OH).

Copyright ©1999 Medical Device & Diagnostic Industry