By Matthew J. Palazzolo, PhD
For more than a decade, concern has been mounting over the declining survival of rats used in carcinogenicity studies, as well as over gains in average weight and an increased incidence of spontaneous cancer and other diseases in these animals. Efforts to understand the causes of the drop in survival, and to reduce its impact on studies evaluating the safety of drugs and other compounds, are under way in a number of laboratories (see sidebar below).
A variety of studies have been and are being conducted in laboratories throughout the U.S. to determine the longevity and incidence of tumors for various strains of rodents used in carcinogenicity bioassays. In a 2-year study under way at Corning Hazleton, 50 animals per sex of each of four strains of rats are being housed under identical conditions to determine comparative survival, incidence of tumors, frequency of non-neoplastic lesions, food consumption, and body weight fluctuations. Halfway through the study, the CD group ("fat rats") shows an increase in food consumption and body weight over the SD group. Other changes being monitored are ocular opacities in the Fischer 344 rats and detectable masses, especially in the CD and SD females.
The standard carcinogenicity bioassay includes 50 animals of each sex for each species. Many countries have regulations that require, or strongly recommend, a 50% survival, or 25 animals per sex per group, at the end of a 2-year study with rats if the results are to be accepted (see examples below). This target may no longer be obtainable with certain strains of laboratory rats. A number of laboratories are now reporting survival rates below 50% (see Table 1), and survival rates as low as 7% at the end of 104-week studies have been reported. As a result, some completed studies may face challenges by regulatory agencies.
Problems are reported
Diminishing life spans of CD rats were documented in the early 1980s, but the problem gained considerable attention in 1988 when the Pharmaceutical Manufacturers Association (PMA) distributed a questionnaire to member companies. Results of the questionnaire showed that 9 of 30 responding companies had encountered regulatory difficulties due to low survival rates in rodent studies.
The PMA findings were bolstered in 1989 when Charles River Laboratories published information indicating unexpectedly low survival of CD rats at the end of 2-year studies in many laboratories in the U.S. and suggesting that the problem was inherent in the CD strain.
About 5 years ago, decreasing life spans of other rat strains (such as the Fischer 344) were detected, but since these strains always have lived longer than CD rats, the drop in their survival rates did not pose an immediate regulatory concern.
Changes in breeding practices, including selection for rapid growth and reproductive performance, are the primary factors that have led to the current situation. Possible contributing influences include ad libitum feeding and individual housing of the animals. These two environmental factors may be the cause of increasing average body weight in CD rats, a circumstance that has led them to be dubbed "fat rats."
Study Year Sex Percent surviving
18 24
months months
1 1979 M 92 70
F 86 56
2 1980 M 92 63
F 91 61
3 1981 M 96 58
F 96 58
4 1981 M 96 70
F 84 60
5 1981 M 90 67
F 91 68
6 1982 M 95 67
F 91 68
7 1983 M 93 69
F 87 60
8 1984 M 93 75
F 85 53
9 1985 M 93 74
F 93 54
1979-1985 means at 24 months: males - 68%; females - 60%
The VAF variant of the CD strain was used after 1987.
10 1988 M 75 33
M 71 27
F 84 45
F 76 51
11 1989 M 80 47
M 62 27
F 69 36
F 67 35
12 1990 M 84 56
M 86 56
F 82 46
F 92 50
1988-1990 means at 24 months: males - 41%; females - 44%
Source: Nohynek et al. 1993
A variety of remedies have been proposed to the longevity dilemma, and some are being tried. Still, a definitive method of dealing with declining survivability has yet to be developed.
Restricting or controlling the animals' diet. One possible solution to the problem of "fat rats" with reduced life spans is to adjust their dietary intake. When the original guidelines on using rats in carcinogenicity studies were drawn up, the importance of eliminating impurities in the animals' diet was recognized, but no mention was made of the optimum number of calories. Recent studies have shown that restricting or controlling the diet, rather than allowing animals to eat as much as they like, can increase longevity and decrease age-related diseases, including cancer.
If diets are controlled and the incidence of cancer is thereby reduced, a new question arises: Does decreased occurrence of disease also mean decreased sensitivity of the assay to detect carcinogenicity? Other potential limitations to dietary restriction include the possibility that this solution would not be accepted by regulatory authorities and that it would increase the effort required to maintain the animals. Even if dietary restriction is successful in improving longevity, the underlying problem is almost certainly genetic, so just changing the diet may still not offer the ultimate solution.
Group housing. The decrease in longevity of CD rats has not been observed universally. Some reports indicate that no significant reduction in the life span occurs if these animals are housed in groups of three or more, rather than individually. Group housing, however, may simply be producing a behavioral food restriction.
Rederiving the CD strain . In 1993, Charles River Laboratories addressed the concern over the reduced longevity of its so-called CD rats. The CD variety, the company noted, was developed decades earlier, having been derived from an outbred stock created in the 1920s by Sprague Dawley. The CD animals of the 1990s, however, most likely have changed considerably from their ancestors. The company concludes that these changes in longevity have come about gradually since the mid-1970s. Charles River maintains over 20 colonies of CD rats around the world, and differences most likely occur from colony to colony as well as between the current and the ancestral stocks.
As a long-term solution to the longevity problem with CD rats, Charles River plans to institute "a program to modify them in such a way as to reduce selection pressures which may be linked to the reported problems in longevity." The goal, the company stresses, is not to replace the old strain with a new strain of rats, "but to maximize and stabilize the existing genetic diversity" within the CD rats.
Choosing a different strain of rat . A number of other rat strains are available for use in carcinogenicity studies, but most do not have the years of historical background data that are available for the CD rat strain. Historical data are valuable for determining, for example, the incidence of spontaneous tumors in a specific strain of rats.
The SD rat, an outbred strain originated in 1925, and the Fischer 344 rat, an inbred strain, both have better survival rates than CD rats in chronic bioassays but also have drawbacks that keep them from being more widely used in carcinogenicity studies.
Increasing the size of groups . To improve the chance of having 25 surviving CD rats per sex in each group after 104 months, the original number could be increased. The FDA is considering allowing the number of rats in control and treated groups to rise to 60 per sex. This change would increase the cost of carcinogenicity bioassays and may not satisfy the regulatory requirements of the European Union (EU).
Reducing the length of carcinogenicity bioassays . Based on statistical considerations, the FDA appears to be giving some thought to allowing a 50% survival rate among CD rats after 80 to 90 weeks of a chronic bioassay, rather than after 104 weeks. An official guidance on this point has not been issued, however. Furthermore, it is far from certain that EU and Japanese regulatory bodies are considering a similar course of action.
Regulatory directions
Representatives of the National Center for Toxicological Research, a branch of the U.S. Food and Drug Administration (FDA), have indicated recently that more emphasis will be placed on controlling and restricting the diets of rodents used in bioassays. The aim will be to reduce the variability of test results and to enhance survival in chronic studies. The FDA plan may call for two control groups, one in which the diet matches that of the test animals and another in which the diet is manipulated so that body weights remain similar to those of the dosed animals.
Harmonization
For a number of years, the FDA has been taking part in meetings designed to promote harmonization of technical requirements for drug development among regulatory agencies in various countries. The International Conference on Harmonisation (ICH) has been focusing its attention on registration requirements for pharmaceutical products in the EU, Japan, and the U.S. Eventually, the ICH may be able to produce uniform guidelines to be followed when conducting rodent bioassays for carcinogenesis. The issue of carcinogenicity bioassays is scheduled for consideration at the November 1995 meeting of the ICH in Yokohama, Japan. Currently, however, fundamental differences exist between nations, and even between different agencies within a single country, over the acceptable rate of survival among rats in chronic bioassays.
Alternatives to current bioassays
Our understanding of carcinogenesis has progressed significantly during the past half century. This new knowledge has led some researchers to conclude that the basic 2-year rodent bioassay for detecting potential carcinogens is no longer sufficient for providing the complex information needed to make decisions about risk management. Traditional rodent bioassays, critics contend, are fine for detecting overtly carcinogenic chemicals (genotoxic compounds), such as nitrosamines, but are inappropriate for nongenotoxic substances that do not necessarily cause cancer as a result of obvious genetic damage.
The need, therefore, is to devise specific methods for testing pharmaceuticals and other compounds for their potential to produce cancer by genotoxicity, immunosuppression, hormone disturbance, or chronic irritation. Methods for obtaining this information may include rodent studies of shorter duration and involving only one species. Such studies could yield toxicokinetic and mechanistic data needed to make predictions of a compound's carcinogenic potential early in the evaluation process of the chemical.
One possible change in the current bioassay methodology is to use transgenic animals, which are genetically controlled to be sensitive to genotoxic carcinogens. Transgenic mice are already being tested as models in cancer studies, although a transgenic rat model has yet to be created. The sensitivity of transgenic animals means that they will show the effects of a cancer-causing compound sooner than other animal models, and the effects of age, immune status, and hormone levels will be lessened. If and when a transgenic rat model is developed, more compounds can be tested more quickly with fewer animals and at less cost than with traditional bioassays. It will certainly be several years, however, before researchers validate transgenic models and accumulate enough historical data to make these animals attractive substitutes in carcinogenicity bioassays.
In the next several years, changes will occur in the way we test new compounds for oncogenicity. These methods will depend on our understanding of the carcinogenic processes of initiation, promotion, and progression. Animal models will become more sensitive and have well characterized genetic composition. New carcinogenic models will make use of the quantitative assessments of dose and will incorporate endpoints that reflect mechanisms that relate to the oncogenic response. Some advances have begun with the use of transgenic animal models, but we are a long way from regulatory acceptance of these new models.
Eventually, additional sources of data, including human epidemiology and in vitro studies, will play greater roles in hazard assessments. Currently, however, no other tests offer more useful results than chronic rodent bioassays. Most toxicologists agree that rodent bioassays are the only proven tool for detecting carcinogens. After all, virtually all human carcinogens also cause cancer in one or more rodent species. With careful attention paid to the genetic condition of the rodents and to their diet and housing, these animals will continue to play a prominent role in protecting humans from exposure to carcinogens for many years to come.
References
Boorman, G.A., R.A. Maronpot, and S.L. Eustis. Rodent carcinogenicity bioassay: Past, present, and future. Toxicologic Pathology 22:105-111, 1995.
Charles River Laboratories. Longevity and fertility in the CD® rat. Genetic management of the CD® rat: Minimizing inbreeding, genetic drift and colony divergence by systematic outbreeding and animal migration. 5 pages, March 9, 1993.
Food and Drug Administration (FDA). Toxicological Principles for the Safety Assessment of Direct Food Additives and Color Additives Used in Food ("Redbook II"), 1993 draft.
Food Chemical News , March 7, 1994, pp. 29-30; June 26, 1995, pp. 54-56; July 17, 1995, pp. 3-6, August 14, 1995, pp. 4-7.
Lang, P.L. Changes in life span of research animals leading to questions about validity of toxicologic studies. Chemical Regulation Reporter , January 18, 1991, pp. 1518-1520.
Lin, K.K., and M.W. Ali. Statistical review and evaluation of animal tumorigenicity studies. In: C.R. Buncher and J.-Y. Tsay (eds.), Statistics in the Pharmaceutical Industry (second edition), Marcel Dekker, pp. 19-57, 1994.
Monro, A. How useful are chronic (life-span) toxicology studies in rodents in identifying pharmaceuticals that pose a carcinogenic risk in humans? Adverse Drug Reactions and Toxicological Reviews 12:5-34, 1993.
Nohynek, G.J., et al. Fat, frail and dying young: Survival, body weight and pathology of the Charles River Sprague-Dawley-derived rat prior to and since the introduction of the VAF® variant in 1988. Human and Environmental Toxicology 12:87-98, 1993.
As director of the Toxicology department at Corning Hazleton, Dr. Matthew Palazzolo coordinates the activities of the scientific staff and works closely with business development and client services personnel. He has a PhD in physiology and pharmacology and is a Diplomate of the American Board of Toxicology. Dr. Palazzolo has been at Hazleton since 1989. He is a member of the Society of Toxicology, the American College of Toxicology, and the American Industrial Hygiene Association and has given numerous presentations on various aspects of toxicology at professional meetings.
About the author
John Wolf, editor
© Corning Hazleton Inc. 1995