The fifth plot of the Carcinogenic Potency Database: results of animal bioassays published in the general literature through 1988 and by the National Toxicology Program through 1989.

This paper is the fifth plot of the Carcinogenic Potency Database (CPDB) that first appeared in this journal in 1984 (1-5). We report here results of carcinogenesis bioassays published in the general literature between January 1987 and December 1988, and in technical reports of the National Toxicology Program between July 1987 and December 1989. This supplement includes results of 412 long-term, chronic experiments of 147 test compounds and reports the same information about each experiment in the same plot format as the earlier papers: the species and strain of test animal, the route and duration of compound administration, dose level and other aspects of experimental protocol, histopathology and tumor incidence, TD50 (carcinogenic potency) and its statistical significance, dose response, author's opinion about carcinogenicity, and literature citation. We refer the reader to the 1984 publications (1,5,6) for a guide to the plot of the database, a complete description of the numerical index of carcinogenic potency, and a discussion of the sources of data, the rationale for the inclusion of particular experiments and particular target sites, and the conventions adopted in summarizing the literature. The five plots of the database are to be used together, as results of individual experiments that were published earlier are not repeated. In all, the five plots include results of 4487 experiments on 1136 chemicals. Several analyses based on the CPDB that were published earlier are described briefly, and updated results based on all five plots are given for the following earlier analyses: the most potent TD50 value by species, reproducibility of bioassay results, positivity rates, and prediction between species.(ABSTRACT TRUNCATED AT 250 WORDS)

This paper is the fifth plot ofthe Carcinogenic Potency Database (CPDB) that first appeared in thisjournal in 1984 (1)(2)(3)(4)(5). We report here results ofcarcinogenesis bioassays published in the general literature between January 1987 and December 1988, and in technical reports of the National Toxicology Program between July 1987 and December 1989. This supplement includes results of412 long-term, chronic experiments of 147 test compounds and reports the same information about each experiment in the same plot format as the earlier papers: the species and strain of test animal, the route and duration of compound administration, dose level and other aspects of experimental protocol, histopathology and tumor incidence, TDs. (carcinogenic potency) and its statistial significance, dose response, author's opinion about carcinogenicity, and literature citation. We refer the reader to the 1984 publications (1,5,6) for a guide to the plot of the database, a complete description of the numerical index ofcarcinogenic potency, and a discussion ofthe sources of data, the rationale for the inclusion of particular experiments and particular target sites, and the conventions adopted in summarizing the literature. The five plots of the database are to be used together, as results ofindividual experiments that were published earlier are not repeated. In all, the five plots include results of 4487 experiments on 1136 chemicals. Several analyses based on the CPDB that were published earlier are described briefly, and updated results based on all five plots are given for the following earlier analyses: the most potent TDs. value by species, reproducibility of bioassay results, positivity rates, and prediction between species. A new feature of this supplement is that Appendix 14 now provides a summary compendium of positivity and potency, as well as an index to all chemicals in the five plots of the CPDB. It provides the following summary data for each chemical: (a) whether it has been tested in each sex ofrats and mice, and positivity results in each group; (b) for positive chemicals, a summary ofcarcinogenic potency for rats and for mice; (c) an index to the CPDB sorted by chemical name that reports synonyms, CAS number, and the plot numbers that include experiments on the chemical. For readers using the CPDB more extensively, a combined plot of all results from the five separate plot papers, ordered alphabetically by chemical is available from the first author in printed form or on computer tape or diskette. A SAS database is also available.

Background
The Carcinogenic Potency Database (CPDB) is a widely used, standardized resource of results of chronic, long-term carcinogenesis bioassays. The CPDB has been published in plot format in this and four earlier papers. To facilitate its use by other researchers, we have prepared a printed version of a combined plot that merges results from all five plots organized by chemical, as well as a computer-readable (SAS) database. These are obtainable from the first author.
In this paper we a) briefly describe the CPDB and the plot included in this fourth supplement; b) refer the reader to our earlier papers; c) update some of the earlier published findings using results from all five plots; and d) report errata to earlier papers.
Our goal in developing the CPDB over the past 12 years has been to provide a single, standardized and easily accessible resource that includes sufficient information on each experiment to permit investigations in many research areas of carcinogenesis. Therefore, the CPDB provides both qualitative and quantitative information on positive and negative tests (1)(2)(3)(4), including for each experiment, the species, strain, and sex of test animal; features of experimental protocol such as route of administration, duration of dosing, dose level(s) in mg/kg body weight/day, and duration of experiment; histopathology and tumor incidence; carcinogenic potency and its statistical significance; shape of the dose-response curve; author's opinion as to carcinogenicity; and literature citation. All experiments in the CPDB meet a specific set of inclusion criteria that are designed to permit the estimation of carcinogenic potency; therefore, reasonable consistency of experimental protocols is assured. Rodent bioassays are included in the database only ifthe test agent was administered alone, rather than in combination with other substances; ifthe bioassay included a control group; ifthe route of administration was diet, water, gavage, inhalation, IV injection or IP injection; and if the length of experiment was at least I year with dosing for at least 6 months. Many cancer tests do not meet these rules and are not included, e.g., ifroute ofadministration was skin painting or SC injection, or if dosing was not chronic. We do not evaluate whether the results in each experiment provide evidence for carcinogenicity; rather, we report the published opinions of the investigators and the statistical significance of the dose response. The CPDB includes results of all NCI/NTP technical reports published through 1989, with a few exceptions where the chemicals were particulates or the route of administration was skin painting.
A detailed guide to the plot of the database was included in the first published plot in 1984 (1); it described the contents, field by field, and discussed the sources ofdata, the criteria for the inclusion of particular experiments and particular target sites, and the conventions adopted in summarizing the literature. It is our intention that readers who are not familiar with the CPDB will first read the 1984 paper when using the plot in this paper. The TD5o, our numerical index of carcinogenic potency, has been fully described (1,5,6) and may be briefly defined as follows: Fora given target site(s), if there are no tumors in control animals, then TDso is the chronic dose rate in mg/kg body weight/day that would induce tumors in half the test animals at the end of a standard lifespan for the species. Because the tumor(s) of interest often does occur in control animals, TD5o is more precisely defined as the chronic dose rate that will halve the probability of remaining tumor-free throughout the standard life span. One reason for choosing TD5o is that it is easy to understand the concept, particularly because of the analogy to LD5o. Importantly, TD5o is often within the range of doses tested; thus the experimental results do not have to be extrapolated far to estimate TD5o. The TD5o does not indicate anything about carcinogenic effects at low doses because carcinogenesis bioassays are generally conducted at doses at or near the maximum tolerated dose (MTD). In the CPDB, the range of statistically significant TDso values for chemicals that are carcinogenic in rodents is more than 10 millionfold (1).
A new compendium has been prepared for this paper in Appendix 14, which includes summary evaluations of positivity and carcinogenic potency in rats and mice for each chemical in all five plots of the CPDB. This tabulation can be used to investigate associations between rodent potency and other factors such as mutagenicity, teratogenicity, chemical structure, and human, exposure, as well as to obtain summary information on individual compounds. Methods are described in Appendix 14, and are the same as were used in our earlier publication (7). Appendix 14 lists alphabetically the 1136 chemicals that appear in any of the five plots and indicates which plot includes results ofexperiments on each chemical. It also lists CAS (Chemical Abstracts Service registry) numbers and common synonyms. In this Appendix 14, four columns have been added that summarize for each chemical whether there are tests in the CPDB in male rats, female rats, male mice, and female mice; for each group we report the strongest level ofevidence for carcinogenicity as defined by the opinion ofthe published author. For chemicals that are classified as positive, two columns report the most potent TDso value in each species that has a positive test in the CPDB. Several footnotes in Appendix 14 give additional information about individual chemicals: e.g., that there is more than one positive test in the species (footnote a); that the TD50 values from different positive experiments ofthe chemical vary by more than 10-fold from one another (footnote f); and that the CPDB includes results in a species other than rats or mice and at least one test is positive (footnote g) or that none are positive (footnote h). In each ofthe five plot papers, Appendices 1-13 are in the same format and provide information for the data in that publication. In this paper, Appendices 1-13 apply only to the plot presented here. Appendix 1 lists alphabetically the compounds included in the current plot, their common synonyms, and Chemical Abstracts Service (CAS) registry number; Appendix 2 provides a list of those same compounds ordered by CAS number. The next several appendices provide codes and definitions required for using the plot: strains oftest animal (Appendix 3); routes of administration (Appendix 4); sites oftumor induction (Appendix 5); histopathology (Appendix 6); notecodes (Appendix 7); dose-response curve symbols (Appendix 8); reference codes (Appendix 9); NCI/NTPbioassays evaluated as inadequate (Appendix 10); and author's opinion codes (Appendix 11). Appendices 12 and 13 give full bibliographic information for all experiments reported in this plot: a bibliography for the general literature (Appendix 12); and a list of the NTP technical reports (Appendix 13).

Plot in this Supplement
This fifth plot ofthe CPDB includes results of412 long-term, chronic experiments on 147 chemicals. It reports results for 47 compounds from technical reports ofthe NTP published between July 1987 and December 1989, and results for 101 compounds published in the general literature between January 1987 and December 1988. Experiments in rats, mice, and hamsters are reported here for compounds representing a variety ofchemical classes and a variety of uses. Some are naturally occurring substances (e.g., catechol, 8-methoxypsoralen, and malonaldehyde); food additives (e.g., potassium bromate and geranyl acetate); industrial chemicals (e.g., 1,3 butadiene, styrene, and pentachlorophenol); and drugs (e.g., ciprofibrate, salbutamol, and diphenhydramine * HCI). Sixty-four ofthe 147 chemicals in this plot were also included in an earlier plot, and we have flagged these names in this plot with a triple asterisk (***). For some substances, only a few experiments are reported here, but several experiments were reported in earlier plots (e.g., benzene and formaldehyde). The TDsovalues for the compounds in this plot fall within the 10 million-fold range reported earlier.
Overview and Update of Our Papers That Use the CPDB The CPDB is exhaustive in that it includes all published tests that meet a set of experimental criteria. There is great diversity in the testing of chemicals reported in the database; while most chemicals have been tested in rats or mice, some have been tested in hamsters, dogs, or monkeys. Experiments with 101 different mouse strains and 74 rat strains are included. For a given chemical, the database may contain only a single experiment or several experiments. For example, among the 857 chemicals tested in rats, 29% have only one rat test and 53% have two tests; however, 15 chemicals have more than 10 tests.
Our group has used the CPDB to address many issues relevant to chemical carcinogenesis and interspecies extrapolation. Below we refer the reader to the appropriate papers. Additionally, because the CPDB now includes many more tests and chemicals than were used in the earlier papers, we have updated several of the original tables from our earlier analyses. Specifically, updated results are reported for the proportion of chemicals that are positive for several datasets, the association between mutagenicity and carcinogenicity, prediction of positivity between species, reproducibility of results in "near-replicate" experiments, carcinogen identification on the basis of two versus four sex-species groups. In each case, the updated findings are similar to those reported earlier, and we refer the reader to the earlier papers for methods and discussion.

Carcinogenic Potency (TD50)
With respect to the measurement ofcarcinogenic potency, two methods for estimating TDso from animal bioassays were compared, one based on lifetable data and one based on summary incidence data (8). There is substantial agreement between these two methods ofanalysis. Second, we have shown that the potency calculated from experimental results (given the usual experimental design and the lack of 100% tumor incidence in dosed animals) is restricted to an approximately 30-fold range surrounding the maximum dose tested in a standard bioassay (9). Third, correlation studies have been conducted ofcarcinogenic potency between rats and mice (9) and of mutagenic and carcinogenic potencies (10). Fourth, we have shown that, with few exceptions, among chemicals that are positive in more than one test in a species, the most potent TD5o value from among all positive tests is similar to other measures that average TD5o values [harmonic mean, geometric mean, or arithmetic mean] (7). Using the most potent TD5o in rats and in mice, we presented a concise tabulation of TD5ovalues for positive chemicals, which also includes a summary of positivity in each sex-species group (7). These results are updated in Appendix 14 of this paper and include results for all five plots of the CPDB.
In addition to positivity and potency, other bioassay measures of carcinogenic hazard that we have investigated are whether tumors were induced at more than one site, whether tumors may have caused the death ofthe animal or instead were found at sacrifice, and whether metastases of induced tumors occurred (11,12).

Reproducibility
Reproducibility of results in animal bioassays has been investigated in "near-replicate" comparisons consisting of two or more tests ofthe same chemical administered by the same route and using the same sex and strain of rodent (13). The updated results continue to show good reproducibility. Among 132 comparisons 86% (114/132) have concordant authors' opinions about whether tumors were induced in the individual experiments. In all but 3 of the 69 positive comparisons, at least 1 target site is identical. TDso values are within a factor of 2 of each other in 51% ofthe positive comparisons, within a factor of4 in 77 %, and within a factor of 10 in 91%.

Positivity
In several papers we have shown that approximately half the chemicals tested in rats or mice are positive in at least one test, according to the opinion of the published author. Using all data currently in the CPDB, positivity rates are reported in Table 1 separately for chemicals tested in NCI/NTP bioassays, in the general literature, and in either ofthese sources. Table 2 reports a similar positivity rate for several additional subsets of the CPDB: naturally occurring chemicals, synthetic chemicals, natural pesticides, mold toxins, and chemicals in roasted coffee. We have discussed why it is unlikely that the 50% positivity rate is due simply to selection of suspicious chemical structures (14)(15)(16), and show in Table 2 that this rate is similar for chemicals tested before 1979 by NCI and those tested later by NCI/NTP.

Mitogenesis
We have postulated that the high positivity rate is to be expected because the administration of chemicals at the maximum Literature 448/894 (50%) 321/608 (53%) 216/498 (43%) aA chemical is classified as positive if the author ofat least one published experiment has evaluated the compound as carcinogenic in that species.
'The number of chemicals in the "NCI/NTP or literature" is smaller than the sum ofeach source separately because some ofthe chemicals have been reported by both sources.  19/26 (73%) 'A chemical is classified as positive ifthe author of at least one published experiment evaluated results as evidence that the compound is carcinogenic. b94% (296/315) are tested by NCI/NTP in both rats and mice. tolerated dose (MTD) in standard animal cancer tests increases cell division (mitogenesis), which in turn increases rates of mutagenesis and thus carcinogenesis (15,17). The high rate of endogenous DNA damage contributes to the importance of mitogenesis. A variety of studies on mechanisms of carcinogenesis are consistent with this explanation (17)(18)(19). We conclude that at the low doses of most human exposures where cell killing does not occur, the hazards to humans of rodent carcinogens may be much lower than is commonly assumed. Thus, understanding the role ofmitogenesis in mutagenesis is critical for clarifying the mechanisms ofcarcinogenesis and interpreting the results of animal cancer tests (15,(17)(18)(19).

Mutagenicity
We have also examined mutagenicity rates in the CPDB (14)(15)(16) and have updated the results in Table 3. Of the 384 chemicals tested in both rats and mice and for which mutagenicity data in Salmonella are available, 72 % are either mutagens or carcinogens or both. Overall, mutagens are more often carcinogenic than nonmutagens; however 45 % of carcinogens tested in rats and mice are not mutagenic, suggesting the importance of mitogenesis in animal tests at the MTD.

Interspecies Extrapolation
The issue of exrapolating carcinogenesis results from one species to another has been addressed in analyses of prediction between two closely related species, rats and mice (12,14). We have examined how well one can predict carcinogenicity from rats to mice and from mice to rats. The updated results in Table  4 indicate that among chemicals tested in both species, 74 % of rat carcinogens are positive in mice, and 72 % of mouse carcino-   a For chemicals tested in both sexes of rats and mice that were evaluated as carcinogenic in at least one experiment.
b The total number of positive chemicals for "NCI/NTP or literature" in this table is 212, while the number in Table 4 is 28& This difference is due to the fact that 76 positive chemicals were tested in both rats and mice, but not in both sexes of rats and mice.
c Percentage indicates the proportion that would be correctly identified as carcinogens using results from experiments only in the two sex-species groups, considering as positive an evaluation of carcinogenic in either sex-species group. gens are positive in rats. We earlier discussed three factors that affect the accuracy ofprediction: chemical class, mutagenicity, and the dose level at which a chemical is toxic (14).

TIrget Organ
We have presented a compendium ofbioassay results organized by target organ for chemicals that are carcinogenic in at least one species. This compendium reports on 35 target sites and can be used to identify chemicals that induce tumors at particular sites and to determine whether target sites are the same for each chemical that is positive in more than one species (12). Sitespecific prediction between rats and mice is less accurate than overall prediction of positivity. Knowing that a chemical is positive at any site in one species gives about a 50% chance that it will be positive at the same site in the other species. Among chemicals with a target site in common between rats and mice, the liver is the most frequent site in common (12). Because the liver is the most common site in both species, we have studied liver carcinogenesis in detail (11,12,14).

Carcinogen Identification by Tlwo Versus Four Sex-Species Groups
We have also addressed the question of how many rodent carcinogens currently identified by performing tests in four sex-' species groups would be identified iftests were conducted in only two sex-species groups. The updated results in Table S continue to show that few carcinogens would be missed by testing one sex ofeach species. The greatest number (91-92%) would have been identified by conducting tests only in male rats and male mice/or in male rats and female mice.

Chemicals Selected for Testing
The natural world makes up the vast bulk of chemicals that humans consume each day in both weight and number. Yet, the natural chemicals have never been tested systematically; synthetic chemicals account for 79% (378/479) ofthe chemicals adequately tested in both rats and mice ( of natural chemicals and half of natural pesticides are positive in animal tests (Table 2), we conclude that our diet is filled with rodent carcinogens as defined by high-dose tests. We have described the concentrations in common foods of natural pesticides that are rodent carcinogens (16). Additionally, we have discussed the toxicological significance ofexposures to synthetic chemicals in the context of exposures to naturally occurring chemicals, and we argue that animals have a broad array of inducible general defenses that at low dose are effective against both natural and synthetic toxins (16,20,21). The relatively high and widespread exposure to natural chemicals that are rodent carcinogens, and the 50% positivity rate among natural chemicals that have been tested, indicate that cancer-prevention strategies aimed at chemical carcinogens need to take a broad overview of chemicals, whether synthetic or natural.

Ranking Possible Carcinogenic Hazards
We have proposed a rough index of possible carcinogenic hazard to humans, HERP (Human Exposure Rodent Potency). HERP compares for a given chemical the chronic dose rate at which humans are exposed (mg/kg/day) to the TD50 (mg/kg/day) in rodents. To put possible carcinogenic hazards in perspective, we have used the HERP index to rank a variety ofman-made and naturally occurring chemical exposures to humans (22). In a separate analysis a similar index, PERP (Permitted Exposure Rodent Potency) was calculated by using the U.S. Occupational Safety and Health Administration Permitted Exposure Limit (OSHA PEL), and assuming a daily worklife exposure at that limit (23). Permitted worker exposure levels for several rodent carcinogens are close to the dose rate that induces tumors in half the test animals. For high occupational exposures, comparatively little extrapolation is required from the doses used in rodent bioassays, and therefore assumptions about extrapolation from high to low dose are less important.

Errata in Earlier Plots
Some errors and additional information about results reported in earlier plots ofthe CPDB (1)(2)(3)(4) have come to our attention. For two NCI/NTP bioassays in the second plot (2), the route of administration was reported incorrectly: for cytembena the route was reported as diet and should have been IP injection; for vinylidene chloride the reported route was diet and should have been gavage. All other information including dose rates and TDso values was reported correctly for these two chemicals.
Issues related to the purity of test compounds have resulted in some name changes. For two NCI/NTP bioassays, we have added "technical grade" to the chemical name in the CPDB (Appendix 14) because of impurities in the test agent: 1,1,1-trichloroethane and trifluralin. For 2,3,4,5,6-pentachlorophenol two papers reported results for the technical grade (cited in the plot as Innes et al., 1968al., /1969al., and Schwetz et al., 1975 . We now report 2,3,4,5,6-pentachlorophenol (Dowicide EC-7) in Appendix 14 for these papers. The chemical name for the paper of Boberg et al., 1983, remains 2,3,4,5,6-pentachlorophenol. Two  CAS numbers have been changed for a few chemicals. The CAS number for sodium hypochlorite phosphate has been changed by the American Chemical Society to 11084-85-8. The corrected CAS number for DL-a-tocopheryl acetate is 58-95-7, and the corrected synonym is vitamin E acetate. The corrected CAS number for 1,2-di-N-butylhydrazine * 2HCl is 78776-28-0. The corrected CAS number for propanolol *HCl is 525-66-6.
For several NTP chemicals in the third plot listed below, evaluations for some target tissues were reported as "a," and we have re-assigned the evaluation as "p" indicating the NTP category, "some evidence of carcinogenicity," which NTP defines as a positive category. Our earlier assignment of "a" had been made before current descriptions of the evaluation categories were published by NTP. These sites with opinion "p"" should be interpreted as positive. For the chemicals in italics, "p" is the highest level ofevidence in the NTP evaluation: benzene, HC Blue No. 1, chlorobenzene, chlorodibromomethane, decabromodiphenyl oxide, dimethyl morpholinophosphoramidate, isophorone, methylene chloride, 1,2-propylene oxide, Telone II, tetrachloroethylene, tris(2-ethylhexyl)phosphate.
For bisphenol A in female mice, the dose used was incorrect.
Therefore, other values such as the TD5o were also incorrect. The corrected plot for this experiment is shown in Figure 1.
There are many people who have provided us with valuable assistance in the course ofour work. We thank Jerrold Ward and Leslie Bernstein for their advice on pathology and statistics and Bonnie Stern for suggestions on the manuscript. We also thank Leah Slyder for technical assistance.        Appendix 14 is both an index to chemicals in the CPDB and a tabular compilation of results on positivity and potency in rats and mice. Chemical names and common synonyms are listed alphabetically for the 1136 chemicals in the database, Chemical Abstracts Service registry (CAS) number is reported, and the plots that include experimental results on the chemical are listed by plot number.
Positivity. For each chemical, a result is reported in male rats (MR), female rats (FR), male mice (MM), and female mice (FM). Ifthere is no experiment in the CPDB for that sex-species group, this is indicated by "NT." When all four sex-species groups are NT, the chemical was tested only in a species other than rats or mice (see footnotes "g" and "h" below). The classification of positivity is based on a positive result in at least one experiment, and we classify an experiment as either positive or negative on the basis of the author's opinion in the published paper. We use the author's opinion to determine positivity because it often takes into account more information than statistical significance alone, such as historical control rates for particular sites, survival and latency, and/or dose response. Generally, this designation by author's opinion corresponds well with the results of statistical tests for the significance of the dose-response effect. The strongest level of evidence of carcinogenicity in any experiment in the sex-species group is reported in Appendix 14 for each chemical. We indicate whether the compound was tested in each group and list the strongest level of evidence for carcinogenicity based upon any author's evaluation in either the general literature or the NCI/NTP. In the general literature, a (+) indicates a positive author's opinion, and a (-) indicates either that "no opinion" was reported for this experiment or that the opinion was negative. In the NCI/NTP the strongest evaluation is clear evidence of carcinogenicity (+). When there was no such evaluation in one of the sex-species groups, but the compound was tested by NCI/NTP and their evaluation was stronger than "no evidence of carcinogenicity" (-), we indicate whether that NCI/NTP evaluation was "some evidence of carcinogenicity" (P), " equivocal" (E) or "inadequate bioassay" (I). For older NCI/NTP tests the evaluation (A) indicates "associated with carcinogenicity," and we do not interpret this as positive. These evaluations correspond to the opinions reported in our published plots. The abbreviations for positivity in Appendix 14 are as follows: NT = No Test in the CPDB in this group + = The CPDB contains at least one experiment in which the compound was evaluated as a carcinogen by the published author. For NCIINTP tests, the evaluation was "clear evidence of carcinogenicity." P = The strongest level of evidence in the CPDB was an NTP evaluation of "some evidence ofcarcinogenicity." I = No tests in the CPDB in this sex-species group were evaluated as positive; however, the NCI/NTP test was evaluated as inadequate. A = The strongest level of evidence in the CPDB was an NCI/NTP evaluation of "associated with carcinogenicity." E = The strongest level of evidence in the CPDB was an NTP evaluation of "equivocal." -= All tests in this group were negative. B+ = In the only positive test in the sex-species, results were reported only for males and females combined. B-= In the only test in the sex-species, results were reported only for males and females combined, and the test was negative.
Carcinogenic Potency. For the purposes of Appendix 14, TD50 values for a chemical are reported only for a species with a positive evaluation ofcarcinogenicity in at least one test. In any given positive experiment we select the lowest TD50 value from among positively evaluated target sites with a statistically significant dose response (two-tailed p < 0.1). If no positive sites have a significant dose response, then we select the most potent (lowest TD50 from among positively evaluated sites withp > 0.1. This method provides a single TD50 to represent an experiment.