We have previously reported the profile of target organs (defined as organs showing histopathological changes) in rodent and non-rodent toxicity studies conducted prior to first-time-in-man (FTiM) for 77 AstraZeneca candidate drugs (CDs). Here, we test the assumption that toxicity is exacerbated by dosing duration by comparing the incidence and severity of target organ toxicities in these ≤6 week FTiM studies with those observed in subsequent subchronic/chronic (≥3 month) studies. Looking at the effect of dosing duration on severity (pathological score) and incidence (percentage of animals within the group) for the 39 CDs that met the criteria for inclusion (comparable doses between FTiM and subchronic/chronic studies), new toxicities appeared for 31 target organs but existing ones resolved for 29 target organs. Increased severity was more frequent for rodent (16 target organs) than for non-rodent (4 target organs). Most notable changes were a large increase in severity/incidence in liver and in non-rodent lung in contrast to a large decrease in severity and incidence for kidneys/ureter and for the non-rodent thymus. Overall this analysis shows that, even with continued exposure, target organ toxicities of CDs are as likely to show partial or complete recovery as they are to progress in severity.

We have previously reported the profile of toxic effects with respect to target organs (defined as organs showing histopathological changes) observed in rodent and non-rodent toxicity studies conducted prior to first time in man (FTIM) for 77 AstraZeneca candidate drugs (CDs) across a range of therapy areas. The main objectives of the current study were twofold; to determine which target organs observed in the FTIM studies recovered after a dose free recovery period and to determine which additional target organs were observed in subsequent chronic (⩾3month) studies required to support longer term clinical dosing. The analysis showed that ⩾86% of findings in studies supporting FTIM either fully or partially resolved at the end of the recovery period, with profiles of recovery that were similar whether the CD progressed into man or not and across different therapy areas. Compared to observations in FTIM studies, chronic studies identified toxicities in an additional 39% of target organs. Overall these data demonstrate that chronic studies in both rodents and non-rodents provide valuable information for the risk assessment for longer term dosing in humans. In addition, the high levels of recovery demonstrated in this analysis suggest that inclusion of recovery assessments on FTIM studies should be on a case-by-case basis driven by a positive indication of need. This is in line with ICH non-clinical guidance that states that reversibility of severe nonclinical toxicities of potential clinic relevance should be assessed 'when appropriate', but that the evaluation can be based on a study of reversibility or on a scientific assessment.

An analysis of target organ toxicities in first time in man (FTiM) toxicity studies for 77 AstraZeneca candidate drugs (CDs) was conducted across a range of therapy areas. In the rodent, the most frequently affected organ was the liver followed by adrenal glands, kidney, spleen, bone marrow and thymus. In non-rodent, liver and thymus were the most frequently affected organs, followed closely by the testis and GI tract. The profile of affected organs was largely similar across the therapy areas of respiratory and inflammation, cardiovascular/gastrointestinal and CNS/pain. The oncology/infection therapy area differed with a larger range of organs affected. For the 75 CDs for which both rodent and non-rodent studies were conducted, new target organs were identified in non-rodents for 43 of the CDs. Notably, the changes seen only in non-rodents included organ systems of high relevance for human risk assessment such as the liver, male reproductive tissues and CNS. Additionally, profiles were similar for those CDs that progressed into human trials and those that did not. Overall, our data provide new insights into drug toxicity profiles in pre-clinical species and additionally confirm the value of using non-rodents as a second species in toxicity testing to support human safety.

Tertiary-Butyl alcohol (TBA), tertiary-butyl acetate™ (TBAc™) and methyl tertiary-butyl ether (MTBE) are chemicals to which the general public may be exposed either directly or as a result of their metabolism. There is little evidence that they are genotoxic; however, an earlier publication reported that significant results were obtained in Salmonella typhimurium TA102 mutagenicity tests with both TBA and MTBE. We now present results of testing these chemicals and TBAc™ against S. typhimurium strains in two laboratories. The emphasis was placed on testing with S. typhimurium TA102 and the use of both dimethyl sulphoxide and water as vehicles. Dose levels up to 5000μg/plate were used and incubations were conducted in both the presence and absence of liver S9 prepared from male rats treated with either Arochlor 1254 or phenobarbital-β-naphthoflavone. The experiments were replicated, but in none of them was a significant mutagenic response observed, thus the current evidence indicates the TBA, TBAc™ and MTBE are not mutagenic in bacteria.

The human and rabbit teratogen thalidomide has been tested for mutagenicity in a wide range of assays, ranging from bacterial gene mutation assays conducted in vitro to in vivo cytogenetic assays conducted using rabbits, and including a variety of human-derived tissues. Thalidomide was not mutagenic to 6 strains of Salmonella when tested both in the presence and absence of Aroclor-induced rat liver S9 mix. This inactivity was confirmed in strains TA98 and TA100 using a 1-h pre-incubation assay protocol with the same S9 mix (10% S9), and additionally, in strain TA98 using 3 concentrations of S9 (4%, 10% and 30% S9 in S9 mix). Thalidomide was not clastogenic either to cultured human lymphocytes (whole blood cultures, minus S9 mix) or to Chinese hamster ovary (CHO) cells treated in vitro. Further, no cytotoxicity was observed in purified human lymphocytes when exposed to thalidomide up to the limit of its solubility in the medium in the presence and absence of liver S9 from Aroclor-induced pregnant rabbit. The CHO assays were conducted without metabolic activation and in the presence of a variety of sources of auxiliary metabolic activation (PB/βNP-induced rat liver S9 mix, pooled male and female human liver S9 mix, uninduced and Aroclor-induced pregnant rabbit liver S9 mix and foetal rabbit S9 mix). Thalidomide did not induce micronuclei in isolated human lymphocytes (minus S9 mix) and it was non-mutagenic to mouse lymphoma L5178Y TK+/− cells when tested to the limits of its solubility in the culture medium (±S9 mix). No indication of recombinogenic or clastogenic activity was observed for thalidomide when tested in Drosophila. In addition, it failed to induce chromosome aberrations in grasshopper neuroblasts when tested in the presence and absence of Aroclor-induced rat liver S9 mix. Some unusual chromosome morphologies were observed in the grasshopper cytogenetic preparations indicating a potential of thalidomide to interact with chromosomal proteins. However, this potential was not evident in the human lymphocyte micronucleus assay, and thalidomide was apparently not reactive to the proteins of the mouse skin, as it gave negative results in a mouse local lymph node assay for skin sensitizing agents. Thalidomide was inactive in bone marrow micronucleus assays conducted using males and females from two strains of mice, and female New Zealand white rabbits. It is concluded that thalidomide is neither a mutagen nor an aneugen. This conclusion is discussed within the context of the results of earlier mutagenicity studies, the recent claim that thalidomide may be a heritable germ cell mutagen to humans, and the current interest in thalidomide for the treatment of immune system-related diseases.

A series of bacterial mutation, mammalian cell (L5178Y) gene mutation and in vitro cytogenetic assays were performed to compare the efficacy of using S9 fractions prepared from rats induced with a combination of phenobarbital (PB) and beta-naphthoflavone (beta NF), with S9 fractions from rats treated with the general enzyme inducer Aroclor 1254. Although some quantitative differences in the magnitudes of the mutagenic/clastogenic effects were observed between the two induction regimes, no qualitative differences were observed. The use of a combined PB/beta NF induction regime using oral dosing is therefore considered to be a suitable substitute for Aroclor 1254.

The rodent liver carcinogen and hepatic peroxisome proliferator methylclofenapate (MCP) has been evaluated for genetic toxicity in a range of in vitro and rodent genotoxicity assays. It gave a negative response in each of the following assays: mutagenicity to S.typhimurium and E.coli (± S9 mix, plate and pre-incubation assays), clastogenicity to cultured human lymphocytes and CHO cells (± S9 mix), a mouse bone marrow micronucleus assay (24h and 48h sampling), a rat liver assay for UDS in vivo (12h sampling), assays for lac I (Big Blue™) and lac Z (Muta™ Mouse) mutations in the liver of transgenic mice, and an assay of the ability of MCP to modify the mutagenicity to the liver of dimethylnitrosamine in both transgenic mutation assays. The micronucleus and UDS assays were conducted using a single administration of MCP at its maximum tolerated dose, while the transgenic assays were conducted using nine daily administrations of MCP at its cancer bioassay dose level. These nine daily administrations were shown to double the weight of the liver of non-transgenic, Big Blue™ and Muta™ Mice, as well as leading to a dramatic proliferation of peroxisomes (electron microscopy) in the livers of each strain. These changed parameters had returned to control levels when the mutation analyses were conducted (10 days after the final dose of MCP). Despite the liver enlargement observed following MCP administration, no evidence of mitotic activity was observed in treated livers, although an increased number of cells were undergoing replicative DNA synthesis during the final 3 days of the 9 days of administration (BUdR assessment of S-phase). Liver biochemistry parameters (ALT, AST, AP, CK, GGT and albumin) were unaffected by the chronic (9 day) administration of MCP indicating an absence of hepatic toxicity, These combined observations favour a non-genotoxic mechanism of action for the hepatic carcinogenicity of MCP. The clastogenicity in vitro of the peroxisome proliferator Wyeth 14,643 has been confirmed in CHO cells, but it is noted that this chemical is more soluble than is MCP. In particular, at the highest dose level at which MCP could be tested, Wy 14,643 was also nonclastogenic.

Methylene chloride was less mutagenic in Salmonella typhimurium TA100/NG-11 (glutathione-deficient) compared to TA100, indicating that glutathione is involved in the activation of methylene chloride to a mutagen in bacteria. In rodents, the pathway of methylene chloride metabolism utilizing glutathione produces formaldehyde via a postulated S-chloromethylglutathione conjugate (GSCH(2)Cl). Formaldehyde is known to cause DNA-protein cross-links, and GSCH(2)Cl may act as a monofunctional DNA alkylator by analogy with the glutathione conjugates of 1,2-dihaloalkanes. The lack of sensitivity of Salmonella TA100 towards formaldehyde (Schmid et al., Mutagenesis, 1 (1986) No. 6, 427-431) suggests that GSCH(2)Cl is responsible for methylene chloride mutagenicity in Salmonella. In Escherichia coli K12 (AB1157), formaldehyde was mutagenic only in the wild-type, a characteristic shared with cross-linking agents, whereas 1,2-dibromoethane (1,2-DBE) was more mutagenic in uvrA cells (AB1886). Methylene chloride, activated by S9 from mouse liver, was mutagenic only in wild-type cells, suggesting a mutagenic role for metabolically derived formaldehyde in E. coli. Mouse-liver S9 also enhanced the cell-killing effect of methylene chloride in the uvrA, and a recA/uvrA double mutant (AB2480) which is very sensitive to DNA damage. This pattern was consistent with formaldehyde damage. However, a mutagenic role in bacteria for the glutathione conjugate of methylene chloride cannot be ruled out by these E. coli experiments because S9 fractions did not increase 1,2-DBE mutagenicity, suggesting lack of cell wall penetration by this reactive species. Rat-liver S9 did not activate methylene chloride to a bacterial mutagen or enhance methylene chloride-induced cell-killing, which is consistent with the carcinogenicity difference between the species.

The suspect human carcinogen, etoposide, is known to be genotoxic, producing both gene and chromosomal mutations, probably by virtue of its ability to inhibit topoisomerase II activity. The present paper describes assays conducted using the Salmonella assay, the mouse lymphoma tk+/- assay (gene and chromosomal mutation analysis and molecular analysis of tk-/- mutants) and the mouse bone marrow micronucleus assay. Nonreproducible, weak, dose-related increases in mutation frequency in strain TA98 (but not TA1538 or TA1537) of Salmonella typhimurium were observed. Etoposide was highly mutagenic at the heterozygous thymidine kinase (tk+/-) locus of L5178Y mouse lymphoma cells at concentrations below 0.1 micrograms/ml. Mostly small colony mutants were induced, consistent with the potent clastogenicity also observed. Molecular analysis of mutants indicated that 83% and 92% of large and small colony mutants, respectively, had lost the entire target gene sequence. Chromosomally aberrant L5178Y cells were approximately 2 to 600-fold more prevalent than small tk-/- mutant colonies. This suggests that the viable target for etoposide-mediated clastogenesis in the selective assay is approximately one-fifth of chromosome 11b, itself being approximately one-fortieth of the mouse genome. An unusually potent response was observed for etoposide in the mouse bone marrow micronucleus assay (63.1 +/- 18 MPE/1,000 PE 24 hours after an oral dose of 1 mg/kg). The minimum detectable dose level in the assay was between 0.01 and 0.1 mg/kg. At dose levels between 1 and 15 mg/kg, an inverse dose response was observed. This reduction in assay response was not due to the small concommitant decrease in the incidence of polychromatic erythrocytes, a conclusion based on studies with N-methyl-N-nitrosourea. Animals sampled 48 hours after dosing with etoposide (10 mg/kg) had no polychromatic erythrocytes in the bone marrow. These observations for the micronucleus assay await explanation. The chemical structure of etoposide is displayed and discussed within the context of such strong mutagenic activity being associated with a nonelectrophilic agent.

The antineoplastic agent fotemustine is shown to be a base-pair mutagen to Salmonella. Activity is more marked in the uvrB-proficient strain G46 than in the repair-deficient strain TA1535. This is consistent with its ability to cross-link DNA. Potent activity as a somatic and germ-cell mutagen to Drosophila was also observed. A potent clastogenic response was given by fotemustine in the mouse bone marrow following either oral gavage or intraperitoneal injection of a single dose of 5 mg/kg. In each of these respects it is shown to be indistinguishable from the structurally related antineoplastic agent and human carcinogen MeCCNU. It is concluded that fotemustine should be regarded as having clear potential to induce cancer in humans. Based on these data, including the preponderance of chromosome breakages over recessive lethal mutations in Drosophila, an estimated rodent carcinogenic potency (TD50) of between 15-150 mg/kg is suggested for fotemustine.

Twenty organic Salmonella mutagens, seven of which (including benzo[a]pyrene) are established skin carcinogens, and one of which (2-chloroethanol) is a well-defined noncarcinogen to skin, have been evaluated for skin-sensitizing activity using the local lymph node assay. The relative mutagenicity of the agents to Salmonella was also established. Fourteen of the chemicals were positive in the local lymph node assay, including the seven skin carcinogens. 2-Chloroethanol was inactive as a sensitizing agent. We suggest that a variety of factors contributes to the lack of sensitizing activity of the remaining six bacterial mutagens: extremes of intrinsic chemical reactivity, high water solubility reducing dermal translocation, and inappropriate dermal metabolism. Two reference skin-sensitizing agents (an oxazolinone and fluorescein isothiocyanate) were established as in vitro clastogens after their recognition as nonmutagens to Salmonella. These data imply that mutagenicity, rather than simply activity in the Salmonella assay, is a primary stimulus for electrophilic sensitization and carcinogenic initiation in the skin. We conclude that genotoxicity data for an agent can provide indications of the agent's potential to induce skin sensitization and that genotoxins which are skin-sensitizing agents have an enhanced potential to initiate skin carcinogenesis. We suggest that common, albeit individually distinct, structure-activity relationships underpin genotoxicity, skin sensitization, and the initiation of skin carcinogenesis. These relationships should simplify the hazard evaluation of chemicals and contribute to a reduction in animal usage. Several predictions of skin carcinogenicity are made based on the data presented.

The genotoxicities in vitro and in vivo of the mouse-skin carcinogen 7,12-dimethylbenz[a]anthracene (DMBA) have been compared with those of its weakly carcinogenic 4,5-sulphyr analogue, 6,11-dimethylbenzo[b]naphtho-[2,3-d]thiophene (S-DMBA). The only datasets that correlated with the relative carcinogenicity of these agents to the skin were those conducted using topically exposed mouse skin. Thus, both chemicals induced lacZ− mutations in the skin of lacZ+ transgenic mice, and both produced DNA adducts on mouse-skin DNA as assessed using the 32P-postlabeling technique. In each case, DMBA gave a stronger response than did S-DMBA. In contrast to these responses, only DMBA was active in the mouse bone-marrow micronucleus assay and in the C3H10T1/2 in vitro cell transformation assay. Both chemicals were mutagenic to Salmonella and of approximately equal potency. The molecular geometry of DMBA and S-DMBA are compared, and divergent CASE predictions of activity in the Salmonella assay and skin-painting bioassay are discussed. The importance of conducting predictive genotoxicity assays in systems close to those in which carcinogenicity is to be assessed is emphasized by these data.

The methylating agent streptozotocin is active in the mouse bone marrow micronucleus assay following a single intraperitoneal injection of 150 - 180 mg/kg. This correlates with its previously reported toxicity to mouse germ cells when administered by the same route of exposure. The potent mutagenicity of streptozotocin to strain G46 of Salmonella typhimurium is compared with its much weaker activity in strain TA1535. The genotoxicity of streptozotocin in vivo is reviewed.

Sulphur mustard is clearly mutagenic to the repair-proficient (uvrB+) strain G46 of S. typhimurium. It is also a micronucleus-inducing agent to the mouse bone marrow. These data strengthen the database indicating that most human carcinogens are genotoxic.

There is limited evidence that beryllium is a lung carcinogen to man, and several compounds of beryllium are carcinogenic to the lungs of the rat, rabbit and monkey. One such compound is beryllium sulphate (BeSO4.4H2O). This soluble salt has been evaluated in a range of genotoxicity tests. It was non-mutagenic to Salmonella typhimurium (strains TA1535, 1537, 1538, 98 and 100) when evaluated in the plate-incorporation assay at dose levels up to 5 mg/plate (+/- induced rat-liver S9 mix). It was also non-clastogenic to Chinese hamster lung (CHL) cells cultured in vitro. When dosed to male CBA mice via oral gavage at dose levels of 80% and 50% of the medium lethal dose (2.3 and 1.4 g/kg, respectively) it failed to increase the incidence of micronucleated polychromatic erythrocytes in the bone marrow (sampled at 24, 48 and 72 h post-dosing). However, a marked depression of erythropoiesis was evident 24 h after dosing suggestive of beryllium-mediated bone-marrow toxicity. When tested in the strain A mouse lung tumour bioassay, BeSO4 induced a significant increase in the number of tumour-bearing animals but not in the number of lung tumours per animal. These findings are discussed within the contexts of other genotoxicity data published for BeSO4, and of current strategies for the detection of possible human carcinogens.

Urethane is shown to be active in the mouse bone-marrow micronucleus assay when administered as a single dose by either gavage or intraperitoneal injection. The magnitude of the response using the two routes was not statistically significantly different. N,N-Dimethylurethane (DMU) is shown to be mutagenic to Salmonella and active in the bone-marrow micronucleus assay by both routes of administration. The activity of DMU in the bone marrow precludes elimination of ethanol, yielding cyanate ion, as an explanation for the micronucleus-inducing activity of urethane.

1-Chloromethylpyrene (1-CMP) has been evaluated as a model mutagen and toxin related to the ultimate electrophiles derived from benzo[a]pyrene and 1-nitropyrene. It was mutagenic to Salmonella (> 100 pg/plate) an exceptionally reactive to DNA when assessed by the 32P-postlabelling technique. 1-CMP was inactive in a mouse bone micronucleus assay when administered by gavage, probably due to hydrolysis, whose kinetics have been studied (t12 ∼ 23 min at 37°C). However, as expected, it was a potent skin toxin as determined by its activity as a mitogen to mouse skin and its contact allergenicity, as determined using the local lymph node proliferative assay. It is concluded that 1-CMP will probably be a potent human skin carcinogen and contact allergen.

N-Nitrosodibenzylamine (NDBzA) is mutagenic to Salmonella typhimurium and induces DNA strand breaks in isolated rat hepatocytes, yet it is reported to be non-carcinogenic to the rat. Here we report that it is inactive in both the rat and mouse bone marrow micronucleus assays and in a rat liver autoradiographic assay for unscheduled DNA synthesis. It is, however, clearly active as a micronucleus-inducing agent and mitogen in the rat liver and is capable of inducing single-strand breaks in the DNA of rat liver. The origin and implications of this curious conflict of in vivo genotoxicity data are discussed. Irrespective of that discussion, it is concluded that NDBzA is genotoxic to the rat liver in vivo.