Bioassay of thio-TEPA for possible carcinogenicity

A bioassay of thio-TEPA for possible carcinogenicity was conducted by administering the test chemical by intraperitoneal injection to Sprague-Dawley rats and B6C3F1 mice. Groups of 31-39 rats of each sex were administered thio-TEPA in phosphate-buffered saline at one of three doses, either 0.7, 1.4, or 2.8 mg/kg body weight, three times per week for a maximum of 52 weeks, then observed for additional periods of time. The maximum time on study (administration of chemical and observation) was 86 weeks. The groups at the low dose were started 69 weeks after those at the mid and high doses, because of high mortalities observed in the groups at the higher doses. Matched controls consisted of groups of 10 untreated rats and 10 vehicle-control rats of each sex. Pooled-control groups also were used. Surviving control rats were killed at 82-87 weeks; surviving dosed rats were killed at 81 or 82 weeks. Groups of 35 mice of each sex were administered thio-TEPA at one of two doses, either 1.15 or 2.3 mg/kg body weight, three times per week for a maximum of 52 weeks, then observed for a maximum additional period of 34 weeks. Matched controls consisted of groups of 15 untreated mice and 15 vehicle-control mice of each sex. Pooled controls also were used. Surviving control and dosed mice were killed at 86 or 87 weeks. Thio-TEPA was toxic to both rats and mice, causing decreased mean body weight gains and early deaths in the mid- and high-dose rats and in the high-dose mice. Because of the early deaths, statistical analyses were based only on time-adjusted incidences of tumors. Since all high-dose male and female rats had died by 21 weeks, microscopic evaluation of tissues was performed only on the low- and mid-dose animals. In rats, the incidence of combined neoplasms of the hematopoietic system (lymphoma, lymphocytic leukemia, or granulocytic leukemia) was significant in the males in both the low-dose (P=0.020) and mid-dose (P=0.001) groups, using pooled controls (pooled controls 0/29, low-dose 6/34; pooled controls 0/30, mid-dose 6/16). Squamous-cell carcinoma of the skin or ear canal occurred at a significant incidence in the male rats in both the low-dose (P=0.009) and mid-dose (P=0.023) groups, using pooled controls (pooled controls 0/29, low-dose 7/33; pooled controls 0/30, mid-dose 3/13) and in the mid-dose females (P<0.001), using pooled controls (pooled controls 0/28, mid-dose 8/21); in addition, two low-dose females had such tumors, with none occurring in the corresponding low-dose controls. The incidence of adenocarcinoma of the uterus was significant in the mid-dose female rats (P=0.001), using pooled controls (pooled controls 0/28, mid-dose 7/21); in addition, two low-dose females had adenocarcinoma of the uterus, with no such tumor occurring in the corresponding low-dose controls. In rats, neuroepitheliomas (neuroblastomas) or nasal carcinomas occurred in three low-dose males, two low-dose females, and two mid-dose females. Although these are not statistically significant incidences, these tumors did not occur among control animals and no such tumors have occurred in 380 Sprague-Dawley control rats of each sex in other bioassays at the same laboratory. Thus, they may be associated with administration of the chemical. In the high-dose groups of both male and female mice, but not in the low-dose groups, the incidences of lymphoma or lymphocytic leukemia were significantly higher (P<0.001) for each sex than those of either the vehicle or pooled controls (males: vehicle controls 1/8, pooled controls 1/18, low-dose 2/24, high-dose 26/28; females: vehicle controls 0/14, pooled controls 0/29, low-dose 5/26, high-dose 32/32). In the low-dose male mice squamous-cell carcinoma was found in the skin of seven animals, in the preputial glands of six animals, and in the ear canal of two animals. A carcinoma of the preputial gland was also found in a high-dose male. When the incidences of the tumors at the different sites were combined, the incidence in the low-dose group was statistically significant using either the vehicle (P=0.004) or the pooled (P<0.001) controls (vehicle controls 0/8, pooled controls 0/18, low-dose 14/24, high-dose 1/2). It is concluded that under the conditions of this bioassay, thio-TEPA was carcinogenic in both Sprague-Dawley rats and B6C3F1 mice. In the rats, the chemical induced squamous-cell carcinoma of the skin or ear canal in both males and females, and hematopoietic neoplasms in the males; in the mice, it induced lymphoma or lymphocytic leukemia in both sexes and squamous-cell carcinoma in the skin and associated glands of males.