Abstract for TR-345

Toxicology and Carcinogenesis Studies of Roxarsone in F344/N Rats and B6C3F1 Mice (Feed Studies)

CASRN: 121-19-7
Chemical Formula: C6H6As NO6
Molecular Weight: 263
Synonyms/Common Names: 4-Hydroxy-3-nitrophenylarsonic acid; 4-hydroxy-3-nitrobenzenearsonic acid; 2-nitro-1-hydroxybenzene-4-arsonic acid; nitrophenolarsonic acid; 3-nitro-4-hydroxybenzenearsonic acid; 3-nitro-4-hydroxyphenylarsonic acid
Report Date: March 1989

Full Report PDF


Roxarsone is a veterinary drug used as a growth promoter and as an anticoccidial agent and for treatment of swine dysentery. Toxicology and carcinogenesis studies were conducted by administering roxarsone (greater than 99.4% pure) in feed to groups of F344/N rats and B6C3F1 mice of each sex for 14 days, 13 weeks, or 2 years.

Fourteen-day and thirteen-week studies

In the 14-day studies, the diets fed to rats contained 0 or 100-1,600 ppm roxarsone, and those fed to mice contained 0 or 60-1,000 ppm. Deaths occurred in rats and mice that received the highest doses. Rats that received 800 or 1,600 ppm lost weight. Male mice that received 1,000 ppm and female mice that received 500 ppm lost weight.

In the first 13-week studies, roxarsone was fed to rats and mice at dietary concentrations of 0 or 50-800 ppm. Decreases (more than 10%) in final mean body weights of dosed rats relative to those of controls were observed for males that received 200, 400, or 800 ppm and for females that received 400 or 800 ppm. Deaths occurred in groups that received 800 ppm. Clinical signs of toxicity (trembling, ataxia, and pale skin) were seen primarily in rats that received 800 ppm. Kidney lesions were observed in rats that received 800 ppm. These lesions were characterized by tubular necrosis and mineralization in the rats that died during the studies and by tubular dilatation and casts, interstitial inflammation, and tubular epithelial cell regeneration in the rats that lived to the end of the studies.

Additional 13-week studies were conducted in rats at dietary concentrations of 0, 100, or 400 ppm to demonstrate the absorption of roxarsone from the gastrointestinal tract; to determine its distribution in liver, kidney, and blood; and to study its effects on various hematologic and clinical chemical values. No deaths occurred. Renal lesions of minimal severity observed in male rats that received 400 ppm were characterized by tubular epithelial cell degeneration and regeneration, tubular casts, and mineralization. Arsenic levels in urine, blood, kidney, and liver of dosed rats increased (140%-300%) with time on study and were proportional to the dietary concentrations of roxarsone. No compound-related hematologic or clinical chemical effects were observed in rats.

In the first 13-week studies, final mean body weights of mice that received 800 ppm were 11%-18% lower than those of controls. Deaths occurred in males and females receiving 400 and 800 ppm. No compound-related gross or histopathologic lesions were observed.

In the second 13-week studies in mice, no compound-related hematologic or clinical chemical effects were observed. At the end of the studies, arsenic concentrations in dosed mice ranged from 0.45 to 0.99 ug/g of liver and from 0.85 to 2.98 ug/g of kidney. No arsenic was detected in the liver or kidney of control mice.

Because of kidney lesions, lower body weight gain, and increased mortality in rats and lower body weight gain and increased mortality in mice in the short-term studies, dietary concentrations of roxarsone selected for the 2-year studies were 0, 50, or 100 ppm for rats and 0, 100, or 200 ppm for mice.

Two-year studies

Body weight and survival

Mean body weights of dosed rats were generally within 5% of those of controls. No significant differences in survival were observed between any groups of rats of either sex, although survival in males was lower than usual (final survival--male: control, 24/50; low dose, 18/50; high dose, 18/50; female: 27/50; 35/50; 32/50). The average feed consumption by high dose rats was 95% that of controls for males and 88% for females. The average amount of roxarsone consumed per day was approximately 2 mg/kg for low dose rats and 4 mg/kg for high dose rats. Mean body weights of high dose male mice were generally 5%-8% higher than those of the controls, whereas those of female mice were generally 6%-15% lower than those of the controls. The survival of the control group of male mice was lower than that of the low dose group after month 22; survival for females was low (final survival--male: 27/50; 40/50; 33/50; female: 14/50; 18/50; 17/50). The low survival in females was due in part to utero-ovarian infection, with more than 50% of the animals in each dose group having suppurative inflammation at this site. The average daily feed consumption by dosed mice was 105%-110% that by the controls. The average amount of roxarsone consumed per day was approximately 21 or 43 mg/kg for low dose or high dose male mice and 27 or 54 mg/kg for low dose or high dose female mice.

Nonneoplastic and neoplastic effects

Although the incidence of adenomas of the exocrine pancreas in high dose male rats was not statistically greater than that in the controls (control, 1/50; low dose, 1/50; high dose, 5/50), it was greater than that seen in any historical control group of male F344/N rats. The historical rate is 1/437 (0.2%) for the study laboratory and 5/1,871 (0.3%) throughout the Program. The incidences of hyperplasia were 2/50; 0/50; 3/50. No hyperplasia oradenomas were observed in the exocrine pancreas of female rats.

Clitoral gland adenomas in female rats occurred with a marginally positive trend (1/44; 3/47; 6/48; P=0.049). One carcinoma was also observed in each of the groups. The incidences of adenomas or of adenomas or carcinomas (combined) in the dosed groups were not significantly different from those in the controls. This marginal effect was not considered to be related to roxarsone administration.

No chemical-related increases in neoplastic or nonneoplastic lesions occurred in male or female mice. Lymphomas in female mice occurred with a negative trend; the incidences in the dosed groups were lower than that in the controls (13/50; 2/50; 3/50; P<=0.01).

Genetic toxicology

Roxarsone was not mutagenic in Salmonella typhimurium strains TA98, TA100, TA1535, or TA1537 with or without metabolic activation. Roxarsone induced trifluorothymidine (Tft) resistance in mouse lymphoma L5178Y cells in the absence of metabolic activation; it was not tested with activation. Exposure of adult male Drosophila melanogaster to roxarsone by injection or by feeding did not cause an increase in sex--linked recessive lethal mutations.

Data audit

The data, documents, and pathology materials from the 2-year studies of roxarsone have been audited. The audit findings show that the conduct of the studies is documented adequately and support the data and results given in this Technical Report.


Under the conditions of these 2-year feed studies, there was equivocal evidence of carcinogenic activity of roxarsone for male F344/N rats, as indicated by a marginally increased incidence of adenomas of the exocrine pancreas. There was no evidence of carcinogenic activity for female F344/N rats fed diets containing 50 or 100 ppm roxarsone for 2 years. There was no evidence of carcinogenic activity for male or female B6C3F1 mice fed diets containing 100 or 200 ppm roxarsone for 2 years.