Theophylline

Endpoint summary

Description of key information

Theophylline was given to rats and mice by feed or by gavage. In rats theophylline caused nephropathy in all fed male rats and a dose-dependent periarteritis in all treated groups.
Oral: 
subchronic
(1) LOAEL rat, female/male, feed, 90 days: 75 mg/kg bw/d [nephropathy (males); periarteritis (females)]; (Collins et al., 1988/NTP 1998) 
(2) LOAEL rat, female/male gavage, 90 days: 37.5 mg/kg bw/d (periarteritis); (Collins et al., 1988/NTP 1998)
(3) LOAEL mouse, male, feed, 90 days: 175 mg/kg bw/d; (reduced body weight); (Collins et al., 1988/NTP 1998)
      LOAEL mouse, female, feed, 90 days: 225 mg/kg bw/d; (reduced body weight); (Collins et al., 1988/NTP 1998)
(4) NOAEL mouse, male, gavage, 90 days: 75 mg/kg bw/d; (reduced body weight); (Collins et al., 1988/NTP 1998)
      NOAEL mouse, female, gavage, 90 days: 150 mg/kg bw/d; (mortality); (Collins et al., 1988/NTP 1998) 
chronic
LOAEL rat, female/male, gavage, 2 yrs: 7.5 mg/kg bw/d; (reduced body weight); (Nyska et al., 1998/NTP 1998)
NOAEL mouse, female, gavage, 2 yrs: 7.5 mg/kg bw/day; (reduced body weight); (NTP 1998)
NOAEL mouse, male, gavage, 2 yrs: 50 mg/kg bw/day; (reduced body weight, mortality); (NTP 1998)   

Additional information

Valid experimental data for oral repeated dose were available to assess the repeated dose toxicity.

Oral:

subchronic:

Male and female F344/N rats and B6C3F1 mice were given theophylline in feed or in corn oil by gavage for 16 days or 14 weeks or in corn oil by gavage for 2 years (NTP 1998 (Technical report 473; NIH-Publication No. 98-3963; NTIS, PB99-113342), Collins J.J. et al., 1988 (Fund. Appl. Toxicol. 11, 472-484), Lindamood III C. et al. 1988 (Fund. Appl. Toxicol. 10, 477-489), Nyska A. et al. 1998 (Arch. Toxicol. 172, 731-737).

 

(1) In a 13 weeks subchronic feeding study F344/N rats were treated with diets containing the substance at doses of 0, 1000, 2000 and 4000 ppm (ca. 75, 125, and 250 mg/kg bw/d for male rats and ca. 75, 125, and 275 mg/kg bw/d for female rats). Each group consisted of 10 animals per sex (Collins, 1988, NTP, 1998).

There was no mortality or significant weight-gain depression in any of the treated groups. The food consumption was unaffected. No substance related gross lesions were observed. Mean cell volume of red blood cells and mean cell hemoglobin of red blood cells was increased in mid dose and high dose males.

The platelet count of high dose males was increased. Segmented neutrophil counts were increased in all females fed the substance.

A dose-dependent increase in the incidence of mesenteric and/or pancreatic periarteritis was observed in the mid and high dose group of males (2 and 3 animals respectively) and in all dose groups of females (1, 1, 5 animals).

Periarteritis was characterized by infiltration of mononuclear and polymorph nuclear leukocytes into the media and adventitia, and the more severe lesions included degeneration of medial smooth muscle and periarterial fibrosis (NTP 1998). These effects were also observed in a gavage dose-finding study after 16 days in males of the highest dose group (400 mg/kg bw/d, lethal dose) and in the 2-year study in males of the highest dose group (75 mg/kg bw/d) (see Chapter 7.7).

Theophylline is a non-specific phosphodiesterase inhibitor leading to elevated cyclic AMP (cAMP) related to arteritis.

Hanton et al. (1995, Arch. Toxicol. 69, 698 -704) suggested that increased cAMP in vascular smooth muscle cells may be a critical event in the development of the arteritis induced by PDE III inhibitors. On the other side, according to Nyska et al. (1998) theophylline may cause excessive vasodilatation and the mesenteric and pancreatic arteries in rats may be particularly sensitive to the excessive vasodilator-pharmacological activity of theophylline, caffeine and other vasodilator drugs. Similar to other vasodilatory chemicals (Hanton et al., 1995), the periarteritis caused by theophylline may be a consequence of hemodynamic changes induced in the vascular wall (Nyska et al., 1998).

Kidney weights were increased in males fed 250 mg/kg bw/d. A nephropathy, characterized by randomly distributed foci of tubular degeneration, dilated tubules containing eosinophilic protein casts and focal interstitial mononuclear cell infiltrates was observed in all males, including the controls. The severity of lesions progressed with dose (controls and low dose: minimal, mid-dose: mild, high-dose: moderate). The mechanism of this effect is unclear; it may relate to the fact that theophylline, as well as other xanthines, acts as a potent vasoconstrictor in the kidney, in contrast to all other vascular beds in which these compounds induced vasodilatation (Osswald H., 1983 (in Regulatory function of Adenosine, Berne et al: Eds. 399 - 415); Williams M., 1987, Annu. Rev. Pharmacol. Toxicol . 27, 315 -345; both cited in Collins et al., 1988). However, also the relevance of this finding is unclear, since these effects (though less severe) have also been observed in the controls, but were not found in other repeated dose studies with theophylline.

No significant differences were found between control and exposed rats in sperm morphology, and in vaginal cytology parameters nor at the microscopic evaluation of sex organs (NTP 1998).

Based on the nephropathy in males and the periarteritis in females the LOAEL for this study is considered to be 75 mg/kg bw/day.

(2) In a 13 weeks subchronic gavage study F344/N rats were administered 0, 37.5, 75, and 150 mg/kg bw/d in corn oil. Each group consisted of 10 animals per sex (Collins, 1988, NTP, 1998). One high dose male rat and one high-dose female died before the end of the study. The mean body weight gain of the 150 mg/kg bw/d females was significantly greater than that of the controls. Food consumption of high dose females and high dose males appeared to be slightly higher than that of the control group. No treatment related gross observations were noted at necropsy. Several organ weight changes were found in an unspecific manner. Thymus weights were decreased in high dose rats, and liver weights were increased in mid and high dose females. The observed findings had no histopathological correlations.

Microscopic examinations revealed a slight dose-dependent increase in the incidence of periarteritis of the small- to medium-size arteries adjacent to the mesenteric lymph nodes of male and female rats (periarteritis incidences males: control: 1/10, 37.5: 1/10, 75: 2/10; 150: 5/10; females: 0/10, 2/10, 2/10, 3/10). The periarteritis observed in one control male was more consistent with that commonly observed in aged rats and consisted of minimal, focal lymphocyte accumulation adjacent to the artery. The periarteritis was focal or circumferential and was characterized by infiltration of mononuclear and polymorph nuclear leukocytes into the media and adventitia. As described in the feeding study above the periarteritis may be a consequence of hemodynamic changes induced in the vascular wall.

Mean cell volume of red blood cells was increased in high dose males and mean cell hemoglobin of red blood cells was increased in all treated males. No significant differences were found between control and exposed rats in sperm morphology, and in vaginal cytology parameters nor at the microscopic evaluation of sex organs (NTP 1998, Collins1988).

Based on the periarteritis the LOAEL for this study is considered to be 37.5 mg/kg bw/day. 

 

(3) In a 13 weeks subchronic feeding study B6C3F1 mice were fed diets containing the substance at doses of 0, 1000, 2000 and 4000 ppm (ca. 175, 400, and 800 mg/kg bw/d for male mice; ca. 225, 425, and 850 mg/kg/d bw for female mice). Each group consisted of 10 animals per sex (Collins, 1988, NTP, 1998).

No deaths were observed in the feeding study. Final mean body weights and body weight gains were reduced in all treated mice as compared to the control. The final weight in comparison to the control was significantly reduced in males by 13%, 15% and 14% and in females by 8%, 7% and 7% in the low, mid and high doses, respectively. Relative thymus weights were decreased in females at mid- and high-dose group.

Leukocyte, segmented neutrophil and lymphocyte counts were increased in high dose males and females and in mid dose females. No significant exposure-related lesions were observed at necropsy. Histological examinations revealed a hepatocyte glycogen depletion in all treated groups, which is considered as a result of lower body weights. No other findings were noted in the inner organs or in the sex organs.

No significant differences in sperm morphology or vaginal cytology parameters were observed between control and exposed mice (NTP 1998, Collins1988). Because of the reduced body weights, a NOAEL could not be achieved.

Based on reduced body weight, the LOAEL in this study for males is 175 mg/kg bw/day and for females 225 mg/kg bw/day.

 

(4) In a 13 weeks subchronic gavage study B6C3F1 mice received 0, 75, 150, and 300 mg/kg bw/d in corn oil. Each group consisted of 10 animals per sex (Collins, 1988, NTP, 1998).

3 high dose males, all high- dose females, one low dose male, and one control female died.

Final mean body weights and body weight gains were reduced in male mice in the mid and high dose level.

Mean cell volume of red blood cells and mean cell hemoglobin increased in males of the 300 mg/kg bw/d dose group. Like in the feeding study hepatocyte glycogen depletion was observed (in females only); lymphoid depletion (minimal to moderate) was observed in the thymus and spleen of high-dose male and was considered to be related to stress associated with theophylline administration. There were no histopathological findings attributed directly to theophylline treatment, no changes were found in the sex organs. No significant differences in sperm morphology nor vaginal cytology parameters were observed between control and exposed mice (NTP 1998).

Based on the reduced body weight the NOAEL for males is 75 mg/kg bw/d. For females the NOAEL is 150 mg/kg bw/d, based on the complete mortality in the high dose females.

chronic

In a 2-year bioassay, the substance was administered in corn oil by gavage to Fischer 344 rats and B6C3F1 mice (50 animals per sex). The rats were given 0, 7.5, 25, and 75 mg/kg bw/d (males and females); male mice received 0, 15, 50, and 150 mg/kg bw/d; female mice received 0, 7 .5, 25, and 75 mg/kg bw/d (Nyska et al., 1998, NTP, 1998).

In rats, mortality was similar in all groups. Body weights were reduced in all dose groups between 10 and 18 %. No increase in the incidence of neoplasms was observed. Increased incidence of periarteritis was found in high-dose males. This observed periarteritis may be a rat- specific response to vasodilators (see 13 weeks feeding study).

In mice, mortality was increased in high-dose males. Body weights were reduced in high-dose males and females and in mid-dose females. No significantly increased incidences of neoplasms or non-neoplastic lesions were observed.

There was no evidence of carcinogenic activity of the substance in both, rats or mice (NTP, 1998).

For rats, based on the reduced body weight a NOAEL could not be achieved. The LOAEL in this study is 7.5 mg/kg bw/day.

For mice, based on the reduced body weight and mortality the NOAEL for males is 50 mg/kg bw/day and based on the reduced body weight, the NOAEL for females is 7.5 mg/kg bw/day.

In humans chronic toxicity was observed in patients receiving the drug therapeutically (Woo O.F. et al., Vet. Hum. Toxicol, 22, 48-51, 1980, Powell E.C., Pediatrics Emerg. Care, 9, 129 -133, 1993) and in newborns due to placental tansfer (Labovitz E. and Spector S., J. Am. Med. Assoc., 247, 786 -788, 1982). The features of chronic theophylline intoxication are qualitatively similar to those of acute overdose (see Chapter 7.2). The accepted therapeutic serum concentration of theophylline ranges from 10 - 20 µg/mL. Improved forced respiratory volume in 1 s (FEV1), vital capacity and airways resistance was observed at plasma concentrations as low as 4 - 5 µg/mL (Minton and Henry 1996 Human and Experimental Toxicology, 15, 471 -481). Therefore, independently of the exposure way, the lower therapeutic level is taken as the NOAEL in humans. However, the plasma concentrations correlate poorly with the ingested dose and are dependent on factors like age, co-medication, bioavailability and smoking habits (Helliwell M. and Berry D., Br. Med. J., II, 1114, 1979). Therefore, the exact determination of the applied amount of the compound corresponding to the effect level is not possible. 

Conclusion:

Theophylline was given by feed or by gavage to rats and mice. In the 13 weeks study, theophylline caused nephropathy in male rats and a dose dependent periarteritis in all treated groups. No histopathological changes were observed in mice.

In the two-year study, nephropathy was not observed and periarteritis only occurred in male rat of the highest dose group. The observed effects were considered as secondary effects, due to the pharmacological properties, such as vasodilatation and vasoconstriction of methylxanthines. Accordingly, the periarteritis may be a consequence of hemodynamic changes induced in the vascular wall (Nyska et al. 1998) At high dosis haematoligical parameters were changed in mice and rats.

No histopathological changes were found in other organs including sex organs of rats or mice.

Dermal

No subacute or subchronic repeated dose studies with dermal application are available for theophylline.

Inhalation

No subacute or subchronic repeated dose inhalation studies are available for theophylline.

Justification for classification or non-classification

Based on the repeated dose studies, theophylline has not to be classified according to requirements of the EU Annex VI of directive 67/548/EEC and GHS for the oral route.