IPCS INCHEM Home

WHO FOOD ADDITIVES SERIES: 51

TRICHLORFON (addendum)

First draft prepared by
Dr Pamela L. Chamberlain
Center for Veterinary Medicine, Food and Drug Administration, Rockville, Maryland, USA

Explanation

Biological data

Biochemical aspects

Absorption, distribution and excretion

Toxicological studies

Genotoxicity

Reproducitive toxicity

Multigeneration studies

Developmental toxicity

Special studies on toxicity to mammalian germ cells

Special studies: Neurotoxicity

Observations in humans

Comments

Evaluation

References

1. EXPLANATION

Trichlorfon (metrifonate) was evaluated by the Committee at its fifty-fourth meeting (Annex 1, reference 147), when it established an ADI of 0–20 µg/kg bw on the basis of a NOEL of 0.2 mg/kg bw per day for inhibition of erythrocyte acetylcholinesterase activity in humans treated orally, applying a safety factor of 10.

A re-evaluation of the ADI for trichlorfon was requested on the basis of the availability of new data that were not reviewed by the previous Committee. The Committee at its present meeting considered the results of additional studies on the pharmacokinetics of trichlorfon and on genotoxicity, reproductive toxicity, developmental toxicity, toxicity to mammalian germ cells and studies in humans.

2. BIOLOGICAL DATA

2.1 Biochemical aspects

2.1.1 Absorption, distribution, and excretion

The pharmacokinetics of trichlorfon and its metabolite dichlorvos was studied in healthy human volunteers and patients with various levels of renal impairment defined by creatinine clearance. The study was conducted in accordance with European Community Good Clinical Practice Guidelines, the Declaration of Helsinki (1989) and the German Drug Laws. The 24 participants (15 men and 9 women aged 45–75 years and weighing 45–94 kg) were divided into four groups according to their creatinine clearance rates. Healthy volunteers (group A) had a creatinine clearance > 90 ml/min per 1.73 m2; individuals with renal impairment were assigned to groups with creatinine clearance = 60–90 ml/min per 1.73 m2 (group B), 30–60 ml/min per 1.73 m2 (group C) or < 30 ml/min per 1.73 m2 (group D). There were four men and two women in groups A, C and D and three men and three women in group B. Treatment consisted of a single 50-mg tablet of trichlorfon administered orally. Clinical evaluations were performed 0 (before dosing),1, 2, 3, 4, 12 and 24 h after administration. Clinical laboratory end-points were evaluated before dosing and during the final examination. A 12-lead electrocardiogram (ECG) was recorded before dosing and 1, 4 and 24 h afterwards. Blood samples for measurements of pharmacokinetics were collected 0, 10, 20, 30, 45 and 60 min and 1.5, 2, 3, 4, 6, 8, 10, 12, 16 and 24 h after dosing. Plasma and erythrocyte cholinesterase activity was measured before dosing and 1, 24 and 168 h afterwards.

The adverse clinical effects observed included diarrhoea, headache, transient increases in amylase and lipase activity and weakness. No treatment-related effects were found in clinical laboratory and ECG end-points. The differences in the pharmacokinetics of trichlorfon and dichlorvos in the various groups were not statistically significant. The geometric mean half-life for trichlorfon in healthy volunteers was 3 ± 1 h, whereas those in groups B, C and D were 3 ± 1 h, 3 ± 1 h and 2 ± 1 h, respectively. The renal clearance of trichlorfon and dichlorvos decreased in proportion to the severity of renal impairment; however, as renal excretion of unchanged drug or dichlorvos contributed only 1–2% of the total dose to the overall clearance of trichlorfon, the systemic concentrations of trichlorfon and its active metabolite did not affect renal function. No significant inhibition of erythrocyte cholinesterase was found in healthy or renally impaired participants. The investigators explained that trichlorfon causes a gradual increase in inhibition of erythrocyte cholinesterase activity, steady-state inhibition being reached within a few weeks at the daily doses used in the treatment of Alzheimer disease. The authors did not specify the doses used in this indication, but the dosage regimens used in reported clinical trials consist of a loading dose of 0.5, 0.9 or 2 mg/kg bw per day for 2 weeks (Cummings et al., 1998), followed by a maintenance dose of 0.2, 0.3 or 0.65 mg/kg bw per day for 10 weeks. In the present study, plasma cholinesterase activity was inhibited by 60–80% 1 h after dosing, with greater inhibition seen in healthy volunteers. The investigators concluded that renal function had no significant effect on the pharmacokinetics of trichlorfon and that doses need not be adjusted for patients with renal impairment (Dingemanse et al., 1999; Heinig & Dietrich, 1999).

The pharmacokinetics of trichlorfon was studied in healthy volunteers given the drug alone or in combination with a magnesium- and aluminum hydroxide-containing antacid, cimetidine or ranitidine, in two studies conducted in accordance with good clinical practice guidelines and the 1975 Declaration of Helsinki and its revisions. In the first study, 12 female and six male volunteers aged 55–75 years and of normal body weight were given three single 50-mg oral doses of trichlorfon 1week apart. Six participants were dosed orally 5 min later with 10 ml of a magnesium- and aluminum hydroxide-containing suspension. Another six participants were dosed 90 min later with 10 ml of the same suspension. The remaining six received no additional treatment. Blood samples for assay of trichlorfon, dichlorvos and cholinesterase activity were collected before dosing and 0.5, 1, 1.5, 2, 3, 6 and 8 h after dosing. An additional sample was collected at 24 h for determination of cholinesterase activity. In the second study, six female and 10 male volunteers aged 45–58 years and of normal body weight were given each of the following treatments separated by a 1-week washout period: cimetidine at 400 mg in a tablet twice daily for 4 consecutive days, with 400 mg of cimetidine and a 50-mg trichlorfon tablet administered on day 5; ranitidine at 150 mg in a tablet twice daily for 4 consecutive days, with 150 mg of rantidine and a 50-mg trichlorfon tablet administered on day 5; or a single oral dose of 50 mg trichlorfon. Blood samples for assay of trichlorfon and dichlorvos were collected before administration and 10, 20, 30, 45 min and 1, 1.5, 2, 3, 4, 6, 8, 10 and 12 h afterwards. Blood samples for determination of cholinesterase activity were collected before dosing and 1, 2 and 24 h after dosing on day 5 of each period and in the follow-up phase of the study 7 and 14 days after administration of trichlorfon. Plasma samples for assay of cholinesterase activity were collected on days 1, 4 and 5 before dosing and on day 5, 24 h after dosing, in each period and 7 and 14 days after administration of trichlorfon in the last study period. In both studies, all volunteers were given a complete medical examination including medical history, physical examination, chest X-ray, laboratory investigations, vital signs and an ECG, before entering and after completion of the study.

Nausea (first study) and headache (second study) were the only adverse events considered possibly to be related to treatment. There were no clinically relevant changes in ECG or clinical laboratory end-points. The pharmacokinetics of trichlorfon and dichlorvos were not significantly affected by concomitant administration of the antacid preparation or by intake of cimetidine or ranitidine. The authors explained that the terminal elimination half-lives of trichlorfon and dichlorvos could not be determined in the first study because many samples contained insufficient concentrations of these compounds during the elimination phase. The geometric mean elimination half-life of trichlorfon in the second study was 2–3 h. In the first study, erythrocyte cholinesterase activity was only slightly inhibited and was not distinguishable from baseline values. The authors explained that multiple doses must be administered to attain significant inhibition of this enzyme and, eventually, therapeutically relevant, steady-state inhibition. In the second study, the authors reported that the cholinesterase activity in plasma was constant on days 1, 4 and 5 of the study with the first two treatments, but the level of inhibition was not given. After the third treatment period, the geometric mean inhibition of plasma cholinesterase activity 24 h after dosing was 35% that of the baseline value on day 5. The authors concluded that concomitant administration of a magnesium- and aluminum hydroxide-containing antacid, cimetidine or ranitidine, had no effect on the pharmacokinetics of trichlorfon or dichlorvos in a healthy volunteer population (Heinig et al., 1999).

The effects of food and time of administration on the pharmacokinetics of trichlorfon were studied in healthy volunteers. The studies were conducted according to good clinical practice guidelines and the 1975 Declaration of Helsinki and its revisions. In the first study, 12 male and two female participants aged 50–68 years and of normal body weight were given two single oral doses of 50 mg of trichlorfon 1 week apart. After an overnight fast (10 h), the volunteers received the following treatments, in random order: a 50-mg tablet of trichlorfon orally or a 50-mg tablet within 5 min of eating a breakfast consisting of 22 g of protein, 76 g of fat, 54 g of carbohydrate and 1015 total calories. Blood samples for assay of trichlorfon and dichlorvos were collected before administration and 10, 20, 30 and 45 min and 1, 1.5, 2, 3, 4, 6, 8 and 10 h after dosing. Blood samples for determination of erythrocyte and plasma cholinesterase activity were collected before administration and 1, 2, 3, 4, 6, 8, 10 and 24 h after dosing. In the second study, 12 healthy male volunteers aged 24–45 years and of normal body weight were given the following treatments, separated by a 1-week washout period: a tablet containing 80 mg of trichlorfon at 8:00 after an overnight fast (10 h), with a meal 4 h after dosing; a tablet containing 80 mg of trichlorfon at 19:00, after a meal at 12:00; and a tablet containing 80 mg of trichlorfon at 22:00 after a meal at 18:00. Blood samples were collected for assay of trichlorfon and dichlorvos before administration and 10, 20, 30 and 45 min and 1, 1.5, 2, 3, 4, 6, 8 and 12 h after dosing. Blood samples for determination of erythrocyte and plasma cholinesterase activity were collected before administration and 0.5, 1, 2, 3, 4, 6, 8 and 12 h after dosing.

Nausea and diarrhoea (first study) and headache (second study) were the only adverse events considered possibly to be related to treatment. The bioavailability of trichlorfon and dichlorvos after breakfast was approximately 95% of that after fasting. The maximum concentrations of trichlorfon and dichlorvos were reduced to approximately 56% with food, and the time to maximal plasma concentration was increased. The terminal elimination half-lives of trichlorfon or dichlorvos were not affected; the geometric mean elimination half-life of trichlorfon in the first and second study was 2 ± 1 h. In the first study, erythrocyte cholinesterase activity was not significantly changed from the values before administration . Plasma cholinesterase activity was almost completely inhibited 1 and 3 h after administration. In the second study, inhibition of erythrocyte cholinesterase activity was low and erratic, with high standard deviations, after all treatments. The mean plasma cholinesterase activity was inhibited maximally (> 95%) 1 and 3 h after treatment in the first and second treatment phases and 3 h after treatment in the third treatment phase. The enzyme was still measurably inhibited (by approximately 50%) 24 h after dosing. The authors concluded that intake of food and time of administration had no relevant effect on the pharmacokinetics of trichlorfon or dichlorvos or the pharmacodynamics of erythrocyte or plasma cholinesterase activity (Heinig & Sachse, 1999).

2.2 Toxicological studies

2.2.1 Genotoxicity

The only information on the genotoxicity of trichlorfon that was not reviewed by the Committee at its fifty-fourth meeting was the results of two tests for sister chromatid exchange reported in the same paper. In a test conducted in human lymphocytes in vitro, trichlorfon at a concentration of 10, 20, 30, 40, 50 or 60 µg/ml did not induce sister chromatid exchange; however, when mice were given trichlorfon at a dose of 30, 60 or 120 mg/kg bw, sister chromatid exchange was observed in bone-marrow cells (Madrigal-Bujaidar et al., 1993).

2.2.2 Reproductive toxicity

(a) Multigeneration studies

Groups of 40 male and 40 female Sprague-Dawley rats received diets containing trichlorfon (purity, 98%) at a concentration of 150, 500 or 1750 mg/kg, equivalent to 7.5, 25 and 88 mg/kg bw per day, for two generations, with one mating per generation. The parameters evaluated in adults and pups included clinical observations, body weight, food consumption and gross and tissue lesions. In addition, plasma and erythrocyte cholinesterase activities were determined in 10 F0 and F1 animals of each sex per dose after 8 weeks of treatment, during the premating phase and again at termination. Plasma, erythrocyte and brain cholinesterase activities were determined in F1 and F2 generation pups at the time of culling (day 4 post partum) and at weaning (day 21 post partum). One pup of each sex per litter was chosen randomly from each litter and killed on lactation day 4 and another pair on lactation day 21, until 10 pups of each sex per dose had been obtained. This study was conducted in accordance with the United States Environmental Protection Agency Pesticide Assessment Guidelines and Health Effect Guidelines and the OECD guidelines for testing of chemicals.

No remarkable clinical signs were observed in adults or pups of either generation. Significantly decreased body weights were observed in males and females at the highest dietary concentration during the premating phase. No meaningful effects were observed on food consumption by either sex. In the gestational phase of the second generation, the body weights of animals at the highest dose were significantly decreased on days 0 and 6 of gestation. The body weights of animals at the highest dose were also decreased during lactation, but the difference was not statistically significant. Treatment had no effect on food consumption during gestation in either generation. Food consumption during the lactation phase was significantly reduced in F0 dams at the highest dose during week 2 of lactation. In animals of the second generation at the highest dose, food consumption during lactation was significantly reduced in weeks 1, 2 and 3.

Dams in the first generation at the highest dose had a significantly reduced lactation index (live pups per litter on lactation day 21/live pups per litter on lactation day 4 after culling × 100). In the second generation, significant decreases in the birth index (pups born per litter/implantation sites per litter) and mean litter size were observed at the lowest and highest doses, but there was no dose–response relationship. Treatment-related decreases in pup body weight were observed in both generations at the highest dose. Gross and histopathological findings were unremarkable in adults and pups of both generations.

Significant decreases in plasma cholinesterase activity were observed in F0 generation adult females at the two higher doses and in males at the highest dose. Erythrocyte cholinesterase activity was significantly reduced in adult F0 males and females at the highest dose after 8 weeks of dietary intake of the test compound. In the F1 generation, plasma cholinesterase activity was significantly reduced in adult females at the two higher doses and in males at the highest dose. Erythrocyte cholinesterase activity was significantly reduced in adult males and females at the two higher doses. At termination of the F0 generation, plasma cholinesterase activity was significantly reduced in females at all doses, erythrocyte cholinesterase activity was significantly reduced in males and females at the two higher doses, and brain cholinesterase activity was significantly reduced in females at all doses and in males at the highest dose. At termination of the F1 generation, plasma cholinesterase activity was significantly decreased in females at the highest dose, erythrocyte cholinesterase activity was significantly decreased in females at all doses and in males at the highest dose, and brain cholinesterase activity was significantly decreased in females at all doses and in males at the highest dose. Assessment of neonatal cholinesterase activity on lactation day 4 showed significantly decreased erythrocyte cholinesterase activity in F1 males at the two higher doses. On lactation day 21, significant decreases in plasma and brain cholinesterase activity were observed in males and females at the highest dose. In the F2 generation, no significant decreases in cholinesterase activity were observed in males or females on lactation day 4. On lactation day 21, significant decreases in plasma cholinesterase activity were observed in males and females at the highest dose, and significant decreases in brain cholinesterase activity were observed in males at the two higher doses and in females at the highest dose. The treatment-related effects observed at the lowest dose in this study included significantly decreased plasma cholinesterase and brain cholinesterase activities in F0 adult females at term and significantly decreased erythrocyte and brain cholinesterase activities in F1 adult females at term. A NOEL could not be identified in this study (Astroff et al., 1998).

(b) Developmental toxicity

The sensitive period and the doses of trichlorfon required to produce brain hypoplasia in offspring were examined in pregnant white guinea-pigs. The number of animals in each treatment group was not stated. Trichlorfon was administered either by stomach tube or by subcutaneous injection. Three groups received an oral dose of 25, 50 or 100 mg/kg bw on days 42, 43 and 44 of gestation. One group received oral doses of 200, 100 and 200 mg/kg bw on days 42, 43 and 44 of gestation, respectively. One group received oral doses of 150, 150, 100 and 150 mg/kg bw on days 40, 41, 42 and 43 of gestation, respectively. One group received a single oral dose of 165 mg/kg bw on day 44 of gestation. Only one group was treated subcutaneously, with a dose of 125 mg/kg bw. An untreated control group was included. The pups were decapitated within 24 h after natural birth or immediately after surgical removal around day 64, and various brain sections were separated out and weighed.

In an earlier study, a significant reduction in brain weight was observed in offspring born to dams treated orally with trichlorfon at a dose of 125 mg/kg bw on days 42, 43 and 44 of gestation (Mehl et al., 1994). In the more recent study, no reduction in the weight of the cerebellum or other brain structures was found in offspring of dams at 25 or 50 mg/kg bw , but significant weight reductions were observed in the offspring of dams given 100 mg/kg bw. A dose–response related increase in brain weight reduction was observed in a comparison of the groups given 100 mg/kg bw, 200–100–200 mg/kg bw and 150–150–100–150 mg/kg bw. Clinical signs of toxicity were observed in dams treated with 100, 150 or 200 mg/kg bw. The cerebellum appeared to be most sensitive brain structure to exposure on days 42–44, the diencephalon and medulla oblongata to treatment on days 45–50 and the colliculus to treatment on days 51–53 of gestation. The NOEL was 50 mg/kg bw (Hjelde et al., 1998).

The brain hypoplasia seen after exposure to trichlorfon has been attributed to DNA alkylation damage and inhibition of DNA alkyltransferase repair (Badawi, 1998; Mehl et al., 2000).

In a study of cytogenetic and developmental effects on pre-implantation, mid-gestation and near-term mouse embryos treated with trichlorfon during the zygote stage, female mice were given an intraperitoneal dose of 100 mg/kg bw (18 animals) or 200 mg/kg bw (21 animals) 6 h after presumed conception. A control group of 18 pregnant animals received distilled water. Developmental outcomes and micronucleus formation during the pre-implantation phase were assessed on day 3 of gestation; and developmental and aneugenic outcomes at mid-gestation were assessed on day 9. On day 17 of gestation, the embryos were removed, sexed, weighed and examined for external malformations.

No signs of toxicity related to treatment with trichlorfon were observed. The mean number of cells in embryos in both treated groups was significantly lower than in the control group, and the mean number of micronuclei was significantly increased in both treated groups compared with the control group. The number of live embryos per dam was significantly lower in the group given 100 mg/kg bw than in controls. The mean number of somites in the treated groups was significantly lower than in the control group. A significant increase in mosaic aneuploidy, including monosomic or trisomic cell lines, was associated with treatment. The percentage of dead or resorbed embryos tended to be higher in treated than in the control group, but the differences were not significant. There was no increase in the incidence of external malformations in the treated groups, and the body weights of male and female fetuses exposed to trichlorfon were comparable to those of controls. The authors concluded that exposure to trichlorfon around the time of fertilization induces a high frequency of micronuclei, aneuploidy and developmental retardation in embryos from the pre-implantation to the mid-gestation stage. Thereafter, embryos with micronuclei or chromosomal damage appeared to develop and could appear normal by near term (Tian et al., 2000).

A study of the effects of trichlorfon on spindle morphology and chromosomal segregation was conducted in fertilized mouse embryos in vitro. Epididymal sperm obtained from male mice was capacitated and added to 385 oocytes collected from superovulated female mice, and the mixture was incubated for 2 h. By that time, most of the fertilized eggs had reached anaphase II of the second maturation division. Some fertilized embryos were incubated further to the first mitotic division. Chromosomal analysis was performed on one-cell embryos arrested in metaphase. To study the effect of trichlorfon on fertilization, chromosomal segregation and spindles at anaphase II, 73 oocytes were incubated in 50 µg/ml of trichlorfon for 1 h before fertilization, then transferred to a chemical-free medium and fertilized. In addition, 300 oocytes were exposed for a total of 3 h: 1 h before fertilization and during the 2-h fertilization. The treated embryos were then processed for chromosomal analysis or immunofluorescence. To study the effects of trichlorfon on spindles and chromosomes of maturing oocytes, the oocytes were incubated for 8 or 16 h in the presence or absence of 50 µg/ml trichlorfon. By that time, most of the oocytes had emitted a first polar body and were arrested at metaphase II.

Trichlorfon had no significant effect on the rate of fertilization, nor did it affect the separation of chromatids or the distribution of chromosomes at anaphase II. With regard to effects on maturing oocytes, aberrant spindles were observed in a large percentage of oocytes after 8 h of exposure to trichlorfon. During the advanced stages of meiosis I, chromosomes appeared to be incapable of proper alignment at the equator. Polar body formation occurred at about the same time in trichlorfon-treated oocytes and controls. After 16 h, most of the treated and control oocytes had reached metaphase II. The spindles were highly aberrant in a large percentage of metaphase II stages, and chromosomes were frequently unaligned and located at different distances from the poles. The authors considered that the effects observed on spindle formation and chromosomal alignment placed oocytes at high risk for errors in chromosomal segregation and might make them more prone to nondisjunction, predisposing the fertilized egg and embryo to trisomy (Yin et al., 1998).

(c) Toxicity to mammalian germ cells

A study of aneuploidy induction in male mouse germ cells was conducted in vivo in groups of five male F1 mice given a single intraperitoneal injection of trichlorfon at 200, 300 or 400 mg/kg bw. The control group was given the solvent, dimethyl sulfoxide. The mice were killed 22 days after treatment, the time it normally takes for mouse spermatocytes to develop into mature sperm, and sperm were collected from the cauda epididymis. As the same investigators had established that trichlorfon has no effect on the duration of meiosis in male mouse germ cells, this sampling schedule was appropriate. A method involving multicolour fluorescence hybridization in situ with chromosome-specific DNA probes was used to identify gain or loss of individual chromosomes during meiosis. Approximately 5000 sperm cells were scored from each group. A significant, dose-related increase in the percentage of disomic (X-X-8, Y-Y-8, X-Y-8, X-8-8 or Y-8-8 phenotype) cells was observed at all doses over that in controls. Trichlorfon also caused spindle disturbances in Chinese hamster V79 cells in vitro. The authors noted that aneuploidy can result from non-disjunction or chromosome loss during meiosis. Chromosomal non-disjunction can result from malfunctioning of the processes involved in chromosomal segregation. The integrity of the spindle apparatus is considered to be of central importance in chromosomal segregation. The authors concluded that trichlorfon could be regarded as a germ-cell aneugen in vivo (Sun et al., 2000; Schmid et al., 2001).

2.2.3 Special studies: Neurotoxicity

The maximal effects of trichlorfon on soluble neuropathy target esterase (NTE) and the regional distributions of NTE and acetylcholinesterase were studied in a group of 16 adult hens aged 20–24 months. The hens were given a single intravenous dose of 200 mg/kg bw. To prevent acute death, each was given a subcutaneous injection of atropine sulfate at 20 mg/kg bw, 10 min before and 15 min after dosing with trichlorfon. Clinical signs of delayed neuropathy were studied in four of the hens for 28 days. The remaining 12 were used to measure the activities of NTE and acetylcholinesterase in three regions of the brain (cerebrum, midbrain and cerebellum) and three regions of the spinal cord (cervical, thoracic and lumbar) after 6, 24 and 48 h, with four hens per time. A group of six hens served as untreated controls.

No signs of delayed neuropathy were observed in the treated hens. NTE activity in all the regions measured was inhibited by 14–45% 6 h after dosing, 6–15% 24 h after dosing and 5–10% 48 h after dosing. Peak inhibition thus occurred at 6 h in all regions and ranged from 15% to 44%. The greatest inhibition was found in the midbrain (44%) and thoracic cord (36%). The average inhibition of NTE during the 6–48-h period after dosing varied from 10% in the cerebellum to 23% in the midbrain (Tian et al., 1998).

2.3 Observations in humans

A cluster of congenital abnormalities was identified in a Hungarian village in 1989–90. Of 15 live births, 11 (73%) were affected by abnormalities, six were twin births, and four of the 11 had trisomy 21 (Down syndrome). Two of the cases of Down syndrome also had endocardial cushion defect. Other abnormalities observed were a ventricular septal defect with pulmonary atresia, an inguinal hernia, stenosis of the left bronchus, anal atresia, choanal atresia, cleft lip and Robin sequence. Of 61 children born in this village between 1980 and 1988, only three had had malformations. The incidence of Down syndrome in the reported cluster was approximately 200 times greater than that in the general Hungarian population. A case–control study revealed that the mothers of all the infants with abnormalities reported having eaten ‘contaminated’ fish during the index pregnancy. It was found that several ponds around the village used for fish farming had been treated with a 40% trichlorfon formulation at a level of 500 mg/l. The composition of the formulation was not described. The average concentrations of trichlorfon in 12 carp, 12 amur and 10 European wels collected from treated ponds, frozen and analysed 10 days later were 0.15, 0.13 and 0.26 mg/kg, respectively. Because trichlorfon rapidly degrades in biological systems, the initial trichlorfon content of the fish was estimated to have been as high as 100 mg/kg. The estimated daily fish consumption was 250 g, resulting in a daily ‘worst-case’ estimated intake of 25 mg per person or 400 µg/kg bw. The authors pointed out that the non-specificity of the observed adverse outcomes argued against a single cause. Nevertheless, because mutagenic and teratogenic effects of trichlorfon have been observed in vitro, in vivo and in studies in experimental animals, the authors suggested that reproductive hazards associated with exposure to high doses of trichlorfon should be further explored (Czeizel et al., 1993).

Chromosomal effects in lymphocytes were studied in 31 people who had attempted suicide by self-poisoning with trichlorfon. The actual or estimated doses were not stated. Three blood samples were collected 3–6, 30 and 180 days after the poisoning incidents. The controls were patients who had undergone surgery for a hernia or appendicitis in the surgical department of the same hospital, and the authors reported that the rate of chromosomal breakage in these controls corresponded to that in the general population.

Aneuploidy was found in 14% of 500 lymphocytes from the first blood sample, 26% of 400 cells from the second sample and 16% of 480 cells from the third sample taken from the victims, with 4% found in controls (Czeizel, 1994).

The results of a double-blind, placebo-controlled, single-centre study of the safety, tolerability and pharmacokinetics of trichlorfon in patients with Alzheimer disease were used to calculate the NOEL for inhibition of erythrocyte cholinesterase activity. A total of 27 patients were given an oral loading dose of trichlorfon by capsule containing 1.5, 2.5, 4 or 4 mg/kg bw per day for 6 days, followed by a daily oral maintenance dose of 0.25, 0.4, 0.65 or 1 mg/kg bw for 21 days. The mean inhibition of erythrocyte cholinesterase activity at the end of the treatment was 14%, 35%, 66%, 77% and 82% with the placebo and the four treatments, respectively. A linear extrapolation of the data on inhibition of erythrocyte cholinesterase activity resulted in an estimated NOEL of 0.1–0.2 mg/kg bw (Bieber et al., 1996).

3. COMMENTS

The pharmacokinetics of trichlorfon was studied in 24 volunteers with renal disease who had various levels of impairment of renal clearance, in 34 healthy volunteers who also received magnesium- and aluminium hydroxide-containing antacids or H2 receptor antagonists (cimetidine or ranitidine) and before and after a meal in 26 healthy volunteers. Trichlorfon was administered as a single oral dose of 50 mg to the volunteers with renal disease, as three oral doses of 50 mg given 1 week apart to the volunteers also receiving antacids and as a single oral dose of 50 or 80 mg to the volunteers before or after a meal. The pharmacokinetics of trichlorfon was not significantly altered in any of these studies. The reported elimination half-life was approximately 2 h, similar to the value found by the Committee at its fifty-fourth meeting. In all studies, trichlorfon caused significant reductions in plasma cholinesterase activity, while erythrocyte cholinesterase activity was relatively unaffected. The authors concluded that multiple doses were required to attain significant inhibition of erythrocyte cholinesterase activity and a steady state of therapeutically relevant inhibition. The data available to the Committee were insufficient to determine whether significant species differences exist with regard to the pharmacokinetics of trichlorfon. The Committee noted that higher NOELs were observed in studies in which trichlorfon was administered in feed rather than by direct oral administration in tablets or by gavage. Therefore, differences in pharmacokinetics may result from differences in the bioavailability of the dosage form administered.

Trichlorfon has been tested in a large number of studies for genotoxicity covering a wide range of end-points, with considerable variation in the results for most end-points. Both positive and negative results were obtained in tests for bacterial mutations and for gene mutation in mammalian cells in vitro, but the results of studies of effects on chromosomes in mammalian cells in vitro (chromosomal aberrations or sister chromatid exchanges) were uniformly positive. Mostly negative results were found in assays in mammals in vivo assessed by the Committee at its fifty-fourth meeting, comprising tests for somatic cell mutations in bone marrow (sister chromatid exchange, negative result in a single study), micronucleus formation (negative results in five of six studies) and chromosomal aberrations (negative results in three of five studies). Mostly negative results were also found in assays for germ cell mutagenicity in vivo evaluated by the Committee at its fifty-fourth meeting, comprising dominant lethal mutations (negative results in six of nine studies) and chromosomal aberrations in spermatogonia or spermatocytes (negative results in three of four studies). The Committee at its present meeting received further data on mutagenicity, comprising positive results in studies of sister chromatid exchange in vivo but not in vitro. Trichlorfon was a germ cell aneugen in laboratory animals in vivo. There was also limited evidence from observations in poisoned humans that trichlorfon caused aneuploidy and chromosome damage in lymphocytes. A study involving pregnant women suggested that exposure to uncertain concentrations of residues of trichlorfon in fish may have caused trisomy 21 (Down syndrome) in their offspring as a result of germ cell aneugenicity. The Committee at its fifty-fourth meeting noted that bioassays for carcinogenicity in rats and mice gave negative results and identified a NOEL for developmental toxicity. The Committee at its present meeting concluded that the weight of the evidence from the assays for mutagenicity in vivo indicated that trichlorfon residues in animal-derived foods would not present a carcinogenic hazard to consumers.

In a two-generation study of reproductive toxicity, trichlorfon was administered in the diet to groups of rats at concentrations providing a dose equivalent to 0, 7.5, 25 or 88 mg/kg bw per day. The parameters evaluated in adults and pups included clinical end-points, body weight, food consumption and gross and histological appearance. The body weights of F0 males and females at the highest dose were significantly decreased during the pre-mating phase, although feed consumption appeared to be unaffected by treatment. The body weights of F1 dams were decreased during gestation and lactation. The feed consumption of pups in the first and second generation at the highest dose was decreased during the lactation phase. The lactation index (live pups per litter on lactation day 21/live pups per litter on lactation day 4 after culling × 100) was significantly decreased in the first generation at the highest dose. In the second generation, significant decreases in birth index (pups born per litter/implantation sites per litter) and mean litter size were observed at the lowest and highest doses, but there was no dose–response relationship. The body weight of pups at the highest dose was decreased in both generations on day 21. No abnormalities were observed at gross and histological examination of adults and pups of either generation. At termination, significant decreases in plasma and brain cholinesterase activity were reported in F0 females at all doses. In addition, erythrocyte cholinesterase activity was significantly decreased in females and males at the two higher doses. Brain cholinesterase activity was significantly decreased in F0 males at the highest dose. At termination of the F1 generation, plasma cholinesterase activity was significantly decreased in adult males and females at the two higher doses. Erythrocyte cholinesterase activity was significantly decreased in females at all doses and in males at the highest dose. Neonatal erythrocyte cholinesterase activity was significantly decreased in F1 males at the two higher doses on day 4 of lactation. On day 21 of lactation, significant decreases in brain and plasma cholinesterase activity were observed in male and female pups at the highest dose. In the F2 generation, significant decreases in plasma cholinesterase activity were observed in male and female pups, and significant decreases were found in brain cholinesterase activity in females at the highest dose and in males at the two higher doses on day 21 of lactation. A NOEL could not be identified in this study, as significantly decreased plasma and brain cholinesterase activities were seen at term in adult F0 females at the lowest dose and significantly decreased erythrocyte and brain cholinesterase activities in F1 adult females at the lowest dose. The Committee noted that the LOEL in this study (7.5 mg/kg bw per day) was lower than the NOEL of 30 mg/kg bw per day identified in a three-generation study of reproductive toxicity in rats by the Committee at its fifty-fourth meeting. It also noted, however, that cholinesterase inhibition was not evaluated in that study. Had that been done, it is reasonable to assume that the NOEL for that study would have been lower than 30 mg/kg bw per day. This assumption is supported by the NOEL of 5 mg/kg bw per day for inhibition of erythrocyte cholinesterase activity identified by the Committee at its fifty-fourth meeting in a 16-week study of toxicity in rats treated orally. On the basis of these considerations and the fact that inhibition of cholinesterase activity was the most sensitive effect in offspring in the present study, with an NOEL of 7.5 kg/kg bw per day, the Committee concluded that the reproductive toxicity of trichlorfon in rats had been adequately assessed.

In a study of developmental toxicity in guinea-pigs, designed to evaluate brain hypoplasia in offspring exposed in utero, trichlorfon was administered either by stomach tube or by subcutaneous injection. Oral doses ranging from 25 to 200 mg/kg bw per day were administered to groups of rats on days 40–44 of gestation in various regimens. Clinical signs of toxicity typical for this substance, including hypersalivation and hindlimb weakness, were observed in dams given trichlorfon at 100, 150 or 200 mg/kg bw. The NOEL for brain hypoplasia in offspring was 50 mg/kg bw.

In a study of the cytogenetic and developmental effects of trichlorfon on pre-implantation, mid-gestation and near-term mouse embryos and fetuses in vivo, groups of mice were given an intraperitoneal injection of 100 or 200 mg/kg bw 6 h after presumed conception. Developmental outcomes and micronucleus formation during the pre-implantation phase were assessed on day 3 of gestation, and developmental and aneugenic outcomes during the mid-gestation period were assessed on day 9 of gestation. On day 17 of gestation, the embryos were removed, sexed, weighed and examined for external malformations. No clinical signs of toxicity were observed in the treated mice. The mean number of cells in embryos in both treated groups was significantly lower than that in the control group, and the mean number of micronuclei was significantly increased in both treated groups compared with controls. The number of live embryos per dam was significantly lower among those given 100 mg/kg bw than in controls. The mean number of somites in the trichlorfon-treated groups was significantly lower than in controls. Significantly more fetuses with abnormal numbers of chromosomes were found to be associated with trichlorfon treatment. The incidence of external malformations and the body weights of male and female fetuses in the trichlorfon-treated groups were comparable to those of controls. This study provides evidence that exposure to trichlorfon around the time of fertilization could result in induction of micronuclei, aneuploidy and developmental retardation in embryos from pre-implantation to mid-gestation. Embryos with micronuclei or aneuploidy may no longer show abnormalities at term. The Committee noted that these effects resulted from intraperitoneal injection of doses 5000 to 10 000 times greater than the current ADI established for orally administered trichlorfon by the Committee at its fifty-fourth meeting.

The effects of trichlorfon on fertilization, spindle morphology and chromosomal segregation were studied in mouse oocytes exposed in vitro to a concentration of 50 µg/ml. Aberrant spindles were observed in maturing oocytes after 8 h of exposure, and the chromosomes appeared to be incapable of proper alignment at the equator. After 16 h, most of the treated oocytes had highly aberrant spindles, and the chromosomes were frequently unaligned and located at different distances from the poles. Such effects on spindles and chromosome alignment could have aneugenic effects on fertilized eggs and embryos. The Committee noted that effects on spindles leading to aneuploidy have thresholds. The Committee also noted that, assuming 100% bioavailability, systemic exposure to trichlorfon resulting from consumption of the entire ADI would be orders of magnitude lower than the dose used in this study. Exposure to trichlorfon at the ADI would therefore pose a negligible risk to human oocytes.

Aneuploidy induction was studied in sperm cells collected from groups of male mice 22 days after a single intraperitoneal injection of trichlorfon at a dose of 200, 300 or 400 mg/kg bw. A significant, dose-related increase in the percentage of sperm cells with an extra chromosome was observed at all doses. On the basis of this experiment, the Committee concluded that trichlorfon is a male mouse germ-cell aneugen in vivo. The Committee noted that these effects resulted from intraperitoneal injection of doses several orders of magnitude higher than the current ADI for trichlorfon established by the Committee at its fifty-fourth meeting.

The maximal effects of trichlorfon on soluble neuropathy target esterase activity and regional distribution of neuropathy target esterase and acetylcholinesterase were studied in brain and spinal cord of hens given trichlorfon at a single intravenous injection of 200 mg/kg bw. Peak inhibition of neuropathy target esterase activity occurred 6 h after dosing and ranged from 15 to 44%. No signs of delayed neuropathy were found during the 28-day observation period in four treated hens. Having reviewed several contemporary studies, the Committee at its fifty-fourth meeting also concluded that trichlorfon did not cause delayed neuropathy in hens.

A cluster of congenital anomalies was identified in a Hungarian village between 1989 and 1990. Of 15 live births, 11 (73%) were affected by abnormalities, and six were twin births. Four of the 11 affected infants had trisomy 21 (Down syndrome). Nine different physical abnormalities were observed. The mothers of all the affected infants reported having eaten fish during pregnancy. Several ponds around the village used for fish farming had been treated with a trichlorfon formulation at a level of 500 mg/l. The average concentrations of trichlorfon measured or estimated in the types of fish consumed ranged from 0.15 to 100 mg/kg. The exposure of fathers to trichlorfon was not evaluated in this study. This is the only known report of reproductive effects in humans possibly associated with oral exposure to trichlorfon, despite its widespread use as an anthelmintic.

The Committee acknowledged that the results of this study were not conclusive and provided limited evidence of a possible association between birth defects in humans and oral intake of trichlorfon in food. In addition, the published report did not include confirmation of the magnitude or frequency of intake or even whether intake had occurred. The Committee at its fifty-fourth meeting evaluated studies of developmental toxicity with trichlorfon conducted in four animal species. In these studies, teratogenic effects were seen only at very high, maternally toxic doses. In addition, as multigeneration studies of reproductive toxicity did not provide evidence of paternally transmitted teratogenicity, the Committee considered that the effect on exposed males had been assessed. The Committee at its present meeting reviewed the assessment by the Joint Meeting on Pesticide Residues in 1993 (WHO, 1994) of dichlorvos, the major metabolite of trichlorfon. That Meeting concluded that dichlorfos was not teratogenic in mice, rats or rabbits, even at doses that were toxic to maternal animals. In addition, at 12 mg/kg bw per day, dichlorvos had no reproductive effects in rats in a three-generation study. On the basis of these considerations, the Committee concluded that the information from the study in humans would not significantly affect its risk assessment of trichlorfon.

Chromosomal effects were studied in the lymphocytes of 31 people who had attempted suicide by taking unknown doses of trichlorfon. There appeared to be an increase in per cent aneuploidy in blood samples collected 3–6, 30 and 180 days after the incidents. An increase was also found in the rate of chromatid and chromosome-type aberrations. The Committee concluded that the intake that had resulted in these effects far exceeded the ADI established for trichlorfon by the Committee at its fifty-fourth meeting.

4. EVALUATION

The additional information reviewed by the Committee for this re-evaluation of trichlorfon included further information on its pharmacokinetics and genotoxic, reproductive and developmental toxicity. In addition, a NOEL for teratogenicity in guinea-pigs was established. The information did not provide evidence that any of these effects was more sensitive than inhibition of acetylcholinesterase activity, and the Committee concluded that inhibition of acetylcholinesterase activity is the most relevant end-point for establishing an ADI for trichlorfon.

The Committee at its fifty-fourth meeting established the ADI for trichlorfon on the basis of a NOEL of 0.2 mg/kg bw per day in a study of volunteers with Alzheimer disease treated with trichlorfon. In this study, the volunteers were given a loading dose of 0.5 mg/kg trichlorfon daily for 2 weeks, followed by a maintenance dose of 0.2 mg/kg per day for 8 weeks. The Committee at its fifty-fourth meeting concluded that the maintenance dose had not significantly enhanced the inhibition of erythrocyte cholinesterase activity established in patients by the loading dose. Therefore, it concluded that the maintenance dose was the NOEL in this study. Because the NOEL was derived from a study in humans, a safety factor of 10 was applied to the NOEL to derive the ADI.

The present Committee re-examined the basis on which it had established the ADI for trichlorfon at its fifty-fourth meeting and concluded that the dose it had identified then as the NOEL was nevertheless effective in maintaining the steady-state level of inhibition of erythrocyte cholinesterase activity and was therefore more appropriately considered a LOEL. The Committee concluded, however, that an ADI could be derived from this study by applying an additional factor of 10 to this LOEL. This conclusion is supported by supplemental information that included a linear dose extrapolation of data from a study of 27 volunteers with Alzheimer disease, who were treated with loading doses of 1.5–4 mg/kg bw per day for 6 days, followed by maintenance doses of 0.25–1 mg/kg bw per day for 15 days. The linear extrapolation resulted in an estimated NOEL for inhibition of cholinesterase activity in the range 0.1–0.2 mg/kg per day. This provides further support that the NOEL for inhibition of erythrocyte cholinesterase activity in humans is very close to 0.2 mg/kg bw per day. Furthermore, the Committee recalled that a clear NOEL of 0.2 mg/kg per day for inhibition of erythrocyte cholinesterase activity was established in a 10-year study of toxicity in monkeys treated orally, which it had evaluated at its fifty-fourth meeting. If that study had been selected as the basis for setting the ADI, a safety factor of 100 would have been applied, resulting in an ADI of 0–2 µg/kg bw. The present Committee thus amended the ADI for trichlorfon from 0–20 µg/kg bw to 0–2 µg/kg bw on the basis of the LOEL of 0.2 mg/kg bw per day for inhibition of erythrocyte acetylcholinesterase activity in humans treated orally and a 100-fold safety factor.

5. REFERENCES

Astroff, B.A., Freshwater, K.J. & Eigenberg, D.A. (1998) Comparative organophosphate-induced effects observed in adult and neonatal Sprague-Dawley rats during the conduct of multigeneration toxicity studies. Reprod. Toxicol., 8, 619–645.

Badawi, A.F. (1998) O6-Methylguanine and O6-methylguanine-DNA methyltransferase activity in tissues of BDF-1 mice treated with antiparasitic drugs. Toxicol. Lett., 94, 199–208.

Bieber, F., Orazem, J. & Mas, J. (1996) A double-blind, placebo-controlled, randomized, single-center study of the safety, tolerability and pharmacokinetics of metrifonate (BAY a 9826) in patients with probable Alzheimer’s disease. Unpublished report No. MMRR-1340 from Bayer, Wuppertal-Elberfeld, Germany. Submitted to WHO by Bayer Animal Health, Germany.

Cummings, J.L., Cyrus, P.A., Bieber, F., Mas, J., Orazem, J. & Gulanski, B. (1998) Metrifonate treatment of the cognitive deficits of Alzheimer’s disease. Metrifonate study group. Neurology, 51, 1214–1221.

Czeizel, A.E. (1994) Phenotypic and cytogenetic studies in self-poisoned patients. Mutat. Res., 313, 175–181.

Czeizel, A.E., Elek, C., Gundy, S., Metneki, J., Nemes, E., Reis, A., Sperling, K., Timar, L., Tusnady, G. & Viragh, Z. (1993) Environmental trichlorfon and cluster of congenital abnormalities. Lancet, 341, 539–542.

Dingemanse, J., Halabi, A., Kleinbloesem, C.H., Heinig, R. & Blume, H. (1999) Pharmacokinetics and pharmacodynamics of the acetylcholinesterase inhibitor metrifonate in patients with renal impairment. Ther. Drug Monit., 21, 310–316.

Heinig, R. & Dietrich, H. (1999) Pharmacokinetics of metrifonate and its metabolite dichlorvos in healthy volunteers and patients with renal impairment. Clin. Drug Invest., 18, 35–46.

Heinig, R. & Sachse, R. (1999) The effect of food and time of administration on the pharmacokinetic and pharmacodynamic profile of metrifonate. Int. J. Clin. Pharmacol. Ther., 37, 456–464.

Heinig, R., Boettcher, Z., Herman-Gnjidic, Z. & Pierce, C.H. (1999) Effects of a magnesium/aluminum hydroxide-containing antacid, cimetidine or ranitidine on the pharmacokinetics of metrifonate and its metabolite DDVP. Clin. Drug Invest., 17, 67–77.

Hjelde, T., Mehl, A., Schanke, T.M. & Fonnum, F. (1998) Teratogenic effects of trichlorfon (metrifonate) on the guinea-pig brain. Determination of the effective dose and the sensitive period. Neurochem. Int., 32, 469–477.

Madrigal-Bujaidar, E., Cadena, R.S., Trujillo-Valdez, V.M. & Cassani, M. (1993) Sister-chromatid exchange frequencies induced by metrifonate in mammalian in vivo and in vitro systems. Mutat. Res., 300, 135–140.

Mehl, A., Schanke, T.M., Johnsen, B.A. & Fonnum, F. (1994) The effect of trichlorfon and other organophosphates on prenatal brain development in the guinea pig. Neurochem. Res., 19, 569–574.

Mehl, A., Rolseth, V., Gordon, S., Bjoraas, M., Seeberg, E. & Fonnum, F. (2000) Brain hypoplasia caused by exposure to trichlorfon and dichloros during development can be ascribed to DNA alkylation damage and inhibition of DNA alkyltransferase repair. NeuroToxicology, 21, 165–174.

Schmid, T.E., Attia, S., Baumgartner, A., Nuesse, M. & Adler, I.D. (2001) Effect of chemicals on the duration of male meiosis in mice detected with laser scanning cytometry. Mutagenesis, 16, 339–343.

Sun, F.Y., Schmid, T.E., Schmid, E., Baumgartner, A. & Adler, I.-D. (2000) Trichlorfon induces spindle disturbances in V79 cells and aneuploidy in male mouse germ cells. Mutagenesis, 15, 17–24.

Tian, Y., Xie, X., Piao, F. & Yamauchi, T. (1998) Delayed neuropathy and inhibition of soluble neuropathy target esterase following the administration of organophosphorus compounds to hens. Tohoku J. Exp. Med., 185, 161–171.

Tian, Y., Ishikawa, H. & Yamauchi, T. (2000) Analysis of cytogenetic and developmental effects on pre-implantation, mid gestation and near-term mouse embryos after treatment with trichlorfon during zygote stage. Mutat. Res., 471, 37–44.

WHO (1994) Pesticide residues in food—1993 evaluations. Part II. Toxicology, WHO/PCS/94.4, Geneva.

Yin, H., Cukurcam, S., Betzendahl, I., Adler, I.-D. & Eichenlaub-Ritter, U (1998) Trichlorfon exposure, spindle aberrations and nondisjunction in mammalian oocytes. Chromosoma, 107, 514–522.



    See Also:
       Toxicological Abbreviations
       Trichlorfon (EHC 132, 1992)
       Trichlorfon (HSG 66, 1991)
       Trichlorfon (WHO Food Additives Series 45)
       TRICHLORFON (JECFA Evaluation)
       Trichlorfon (WHO Pesticide Residues Series 1)
       Trichlorfon (WHO Pesticide Residues Series 5)
       Trichlorfon (Pesticide residues in food: 1978 evaluations)
       Trichlorfon (IARC Summary & Evaluation, Volume 30, 1983)