Emea 2004
Emea 2004
Emea 2004
EMEA/MRL/915/04-FINAL
November 2004
1. Ivermectin belongs to the macrocyclic lactone class of endectocides and consists of a mixture of
two homologous compounds, 22,23-dihydroavermectin B1a (H2B1a, not less than 80%) and
22,23-dihydroavermectin B1b (H2B1b, not more than 20%). Avermectins are potent anthelmintic,
insecticidal and acaracidal compounds which, by increasing the membrane permeability to
chloride ions, mediate the paralysis of the nematodes and certain classes of ectoparasites.
Initially, the Committee established an ADI of 0.2 µg/kg bw/day, i.e. 12 µg/person/day based on a
NOEL of 0.1 mg/kg for maternotoxicity effects in a mouse teratogenicity study applying a safety
factor of 500. In 1993, the Joint FAO/WHO Expert Committee on Food Additives (JECFA) re-
evaluated the ADI for ivermectin and concluded, on the basis of new human data, that the safety
factor applied to this same NOEL could be reduced to 100, resulting in an ADI of 1 µg/kg bw/day
corresponding to 60 µg/person/day. After reviewing this re-evaluation the CVMP established the
same revised ADI.
Ivermectin is currently included in Annex I of Council Regulation (EEC) No 2377/90 with MRLs
for bovine, porcine, ovine, Equidae and deer species as follows:
2/7
EMEA 2005
These results corresponded well with those from in vivo liver metabolism studies. In addition, a
group of non-polar metabolites was detected in fat, which yielded polar products on hydrolysis
that were similar to the ivermectin metabolites present in liver.
4. Acute studies were conducted in mice, rats, rabbits, dogs and monkeys. The typical signs of acute
ivermectin toxicity were all attributed to effects on the central nervous system. These were most
severe in CF-1 mice, which exhibited ataxia, bradypnoea and tremors. Death occurred from
approximately one to six hours after treatment. Ivermectin was more toxic in neonatal rats than in
young adults. This was believed to be due to the postnatal completion of the blood-brain barrier
in this species. In beagle dogs, mydriasis was the most sensitive indicator of toxicity. More severe
signs included ataxia and tremors. Deaths were preceded by a comatose state. Approximately
30% of collie dogs were highly sensitive to ivermectin (as estimated from reports from non-
recommended use of the drug). In immature rhesus monkeys no tremors or convulsions occurred.
The most sensitive indicator was vomiting, which occurred in one of four animals given
2.0 mg/kg bw. The steep dose response curve in rodents was not reproduced in monkeys.
5. Short-term studies were conducted in rats, dogs and monkeys. In a 14-week study in rats
ivermectin was administered orally to pregnant dams, splenic enlargement and bone-marrow
hyperplasia was noted in offspring at 0.8 and 1.6 mg/kg bw. The NOEL was 0.4 mg/kg bw. These
effects did not occur in other species.
In a 14-week oral study in beagle dogs (4/sex/group), mydriasis and slight weight loss were
observed at 1.0 and 2.0 mg/kg bw. Four dogs in the 2.0 mg/kg bw group developed tremors,
ataxia, anorexia and dehydration and were killed prior to the end of the study. The NOEL was
0.5 mg/kg bw.
In a two-week oral study, ivermectin was administered to neonatal monkeys at 0.04 and
0.1 mg/kg bw and to immature monkeys at 0.3, 0.6 and 1.2 mg/kg bw. No treatment-related
effects were observed. In a short-term escalating dose study in monkeys, an NOEL of 1 mg/kg
was identified.
6. Ivermectin is generally well-tolerated in the intended target species, with occasional coughing in
sheep and goats after oral administration, occasional oedema and pruritis in horses and transient
injection site pain in sheep. Neurotoxic effects similar to those seen in laboratory species may
occur in overdosage.
7. Three multigeneration studies were conducted in rats, but the first two were terminated because
of neonatal toxicity at all doses. In the third (three-generation study), neonatal toxicity was
observed at 0.4 mg/kg bw with increased neonatal mortality up to about 10 days post-partum and
decreased bodyweight in the survivors. A cross-fostering study indicated that the neonatal
toxicity was not related to in utero exposure but post natal exposure via maternal milk. There is
evidence that neonatal rats are hypersusceptible to avermectin toxicity
8. The developmental toxicity of ivermectin was investigated in mice, rats, rabbits and dogs. The
results demonstrated that teratogenic effects (cleft palates in mice, rats and rabbits, and clubbed
fore-feet without skeletal alterations in rabbits) were produced only at doses similar to those
causing severe maternal toxicity. The NOEL for teratogenicity in the most sensitive species and
strain, the CF-1 mouse was 0.2 mg/kg bw, while the NOEL for maternal toxicity was 0.1 mg/kg
bw. The CF-1 mouse has a genetic predisposition to avermectin toxicity. No teratogenic or
maternotoxic effects were observed in dogs given oral doses of 0.5 mg/kg bw every five or ten
days from days 5 to 40 of gestation.
9. Ivermectin was negative for mutagenic effects in a bacterial gene mutation study (up to
2000 µg/plate), an L5178Y mouse lymphoma assay (up to 1000 µg/ml), and a Unscheduled DNA
(deoxyribonucleic acid) Synthesis (UDS) study in human IMR-90 fibroblasts (to
1000 µg/ml). The two components were negative in a bacterial gene mutation study.
10. No carcinogenicity studies were performed. However, such data were not considered necessary
on the basis of the absence of structural alerts and the results of the mutagenicity studies.
3/7
EMEA 2005
11. No specific studies were provided concerning potential immunotoxicity. The results of laboratory
animals studies and clinical use in humans gave no indications of any effect on the immune
system.
12. No data on the effects of ivermectin of the human gut flora or micro-organisms used in food
processing were available. However, such data were not considered necessary for this class of
compound.
13. Ivermectin is widely used in humans for treatment of onchocerciasis and other parasitic diseases
at single or repeated doses of 0.15 to 800 mg/kg bw. Tolerance to the compound has been
assessed in healthy volunteers and in patients; adverse effects are usually mild and transient. In
particular, no effects on the central nervous system were observed. The main effects noted in field
and community based trials have been those arising from the death of the parasites, the so-called
Mazzotti reaction, which is characterised by arthralgia, pruritis, fever, hypertension, tachycardia,
headache and ocular changes. Neither in these studies nor during treatment for other parasitic
diseases has a subset of atypically sensitive individuals been detected. Furthermore, the adverse
effects experienced by the small number of persons accidentally exposed to doses (often of
veterinary preparations) higher than customary human doses are in keeping with those noted in
test animals.
A double blind, randomised, placebo controlled clinical trial was conducted to assess the safety
and tolerability of oral subacute repeat doses of up to 1.2 mg/kg bw and an acute dose of
2.0 mg/kg bw in 68 healthy adult male and female human subjects as a treatment for headlice. No
treatment related signs of toxicity were observed and in particular, there were no signs of
neurotoxicity such as emesis, mydriasis (measured by pupillometry) or ataxia, that were predicted
to be the most sensitive potential adverse effects based on the laboratory animal studies. A
conservative NOEL of 420 µg/kg bw was identified.
However, it was considered that the human study could not be used to establish the ADI because
the dosing regimen used. As indicated in paragraph 3 above, the time to reach steady state would
be much longer than the maximum treatment period used (7 days). The plasma steady state
concentration would be at least 1.5-fold higher than the levels achieved after administration on
days 1, 3, and 7 only. The relation of these kinetics to central nervous system concentrations is
unknown.
14. As indicated in the CVMP Summary Reports addressing the revision of the ADIs for abamectin
(EMEA/MRL/838/02-FINAL) and eprinomectin (EMEA/MRL/520/98-FINAL), the CF-1 mouse
is not directly relevant for human risk assessment due to the genetic predisposition of this strain
to avermectin toxicity. The data from reproduction studies in the rat is also considered an
unsuitable basis for setting an ADI, as there is evidence that neonatal rats are hypersusceptible to
avermectin toxicity.
Therefore, because the rodent data are now considered inappropriate for ivermectin risk
assessment, and the available human data were unsuitable, the dog toxicity data was considered
the most relevant for the establishment of the toxicological ADI. In considering the
pharmacokinetic data across species, the dog is seen as conservative – more sensitive than
humans and non-human primates – in predicting ivermectin toxicity. A possible explanation for
the greater sensitivity of the dog to ivermectin toxicity than humans is that there are clear
differences in the oral pharmacokinetic profiles. In contrast to humans, a four-fold increase in
dose from 0.5 to 2.0 mg/kg bw, results in over an 8.5-fold increase in plasma ivermectin
concentrations. In healthy humans, single dose pharmacokinetic parameters show near linear
behaviour at doses of 0.1 to 2.0 mg/kg bw. Ivermectin elimination in dogs is much slower than
humans with a t½ of 1.6 to 1.8 days, compared to 11.8 to 20.1 hours in humans. These data
suggest that ivermectin may reach threshold levels for overt toxicity more readily in dogs than
humans based on comparison of pharmacokinetics. Because of this it is considered that a safety
factor of 50 is adequate for the establishment of an ADI based on the NOEL of 0.5 mg/kg bw/day
in the 14-week dog study. This ADI is 10 µ g/kg bw/day (600 µ g/person per day).
4/7
EMEA 2005
The NOEL of 0.5 mg/kg bw/day from a 14-week repeat-dose study in dogs supports a new higher
ADI in combination with an ivermectin-specific uncertainty factor that appropriately accounts for
the uncertainties and species-specific differences associated with the use of the dog NOEL. The
available human and non-human primate toxicological data supported the established uncertainty
factor of 50.
The availability of new human safety data that accounts for the same central nervous system
neuropharmacological interactions that are present in the dog, provides reassurance that the ADI
established from the dog study is appropriate.
No new residue studies were submitted, a summary of those already available and previously
evaluated in the different animal species are enclosed:
15. Four residue studies in cattle were presented. In three studies cattle were dosed percutaneously
with 0.5 or 1.0 mg ivermectin/kg bw and in one study cattle were dosed subcutaneously with
0.2 mg/kg bw. These studies show that the highest concentrations of residues in tissues of cattle
were found in liver followed by fat, kidney and muscle except for the injection site muscle. Joint
FAO/WHO Expert Committee on Food Additives has calculated the percentage of the marker
residue (22,23-dihydroavermectin B1a) of the total residues in cattle 28 days post treatment. The
marker accounted for 67% in muscle, 37% in liver, 54% in kidney and 18% in fat of total
residues and the distribution of residues between tissues was 1:2:11:27 for muscle, kidney, fat
and liver, respectively.
16. In a radiometric residue depletion study, 12 pigs were given a single tritium labelled (C22, 23
positions) ivermectin dose of 0.4 mg/kg bw subcutaneously. 3 animals in each group were killed
on days 1, 7, 14 and 28 days after treatment. Fat total residue levels were the highest followed by
liver. Total residue levels in fat were 384, 152, 28 and 6; in liver were 199, 112, 22, and 3 µg/kg
on 1, 7, 14 and 28 days after treatment respectively. Kidney total residue levels were 106, 55 and
10 whereas muscle total residue levels were 43, 25 and 4 µg/kg after 1, 7 and 14 days
respectively. The average marker to total residue ratios were 0.27, 0.41, 0.3 and 0.39 for liver,
kidney, fat and muscle respectively.
In another radiometric residue depletion study, 15 pigs (male and female) were given tritium
labelled (C22, 23 positions) ivermectin at a dose rate of 0.1 mg/kg bw/day for 7 days in the feed.
Total radioactive residue levels in liver were 237.1, 43, 10.7, 4.1 and 2.7 µg/kg and those in fat
were 207.2, 63.6, 18, 8 and 4.5 µg/kg at 4 hours, 3, 7, 14 and 21 days after treatment
respectively.
In a residue depletion study where pigs were administered a single subcutaneous dose of
0.4 mg/kg bw, the highest residue levels were found at the injection site samples, followed by fat,
liver, kidneys and then muscle at all time points. Mean injection site residues were 12500, 5100,
1110, 2300, 2500 and 230 µg/kg at 1, 3, 5, 7, 10 and 14 days after treatment respectively. Peak
residue levels were found after 3 days in all of the other tissues. Liver residue levels were 67, 69,
53, 41, 23 and 13 µg/kg and fat residue levels were 74, 110, 91, 73, 47 and 24 µg/kg on days 1, 3,
5, 7, 10 and 14 days after treatment respectively.
17. In a radiometric residue depletion study, 12 sheep were given a single oral (intraruminal) tritium
labelled (C22, 23 positions) ivermectin dose of 0.3 mg/kg bw. Three animals were killed on days 1,
3, 5 and 7 days after treatment. The average total concentration in liver were 238, 125, 25,
44 µg/kg at 1, 3, 5 and 7 days after treatment respectively. At the same time points, average total
residue concentrations in fat were 307, 153, 63, 73; in kidney 72, 46, 12, 13 and in muscle were
55, 50, 9 and 10 µg/kg. The ratios of marker (H2B1a) to total residues at 3 days post treatment
were 0.51, 0.51, 0.44 and 0.52 for liver, fat, kidney and muscle, respectively.
In a residue depletion study, 30 sheep (male and female) were given oral doses of 0.3 mg
ivermectin/kg bw in a micellar vehicle. The highest residues (marker residue, H2B1a) were found
in fat followed by liver. Residue levels in liver were 72, 12, 11 and 8 µg/kg; in fat were 145, 32,
11 and 9 µg/kg; in kidneys were 30, 5, 2 and 1 µg/kg and in muscle were 20, 4, 2 and 2 µg/kg at
1, 3, 5 and 7 days after treatment respectively.
5/7
EMEA 2005
In a residue depletion study where sheep were administered a three subcutaneous doses of
0.3 mg/kg bw at weekly intervals, the highest residue levels were found at the injection site
samples, followed by fat, liver, kidneys and then muscle at all time points. Mean injection site
residues were 17000, 2900, 2300, 460 and 220 µg/kg at 3, 7, 10, 14 and 28 days after last
treatment respectively. Peak residue levels were found after 7 days in all of the other tissues.
Liver residue levels were 160, 190, 97, 55 and 7.2 µg/kg and fat residue levels were 230, 310,
180, 99 and 13 µg/kg at 3, 7, 10, 14 and 28 days after treatment respectively.
18. In a radiometric residue depletion study, 3 horses were given tritium labelled (C22, 23 positions)
ivermectin doses of 0.3 mg/kg bw. Two of the animals were administered orally and the
remaining one was administered intramuscularly. The animals were then slaughtered after 28
days and the mean total radioactivity levels in those administered orally was 2.64, 3.02, 3.1, 4.26,
4.11 and 3.52 µg/kg and in those administered intramuscularly was 43.2, 17.1, 14.4, 54.2, 47.4
and 36.1 for liver, kidneys, muscle, perirenal fat, omental fat and subcutaneous fat respectively.
Total radioactivity levels at the injection site was 64.4 µg/kg. The ratios of marker to total
residues were 0.12, 0.22 and 0.36 for liver, kidney and fat, respectively.
In a residue depletion study where horses were administered a single oral dose of 0.3 mg/kg bw,
the highest residue levels were found fat followed by liver. Fat residues were 80 and 10.9 µg/kg
and liver residues were 31 and 4 µg/kg on 7 and 14 days after treatment. Residue levels were only
detected after 7 days in kidney (15 µg/kg) and muscle (8.3 µg/kg).
19. Following a single percutaneously administered dose of 1 mg/kg (twice the recommended dose)
to red deer, the tissue residues decreased progressively with time with highest concentrations
found in fat followed by liver, muscle and kidney. The ratio of 22,23-dihydroavermectin B1a in
tissues at 28 days withdrawal, were in this study 6.8:2.5:1.3:1 for fat, liver, kidney and muscle. At
7 and 28 days after treatment the mean concentrations of 22,23-dihydroavermectin B1a of
residues in fat were 292 µg/kg and 13.2 µg/kg, respectively, in liver 180 µg/kg and 9.3 µg/kg,
respectively, in muscle 78 µg/kg and 1.4 µg/kg respectively and in kidney 78 µg/kg and
3.6 µg/kg, respectively.
A single subcutaneous dose of 0.2 mg ivermectin/kg bw was given to reindeer. The ratio of
22,23-dihydroavermectin B1a in tissues at 17 days withdrawal, were in this study 9.5:6.6:2.6:1
for fat, liver, kidney and muscle, respectively. The highest residue concentrations of 22,23-
dihydroavermectin B1a 10 and 17 days after treatment, respectively, were 362 µg/kg and 68
µg/kg in back fat, 71 µg/kg and 28 µg/kg in liver, 54 µg/kg and 13 µg/kg in kidney, 40 µg/kg and
11 µg/kg in muscle, and 44 µg/kg and 9 µg/kg in injection site muscle. The half-lives in the
tissues after a single subcutaneous treatment were 7.1 days in back fat, 2.9 days for the injection
site, 4.9 days for muscle, 5.8 days for liver and 5.7 days for kidney.
20. A routine analytical method validated in accordance with Volume 8 of the Rules Governing
Medicinal Products in the European Union based on HPLC with fluorescence detection was
presented in an internationally recognised format for quantifying 22,23-dihydro-avermectin B1a
residues in tissues from bovine, porcine, ovine, Equidae and deer. The limit of quantification for
all tissues for pigs, Equidae and ovine species was 5 µg/kg. For bovine species the limit of
quantification was 3 µg/kg for muscle, 5 µg/kg for fat and kidney and 3.6 µg/kg for liver. For
deer the limit of quantification was 2 µg/kg for all tissues. This method should be applicable to
other mammalians.
21. In accordance with the Notes for Guidance on Risk Analysis Approach for Residues of
Veterinary Medicinal Products in Food of Animal Origin (EMEA/CVMP/187/00-FINAL) and on
the Establishment of Maximum Residue Limits for Minor Animal Species
(EMEA/CVMP/153a/97-FINAL) it was possible to extrapolate the MRLs to all mammalian food
producing species.
6/7
EMEA 2005
Conclusions and recommendation
Having considered that:
• a revised toxicological ADI of 10 µg/kg bw/day (600 µg/person per day) was established,
• residues in edible tissues in cattle, pigs, sheep, horses and deer are below the ADI from early
time points,
• 22,23-dihydroavermectin B1a (H2B1a) is the marker residue in all tissues and species,
• the tissue distribution of residues and the overall ratios of marker to total residues were
generally similar with residue levels being the highest in fat and liver tissues. Horses and
pigs had slightly different marker/total residue ratios,
• residue concentrations were persistently low in non injection site muscle, therefore muscle
was considered unsuitable for monitoring purposes and no MRLs were established for
muscle,
• an analytical method for the monitoring of residues in tissues of all mammalian food
producing species was available;
the Committee recommends the modification of the current entries for ivermectin for bovine,
porcine, ovine, Equidae and deer, including reindeer, in Annex I of Council Regulation (EEC) No
2377/90 in accordance with the following table:
7/7
EMEA 2005