Registration Dossier

Toxicological information

Carcinogenicity

Currently viewing:

Administrative data

Description of key information

The following robust study summary is based on the data presented in Carcinogenicity section of this dossier and adresses solely sodium flouride. The genetic and carcinogenic relevant concern of zinc hexafluorosilicate originate exclusively from the flouride. Refer to read-across justification for details.

Under the conditions of these 2-year dosed water studies, there was equivocal evidence of carcinogenic activity of sodium fluoride in male F344/N rats, based on the occurrence of a small number of osteosarcomas in dosed animals. Equivocal evidence is a category for uncertain findings defined as studies that are interpreted as showing a marginal increase of neoplasms that may be related to chemical administration. There was no evidence of carcinogenic activity in female F344/N rats receiving sodium fluoride at concentrations of 25, 100, or 175 ppm (11, 45, or 79 ppm fluoride) in drinking water for 2 years. There was no evidence of carcinogenic activity of sodium fluoride in male or female mice receiving sodium fluoride at concentrations of 25, 100, or 175 ppm in drinking water for 2 years.

Key value for chemical safety assessment

Carcinogenicity: via oral route

Link to relevant study records

Referenceopen allclose all

Endpoint:
carcinogenicity: oral
Type of information:
other: evidence from degradation product
Adequacy of study:
supporting study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
comparable to guideline study with acceptable restrictions
Qualifier:
no guideline followed
Principles of method if other than guideline:
Shon-term toxicity studies of sodium fluoride were
conducted in rodents to determine appropriate doses
for Ionger studies. Male and female F344/N rats
were obtained from the Charles River Labaratones
(Portage, MI) and were observed for 6 days; male
and female B6C3F1 mice were obtained from the
same source and were observed for 25 days. The
rats and mice were 5 weeks old when placed on
study. Animals were assigned to cages, then cages
were assigned to control or dose groups by random
number tables.
Groups of five rats and mice of each sex received
0, 50, 100, 200, 400, or 800 ppm sodium fluoride in
deionized water ad libitum for 14 consecutive days.
Animals were housed five per cage, witb semisyntbetic
low fluoride (12.7 to 14 ppm) feed available
ad libitum. Water consumption was recorded every
3 days for rats, every 4 days for mice. The rats and
mice were observed twice daily for morbidity,
mortality, and signs of toxicity; they were weighed at
tbe beginning of the studies, at the end of tbe first
week, and at necropsy. Details of study design and
animal maintenance are presented in Table 1. All
animals were necropsied, and tissues were examined
for gross Iesions. Microscopic examination of
tissues was not performed.
GLP compliance:
not specified
Specific details on test material used for the study:
Sodium fluoride was obtained from Apache Chemical,
Inc. (Seward, IL) in two lots. One Iot was used
for the 14-day and 6-montb studies (Iot no. A-
06255) and anotber for tbe 2-year studies (Iot no.
A022085). Identity, purity, and stability analyses
were conducted on tbese lots by Midwest Research
Institute (Kansas City, MO). Details of these
analyses are presented in Appendix J. Tbe study
chemical, a white crystalline powder, was identified
as sodium fluoride and was found to be at least
99% pure, as determined by elemental analysis, Karl
Fiseber water analysis, spark source mass spectrometry,
and titration of acidic components. During
the 2-year studies, the stability of tbe bulk chemical
was monitored by Galbraitb Labaratories (Knoxville,
TN). No degradation of the study material was
detected througbout tbe studies.
Species:
other: rats and mice
Strain:
other: F344/N rats and B6C3F1 mice
Details on species / strain selection:
Cbarles River Laboratories
(Ponage, MI)
Sex:
male/female
Details on test animals and environmental conditions:
Temperature: 21 -23 °C
Humidity: 40%-60%
Fluorescent light: 12 hours/day
Room air changes: 1S/hour
Route of administration:
oral: drinking water
Vehicle:
water
Details on exposure:
0, 5O, 100, 200, 400, or 800 ppm sodium
fluoride in deionized water, available ad
Libidum
Analytical verification of doses or concentrations:
not specified
Details on analytical verification of doses or concentrations:
Througbout all studies, dose formulations were
prepared by mixing appropriate amounts of sodium
fluoride with deionized water. Stability studies
indicated tbat sodium fluoride at a concentration of
25 ppm in deionized water was stable under Simulated
animal dosing conditions for up to 3 weeks
when stored in the dark at room temperature
(25° C). Details of the preparation and storage of
dose formulations of sodium fluoride are presented
in Appendix J.

0, 50, 100, 200, 400, or 800 ppm sodium
fluoride in deionized water, available ad
Libidum
Duration of treatment / exposure:
14 days (7 da)'llweek)
Frequency of treatment:
0, 50, 100, 200, 400, or 800 ppm sodium
fluoride in deionized water, available ad
Libidum
Dose / conc.:
50 ppm
Dose / conc.:
100 ppm
Dose / conc.:
200 ppm
Dose / conc.:
400 ppm
Dose / conc.:
800 ppm
Control animals:
yes, concurrent no treatment
Details on study design:
-
Positive control:
-
Observations and examinations performed and frequency:
Observed twice daily for mortality and
morbidity; weighed initially, at the end of
the first week, and at termination; clinical
Observations recorded daily. Water
consumption by cage recorded every 3
days for rats, every 4 days for mice.
Other examinations:
Necropsy and tissue collection performed
on all animals.
Clinical signs:
effects observed, treatment-related
Description (incidence and severity):
RAT: The following signs of toxicity were noted in all
animals in the two highest dose groups: dehydration
and lethargy by day 4 and hunched posture by
day 5. In addition, reduced water consumption was
recorded among the two highest dose groups. Daily
water consumption recorded by cage for males and
females in the second highest dose group (400 ppm)
was approximately 70% that of controls; for high-dose males (800 ppm), water consumption was
approximately 50% that of controls, and for highdose
females, approximately 25%.
Dermal irritation (if dermal study):
not specified
Mortality:
mortality observed, treatment-related
Description (incidence):
MICE:
All mice
survived to scheduled termination, except two highdose
males that died on days 4 and 6.

RAT:One female
from the second highest dose group (400 ppm) died
on day 6; all male rats in the high-dose group
(800 ppm) died by day 7; and all female rats in the
high-dose group died by day 10. All groups, male
and female, surviving to the end of the studies
gained weight except the group receiving 400 ppm
Body weight and weight changes:
effects observed, treatment-related
Description (incidence and severity):
MICE:
two highdose
males that died on days 4 and 6. Among the
male mice, weight changes were variable, with the
high-dose group having a significant decrease in
body weight. Among female mice, weight Iosses
occurred only in the high-dose group.


RAT:
All groups, male
and female, surviving to the end of the studies
gained weight except the group receiving 400 ppm.
In this group, 4/5 males lost from 5%-31% of their
initial body weight, and 3/4 females lost from 10%-
29% of their initial body weighL
Food consumption and compound intake (if feeding study):
not specified
Food efficiency:
not specified
Water consumption and compound intake (if drinking water study):
effects observed, treatment-related
Description (incidence and severity):
MICE:
Reduced water
consumption was recorded for high-dose males and
females. Daily water consumption averaged approximately
30% for high-dose males and 60% for highdose
females in comparison with controls.


RAT:
In addition, reduced water consumption was
recorded among the two highest dose groups. Daily
water consumption recorded by cage for males and
females in the second highest dose group (400 ppm)
was approximately 70% that of controls; for high-dose males (800 ppm), water consumption was
approximately 50% that of controls, and for highdose
females, approximately 25%
Ophthalmological findings:
not specified
Haematological findings:
not specified
Clinical biochemistry findings:
not specified
Urinalysis findings:
not specified
Behaviour (functional findings):
not specified
Immunological findings:
not specified
Organ weight findings including organ / body weight ratios:
not specified
Gross pathological findings:
no effects observed
Description (incidence and severity):
MICE:
No consistent significant gross lesions were noted in
any of the surviving mice at scheduled necropsy.
The tissues of these animals were not evaluated
microscopically.

RAT:
No significant gross lesions were seen at necropsy in
any groups of rats. Tissues from the animals were
not evaluated microscopically
Neuropathological findings:
not specified
Histopathological findings: non-neoplastic:
not specified
Histopathological findings: neoplastic:
not specified
Other effects:
not specified
Conclusions:
In the 14-day studies, all male and female rats and
\ several male mice given water containing 800 ppm
died ( concentrations are expressed as sodium
ßuoride; tluoride ion is abaut 45% of the sodium
salt on a weight basis); 115 female rats given 400
ppm also died. No gross lesions were observed at
necropsy; tissues were not examined microscopically.
Executive summary:

Rats and mice received sodium

fluoride in drinking water at concentrations as high

as 800 ppm. (Concentrations are expressed as

sodium fluoride; fluoride ion is 45% of the sodium

salt by weight) In the high-dose groups, 5/5 male

and 5/5 female rats and 2/5 male mice died; one

female rat given 400 ppm in the drinking water also

died before the end of the studies. No gross lesions

were attributed to sodium tluoride administration.

Endpoint:
carcinogenicity: oral
Type of information:
other: evidence from degradation product
Adequacy of study:
key study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
comparable to guideline study with acceptable restrictions
Qualifier:
no guideline followed
Principles of method if other than guideline:
Groups of 100 rats and mice of each sex received 0 or 175 ppm sodium fluoride and groups of 70 rats and mice of each sex received 25 or 100 ppm sodium fluoride in deionized drinking water ad libitum for up to 103 weeks. Interim sacrifices of ten animals of each sex per dose group from each species occurred at 24 weeks for mice, at 27 weeks for rats, and at 66 weeks for both species. An additional group of 50 animals of each sex and species was included to provide paired (agematched) controls. These animals received deionized drinking water. During every study week that one or more animals from any group receiving
sodium fluoride-supplemented water was found dead or killed in a moribund condition, one animal of the same species and sex was chosen at random from
the paired control group and killed.
GLP compliance:
not specified
Specific details on test material used for the study:
Sodium fluoride was obtained from Apache Cbemical, Inc. (Seward, IL) in two lots. One Iot was used for the 14-day and 6-month studies (Iot no. A-06255) and anotber for the 2-year studies (Iot no. A022085). Identity, purity, and stability analyses were conducted on ttese lots by Midwest Research
Institute (Kansas City, MO). Details of these analyses are presented in Appendix J. Tbe study chemical, a white crystalline powder, was identified as sodium fluoride and was found to be at least 99% pure, as determined by eiemental analysis, Karl Fisher water analysis, spark source mass spectrometry, and titration of acidic components. During the 2-year studies, the stability of the bulk cbemical was monitored by Galbraitb Labaratories (Knoxville, TN). No degradation of the study material was detected throughout tbe studies.
Species:
other: rat and mice
Strain:
other: male and female F344/N rats and B6C3F1 mice
Details on species / strain selection:
24-week interim sacrifice: Mice, 30 weeks
27-week interim sacrifice: Rats, 32 weeks
66-week interim sacrifice: 71 weeks
105-week scheduled termination: Rats,
110 weeks; Mice, 111 weeks
Base studies: 80, 50, 50, and 80 males
and females of each species for control,
low-dose, mid-dose, and high-dose groups;
10 additional animals per sex for each
species and dose group for each interim
evaluation (24 weeks for mice, 27 weeks
for rats, 66 weeks for both rats and
mice).
Paired (age-matched) controls: 50 males
and 50 females of each species
Animals assigned to weight classes, then
randomized to test and control groups by
partitioning algorithm using Xybione
Pathology/Toxicology Data System.
Sex:
male/female
Details on test animals and environmental conditions:
The male and female F344/N rats and B6C3F1 mice used in these studies were obtained from the National Cancer Institute's Frederick Cancer
Research Facility (Frederick, MD). Rats and mice were 4 weeks old upon arrival at the study laboratory. After all animals were quarantined for 12 to
13 days, a complete necropsy was performed on 20 rats of each sex and on three male and two female mice to assess their health status. Serologie
analyses were performed on samples drawn from five animals of each sex and species on two different dates during quarantine and on sentinel animals at 6, 12, and 18 months. Details of animal health monitoring are presented in Appendix N. The rodents were placed on study when they were 6 weeks old. Rats were housed five per cage; mice were housed individually. Specially formulated low fluoride feed and deionized water were available ad libitum.
Route of administration:
oral: drinking water
Vehicle:
water
Details on exposure:
0, 25, 100, or 175 ppm sodium fluoride in deionized water, available ad libitum
Analytical verification of doses or concentrations:
not specified
Details on analytical verification of doses or concentrations:
Throughout all studies, dose formulations were prepared by mixing appropriate amounts of sodium fluoride with deionized water. Stability studies indicated that sodium fluoride at a concentration of 25 ppm in deionized water was stable under Simulated animal dosing conditions for up to 3 weeks when stored in the dark at room temperature (25° C). Details of the preparation and storage of dose formulations of sodium fluoride are presented in Appendix J.
During the 2-year studies, the study Iabaratory conducted periodic dose formulation analyses, utilizing a potentiometric method with a fluoride ion electrode as described in Appendix J. These analyses were performed weekly on all dose formulations during approximately the first 6 months of the 2-year studies and then every 8 weeks for the duration of these studies (Appendix J, Tables J2, J3, and J4). These analyses indicated that all dose formulations were within ± 10% of target concentrations throughout the studies. Results of periodic referee analyses performed by Midwest Research Labaratory were in agreement with the results from the study Iabaratory (Appendix J, Table J5). Analyses
of deionized water for pH and fluoride concentration were all within acceptable Iimits of pH~5 (except on four occasions) and fluoride concentration <=0.1 ppm.

0, 25, 100, or 175 ppm sodium fluoride in deionized water, available ad libitum
Duration of treatment / exposure:
103 weeks (7 days/week)
Frequency of treatment:
0, 25, 100, or 175 ppm sodium fluoride in deionized water, available ad libitum
Post exposure period:
-
Dose / conc.:
25 ppm
Dose / conc.:
100 ppm
Dose / conc.:
175 ppm
No. of animals per sex per dose:
Base studiea: 80, 50, 50, and 80 males and females of each species for control, low-dose, mid-dose, and high-dose groups;
10 additional animals per sex for each species and dose group for each interim evaluation (24 weeks for mice, 27 weeks for rats, 66 weeks for both rats and
mice).
Paired (age-matched) controls: 50 males and 50 females of each species
Control animals:
yes, concurrent no treatment
Positive control:
-
Observations and examinations performed and frequency:
Observed twice daily for mortality and morbidity; weighed initially, weekly through week 13, monthly thereafter, clinical observations recorded weekly through week 13, monthly thereafter.
Every 4 weeks, feed and water consumption rec:rded for a 1-week period.
Other examinations:
Necropsy performed on all animals; complete histopathologic examination performed on all animals dying spontaneously or terminated because of
moribund condition. in addition to tissue masses, gross lesions, and associated regional lymph nodes. the following organs and/or tissues were included in
complete histopathological examinations: adrenals, bone (femur, humerus, mandible, maxilla, tibia, and vertebra), bone marrow, brain (frontal cortex and
basal ganglia, parietal corta and thalamus, cerebellum and pons), clitoral gland, epididymis, esophagus, eyes (when grossly abnormal), gallbladder (mice
only), heart, kidney, large intestines (cecum, colon, rectum), liver, lung with bronchi, lymph nodes (mandibular, mesenteric), mammary glands, nasal cavity
and turbinates, ovarics, pancreas, parathyroid, pharynx (when grossly abnormal), pituitary, preputial gland, prostate, salivary gland, sciatic nerve (when neurologic signs were present), seminal vesicles, skeletal muscle (thigh). skin, small intestincs (duodenum, ileum. jejunum), spinal cord (when neurologic signs were present), spleen, stomach (including forestomach and glandular stomach), teeth, testes, thymus, thyroid, trachea, urinary bladder, and uterus.
For all interim sacrifices, weights were recorded for liver, right kidney, left kidney, and brain. At 24-week and 27-week interim sacrifices, complete
histopathologic examination (including supplemental examination of right tibia, right humerus, thoracic vertebrae [7 ,8,9), maxilla, mandible, and incisors) performed on all animals in the 0 and 175 ppm dose groups; gross lesions examined in animals in the 25 and 100 ppm dose groups. At 66-week interim sacrifice, complete histopathologic examination (including supplemental examination of bones and teeth) performed on all animals. At 105-week scheduled termination, complete histopathologic examination (including supplemental examination of bones and teeth) performed on all animals.
Clinical signs:
effects observed, non-treatment-related
Description (incidence and severity):
MICE:
While numerous clinical signs were recorded over
the course of these studies, most occurred with such
low frequency or with such similarity across dosed
and control groups that they were not considered
related to treatment. The exception was white
discoloration of the teeth at the higher exposure
Ievels (Table 18). This abnormality occurred earlier
in the high-dose groups (day 74), later in mice
receiving lower concentrations (from day 81 to 200),
and much later in control animals (day 508).



RAT:
While numerous clinical signs were recorded during
these studies, most occurred with such low frequency
or with such similarity across dosed and control
groups that they were not considered related
to treatmenL The exceptions were abnormalities in
the teeth of rats at the two highest exposure Ievels
(Table 7).
Dermal irritation (if dermal study):
not specified
Mortality:
mortality observed, non-treatment-related
Description (incidence):
MICE:
Estimates of the probabilities of survival of male
and female mice administered sodium fluoride in
drinking water at the doses used in these studies
and those of the vehicle conttols are illustrated in
Kaplan-Meier curves (Figure 6). Overall survival
Information is given in Table 19. No significant
chemical-related effects on survival were observed.


RAT:
Estimates of the probabilities of survival of male
and female rats administered sodium fluoride in
drinking water at the doses used in these studies
and those of tbe vebicle controls are illustrated in
Kaplan-Meier curves (Figure 3). Overall survival
information is given in Table 8. No significant
chemical-related effects on survival were observed
Body weight and weight changes:
effects observed, non-treatment-related
Description (incidence and severity):
MICE:
No notable chemical-related effects were
observed; however, the maximum mean
body weights achieved (51.4 g for control males;
57.4 g for control females) were markedly higher
than the average peak body weights achieved by
historical controls ( 42.5 g for males and 41.7 g for
females; see Haseman et al., 1985).


RAT:
No significant chemically related
differences in body weights were observed.
Food consumption and compound intake (if feeding study):
effects observed, non-treatment-related
Description (incidence and severity):
MICE:
Average daily feed consumption for control and
treated groups ranged from 4.9 to 5.2 g for males
and 5.4 to 5.8 g for females (data on file at
NIEHS). Administration of sodium fluoride in
drinking water at the concentrations used in these
studies had no effect on feed consumption.
Deionized drinking water was the vehicle for administering
sodium fluoride to mice. Average daily
water consumption for control and treated groups
ranged from 4.1 to 4.2 g for males and 4.4 to 4.6 g
for females (Appendix L, Tables I.J and U).
Administration of sodium fluoride in drinking water
at the concentrations used in these studies had no
effect on water consumption. Estimated daily
ingestion of the chemical throughout the studies is
presented in Appendix L (Tables I.J and U ).
When averaged over the 2-year studies, the daily amounts of sodium fluoride ingested were 2.4 mg/kg
for low-dose males, 9.6 mg/kg for mid-dose males,
16.7 mg/kg for high-dose males, 2.8 mg/kg for lowdose
females, 11.3 mg/kg for mid-dose females, and
18.8 mg/kg for high-dose females.



RAT:
Average daily feed consumption for control and
treated groups ranged from 17.2 to 17.4 g for males
and 11.2 to 11.3 g for females ( data on file at
NIEHS). Administration of sodium fluoride in
drinking water at the concentrations used in these
studies bad no effect on feed consumption.
Deionized drinking water was the vehicle for administering
sodium fluoride to rats. Average daily water
consumption for control and treated groups ranged
from 19.8 to 21.2 g formales and 13.1 to 13.6 g for
females (Appendix L, Tables L1 and L2). Administration
of sodium fluoride in drinking water at the
concentrations used in these studies bad no effect
on water consumption. Estimated daily ingestion of
the chemical throughout the studies is presented in
Table L1 for male rats and in Table L2 for female
rats. When averaged over the. 2-year studies, the
daily amounts of sodium fluoride ingested were
1.3 mg/kg for low-dose males, 5.2 mg/kg for middose
males, 8.6 mg/kg for high-dose males,
1.3 mg/kg for low-dose females, 5.5 mg/kg for middose
females, and 9.5 mg/kg for high-dose females
Food efficiency:
not specified
Water consumption and compound intake (if drinking water study):
not specified
Ophthalmological findings:
not specified
Haematological findings:
effects observed, non-treatment-related
Description (incidence and severity):
MICE:
Hematology and clinical chemistry data were collected
for an animals sacrificed at the 24-week and
66-week interim sacrifices. Bone fluoride concentrations
were measured for all animals sacrificed at
24 weeks and 66 weeks and for selected animals at
the termination of the studies. Results of these
studies are presented in Appendix I and in Figure 5.
There were no biologically significant differences in
hematologic parameters or in serum concentrations
of calcium or phosphorous in dosed versus control
male or female mice at the 24-week or 66-week
interim evaluations (Tables 19 and 110). Serum
alkaline phosphatase activity was increased mildly at
24 weeks and moderately at 66 weeks in high-dose
female mice (Tables 19 and 110). For an treated
groups, fluoride concentrations in bone were dose
and age related, and were significantly increased
over control values for an evaluation periods




RAT:
Hematology, clinical chemistry, urinalysis and urine
concentrating abillty, and fluoride concentrations in
serum and urine were measured in all animals
sacrificed at 27 and 66 weeks. Fluoride concentrations
in bone were measured for all animals sacrificed
at 27 and 66 weeks and for selected animals at
termination of the studies. Results of these
measurements are presented in Appendix I and in
Figure 2.
There were no biologically significant differences in
hematologic indices, serum concentrations of phosphorus
or calcium, or alkaline phosphatase activity
among dosed and control male or female rats at the
TABLE 7
Sodium Fluoride, NTP TR 393
27-week or 66-week interim evaluations (Tables 17
and 18). Serum fluoride concentrations were
increased over control values in females receiving
drinking water containing 100 or 175 ppm sodium
fluoride at 27 weeks and in all exposed males and
females at 66 weeks (Table 15). These increases
ranged as high as almost threefold over control
values in high-dose rats. Urinalysis results did not
indicate biologically significant effects related to
fluoride administration with the possible exception
of a small increase in calcium excretion in highdose
female rats at both time points (Tables 111 and
112). A dose-related increase was observed in the
fluoride concentration of urine from male and
female rats at both the 27-week and 66-week interim
evaluations (Table 16). For all treated groups, doserelated
tluoride concentrations in bone were significantly
increased over control values for all evaluation
periods. Fluoride content of bone also
increased as a function of age (Table 13 and
Figure 2)
Clinical biochemistry findings:
not specified
Urinalysis findings:
effects observed, non-treatment-related
Description (incidence and severity):
Hematology, clinical chemistry, urinalysis and urine
concentrating abillty, and fluoride concentrations in
serum and urine were measured in all animals
sacrificed at 27 and 66 weeks. Fluoride concentrations
in bone were measured for all animals sacrificed
at 27 and 66 weeks and for selected animals at
termination of the studies. Results of these
measurements are presented in Appendix I and in
Figure 2.
There were no biologically significant differences in
hematologic indices, serum concentrations of phosphorus
or calcium, or alkaline phosphatase activity
among dosed and control male or female rats at the
TABLE 7
Sodium Fluoride, NTP TR 393
27-week or 66-week interim evaluations (Tables 17
and 18). Serum fluoride concentrations were
increased over control values in females receiving
drinking water containing 100 or 175 ppm sodium
fluoride at 27 weeks and in all exposed males and
females at 66 weeks (Table 15). These increases
ranged as high as almost threefold over control
values in high-dose rats. Urinalysis results did not
indicate biologically significant effects related to
fluoride administration with the possibte exception
of a small increase in calcium excretion in highdose
female rats at both time points (Tables 111 and
112). A dose-related increase was observed in the
fluoride concentration of urine from male and
female rats at both the 27-week and 66-week interim
evaluations (Table 16). For all treated groups, doserelated
tluoride concentrations in bone were significantly
increased over control values for all evaluation
periods. Fluoride content of bone also
increased as a function of age (Table 13 and
Figure 2)
Behaviour (functional findings):
not specified
Immunological findings:
not specified
Organ weight findings including organ / body weight ratios:
effects observed, non-treatment-related
Description (incidence and severity):
MICE:
Organ weights for the brain, right kidney, left
kidney, and liver were recorded for mice sacrificed
at 24 weeks and 66 weeks. Group mean organ
weights and organ-weight-to-body-weight ratios are
presented in Appendix 0. No changes in organ
weights were observed that were attributed to
sodium ßuoride administration


RAT:
Organ weights for tbe brain, right . kidney, left
kidney, and liver were recorded for rats sacrificed at
27 weeks and 66 weeks. Group mean organ weights
and organ-weight-to-body-weight ratios are presented
in Appendix G. There were no changes in organ
weights that appeared related to sodium fluoride
administration.
Gross pathological findings:
not specified
Neuropathological findings:
not specified
Histopathological findings: non-neoplastic:
not specified
Histopathological findings: neoplastic:
not specified
Other effects:
effects observed, non-treatment-related
Description (incidence and severity):
MICE:
Summaries of the incidence or neoplasms and
nonneoplastic lesions, individual animal tumor
diagnoses, statistical analyses of primary tumors that
occurred with an incidence of at least 5% in at least
one animal group, and historical control incidence
for the neoplasms identified by the Pathology
Working Oroup are presented in Appendix C for
male mice and Appendix D for female mice.
Summaries of the incidence of neoplasms and
nonneoplastic lesions in male and female mice at
the interim evaluations are presented in Appendix E
5'
for the 24-week sacrifice and in Appendix F for the
66-week sacrifice. Findings of note were in the
hematopoietic system, harderian gland, Jung, pars
distalis of the pituitary gland, and liver.
Hematopoietic System: The incidence of malignant
Iymphoma (an types) and malignant Iymphoma and
histiocytic sarcoma combined were marginally
increased in female mice receiving 175 ppm sodium
fluoride (Table 20). Histiocytic sarcoma is a term
used synonymously with malignant Iymphoma,
histiocytic type, in previous NTP studies. The
precise origin of this neoplasm is uncertain, but
both terms suggest a histiocytic origin. Although
the incidence in the high-dose group was increased
relative to controls, malignant Iymphoma has
occurred with a variable incidence rate in NTP
historical controls. Moreover, the incidence in the
high-dose group is similar to the mean rate and well
within the range of historical untreated controls at
the study laboratory (145/419, 34.6%, range
18%-48%) and all NTP laboratories combined
(639(2209, 31.4%, range 10%-74%). Therefore, the
marginal increase in malignant Iymphomas in mice
was not considered related to sodium fluoride.
Liver: The incidences of hepatoceUular neoplasms
in an groups of dosed and control male and female
mice were higher than those of historical controls in
previous NTP studies (Table 21, Appendixes C4.
D4b ). Five liver neoplasms in dosed male mice and
four in dosed female mice were diagnosed by the
laboratory pathologist as hepatocholangiocarcinoma.
Although one of these and one other lesion identified
as an hepatocholangiocarcinoma in a paired
control female (Table 21) clearly demonstrated areas
of biliary differentiation, the others contained welldefined
populations of cells which more closely
resembled embryonal liver cells than biliary cells.
1be PWO thought that the latter neoplasms were
more appropriately diagnosed as hepatoblastoma. In
both types of neoplasms the biliary or embryonal
ceu populations represent phenotypic variants within
a primary liver neoplasm that is otherwise characteristic
of a hepatocellular carcinoma. Malignant
neoplasms commonly contain a heterogenaus population
of cells, some of which may be relatively
undifferentiated and therefore resemble stem cells or
which demoostrate divergent differentiation.
Although hepatocellular neoplasms with the
embryonal cell type (Le. hepatoblastomas) occur
rarely (historical control incidences of 0/2197 in males and 1/2202 in females) and occurred more
frequently in mice receiving sodium fluoride, the
overall incidences of primary hepatocellular neoplasms
in males were similar or somewhat decreased
among dosed groups compared to controls. Thus,
the slight numerical increase in hepatoblastomas was
not considered biologically significant.
Negative Trends: Among male mice, there was a
dose-related decrease in harderian gland adenomas
(1(19; 2150; 0/51; 1/80); the incidence in the control
group was nearly threefold higher than is typically
seen in historical control groups (3.2%) (Haseman
et a/., 1984). In female mice, a negative dose related
trend was seen in adenomas of the pituitary
gland (pars distalis, 25/80; 7/51; 8/50; 13(19).
Neither of these decreases was considered to be
related to chemical administration.
Teeth: Dentine dysplasia occurred in bath dosed
and control groups of male and female mice
(Appendixes C5 and 05). The incidence of this
lesion was significantly greater in high-dose than in
control male mice (62(19; 44/50; 43/51; 73/80;
P=0.016).
Other lesions were incidental or part of spontaneaus
disease complexes of mice. There was no alteration
in the incidence or severity of these lesions in the
treated versus control animals, and they were
histopathologically typical of those commonly seen
in this strain of Iabaratory mouse.




RAT:
Pathology and Statistical Analyses of Results
Summaries of the incidence of neoplasms and
nonneoplastic lesions, individual animal tumor
diagnoses, statistical analyses of primary tumors tbat
occurred with an incidence of at least 5% in at least
one animal group, and bistorical control incidence
for neoplasms of interest are presented in
Appendix A for male rats and Appendix B for
female rats. Summaries of tbe incidence of neoplasms
and nonneoplastic lesions in male and female
rats at the interim evaluations are presented in
Appendix E for. the 27 -week sacrifice and in
Appendix F for tbe 66-week sacrifice. Findings of
note were in bone, mesenchymal soft tissue, oral
mucosa, teeth, tbyroid gland, skin, and uterus.
Bone: Tbe bones examined microscopically included
tbe proximal portians of tbe humerus, femur, and
tibia; a tboracic vertebra; tbe maxilla, incisive, and
nasal bones included in tbe sections of nose; and
the mandible. All but tbe mandible were decalcified,
routinely processed, and stained witb bematoxylin
and eosin; the mandible was processed and
sectioned witbout decalcification. Lesions observed
by gross examination at necropsy also were
sectioned and examined microscopically.
Nonncoplastic lesions of bone occurring in control
and exposed male and/or female rats included
fibrous osteodystropby and osteosclerosis. Fibrous
osteodystropby was always associated with advanced
nephropatby, principally in male rats, and was
considered to be due to renal secondary hyperparatbyroidism.
Osteosclerosis is a spontaneaus bone
disease of unkDown cause that occurs in aging
Sodium Fluoride, NTP TR 393
F344/N rats, primarily females. Histologically, it is
similar to a congenital, hereditary disease called
osteopetrosis, wbicb occurs in some strains of rat as
weil as in otber animals. Tbe incidence of osteosclerosis
was increased in female rats receiving
175 ppm sodium tluoride relative to untreated
controls (6/80 control, 18/81 high-dose, P=0.04). In
every rat with osteoscleroshis in these studies, most
of the bones examined microscopically were affected
to some degree. There was an increase in the
amount of trabecular bone in the diaphysis and
occasionally extending into the diaphysis and/or
epiphysis of the vertebrae and long bones, which
varied from immature woven bone with thick
osteoid seams to dense lamellar bone. The more
severe lesions were detected in the radiographs of
affected animals.
Osteosarcomas of the bone were observed in one
male receiving 100 ppm and in three males receiving
175 ppm (Table 9). None occurred in control or
low-dose male rats or in female rats. The osteosarcomas
occurred with a significant dose response
trend; pairwise comparison of the incidences in the
dosed groups versus control were not significant.
One male rat (175 ppm, CID#0713) with a vertebral
osteosarcoma exhibited posterior paralysis due
to invasion of the spinal cord by the neoplasm.
There were no other clinical signs attributable to
the bone neoplasms. All osteosarcomas but one
(175 ppm, CID#0745) were seen in the radiographs.
Tbe osteosarcoma in the male rat that received
100 ppm (CID#0495) was a 25x20x20 mm mass
surrounding the first and second coccygeal vertebrae.
The peripheral margin of the neoplasm was well
defined and the vertebral body was largely intact
within the mass. The neoplasm consisted of an
abundant osteoid matrix with interspersed single and
small nests of osteoblasts within lacunae (Piate 3).
Tbe osteoblasts were more abundant at the periphery
of the neoplasm where active growth was occurring.
They were generally polygonal, with a large
nucleus and prominent nucleolus, and were relatively
uniform in size and shape. This animal bad a
metastatic lesion in the lungs with the same morphological appearance as the primary lesion in
the coccygeal vertebra (Plate 4). Of the three male rats receiving 175 ppm sodium
fluoride, two bad osteosarcomas involving vertebrae,
and the third bad an intramedullary neoplasm in the
proximal portion of the humerus. The first of these
male rats (CID#0775) had a 26x17x9 mm mass
involving primarily the twelfth and thirteenth
thoracic vertebrae. The compact bone forming tbe
wall of the vertebral body was thin and discontinuous,
apparently destroyed by neoplastic tissue
extending from the medullary cavity to the dorsal
and lateral aspects of the vertebra. This osteosarcoma
was highly cellular and less differentiated
than that observed in the male that received
100 ppm, and the neoplastic cells were present in
solid sheets with occasional islands of osteoid
(Piate 5). In some areas, multinucleated cells were
abundant (Piate 6).
The second male rat with a vertebral osteosarcoma
(CID#0713) had a 10x10x10 mm mass that appeared
to involve the seventh cervical vertebra. The
largest portion of the neoplasm was ventral to the
vertebral body, although neoplastic tissue was
present in the spinal canal and had invaded the
spinal cord (Plate 7). The third high-dose male rat
(CID#0745) had an osteosarcoma in the humerus
that was not observed radiographically or at
necropsy. It was intramedullary and located primarily
on the metaphyseal side of the epiphysis. In
the section of humerus examined the epiphyseal
plate was intact, although large clusters of neoplastic
cells were interspersed among the trabecular bone
on tbe epiphyseal side of the growtb plate. This
neoplasm was also highly cellular with occasional
islands of osteoid interspersed among the neoplastic
cells (Plate 8).
The osteosarcomas of the bone in male rats receiving
sodium fluoride are notable because of their
rarity in untreated control groups from NTP studies.
Osteosarcomas at any site, including extraskeletal
tissues (see below), have been seen in 10/2,106
(0.5%) male untreated historical control rats. The
greatest incidence observed in any one control group
was 6%. Thus, the incidence rate of osteosarcoma
(any site) in high-dose male rats receiving sodium
fluoride is within the range of historical controls. lt
should be noted tbat the fluoride Ievels in the diets
used in the previous studies were higher than in
that used in the current studies (see the Discussion
for further information).
The quality assessment pathologist confirmed the
increased incidence of osteosclerosis in female rats
as reported by the laboratory pathologist, and the
Pathology Working Group (PWG), which reviewed
selected examples of these lesions, concurred with
these findings. The PWG also concurred with the
diagnoses of osteosarcoma in the bone of four male
rats receiving sodium fluoride. Mesenchymal Soft Tissue: An extraskeletal osteosarcoma
in the subcutis of the flank occurred in a
male rat receiving 175 ppm sodium fluoride. The
radiographs of this rat showed no evidence of a
primary neoplasm of the bone. The neoplasm was
a 45x21X21 mm mass with a large necrotic center
and a more varied morphology than that of the
other osteosarcomas (Plate 9). There was a mixture
of canilaginous and osteoid matrix and highly
cellular areas with no intercellular matrix (Plate 10).
Although this neoplasm did not arise from bone,
the PWG thought that is was appropriately diagnosed
as an osteosarcoma because of the cellular
differentiation.
A lesion found in the subcutis of a control male rat
is also notable because it consisted of collagenous
connective tissue with well-defined islands of osteoid
and/or woven bone. It was identified in the radiographs
and was subsequently examined microscopically.
The lesion was unusual and is not easily
categorized in current rodent or human classification
schemes. The medical and veterinary pathologists
who reviewed the lesion concluded that it was not
TADLE 10
Sodium Fluoride, NTP TR 393
an osteoma or osteosarcoma but a benign
mesenchymal growth with osseous metaplasia of
uncertain histogenesis and biological potential
(Piate 11).
Teeth: Alterations of the teeth associated with the
administration of sodium fluoride were observed in
the incisors and were more frequent in males than
females (Table 10). The lesions identified were
similar to those previously reponed in the literature.
In general the PWG concurred with the findings of
the Iaboratory pathologist, although there was a
minor difference of opinion regarding the most
appropriate terminology for the lesion involving the
ameloblastic epithelium. The lesion occurred
primarily in the maturation and transitional zones.
Although the laboratory pathologist bad used the
term "squamous metaplasia," the PWG thought that
"degeneration" was more appropriate. Degeneration
of the ameloblastic epithelium varied in severity and
extent from loss of the surface columnar cells to
marked reduction in cellularity (atrophy) of the
surface and papillary layers with only two to three
cell layers of flattened, squamous-like cells remain- ing (Plate 12). Since there was no evidence of
proliferation of these cells and no keratin production,
it was not considered analogaus to squamous
metaplasia as it occurs in a variety of argans in
response to prolonged injury due to viral or bacterial
infections or chemicals. Dentine dysplasia was
characterized by variable degrees of focal flattening
(atrophy) of the odontoblasts and irregularities in
thickness of the dentine. In some animals there was
disorganization of the dentine, with predentine-like
material among the layer of odontoblasts or within
the pulp and irregularities in the contour of the
dentine.
Oral Mucosa (Tongue, Pharynx, Gingiva and Tooth):
Squamous cell papilloma or carcinoma arising from
the epithelium of the oral mucosa occurred in
several dosed and control rats (Table 11). The
incidence of papilloma or carcinoma combined was
marginally increased in male and female rats
receiving 175 ppm sodium fluoride, but it was not
significantly greater than that of the control groups.
Squamous cell neoplasms of the oral mucosa are
relatively uncommon in F344/N rats, occurring in
14/2,106 (0.7%) historical male untreated controls
and 12/2,153 (0.6%) female untreated controls. The
highest incidence observed in any single control
group was 4%. The squamous cell neoplasms in
rats receiving sodium fluoride were not considered
chemical related because a squamous cell carcinoma
was observed in one control male (paired control
group) and in one control female, the incidences in
the dosed groups were not significantly greater than
in concurrent controls and were within the range of
historical controls, and there was no supporting
evidence of focal hyperplasia of the oral mucosa.
Thyroid Gland: Follicular cell adenomas were
observed in 1/49 mid-dose male and 3/80 high-dose
male rats. Follicular cell carcinomas were observed
in 1/80 control, 1/51 low-dose, and 1/80 high-dose
male rats (Appendix A, Table Al). Further, a
follicular cell carcinoma was seen in a high-dose
male at the 66-week interim evaluation. Although
there is a marginal numerical increase in follicular
cell neoplasms in male rats receiving 175 ppm
sodium fluoride, the incidence is not significantly
greater than that in controls (Table A3). Moreover,
the incidcnce of follicular cell neoplasms in the
47
high-dose group is within the range of historical
untreated controls (26/2,086, 1.2%, range 0%-6%)
(Table A4c), and the incidence of follicular cell
hyperplasia is not increased in dosed rats (Table
AS). Thus, the marginal increase in follicular cell
neoplasms was not considered related to
administration of the chemical.
Skin: Keratoacanthomas were seen in three highdose
female rats; none occurred in lower dose
groups or in controls (Appendix B, Table Bl).
However, other benign neoplasms arising from the
stratified squamous epithelium were observed in one
control female (trichoepithelioma) and one paired
control female (squamous papilloma) (Table Bl).
The incidence of squamous cell neoplasms of the
skin (keratoacanthoma, trichoepithelioma, or squamous
cell papilloma combined) in high-dose female
rats was not significantly greater than that in controls
and was not considered related to
administration of the chemical (Table B3).
Keratoacanthomas also occurred in male rats, but
they were not dose related (Table A3). The
incidence in the high-dose group was similar to that
in controls (control males, 11 %; low-dose males,
4%; mid-dose males, 2%; high-dose males, 10% ).
Uterus: Uterine stromal polyps were seen in 12/80
(15%) control, 4/50 (8%) low-dose, 6/50 (12%) middose,
and 2/81 (2%) high-dose female rats (Appendix
B, Tables Bl and B3). A stromal sarcoma
occurred in one high-dose female. The incidence of
stromal polyp or stromal sarcoma combined in the
high-dose females was significantly less than that in
controls (P=0.014) (Table B3). However, the
incidence of stromal polyps in historical untreated
control groups is quite variable and ranges from 8%
to 36% with a mean of 21% (Table B4c).
Therefore, it is uncertain whether the decreased
incidence is related to administration of the
chemical.
Other lesions were incidental or part of spontaneaus
disease complexes of rats. There was no alteration
in the incidence or severity of these lesions in the
treated and control animals, and they were histopathologically
typical of those commonly seen in this
strain of Iabaratory rat.

Dose Selection for the 2-Year Studies

The sodium fluoride drinking water concentrations selected for the first 2-year studies using the low fluoride, semisynthetic diet were 0, 10, 30, and 100 ppm for both sexes of rats and mice. Higher concentrations were not chosen based on the decreased body weight gain of rats receiving 300 ppm and of mice receiving 200 to 600 ppm in the 6-month studies, and because of the severity of gastric lesions seen in rats at the 300 ppm dose Ievel. Histopathologie examinations were performed on ll early death animals and on a ponion of the animals that survived to the end of the first 2-year studies (with the low fluoride, semisynthetic diet) to assess the cumulative toxic effects of sodium fluoride at these concentrations. The results of this evaluation revealed no significant toxic effects that could be attributed to sodium fluoride administration, suggesting that higher doses could be tolerated by both sexes of rats and mice. Therefore, concentrations selected for the second 2-year studies were 0, 25, 100, and 175 ppm sodium fluoride (equivalent to 0, 11, 45, and 79 ppm fluoride ion). Because of the nutrition problems encountered with the use of the semisynthetic diet during the first 2-year studies, the second set of studies were performed with the low fluoride NIH-07 diet. The NIH-07 diet is customarily used in NTP studies.

In-Life Observations in the 2-Year Studies

During the 2-year studies using the low fluoride NIH-07 diet, survival and weight gains of all dosed and control rats and mice were similar. The peak weights achieved by rats were similar to those typically observed in other NTP studies, but the

weights of dosed and control mice were greater than historical control average weights. No specific reason for this increased weight gain in mice was determined, but this is one of the first studies completed in which mice were housed singly rather

than in groups and factors associated with this change may be involved in tbe higher weight gains. Food and water consumption did not differ between dosed and control rats or mice. The average daily dose of sodium ßuoride consumed by the dosed animals ranged from 1.3 to 8.6 mg/kg for male rats; 1.3 to 9.5 mg/kg for female rats; 24 to 16.7 mg/kg for male mice; and 28 to 18.8 mg/kg for female mice. Actual doses achieved throughout the studies varied from these average figures because of normal changes in the pattem of water consumption as a function of · animal age and body weight (Appendix L). The highest doses were achieved early in the study when body (and skeletal) growth is at its peak. The average amount of fluoride ion, rather than sodium fluoride, consumed through the water by the dosed animals ranged from 0.6 to 8.5 mg/kg per day. The estimated total fluoride intake from feed and from dosed water for dosed and control animals is given in Table 22 Also given in Table 22 are estimates of the average fluoride intake (primarily from the feed) by historical control groups of animals that have been maintained on the Standard NIH-07 diet, which has not been closely monitored or controlled for fluoride

TABLE ll

content. By comparison, estimates of the total daily fluoride intake by human adults living in areas

served by nonfluoridated water supplies range from

0.4 to 0.9 mg (6 to 13 p.glkg per day), and from

about 1 to 5 mg (15 to 70 p.glkg per day) in areas

with fluoridated water (IPCS, 1984). Doses of

sodium fluoride given therapeutically for

osteoporosis range from 50 to 100 mg/day (0.3 to

0.6 mg fluoride/kg per day) (Farley et al., 1987).

The teeth of both rats and mice were visibly affected

by exposure to sodium fluoride during the 2-year

studies. Rats, primarily males, showed dose-related

increased incidences of whitish discoloration and

deformity leading to malocclusion. Approximately

50% of high-dose male rats showed attrition of the

lower incisors at the end of the studies, and dosed

male and female mice had increased incidences of

mottling relative to the incidence in controls. In

mice, the whitish discoloration occurred to a much

greater extent than in rats, but there was little

evidence of increased attrition, deformity, or malocclusion.

Dosed animals were otherwise similar to

controls in behavior, general heallh, and appearance.

Hematology, Oinical Chemistry, and Tissue

Fluoride Analyses in the 2-Year Studies

Hematology, clinical chemistry (calcium, phospborus,

alkaHne phospbatase), and fluoride analyses of bone,

serum (rats only), and urine (rats only) were conducted

at 24 weeks for mice or 27 weeks for rats

and at 66 weeks for both species. Bone fluoride

concentrations were also determined at the end of

the 2-year studies. There were no biologically

significant changes in hematologic measures or in

serum levels of calcium or phosphorus in rats or in

mice. Serum alkaline phosphatase activity was

elevated at both 24 and 66 weeks in female mice

given 175 ppm. The reason for this is not clear, but

it could indicate increased activity of osteoblasts

(Farley et al., 1987). However, there was no gross

or microscopic evidence of increased bone formation

in the female mice in this dose group.

The concentration of fluoride in both the serum and

the urine of rats increased with the concentration of

sodium fluoride in the drinking water. Serum

fluoride concentrations of high-dose rats were 2- to

3-fold greater than those of control animals. These

concentrations were within the range of values of

the fluoride concentrations determined for the

plasma of rats in the 6-month studies and are in

general agreement with previously published values

for rat plasma obtained by this method (Singer and

Ophaug, 1977). Plasma fluoride concentrations in

humans have been reported to be about 0.01 µg/mL

and vary with the fluoride concentration of the

drinking water (IPCS, 1984). If the serum fluoride

concentrations for rats are directly comparable to

plasma levels for humans, then the fluoride concentrations

for rats in the present studies range from 5-

to 7-fold higher in control animals to 10- to 20-fold

higher in tbe high-dose animals at the times

measured during the studies. However, the current

studies used a procedure for fluoride determination

that employed sample decomposition. Guy (1979)

has reported that methods involving sample decomposition

generally give estimates of fluoride content

that are higher tban metbods using non-decomposed

samples, and may overestimate the actual concentration

of ionic fluoride.

Fluoride concentrations of bone showed expected

dose- and age-related increases that were similar in

both sexes of rats and mice. The maximum concentration

of fluoride in bone ash of high-dose rats and

mice at the end of the 2-year studies ranged from

5.3 to 6.2 }'g/mg asb (5,300 to 5,200 ppm). This

Sodium Fluoride, NTP TR 393

represents an approximate 17-fold increase in mice

and a 29-fold increase in rats over the prestudy

bone fluoride levels measured in 6-week old animals,

and an approximate 7- to 8-fold increase in mice

and 10- to 12-fold increase in rats over the fluoride

levels accumulated in the bones of control animals

during the 2-year studies. Bone fluoride levels

similar to those determined in the high-dose animals

have been reported in human bone samples taken

from people who had lived for at least 10 years in

an area with an average fluoride content of 4 ppm

in drinking water (Zipkin et al., 1958) and from

patients who had taken 50 to 66 mg of sodium

fluoride daily for 5 to 6 years in the treatment of

osteoporosis (Boivin et al., 1988).

Neoplastic and Nonneoplastic Lesions in

the 2-Year Studies

Lesions observed in the incisor teeth of rats receiving

sodium fluoride for 2 years included dysplasia

(malformation) of the dentine layer, degeneration of

ameloblasts, and, less frequently, degeneration of

odontoblasts. The degenerative changes in the

ameloblasts were similar to, but less severe than,

those observed in the 6-month studies. The teeth of

mice apparently were less affected by sodium fluoride

than were those of rats. However, dysplasia or

malformation of the dentine layer occurs with

increasing frequency in untreated control mice as

they age. Thus, the effects of sodium fluoride may

be less discernible in mice.

There was an increased incidence and severity of

osteosclerosis observed in high-dose female rats, but

not in dosed male rats or in mice. This change

occurred during the last 9 months of the 2-year

studies because no increases were observed in

animals examined at 27 or 66 weeks. Osteosclerosis

is characterized by an increase in trabecular bone in

the metaphysis of long bones and the vertebrae and

occurs spontaneously, particularly ·in aging female

rats. The increased incidence and severity of this

lesion are consistent with the demonstrated stimulatory

effects of fluoride on osteoblasts and osteoid

production, although an effect on bone resorption

cannot be ruled out. The stimulatory effects of

fluoride on bone production form the basis for the

use of fluoride in treating osteoporosis in humans

(Farley et al., 1987).

Lesions of the femur and tibia similar to those

identified in the 6-month studies were not observed

in the 2-year studies in mice. The reason for this is

not known, but may be related to the levels of

sodium fluoride administered and changes in the

rates of bone remodeling as the animals age.

Exostoses or other bone changes typically associated

with severe skeletal fluorosis were not found in rats

or mice.

Osteosarcomas of the bone were observed in 3/80

(4%) high-dose and in 1/50 (2%) mid-dose male

rats. An additional osteosarcoma, which was determined

to be of subcutaneous origin, was observed in

a found high-dose male rat. No osteosarcomas

were seen in controls or in male rats receiving

25 ppm. The neoplasms were clearly malignant

( one metastasized to the lung) and there was

complete agreement concerning the diagnoses at

both the Quality Assessment and the Pathology

Working Group stages of histopathology review.

Table 23 summarizes the location, method of first

observation, and the time during the sodium fluoride

studies at which the osteosarcomas were observed.

Many scientific and technical factors enter into the

determination of whether there is an association

between sodium fluoride exposure and the occurrence

of osteosarcomas in male rats. A number of

factors support such an association and others

suggest that there may be no association. To fully

consider these, it is necessary to review certain

TABLE 23

71

aspects of the design of the studies and the nature

of the NTP historical database for osteosarcomas.

The procedures for the examination of bones and

teeth used in these studies differed from those used

in the studies that compose the NTP historical

database. When animals in the dosed or control

groups died or were killed in the sodium fluoride

studies, whole body radiographs were taken and a

complete gross necropsy was conducted. Sections of

bone from the tibia, femur, humerus, thoracic

vertebra, maxilla, incisive bones, nasal bones, and

the mandible were routinely examined microscopically

in addition to any bone lesions observed on

gross or radiographic examination. In a typical NTP

study, sections of bone from the maxilla, rib, or the

femur are routinely taken in addition to any gross

bone lesions. Radiographs are not routinely taken.

Osteosarcomas (in bone or extraskeletal) are not

commonly observed in control male rats in NTP

studies. The historical incidence in control male

rats from dosed feed or water studies is 10/2,106

(0.47%) and in male controls from studies by all

routes of administration, including gavage and

inhalation, the rate is 37/6,131 (0.60%) (includes

5 tumors diagnosed as osteoma ). Of these

37 tumors occurring in control male rats, one was

found in an animal killed at approximately

40 weeks, 25 occurred in animals dying between

weeks 70 and the end of the 2-year experiments, and

the other 11 were observed in animals killed at

2 years. About 20% of the osteosarcomas occurred

in the venebra, 20% in the skull, and about 10%

were found in the rib. The remainder were

observed at various sites in the long bones, pelvis,

subcutaneous tissue, and lung. All but one of these

tumors were observed upon gross examination of the

animal, and of these, two were found in the

subcutis. The other tumor was observed

microscopically in the lung. Thus, the majority of

osteosarcomas and osteomas found in control

animals occurred in bone and were observed grossly

during the latter pan of the 2-year studies.

The number of studies with 0, 1, 2, or 3 osteosarcomas

in control male rats in the 122 studies that

compose the historical control database (6,131 male

rats) fits a Poisson distribution centered on a

historical incidence of 0.5%. As many as 3 osteosarcomas

in a group of 50 control male rats (6%)

was observed on one occasion, in agreement with

the frequency expected based upon the Poisson

distribution.

An imponant consideration that limits the usefulness

of the historical control database in the interpretation

of the current studies is that the diet used

in all other NTP studies has not been closely

controlled or monitored for fluoride content.

Fluoride concentrations in typical batches of the

NIH-07 diet range between 28 and 47 ppm (Rao

and Knapka, 1987). Assuming a maximum bioavailability

of 60%, the historical database animals

actually constitute a group receiving sufficient

tluoride to place them between the low- and

mid-concentration groups in the current 2-year

studies (fable 22). The fact that this fluoride is

available for absorption from the standard diet is

supported by the levels of fluoride found in the

bones of animals maintained on this diet in the

6-month studies (Appendix I). If fluoride is in fact

influencing the "spontaneous" or background incidence

of osteosarcomas in male rats, comparisons of

the incidences observed in the current studies with

those in the historical database may be misleading.

This forces an even greater reliance on the withinstudy

comparisons, i.e. the incidences in the dosed

groups compared with the concurrent control, in the

interpretation of the results of the sodium tluoride

studies.

Sodium Fluoride, NTP TR 393

The Cour osteosarcomas of bone ( one in the middose

and three in the high-dose groups) in the

current studies occurred with a statistically significant

dose-response trend by the logistic regression

test (P=0.027); the pairwise comparison of the

incidence in the high-dose group versus that in

controls was not statistically significant (P=0.099).

The statistical significance of the trend test is

increased (P=0.010) when the subcutaneous osteosarcoma

in the found high-dose rat is included

in the incidence, but the pairwise comparison

remains not significant (P=0.057). The incidence of

bone osteosarcomas of 3/80 and the incidence of all

osteosarcomas of 4/80 in the high-dose male rats are

both significantly greater than the rate of 0.6% for

osteosarcomas and osteomas at all sites in control

male rats in the historical database. Note that one

of the tumors in the current studies was observed

microscopically and was not visible on the radiograph.

lt is likely that the actual occurrence of

microscopic osteosarcomas is underrepresented in

the historical database and possibly in the current

studies because few sections of bone are taken for

routine microscopic analysis.

The analyses of osteosarcomas are considered both

with and without the subcutaneous tumor because it

may not be appropriate biologically to combine

these for statistical comparison with the controls.

Chemical carcinogens typically produce site-specific

increases in tumor incidences (Haseman et al.,

1986); thus tumors are usually combined for analysis

on the basis of the tissue of origin and not simply

because they may have the same histologic diagnosis.

Osteosarcomas of bone contain neoplastic osteoblasts

which presumably are responsible for the

abnormal deposition of osteoid and collagen in

osteosarcomas (Spjut et al., 1971). Osteosarcomas

that originale in bone may metastasize to soft

tissues. On the other band, sarcomas of soft tissues

may occasionally produce osteoid and develop into

an osteosarcoma (Caner, 1973). Careful examination

of the radiograph of the male rat with the

subcutaneous osteosarcoma did not reveal a potential

site of origin within bone for this tumor; thus it

is unlikely that the subcutaneous neoplasm represents

a metastatic bone tumor.

The distinction concerning the site of origin is also

important because, if fluoride were to exert a

neoplastic effect, it is reasonable to expect that this

might be expressed in a tissue that accumulates

fluoride. This would include bone, and, therefore,

there is biological plausibility for an association

between sodium fluoride administration and the

development of bone osteosarcomas. However,

fluoride does not accumulate in soft tissues such as

the subcutis (Smitb et al., 1960), making it perbaps

less likely tbat tbis tumor developed as a oonsequence

of sodium fluoride administration.

Fluoride was found to accumulate in the bone of

female rats and male and female mice to a similar

extent as in male rats. There were no osteosaroomas

observed in female rats, yet high-dose

female rats had tbe clearest evidence of fluorideinduced

osteosclerosis, suggesting that a stimulatory

or mitogenic effect of fluoride on osteoblasts (Farley

et al., 1983; Marie and Hott, 1986) was occurring in

female rat~. Male and female mice had no microsoopic

evidence of osteosclerosis of bone. A total of

three osteosarcomas and one OSteoma were found in

male and female mice in the present studies. An

osteosarcoma occurred in one low-dose male mouse

killed for evaluation at 66 weeks, in one low-dose

female mouse, and one osteosarcoma and one

osteoma occurred among the female mice in the

control group. None of the female rats or male or

female mice in tbe mid- or high-dose groups had an

osteosarcoma.

The Iack of supporting evidence in female rats and

male and female mice for the apparent association

between sodium fluoride administration and osteosarcoma

production in male rats may, however, have

only limited significance. While only one chemical

previously studied by the NTP has been associated

with osteosarcoma formation (Acronycine in

Sprague-Dawley rats, NCI, 1978), the tumor

response in this earlier study was clearly shown in

male and not in female rats. The study in mice was

judgcd inadequate for evaluation. On the other

hand, Litvinov and Soloviev (1973), in a review of

tumors of the bone in rats, stated that the sex of

the animal does not play a role in the tumor

response to chemical inducing agents. Their review

of the literature indicates that most experimentally

induced bone tumors in rats have been found in the

long bones following administration of radioactive

isotopes of bone-seeking elements such as phosphorus-

32, calcium-45, or strontium-90, skeletal

irradiation, or intraosseous administration of chemical

carcinogens. A review of more recent literature

has not added significant information concerning the

potential for a sex-linked response of bone tumor

formation in animals. However, osteosarcomas in

humans occur more frequently in males than in

females (NCI, 1989).

No studies were found in the literature which have

directly assessed the genotoxic potential of sodium

fluoride to osteoblasts. However, sodium fluoride

was found positive in assays for gene mutation

induction in mammalian cells in vitro and in

Drosophila. It is positive in some plant and animal

systems for the induction of chromosomal aberrations,

and it is positive in in vitro assays for morphologic

transformation of Syrian hamster ovary

cells. While the mechanisms for these effects are

not understood, the data suggest that sodium fluoride

has the capability, probably through an indirect

mechanism, for genotoxic activity.

To summarize these considerations, a small number

of osteosarcomas occurred in mid- and high-dose

male rats. These neoplasms occurred with a significant

dose response trend, but at a rate within the

upper range of incidences previously seen in control

male rats in NTP studies. Three of the tumors

arose in the venebra, a site not commonly associated

with chemically induced osteosaroomas. Bone

is known to accumulate fluoride, and fluoride has

been shown to be genotoxic to some mammalian

cells in culture. No osteosarcomas were seen in

female rats, and several osteosarcomas seen in mice

occurred with an incidence that did not soggest a

relationship with sodium Ouoride exposure. Taken

together, the current findings are inconclusive, but

are weakly supportive of an association between

sodium fluoride administration and the occurrence

of osteosarcomas in male rats.

A second potential target site for sodium fluoride

when given in drinking water is the upper digestive

tract and oral cavity. Squamous cell neoplasms of

the oral mucosa (tongue, palate, or gingiva)

occurred with marginally increased incidcnces in

dosed male and female rats over the rates in controls.

The increased incidences of these neoplasms

were not statistically significant when compared with

the incidences in concurrent controls; however, the

incidences in the high-dose groups were significantly

higher than the incidences observed in historical

control animals (0.7% male rats; 0.6% female rats).

As with lesions of the bone, a direct comparison

with the historical rates for oral cavity neoplasms is

not completely accurate because of the increased

attention given to the oral cavity and teeth in the

sodium ßuoride studies compared to previous NTP

studies. Rates for oral cavity neoplasms similar to

tbose observed in bigh-dose male and female rats in

the sodium fluoride studies (4%) have been

observed twice for males and once for females in

the historical control database of 42 dosed feed or

water studies. Neoplasms of the oral cavity were

observed in control male and female rats in the

current studies; one was observed in an age-matched

control male rat and one occurred in a control

female rat in the main study.

An argument could be made for combining the male

and female rat studies for analysis of oral cavity

neoplasms because a marginal increase occurred in

both groups. An analysis for significance of the

combined P values for the logistic regression trend

tests for male and female rats resulted in a nonsignificant

P value of 0.065.

In contrast to osteosarcomas, for which there are no

recognized benign or preneoplastic counterparts

(Utvinov and Soloviev, 1973), squamous cell byperplasias

of the oral cavity are considered preneoplastic

precursor lesions of squamous cell neoplasms

of the oral cavity (Brown and Hardisty, 1990).

Squamous cell hyperplasia occurred in no more than

one animal in any of the dosed or control groups in

the current studies. Thus, based on the absence of

statistical significance versus the concurrent controls,

the occurrence of these tumors in control animals,

and the lack of a dose-related increase in nonneoplastic

precursor lesions, it is concluded that

there is insufficient evidence to relate tumors of the

oral cavity with administration of sodium fluoride to

male or female rats. Glattre and Wiese (1979)

reported an association between a decrease in

human mortality due to oral cavity neoplasia and

increasing ßuoride content in water over the range

of 0 to 0.5 ppm.

Follicular cell neoplasms of the thyroid gland

appeared with a marginally increased incidence in

high-dose male rats compared with controls. This

increase is not statistically significant compared with

controls unless control animals from both interim

groups (27 and 66 weeks) and the age-matched

controls are pooled with tbe main study control

group. If this is done, the logistic regression P

value for the trend is 0.027. Thyroid follicular cell

neoplasms typically occur with an incidence of 1.2%

in historical control animals. Incidences of 6% have

previously been observed in untreated control

groups and incidences as high as 10% have occurred

in control groups for gavage studies. The incidence

of these neoplasms in the high-dose groups was 5/90

(5.5%; includes 10 animals from the 66-week interim

sacrifice, one of which bad a thyroid follicular cell

carcinoma ). Three of these tumors were adenomas.

The incidence of carcinomas did not differ across

the dosed groups and the incidence of follicular cell

hyperplasia was not increased. No increase in the

incidence of these tumors occurred in female rats.

Based on these considerations, follicular cell neoplasms

of the thyroid are not considered related to

sodium fluoride administration.

In mice, the only neoplasm that appeared to be

possibly related to sodium fluoride administration

was lymphoma in females. However, lymphoma is

a common neoplasm in mice, occurring with rates

varying from 10% to 74% in historical controls. In

the current studies, the incidences in control and in

low-dose female mice (14% and 10%) were less

than the lowest incidence observed in the 9 studies

composing the historical database at the study

laboratory. The incidence of 24% in high-dose

female mice (or 30% when considering the combination

of all lymphomas and histiocytic sarcomas) is

similar to the average historical control incidence of

31% in female mice. There was no increase in the

incidence of lymphomas in male mice. For these

reasons, it is considered unlikely that sodium fluoride

administration affected tbe incidence of this

neoplasm in female mice.

One other finding in tbe sodium fluoride studies

deserves mention. The incidence of liver neoplasms

in all groups of dosed and control male and female

mice was higher than has typically been seen in NTP

studies (Appendixes C4 and D4b). A review of

pathology information from NTP studies which

began about the same time as the sodium fluoride

studies, but which bave not yet been completely

evaluated and reported, has revealed a sharp

increase in liver neoplasms, especially in females.

The reasons for this trend toward increasing liver

neoplasms in control mice have not been determined,

but it is worth noting that the weights of all

groups of male and female mice in the sodium

weights attained by previous control groups. A

possible relationship between body weight and the

incidence of liver neoplasms in B6C3F1 mice has

been discussed by Rao et al. (1987).

In these 2-year studies, toxic effects of the concentrations

of sodium fluoride employed were noted in

the teeth and bones of rats and several osteosarcomas

occurred in dosed males. Higher sodium

Ouoride concentrations in drinking water may have

been tolerated by the rats, but it is difficult to

predict the concentration above which effects on

dentition would become so severe as to interfere

with the animals' ability to eat. The effects of these

sodium fluoride concentrations in mice were limited

to discoloration of teeth and a marginal increase in

dentine dysplasia in males. Based on these findings,

it would appear that mice could have tolerated

somewhat higher sodium fluoride concentrations in

the drinking water.

Conclusions:
Under the conditions of these 2-year dosed water
studies, there was equivocal evidence of carcinogenic
activity* of sodium fluoride in male F344/N rats,
based on the occurrence of a small number of
osteosara>mas in dosed animals. •Equivocal
evidence• is a category for uncertain findings defined
as studies that are interpreted as showing a marginal
increase of neoplasms that may be related to chemical
administration. There was no evidence of carcinogenic
activity in female F344/N rats receiving
sodium fluoride at concentrations of 25, 100, or 175
ppm (11, 45, or 79 ppm ßuoride) in drinking water
for 2 years. There was no evidence of carcinogenic
activity of sodium fluoride in male or female mice
receiving sodium fluoride at concentrations of 25,
100, or 175 ppm in drinking water for 2 years.
Dosed rats bad lesions typical of fluorosis of tbe
teeth and female rats receiving drinking water
containing 175 ppm sodium fluoride bad increased
osteosclerosis of long bones.
Executive summary:

The sodium tluoride concentrations selected for the

2-year studies in both rats and mice were 0, 25, 100,

and 175 ppm in the drinking water. These concentrations

were selected based on the decreased weight

gain of rats at 300 ppm and of mice at 200 ppm

and above, on the incidence of gastric lesions in rats

at 300 ppm in the 6-month studies, and on the

absence of significant toxic effects at sodium tluoride

concentrations as high as 100 ppm in an earlier

2-year study.

Body Weighls and Sunival in the 2-Year Studies:

Mean body weights of dosed and control groups of

rats and mice were similar tbroughout the 2-year

studies. Survival of rats and mice was not affected

by sodium tluoride administration. Survival rates

after 2 years were: male rats--control, 42/80;

25 ppm, 25/51; 100 ppm, 23/50; 175 ppm, 42/80;

female rats-59/80; 31/50; 34/50; 54/81; male

mice-ssn9; 39/50; 37/51; 65/80; female mice-53/80;

38/52; 34/50; 52/80.

Neoplastic and Nonneoplastic Effects in the 2-Year

Studies: The teeth of rats and mice bad a dose

dependent whitish discoloration, and male rats bad

an increased incidence of tooth deformities and

attrition leading on occasion to malocclusion. The

teeth of male and, to a lesser degree, female rats

bad areas of microscopic dentine dysplasia and

degeneration of ameloblasts. Dentine dysplasia

occurred in both dosed and control groups of male

and female mice; the incidence of this lesion was

significantly greater in high-dose than in control

male mice. Osteosclerosis of long bones was

increased in female rats given drinking water containing

175 ppm sodium fluoride. No other significant

nonneoplastic lesions in rats or mice appeared

related to sodium fluoride administration.

Osteosarcomas of bone were observed in 1/50 male

rats in the 100 ppm group and in 3/80 male rats in

the 175 ppm group. None were seen in the control

or 25 ppm dose groups. One other 175 ppm male

rat bad an extraskeletal osteosarcoma arising in the

subcutaneous tissue. Osteosarcomas occur in

historical control male rats at an incidence of 0.5%

(range 0-6%). The historical incidence is not

directly comparable with the incidences observed in

this study because examination of bone was more

comprehensive in the sodium fluoride studies than

in previous NTP studies of other chemicals, and the

diet used in previous studies was not controlled for

fluoride contenL In the current study, although the

pairwise comparison of the incidence in the

175 ppm group versus that in the controls was not

statistically significant, osteosarcomas occurred with

a statistically significant dose-response trend, leading

to the conclusion that a weak association may exist

between the occurrence of these neoplasms and the

administration of sodium fluoride. No other neoplastic

lesions in rats or mice were considered

possibly related to chemical administration.

Endpoint:
carcinogenicity: oral
Type of information:
other: evidence from degradation product
Adequacy of study:
supporting study
Reliability:
2 (reliable with restrictions)
Rationale for reliability incl. deficiencies:
comparable to guideline study with acceptable restrictions
Qualifier:
no guideline followed
Principles of method if other than guideline:
Six-month studies were conducted to evaluate the
cumulative toxic effects of continuous exposure to
sodium fluoride and to determine appropriate doses
for the 2-year studies. The male and female rats
and mice for these studies were bred at the study
labaratory. Breeder F344 rats (Harlan Industries Indianapolis, IN) and C57BL/N6 female and
C3H/HeN male mice (Cbarles River Laboratories,
Wilmington, DE) were placed on a low fluoride diet
(S2.1 ppm fluoride) 1 month before monogamaus
pairing (Appendix K). Progeny that survived to
weaning were distributed to weigbt classes and
assigned to cages by a random number table; cages
were then assigned to dose and control groups by a
second table of random numbers. Rats were S to
6 weeks old when placed on study, and mice were
4 to 6 weeks old. The weigbt range of animals used
in the 6-month studies was somewhat larger than is
typical in NTP studies because insufficient numbers
of animals were derived from the breeding program
to allow the usual culling of animals of high or low
weights.
Groups of ten rats of each sex were administered 0,
10, 30, 100, or 300 ppm sodium fluoride in deionized
drinking water ad libitum for 6 months.
Groups of 8 to 12 mice of each sex received 0, 10,
SO, 100, 200, 300, or 600 ppm sodium fluoride in
deionized drinking water ad libitum for 6 months.
The study design called for ten mice of each sex per
dose group; however, on the second day of dosing,
one male and five female mice were found to have
been missexed, and they were placed with the group
of the correct sex at the same dose Ievel. All test
animals receiving water supplemented with sodium
fluoride were provided with a low fluoride
(S2.1 ppm), semisynthetic diet throughout the study.
Three control groups were included in the studies of
male rats and male and female mice: one received
deionized drinking water and a Iow fluoride, semisynthetic
diet; the second received sodium chloridesupplemented
deionized drinking water and a low
fluoride, semisynthetic diet; and the third received
deionized drinking water and a standard NIH-07
diet. The first two of these control groups were
included in the female rat study.
Animals were housed five per cage, with feed and
water available ad libitum. They were observed
twice daily for morbidity and mortality, and signs of
toxicity were recorded. Individual weights were
recorded weekly throughout the studies. Water
consumption was recorded daily by cage; diet consumption
was recorded every other week for the
first 13 weeks and for a 1-week period during each
of the last 3 months. Table 1 summarizes further
experimental details.At termination of the studies, the fluoride concentrations
in urine, blood, and bone were determined
from samples collected from five male and five
female rats and from all surviving mice from all
groups (except the control group given sodium
chloride-supplemented deionized drinking water and
a low fluoride, semisynthetic diet); samples from the
same animals were included in all three evaluations.
Methodological details are presented in Appendix I.
Complete necropsies with tissue collection were
performed on all animals. Microscopic examination
was conducted on tissues from all control animals,
all animals dying spontaneously, and all animals in
the two higbest dose groups.
GLP compliance:
not specified
Specific details on test material used for the study:
Sodium fluoride was obtained from Apache Cbemical,
Inc. (Seward, IL) in two lots. One Iot was used
for the 14-day and 6-montb studies (Iot no. A-
06255) and anotber for tbe 2-year studies (Iot no.
A022085). Identity, purity, and stability analyses
were conducted on tbese lots by Midwest Research
Institute (Kansas City, MO). Details of tbese
analyses are presented in Appendix J. Tbe study
cbemical, a white crystalline powder, was identified
as sodium fluoride and was found to be at least
99% pure, as determined by eiemental analysis, Karl
Fiseber water analysis, spark source mass spectrometry,
and titration of acidic components. During
tbe 2-year studies, the stability of the bulk cbemical
was monitored by Galbraitb Labaratories (Knoxville,
TN). No degradation of the study material was
detected througbout tbe studies.
Species:
other: rat and mice
Strain:
other: F344 rats and C57BL/N6 female mice and C3H/HeN male mice
Details on species / strain selection:
Batteile Columbua Laboratorica
(Columbua, OH)
Rats: Progeny of Fß44 breeden from
Harlan Induatries (Indianapolia, IN)
Mice: Progeny of C57BUN6 female aod
C3HJHEN male breeden from Cwtel
River Laboratorlei (Wdmingtoa, OE
Sex:
male/female
Details on test animals and environmental conditions:
Temperature: 22 -24 °C
Humidity: 40%~%
Fluorescent light: 12 hours/day
Room air changes: 15/hour
Route of administration:
oral: drinking water
Vehicle:
water
Details on exposure:
Rats: 0, 10, 30, 100, or 300 ppm sodium
fluoride in deionized water, available ad
libitum
Mice: 0, 10, SO, 100, 200, 300, or 600
ppm sodium ßuoride in deionized water,
available ad libitum
3 control groups: (1) male and female
rats and mice: deionized water and low
fluoride, semisynthetic diet; (2) male and
female rats and mice: sodium chloride supplemented
water and low fluoride,
semisynthetic diet; (3) male rats and male
and female mice: deionized water and
standard NIH-07 diet.
Analytical verification of doses or concentrations:
not specified
Details on analytical verification of doses or concentrations:
Througbout all studies, dose formulations were
prepared by mixing appropriate amounts of sodium
fluoride with deionized water. Stability studies
indicated tbat sodium fluoride at a concentration of
25 ppm in deionized water was stable under Simulated
animal dosing conditions for up to 3 weeks
when stored in the dark at room temperature
(25° C). Details of the preparation and storage of
dose formulations of sodium fluoride are presented
in Appendix J.

Rats: 0, 10, 30, 100, or 300 ppm sodium
fluoride in deionized water, available ad
libitum
Mice: 0, 10, SO, 100, 200, 300, or 600
ppm sodium fluoride in deionized water,
available ad libitum
3 control groups: (1) male and female
rats and mice: deionized water and low
fluoride, semisynthetic diet; (2) male and
female rats and mice: sodium chloride supplemented
water and low ßuoride,
semisynthetic diet; (3) male rats and male
and female mice: deionized water and
standard NIH-07 diet.
Duration of treatment / exposure:
26 weeks (7 days/week)
Frequency of treatment:
Rats: 0, 10, 30, 100, or 300 ppm sodium
fluoride in deionized water, available ad
libitum
Mice: 0, 10, 5O, 100, 200, 300, or 600
ppm sodium fluoride in deionized water,
available ad libitwn
3 control groups: (1) male and female
rats and mice: deionized water and low
fluoride, semisynthetic diet; (2) male and
female rats and mice: sodium chloride supplemented
water and low fluoride,
semisynthetic diet; (3) male rats and male
and female mice: deionized water and
standard NIH-07 diet.
Dose / conc.:
10 ppm
Remarks:
rat and mice
Dose / conc.:
30 ppm
Remarks:
rats
Dose / conc.:
50 ppm
Remarks:
mice
Dose / conc.:
100 ppm
Remarks:
rats and mice
Dose / conc.:
200 ppm
Remarks:
mice
Dose / conc.:
300 ppm
Remarks:
rats and mice
Dose / conc.:
600 ppm
Remarks:
mice
No. of animals per sex per dose:
Rats: 10 males and 10 females
Mice: 8-12/sex/group
Control animals:
yes, concurrent no treatment
Positive control:
-
Observations and examinations performed and frequency:
Observed twice daily for mortality and
morbidity; weighed initially, once weekly,
and at termination; clinical Observations
recorded daily. Feed consumption
recorded every other week for first 13
weeks and for 1 week during each of last
3 months. Water consumption recorded
daily.
Other examinations:
Necropsy and tissue collection performed
on all animals. in addition to tissue
masses, grosa lesions, and associated
regional lymph nodes. the following
organs and/or tissues were examined
histologically for all males and females in
the control groups and in the two highend
dose groups of both species (100 ppm,
300 ppm in rats; 300 ppm, 600 ppm in
mice): adrenals, bone (femur, tibia), brain
(frontal corta and basal ganglia, parietal
cortex and thalamus, cerebellum and
pons), esophagus, eyes (when grossly
abnormal), gallbladder (mice only), heart,
kidney, large intcstines (colon, liver, Jung
with bronchi, lymph nodcs (mandibular),
mammary glands, ovarics, pancreas,
parathyroid, pituitary, prostate, salivary
gland, small intcstines, spinal cord (when
neurologic signs were present), spieen,
stomach, teeth (incisors), tcstes, thymus,
thyroid, tracbea, urinary bladder, and
Uterus
Clinical signs:
effects observed, treatment-related
Description (incidence and severity):
MICE:
Among the 13 high-dose animals that died before
scheduled sacrifice, six were killed because they were
moribund. Signs of toxicity (thin appearance,
hunched posture, weakness) were observed in only
two of these before they became moribund. Mice
exposed to the four highest doses of sodium fluoride
had chalky white teeth; the lower incisors were more
affected than upper incisors, and some teeth in mice
in the two highest dose groups were chipped. No
other signs of toxicity were observed in any of the
animals that died early or that survived to the end
of the studies.



RATS:
Signs of dental fluorosis were observed in all highdose
animals. From week 6 to the end of the
studies, their teeth appeared chalky white and bad
an unusual wear pattem. During weeks 6 to 17, the
upper incisors grew quite long, while the occlusal
surface of the lower incisors was wom to the gum
line. So that feed consumption would not be
affected, the upper incisors were trimmed periodically,
allowing the lower incisors to grow to a
normal length. Unusual chipping of incisors was
observed from week 17 through the end of the
studies. In addition, all high-dose animals bad
rough hair coats during the last 9 weeks of the
studies.
Dermal irritation (if dermal study):
not specified
Mortality:
not specified
Description (incidence):
MICE;
All but one early death occurred in the
high-dose groups: four high-dose males died during
weeks 13 and 14; one male mouse in the second
highest dose group died during week 19; nine highdose
females died during weeks 8 to 18. All other
mice survived to scheduled termination.


RAT:
There were no deaths throughout these studies
Body weight and weight changes:
not specified
Description (incidence and severity):
MICE:
Body weight gain was depressed in the three highest dose
groups for both sexes.


RAT:
Body weight gain was depressed only in the highest
dose groups
Food consumption and compound intake (if feeding study):
not specified
Description (incidence and severity):
MICE:
Average weekly feed consumption was within 20%
of control values for all groups, except high-dose
males which consumed only 77% of that consumed
by controls. Average weekly water consumption was
within approximately 20% of control values for all
dosed groups.


RAT:
Average weekly feed consumption was approximately
13% less in high-dose males and 18% less in highdose
females compared to controls. Somewhat reduced water consumption averages were also
recorded (8% Iess than controls in high-dose males
and 19% Iess than controls in high-dose females).
Food efficiency:
not specified
Water consumption and compound intake (if drinking water study):
not specified
Description (incidence and severity):
Average weekly feed consumption was approximately
13% less in high-dose males and 18% less in highdose
females compared to controls. Somewhat reduced water consumption averages were also
recorded (8% Iess than controls in high-dose males
and 19% Iess than controls in high-dose females).
Ophthalmological findings:
not specified
Haematological findings:
not specified
Clinical biochemistry findings:
not specified
Urinalysis findings:
not specified
Description (incidence and severity):
MICE:
The fluoride content of bone and urine was
increased in a dose-related fashion with increasing
fluoride concentrations in the drinking water. The
fluoride concentration in plasma appeared to
increase with the dose of fluoride, but the necessity
of pooling samples to obtain sufficient material for
analysis prevented performance of meaningful
statistical analyses of these data. The fluoride
content of urine and bone for control mice fed the
standard NIH-07 diet was greater than the values
obtained for mice given the semisynthetic diet and
drinking water containing 10 ppm sodium fluoride.


RAT:
The fluoride content of bone and urine increased
with increasing fluoride concentration in the drinking
water. The fluoride content of plasma was
significantly increased over that in control rats
maintained on the low fluoride, semisynthetic diet
only in the high-dose groups (300 ppm) andin the
group of male rats maintained on the standard NIH-
07 dict (Appendix I).
Behaviour (functional findings):
not specified
Immunological findings:
not specified
Organ weight findings including organ / body weight ratios:
not specified
Gross pathological findings:
not specified
Description (incidence and severity):
On gross examination, the mucosa of the glandular
stomach of most male rats receiving 300 ppm
sodium fluoride appeared thickened, and focal or
multifocal punctate hemorrhages were observed in
4/10 males and 1/10 females. Similar but less severe
alterations were observed in some rats receiving
100 ppm sodium fluoride. A perforated ulcer of the
glandular stomach was seen in a 300 ppm female,
and multiple, small, nonperforated ulcers were seen
in one 300 ppm male. Histologically, a subtle focal
to diffuse hyperplasia of the mucosal epithelium of
the glandular stomach was observed in 10/10 male
and 9/10 female rats receiving 300 ppm (Table 4).
It was accompanied by minimal individual cell
necrosis (apoptosis) and was most evident in the
pyloric region. In affected rats, the nurober of
mucous cells in the epithelium was slightly
decrcased relative to that in controls, the columnar
cells stained more basophilic, and the nurober of
mitotic figures at the base of the gastric pits was
increased relative to that in controls. The epithelium
lining the gastric pits contained one or several
cells with pyknotic nuclei, fragments of nuclear
debris, or residual bodies. Nearly all rats receiving
300 ppm sodium fluoride had focal basal cell hyperplasia
of the stratified squamous epithelium adjacent
to the limiting ridge (junction of the glandular
starnach and forestomach). Hyperplasia of the
mucosal epithelium of the glandular starnach also
was observed in half the males and in two females
receiving 100 ppm sodium fluoride, but individual
cell necrosis was not.
Neuropathological findings:
not specified
Histopathological findings: non-neoplastic:
not specified
Histopathological findings: neoplastic:
not specified
Other effects:
effects observed, treatment-related
Description (incidence and severity):
MICE:
A number of histological alterations were identified
in the kidney, liver, testes, and/or myocardium of
mice dying early or sacrificed while moribund
(Table 14). The acute nephrosis in three male and
two female mice was characterized by extensive
multifocal degeneration and necrosis of the tubular
epithelium. The proximal convoluted tubules in the
cortex and straight portians of the nephron in the outer mcdulla were affected. Nephrosis was likely
the principal cause of death in these mice. Multifocal
myocardial degeneration was seen in two highdose
fcmale mice, and scattered focal accumulations
of mineral were seen in the myocardium of several
others. The changes in the liver consisted of widely
scattered, individual, enlarged cells with multiple
nuclei (megalocytosis and syncytial alteration).
Generally, only a few affected cells were observed in
the field of view at 200x magnification. Degeneration
and/or necrosis of the germinal epithelium in
the seminiferous tubules, noted in dosed males,
often occurs in debilitated mice or those dying from
other toxic lesions and was not considered a direct
compound-related effect.
Compound-related effects were observed in the
femur and, to a lesser extent, in the tibia of nearly
all male and female mice receiving 100 to 600 ppm
sodium fluoride and 5/10 males receiving 50 ppm
(Table 15). In control mice the circumferential
lamcllae of the cortical bone were relatively uniform
in thickness and the cement lines were regularly
spaced. In mice receiving 600 ppm some lamellae
appeared thicker and more irregular with cement
lines that were less prominent and smooth in
contour. The osteoid seams lining some osteons
(havcrsian canals) of the cortical bone were
increased in thick.ness (Plates 13 and 14). These
changcs were not uniform or diffuse. In mice
receiving 50 or 100 ppm only occasional prominent
osteoid seams were evident. The spectrum of
changes are indicative of altered rates of bone
dcposition and remodeling.
Paraffin-embedded sagittal sections of the upper
incisors and incisive bone were unsatisfactory for
critical examination of the teeth. Therefore, the
lower incisors were embedded in glycol methacrylate,
sectioned, and stained with hematoxylin and eosin.
Thc rcsults of this evaluation are shown in Table 15.
The lesions were generally more extensive in the
mice receiving 300 or 600 ppm than in mice receiving
lowcr doses. The enamel organ from the
affectcd mice that were examined bad focal or
multifocal irregularity of the layer of ameloblasts,
with projections and folds that sometimes surrounded
isolated islands of enamel. In some mice there was loss of the surface columnar cells and
variable loss of cells from the stratum intermedium.
The remaining cells were reduced in size and
disorganized. These changes collectively were
diagnosed as dysplasia by the study pathologist


RAT:
The principal pathological effects associated with the
administration of sodium fluoride for 6 months were
observed in the incisor teeth and stomach. The
incisor teeth were chosen for examination because
they are continuously growing and, therefore, contain
all specialized components of the dental epithelium.
Paraffin-embedded sagittal sections of upper
incisors were unsatisfactory for critical examination.
Therefore, the incisors were embedded in glycol
methacrylate, sectioned sagittally, and stained with
hematoxylin and eosin. Five male rats receiving
300 ppm sodium fluoride had focal or multifocal
degeneration of the enamel organ, primarily in the
maturation zone near the apical end of the incisor
tooth (Table 4). The columnar ameloblasts were
tlattened or lost (atrophy), and the cells of the
stratum intermedium were disorganized and contained
less cytoplasm and fewer secretory vacuoles
(Plates 1 and 2). In a few animals, small aggregates of enamel-like material were trapped within the
cell layers. These changes collectively were diagnosed
as dysplasia by the Iabaratory pathologist.

In the 6-month studies in mice, 4/9 males and

9/11 females receiving 600 ppm sodium flluoride and

1/8 males receiving 300 ppm died. No rats given

water containing as much as 300 ppm sodium

fluoride died. Male and female rats given 300 ppm

and mice given 200 to 600 ppm gained notably less

weight than did controls. Weight gains of control

rats and mice maintained on the low fluoride,

semisynthetic diet were greater than those of control

animals fed the NIH-07 diet, suggesting that the diet

was sufficient to support growth at normal rates.

The fluoride content of urine and bane increased

with the concentration of sodium fluoride in the

drinking water in bath sexes of rats and mice. Bone

fluoride concentrations were as high as 14.8

(14,800 ppm) of ashed bone in male mice receiving

600 ppm sodium fluoride in the water. The bone

fluoride content found in mice was somewhat

greater than that found in rats given comparable

sodium fluoride concentrations in the water. This

may be partly due to a greater water intake on a

body weight basis by mice than by rats, resulting in

higher exposures. Plasma fluoride concentrations

in dosed rats appeared clearly elevated over that of

controls only in the groups receiving water containing

300 ppm sodium fluoride. The plasma fluoride

Ievels of mice showed a better dose relationship and

appeared increased in groups receiving water containing

50 ppm of sodium fluoride or higher concentrations.

Clinical signs attributed to sodium fluoride administration

were limited to changes in the appearance of

the teeth in rats given water containing 300 ppm

and in mice given water containing 100 to 600 ppm;

the teeth bad a chalky white appearance instead of

the dull yellow color seen in normal rats and mice.

The teeth of affected rats also showed unusual wear

patterns. lncreased wear of molars of rats of the

Sabra strain following ingestion of tluoridated

drinking water (25 ppm) for 40 days was reported by

Markitziu et al. (1985).

Histopathologie findings for rats and mice are

consistent with previously recognized to:xic effects of

sodium fluoride in Iabaratory rodents. Rats receiving

100 or 300 ppm sodium fluoride in the drinking

water for 6 months had slight hyperplasia of the

mucosal epithelium of the glandular stomach with

individual cell necrosis and hyperplasia of the

stratified epithelium of the forestomach near the

limiting ridge. These lesions are consistent with

low-grade cytotoxicity and increased cell tumover,

perhaps due to the formation of hydrogen fluoride

at the low pH in the gut (IPCS, 1984). However,

the lesions in the gastric mucosa occurred only in

rats, despite the fact that mice were given up to

twice the concentration given to rats. Hemorrhagic

gastroenteritis has been reponed in acute oral

fluoride toxicity in humans, and epithelial cell

degeneration and other changes in the duodenal

mucosa of rabbits receiving 10 mg/kg sodium Fluoride

per day for 24 months have been described

(Susheela and Das, 1988).

Mice that died during the 6-month studies had

microscopic lesions in the kidney, myocardium, liver,

and/or testis. The acute nephrosis observed in the

kidneys was considered a likely cause of death.

These findings were similar to the kidney injury

attributed to the administration of sodium fluoride

in rats of the Rochesterstrain (Taylor et al., 1961).

However, the F344 rats in the current 6-month

studies did not develop kidney lesions, perhaps due

to strain differences. The myocardial degeneration

and accumulation of mineral in degenerate

myofibers are also indicative of cytotoxicity.

Whether the lesions in the germinal epithelium of

the testis were due to a direct effect of fluoride on

the cells or secondary to debilitation is unknown,

but they are commonly seen in mice dying from any

one of a variety of unrelated causes.

 

The lesions of the incisor teeth were similar in rats

and mice in these 6-month studies and were consistent

with the findings of others (Yaeger, 1966;

Walton and Eisenmann, 1974). These changes were

observed in the incisor teeth, which are continuously

growing and erupting and thus retain a functionally

active enamel ("odontogenic") organ. Unlike the

incisor teeth, the molars of rodents do not continuously

grow and the enamel organ regresses.

Fluoride seems to exen its primary effects on the

secretory and maturation stages of the ameloblasts,

resulting in an increased organic :ontent and

decreased mineral content of the dental enamel

(Denbesten et al., 1985; Nordlund et al., 1986).

Lesions were observed in the femur and tibia of

mice receiving sodium Fluoride for 6 months in the

current studies, but not in rats. However, more

sensitive morphometric techniques might have

demonstrated changes in the bones of rats. The

lesions observed in mice are consistent with the

findings of others and are indicative of altered rates

of bone deposition and remodeling. The effects of

Fluoride on the bone have been studied in a variety

of species of varying ages. Baylink et al. (1970)

previously demonstrated increases in periosteal

matrix and bone formation, an inhibition of mineralization

at the periosteum, and an increase in

endosteal bone resorption in young Sprague-Dawley

rats receiving as little as 100 to 125 ppm Fluoride in

the drinking water for periods as shon as 2 weeks.

In contrast, Marie and Hott (1986) demonstrated

rapid stimulation of bone formation without detectable

change in resorption in C57BL/6J mice receiving

4 mg/L sodium Fluoride in the drinking water for

4 weeks.

Conclusions:
The teeth of rats and mice receiving the higher
doses of sodium fluoride were chalky white and
chipped or showed unusual wear patterns. Mice and
male rats given the higher concentrations had
microscopic focal degeneration of the enamel organ.
Rats receiving 100 or 300 ppm sodium fluoride had
minimal hyperplasia of the gastric mucosa of the
stomach, and one high-dose rat of each sex had an
ulcer. Acute nephrosis and/or lesions in the liver
and myocardium were observed in mice that died
early, and minimal alterations in bone
growth/remodeling were observed in the long bones
of mice receiving sodium fluoride at concentrations
of 50 to 600 ppm.
Executive summary:

Rats received concentrations of

sodium fluoride in drinking water as high as

300 ppm, and mice as high as 600 ppm. No rats

died during the studies; however, among the mice,

4/9 high-dose males, 9/11 high-dose females, and

1/8 males in the 300 ppm group died before the end

of the studies. Weight gains were less than those of

controls for rats receiving 300 ppm and mice receiving

200 to 600 ppm.

The tceth of rats and mice receiving tbe higher

doses of sodium fluoride were chalky white and

chipped or showed unusual wear patterns. Mice and

male rats given the higher concentrations bad

microscopic focal degeneration of the enamel organ.

Rats receiving 100 or 300 ppm sodium fluoride had

minimal hyperplasia of the gastric mucosa of the

stomach, and one high-dose rat of each sex had an

ulcer. Acute nephrosis and/or lesions in the liver

and myocardium were observed in mice that died

early, and minimal alterations in bone

growth/remodeling were observed in tbe long bones

of mice receiving sodium tluoride at concentrations

of 50 to 600 ppm.

Endpoint conclusion
Endpoint conclusion:
no adverse effect observed
Study duration:
chronic
Species:
other: rat and mice
Quality of whole database:
good quality

Carcinogenicity: via inhalation route

Endpoint conclusion
Endpoint conclusion:
no study available

Carcinogenicity: via dermal route

Endpoint conclusion
Endpoint conclusion:
no study available

Additional information

Justification for classification or non-classification

There was no evidence of carcinogenic activity in female F344/N rats receiving sodium fluoride at concentrations of 25, 100, or 175 ppm (11, 45, or 79 ppm fluoride) in drinking water for 2 years. There was no evidence of carcinogenic activity of sodium fluoride in male or female mice receiving sodium fluoride at concentrations of 25, 100, or 175 ppm in drinking water for 2 years.