Review

Tooth whitening products and the risk of oral cancer

I.C. Munro

a,*, G.M. Williams b, H.O. Heymann c, R. Kroes d

a Cantox Health Sciences International, Suite 308, 2233 Argentia Road, Mississauga, ON, Canada L5N 2X7

b Department of Pathology, New York Medical College, Valhalla, Basic Science Building, NY 10595, USA

c University of North Carolina, Chapel Hill, NC 27599-7450, USA

d Utrecht University, P.O. Box 80176, NI-3508 TD Utrecht, The Netherlands

Received 13 April 2005; accepted 21 July 2005

Abstract

Tooth whitening products (TWP) containing hydrogen peroxide (HPO) or carbamide peroxide (CPO) were evaluated in relation

to potential oral cancer risk from their use. HPO is genotoxic in vitro, but such activity is not expressed in vivo. The genotoxic risk
of HPO exposure of the oral mucosa encountered from TWP use is likely therefore to be vanishingly small. Available animal data on
the carcinogenicity of HPO are of limited relevance to risk assessment of oral hazard of HPO exposure from TWP, and where
relevant, do not indicate that there is an increased oral cancer risk for people using TWP. Clinical data on HPO-containing
TWP only show evidence of mild, transient gingival irritation and tooth sensitivity, with no evidence for the development of pre-
neoplastic or neoplastic oral lesions. Exposures to HPO received by the oral cavity, including areas commonly associated with oral
cancer, are exceedingly low and do not plausibly pose a risk for the promotion of initiated cells or for induction of co-carcinogenic
effects in conjunction with cigarette smoke or alcohol. The use of TWP was concluded not to pose an increased risk for oral cancer
in alcohol abusers and/or heavy cigarette smokers. Furthermore, TWP were concluded to be safe for use by all members of the
population, including potential accidental use by children.
Ó 2005 Elsevier Ltd. All rights reserved.

Keywords: Tooth whitening; Hydrogen peroxide; Oral cancer

Contents

1.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

2.

Genotoxicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

2.1.

In vitro data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

2.2.

In vivo data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

3.

Carcinogenicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

3.1.

Standard bioassays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

3.2.

Sequential exposure studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

3.3.

Combined exposure studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

0278-6915/$ - see front matter

Ó 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.fct.2005.07.012

Abbreviations: bw, body weight; CHO, Chinese hamster ovary; CPO, carbamide peroxide; DMBA, 7,12-dimethylbenza[a]anthracene; DNA,

deoxyribonucleic acid; GI, gastrointestinal; GLP, Good Laboratory Practice; HPO, hydrogen peroxide; MNNG, N-methyl-N0-nitro-N-
nitrosoguanidine; i.p., intraperitoneal; i.v., intravenous; MAM, methylazoxymethanol acetate; MTD, maximum tolerated dose; SCCP, European
UnionÕs Scientific Committee on Consumer Products; SCE, sister-chromatid exchanges; TWP, tooth whitening products; UDS, unscheduled DNA
synthesis.

* Corresponding author. Tel.: +1 905 542 2900; fax: +1 905 542 1011.

www.elsevier.com/locate/foodchemtox

Food and Chemical Toxicology xxx (2005) xxx–xxx

ARTICLE IN PRESS

4.

Clinical studies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

5.

Risk factors for the development of oral cancer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

6.

Discussion and conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 00

1. Introduction

Tooth whitening products (TWP) (e.g., strips, gels,

varnishes) that contain hydrogen peroxide (HPO), or
carbamide peroxide (CPO), a product that degrades to
form urea and HPO, have been in common use through-
out North America, particularly over the past 15 years.
Even though tooth whitening products have been in use
for over 100 years, heightened interest in tooth whiten-
ing arose following the introduction in 1989 of a partic-
ularly popular form of dentist-supervised bleaching,
called

nightguard

vital

bleaching

(Haywood

and

Heymann, 1989). Moreover, in North America, TWP
have been available directly to the consumer since early
2001. During this time no significant health effects from
use of TWP have been noted. In Europe, by contrast,
TWP containing HPO or CPO, are available to consum-
ers only from a dental practitioner. The legal status of
TWP, with respect to availability directly to the con-
sumer as cosmetic products was recently assessed by
the European UnionÕs Scientific Committee on Con-
sumer Products (SCCP, 2005). The Committee was of
the opinion that TWP containing from >0.1% to 6.0%
were safe for use upon consultation and approval of
the consumerÕs dentist. The SCCP raised concerns with
respect to the potential for HPO, including HPO gener-
ated from CPO, to be associated with an increased risk
of oral cancer, especially in smokers and alcohol abusers
(SCCP, 2004, 2005). Smokers and alcohol abusers have
a significantly elevated risk for the development of oral
cancer, with a reported synergistic effect of these 2 fac-
tors (Blot et al., 1988; Maier et al., 1992; Baron et al.,
1993).

Given the SCCP (2004, 2005) opinion, we undertook

a review of the available safety data on various TWP,
and HPO in particular, to assess the genotoxic and/or
carcinogenic risks posed by HPO exposures from the
use, both intended and exaggerated, of TWP. As part
of this evaluation, in vitro and in vivo genotoxicity stud-
ies, experimental animal studies, clinical tolerance stud-
ies involving TWP and human pharmacokinetic studies
were reviewed and assessed. In addition to these data,
the results of a large number of unpublished, and several
published, short- and longer-term clinical trials were
critically analyzed. The following presents a review of
the above safety data and conclusions with respect to
the potential for HPO to influence the development of
oral cancer in humans.

2. Genotoxicity

2.1. In vitro data

HPO generates reactive hydroxyl radicals that can

oxidize lipid (Kanner et al., 1987; OÕBrien, 1988) and
produce oxidative deoxyribonucleic acid (DNA) dam-
age (Williams and Jeffrey, 2000; Cadet et al., 2003). In
particular, the hydroxyl radical formed from HPO
reacts with deoxyguanosine to form 7,8-dihydro-8-
oxo-20-deoxyguanosine

(8-oxo-dG)

DNA

adducts

(Rosen et al., 1996). The 8-oxo-dG adducts are poten-
tially promutagenic adducts and mispair during DNA
replication to yield point mutations (Wood et al.,
1992; Kamiya, 2003). However, for mutagenicity to
occur, the DNA adducts must escape the effective
DNA repair process (Asagoshi et al., 2000; Slupphaug
et al., 2003), which is continuously dealing with the sub-
stantial levels of endogenous DNA oxidation that arise
from cellular metabolic activity (Williams and Jeffrey,
2000; Cooke et al., 2003). In mammalian cells, the deg-
radation of HPO is carried out by catalase and hydroxyl
radicals formed from HPO are scavenged by peroxidase
and the cellular stores of nucleophiles such as glutathi-
one and protein (Griffith and Mulcahy, 1999). As noted,
any 8-oxo-dG adducts that may be formed as a result of
exceeding the free radical scavenging capacity of the
cells, including cells of the oral mucosa, are known to
be excised by DNA repair enzymes. In particular, in hu-
mans 8 oxo-dG adducts are readily repaired by such
enzymes to the point that these adducts are not easily
converted into a mutagenic lesion (Asagoshi et al.,
2000; Lunec et al., 2002).

As expected, the in vitro genetic toxicity data clearly

show genotoxic effects of HPO. In the bacterial mutage-
nicity assays, positive results have been reported in Sal-
monella typhimurium strains TA102 and TA104, strains
that are known to be sensitive to oxidative DNA dam-
age (Levin et al., 1982; De Flora et al., 1984; Carlsson
et al., 1988; Glatt, 1989; Kensese and Smith, 1989;
Abu-Shakra and Zeiger, 1990; Wilcox et al., 1990; Li
et al., 1992; Nakayama et al., 1993).

The results of in vitro mammalian gene mutation as-

says, including the Chinese hamster ovary (CHO) V-79
hprt and the mouse lymphoma L5178Y hprt locus
assays, are mixed. One (Ziegler-Skylakakis and Andrae,
1987) of 6 studies (Bradley et al., 1979; Tsuda, 1981;
Bradley and Erickson, 1981; Nishi et al., 1984; Speit,

2

I.C. Munro et al. / Food and Chemical Toxicology xxx (2005) xxx–xxx

ARTICLE IN PRESS

1986; Ziegler-Skylakakis and Andrae, 1987) in CHO
cells reported a mutagenic effect at an HPO concentra-
tion of 17 lg/ml. The available mouse lymphoma assays
reported weak mutagenic activity of HPO at concentra-
tions of 0.17 and 0.34 lg/ml (Kruszewski et al., 1994).

HPO has been reported to induce sister-chromatid

exchanges (SCE) in several mammalian cell types,
including Chinese hamster V-79 cells (Bradley et al.,
1979; MacRae and Stich, 1979; Sasaki et al., 1980; Speit
et al., 1982; Estervig and Wang, 1984; Mehnert et al.,
1984a,b; Speit, 1986; Tucker et al., 1989; Diaz-Llera
et al., 2000).

In vitro exposures to HPO have also produced single

stranded DNA breaks (Bradley et al., 1979; Cantoni
et al., 1986; Prise et al., 1989; Kleiman et al., 1990; Dju-
ric et al., 1993), chromosomal aberrations (e.g., Hanham
et al., 1983; Estervig and Wang, 1984; Ishidate et al.,
1984; Oya et al., 1986; Fenech et al., 1999), induction
of unscheduled DNA synthesis (UDS) (Regnier et al.,
1996), and the induction of micronuclei (Sasaki et al.,
1980; Stich and Dunn, 1986).

The in vitro mutagenicity and clastogenicity data

must be interpreted in light of the fact that these test sys-
tems do not contain the in vivo levels of the enzymes
responsible for the detoxification of HPO. For example,
the inclusion of catalase enzymes in the test preparations
prevented the production of clastogenic effects (Hanham
et al., 1983; Estervig and Wang, 1984; Stich et al., 1978;
Tsuda, 1981). Finally, the inclusion of a metabolic acti-
vation system in the in vitro assays had the effect of
reducing or negating the effects of HPO (summarized
in IARC, 1999). This is due to either the direct detoxifi-
cation of HPO or the reaction of hydroxyl radicals
with the thiols or proteins in the metabolic activation
system.

2.2. In vivo data

In vivo data on the genotoxicity of HPO are limited

to a SCE assay in Chinese hamsters (Li et al., 1993),
an UDS assay in rats (Regnier et al., 1997), and a bone
marrow micronucleus assay in mice (Regnier et al.,
1996).

In the SCE assay, Li et al. (1993) exposed groups of

20 Chinese hamsters (sex not specified) to daily intuba-
tions of HPO at a dose of 70 mg/kg body weight (bw),
5 days per week, for a period of either 15 or 26 weeks.
Control groups of 20 received water. The frequency of
SCE in bone marrow cells was used to evaluate genotox-
icity. There was no difference (p > 0.05) in the SCE
frequency between the groups exposed to HPO or to
water for either 15 or 26 weeks. In a parallel experiment,
Rembrandt dental whitening gel, containing CPO, was
intubated to groups of 20 Chinese hamsters, at doses
of either 500 or 2000 mg/kg bw for 5 days/week, for
15 or 26 weeks. As with HPO, the CPO-containing gel

had no effect on SCE frequency in bone marrow cells
in comparison to water-exposed controls.

Regnier et al. (1997) compared the activity of HPO in

an ex vivo and an in vivo UDS assay in rat liver tissue.
Although only reported in an abstract, all experiments
were reportedly conducted according to OECD and
Good Laboratory Practice (GLP) standards. In the
in vivo assay, groups of five male Wistar rats were
administered HPO by intravenous (i.v.) infusion at
doses of 25 or 50 mg/kg bw, the maximum tolerated
dose. Animals were killed and hepatocytes isolated
and cultured for either 2–4 or 12–14 h after dosing.
Based on a mean net grain count of <

2.1, and the find-

ing that <0.7% of hepatocytes from the treated animals
were in a state of DNA repair, Regnier et al. (1997)
concluded that HPO was negative in the in vivo UDS as-
say. In comparison, in the ex vivo assay, incubation of
HPO with rat hepatocytes at concentrations of 0.8 to
50 lg/ml for either 2–4 or 12–14 h, elicited UDS, as
expected.

Regnier et al. (1996) also reported, in abstract form,

that HPO was inactive in a mouse bone marrow micro-
nucleus assay. In this study, 35% HPO (of a purity and
grade approved by the FDA for use in food processing)
in water was administered by intraperitoneal (i.p.) injec-
tion as a single 0, 250, 500, and 1000 mg HPO/kg bw
dose to groups of 5 Swiss OF1 mice. Cyclophosphamide
was used as a positive control. In an accompanying
oral study, a group of 10 male and female C57BL/6N
mice were administered HPO at a concentration of
6000 ppm in the drinking water (equivalent to doses of
536–774 mg/kg bw/day) as part of a 14-day subacute
toxicity study. As in the i.p. study, cyclophosphamide
was used a positive control. In the i.p. study, 24 and
48 h after dosing, and in the oral study, on day 14 of
dosing, after sacrifice, bone marrow was harvested and
2000 polychromatic erythrocytes were counted. Scoring
of 1000 total erythrocytes was conducted to establish the
polychromatic: normochromatic erythrocyte ratio. By
the i.p. route, all 3 dose levels at the 24-h harvest, and
the 500 and 1000 mg/kg bw dose levels at the 48-h har-
vest, caused a decrease in the polychromatic erythrocyte:
normochromatic erythrocyte ratio, indicative of toxicity
of HPO to the bone marrow. No effect on the polychro-
matic erythrocyte: normochromatic erythrocyte ratio
was reported after 14 days oral exposure to 6000 ppm
HPO in the drinking water. The frequency of micronu-
cleated erythrocytes was not increased relative to water
controls in either the i.p. or the drinking water studies.

The in vivo genotoxicity studies indicate that activity

demonstrated in vitro is not expressed in vivo. This is
likely related to the rapid detoxification of HPO and
scavenging of radicals prior to any opportunity to inter-
act with DNA.

Overall, the genotoxicity data indicate that while

HPO is predictably genotoxic under conditions that

I.C. Munro et al. / Food and Chemical Toxicology xxx (2005) xxx–xxx

3

ARTICLE IN PRESS

allow oxidative attack on DNA (i.e., high concentra-
tions and lack of detoxification systems), such activity
is not expressed in vivo. Taking into consideration the
foregoing, the genotoxic risk of exposures of the oral
mucosa (having considerable catalase activity in saliva
as well as the oral mucosa) to HPO encountered from
TWP under recommended conditions of use is likely to
be vanishingly small.

3. Carcinogenicity

Studies to assess the carcinogenicity of HPO in ro-

dents include an unpublished drinking water study of
the carcinogenicity of HPO in F344 rats (Takayama,
1981), oral administration to several strains of mice
(Ito et al., 1981a,b, 1982, 1984), several dermal skin
painting assays (Klein-Szanto and Slaga, 1982; Kurok-
awa et al., 1984), and an oral initiation–promotion study
in rats (Takahashi et al., 1986). Since these studies uti-
lized various designs and protocols, they are summa-
rized below according to study type.

3.1. Standard bioassays

In an unpublished study (Takayama, 1981), F344 rats

(50 per sex per group) were administered HPO in the
drinking water at dose levels of 0, 0.3 (195–306 mg/kg/
day), or 0.6% (433–677 mg/kg/day) for 18 months,
followed by a 6-month recovery period. This study
was thorough and collected data pertaining to mortality,
serum biochemistry, as well as of the histopathology of
all key organs (skin, mammary glands, pituitary,
thyroid, lung, pancreas, liver, adrenal, kidney, small
intestine, testis, muscle, peritoneum, eye spleen, stom-
ach, uterus, vagina, lymph nodes, as well as the oral cav-
ity and esophagus). Survival of treated rats was similar
to the controls (41/50), except for males treated at
0.3% (36/50). Following 45 weeks of treatment, body
weights were reduced by approximately 10% in the
high-dose groups, and by 6% in the low-dose group.
During the course of the study some nasal bleeding
was observed, the significance of which was not clear
since no additional data were reported regarding the his-
topathology of the nasal epithelium. There were no re-
ported statistically significant differences in tumor
incidence between the treated and control animals for
animals that died prior to termination or for animals
killed at the end of the recovery period. No tumors, or
any other adverse effects were reported to occur in the
oral cavity or esophagus, the 2 sites having initial con-
tact with HPO in drinking water. The authors concluded
that HPO was not carcinogenic to F344 rats. Findings of
reduced body weight gain of about 6 and 10% in the 0.3
and 0.6% HPO groups, respectively, indicate that an

adequate, near maximum tolerated dose (MTD), was
delivered in the study.

In a series of studies in mice, which included catalase-

deficient strains, Ito et al. (1981a,b, 1982, 1984) reported
weak pre-carcinogenic and carcinogenic effects of HPO
following administration in the drinking water for peri-
ods of up to 2 years. Administration of HPO to groups
of 48–50 male and female C57BL/6J mice (Ito et al.,
1981a,b, 1982) at 0%, 0.1%, or 0.4% (w/v) in the drink-
ing water (approximately 0, 250, or 1000 mg/kg/day) for
2 years yielded a slight increase in the incidence of
duodenal adenocarcinomas, but only when the results
for both sexes were combined (5/99, males and females)
(p = 0.05). The study authors concluded that the oral
administration of HPO to mice induced gastric erosion,
duodenal hyperplasia, and, at the high dose, duodenal
carcinoma. These results must be interpreted with cau-
tion as C57BL mice have low levels of catalase, and
may, therefore, be especially susceptible to HPO (Ito
et al., 1984). This fact was highlighted by the finding
in further studies (Ito et al., 1984) that preneoplastic/
neoplastic lesion development in the duodenum follow-
ing HPO treatment was inversely correlated with the
catalase activity of each strain of mouse tested; mice
with the highest catalase activity developed a low inci-
dence of duodenal hyperplastic/preneoplastic lesions.
No pathological findings in the oral cavity or in the
esophagus were reported despite the oral administration
of high HPO concentrations. The incidence of duodenal
tumors in four different strains of female mice adminis-
tered 0.4% HPO in the drinking water is shown below in
Table 1.

The relevance of the Ito et al. studies to humans is

limited because healthy humans are likely to have suffi-
cient peroxidase/catalase activity in saliva (Tipton et al.,
1995) and in the oral mucosa to deal with the extremely
low amounts of HPO released from TWP. Tipton et al.
(1995) have reported that human saliva effectively inhib-
its the cytotoxic effects of 0.05% HPO on human gingi-
val fibroblasts in vitro.

Table 1
Incidence of duodenal tumours in 4 strains of female mice adminis-
tered HPO in the drinking water at a concentration of 0.4% for about 6
months

Mouse
strain

Number

Catalase
activity

a

Number
with tumours
(%)

Total
number
of tumours

C3H/HeN

18

5.3 ± 1.4

2 (11.1)

2

B6C3F1

22

1.7 ± 0.2

7 (31.8)

8

C57BL/6N

21

0.7 ± 0.3

21 (100)

82

C3H/C

24

0.4 ± 0.1

22 (91.7)

63

Source: Ito et al. (1984) and SCCP (2004).

a Duodenal catalase activity expressed as 104 k/mg proteinin 6–

8 week old mice.

4

I.C. Munro et al. / Food and Chemical Toxicology xxx (2005) xxx–xxx

ARTICLE IN PRESS

3.2. Sequential exposure studies

In an initiation–promotion experiment in rats, in

which one of the study groups consisted of non-initiated
control rats dosed with 1.0% HPO in the drinking water
for 32 weeks, forestomach papillomas developed in 5/10
of the treated rats as compared to 0/10 in the non-initi-
ated controls not administered HPO (Takahashi et al.,
1986). No tumors of the oral cavity or esophagus were
reported.

In this study, enhanced tumor development was

reported in rats that were pre-treated with N-methyl-
N-nitro-N-nitrosoguanidine

(MNNG)

and

sodium

chloride, and then exposed to 1.0% HPO in the drinking
water for 32 weeks. The HPO-treated rats had a 100%
incidence of forestomach papillomas compared to 0%
in the initiated comparison group. The incidence of ade-
nocarcinoma of glandular stomach and duodenum was
not increased by HPO in comparison to initiation-only
controls, although the incidence of adenomatous hy-
perplasia in the glandular stomach was increased in
the initiated and HPO-treated group (8/21 or 38%) com-
pared to the initiated controls (0%).

The significance of forestomach tumors is question-

able given the fact that humans have no corresponding
organ. In the rat, the forestomach acts as a storage or-
gan rather than a digestive one. As a result, locally high
and longer duration exposures to forestomach epithe-
lium/mucosa would be expected. For this reason,
tumors of the forestomach, especially if related to
chronic tissue irritation, are generally considered to be
of little relevance to human carcinogenic risk (Wester
and Kroes, 1988; Grasso et al., 1991; Wurtzen, 1993;
IARC, 2003). Moreover, the effects in the forestomach
may represent a sequential syncarcinogenic effect of
DNA-reactive agents given in sequence rather than a
true promoting effect (Williams and Iatropoulos, 2001).

Despite the use of a strong alkylating agent and high

drinking water concentrations of HPO over a 32-week
span, no tumors or other adverse effects were reported
to occur in tissues proximal to the forestomach (i.e.,
the oral cavity and the esophagus). Strong alkylating
agents and other carcinogens have been shown to pro-
duce tumors of the oral cavity and esophagus, thus indi-
cating that these sites in rodent respond to genotoxic
insult (Gold et al., 2001).

The available skin painting initiation–promotion

studies in which mice were pre-treated with 7,12-dimeth-
ylbenz[a]anthracene (DMBA) followed by treatment
with HPO, failed to elicit any clear evidence of a tumor
promoting effect (Shamberger, 1972; Bock et al., 1975;
Klein-Szanto and Slaga, 1982; Kurokawa et al., 1984).
Although the dermal studies were negative, it should be
acknowledged that mouse skin, although a standard
assay (Enzmann et al., 1998), is not a perfect surrogate
for oral mucosa. Both are squamous epithelia, but mouse

skin has a greater degree of keratinisation as compared
to the oral mucosa. Thus, oral mucosa could be more
sensitive due to a greater degree of HPO penetration.
The gingiva, however, is more highly keratinized than
the floor of the mouth, and thus is more similar to mouse
skin. The studies cited above generally used acetone as
the dosing vehicle for HPO. This vehicle, based on data
available for benzoyl peroxide (Binder et al., 1997) likely
increased the absorption of HPO into the skin.

3.3. Combined exposure studies

In another rat experiment, Hirota and Yokoyama

(1981) studied the interactive effects of methylazoxy-
methanol acetate (MAM) treatment in conjunction with
HPO. Male Fischer 344 rats were exposed to 1.5% HPO
in the drinking water for 8 or 21 weeks, during which
time MAM was administered by i.p. administration
(25 mg/kg) in weeks 4, 6, and 8 of the study. No gastro-
intestinal (GI) tract tumors were observed in rats treated
with HPO alone for 21 weeks or in untreated controls.
Treatment of rats with HPO for 8 weeks, during which
time MAM was administered, and for a further
25 weeks resulted in a 100% incidence of duodenal
carcinomas. In rats exposed only to HPO during the first
8 weeks plus the MAM treatments, a 25% incidence of
duodenal carcinomas was reported. The lack of a
MAM only initiation control group limits the interpre-
tation of the study; however, the data are suggestive of
a syncarcinogenic effect (Williams and Iatropoulos,
2001) of HPO given following co-administration of
HPO and MAM. In a small group (n = 3) of rats given
only HPO, no tumors were recorded. As in the other
oral studies, including protocols for carcinogenicity
(Ito et al., 1981a,b, 1982, 1984; Takayama, 1981) and
initiation–promotion (Takahashi et al., 1986), no evi-
dence of any tumors in the upper alimentary tract was
reported in rats treated with HPO and MAM, or HPO
alone.

In a study in hamsters, the buccal pouches were

painted with 0.25 DMBA 2 times per week for 19–
22 weeks together with 30% HPO applied 2 times
per week the day following DMBA (Weitzman et al.,
1986). A marginally significant (p = 0.054) increase
was reported in the trend for cheek carcinoma incidence
in hamsters treated with DMBA and HPO for 22 weeks
(5/5 = 100%) versus DMBA alone (3/7 = 43%). These
results are uncertain given their marginal significance
due to the low numbers of animals used in the experi-
ment. Co-treatment of the buccal pouches with DMBA
and 3% HPO did not increase the incidence of carci-
noma (6/11 = 55%) in comparison to the DMBA only
controls. Also, no tumors were seen in hamsters
(n = 9) treated with 30% hydrogen peroxide alone twice
per week. Treatment at the high-concentration of 30%
HPO in uninitiated controls resulted in clear evidence

I.C. Munro et al. / Food and Chemical Toxicology xxx (2005) xxx–xxx

5

ARTICLE IN PRESS

of tissue irritation and toxicity as shown by the observa-
tion of chronic inflammation and cellular dysplasia. The
fact that tumors developed within the time frame of the
study (19–22 weeks) with DMBA application indicates
that the negative finding with HPO alone reflects non-
carcinogenicity.

In a later study, using groups of 25 hamsters of each

sex, Marshall et al. (1996) exposed the cheek pouches to
a solution containing 0.75% HPO along with 5% baking
soda 5 times per week for 20 weeks. Another group
received the same solution along with 0.5% DMBA. A
third group received a commercial dentifrice containing
3% HPO along with 0.5% DMBA. A control group re-
ceived 0.1 ml mineral oil. At the end of treatment, 0/
50 hamsters receiving only the 0.75% HPO/5% baking
solution presented with cheek pouch masses. In con-
trast, 40/50 of those exposed to DMBA alone had cheek
pouch masses. The incidence of cheek pouch masses in
the hamsters treated with DMBA and the 0.75%
HPO/5% baking soda solution (37/50) or the DMBA
and 0.5% HPO commercial dentifrice preparation (41/
50) were not significantly different from the group trea-
ted with DMBA alone.

In a second phase of the previous study, Marshall

et al. (1996) applied HPO (1.5%) in dentifrice formula-
tion (single or dual phase) mixtures with sodium bicar-
bonate (7.5%) to the cheek pouches of groups of 25
male and 25 female hamsters. Another group of ham-
sters was exposed to a solution of 3% HPO/7.5% baking
soda. In each of these groups, the cheek pouches were
co-treated with DMBA at concentrations of either
0.25% or 0.5%. DMBA applications were made 3 times
per week while the dentifrice preparations/solutions
were administered 5 times per week. In this second
phase, there was no HPO/baking soda exposure only
group. Since the cheek pouch carcinoma incidence was
close to 100% in the DMBA-only groups as well as in
the DMBA/HPO co-exposed groups, this phase of the
study was not capable of detecting any potential enhanc-
ing effect of HPO, if it existed.

Overall, the combined exposures studies, including

the hamster cheek pouch studies, can be considered to
provide limited evidence of a weak interactive effect of
relatively high concentrations of HPO (1.5–30% in aque-
ous solution) with co-treatment with potent DNA-reac-
tive initiating agents such as MAM and DMBA. It is
noteworthy that no carcinogenic effect of HPO alone
in the oral cavity was noted in the hamster cheek pouch
assays or in the oral dosing studies in mice and rats.

4. Clinical studies

There are over 100 clinical studies, most as yet

unpublished, comprising approximately 4000 subjects
in total, that have been conducted on HPO-containing

(5.33–16%) TWP. In addition, Leonard et al. (2003) re-
ported a 7.5-year follow-up study on a small group of
TWP users. In this follow-up study of 15 subjects who
received 6 months of continuous HPO treatment for tet-
racycline stains, no evidence of adverse effects in the oral
cavity were noted in 9 of the 15 who agreed to a clinical
examination. While the study is small in terms of num-
ber of subjects, thus limiting the value of statistical anal-
yses, none of the 15 participants reported any side effects
that they believed to have been treatment-related.
Studies have evaluated the effects of TWP under recom-
mended use conditions (1–2 weeks) and under condi-
tions of extended (up to 6 months) and exaggerated
use (four times application per day). Summaries of the
longer-term (i.e., 90–180 days of HPO exposure through
use of TWP) and of studies that incorporated long-term
follow-up of subjects following TWP product use are
provided in Tables 2 and 3, respectively.

The incidence of adverse effects, while quite variable,

is in all cases mild and transient and limited to gingival
irritation and tooth sensitization. These effects resolve
within a few days of ending product use.

Mild gingival irritation has not been reported to be a

risk factor for the development of oral cancer. More-
over, the gingiva is a very rare site for the development
of oral cancers. The most common sites, the floor of the
mouth and the lateral edge of the tongue (Cawson, 1975;
Mashberg and Meyers, 1976) have not been reported to
be adversely affected in any of the clinical studies on
TWP. Also, at these sites, HPO concentrations in saliva
[maximum concentration of 0.03% 1-min post applica-
tion (Slezak et al., 2002)] are very low in comparison
to HPO concentrations achieved on the gingiva [maxi-
mum median concentrations of 0.65% within 5 min of
application of 10% HPO TWP strips (unpublished clin-
ical trial)], the site adjacent to the application of TWP.
Even the highly variable incidence of gingival irritation
reported in the clinical studies may not entirely be the
result of HPO since many TWP contain dehydrant vehi-
cles such as glycerol. In addition, subjects in clinical
trials often traumatize gingival tissues through over
zealous brushing prior to dental visits.

Beyond the published and unpublished clinical data

accumulated to support the safety of TWP, over the last
4–5 years, millions of tooth whitening kits have been
sold directly to consumers, and, bleaching procedures
have been extensively conducted under the supervision
of dental professionals for the last 15 years, yet no pub-
lished reports of preneoplastic or neoplastic lesions asso-
ciated with their use have appeared in the scientific
literature to date. Since the incidence of gingival cancer
is likely less than 1 per 100,000 population (Sugerman
and Savage, 2002), if TWP were to cause neoplasms of
the gingiva, it would be expected that changes in the
background incidence rate would be relatively easy to
detect.

6

I.C. Munro et al. / Food and Chemical Toxicology xxx (2005) xxx–xxx

ARTICLE IN PRESS

Table 2
Clinical studies on the use of HPO-containing TWP for 31–180 days

Reference or
unpublished
study number

Number of
subjects enrolled

Formulation tested

Exposure

Safety outcomes

Study number 1998073

108

5.3% HPO gel strips,
pH 5.5, low viscosity

3-Phases

Tooth sensitivity was most frequent adverse event

Bleaching
regimen
comparison

Treatment phase:

4 weeks, 30 min per day

Tooth sensitivity incidence was not related to treatment
duration

8 weeks, 30 min per day

Maintenance phase:

4-Week treatment: 0.33 adverse events/subject;
14% of subjects had tooth sensitivity

30 min, once per week
30 min, 4 consecutive days in
the month

8-Week treatment: 0.35 adverse events/subject;
15% of subjects had tooth sensitivity

Re-treatment phase:

4 weeks, 30 min per day
8 weeks, 30 min per day

Study number
2000101

40

6.0% HPO gel strip

Maxillary application only

6% HPO strip: 20% of subjects had oral soft tissue adverse
events, 40% had tooth sensitivity

University of
North Carolina

Placebo

30 min per application, twice daily,
every day for 6 weeks

Placebo strip: 10% of subjects had oral soft tissue adverse
events, 10% had tooth sensitivity

No subjects experienced any adverse events other than oral
soft tissue irritation and tooth sensitivity

Karpina et al.
(2003)

50

5.3% HPO paint-on gel

Subjects used product once daily,
with the first dayÕs use
under supervision

All adverse events were mild in severity and resolved during
treatment or upon completion of treatment

Placebo paint-on gel

Product used overnight only,
on days 1–42

7 (14%) subjects reported possible/probable treatment
related oral symptom adverse events in the 5.3% HPO
paint-on gel group, of which 5 (20%) were tooth sensitivity
and 2 (8%) were oral soft tissue adverse events

One of the subjects discontinued treatment after week
4 due to tooth sensitivity

Placebo: 4 (16%) subjects reported oral soft tissue
adverse events

Study number 2002096

152

5.3% HPO paint-on gel

Used overnight for 6 weeks

All adverse events were mild in severity, except 1 oral soft
tissue event which was moderate in severity

Resolution of all adverse events occurred during treatment
or upon cessation of treatment

All oral soft tissue adverse events that were deemed possibly
or probably treatment related were reported by subjects,
with 1 exception, a mild aphthous stomatitis, which the
examiner observed

(continued on next page)

I.C.

Munro

et
al.
/

Food

and

Chemical

Toxicolog

y

xxx

(2005)

xxx–xxx

7

ARTICLE

IN

PRESS

Table 2 (continued)

Reference or
unpublished
study number

Number of
subjects enrolled

Formulation tested

Exposure

Safety outcomes

Overall, 8 (5.3%) subjects reported oral soft tissue adverse
events determined by the investigator to be possibly or
probably treatment related

4 (2.6%) subjects reported tooth sensitivity determined
by the investigator to be possibly or probably treatment
related

Kugel et al. (2002) and
Study number 2000043

40

6.3% HPO gel strips (N = 30)

Gel strips: 30 min,
twice daily for 6 months

6.3% HPO strip: 43% of subjects had oral soft tissue adverse
events, 47% had tooth sensitivity

10% CPO (

3.3% HP)

Opalescence in dental
tray (N = 10)

10% CPO dental tray:
overnight use

10% CPO dental tray: 30% of subjects had oral soft tissue
adverse events, 40% had tooth sensitivity

All of the subjects had
tetracycline stained teeth

All adverse events were reported; none were observed by
the examiner
No subject left the study due to adverse events

Study number 20022063

35

Crest White Strips Retail Kit
(6% HPO gel strips gel strips)

Maxillary application only

Crest White Strips: 6% of subjects had oral soft tissue
adverse events, 44% had tooth sensitivity

Safety/efficacy on Tetracycline
Dental Stain Loma Linda Univ.

9.5% HPO gel strips

30 min per application,
2 applications per day
for 3 months

9.5% HPO gel strips: 6% of subjects had oral soft tissue
adverse event, 59% had tooth sensitivity

Overall, 73% adverse events were mild in severity,
23% moderate, and 4% severe. No serious adverse events

One severe case of tooth sensitivity occurred in the 9.5% group

Haywood et al. (1997)

10

10% CPO in a custom-fitted
dental tray

Overnight use for
a period of 6 months

Four subjects discontinued product use during the first
2 weeks due to gingival irritation (oral soft tissue), tooth
sensitivity, and/or throat irritation/taste

All symptoms resolved within 24 h

8

I.C.

Munro

et
al.
/

Food

and

Chemical

Toxicolog

y

xxx

(2005)

xxx–xxx

ARTICLE

IN

PRESS

Table 3
Clinical studies on the use of HPO-containing TWP that included long-term follow-up

Reference or
unpublished
study number

Number of
subjects enrolled

Formulation tested

Exposure

Safety outcomes

Kugel and
Kastali (2000)

70

5.3% HPO gel strips

Maxillary application only

Tooth sensitivity was the only adverse event reported

Placebo

30 min per application, twice daily
for 2 weeks

5.3% HPO strip: 6% of subjects had oral soft tissue adverse
events

Follow-up visits at approximately
3- and 6-months post-treatment

Placebo strip: No adverse events were reported

Gingival Index and Plaque Index were the same for both groups
at the 3- and 6-month visits

Study number
1999091

70

5.3% HPO gel strips

Maxillary application only

5.3% HPO strip: 3% of subjects had oral soft tissue adverse events,
0% had tooth sensitivity

Hill Top, Florida

Placebo

30 min per application, twice daily
for 4 weeks

Placebo strip: No subjects experienced any oral soft tissue
irritation or tooth sensitivity

Follow-up visits at approximately
3- and 6-months post-treatment

No subjects experienced any effects other than oral soft tissue
irritation and/or tooth sensitivity

Gingival Index and Plaque Index were the same for both groups
at the 3 and 6-month visits

Study number
1999113

95

5.3% HPO gel strips

Maxillary application only

5.3% HPO strip (2 weeks): 39% of subjects had oral soft tissue
adverse events, 3% had tooth sensitivity, 3% had other, non-oral,
types of adverse events

Hill Top, Ohio

Placebo

30 min per application, twice daily
for 2 or 4 weeks

Follow-up visits at about
3- and 6-months post-treatment

5.3% HPO strip (4 weeks): 15% of subjects had oral soft tissue
adverse events, 15% had tooth sensitivity, but no other types of
adverse events

Placebo strip: 3% of subjects had oral soft tissue adverse events,
0% had tooth sensitivity or other types of adverse events

Gingival Index and Plaque Index were the same for both groups
at the 3 and 6-month visits

Leonard et al.
(1999)

21 initial, 12 completed
the follow-up period

10% CPO gel in
custom-fitted mouth
guard

10% CPO gel in custom-fitted
mouth guard, overnight for 6 months

Eighty percent of subjects reported adverse events during the
6 month treatment period

Follow-up visits at 6, 12,
and 54 months post-treatment

There were no reports of tooth whitener related adverse events
at the 6-month post-treatment visit

One subject reported having tooth sensitivity or gingival irritation
at the 12 month post-treatment visit

Three subjects reported having tooth sensitivity or gingival
irritation at 54 months

(continued on next page)

I.C.

Munro

et
al.
/

Food

and

Chemical

Toxicolog

y

xxx

(2005)

xxx–xxx

9

ARTICLE

IN

PRESS

An unpublished meeting abstract (Burningham et al.,

2004) suggested a possible association between the
development of oral cancer and use of TWP in younger
adults (<45 years of age). This abstract recorded the use
of TWP in 19 young adult patients with primary oral
cancer. Three persons (16%) reported a history of use
of TWP. The ages of the three persons who used TWP
(mean of 34.3 years) tended to be lower than those
who did not (mean of 52.4 years). The use of alcohol
and tobacco products was similar between users and
non-users of TWP. The oral cancers in three subjects
who used TWP had spread to regional lymph nodes,
while in the remaining 16 patients who did not use
TWP, three of the cancers were reported to have spread
to regional lymph nodes. The authors concluded that
their case reports do not show any causative association
between TWP use and oral cancer risk. Among the three
cases of primary oral cancer in young adults who had
used TWP, two were cases of tongue cancer in patients
who reported using TWP 2–3 years prior to diagnosis;
that is not a sufficient interval for induction of malig-
nant, metastatic tumors. The abstract provides no
biologically plausible basis for any potential association
between the use of TWP and the two cases of
tongue cancer. Moreover, in no clinical studies on
TWP have adverse effects on the tongue been reported
to occur.

5. Risk factors for the development of oral cancer

Oral cancer, more specifically squamous cell carci-

noma of the oral mucosa, is a multifactorial disease,
the primary risk factors for which are smoking and alco-
hol consumption/abuse. Other contributory factors that
have been identified include poor nutrition, poor oral
hygiene, viral infections, and exposure (occupational)
to carcinogens. The most common sites at which squa-
mous cell carcinoma is reported to occur in the oral cav-
ity is along the floor of the mouth and on the soft palate
(Cawson, 1975). The gingiva is only rarely identified as a
site for oral cancer development (Lesch et al., 1989;
Mashberg and Meyers, 1976).

More than 90% of persons who develop oral cancer

are smokers (Blot et al., 1988; Merletti et al., 1989; Bar-
on et al., 1993). In persons who smoke more than 1 pack
of cigarettes per week, the relative risks for the develop-
ment of oral cancer (7.3) are especially high in compar-
ison to persons who smoke less than 1 pack per week
(Maier et al., 1992). Life-long smokers of unfiltered
cigarettes have been reported to have approximately
double the risk for development of oral cancer
compared to life-long smokers of filtered cigarettes
(Mashberg et al., 1993; Andre et al., 1995). Cessation
of smoking results in a sharp decline in the risk for oral
cancer development, with no increased risk detectable

Ta

ble

3

(con

tinue

d

)

Refe

rence

or

unpu

blished

stud

y

numbe

r

Numbe

r

o

f

subjects

enrolled

Form

ulatio

n

tested

Expo

sure

Safe

ty

outco

mes

Leon

ard

et

al.

(2003

)

21

initial,

15

assessed

at

follow

-up

10%

CPO

in

cu

stom-fitted

mo

uth

gua

rd

10%

CPO

gel

in

custom

-fitted

mouth

guard,

overn

ight

for

6

month

s

One

sub

ject

rep

orted

having

too

th

sensitivity

at

the

90-mont

h

post

-treatm

ent

visit

;

howev

er,

this

subject

also

repor

ted

pre-

treatmen

t

too

th

sensitivity

Follo

w-up

visit

90

month

s

post-

treatmen

t

N

o

path

ologic

al

altera

tions

w

ere

seen

on

the

rad

iograp

hs

for

thes

e

sub

jects

Follo

w-up

was

an

extension

of

the

Leonard

et

al.

(1999

)

study

SEM

photom

icrogra

phs

ind

icated

no

ob

vious

differen

ces

between

the

facia

l

surfa

ces

of

the

treated

max

illary

teeth

and

the

un

treated

surfa

ces

of

the

mand

ibular

teeth

at

90

month

s

post-

treatmen

t

Ritt

er

et

al.

(2002

)

30

of

38

particip

ants

origina

lly

in

the

st

udy

were

asse

ssed

at

post-tr

eatment

follow

-up

10%

CPO

gel

in

a

cu

stom-fitted

night

guard

10%

CPO

gel

in

a

custom

-fitted

night

guard

tray,

overn

ight

for

6

w

eeks

For

the

exa

mined

teeth

,

93%

had

a

norm

al

gingival

ind

ex

sco

re,

5%

had

a

ging

ival

index

=

1

(mild

inflam

mation

)

and

1%

had

a

ging

ival

index

=

2

(mo

derate

inflam

mation

).

No

evidenc

e

o

f

ECR

was

fou

nd

durin

g

a

n

evalua

tion

of

the

X-rays

and

no

apical

lesions

were

obser

ved.

Gingival

ind

ex

and

ECR

fin

dings

were

consid

ered

norm

al

sugges

ting

minimal

post

w

hitening

eff

ects

at

10

years

po

st

treatm

ent

Subjects

were

eva

luated

for

gingival

ind

ex

and

externa

l

cervical

resorp

tion

(ECR)

by

radiog

raphic

examinat

ion

10

years

post

-treatm

ent

10

I.C. Munro et al. / Food and Chemical Toxicology xxx (2005) xxx–xxx

ARTICLE IN PRESS

after 10–15 years (Merletti et al., 1989; Franceschi et al.,
1992; Andre et al., 1995).

Excessive consumption of alcohol is a pronounced

risk factor for oral cancer development of nearly the
same order of magnitude as smoking status (Blot
et al., 1988; Maier et al., 1992; Mashberg et al., 1993;
Andre et al., 1995). As with smoking, a distinct linear
dose-response relationship between the amount of alco-
hol consumed and the relative risk for the development
of oral cancer has been reported. The type of alcoholic
beverage (e.g. wine, beer, spirits) consumed appears to
have little influence on the degree of risk for the develop-
ment of oral cancer (Blot et al., 1988; Kabat and
Wynder, 1989).

In contrast to cigarette smoke, which is known to

contain a number of carcinogenic chemicals, alcohol is
itself not a carcinogen, but functions as a co-carcinogen
or promoter (Seitz et al., 1998). The co-carcinogenic
effect of alcohol may be mediated by CYP 2E1 metabo-
lism to acetaldehyde. Although most metabolism of
ingested alcohol occurs in the liver, the oropharyngeal
mucosa contains relatively high concentrations of
metabolizing enzymes (Seitz et al., 1998).

The combination of smoking and heavy alcohol con-

sumption appears to have a synergistic, or greater than
additive (i.e., multiplicative), effect on the risk for devel-
opment of oral cancer (Blot et al., 1988; Maier et al.,
1992; Baron et al., 1993).

Given the potential, if not likely, use of TWP by

smokers and/or alcohol consumers, it is of interest to
evaluate the potential (exacerbation of) oral cancer risks
by TWP use, and HPO exposure, in these subjects. As
with smoking and alcohol consumption, increased can-
cer risk from combined exposures can arise when both
exposures each convey a cancer risk. For example, com-
bined smoking and asbestos exposures, which individu-
ally present cancer risks (IARC, 1977, 2002), convey
greatly increased risks for lung cancer (IARC, 1977).
However, since there is no established human cancer
risk from TWP or HPO, there is no basis to postulate
that there would be an increased risk from use of
TWP by individuals with exposure to products associ-
ated with risk of oral cancer, such as in smokers and/
or heavy drinkers.

The clinical studies on TWP, many of which would

have included smokers and/or alcohol consumers, pro-
vide no evidence to indicate that the rate or severity of
the adverse effects of TWP, namely mild, transient gingi-
val irritation and tooth sensitivity are significantly differ-
ent from non-smokers/alcohol consumers. Although,
there is no long-term follow-up (e.g., greater than
7.5 years) in smokers and non-smokers, no visible path-
ological changes that could plausibly be related to future
preneoplastic or neoplastic lesion development were
seen in any of the subjects in the over 100 clinical
trials.

6. Discussion and conclusions

As would be expected, HPO is genotoxic in vitro, un-

der conditions that allow oxidative attack on DNA (i.e.,
high concentrations and lack of detoxification systems).
Nevertheless, such activity is not expressed in vivo. HPO
at high concentrations is weakly carcinogenic to the
duodenum of mice, especially those that are catalase
deficient (Ito et al., 1981a,b, 1982, 1984). This animal
model is of limited relevance to humans because humans
have high levels of catalase activity, especially in the oral
cavity where exposure to HPO would occur with the use
of TWP. Similarly, the relevance of forestomach tumors
induced by a high drinking water concentration of 1%
HPO in rats (Takahashi et al., 1986) is highly question-
able given the lack of a human correlate for this organ
and the fact that chronic tissue irritation over a sus-
tained period often underlies tumor development in this
organ (Wester and Kroes, 1988; Grasso et al., 1991;
Wurtzen, 1993; Kraus et al., 1995; IARC, 2003).

The available sequential exposure study (Takahashi

et al., 1986) and combined exposure studies (Hirota
and Yokoyama, 1981; Weitzman et al., 1986) document
interactive effects of HPO with DNA-reactive carcino-
gens. The effects in the forestomach (Takahashi et al.,
1986) and duodenum (Hirota and Yokoyama, 1981)
are not relevant to risk assessment since the experimen-
tal conditions (i.e., use of potent DNA-reactive car-
cinogens for stomach and duodenum) are highly
artificial. The hamster cheek pouch experiments (Weitz-
man et al., 1986; Marshall et al., 1996) are conceptually
more relevant, but did not reveal a clear effect of
HPO.

Beyond the limitations of the design of these studies

with respect to human relevance, the results must be
interpreted in light of the exposure conditions experi-
enced by humans with TWP use. Based on available
data, salivary concentrations of HPO following applica-
tion of a TWP rapidly decline to near undetectable levels
within 15–60 min (Slezak et al., 2002; Mahony et al.,
2003). Moreover, based on a surface area exposure anal-
ysis (to account for the fact that effects in the human
oral mucosa would, if they were to occur, be associated
with site of contact concentrations, not systemic mg/kg
body weigh/day exposure rates), exposures in the carcin-
ogenicity/tumor

promotion/interaction

studies

are

orders of magnitude higher than would be experienced
by humans using TWP. Specifically, exposure of the
floor of the mouth to HPO from TWP use was calcu-
lated to be >400-fold lower than the dose used in mouse
dermal tumor initiation–promotion skin painting stud-
ies (Shamberger, 1972; Bock et al., 1975; Kurokawa
et al., 1984) and >100-fold lower than the dose used in
a hamster check pouch tumor initiation and promotion
studies in which no carcinogenic effects were observed
(Marshall et al., 1996).

I.C. Munro et al. / Food and Chemical Toxicology xxx (2005) xxx–xxx

11

ARTICLE IN PRESS

A comparison of the maximal drinking water HPO

concentrations of 0.4%, 1.0%, and 1.5% utilized in the
Ito et al. (1981a,b, 1982, 1984) mouse studies, in the
Takahashi et al. (1986) rat initiation–promotion study,
and in the Hirota and Yokoyama (1981) combined
exposure (with MAM) study, respectively, with HPO
concentrations in saliva after application of TWP also
reveals large differences in exposure. The above drinking
water HPO concentrations (continuous exposure) are
from 13-fold to 33-fold-greater than the peak concentra-
tions of HPO in the saliva of 0.03% (representative data)
achieved 1 min after application of TWP (Slezak et al.,
2002). Given the rapid disappearance of HPO in the sal-
iva (undetectable within 15–60 min), daily exposures in
the animal studies were in fact likely 1000 s fold greater
than exposure to HPO from TWP since during the time
that the animals consumed water, and during residency
time in the stomach and duodenum, HPO concentra-
tions would remain near the nominal concentrations
used in each study (i.e., constant exposure to 0.4–1.0%
HPO concentrations during periods of water consump-
tion and storage/transit through the stomach and
duodenum).

In addition to the low rates of exposure of the human

oral mucosa to HPO from the use of TWP, exposures
are generally short-term (minutes post-application) and
intermittent in nature (e.g., exposure periods of up to
14 days 2 or 3 times per year). Accordingly, the weak
carcinogenic, promoting and or enhancing effect of re-
peated or sustained exposures to much higher concen-
trations of HPO as was the case in the development of
duodenal, gastric and forestomach tumors in rodents
are not at all comparable to the very low, short-term
and intermittent exposures to HPO from TWP use in
humans.

The more than the 100 published and unpublished

clinical studies on HPO-containing TWP demonstrate
that the only findings of clinical significance are variable
incidences of tooth sensitivity and of mild temporary
gingival irritation. In each case the effects are transient
and usually disappear within a few days. No long-term
sequelae or pathology has been reported in any of these
studies. Similarly, a 7.5-year follow-up study on a small
group of TWP users (Leonard et al., 2003) found no evi-
dence of any adverse effect, even after use of TWP for six
continuous months. The clinical data provide no evi-
dence to indicate that HPO in TWP has preneoplastic
or neoplastic potential in humans.

Mild gingival irritation has not been reported to be a

risk factor for the development of oral cancer and the
gingiva is a very rare site for the development of oral
cancers (Cawson, 1975; Mashberg and Meyers, 1976).
The most common sites, the floor of the mouth and
the lateral edge of the tongue have not been reported
to be adversely affected in any of the clinical studies
on TWP.

While smoking and alcohol use have a synergistic

effect on the risk for development of oral cancer, the
theoretical risk from the use of TWP, even under exag-
gerated use conditions, to smokers and/or heavy drink-
ers must be put in perspective with the fact that risks for
oral cancers are significantly increased only after pro-
longed and sustained high-level usage of known carcin-
ogenic agents. Exposures of gingival cells, and cells of
the oral mucosa at other sites as well, to HPO are
exceedingly low and very brief, thus are highly unlikely
to pose an added risk for the development of cancer by
cells initiated by tobacco smoke carcinogens or by the
co-carcinogenic effects of alcohol.

A classical ‘‘promoting’’ effect of TWP can be dis-

counted because such an effect in animals typically in-
volves high and sustained exposures. For promoters to
be effective, continuous long-term sustained high-level
exposures are required and, interruption of which gener-
ally results in the lack of initial development of preneo-
plastic/neoplastic lesions or regression of any lesions
formed (Burns et al., 1976; Williams and Whysner,
1996). Similarly, such exposures typically produce clear
signs of tissue injury at the affected site [e.g., forestom-
ach (Wester and Kroes, 1988; Grasso et al., 1991; Wurt-
zen, 1993; IARC, 2003) and skin (summarized in Kraus
et al., 1995)]. TWP use is not associated with any clinical
signs of sustained tissue injury. Therefore, the results of
initiation–promotion or combined exposure studies,
typically conducted to address mechanisms of carcino-
genesis and generally not used in human cancer risk
assessment (Kraus et al., 1995), should not be extrapo-
lated to suggest a potential risk of HPO to the oral
mucosa from TWP under recommended, exaggerated
or extended, conditions of use. This conclusion is also
supported by the fact that there are many common
rodent tumor promoters, including food ingredients
such as sodium chloride, ascorbates, butylated hydroxy-
anisole, glycerine, and sucrose, but very few, if any, are
known human tumor promoters (Kraus et al., 1995).

There exists the possibility of accidental use of TWP

by children. Such use however, would be rare and for
the most part would be a ‘‘one-time’’ occurrence. No
excess cancer risk due to tumor promoting effects would
occur since long-duration and continuous exposures are
required for the interaction of tumor promoters with ini-
tiating carcinogens.

In conclusion, the available genetic toxicity and ani-

mal toxicology data do not indicate that HPO poses a
carcinogenic risk to the human oral mucosa. This
conclusion is further bolstered by the results of the dosi-
metric exposure analyses from TWP users showing mar-
gins-of-safety on the order of 100–1000 s of fold between
no effect levels in animal studies and transient peak HPO
concentrations in saliva at the floor of the mouth. More-
over, HPO concentrations are highest in the gingiva, a
site where oral cancer is rarely found and humans have

12

I.C. Munro et al. / Food and Chemical Toxicology xxx (2005) xxx–xxx

ARTICLE IN PRESS

sufficient catalase activity in saliva and oral mucosa to
effectively detoxicate HPO at such low exposure levels.

Acknowledgements

This review was funded by a consortium of compa-

nies belonging to the European Cosmetic Toiletry and
Perfumery Association (COLIPA), Avenue Herrmann
Debroux 15A B-1160 Auderghem—Brussels, Belgium.

The authors would like thank Dr. Donna McMillan

of Procter & Gamble (Mason, OH USA 45040-9762)
for providing the cited unpublished clinical trial data.

References

Abu-Shakra, A., Zeiger, E., 1990. Effects of Salmonella genotypes and

testing protocols on H2O2-induced mutation. Mutagenesis 5, 469–
473.

Andre, K., Schraub, S., Mercier, M., Bontemps, P., 1995. Role of

alcohol and tobacco in the aetiology of head and neck cancer: a
case-control study in the Doubs region of France. Eur. J. Cancer B.
Oral Oncol. 31B, 301–309.

Asagoshi, K., Yamada, T., Terato, H., Ohyama, Y., Monden, Y.,

Arai, T., Nishimura, S., Aburatani, H., Lindahl, T., Ide, H., 2000.
Distinct repair activities of human 7,8-dihydro-8-oxoguanine DNA
glycosylase and formamidopyrimidine DNA glycosylase for foram-
idopyrimidine 8-dihydro-8-oxoguanine. J. Biol. Chem. 275, 4956–
4964.

Baron, A.E., Franceschi, S., Barra, S., Talamini, R., La Vecchia, C.,

1993. A comparison of the joint effects of alcohol and smoking on
the risk of cancer across sites in the upper aerodigestive tract.
Cancer Epidemiol. Biomarkers Prev. 2, 519–523.

Binder, R.L., Firriolo, J., Totman, L.C., Goldenthal, E.I., Nash, J.F.,

Kraus, A.L., 1997. Influence of vehicle on skin response to benzoyl
peroxide (BPO) in F344 rats and B6C3F1 mice. Fund. Appl.
Toxicol. Suppl. 36, 188 [Abstract No. 955].

Blot, W.J., McLaughlin, J.K., Winn, D.M., Austin, D.F., Greenberg,

R.S., Preston-Martin, S., Bernstein, L., Schoenberg, J.B., Stemha-
gen, A., Fraumeni Jr., J.F., 1988. Smoking and drinking in relation
to oral and pharyngeal cancer. Cancer Res. 48, 3282–3287.

Bock, F.G., Myers, H.K., Fox, H.W., 1975. Cocarcinogenic activity of

peroxy compounds. J. Natl. Cancer Inst. 55, 1359–1361.

Bradley, M.O., Erickson, L.C., 1981. Comparison of the effects of

hydrogen peroxide and X-ray irradiation on toxicity, mutation,
and DNA damage/repair in mammalian cells (V-79). Biochim.
Biophys. Acta 654, 135–141.

Bradley, M.O., Hsu, I.C., Harris, C.C., 1979. Relationship between

sister chromatid exchange and mutagenicity, toxicity and DNA
damage. Nature 282, 318–320.

Burningham, A.R., Davidson, B.J., Malekzadeh, S., Dasgupta, R.,

Yoder, E., Newkirk, K.A., 2004. Do Teeth Whiteners Lead to Oral
Cancer? [Press Release Summarizing ‘‘Tooth Whiteners as a Risk
Factor for Oral Cavity Squamous Cell Carcinoma: A Report of
Cases’’. Presented at the 6th International Conference on Head and
Neck Cancer, Aug. 7–11, 2004, Washington, DC]. Available from:
<http://www.innovations-report.com/html/reports/medicine_health/
report-32206.html>.

Burns, F., Vanderlaan, M., Sivak, A., Albert, R., 1976. The regression

kinetics of mouse skin papillomas. Cancer Res. 36, 1422–1427.

Cadet, J., Douki, T., Gasparutto, D., Ravanat, J.L., 2003. Oxidative

damage to DNA: formation, measurement and biochemical
features. Mutat. Res. 531, 5–23.

Cantoni, O., Murray, D., Meyn, R.E., 1986. Effect of 3-aminobenza-

mide on DNA strand-break rejoining and cytotoxicity in CHO cells
treated with hydrogen peroxide. Biochim. Biophys. Acta 867, 135–
143.

Carlsson, J., Berglin, E.H., Claesson, R., Edlund, M.B., Persson, S.,

1988. Catalase inhibition by sulfide and hydrogen peroxide-induced
mutagenicity in Salmonella typhimurium strain TA102. Mutat.
Res. 202, 59–64.

Cawson, R.A., 1975. Premalignant lesions in the mouth. Br. Med. Bull.

31, 164–168.

Cooke, M.S., Evans, M.D., Dizdaroglu, M., Lunec, J., 2003. Oxidative

DNA damage: mechanisms, mutation, and disease. FASEB J. 17,
1195–1214.

De Flora, S., Camoirano, A., Zanacchi, P., Bennicelli, C., 1984.

Mutagenicity testing with TA97 and TA102 of 30 DNA-damaging
compounds, negative with other Salmonella strains. Mutat. Res.
134, 159–165.

Diaz-Llera, S., Podlutsky, A., Osterholm, A.M., Hou, S.M., Lambert,

B., 2000. Hydrogen peroxide induced mutations at the HPRT locus
in

primary

human

T-lymphocytes.

Mutat.

Res.

469,

51–

61.

Djuric, Z., Everett, C.K., Luongo, D.A., 1993. Toxicity, single-strand

breaks, and 5-hydroxymethyl-20-deoxyuridine formation in human
breast epithelial cells treated with hydrogen peroxide. Free Radic.
Biol. Med. 14, 541–547.

Enzmann, H., Bomhard, E., Iatropoulos, M., Ahr, H.J., Schlueter, G.,

Williams, G.M., 1998. Short- and intermediate-term carcinogenic-
ity testing–A review. Part 1: The prototypes mouse skin tumor
assay and rat liver focus assay. Food Chem. Toxicol. 36, 979–
995.

Estervig, D., Wang, R.J., 1984. Sister chromatid exchanges and

chromosome aberrations in human cells induced by H2O2 and
other photoproducts generated in fluorescent light-exposed med-
ium. Photochem. Photobiol. 40, 333–336.

Fenech, M., Crott, J., Turner, J., Brown, S., 1999. Necrosis, apoptosis,

cytostasis and DNA damage in human lymphocytes measured
simultaneously within the cytokinesis-block micronucleus assay:
description of the method and results for hydrogen peroxide.
Mutagenesis 14, 605–612.

Franceschi, S., Barra, S., La Vecchia, C., Bidoli, E., Negri, E.,

Talamini, R., 1992. Risk factors for cancer of the tongue and the
mouth. A case-control study from northern Italy. Cancer 70, 2227–
2233.

Glatt, H., 1989. Mutagenicity spectra in Salmonella typhimurium

strains of glutathione, L-cysteine and active oxygen species.
Mutagenesis 4, 221–227.

Gold, L.S., Manley, N.B., Slone, T.H., Ward, J.M., 2001. Compen-

dium of chemical carcinogens by target organ: Results of chronic
bioassays in rats, mice, hamsters, dogs, and monkeys. Toxicol.
Pathol. 29, 639–652.

Grasso, P., Sharratt, M., Cohen, A.J., 1991. Role of persistent, non-

genotoxic tissue damage in rodent cancer and relevance to humans.
Annu. Rev. Pharmacol. Toxicol. 31, 253–287.

Griffith, O.W., Mulcahy, R.T., 1999. The enzymes of glutathione

synthesis: gamma-glutamylcysteine synthetase. Adv. Enzymol.
Relat. Areas Mol. Biol. 73, 209–267.

Hanham, A.F., Dunn, B.P., Stich, H.F., 1983. Clastogenic activity of

caffeic acid and its relationship to hydrogen peroxide generated
during autooxidation. Mutat. Res. 116, 333–339.

Haywood, V.B., Heymann, H.O., 1989. Nightguard vital bleaching.

Quintessance Inter. 20, 173–176.

Haywood, V.B., Leonard, R.H., Dickinson, G.L., 1997. Efficacy of six

months of night guard vital bleaching of tetracycline-stained teeth.
J. Esthet. Dent. 9, 13–19.

Hirota, N., Yokoyama, T., 1981. Enhancing effect of hydrogen

peroxide upon duodenal and upper jejunal carcinogenesis in rats.
GANN 72, 811–812.

I.C. Munro et al. / Food and Chemical Toxicology xxx (2005) xxx–xxx

13

ARTICLE IN PRESS

IARC, 1977. Asbestos. International Agency for Research on Cancer

(IARC), Lyon, France. IARC Monographs on the Evaluation of
Carcinogenic Risks to Humans, Vol. 14.

IARC, 1999. Hydrogen peroxide. In: Re-Evaluation of Some Organic

Chemicals, Hydrazine and Hydrogen Peroxide. [IARC Working
Group Meeting], Feb. 17–24, 1998, Lyon, France. International
Agency for Research on Cancer (IARC), Lyon, France. IARC
Monographs on the Evaluation of Carcinogenic Risks to Humans,
Vol. 71, Part 2, pp. 671–689.

IARC, 2002. Tobacco Smoking And Tobacco Smoke. International

Agency for Research on Cancer (IARC), Lyon, France. IARC
Monographs on the Evaluation of Carcinogenic Risks to Humans,
Vol. 83.

IARC, 2003. Predictive Value of Rodent Forestomach and Gastric

Neuroendocrine Tumors in Evaluating Carcinogenic Risk to
Humans. International Agency for Research on Cancer (IARC),
Lyon, France. IARC Technical Publication No. 39.

Ishidate Jr., M., Sofuni, T., Yoshikawa, K., Hayashi, M., Nohmi, T.,

Sawada, M., Matsuoka, A., 1984. Primary mutagenicity screening
of food additives currently used in Japan. Food Chem. Toxicol. 22,
623–636.

Ito, A., Naito, M., Watanabe, H., 1981a. Implication of chemical

carcinogenesis in the experimental animal - tumorigenic effect of
hydrogen peroxide in mice. Nenpo (Hiroshima Daigaku Genbaku
Hoshano Igaku Kenkyujo) 22, 147–158.

Ito, A., Watanabe, H., Naito, M., Naito, Y., 1981b. Induction of

duodenal tumors in mice by oral administration of hydrogen
peroxide. GANN 72, 174–175.

Ito, A., Naito, M., Naito, Y., Watanabe, H., 1982. Induction and

characterization of gastro-duodenal lesions in mice given continu-
ous oral administration of hydrogen peroxide. GANN 73, 315–322.

Ito, A., Watanabe, H., Naito, M., Naito, Y., Kawashima, K., 1984.

Correlation between induction of duodenal tumor by hydrogen
peroxide and catalase activity in mice. GANN 75, 17–21.

Kabat, G.C., Wynder, E.L., 1989. Type of alcoholic beverage and oral

cancer. Int. J. Cancer 43, 190–194.

Kamiya, H., 2003. Mutagenic potentials of damaged nucleic acids

produced by reactive oxygen/nitrogen species: Approaches using
synthetic oligonucleotides and nucleotides: survey and summary.
Nucleic Acids Res. 31, 517–531.

Kanner, J., German, J.B., Kinsella, K.E., 1987. Initiation of lipid

peroxidation in biological systems. Crit. Rev. Food Sci. Nutr. 25,
317–364.

Karpina, K.A., Magnusson, I., Barker, M.L., Gerlach, R.W., 2003.

Placebo-controlled clinical trial of a19% sodium pentacarbonate
whitening film: initial and sustained whitening. Am. J. Dent. (Spec.
No.), 12B–16B.

Kensese, S.M., Smith, L.L., 1989. Hydrogen peroxide mutagenicity

towards Salmonella typhimurium. Teratog. Carcinog. Mutagen. 9,
211–218.

Kleiman, N.J., Wang, R.R., Spector, A., 1990. Hydrogen peroxide-

induced DNA damage in bovine lens epithelial cells. Mutat. Res.
240, 35–45 [Erratum, 241(4):395].

Klein-Szanto, J.P., Slaga, T.J., 1982. Effects of peroxides on rodent

skin: Epidermal hyperplasia and tumor promotion. J. Invest.
Dermatol. 79, 30–34.

Kraus, A.L., Munro, I.C., Orr, J.C., Binder, R.L., LeBoeuf, R.A.,

Wiliams, G.A., 1995. Benzoyl peroxide: An integrated human
safety assessment for carcinogenicity. Regul. Toxicol. Pharmacol.
21, 87–107.

Kruszewski, M., Green, M.H.L., Lowe, J.E., Szumiel, I., 1994. DNA

strand breakage, cytotoxicity and mutagenicity of hydrogen
peroxide treatment at 4

°C and 37 °C in L5178Y sublines. Mutat.

Res. 308, 233–241.

Kugel, G., Kastali, S., 2000. Tooth-whitening efficacy and safety: a

randomized and controlled clinical trial. Compend. Contin. Educ.
Dent. 29, S16–S21.

Kugel, G., Aboushala, A., Zhou, X., Gerlach, R.W., 2002. Daily use of

whitening strips on tetracycline-stained teeth: comparative results
after 2 months. Compend. Contin. Educ. Dent. 23, 29–34.

Kurokawa, Y., Takamura, N., Matsughima, Y., Imazawa, T.Y.,

Hayashi, Y., 1984. Studies on the promoting and complete
carcinogenic activities of some oxidizing chemicals in skin carci-
nogenesis. Cancer Lett. 24, 299–304.

Leonard Jr., R.H., Haywood, V.B., Eagle, J.C., Garland, G.E.,

Caplan, D.J., Matthews, K.P., Tart, N.D., 1999. Nightguard vital
bleaching of tetracycline-stained teeth: 54 months post treatment.
J. Esthet. Dent. 11, 265–277.

Leonard Jr., R.H., Van Haywood, B., Caplan, D.J., Tart, N.D., 2003.

Nightgaurd vital bleaching of tetracycline-stained teeth: 90 months
post treatment. J. Esthet. Restor. Dent. 15, 142–152.

Lesch, C.A., Squier, C.A., Cruchley, A., Williams, D.M., Speight, P.,

1989. The permeability of human oral mucosa and skin to water. J.
Dent. Res. 68, 1345–1349.

Levin, D.E., Hollstein, M., Christman, M.F., Schwiers, E.A., Ames,

B.N., 1982. A new Salmonella tester strain (TA102) with A X T
base pairs at the site of mutation detects oxidative mutagens. Proc.
Natl. Acad. Sci. USA 79, 7445–7449.

Li, Y., Noblitt, T., Dunipace, A., Stookey, G., 1992. Evaluation of

genotoxicity of a tooth whitener. J. Dent. Res. 71, 157 [Abstract
No. 413].

Li, Y., Noblitt, T., Zhang, A., Origel, A., Kafrawy, A., Stookey, G.,

1993. Effect of long-term exposure to a tooth whitener. J. Dent.
Res. 72, 248 [Abstract No. 1162].

Lunec, J., Holoway, K.A., Cooke, M.S., Faux, S., Griffiths, H.R.,

Evans, M.D., 2002. Urinary8-oxo-20-deoxyguanosine: redox regu-
lation of DNA repair in vivo? Free Radic. Biol. Med. 33, 875–885.

MacRae, W.D., Stich, H.F., 1979. Induction of sister-chromatid

exchanges in Chinese hamster ovary cells by thiol and hydrazine
compounds. Mutat. Res. 68, 351–365.

Mahony, C., Barker, M.L., Engel, T.M., Walden, G.L., 2003. Peroxide

degradation kinetics of a direct application percarbonate bleaching
film. Am. J. Dent. 16 (Spec No), 9B–11B.

Maier, H., Dietz, A., Gewelke, U., Heller, W.D., Weidauer, H., 1992.

Tobacco and alcohol and the risk of head and neck cancer. Clin.
Investig. 70, 320–327.

Marshall, M.V., Kuhn, J.O., Torrey, C.F., Fischman, S.L., Cancro,

L.P., 1996. Hamster cheek pouch bioassay of dentifrices containing
hydrogen peroxide and baking soda. J. Am. Coll. Toxicol. 15, 45–
61.

Mashberg, A., Meyers, H., 1976. Anatomical site and size of 222 early

asymptomatic oral squamous cell carcinomas: a continuing pro-
spective study of oral cancer. II. Cancer 31, 2149–2157.

Mashberg, A., Boffetta, P., Winkelman, R., Garfinkel, L., 1993.

Tobacco smoking, alcohol drinking, and cancer of the oral cavity
and oropharynx among US veterans. Cancer 72, 1369–1375.

Mehnert, K., During, R., Vogel, W., Speit, G., 1984a. Differences in

the induction of SCEs between human whole blood cultures and
purified lymphocyte cultures and the effect of an S9 mix. Mutat.
Res. 130, 403–410.

Mehnert, K., Vogel, W., Benz, R., Speit, G., 1984b. Different effects of

mutagens on sister chromatid exchange induction in three Chinese
hamster cell lines. Environ. Mutagen. 6, 573–583.

Merletti, F., Boffetta, P., Ciccone, G., Mashberg, A., Terracini, B.,

1989. Role of tobacco and alcoholic beverages in the etiology of
cancer of the oral cavity/oropharynx in Torino, Italy. Cancer Res.
49, 4919–4924.

Nakayama, T., Hiramitsu, M., Osawa, T., Kawakishi, S., 1993. The

protective role of gallic acid esters in bacterial cytotoxicity and SOS
responses induced by hydrogen peroxide. Mutat. Res. 303, 29–34.

Nishi, Y., Hasegawa, M.M., Taketomi, M., Ohkawa, Y., Inui, N.,

1984. Comparison of 6-thioguaninine-resistant mutation and sister
chromatid exchanges in Chinese hamster V79 cells with forty
chemical ands physical agents. Cancer Res. 44, 3270–3279.

14

I.C. Munro et al. / Food and Chemical Toxicology xxx (2005) xxx–xxx

ARTICLE IN PRESS

OÕBrien, P.J., 1988. Radical formation during the peroxidase catalyzed

metabolism of carcinogenic and xenobiotics: the reactivity of these
radicals with GSH, DNA, and unsaturated lipid. Free Radic. Biol.
Med. 4, 169–183.

Oya, Y., Yamamoto, K., Tonomura, A., 1986. The biological activity

of hydrogen peroxide. I. Induction of chromosome-type aberra-
tions susceptible to inhibition by scavengers of hydroxyl radicals
in

human

embryonic

fibroblasts.

Mutat.

Res.

172,

245–

253.

Prise, K.M., Davies, S., Michael, B.D., 1989. Cell killing and DNA

damage in Chinese hamster V79 cells treated with hydrogen
peroxide. Int. J. Radiat. Biol. 55, 583–592.

Regnier, J.-F., Molinier, B., Bentley, K.S., de Gerlache, J., Malin-

verno, G., Mayr, W., Weiner, M.L., Trochimowicz, H., Brock, W.,
1996. Micronucleus tests in mice with hydrogen peroxide. Fund.
Appl. Toxicol. Suppl. 30, 233.

Regnier, J.-F., Clare, C., de Gerlache, J., Malinverno, G., Mayr, W.,

Weiner, M.L., Trochimowicz, H., 1997. Ex vivo and in vitro
unscheduled DNA synthesis (UDS) assays in rat liver with
hydrogen peroxide (H2O2). Mutat. Res. 379, S168–S169.

Ritter, A.V., Leonard Jr., R.H., StGeorges, A.J., Caplan, D.J.,

Haywood, V.B., 2002. Safety and stability of Nightguard vital
bleaching: 9 to 12 years post-treatment. J. Esthet. Restor. Dent. 14,
275–285.

Rosen, J.E., Prahalad, A.K., Wiliams, G.M., 1996. 8-Oxodeoxygu-

anosine formation in the DNA of cultured cells after exposure to
H2O2 alone or with UVB or UVA irradiation. Photochem.
Photobiol. 64, 117–122.

Sasaki, M., Sugimura, K., Yoshida, M.A., Abe, S., 1980. Cytogenetic

effects of 60 chemicals on cultured human and Chinese hamster
cells. Kromosomo 2, 574–584.

SCCP, 2005. Opinion on Hydrogen Peroxide in Tooth Whitening

Products. The Scientific Committee on Consumer Products
(SCCP). Adopted by the SCCP during the 3

rd Plenary of 15

March 2005. SCCP/0844/04. Available at http://europa.eu.int/
comm/health/ph_risk/committees/04_sccp/docs/sccp_o_022.pdf
(accessed on 21 July 2005).

SCCP, 2004. Public Consultation on a Preliminary Opinion on

Hydrogen Peroxide in Tooth Whitening Products. The Scientific
Committee on Consumer Products (SCCP). Approved by the 2

nd

Plenary of 7 December 2004. SCCP/0844/04. Available at http://
europa.eu.int/comm/health/ph_risk/committees/04_sccp/docs/sccp_
mi_002.pdf (accessed on 21 July 2005).

Seitz, H.K., Simanowski, U.A., Poschl, G., 1998. Alcohol and Cancer.

In: Galanter, M. (Ed.), The Consequences of Alcoholism, Recent
Developments in Alcoholism, vol. 14. Plenum Press, New York,
pp. 68–95.

Shamberger, R.J., 1972. Increase of peroxidation in carcinogenesis.

J. Natl. Cancer Inst. 48, 1491–1497.

Slezak, B., Santarpia, P., Xu, T., Monsul-Barnes, V., Heu, R.T.,

Stranick, M., Sullivan, R., Petrou, I., Bagley, D., Li, Y., 2002.
Safety profile of a new liquid whitening gel. Compend. Contin.
Educ. Dent. Suppl. 23, 4–11.

Slupphaug, G., Kavli, B., Krokan, H.E., 2003. The interacting

pathways for prevention and repair of oxidative DNA damage.
Mutat. Res. 531, 231–251.

Speit, G., 1986. The relationship between the induction of SCEs and

mutations in Chinese hamster cells. I. Experiments with hydrogen
peroxide and caffeine. Mutat. Res. 174, 21–26.

Speit, G., Vogel, W., Wolf, M., 1982. Characterization of sister

chromatid exchange induction by hydrogen peroxide. Environ.
Mutagen. 4, 135–142.

Stich, H.F., Dunn, B.P., 1986. Relationship between cellular levels of

beta-carotene and sensitivity to genotoxic agents. Int. J. Cancer 38,
713–717.

Stich, H.F., Wei, L., Lam, P., 1978. The need for a mammalian test

system for mutagens: action of some reducing agents. Cancer Lett.
5, 199–204.

Sugerman, P.B., Savage, N.W., 2002. Oral cancer in Australia: 1983–

1996. Aust. Dent. J. 47 (1), 45–56.

Takahashi, M., Hasegawa, R., Furakawa, F., Toyoda, K., Sato, H.,

Hayashi, Y., 1986. Effects of ethanol, potassium metabisulfite,
formaldehyde and hydrogen peroxide on gastric carcinogenesis in
rats after initiation with N-methyl-N0-nitro-N-nitrosoguanidine.
Jpn. J. Cancer Res. 77, 118–124.

Takayama, S., 1981. Report on a Carcinogenicity Study. [Unpublished

Work].

Tipton, D.A., Braxton, S.D., Dabbous, M.K., 1995. Role of saliva and

salivary components as modulators of bleaching agent toxicity to
human gingival fibroblasts in vitro. J. Periodontol. 66, 766–774.

Tsuda, H., 1981. Chromosomal aberrations induced by hydrogen

peroxide in cultured mammalian cells. Jpn. J. Genet. 56, 1–8.

Tucker, J.D., Taylor, R.T., Christensen, M.L., Strout, C.L., Hanna,

M.L., Carrano, A.V., 1989. Cytogenetic response to 1,2-dicarbo-
nyls and hydrogen peroxide in Chinese hamster ovary AUXB1 cells
and human peripheral lymphocytes. Mutat. Res. 224, 269–279.

Weitzman, S.A., Weitberg, A.B., Stossel, T.P., 1986. Effects of

hydrogen peroxide on oral carcinogenesis in hamsters. J. Period-
ontol. 57, 685–688.

Wester, P.W., Kroes, R., 1988. Forestomach carcinogens: pathology

and relevance to man. Toxicol. Pathol. 16, 165–171.

Wilcox, P., Naidoo, A., Wedd, D.J., Gatehouse, D.G., 1990.

Comparison of Salmonella typhimurium TA102 with Escherichia
coli WP2 tester strains. Mutagenesis 5, 285–291.

Williams, G.M., Jeffrey, A.M., 2000. Oxidative DNA damage:

endogenous and chemically induced. Regul. Toxicol. Pharmacol.
32, 283–292.

Williams, G.M., Whysner, J., 1996. Epigenetic carcinogens: Evaluation

and risk assessment. Exp. Toxicol. Pathol. 48, 189–195.

Williams, G.M., Iatropoulos, M.J., 2001. Principles of testing for

carcinogenic activity. In: Hayes, W. (Ed.), Principles and Methods
of Toxicology, Fourth Ed. Taylor and Francis, pp. 959–1000.

Wood, M.L., Esteve, A., Morningstar, M.L., Kuziemko, G.M.,

Essigmann, J.M., 1992. Genetic effects of oxidative DNA damage:
comparative

mutagenesis

of

7,8-dihydro-8-oxoguanine

and

7,8=dihydro-8-oxoadenine in Escherichia coli. Nucleic Acids Res.
20, 6023–6032.

Wurtzen, G., 1993. Scientific evaluation of the safety factor for the

acceptable daily intake (ADI). Case study: butylated hydroxyani-
sole (BHA). Food Addit. Contam. 10, 307–314.

Ziegler-Skylakakis, K., Andrae, U., 1987. Mutagenicity of hydrogen

peroxide in V79 Chinese hamster cells. Mutat. Res. 192, 65–67.

I.C. Munro et al. / Food and Chemical Toxicology xxx (2005) xxx–xxx

15

ARTICLE IN PRESS