Squamous cell carcinoma and precursor lesions of the oral cavity

19 Pages·2011·0.57 MB·English

Squamous cell carcinoma and precursor lesions of the oral
cavity: epidemiology and aetiology
Author
Johnson, Newell W, Jayasekara, Prasanna, Amarasinghe, AA Hemantha K
Published
2011
Journal Title
Periodontology 2000
DOI
https://doi.org/10.1111/j.1600-0757.2011.00401.x
Copyright Statement
© 2011 John Wiley & Sons A/S. This is the pre-peer reviewed version of the following article:
Squamous cell carcinoma and precursor lesions of the oral cavity: epidemiology and aetiology,
Periodontology 2000, Vol. 57(1), 2011, pp. 19-37, which has been published in final form at
http://dx.doi.org/10.1111/j.1600-0757.2011.00401.x.
Downloaded from
http://hdl.handle.net/10072/45454
Griffith Research Online
https://research-repository.griffith.edu.au
Periodontology2011,Vol.57,2011,1–19
PrintedinSingapore.Allrightsreserved
Squamous cell carcinoma and
precursor lesions of the oral
cavity: epidemiology and
aetiology
N W. J , P J & A. A. H K. A
EWELL OHNSON RASANNA AYASEKARA EMALATHA MARASINGHE
People diagnosed with malignant neoplasms often lous), the data may come from localized atypical
feel afraid and alone. However, cancer is largely an regions. Hospital-based cancer registries naturally
avoidabledisease,andstudieshaveshownthatmore gather biased information, i.e. only those cases that
than two-thirds of cancer deaths may be prevented present to hospital; thus, in many developing coun-
through lifestyle modification, early detection and tries, cases may not come to attention at all, either
effective treatment. The World Health Organization because of fear or the inability of poor people to
(WHO) estimates that nearly 12.7 million new cases access hospital services. Data may be even more
and7.6 millioncancerdeathsoccurredworldwidein unreliablebecausefollow-up,evenoftreatedcases,is
2008 (27). impossible in many developing countries. Death
Oral cancers are malignant neoplasms that affect certification is not always compulsory, and there is
thestructuresortissuesofthemouth.Theymaybea limited international standardization with regard to
primary lesion that originated in the mouth, a the categories for cause of death, nor consistency
metastasis from a distant site, or an extension from among of those signing death certificates.
anadjoiningsite.Inquotingdetailedepidemiological Figure 1 shows the estimated incidence and mor-
data, it is necessary to distinguish between these talityratesofthemostcommoncancersbyanatomic
origins, and to ascertain which structures are inclu- site in males and females. For both sexes combined,
ded, e.g. the vermillion borders of the lips are cancerofthemouthandpharynx[ICD-10(C01–C14)]
sometimes included and sometimes not. Globally, is placedninth overall, behind breast, prostate, lung,
(cid:2)oral cancer(cid:3) is the eighth most common cause of colon ⁄ rectum, cervix uteri, stomach, liver and cor-
cancer-related deaths, although many people are pus uteri, in that order. In the published data for
unaware of its existence (91). Of these oral cancers, 2008,oral andpharyngeal sites were theeighthmost
morethan90%aresquamouscellcarcinomasarising commonneoplasmswithintheEuropeanUnion(28).
in the mucous membranes of the mouth and oro- However, the mouth and pharynx are the sixth most
pharynx. common site for malignant disease among men in
developing countries and the eighth most common
Epidemiology of oral cancer site among women. According to the GLOBOCAN 2008
database produced by the International Agency for
Research on Cancer, there were some 399,546 new
Incidence rates worldwide
cases worldwide of oral cancer (C00–C08) and other
Cancer registries play a vital role in monitoring the pharyneal cancers (C09–C10, C12–C14) (27). Esti-
incidenceofcancers.However,thequalityofthedata mated age-standardized rates for the incidence and
available in such registries is far from ideal. Fur- mortalityofcancerofthelipandmouth(C00–08)and
thermore,manypartsoftheworldproducenodataat the pharynx (C09–C10, C12–C14), excluding the sali-
all, whereas in others (often among the most popu- vary gland, 2008, by sex and geographic area, are
1
Johnsonetal.
(a) Male (b) Female
Lung Breast
Prostate Cervix uteri
Colorectum Colorectum
Stomach Lung
Liver Stomach
Oesophagus Corpus uteri
Bladder Ovary
Non-Hodgkin’s lymphoma Liver
Leukaemia Thyroid
Lip, oral cavity Leukaemia
Kidney Non-Hodgkin’s lymphoma
Pancreas Oesophagus
Larynx Pancreas
Brain, nervous system Brain, nervous system
Incidence Incidence
Other pharynx Mortality Kidney Mortality
0 10 20 30 40 0 10 20 30 40
ASR (W) rate per 100 000 ASR (W) rate per 100 000
Fig.1. Estimated global incidence and mortality rates (standardized to a world population age distribution) of the 14
mostcommoncancersinmen(left)andthe15mostcommoncancersinwomen(right).Modifiedfrom(27).Ratesare
standardisedtotheagestructureofthetotalworldpopulation(ASR(W))
Incidence Mortality
Male Female
29.1 12.4
Melanesia 16.6 6.2
16.7 7.3
South-Central Asia 12.7 5.1
13.3 4.1
Western Europe 4.9 1.1
14.3 2.4
Central and Eastern Europe 8.8 1.1
10.8 4.5
Australia/New Zealand
2.8 0.9
10.9 3.7
Northern America 2.2 0.7
11.3 3.1
More developed regions 4.5 0.9
11.3 3
Southern Europe 4.4 0.9
9.5 3.5
Caribbean 4.5 1.3
8.3 3.7
Northern Europe 3 1
8.7 3.3
World 5 1.7
7.7 3.4
Less developed regions 5.1 2.1
8.4 2.5
Southern Africa 5.4 1.5
8.2 2.4
South America 4.5 1.1
9.3 0.9
Polynesia 4.2 0.9
6.2 3
Eastern Africa 3.7 1.7
South-Eastern Asia 5.6 3.5
3.4 2
4.2 2.2
Central America 1.7 0.8
Western Asia 3.72 12.1.1 Fig.2. Estimated age-standardized
3.2 2.4 ratesfortheincidenceandmortality
Northern Africa 1.9 1.4 of cancers of the lip and mouth
3 2.4
Western Africa 1.7 1.3 (C00–C08) and pharynx (C09–C10,
Middle Africa 3.22 11.8 C12–C14), excluding the salivary
3.2 1.1 gland, by sex and WHO geographic
Eastern Asia 1.6 0.5
2.4 1.3 area.ValuesareASR(W)per100000
Micronesia 1.1 population (all age groups). Modi-
40 30 20 10 0 10 20 30 40 fiedfrom(27).
shown in Fig. 2 (27). The rates range from nations)toahighof29.1 ⁄ 100,000perannum(males
< 1 ⁄ 100,000perannum(forfemalesubjectsinparts in Melanesia). These rates illustrate the marked
of Africa, Eastern Asia and some Pacific Island global variability in oral cancer incidence, (136) and
2
Oralcancerepidemiology⁄aetiology
<2.8 <4.2 <7.6 <11.8 <38.7
Fig.3. Estimated age-standardized
oral and pharyngeal cancer inci-
dence rates per 100000 population
throughout the world, showing
marked geographical variation in
incidence. Care is needed in inter-
pretation because some large
groupings disguise marked regional
variation. The top part of the figure
shows the data for males; the lower
part shows the data for females.
<1.5 <2.1 <2.6 <3.7 <16.3 Modifiedfrom(27).
are illustrated in Fig. 3 (27). The highest crude rates and these trends are likely to continue as the Indian
in the world are found in Melanesia, Hungary, economy grows and habits change from betel quid
France, Sri Lanka and Croatia (27).There are marked use to use of other forms of tobacco, especially
differences between countries in the same geo- smoking. On the other hand, there is growing con-
graphical region (77, 78). cern that supply of commercial areca nut and
The extremely high rates in the relatively small tobacco products for chewing will contribute to
populationofMelanesiaarethoughttorelatetoareca futureincreasesintheincidenceoforalsub-mucous
nut (betel) chewing and smoking habits. It has been fibrosis and subsequent oral cancer (38).
apparent for decades that there is a very high inci- DatainJapanshowadramaticincreaseinoraland
dence of oral cancer in southern Asia and of oral pharyngeal cancer incidence (ICD-10 C01–C14) for
cavityplusnasopharyngealcancerinSouth-EastAsia. both sexes: a 4.4-fold increase for males and 3.8-fold
InIndia,Bangladesh,PakistanandSriLanka,theoral increaseforfemaleswererecordedbetween1965and
cavity is the most common site for cancer, and 1999intheOsakaCancerRegistry(cid:3)sdatabase(54).The
accounts for approximately one-third of all cancers estimates oforal cancerfortheyear2008in theUSA
(104). Even within the sub-continent, there are indicate that there were 35,310 new cases and 7,590
striking differences in incidence rates. More than deaths(54).Thisrepresents1.34%ofalldeathsdueto
180,000casesoforalcanceroccureveryyearinSouth cancerinthatcountry.IntheUSA,Theincidencerates
and South-East Asia, with poor prospect of survival: forcancersoftheoralcavityandpharynxwerestable
approximately 90% of these cases are attributable to ordecreasedformenandwomeninmostagegroups
smoking and chewing habits (142). However, overall during the period 1973–2003. Unfortunately, there is
rates in India are showing a decreasing trend in an upward trend for both males and females in
successive birth cohorts: decreasing trends were Australia and among the non-Maori population in
observed for mouth (ICD-10 C03–C06) and tongue NewZealand.Lipcancerinfair-skinnedpopulations,
(C01–C02) cancers among females and for tongue largelyduetoultravioletlightandespeciallynearthe
cancers among males between 1982 and 2000 (107), equator, is a growing problem (76). In Europe,
3
Johnsonetal.
Hungaryhasthehighestincidenceandmortalityoforal oralcancerinyoungerpatientswithoralcancer(43).
and pharyngeal cancer for both sexes (28). Between Anincreaseinincidenceoforopharyngealcancersin
1984and1994,theHungarianmortalityratesfororal the USA, especially of the tongue base and tonsils,
cancers increased by 84% and 72% in males and has occurred in individuals aged 40–45 years: these
females, respectively. The mortality rates among tumoursarefrequentlypositiveforhigh-risktypesof
Italian and French males peaked in the 1980s and human papillomavirus, especially type 16 (70).
decreased after 1990 (126). However, continuing
upward trends have been recorded for Belgium,
Sex distribution
Denmark,Greece,PortugalandScotland(11).
Worldwide, the incidence of head and neck cancers
is higher for males than females. According to the
Age distribution
InternationalAgencyforResearchonCancer(27),the
Oral cancer is usually a disease that occurs in males mean global age-specific incidence of mouth and
after the 5th decade of life. The mean age at pre- pharyngeal cancer excluding the nasopharynx for
sentation is in the fifth and early sixth decades in malesin2008was8.7 ⁄ 100,000 population,compared
Asian populations, compared with the seventh and to3.3forfemales.Thismaybebecauseofthegreater
eighth decades in the North American population indulgencebymeninthemostimportantriskfactors,
(14,23,53,61,62,118).StatisticsintheUSAfor2003– such as heavy alcohol and tobacco consumption for
2007 show that the median age at diagnosis for can- intra-oral cancer and sunlight for lip cancer in those
cer of the oral cavity and pharynx was 62 years (2). whoworkoutdoors.However,oralcancerinfemalesis
Studies have suggested that 4–6% of oral cancers increasinginsomepartsoftheworld.Forexample,a
now occur at ages younger than 40 years (64). An meta-analysis from Buenos Aires showed the
alarmingincreaseinincidenceoforalcancersamong male ⁄ female ratio to be 1.24:1 for the period 1992–
younger people has been reported from many parts 2000comparedto7.1:1fortheperiod1950–1970,due
of theworld (5,69, 108,138), a trend thatappears to toanalarmingincreaseinthenumberoffemalecases
be continuing. There was a significant increase in (12). The incidence of tongue and other intra-oral
cancersofthetongueandtonsilsamong20–40-year- cancersforwomenisgreaterthanorequaltothatfor
olds in the USA between 1973 and 2001 (116). In meninhigh-incidenceareassuchassomeregionsof
Germany, Czechoslovakia and Hungary, there has India, where betel quid ⁄ areca nut chewing (and
been an almost 10-fold increase within one genera- sometimessmoking)arecommonamongwomen.
tioninmortalityfromoralcancerinmenaged35–44 Within Europe, the incidence of oral cavity and
(113). Robinson & Macfarlane showed a dramatic pharyngeal cancers (C00–C14 in the most recent
increase in incidence for younger males in Scotland period, i.e. within the GLOBOCAN 2008 database) is
from the 1980s to the 1990s (103), and the increase highest among males in Hungary (46.1 ⁄ 100,000 per
hascontinuedinyoungadultsinmuchoftheUK.UK annum), and lowest among females in Cyprus and
cancer registries currently show that 6% of all oral Greece (1.5 and 2.0, respectively). Between 1990 and
cancers occur in people under the age of 45 years 1999, the UK incidence rates for oral cancers
(138). In the high-prevalence areas of the world, increasedinmalesofallagesfrom6.5to8.3 ⁄ 100,000
many patients are < 40 years old, partly owing to (an increase of 18%) and in females from 2.6 to
heavy abuse of various forms of tobacco, although 3.6 ⁄ 100,000 (an increase of 30%) (16). The male to
this is not the case in parts of India (25). female ratio in UK in 2008 was 2.1:1 (28).
AstudyperformedinsouthernEnglandconcluded IntheUSA,thedeathrateduetocanceroftheoral
thatasubstantialproportionofcasesoforalcancerin cavity and pharynx per 100,000 population in 2005
youngerpeopleoccurredintheabsenceof,oraftera was3.9formalesand1.4forfemales,downfrom6.9
relatively short duration of exposure to, traditional and 2.3, respectively, in 1975. This substantial
riskfactors(65).Thissuggeststhatfactorsotherthan improvement is not reflected in most of the rest of
tobacco and alcohol are implicated in the develop- the world (2). Apart from the typical risk factors, it
mentoforalcancerinasignificantminorityofcases. has been suggested that oestrogen deficiency may
Diets poor in fresh fruits and vegetables were iden- influence susceptibility to oral cancer in women: a
tifiedasconferringsignificantriskinthispopulation. significantly younger mean age at menopause and
It has also been suggested that greater attention higher rates of hysterectomy may influence the
should be paid to family history and a possible higher rates of oral cancer seen among younger
hereditary predisposition in younger patients with females in the West (122).
4
Oralcancerepidemiology⁄aetiology
havepooreroverallanddisease-specificsurvivalthan
Ethnic variations
whites, mainly because of their comparatively poor
Variations by ethnicity may be due to social and access to health care (36). This is especially con-
cultural practices, as well as the influence of dietary cerning because the incidence of oral plus pharyn-
and genetic factors. Variations in outcome are also gealcancerforblackmenintheUSAissohigh,andit
contributedtobydifferencesinaccesstohealthcare. is the sixth most common site for malignant disease
Where cultural practices represent risk factors, their amongst this group (121).
continuation by immigrants from high-incidence Theage-adjustedincidenceoforalandpharyngeal
regions to other parts of the world results in com- cancers is higher for South Asians than non-South
parativelyhighratesinimmigrantcommunities.This Asians resident in England, particularly among
can also influence the sub-sites most commonly females (74). Interestingly, south Asian males have
affected (63). The highest age-adjusted oral cancer significantly better survival than their non-south
rates in USA are found among non-Hispanic black Asian peers in the south-east of England, possibly a
men (4.86 ⁄ 100,000) followed by non-Hispanic black reflection of the slower progress of tobacco ⁄ areca
women (4.71 ⁄ 100,000), then Asian and Hispanic nut-induced lesions (74).
populations,withthelowestincidenceratesinwhite
(Caucasian) ethnic groups. Tongue cancer was the
Mortality rates and trends
commonest type of oral cancer among every ethnic-
ity. Asians were more likely to develop malignancies Global differences in mortality from oral (C00–C08)
in the buccal mucosa, a reflection of continuing and other pharyneal cancers (C09–C10, C12–C14) in
areca- and tobacco-chewing habits. Another study 2008 are illustrated in Fig. 4. According to GLOBOCAN
showed that American Indians and Alaska Natives 2008,thehighestmortalityfromoralcancerisfound
had significantly lower incidence rates overall than inMelanesia(16.6 ⁄ 100 000inmenand6.2 ⁄ 100 000
non-Hispanic whites (102). Several studies from the in women) (27). There was a steady increase in oral
USAhaveshownthatblackpatientswithoralcancer cancer mortality in men from the 1950s to the late
<1.6 <2.4 <3.5 <5.7 <22.6
Fig.4. Estimated age-standardized
oral cancer and pharyngeal mortal-
ityratesper100000population.The
toppartofthefigureshowsthedata
for males; the lower part shows the
<0.7 <0.9 <1.2 <1.7 <9.3 dataforfemales.Modifiedfrom(27).
5
Johnsonetal.
1980s in most Western European countries (57), but use among young adult men and women, especially
this trend has since declined, e.g. in France, which in the form of pan parag ⁄ gutka-type products, and
had very high rates in the past. In most countries in cigarette smoking is increasing. Overall, therefore,
central and Eastern Europe, for example Hungary, upper aero-digestive tract cancers maynotdecrease.
Slovakia, Slovenia and the Russian Federation, oral Population-based survival rates around the world
cancer mortality in men reached exceedingly high show little evidence of improvement over recent
ratesuptothe1990s.Hungary,Ukraine,Estoniaand decades, despite vast improvements in the field of
Bulgaria showed more than 100% increases in medicine. Cure rates and survival rates have
mortality rates for men during the 20-year period improved with advances in surgical and other tech-
from 1980–2000. There are now the beginnings of a niques in highly specialized, high-volume treatment
reversal. institutions. Regrettably, such highly expert man-
Even though the rates of oral cancer are compar- agementisnotyetuniformlyavailable, anditwill be
atively low among women, there is a steady increase manymoredecadesbeforetheseresultsarereflected
in some countries in Europe (notably Hungary, in population trends.
Belgium, Denmark and Slovakia). Hungary also
showeda98%increaseinmortalityratesforwomen
Oral potentially malignant
(Available at http://www-dep.iarc.fr/WHOdb/WHOdb.
disorders
htm).Thesedisturbingtrendsarethoughttorelateto
highlevelsofdrinkingandsmokinginthesesocieties,
together with poor diet in lower socio-economic The term (cid:2)oral potentially malignant disorders(cid:3)
groups. Trends of age-standardized (world popula- (OPMD) was recommended by an international
tion) mortality rates from oral or pharyngeal cancer working group convened by the WHO Collaborating
withinselectedcountriesduringtheperiodfrom1980 Centre for Oral Cancer and Pre-Cancer in London in
to the present are shown in Fig. 5A,B. 2005 (137). It indicates that not all disorders thus
The Surveillance Epidemiology & End Results described will transform to invasive cancer – at least
(SEER) programme in the USA reported an overall not within the lifespan of the affected individual.
decrease in the mortality from oral and pharyngeal Leukoplakia,erythroplakia,oralsub-mucousfibrosis,
cancer between 1998 and 2007 of 1.6% per annum. lichen planus, palatal lesions in reverse smokers,
Table 1showsadecreaseinallmortalityratesfororal actinic keratosis, discoid lupus erythematosus, dys-
andpharyngealcancerintheUSAbetween1998and keratosiscongenitaandepidermolysisbullosaarefall
2007. The decrease is considerable among both His- into the broad definition of (cid:2)oral potentially malig-
panic and black women (annual percentage changes nant disorders(cid:3) (132, 137).
of )1.6% and )3.7%, respectively). Furthermore, the
SEER data show better five-year survival rates for
Global prevalence
white men (62.2%) and for black men (37.5%) diag-
nosed during the period 1999–2006, than for those Estimates of the global prevalence of oral potentially
diagnosed during the period 1974–1976 (when rates malignant disorders range from 1–5% (85), although
for whites and blacks were 55% and 36.3%, respec- much higher prevalences are reported from South
tively) (79). The five-year survival rates in the SEER EastAsia,usuallywithamalepreponderance,e.g.Sri
registriesrangefromalowof53.3%in1975–1977toa Lanka (11.3%) (3), Taiwan (12.7%) (15), and some
highof62.7%in1999–2006(2).Thesestrikingdiffer- Pacific countries (Papua New Guinea 11.7%) (128).
ences may be explained by a number of factors, Wide geographical variations across countries and
including socio-economic condition, age, stage at regions are mainly due to differences in socio-
diagnosis, continued presence or absence of envi- demographic characteristics, the type and pattern of
ronmentalriskfactors,andaccesstohospitalservices. tobacco use and clinical definitions of disease
African-American patients have consistently poorer (Table 2). In Western countries, the overall preva-
survivaloutcomes(75). lence is low and a decrease over time is observed.
A study in Mumbai, India, (88) indicated a Petti(93)performedameta-analysis of23primary
decreasing trend in oral cancer incidence among studies on oral leukoplakia published between 1986
Indian men, which may be due to a decrease in the and 2002. He calculated the point prevalence to be
useofbetelquid ⁄ panandassociatedoralsmokeless 1.49%(95%confidenceinterval1.42–1.56%)or2.6%
tobacco over this period (123). However, there con- (1.72–2.74%) using a random effects model. Leuko-
tinues to be a high prevalence of smokeless tobacco plakiawassignificantlymoreprevalentamongmales
6
Oralcancerepidemiology⁄aetiology
24
22
20
18
16 Australia
China: selected rural areas
China: selected urban areas
014
0 Finland
0
0 France
0
er 112 Hungary
p Russian Federation
Rate 10 SUoKu, tEhn Agflraicnad and Wales
UK, Northern Ireland
8 UK, Scotland
United States of America
6
4
2
0
1980 1990 2000
Year
4
3.5
3
Australia
China: selected rural areas
2.5
China: selected urban areas
0
0 Finland
0
0 France
0
er 1 2 Hungary
p Russian Federation
ate South Africa
R UK, England and Wales
1.5 UK, Northern Ireland
UK, Scotland Fig.5. Trends in age-standardized
United States of America (world standard population)
mortality rates from oral plus
1
pharyngeal cancer per 100000
per annum for selected countries,
from 1980 to present. Top, males;
0.5
bottom, females (59). Although
only approximately a tenth of the
male rate, the rate in Hungarian
0 females remains a challenge.
1980 1990 2000 (Availableathttp://www-dep.iarc.fr/
Year WHOdb/WHOdb.htm).
7
Johnsonetal.
Table1. Mortality rates and trends by race⁄ethnicity and sex for oral and pharyngeal cancer in the USA between
1975 and2007 (2)
Race⁄ethnicity Rate2003–2007⁄100,000 Trend1998–2007(cid:4)
AAPC(%)
Total Males Females Total Males Females
Allraces 2.5 3.9 1.4 )1.6* )1.6* )2.3*
WhiteHispanic 1.6 2.5 0.8 )3.0* )3.8* )1.6*
White 2.5 3.8 1.5 )1.1* )0.9* )1.9*
non-Hispanic
Black 3.6 6.3 1.5 )3.7* )3.7* )3.7*
Asian⁄Pacific 2.1 3.1 1.2 )2.6* )2.8* )1.8*
Islander
American 1.9 2.9 1.2 )1.7* )2.4*
Indian⁄Alaska
Native
TheSEERprogrammecoversregistriesin17areas[SanFrancisco(SF),Connecticut,Detroit,Hawaii,Iowa,NewMexico,Seattle,Utah,Atlanta,SanJose-Monterey
(SJM),LosAngeles(LA),theAlaskaNativeRegistry,ruralGeorgia,CaliforniaexcludingSF⁄SJM⁄LA,Kentucky,LouisianaandNewJersey).Californiaexcluding
SF⁄SJM⁄LA,Kentucky,LouisianaandNewJerseycontributedcasesfordiagnosisyears2000–2006.Theremaining13SEERareascontributedcasesfortheentire
period(1999–2006).Resultsarebasedonfollow-upofpatientsinto2007.
AAPC,averageannualpercentagechange.
*TheAAPCissignificantlydifferentfromzero(P<0.05).
(cid:4)The1998–2007mortalityAAPCsarebasedonaJoinpointanalysisusingyearsofdeath1992–2007.
(prevalence ratio 3.22). Using these data, he calcu- 17.5%(119).Notsurprisingly,ratesforhospital-based
latedthatthecrudeannualoralcancerincidencerate studies are consistently higher than for community-
attributable to leukoplakia is between 6.2 and basedstudiesbecauseofsamplingbias.
29.1 ⁄ 100,000, suggesting that the global number of Petti (93) estimated a mean global prevalence of
oral cancer cases may be under-reported. 2.6% for leukoplakia, and a mean global transfor-
mation rate of 1.36% per year (95% confidence
interval 0.69–2.03). These prevalence and transfor-
Age and gender distribution
mation figures suggest an actual incidence of oral
The age and gender distribution of oral potentially squamous cell carcinoma higher than that currently
malignant disorders varies considerably, mainly reported, possibly due to under-reporting ofcases of
dependent on lifestyle and on geographical location oral cancer in the developing world.
and ethnicity. In the developed world, leukoplakia
is usually found between the fourth and seventh
Aetiology of oral cancer and oral
decades of life, but in the developing world it is
potentially malignant disorders
found some 5–10 years earlier (86). Females are less
commonly affected, largely reflecting greater use of
relevant habits in men. The majority of oral squamous cell carcinomas are
related to tobacco in various forms, areca nut ⁄ betel
quid chewing, heavy alcohol drinking and dietary
Malignant transformation
micronutrient deficiency. In the developing world,
Theriskofmalignanttransformationvariesfromsite use of tobacco and areca nut, either alone or in
to site within the mouth, from population to popu- combination, accounts for the vast majority of oral
lation,andfromstudytostudy(95,100,119).Aclassic cancers and oral potentially malignant disorders (3).
studyofover30,000Indianvillagers,performedinthe The WHO has classified areca nut, a common com-
1970s with follow-up over 7 years, showed transfor- ponent of many different chewing habits, as carcin-
mation rates between 10 and 24 ⁄ 100,000 per year ogenictohumans(51).UVradiationisrelevanttolip
(95). Another classic study from the early 1980s, a cancer, and there is now unequivocal evidence for a
hospital-based study in Californian patients with role of (cid:2)high-risk(cid:3) genotypes of the human papilloma
oral leukoplakia, with a mean follow-up period of virus family, especially for tonsillar and other
7.2 years,revealedamalignanttransformationrateof oro-pharyngeal sites.
8
Oralcancerepidemiology⁄aetiology
Table2. Summaryofthe reportedprevalence oforal potentially malignantdisorders (OPMD)
References Country(year) Sampling Female⁄male Agegroup Diseaseentity Definition Prevalence
method ratio (years) used (%)
Thomasetal. PapuaNew Random 0.8⁄1.0 ‡18 Leukoplakia Axell1984 11.7
(128) Guinea(2008)
Amarasinghe SriLanka MSSC 0.6⁄1.0 ‡30 OPMDasde- WHO1994 11.3(weighted
etal.(3) (2008) finedabove forgenderand
geographical
location)
Chungetal. Taiwan(2005) Random 0.9⁄1.0 ‡15 OPMD Notgiven 12.7
(15) Leukoplakia 7.4
Erythroplakia 1.9
Lichenplanus 2.9
OSF 1.6
Scheifele USA(2003) MSSC 0.9⁄1.0 ‡20 Leukoplakia Kramer1978 0.5–0.3
etal.(109) Kramer1980
Ministryof SriLanka MSSC – 35–44and65– OPMD WHO1994 4.1
Health of Sri (2003) 74 Leukoplakia 2.6
Lanka(73) anderythopla- 0.4
kia
OSF
Garcia-Polo Spain(2002) Stratified,ran- 0.8⁄1.0 ‡30 Leukoplakia WHO1978 1.6
Vallejo et al. dom Axelletal.1984
(34)
Reichartetal. Germany Stratified,ran- 1.0⁄1.0 35–44and Leukoplakia Axell1976 1.6
(101) (2000) dom 0.7⁄1.0 65–74 Leukoplakia Zain1995 1.0
WHO-ICD-DA
Nagaoetal. Japan(2000) Allinvited 0.4⁄1.0 Male>40, Leukoplakia WHO1980 0.19
(80) female>20 Lichenplanus 0.21
Zainetal. Malaysia Stratified,ran- 0.7⁄1.0 ‡25 Leukoplakia WHO1978 0.96
(147) (1997) dom Erythroplakia Axelletal.1984 0.01
OSF 0.06
Lichenplanus 0.38
Schepman TheNether- Waitingroom 0.9⁄1.0 13–93 Leukoplakia Axelletal.1984 0.6
etal.(110) lands(1996) Axell1996
Schepman
1995
Banoczy Hungary(1991) Random 0.7⁄1.0 Allagegroups Leukoplakia Axell1984 1.3
etal.(10) Lichenplanus 0.1
Ikedaetal. Japan(1991) Factorywork- 0.5⁄1.0 18–63 Leukoplakia Axell1984 2.5
(48) ers
Axelletal. Sweden Stratified,ran- Notfound ‡15 Lichenplanus Axell1976 1.9
(8) (1987) dom
Axelletal. Sweden(1987)Allinvitedresi- 0.9⁄1.0 ‡15 Leukoplakia Axell1976 3.6
(7) dents
MSSC,multi-stagestratifiedcluster;OSF,oralsubmancousfibrosis.
eners, are added according to local preference (51).
Betel quid
Betel quid with or without tobacco is considered
A betel quid generally contains betel leaf, areca nut carcinogenic to humans, with a risk of developing
and slaked lime, and may contain tobacco. Other oral cancer and oral potentially malignant disorders
substances,particularlyspices,includingcardamom, (51, 114). Several studies have quantified the risk in
saffron, cloves, aniseed, turmeric, mustard or sweet- terms of the odds ratio in different regions and
9