A B Schütz
Dentist, Pharmacist, Master and Doctor in Oral Pathology by Faculties of Dentistry
at the Federal University of Rio de Janeiro (UFRJ) and University of Sao Paulo
(USP), Brazil
Previous
Page
_____________________________________________________
Seventy cases of oral epidermoid carcinoma of the mouth registered at the Oral
Diagnostic and Pathology Department at the Faculty of Dentistry, Federal University
of Rio de Janeiro (UFRJ) were reviewed at their clinics information such as
age, localization, sex, clinic aspect, time of evolution and clinic suspicion
of malignancy of the lesions. The data obtained were submitted to the statistical
test of the Chi-square (X2) for comparison of mean, and Student's test ("t")
for verification of contrast. These results were compared with the results reported
in the English and Portuguese literature consulted. The anatomic regions more
frequently attacked by the Oral Epidermoid Carcinoma, considering the sequence
by the greatest number of cases, were: floor of the mouth (24.3%), tongue (22%)
and inferior alveolar edge (14.3%). The frequency of cases was more elevate
in males than in females. However, the statistic difference between the sex
(1.9:!) was considered not significant (p>0.05). The age of patients in that
was observed the greatest frequency was from 50 to 59 years old; having been
the average rate of age of females (61.6 years old) and males (53.8 years old)
considered statistically significant (p<0.01).The presence of ulceration
occurred at 62.8% lesions, and was more elevate in males than in females (p<0.05).
However, the difference in the time of evolution of the lesions did not show
significant statistic difference between the sexes (p>0.05%). Fifteen (15,8%)
were ulcerated lesions and localized themselves predominantly in the floor of
the mouth (12.8%) and tongue (12.8%). Among the non-ulcerated (37.2%) lesions,
15,8% were reported as eritroleukoplastic, 7,1% eritroplastic and 4,3% leukoplastic.
Tongue (5,7%) and floor of the mouth (5.7%) were their most frequents localization.
The correct clinic suspicion of malignancy occurred in 65.1% lesions, however,
was not considered statistically significant (p>0.05%). These results suggest
that the Oral Cancer in Brazil still presents elevate rates of incidence and
mortality, being, at least in part, caused by reduced number of cases diagnosed
in initial stages - at the public and private services of health - particularly
at the Brazilian medical and dental schools. For solving this problem, together
with the implantation of governmental actions of improvement of the services
of diagnostic and prevention of the Oral Cancer, is also need the introduction
of one specific discipline of study of the cancer in the medical and dental
courses.
Keywords: oral cancer, epidermoid carcinoma, epidemiology, clinical data
_____________________________________________________
Correspondence to A B Schütz
Dentist, Pharmacist, Master and Doctor in Oral Pathology by Faculties of Dentistry
at the Federal University of Rio de Janeiro (UFRJ) and University of Sao Paulo
(USP), Brazil
358 Conde de Porto Alegre, Santa Maria, RS, Brazil, 97100-010.
E-mail:vaschutz@terra.com.br
INTRODUCTION
According to WHO the oral cancer together with the dental caries, periodontal
disease, malocclusion, and labial and palatine gap are the principals problems
connected to dentistry [6]. In Brazil, the statistics data indicate a increase
crescent of this disease, and the example of the developed countries such as
France and the United States, in spite of, in the last years, having been identified
numberless advancements in the medical specialty, the actual incidence and mortality
rates stays elevated [1, 2 ,3, 4, 5, 11, 12, 16].
For example, in the United States, country where was registered annually 28.000
new cases and around of 11.000 deaths [16], actually, is foreseen about 30.000
new cases and 8.000 deaths, indicating that occurred some improvement in the
methods of diagnostic and treatment in this country. Same so, the survival rate
of 5 years is still of only 53% (database-http://www.cdc.gov/). These data contrast,
same considering the numeric difference of the Brazilian population, of data
collected in Brazil, where is estimated only 2.660 new cases and around 660
deaths in 2001 (database-http://www.inca.org.br/).
The gravity of this disease and the large number of cases diagnosed suggest
that the cancer is one problem of public health, principally, in developing
countries such as Brazil, where 12,11% of the total of deaths is caused by cancer.
In 1996, the National Institute of Cancer (INCA,RJ) estimated 2.945 new cases
of oral cancer (1.875 in males and 1.070 in females). Only in the southwest
region is estimated 2.660 new cases (1.980 males and 680 females) in 2001. Of
these, 1.160 new cases (650 at male and 510 at females), and 300 deaths will
occur in the city of Rio de Janeiro (database-http://www.inca.org.br/). Certainly,
these numbers will be more elevated, considering the poor system of registers
and transmissions of data existent in some Brazil's regions.
In this country, is calculated that the mouth is the anatomic site of 90% of
the malign neoplasm, and around of 38% of the tumors localized in the region
of the head and neck, causing approximately from 3-5% of the total number of
deaths caused for cancer, and 99% of the deaths for oral cancer. Is estimated
that the survival rate, of at least five years, is not superior to 30% in the
majority of the cases [3, 4, 5, 18]. This last data, at least in part, may be
explicated by fact that more of 60% of the Brazilian patients present themselves
in the moment of diagnostic with oral cancer at advanced stage (III-22,1% and
IV-42%) (database-http://www.inca.org.br/).
These data suggest that in Brazil, in spite of big part of population not to
have assess at the private and public services of health, various initials cases
of oral cancer pass unnoticed by professionals of health, presenting themselves,
when of the diagnostic, profoundly invasive and with evidence (clinic and sub-clinical)
of metastases in the regional lymphonodus, causing difficulty in the treatment;
therefore, confirming the observations made by the Professor Shaeffer el al.
[13] in the United States behind years. According to the database of INCA, RJ,
Brazil, only 6,5% of the patients are diagnosed with oral cancer in the clinic
stage I.
So far, considering the crescent number of cases of oral cancer occurring in
Brazil at this time, as well as the pressing need of reporting all statistics
data existent about this disease, with objective of stabilizing one adequate
base for implantation of a serious program of public health at worldwide level,
capable of preventing it, or then, identifying it at initial stage; the purpose
of this study was to report a retrospective study of cases of oral epidermoid
carcinoma registered at the Oral Diagnosis and Oral Pathology Department at
the Faculty of Dentistry, Federal University of Rio de Janeiro (UFRJ), Brazil,
studying the following variables: clinics aspects, anatomic localization, age,
time of evolution and the evaluation clinic of malignity of the lesions.
MATERIALS AND METHODS
This work is one retrospective study of the registers of 70 cases of oral epidermoid
carcinoma registered at the Oral Diagnostic and Oral Pathology Department, Faculty
of Dentistry, Federal University of Rio de Janeiro (UFRJ), whose histopathologic
diagnostic by mean of the routine method (WHO -Whai et al. (1971)- classification
and routines cuts) was squammous cell carcinoma of the mouth, classified in
the tree histologic grades (I, II, III) according to the histopathologic parameters
of proliferation and differentiation cellular (stratified sample), selecting
the cases in that the clinic related to the assault of, or preferential for
one anatomic (p. e. lesion localized in tongue, or lesion localized in tongue
extend itself to the floor of the mouth).
Were discarded the cases in that the origin of the lesion was not clear, such
as when the clinic reported the sudden attack for more of one anatomic region
(p. e. lesion localized in tongue and floor of the mouth). The clinics information
(parametric and non parametric variables) referent to: age, sex, race, anatomic
localization, time of evolution, clinic aspect and evaluation of malignity of
the lesions were submitted at the parametric test Student's test ("T")
for comparison of arithmetic means (S12/S22=1.13; F=1.53), and the non parametric
test of x2 (chi-square) for verification of contrast. The distribution at the
level of 1% (p<0.01) and 5% (p<0.05) were considered significant. Above
of these valor the statistics differences were considered not significant (p>0.05).
RESULTS
Sex Distribution and Male/female ratio
Of 70 cases studied, 46 patients (65.7%) were males and 24 (34.3%) woman. The
male/female ratio was 1.9:1 (fig.2).
Racial and Age Distribution of the
Lesions
Fifty patients (71.5%) were Caucasians, 12 (17.1%) Black and 8 (11.4%) Grays.
The most young patient was 17 years old (male) and the most old 91 years old
(female). Fifth, sixth and seventh decades of life were the more assaulted.
The average age of the patients studied was 61.6 years for the women and 53.8
years for the men, having been this statistic difference (p< 0.01) considered
significant (Student test).
Anatomic Localization of the Lesions
The greatest number of cases 17 (24.3%) occurred in the floor of the mouth,
followed by the tongue 15 (21.4%) and inferior alveolar edge 10 (14.3%). The
remaining cases were distributed in other anatomic regions of the mouth (Superior
Alveolar edge, 3%; Inferior Gingival, 6%; Superior Gingival, 4%; Palate, 4%;
Lip, 4%; Cheek, 11%, Retromolar Region, 7%). These data are presented in the
fig. 1.
Anatomic Localization and sex of
the patients
The distribution of the lesions according to anatomic localization and sex reveled
that in males the lesions occurred more frequently in the tongue, floor of the
Mouth, palate and retromolar Region, while in female they most frequent in the
inferior Alveolar Edge and gingival. In the masculine sex, the greatest number
of cases occurred in the floor of the mouth 15 (21,4%),while in the feminine,
the inferior alveolar edge with 6 cases (8,6%) was the localization more frequent
(fig. 2).
Anatomic Localization and Aspect
Clinic of the Lesions
The distribution of the lesions according to the clinic aspect and anatomic
localization is presented in the fig. 3. Forty-four lesions (62.8%) were reported
as ulcerate and 19 (27.2%) not ulcerate. In 23 lesions (32.9%) was observed
the component leukoplastic and in 27 (38.6%) eritroplastic component. Excepting
the gingiva and mandibular edge, in the besides anatomic regions the ulcerated
aspect was more frequent than the ulcerated. Among the lesions ulcerates aspect,
the greatest frequencies were verified in the floor of the mouth, 11 cases (15.8%)
and tongue, 9 cases (12.8%). Among the non-ulcerated lesions, the aspect predominant
was the eritroleukoplastic with 11 cases (15.8%). Of the 8 cases reaming, tree
lesions (4.3%) were reported as leukoplastic and 5 (7.1%) as eritroplastic.
Floor of the mouth and tongue, with 4 cases (5.7%) each were the anatomic regions
in that occurred the greatest number of cases with the non- ulcerate aspect.
Aspect Clinic of the lesions and
Sex of the Patients
The distribution of lesions according to the sex and clinics aspects of the
lesions demonstrated that for the masculine sex, at 31 cases (44.2%) was observed
the ulcerate aspect, while In the feminine, 13 cases (18.6%) presented this
aspect clinic. Objectifying to analyze the possible association between the
sex of patients and the presence of ulceration in the lesions, the difference
between the sexes in relation the presence of ulceration was submitted at the
statistic test of the chi-square (X2), having bee considered significant at
the level of 5% (p<0.05).
Time of Evolution of the Lesions
and sex of the patients
The analysis of the time of evolution of the lesions revealed that the patients
reported evolution inferior to 12 months at 34 cases (48,5%), and at 18 lesions
(25,7%), evolution of 12 months or more. This difference in relation at the
sexes of the patients was submitted at the chi-square (x2) having been considered
not significant statistically (p>0.05).
Clinic Suspicion of Malignancy of
the lesions
In relation the suspect clinic of malignity of the lesions, 53 cases (75.71%)
possessed this information. In the majority (65.7% lesions), the clinic suspected
of malignity, while in 10% of the cases this suspect did not occur. Applied
the chi-square test (X2), the clinic evaluation of malignity of the lesions,
according to the sex of the patients, was considered not significant (p>0.05).
DISCUSSION
The etiology of oral epidermoide carcinoma is connected to the abusive use of
tobacco and alcohol, having been in various studies demonstrated the effect
synergetic of these agents [7, 15, 19, 20, 21], because the tobacco might act
as initiator, while the alcohol would actuate as promotor of the oral cancer
[11, 12]. Is also well-known that the carcinogens resultants of combustion of
the tobacco not only cause mutations, deletions, substitutions and amplifications
in proto-oncogenes, but also promote the inactivation, or molecular alterations
in onco-supressor (p53) gene, which is one gene activator of the appoptosis
of the cancers cells. Moreover, is also possible that these agents alter the
expression of other molecules regulatory of the cell cycle such as the cyclin
and cyclin-dependent kinases [8]. However, the effects resultant of the abusive
use of tobacco and alcohol, in the genetic expression of the p53 is still controversy,
because while some articles has reported that this abusive use may cause mutations
in p53, others have reported absence of the abusive use of these agents in 33%
of the patients with oral cancer, who presented alterations in the expression
of the p53 [9].
At 24 (34..28%) cases in which was reported the participation of the possible
etiologies of the oral cancer, of these, at 14 cases (58,3%) was identified
the habit of smoking tobacco in its different forms, suggesting to be this agent
the principal etiologic factor of the oral cancer. Among 10 patients (48.3%)
in that was reported the presence concomitant of cancer and ill adapted prosthesis,
2 patients (8,6%) reported to use tobacco. This result suggests that the transformation
carcinomatous caused for adapted-ill prosthesis in smokers patients can not
be rare as the literature indicates. However, we emphasize that for confirmation
of this result is necessary studies more accurate utilizing greater samples
necessaries for a more accurate statistic analyze. In spite of this result revealed
some relation between trauma and oral cancer, we think not to be the trauma
caused by the adapted-ill prosthesis an initiator agent of oral cancer, but
for having the effect irritant chronic might, maybe, contribute for the transformation
carcinomatous in cells previously altered genetically by use abusive of tobacco
and alcohol.
The majority of the reports of literature prove that the greatest frequency
of these lesions in the masculine sex. Meanwhile, in this studies the male/female
ratio changed according to the geographic region in that were realized [8, 2].
Recent works shown that this ratio, in the last decades, is decreasing in various
regions [12]. The results of our study confirmed this decrease. For explicate
its cause among the Brazilian population, has been suggested the increase of
the incidence of oral cancer in the feminine sex, possibly caused by the increase
of the habit of smoking tobacco among, observed in Brazil in the last decades
confirming this tendency worldwide. The decrease of this ratio, is confirmed
by the data of the INCA-Brazil that estimated 1,160 new cases for the city of
Rio de Janeiro in 2001 Of these, 650 (56%) will occur in male and 510 (44%)
at females (1.27:1), according to its database (http://www.inca.org.br/).
Data of the English literature revealed that though the oral cancer can occur
in any decades of life, it is more frequent in adults and oldest (after 40 years
old), whose the explication may be the cumulative exposition to various carcinogenic
stimulus during the lifetime. Results of the present study support these data,
because the 5th., 6th. and 7th. decades of life presented sudden attack, totaling
80% of the cases. The average rate of age of these patients was 56,5 years old.
The arithmetic mean of age of the patients was 61,6 years for the females and
53,8 years for the males, having been this difference considered statically
significant (p<0.01) (Chi-square). This age difference, at least in part,
may be explicated by the large number of lesions localized in the floor of the
mouth and tongue, whose reports of the literature aim smaller age for the masculine
sex than feminine, as well as the greatest male/female ratio [5, 18, 19]. Other
possibility is the anterior exposition of the males at to tobacco and/or alcohol,
or to other carcinogenic agents.
According to the majority of the works reported in the English literature, and
at some Brazilian cities such as in Bauru, State of São Paulo, southwest
region of Brazil, the lip was the more frequent localization of the oral cancer.
This report was not confirmed in our study, because the greatest number of cases
occurred in the floor of the mouth (24,3%) and tongue (22,3%). Believed that
the reduced number of lip cancer verified in our study possibly may be explicated
by the fact of the patients with this kind of cancer had been initially attended
in specialized medic centers such as the Institute National of the Cancer (INCA-RJ),
which is one center of reference for the diagnostic and treatment of the cancer
in Brazil. However, the data of this specialized center, which reported the
tongue as the most frequent localization of the Oral Cancer in some Brazilian
cities such as Fortaleza and Porto Alegre, confirming the results of our study,
which aim this region as the second (15 cases) not presenting significant difference
for the floor of the mouth (17 cases), which was the first. If the cases for
us studied localized in the inferior and superior alveolar edge, and in the
inferior and superior gingiva were grouped in only one anatomic region (gingiva
and alveolar edge), the 19 cases (27,1%) for us identified, are particularly
of according to Onofre et al' study [10] - also realized in the Southwest region
of Brazil - which aim this anatomic as the most affected by the oral cancer.
By analyze of the distribution of the lesions accord to the sex and localization
(fig.2) was observed that for all localization, except in the gingiva, cheek
and alveolar edge, occurred greatest frequency in the masculine sex. In these
last anatomic regions (cheek and superior alveolar edge), the frequency more
approximated between the sexes might indicate the greatest resistance of the
epithelial tissue in the males, or the greatest susceptibility in the females,
considering that the same etiologic agent produced in both the sex identical
effect [20]. Considering this think is possible that in these localization the
resistance or the susceptibility of the epithelial tissue might be genetically
determined, manifesting themselves under the hormonal influence; in spite of
studies accomplished in the southwest of the USA, Ceylon and Philippine Islands,
Panama and Venezuela had reported that the elevated incidence of oral cancer
in these regions was caused by habits as to chew tobacco [13, 20]. In the present
work due to a great number of cases with lack of information about the participation
of etiologic factors of the oral cancer, in large number of cases, was not possible
establish any correlation. However, as these habits are not commons in Brazil,
we think that the similar frequency between the sexes, in those anatomic regions,
must not explicate with base in these habits.
The clinic aspect predominant was the ulcerated in 44 cases (62,8%) of 63 (90%)
lesions in that the clinic aspect was described, confirming the reports of the
specialized literature. The general think that aim the leukoplakia and the eritroplakia
as lesions that, at the microscopic level, may exhibit epithelial alterations
varying since one simple alteration epitelial at to invasive carcinoma, was
confirmed in present study since 3 cases (4,1%) were leuckplastic and 5 cases
(7,1%) eritroplastic. Among the lesions related with not ulcerate predominated
the aspect eritroleukoplastic with 11 cases (15,8%). This find confirmed the
reports [13, 14, 15] that aim the increase of the potential of malign transformation
presented by this kind of lesions. The presence of ulceration was more accentuate
in male sex than in females, having been this difference statistically significant
at the level of 5% of probability (p<0.05). In the attempt of explicating
it with base in the possible relation between the time of evolution and the
presence of ulceration, we submitted this difference at to statistic treatment,
which proved to be this difference not significant between the sexes (p>0.05).
In spite of this resulted we accredit that hormonal factors might cause the
evolution more slow of oral cancer in females.
In relation to the clinic suspicion of malignity of the lesions, the reports
quote that around 50% of the initial cases of oral cancer is not identifiable
by the professionals of health. However, of 70 cases for us studied, in 46 (65,7%)
the clinic suspected of malignity. In spite of this, when applied the chi-square
test (x2), this difference was not statistically significant (p>0.05). This
resulted suggest that in the material that constituted the present study there
was one number expressive of lesions in advanced stage that generally shown
the clinic aspect characteristic of the invasive cancer (ulcerate lesions with
neat edge and deep, or base harden), which not difficult the clinic diagnostic,
such as occurred in the present study in 44 cases (62.8%); therefore, confirming
the big number of cases of oral carcinoma in initial stage that pass unperceived
by the majority of the professional of health in the public and private services
of health, as reported in literature. According to INCA, RJ, the percentage
of patients diagnosed in Brazil at advanced stage increased form 57% in 1986
to 64.3% in 1992. At this time, it is around of 60% (database-http://www.inca.org.br/).
Other data that cause preoccupation was a work realized at Santa Maria city
(south region), Brazil, by Federal University of Santa Maria (UFSM) which verified
in one period of 5 years the mensal rate of 3. 5 biopsies, having been diagnosed
in this period only four cases of oral cancer. This results, same considering
the proportion in relation at the number of patients attended in the public
and private services, as well as the number of Faculty of Dentistry in the region
is one clear signal that lack knowledge, or conscience for the dentist that
work in these region, in relation to importance and simplicity of this auxiliary
exam of diagnostic in the detection of the oral cancer at initial stage. This
last datum, when extrapolate for other regions less developed, is one clear
evidence that the Brazilians educational and health systems must suffer a radical
intervention. with objective of its improvement.
Fig. 1. Distribution of oral carcinoma for localization
Fig.2. Distribution of oral carcinoma for sex and localization
Legend: L-Leukoplastic, E-Eritroplastic, EL- Eritroleukoplastic,
U-Ulceration, UL- Ulceration leukoplástic, EU- Ulceration eritroplastic,
UEL- Ulceration eritroleukoplasti
Fig. 3. Distribution of oral carcinoma for clinic aspect and sex
Acknowledgments
This work was realized with financial recourse furnished by CAPES to the Professor Dr. José Carlos BorgesTeles, Oral Diagnostic and Pathology Department at the Federal University of Rio de Janeiro (UFRJ). I am specially grateful to L. N. Guimarães by dedication that realized the wearisome work of typewritten and revision of the originals in Portuguese, which served of base for realization of this article, and at to Prof. Dr. Pedro Carvalho, Service of Statistic of the Institute National of the Cancer (INCA-RJ), by the statistical treatment of the results.
____________________________________________________
1. ACKERMAN, l. V. & Del REGATO, J. Cancer: diagnosis, treatment and prognosis. 6th. ed., St. Louis Mosby, 1985.
2 BOYLE, P. et al. European school of oncology advisory report to the European Commission for Europe Against Cancer Program: Oral carcinogenesis in Europe. Oral Oncol Eur. J. Cancer, 1995, 31 (2): 75-85.
3. Brasil Ministério da Saúde. Instituto Nacional do Câncer. Câncer de Boca. Rio de Janeiro. Instituto Nacional do Câncer, 1992, 53p.
4. Brasil Ministério da Saúde. Instituto Nacional do Câncer. Estimativa da incidência e mortalidade por câncer no Brasil. Rio de Janeiro, 1996.
5. BRUMINI, R. et al. Câncer no Brasil: dados histopatológicos. Rio de Janeiro, Campanha Nacional de Combate ao Câncer. Ministério da Saúde, 1982.
6. GOODGER, N. M. et al. Cell cycle regulatory proteins- an overview with relevance to oral cancer. Oral Oncology, 1997, 33 (2): 61-75.
7. KAISER, O.B. et al. Estudo das lesões buco-maxilo-faciais submetidas à exame anátomo-patológco no período de 1988 a 1993 na cidade de Santa Maria. Revista Saúde (UFSM), 20 (3-4): 31-37, 1994.
8. LODI, G. et al. Attitude towards smoking and oral cancer prevention among Northern Italian Dentists. Oral oncology, 1997, 33 (2): 100-104.
9. MATTHEWS, J. B. et al. Relationship of tobacco/Alcohol use to p53 expression in patients with lingual squamous cell carcinomas. Oral Oncol Eur. J. Cancer, 1993, 298 (4): 285-289.
10.ONOFRE, M. A. et al. Prevalência do câncer bucal no Serviço de Medicina Bucal da Faculdade de Odontologia de Araraquara/UNESP: 1989-1995. R. G. O., 1997, 45 (2): 101-104.
11.SCHÜTZ, A. B. Carcinoma epidermóide
bucal: aspectos microscópicos e comportamento biológico. R. G.
O., 1997, 45 (2): 71-77.
12. SCHÜTZ, A. B. Estudo comparativo da graduação histológica
do carcinoma epidermóide de boca, por meio das técnicas de cortes
de rotina e semi-seriado. Rio de Janeiro. Faculdade de Odontologia da Universidade
Federal do Rio de Janeiro, 1992, Tese de Mestrado, 166p.
13. SHAEFER, R, W. G. et al. Carcinoma oral in situ. Oral Surg., 1975, 39: 227-238.
14. SHKLAR, G. Oral cancer. Philadelphia,
Saunders, 1984.
15. SHKLAR, G. et al. Correlated study of oral cytology and histopathology.
Report on 2.052 oral lesions. Oral Surg., 1980, 23: 61-70.
16. ZHENG, T. et a;. Risk of tongue cancer associated with tobacco smoking and alcohol consumption: a case-control study Oral Oncology, 1997, 33 (2): 82-85
17. TANAKA, T. Chemoprevention of
Oral Carcinogenesis. Oral Oncol, Eur. J. Cancer,31: 3-15, 1995.
18. TORLONI, E.; BRUMINI, R. Registro Nacional de Tumores. Rio de Janeiro, Ministério
da Saúde, Divisão Nacional de Doenças Crônico-Degenerativas,
1978, 142 p.
19. TIECKE, R. W.; BERNIER, J. L. Statistical and morphological analysis of
four hundred and one cases of intraoral squamous cell carcinoma. J. Amer. Dent.
Ass., 1954, 49: 684-698.
20. VOGLER, W. R. et al. The retrospective study of etiologic factors in cancer of the mouth, pharynx, and larynx, Cancer, 1962, 15: 246-58.
21. WYNDER, E. L. et al. The study of the etiologic factor in cancer of the mouth. Cancer., 1957, 10: 1300-1327.