High Association of Human Papillomavirus Infection with Oral Cancer:
A Case-Control Study
Gabriela Anaya-Saavedra,aVelia Ramı ´rez-Amador,aMa. Esther Irigoyen-Camacho,a
Claudia Ma. Garcı ´a-Cuellar,bMiriam Guido-Jime ´nez,cRocı ´o Me ´ndez-Martı ´nez,b
and Alejandro Garcı ´a-Carranca ´b,c
aUniversidad Auto ´noma Metropolitana, Mexico, D.F., Mexico
bInstituto Nacional de Cancerologı ´a, Mexico, D.F., Mexico
cInstituto de Investigaciones Biome ´dicas, Universidad Nacional Auto ´noma de Me ´xico, Mexico, D.F., Mexico
Received for publication June 11, 2007; accepted August 8, 2007 (ARCMED-D-07-00247).
Background. The aim of the present study was to determine the association of high-risk
human papillomavirus (HR-HPV) in Mexican individuals with oral squamous cell carci-
noma (OSCC) and their association with various risk factors.
Methods. We designed a matched case-control study. Cases were individuals with newly
were obtained; also a self-administered questionnaire about sexual behavior was included.
DNA from oral brushing was purified to amplify HPV-DNA through MY09/MY11 and
GP5þ/GP6þ primers and subsequently subjected to sequencing. Conditional regression
models were built to calculate odds ratios (ORs) and 95% confidence intervals (CI).
Results. Sixty two cases and 248 controls (53.2% males), median age 62 years (Q1eQ35
54e72 years) were included. HPV prevalence was 43.5% in cases and 17.3% in controls
(HR-HPV: 37.1% cases, 9.7% controls). The most frequent types in cases were HPV-16
and HPV-18 (55.6 and 18.5%). The presence of HR-HPV was associated with OSCC
(OR 5 6.2; 95% CI: 2.98e12.97) controlling for the most common risk factors. An inter-
action between smoking and drinking was detected, and family history of cancer was also
significant (OR: 3.61; 95% CI 5 1.44e8.99). Early age at first sexual intercourse and large
number of lifetime sexual partners showed an association with HR-HPV (p 5 0.019 and
p 5 0.033, respectively).
Conclusions. Oral HR-HPV was strongly associated with OSCC, suggesting that HPV-16
and -18 are risk factors for oral cancer in Mexican patients. A significant association of
tobacco and alcohol was confirmed. In addition, family history of cancer was associated
with OSCC. The results underline the role of HPV in OSCC and its multifactorial etiol-
ogy. ? 2008 IMSS. Published by Elsevier Inc.
Key Words: Human papillomavirus, Oral cancer, Tobacco, Alcohol, Mexico.
The causal association between consumption of tobacco
and alcohol and the development of oral squamous cell car-
cinoma (OSCC) is well established (1,2); however, a consid-
erable proportion of OSCC occurs in nonsmokers and
nondrinkers, indicating the presence of other risk factors
During the past two decades, it has been shown that hu-
man papillomavirus (HPV) infection is etiologically linked
with a subset of head and neck carcinomas (HNC) (4e8).
The most common high-risk HPV (HR-HPV) types found
in cervical cancer, and HPV-16 and 18, have also been iden-
tified in HNC (4,9).
HPV prevalence reported in oral cancer studies have
ranged from 3.9 to 74% (6,10). These large variations
Published previously online November 9, 2007.
Address reprint requests to: Velia Ramı ´rez-Amador, Camino Sta. Ter-
esa 277-9. Col. Parques del Pedregal, Mexico D.F. 14010, Me ´xico
E-mail: firstname.lastname@example.org or email@example.com
0188-4409/08 $esee front matter. Copyright ? 2008 IMSS. Published by Elsevier Inc.
Archives of Medical Research 39 (2008) 189e197
may be due to biological and methodological factors such
as the anatomic site, sampling collection techniques (frozen
or fixed biopsies, scrapes or brushing of epithelial cells, or
mouthwash) (11), and the use of different HPV detection
assays with variations in sensitivity and specificity (12).
Also, the sample size studied, the source of the cases, and
the presence of environmental risk factors probably influ-
enced the discrepancies found among studies (13).
A worldwide analysis of 5046 HNC specimens from 60
studies of 26 different countries showed that 23.5% of 2642
oral cavity biopsy samples were HPV positive (13). How-
ever, many of the studies included did not consider vari-
ables such as age, gender, smoking, and alcohol drinking,
and only a few included a control group (4,13).
In Mexico, a cross-sectional study carried out at the In-
stituto Nacional de Cancerologı ´a (INCan) showed that 42%
of oral cancer cases were HPV positive; nevertheless, lack
of a control group in this study did not allow the identifica-
tion of HPV prevalence in normal oral mucosa (14).
Considering the large variation of HPV prevalence
among different populations in the world, lack of a control
group in many of the studies, scarce evidence about OSCC
etiology in Mexico and, particularly, the limited informa-
tion on HPV, the aim of the present study was to determine
the association of HR-HPV in Mexican individuals with
oral cancer and their association with various risk factors.
Materials and Methods
We conducted a matched case-control study from Septem-
ber 2003 to July 2006 at the Instituto Nacional de Cancer-
ologı ´a (INCan), Hospital General ‘‘Dr. Manuel Gea
Gonza ´lez’’ and Clı ´nica Estomatolo ´gica ‘‘Rafael Lozano’’
of the Universidad Auto ´noma Metropolitana-Xochimilco,
(UAM-X) in Mexico City. The ethics review board of each
institution approved the study. Each patient who agreed to
participate in the study signed an informed consent. The de-
sign of the study was according to the STROBE statement
recommendations for case-control studies (15).
Cases. Eligible cases were males or females aged 18 years
or older, with newly diagnosed OSCC, histologically con-
firmed and prior to the application of any kind of therapy
(chemotherapy, radiotherapy, or surgery). Topographic lo-
cations included malignancies of the tongue, gingiva, floor
of mouth, palate, buccal mucosa, and retromolar area (ICD-
10 C01-06) and excluded lip cancer and malignant tumors
of major salivary glands, nasopharynx, oropharynx, and lar-
ynx (ICD-10 C00, C09-C14) (16). Sixty two OSCC cases
were selected, 33 were men (median age 60 years, range:
31e83 years) and 29 women (median age 70 years, range:
31e86 years). All patients were staged in agreement with
the AJCC system of classification (17). None of the OSCC
patients refused to participate in the study.
Controls. Four controls per each cancer case age- (?5
years) and gender-matched were included. They were free
of any premalignant or malignant oral lesions and without
any personal history of cancer. Two controls were recruited
from a medical service and two from a dental service. Hos-
pital General ‘‘Manuel Gea Gonza ´lez’’ and Clı ´nica Estoma-
tolo ´gica ‘‘Rafael Lozano’’, two services of the INCan, were
the source of the cases and are governmental health services
for persons not affiliated with the Mexican social security
system. All selected health services were located close to
each other in the south of Mexico City. Exclusion criteria
included mental illness that impaired the adequate consent
process, personal history of any type of cancer, history or
presence of premalignant or malignant oral lesions, a diag-
nosis associated with the exposure of interest, or subjects
who refused to participate in the study (1.6%). Overall,
there were 248 matched controls: 132 men (median age
59 years, range: 30e81 years), and 116 women (median
age 67 years, range: 27e86 years).
Cases and controls were interviewed by the same investi-
gator who used a standardized questionnaire that contained
demographic (age, gender, marital status, education, and
occupation) and clinical data (familial history of cancer,
and tobacco and alcohol consumption). In addition, OSCC
cases were asked about their disease duration and signs and
symptoms of their oral lesion. Data related to shape, size,
color, and stage of the OSCC lesion were also obtained.
According to tobacco use, patients were classified as
never, current, and former smokers. Never smokers were in-
dividuals who reported that they had smoked less than one
cigarette/day throughout their life. Current smokers were
defined as subjects who had smoked at least one ciga-
rette/day from the previous year to the present. Former
smokers were those individuals who had quit smoking at
least 12 months prior to the study interview. Subjects were
asked about the ages when they began smoking, as well as
when they stopped smoking, duration of the habit, and
number of cigarettes smoked per day. We calculated the cu-
mulative amount of tobacco use in pack-years (number of
cigarettes smoked a day/20 ? 365.25 ? years of smoking).
Smokers were classified in three categories according to
their cumulative tobacco consumption: 0, !25, and $25
Alcohol exposure was quantified in terms of drink units
(DU) per week; one DU corresponded to a glass of wine
(125 mL), a bottle of beer (330 mL), or a glass of tequila
or hard liquor (30 mL). Current drinkers were defined as
those subjects who drank at least one DU weekly during
the last year, non-drinkers were classified as subjects who
never consumed alcohol or had less than one drink per
week. Former users included drinkers who stopped alcohol
Anaya-Saavedra et al./ Archives of Medical Research 39 (2008) 189e197
use at least one year before the study entry. Participants
were asked about the age of starting and stopping alcohol
use and the number of DU per week. The cumulative
amount of DU was calculated in DU-years (DU per week ?
52 ? years of drinking). Drinkers were classified in three cat-
egories: according to their cumulative alcohol consumption: 0,
!16, and $16 DU-years.
Enrolled subjects were asked to complete a self-
administered questionnaire that included questions about
lifetime number of sexual partners, age at first intercourse,
oral-genital contact, and number of oral-genital partners.
Six cases (9.7%) and 17 controls (7.3%) refused to respond
to the questionnaire.
Oral Samples. An incisional biopsy was taken from all
OSCC cases. One section was sent to a specialized labora-
tory for histopathological examination and the other kept in
a Falcon tube with 8 mL of PrecervCyt solution (Cytyc
Corp., Marlborough, MA) for HPVidentification. Addition-
ally, oral specimens were collected from OSCC lesions and
matched mucosal site with controls using a conical cyto-
brush. In order to avoid possible false negative results, three
samples were taken with the cytobrush from the lesion in
cases and oral mucosa in controls. Also, cytological sam-
ples were placed into 8 mL of PrecervCyt solution (Cytyc
Corp.) and stored frozen until testing.
For DNA extraction in biopsies, a 4-mm tissue section
was cut, placed in a microtube with nuclear lysis solution
(Wizard Genomic DNA Purification kit; Promega, Madi-
son, WI) and 17.5 mL of proteinase K (20 mg/mL), and in-
cubated at 65?C for 3 h. Subsequently, all samples (biopsy
specimens in the cases and cytobrush smears in cases and
controls) were centrifuged at 2000 ? g and the supernatant
was removed. DNAwas purified using the Wizard Genomic
DNA Purification kit (Promega) according to the manufac-
turer’s instructions and quantified by spectrophotometry.
HPV DNA Detection and Typing
Three types of controls were included in each reaction se-
ries, water (negative control), DNA of a HPV-18-positive
HeLa cell line and HPV-16-positive CasKi cell line (posi-
tive controls). All control samples were prepared and ana-
lyzed in parallel with the clinical specimens to ensure
that proper reaction conditions were maintained. Standard
precautions for PCR steps (sample preparation, reaction
mix preparation, amplification, and electrophoresis) were
done in separate rooms.
The presence and quality of human genomic DNA was
verified in all clinical samples by amplification of a 268-
bp fragment of the b-globin gene by using the primers
PCO4 (50-CAACTTCATCCACGTTCACC-30) and GH2O
(50-GAAGAGCCAAGGACAGGTAC-30) (18). For detec-
tion of HPV, we used the general degenerate consensus
primers MY09 (CGT CCM ARR GGA WAC TGA TC),
and MY11 (GCM CAG GGW CAT AAY AAT GG; the
degenerate base code was M 5 A or C, W 5 A or T, Y 5
C or T, and R 5 A or G), which spans nucleotides
6722e7170 in HPV type 6 and the corresponding regions
of the other genital HPVs (19). Also, oligonucleotides
GP5þ/GP6þ that extend nucleotides 6764e6902 in HPV
type 6 and the corresponding regions of other genital HPVs
were used (20). PCR conditions included preheating for
5 min at 94?C followed by 40 cycles of 20 sec at 94?C,
20 sec at 39?C (GP5þ/GP6þ), or 55?C (MY09/MY11;
Glob-F/Glob-R) and 40 sec at 72?C, and a final extension
of 5 min at 72?C. PCR was performed in a final volume of
20 mL, which consisted of 3 mL extracted DNA as a template
(100 ng), 2 mL 10X PCR buffer, 1 U Taq polymerase
(Applied Biosystems, Foster City, CA), 10 pmol each primer
and 0.2 mM dNTPs. A 1.5-mM MgCl2concentration was
used for all primers. DNA samples from case and control
subjects were included randomly, blinded to clinical and
demographic characteristics of samples. PCR was performed
in a programmable thermal cycler (Mastercycler gradient;
Eppendorf, Westbury, NY). A 5-mL aliquot of the amplifica-
tion mixture was analyzed by electrophoresis in 1.5 agarose
gel and visualized by UV light after ethidium bromide
Positive PCR products were purified and subsequently
sequenced in a programmable thermal cycler (Mastercycler
gradient; Eppendorf) using the BigDye Terminator v3.1
Cycle Sequencing Kit (Applied Biosystems) by using one
of the PCR oligonucleotides as a sequencing primer
(GP5þ and MY09). The PCR mix and cycling conditions
were used as recommended by the manufacturer. Sequence
analysis was performed in the ABI PRISM 3100 Genetic
Analyzer System (Applied Biosystems). The obtained
sequence was compared with the GenBank database
(National Center for Biotechnology Information, Bethesda,
MD) by using the BLAST program (21).
Descriptive data for cases and controls were summarized
with conventional central tendency statistics. Differences
in demographic and clinical characteristics between high
risk HPV (HR-HPV) positive and negative cases were
calculated using the c2test and Fisher’s exact test, when
necessary. For dimensional variables, the Mann-Whitney
U-test was applied. Univariate analysis between oral cancer
and each risk factor under study was carried, and OR and its
95% confidence intervals (CI) were estimated using condi-
tional logistic regression models. The category of indepen-
dent variable with assuming minimal level of risk of oral
cancer was taken as reference. A two-sided p value
!0.05 was considered significant.
Human Papillomavirus and Oral Cancer in Mexico
To examine independence of the effects, multivariate lo-
gistic regression was performed by using conditional logis-
tic regression functions. The criteria for selection of the
variables entering the model were based on both biological
plausibility and p value !0.20 in the univariate analysis. A
backward stepwise variable selection strategy was used to
build a final model with a significance level $0.1 for
removal and a significance level !0.05 for re-entry into
the model. Biologically meaningful interactions between
OSCC and risk factors were also assessed for inclusion in
the models. The STATA program v.8.0 (Stata Corporation
LP, College Station, TX) was used for data analysis.
Sixty two cases and 248 controls (53.2% males) were in-
cluded in the present study; the participants had a median
age of 62 years (Q1eQ3, 54e72 years). The distribution of
cancer cases and controls according to sociodemographic
ditional regression analysis showed that none of the selected
education (p 5 0.307) or occupation (p 5 0.195). The most
frequently affected OSCC site was the tongue in 33 cases
(53.2%), followed by the gingival in 12 cases (19.4%), and
mucosa accounted jointly for seven of the OSCC cases
Table 2 shows the ORs and their 95% confidence inter-
vals for those clinical variables that showed a significant as-
sociation. Positive familial history of cancer was associated
with the presence of oral cancer (OR 5 4.16, 95% CI 5
2.1e8.1, p !0.0001). The odds ratios for oral cancer in in-
dividuals who were current smokers and drinkers were high
(7.8 and 6.3, respectively), also high odds ratios were found
when the cumulative amount of tobacco and alcohol con-
sumption was analyzed.
As shown in Table 2, in cases, tobacco consumption was
present in 32 (51.6%) and alcohol use in 21 (33.9%), both
exposures were present in 15 (24.2%) cases. Two hundred
and three individuals (65.5%) of the 310 participants were
not exposed to either of these two risk factors (24 cases
[38.7%] and 179 [72.2%] controls).
As described in Table 3, oral HPV infection was identi-
fied in 27 (43.5%) of the 62 cases and 43 (17.3%) of the 248
controls. Six HR-HPV types were detected in 23 cases
(37.1%) and 24 controls (9.7%), of which four were more
frequent in cases than in controls: HPV-16 (55.6 vs.
34.9%), HPV-18 (18.5 vs. 11.6%), HPV-33 (7.4 vs. 2.3%),
and HPV-35 (3.7 vs. 0%). Two high risk HPV types were
found in the controls but not in the cases, HPV-31 (4.7%)
and HPV-52 (2.3%). In contrast, the frequency of low- risk
HPV types was lower in cases than in controls (HPV-1 [3.7
vs. 16.3%], HPV-2 [3.7 vs. 7.0%], HPV-6 [0 vs. 9.3%],
HPV-13 [3.7 vs. 9.3%], with the exception of HPV-32, that
was more frequent in the cases (3.7%) than in the controls
As shown in Table 3, an association between oral cancer
with overall HPV (OR 5 3.41) and HR-HPV types (OR 5
6.21) was found. In contrast, low HPV types did not show
an association with OSCC (OR 5 0.83, p 5 0.747).
The conditional logistic regression analysis showed that
familial history of cancer, smoking, and alcohol consump-
tion was associated with OSCC. An interaction was
detected between alcohol and smoking (Table 4); for
alcohol, distinct ORs were obtained in the strata of non-
smokers, where the OR was 3.19, compared with the
OR 5 28.44 in the strata of smokers (!25 pack-years).
In addition, for tobacco consumption, in the strata of non-
drinkers lower odds were shown (OR 5 2.65) compared
with the odds of the strata of drinkers (OR 5 34.29). Also,
high risk HPV was independently associated with OSCC,
roughly a 6-fold increased odds ratio (Table 4). The pres-
ence of tobacco, alcohol, and/or HR-HPV infection was
identified in 52 (83.9%) of all cases in this study. Intrigu-
ingly, 10/62 patients with OSCC (16.1%) were negative
to tobacco, alcohol use, and oral high-risk papillomavirus
infection; with a median age of 67.5 years (Q1eQ3,
37e67 years), 6 (60%) were females.
Forty six cases (74.2%) were in advanced stages of the
disease (III and IV) and 30 (48.4%) showed well-differen-
tiated carcinomas. When cases were analyzed according to
HR-HPV status, no statistically significant differences were
found in regard to gender, age, clinical stage, histological
grading, anatomic cancer location, and familial history of
cancer. The use of tobacco was higher among the cases
who did not show HR-HPV in oral mucosa, but the differ-
ence was not significant (50.0 vs. 26.1%, p 5 0.153); as
well, 14/39 cases who were negative to HR-HPV were
Table 1. Distribution of demographic characteristics among cases
Cases (n 5 62) Controls (n 5 248)
Farmers and craftsmen
Anaya-Saavedra et al./ Archives of Medical Research 39 (2008) 189e197
drinkers (35.9%), in contrast with 3/23 (13.0%) positive
cases (p 5 0.023). Thirteen of the 23 (56%) HR-HPV-
positive cases did not consume tobacco or alcohol; in con-
trast with 11 of the 39 (28.2%) HR-HPV-negative cases
(p 5 0.027).
The comparison of sexual behavior in OSCC cases
according to HR-HPV status is shown in Table 5. The age
at the first sexual intercourse and the number of lifetime
sexual partners showed a statistically significant difference
between the HR-HPV-positive and -negative oral cancer
cases, the rest of the variables revealed no significant differ-
ences between groups.
The present study is the first case-control study of oral can-
cer performed in a Mexican population. Results showed
a strong association between HR-HPVinfection and OSCC,
independently of tobacco and alcohol. Cases with oral HR-
HPV infection had increased odds ratio (6-fold) for oral
cancer, compared with controls.
A previous report in OSCC completed in the same Mex-
ican cancer referral center as this study showed a compara-
ble HPV prevalence (42%). However, the earlier study did
not include a control group and only estimated the propor-
tion of HPV-16-positive cases (14).
HPV prevalence in oral cancer found in the present study
(43.5%) was in accordance with a recent population-based
case-control study in Sweden, which reported that 40% of
131 oral and oropharyngeal squamous cell carcinomas were
HPV positive (22). In both studies, the identification of vi-
ral DNA sequences was performed with the same sets of
primers and under similar PCR and HPV-DNA sequencing
conditions. In other reports, in which the HPV positivity
varied from 34.9e54.3% (23e26), there were some differ-
ences in the methods applied such as the PCR primers used
Table 2. Conditional logistic odds ratios (OR) and 95% confidence interval for oral cancer of selected risk factors
Cases (n 5 62) Controls (n 5 248)
(%) Crude OR(95% CI)
Familial history of cancer
Tobacco consumption (pack-years)
Alcohol consumption (DU-years)
Table 3. Conditional logistic odds ratios (OR) and 95% confidence interval for oral human papillomavirus infection
Cases (n 5 62)Controls (n 5 248)
(%)Crude OR(95% CI)
Oral low-risk HPV
Human Papillomavirus and Oral Cancer in Mexico
(23) or the use of specific oligonucleotides for HPV-16 and
HPV-18 E6 gene (25).
Consistent with the 36% of HR-HPV-positive oral can-
cer cases identified by Goran-Hansson et al. (22), we found
that 37.1% of the OSCC patients were positive for high-risk
papillomavirus. The most common HPV types detected
were HPV-16 (55.6%), which is consistent with the major-
ity of oral cancer reports (12,13) and followed by HPV-18
(18.5%), which was the second most common type in
numerous reports (8,13,27).
The present study identified HPVas a risk factor for the
development of oral cancer (OR 5 3.4), similar to an
American case-control study that demonstrated an associa-
tion (OR 5 3.0) of HPV infection and laryngeal cancer,
after adjusting for age, smoking, and alcohol consumption
(28). Likewise, a meta-analysis of OSCC that analyzed
independently 94 HPV-OSCC studies found that oral
HPV infection was 4.7-fold more likely to be detected in
oral cancer, compared with normal oral mucosa (29). In ad-
dition, the findings of a number of well-controlled studies
have demonstrated that HPV was associated with an in-
creased risk of OSCC (3e7 times), independent of expo-
sure to alcohol or tobacco (8,30). An earlier study
recognized HPV-16 as a risk for the development of oral
cancer (OR 6.2) in males (31), analogous to the OR 6.2
for HR-HPV found in this Mexican study.
Table 4. Conditional logistic regression adjusted odds ratios of selected risk factors for oral cancer
Variable Adjusted ORa
Familial cancer historyb
aAdjusted OR: odds ratios were derived from a conditional logistic regression model adjusted for all variables in the model.
cEstimates not provided for strata $25 pack-years and alcohol due to small sample in strata.
Table 5. Sexual behavior in OSCC cases according to HR-HPV status
Median age at first sexual intercourse (Q1eQ3)
Median lifetime sexual partners (Q1eQ3)
Categorized lifetime sexual partners
Lifetime oralegenital sexual partners
Q1eQ35 interquartile interval.
aMann-Whitney U test.
bFisher’s exact test.
Anaya-Saavedra et al./ Archives of Medical Research 39 (2008) 189e197
The possible causal association of papillomavirus, par-
ticularly HPV-16, with HPV-DNA positive HNC has been
supported by several studies (7,32,33) and confirmed by
a recent nested case-control study in oropharyngeal cancer
(4). Moreover, evidence of hypermethylation of CpGs is-
lands at the L1 region and part of the LCR of HPV-16 as-
sociated with OSCC recently found by our group suggests
that HPV-16 genomes are chromosomally integrated in oral
malignancies and it affects the host cells in a similar
manner as in anogenital malignancies (34).
In the present study, the frequency of oral HR-HPV in-
fection among controls was low (9.7%), consistent with
most reports that have found a HR-HPV frequency that
varies from 0.6 to 11.1% (8,22,30,35e40). Nevertheless,
the frequency of HR-HPV in normal mucosa has been re-
ported as high as 36% (25) and 81.1%, when all HPV types
have been analyzed (41). The detection of HPV in normal
oral mucosa would suggest that not all HPV infections lead
to malignancy; probably geographic and other exposure-
related factors, as well as behavioral aspects, play a role
in OSCC development (11).
Independent of HPV infection, we found that tobacco
and alcohol use were strongly associated with OSCC, in
agreement with a large number of studies (1,42,43) that
have confirmed the role of these habits as strong oral cancer
determinants. The different measurements of tobacco
smoking and alcohol consumption analyzed in this study
were associated with OSCC; also, the interaction detected
between alcohol and smoking showed a noticeable impact
in the odds of OSCC when both risk factors were present.
The interaction found between alcohol and tobacco use is
in accordance with several studies that have demonstrated
a synergistic effect between the two habits, as risk factors
for all upper aerodigestive tract malignancies (1,42).
In addition, in the present study the highest percentage
of smokers and drinkers was found in the HPV-negative
HNC group, as has been reported (4e7,44,45); in contrast,
a number of studies have found no differences in smoking
between HPV-positive and -negative patients (46,47).
Whether there is a different carcinogenic pathway for
HNSCC patients with HR-HPV compared with those can-
cers related to heavy tobacco and alcohol use is not clear,
and more studies are needed in this area.
Two separate pathways of HNSCC development have re-
cently been proposed; one suggests that carcinogenesis is
mainly promoted by environmental factors such as tobacco
and alcohol, and the other in which the oncogenic process is
favored by HR-HPV (4,6,7). Accordingly, in our findings,
in more than half of the HR-HPV-positive OSCC cases
(56%), there was a negative history of tobacco or alcohol
consumption; therefore, these two factors could not explain
the development of the oral malignancy in these patients.
The precise contribution of HPVin the etiology of oral can-
cer remains to be determined, and this is essential in order
to establish possible preventive measures.
An association between oral cancer and a positive famil-
ial history (first-line relatives) of cancer was found in the
present study. Various investigations have collected infor-
mation on familial background of cancer as a possible risk
factor, but the results are inconclusive (4,48e50). It has
been suggested that environmental factors may be responsi-
ble for the familial effects observed, especially in tobacco-
related sites such as lung/oral cancer (49,51). In addition, it
is necessary to consider the possibility of the presence of
recall bias, frequently shared by case-control studies due
to the fact that cases had a higher motivation to recall their
cancer familial history than the controls (52).
Epidemiological studies have suggested that the oral ex-
posure to HPV could be related to sexual behavior (53,54).
Our data indicated that larger numbers of lifetime sexual
partners and early age at first sexual intercourse were sig-
nificantly associated with the presence of HR-HPV infec-
tion (p 5 0.033 and p 5 0.019, respectively). This
finding is supported by other studies indicating that a high
lifetime number of sexual partners and young age of first
intercourse were associated with oral cancer after adjusting
for alcohol and tobacco exposure (4,5,8,53). Nevertheless,
a potential limitation of the present study is the possibility
that the sexual behaviors were underreported, due to the
personal nature of the questions.
It is important to highlight that in 10 cases (16.1%) of
OSCC, we were unable to find any of the selected risk fac-
tors associated with their malignancy. Considering that oral
carcinogenesis is a multistep process of cumulative genetic
damage (55), a number of unidentified factors may be asso-
ciated as potential initiators and/or promoters of oncogene-
sis. Further research is needed to address other risk factors
such as genetic and epigenetic mechanisms that may be
involved in malignant transformation.
The design of this case-control study allowed us to com-
pare the presence of HR-HPV between patients with cancer
of the oral cavity and matched controls. Each case had four
matched controls, and both groups of individuals came
from comparable sociodemographic areas; also, a high par-
ticipation rate was reached in both groups. In addition, all
patients and controls were examined and interviewed by
the same person who was responsible for the collection of
the laboratory samples.
Another strength of the study was the use of laboratory
procedures with high sensitivity such as PCR for viral
(MY09/11 and GP5þ/6þ) that increased the viral detection
boundary (56). The use of the second set of primers allowed
us to retrieve 10.3% (30/276) of positive samples (22.4% in
ly use of MY09/11. Also, suggested methods for contamina-
would permit the identification of missed HPV types.
Human Papillomavirus and Oral Cancer in Mexico
In conclusion, results of this study have indicated that
HR-HPV was strongly associated with oral cancer in
a group of Mexican patients, suggesting that high-risk pap-
illomavirus plays a role in oral carcinogenesis. Also, the
important role of tobacco and alcohol as cancer determi-
nants was confirmed, and an association of oral HR-HPV
presence with sexual behavior was detected. The precise
understanding of the oncogenic potential of HPV in HNC
will require more extensive knowledge of the natural his-
tory of the virus in normal tissues. In that case, it would
be useful to conduct cohort studies on HPV infection in
the head and neck region that could allow us to characterize
the persistence of HPV in normal and in transformed cells.
This research is the result of Dr. Anaya-Saavedra’s research pro-
ject to obtain the PhD degree in the Doctorado en Ciencias Biolo ´g-
icas of the Universidad Autono ´ma Metropolitana. We are grateful
to the surgeons, oncologists, dental surgeons and Department
Heads of INCan, Hospital General ‘‘Dr. Manuel Gea Gonza ´lez’’,
and the Dental Clinic ‘‘Dr. Rafael Lozano’’ of UAM-Xochimilco,
for its substantial contribution in the recruitment of the study
participants. We also especially thank Dr. Sergio Ponce-de-
Leo ´n-Rosales for his helpful comments on this project. The study
was supported by a grant from CONACYT (no. 25987).
1. Castellsague ´ X, Quintana MJ, Martı ´nez MC, Nieto A, Sa ´nchez MJ,
Amparo J, et al. The role of type of tobacco and type of alcoholic
beverage in oral carcinogenesis. Int J Cancer 2004;108:741e749.
2. Wogan GN, Hecht SS, Felton JS, Conney AH, Loeb LA. Environmen-
tal and chemical carcinogenesis. Semin Cancer Biol 2004;14:
3. Farshadpour F, Hordijk GJ, Koole R, Slootweg PJ. Non-smoking and
non-drinking patients with head and neck squamous cell carcinoma:
a distinct population. Oral Dis 2007;13:239e243.
4. D’Souza G, Kreimer AR, Viscidi R, Pawlita M, Fakhry C, Koch WM,
et al. Case-control study of human papillomavirus and oropharyngeal
cancer. N Engl J Med 2007;356:1944e1956.
5. Smith EM, Ritchie JM, Summersgill KF, Klussmann JP, Lee JH,
Wang D, et al. Age, sexual behavior and human papillomavirus infec-
tion in oral cavity and oropharyngeal cancers. Int J Cancer 2004;108:
6. Herrero R, Castellsague ´ X, Pawlita M, Lissowska J, Kee F, Balaram P,
et al. Human papillomavirus and oral cancer: the International Agency
for Research on Cancer multicenter study. J Natl Cancer Inst 2003;95:
7. Gillison ML, Koch WM, Capone RB, Spafford M, Westra WH, Wu L,
et al. Evidence for a causal association between human papillomavirus
and a subset of head and neck cancers. J Natl Cancer Inst 2000;92:
8. Schwartz SM, Daling JR, Doody DR, Wipf GC, Carter JJ,
Madeleine MM, et al. Oral cancer risk in relation to sexual history
and evidence of human papillomavirus infection. J Natl Cancer Inst
9. Nair S, Pillai MR. Human papillomavirus and disease mechanisms:
relevance to oral and cervical cancers. Oral Dis 2005;11:350e359.
10. Balaram P, Nalinakumari KR, Abraham E, Balan A, Hareendran NK,
Bernard HU, et al. Human papillomaviruses in 91 oral cancers from
Indian betel quid chewers—high prevalence and multiplicity of infec-
tions. Int J Cancer 1995;61:450e454.
11. Ha PK, Califano JA. The role of human papillomavirus in oral carci-
nogenesis. Crit Rev Oral Biol Med 2004;15:188e196.
12. Syrja ¨nen S. Human papillomavirus (HPV) in head and neck cancer.
J Clin Virol 2005;32:59e66.
13. Kreimer AR, Clifford GM, Boyle P, Franceschi S. Human papillomavi-
rus types in head and neck squamous cell carcinomas worldwide:a sys-
tematic review. Cancer Epidemiol Biomarkers Prev 2005;14:467e475.
14. Ibieta BR, Lizano M, Frı ´as-Mendivil M, Barrera JL, Carrillo A, Ma
Ruı ´z-Godoy L, et al. Human papilloma virus in oral squamous cell
carcinoma in a Mexican population. Oral Surg Oral Med Oral Pathol
Oral Radiol Endod 2005;99:311e315.
15. STROBE Group. STROBE Statment. Checklist of essential items
Version 3 (Sept 2005). Available at: http://www.strobe.statement.
16. World Health Organization. International Statistical Classification of
Diseases and Related Health Problems: 10th revision. Geneva: World
17. American Joint Committee on Cancer, AJCC Cancer Staging Manual.
5thed. Chicago: Lippincott Williams and Wilkins;1997.
18. Saiki RK, Scharf S, Faloona F, Mullis KB, Horn GT, Erlich HA, et al.
Enzymatic amplification of beta-globin genomic sequences and
restriction site analysis for diagnosis of sickle cell anemia. Science
19. Manos MM, Ting Y, Wright DK, Lewis AJ, Broker TR, Wolinsky SM.
Use of polymerase chain reaction amplification for the detection of
genital human papillomavirus. Cancer Cells 1989;7:209e214.
20. De Roda Husman AM, Walboomers JMM, van den Brule AJC,
Meijer CJLM, Snijders PJF. The use of general primers GP5 and
GP6 elongated at their 30ends with adjacent highly conserved
sequences improves human papillomavirus detection by PCR. J Gen
21. Altschul SF, Madden TL, Scha ¨fer AA, Zhang J, Zhang Z, Miller W,
et al. Gapped BLAST and PSI-BLAST: a new generation of protein
database search programs. Nucleic Acid Res 1997;25:3389e3402.
22. Go ¨ran Hansson B, Rosenquist K, Antonsson A, Wennerberg J,
Schildt E, Bladstro ¨m A, et al. Strong association between infection
with human papillomavirus and oral and oropharyngeal squamous cell
carcinoma: a population-based case-control study in southern Sweden.
Acta Otolaryngol 2005;125:1337e1344.
23. Tachezy R, Klozar J, Salakova M, Smith E, Turek L, Betka J, et al.
HPV and other risk factors of oral cavity/oropharyngeal cancer in
the Czech Republic. Oral Dis 2005;11:181e185.
24. Chang JY, Lin MC, Chiang CP. High-riskhuman papillomaviruses may
have an important role in non-oral habits-associated oral squamous cell
carcinomas in Taiwan. Am J Clin Pathol 2003;120:909e916.
25. Sugiyama M, Bhawal UK, Dohmen T, Ono S, Miyauchi M,
Ishikawa T. Detection of human papillomavirus-16 and HPV-18
DNA in normal, dysplastic, and malignant oral epithelium. Oral Surg
Oral Med Oral Pathol Oral Radiol Endod 2003;95:594e600.
26. Cruz IB, Snidjers PJ, Steenbergen RD, Meijer CJ, Snow GB,
Walboomers JM, et al. Age-dependence of human papillomavirus
DNA presence in oral squamous cell carcinomas. Eur J Cancer B Oral
27. Adams V, Schmid S, Zariwala M, Schmid M, Kleihues P, Briner J,
et al. Prevalence of human papilloma virus DNA in head and neck can-
cers carrying wild-type or mutant p53 tumor supressor genes. Antican-
cer Res 1999;19:1e6.
28. Smith EM, Summersgill KF, Allen J, Hoffman HT, McCulloch T,
Turek LP, et al. Human papillomavirus and risk of laryngeal cancer.
Ann Otol Rhinol Laryngol 2000;109:1069e1076.
29. Miller CS, Johnstone BM. Human papillomavirus as a risk factor for
oral squamous cell carcinoma: a meta-analysis, 1982e1997. Oral Surg
Oral Med Oral Pathol Oral Radiol Endod 2001;91:622e635.
Anaya-Saavedra et al./ Archives of Medical Research 39 (2008) 189e197
30. Smith EM, Hoffman HT, Summersgill KS, Kirchner HL, Turek LP, Download full-text
Haugen TH. Human papillomavirus and risk of oral cancer. Laryngo-
31. Maden C, Beckmann AM, Thomas DB, McKnight B, Sherman KJ,
Ashley RL, et al. Human papillomaviruses, herpes simplex viruses, and
the risk of oral cancer in men. Am J Epidemiol 1992;135:1093e1102.
32. Smith EM, Ritchie JM, Pawlita M, Rubenstein LM, Haugen TH,
Turek LP, et al. Human papillomavirus seropositivy and risks of head
and neck cancer. Int J Cancer 2006;120:825e832.
33. Wiest T, Schwarz E, Enders C, Flechtenmacher C, Bosch FX. Involve-
ment of intact HPV16 E6/E7 gene expression in head and neck can-
cers with unaltered p53 status and perturbed pRb cell cycle control.
34. Balderas-Loaeza A, Anaya-Saavedra G, Ramirez-Amador VA, Guido-
35. Furrer VE, Benitez MB, Furnes M, Lanfranchi HE, Modesti NM. Bi-
opsy vs. superficial scraping: detection of human papillomavirus 6, 11,
16 and 18 in potentially malignant and malignant oral lesions. J Oral
Pathol Med 2006;35:338e344.
36. Koppikar P, de Villiers E, Mulherkar R. Identification of human pap-
illomaviruses in tumors of the oral cavity in an Indian community. Int
J Cancer 2005;113:946e950.
37. Kurose K, Terai M, Soedarsono N, Rabello D, Nakajima Y, Burk RD,
et al. Low prevalence of HPV infection and its natural history in nor-
mal oral mucosa among volunteers on Miyako Island, Japan. Oral
Surg Oral Med Oral Pathol Oral Radiol Endod 2004;98:91e96.
38. Giovannelli L, Campisi G, Lama A, Giambalvo O, Osborn J,
Margiotta V, et al. Human papillomavirus DNA in oral mucosal
lesions. J Infect Dis 2002;185:833e836.
39. Mork J, Lie AK, Glattre E, Hallmans G, Jellum E, Koskela B, et al.
Human papillomavirus infection as a risk factor for squamous-cell car-
cinoma of the head and neck. N Engl J Med 2001;344:1125e1131.
40. Summersgill KF, Smith EM, Kirchner HL, Haugen TH, Turek LP. P53
polymorphism, human papillomavirus infection in the oral cavity, and
oral cancer. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2000;
41. Terai M, Hashimoto K, Yoda K, Sata T. High prevalence of human
papillomaviruses in the normal oral cavity of adults. Oral Microbiol
42. Johnson N. Tobacco use and oral cancer: a global perspective. J Dental
43. Scully C, Field JK, Tanzawa H. Genetic aberrations in oral or head and
neck squamous cell carcinoma (SCCHN): 1. Carcinogen metabolism,
DNA repair and cell cycle control. Oral Oncol 2000;36:256e263.
44. Hafkamp HC, Speel EJ, Haesevoets A, Bot FJ, Dinjens WNM,
Ramaekers FCS, et al. A subset of head and neck squamous cell car-
cinomas exhibits integration of HPV 16/18 DNA and overexpression
of p16INK4A and p53 in the absence of mutations in p53 exons
5-8. Int J Cancer 2003;107:394e400.
45. Ritchie JM, Smith EM, Summersgill KF, Hoffman HT, Wang D,
Klussmann JP, et al. Human papillomavirus infection as a prognostic
factor in carcinomas of the oral cavity and oropharynx. Int J Cancer
46. Sisk EA, Soltys SG, Zhu S, Fisher SG, Carey TE, Bradford CR. Hu-
man papillomavirus and p53 mutational status as prognostic factors in
head and neck carcinoma. Head Neck 2002;24:841e849.
47. Paz IB, Cook N, Odom-Maryon T, Xie Y, Wilcynski SP. Human pap-
illomavirus (HPV) in head and neck cancer. An association of HPV 16
with squamous cell carcinoma of Waldeyer’s tonsillar ring. Cancer
48. Mork J, Moller B, Glattre E. Familial risk in head and neck squamous
cell carcinoma diagnosed before the age of 45: a population-based
study. Oral Oncol 1999;35:360e367.
49. Ankathil R, Mathew A, Joseph F, Nair MK. Is oral cancer susceptibil-
ity inherited? Report of five oral cancer families. Oral Oncol Eur J
50. Goldstein AM, Blot WJ, Greenberg RS, Schoenberg JB, Austin DF,
Preston-Martin S, et al. Familial risk in oral and pharyngeal cancer.
Oral Oncol Eur J Cancer 1994;30B:319e322.
51. Llewellyn CD, Johnson NW, Warnakulasuriya KAAS. Risk factors for
squamous cell carcinoma of the oral cavity in young people—a com-
prehensive literature review. Oral Oncol 2001;401e408.
52. Herna ´ndez-Avila M, Garrido F, Salazar-Martı ´nez E. Biases in epide-
miological studies. Salud Publica Mex 2000;42. 438e436.
53. Scully C. Oral cancer: the evidence for sexual transmission. Br Dent J
54. Scully C. Oral squamous cell carcinoma: from an hypothesis about
a virus, to concern about possible sexual transmission. Oral Oncol
55. Carbone M, Pass HI. Multistep and multifactorial carcinogenesis:
when does a contributing factor become a carcinogen? Semin Cancer
56. Carrillo A, Mohar A, Meneses A, Frı ´as-Mendivil M, Solorza G,
Lizano M. Usefulness of combining universal oligonucleotides in
detecting human papillomavirus in cervical cancer and premalignant
lesions. Salud Publica Mex 2004;46:7e15.
57. Boy S, Van Rensburg EJ, Engelbrecht S, Dreyer L, van Heerden M,
van Heerden W. HPV detection in primary intra-oral squamous cell
carcinomas—commensal, aetiological agent or contamination? J Oral
Pathol Med 2006;35:86e90.
Human Papillomavirus and Oral Cancer in Mexico