Human Papillomavirus and Oral Cancer: The

International Agency for Research on Cancer

Multicenter Study
Rolando Herrero, Xavier Castellsagué, Michael Pawlita, Jolanta Lissowska,
Frank Kee, Prabda Balaram, Thangarajan Rajkumar, Hema Sridhar, Barbara
Rose, Javier Pintos, Leticia Fernández, Ali Idris, María José Sánchez,
Adoración Nieto, Renato Talamini, Alessandra Tavani, F. Xavier Bosch, Ulrich
Reidel, Peter J. F. Snijders, Chris J. L. M. Meijer, Raphael Viscidi,
Nubia Muñoz, Silvia Franceschi
For the IARC Multicenter Oral Cancer Study Group

cases expected for 2005, if current incidence rates remain un-

Background: Human papillomavirus (HPV), the causal agent changed (1,2). Tobacco and alcohol are established etiologic
of cervical cancer, appears to be involved in the etiology of agents of these cancers (3), with attributable fractions of approx-

Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015

cancer of the oral cavity and oropharynx. To investigate imately 90%. Micronutrient deficiencies (4,5) and poor oral
these associations, we conducted a multicenter case– control hygiene (6) have also been associated with increased risk.
study of cancer of the oral cavity and oropharynx in nine Human papillomavirus (HPV) is etiologically involved in
countries. Methods: We recruited 1670 case patients (1415 virtually all cervical cancers [for review, see (7)], and the early
with cancer of the oral cavity and 255 with cancer of the HPV oncoproteins E6 and E7 are responsible for the malignant
oropharynx) and 1732 control subjects and obtained an phenotype, mainly through inactivation of tumor suppressor
interview, oral exfoliated cells, and blood from all partici- proteins such as p53 and pRB. Previous work evaluating differ-
pants and fresh biopsy specimens from case patients. HPV
ent markers of exposure and viral activity in tumors indicates
DNA was detected by polymerase chain reaction (PCR).
that HPV may also play a role in some cancers of the oral cavity
Antibodies against HPV16 L1, E6, and E7 proteins in plasma
and oropharynx (8). Several studies [(9 –11); for review, see
were detected with enzyme-linked immunosorbent assays.
(12)] have investigated prevalence of HPV in these cancers, but
Multivariable models were used for case– control and case–
case comparisons. Results: HPV DNA was detected in biopsy the prevalence of HPV detection varies broadly, depending on
specimens of 3.9% (95% confidence interval [CI] ⴝ 2.5% to the population, combination of subsites, type of specimen, and
5.3%) of 766 cancers of the oral cavity with valid PCR detection method. HPV is consistently and more frequently
results and 18.3% (95% CI ⴝ 12.0% to 24.7%) of 142 detected in cancers of the oropharynx and tonsil than at other
cancers of the oropharynx (oropharynx and tonsil com-
bined) with valid PCR results. HPV DNA in cancer biopsy
specimens was detected less frequently among tobacco smok- Affiliations of authors: International Agency for Research on Cancer, Lyon,
ers and paan chewers and more frequently among subjects France (RH, NM, SF); Proyecto Epidemiológico Guanacaste, Costa Rican Foun-
dation for Health Sciences, San José, Costa Rica (RH); Institut Català
who reported more than one sexual partner or who prac- d’Oncologia, Barcelona, Spain (XC, FXB); Deutsches Krebsforschungszentrum,
ticed oral sex. HPV16 DNA was found in 94.7% of HPV Heidelberg, Germany (MP, UR); Cancer Center and Institute of Oncology,
DNA–positive case patients. HPV DNA in exfoliated cells Warsaw, Poland (JL); Queen’s University Belfast, Northern Ireland, United
was not associated with cancer risk or with HPV DNA Kingdom (FK); Regional Cancer Center, Trivandrum, India (PB); Cancer Insti-
detection in biopsy specimens. Antibodies against HPV16 L1 tute, Women India Association, Chennai, India (TR); Kidwai Memorial Institute
were associated with risk for cancers of the oral cavity (odds of Oncology, Bangalore, India (HS); Sydney Head and Neck Cancer Institute,
Royal Prince Alfred Hospital and University of Sydney, Sydney, Australia (BR);
ratio [OR] ⴝ 1.5, 95% CI ⴝ 1.1 to 2.1) and the oropharynx McGill University, Montreal, Canada (JP); Instituto Nacional de Oncologia y
(OR ⴝ 3.5, 95% CI ⴝ 2.1 to 5.9). Antibodies against HPV16 Radiobiologia, Havana, Cuba (LF); Toombak and Smoking Research Center,
E6 or E7 were also associated with risk for cancers of the Khartoum, Sudan (AI); Escuela Andaluza de Salud Publica, Granada, Spain
oral cavity (OR ⴝ 2.9, 95% CI ⴝ 1.7 to 4.8) and the oro- (MJS); Facultad de Medicina, Seville, Spain (AN); Centro di Riferimento
pharynx (OR ⴝ 9.2, 95% CI ⴝ 4.8 to 17.7). Conclusions: Oncologico di Aviano, Aviano, Italy (RT); Istituto di Ricerche Farmacologiche
HPV appears to play an etiologic role in many cancers of the “Mario Negri,” Milan, Italy (AT); Vrije Universiteit Medical Center, Amster-
dam, The Netherlands (PJFS, CJLMM); Stanley Division of Developmental
oropharynx and possibly a small subgroup of cancers of the
Neurovirology, Department of Pediatrics, The Johns Hopkins University School
oral cavity. The most common HPV type in genital cancers of Medicine, Baltimore, MD (RV).
(HPV16) was also the most common in these tumors. The Correspondence to: Rolando Herrero, MD, PhD, Proyecto Epidemiológico
mechanism of transmission of HPV to the oral cavity war- Guanacaste, P.O. Box 125-6151, San José, Costa Rica (e-mail: rherrero@
rants further investigation. [J Natl Cancer Inst 2003;95: amnet.co.cr).
1772– 83] See “Appendix” for a list of additional members of the study group.
See “Notes” following “References.”
DOI: 10.1093/jnci/djg107
Cancers of the oral cavity and oropharynx constitute major Journal of the National Cancer Institute, Vol. 95, No. 23, © Oxford University
worldwide public health problems, with more than 400 000 new Press 2003, all rights reserved.

1772 ARTICLES Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003
head and neck sites, and HPV16 tends to be the predominant were also excluded. When patients with cancers with topogra-
type detected (9 –11). Results of a few case– control studies phies not in the exclusion criteria were recruited as control
(10,11) and a cohort study (13) point to a likely role of HPV in subjects, they were recruited before any primary treatment and
some cancers at certain anatomical sites. However, many aspects should not account for more than 20% of the overall comparison
of the association remain to be investigated to better define the group. In India, control subjects were selected among patients
precise contribution of HPV to the etiology of these tumors and without cancer presenting for voluntary cancer checkup or
the potential preventive interventions (14). among visitors of patients with diseases other than cancer of the
We report an International Agency for Research on Cancer oral cavity or oropharynx (16). In Northern Ireland, community
(IARC) multicenter case– control study of cancer of the oral control subjects were selected from county listings, geographi-
cavity and oropharynx carried out in nine countries that used a cally matched to the case patients, and invited by personal letter.
common protocol for data and specimen collection and included We recruited 1670 case patients (1415 with cancer of the oral
several markers of HPV infection that were assayed in central cavity and 255 with cancer of the oropharynx) and 1732 control
laboratories. subjects in nine countries. The study was approved by the IARC
and local ethical review committees, and participants gave writ-
MATERIALS, PATIENTS, AND METHODS ten informed consent according to local regulations.

The study was conducted in Italy, Spain, Northern Ireland, Data and Specimen Collection
Poland, India, Cuba, Canada, Australia, and Sudan from April
1996 through December 1999. Case patients were recruited as Specially trained interviewers administered a standardized

Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015

follows: in Italy, from three cancer referral hospitals in the questionnaire that included demographic characteristics (age,
Friuli–Venezia–Giulia and Lombardia regions (6); in Spain, sex, ethnic group, area of birth and residence, religion, and
from two hospitals in Barcelona, one in Granada, and one in language spoken), education (ever attendance at school, number
Seville (3,5); in Northern Ireland, from the School of Dentistry of years of school attendance, and age when stopped going to
and several general and cancer hospitals in Belfast; in Poland, school), longest occupation, use of tobacco in its different forms,
from one maxillofacial surgery clinic in Warsaw (15); in India, alcohol drinking habits, dietary habits, marital status, sexual
from one cancer referral center each in Madras, Bangalore, and history (number of lifetime sexual partners, visits to prostitutes,
Trivandrum (16); in Cuba, from the main cancer referral center and frequency of oral sex), histories of various diseases, family
in Havana (4); in Canada, from three hospitals in Montreal; in history of cancer, and oral cavity health. A smoker was defined
Australia, from one cancer referral center in Sydney; and in as a subject who reported having smoked tobacco daily for at
Sudan, from a dental hospital in Khartoum. Control subjects least 1 year, and smokers were asked about duration of smoking
were recruited in most centers from the same hospitals or neigh- and amount and type of tobacco smoked (cigarettes, cigars, or
boring general hospitals serving the same populations to which pipes). In India, specific questions were asked about paan chew-
the case patients belonged. In Northern Ireland and India, selec- ing (usually including tobacco, betel leaf, areca nut, and calcium
tion of control subjects was modified so that they were recruited hydroxide). Additional questions were asked about consumption
only if, in the event of a cancer diagnosis, they would have been of snuffed tobacco in some areas (e.g., toombak dipping in
admitted to the same hospitals in which the case patients were Sudan). A drinker was defined as a subject who reported drink-
recruited (see below). ing alcoholic beverages at least once a month, and details were
Case patients with incident cancer of the oral cavity or obtained on type of beverage, amount, and duration.
oropharynx were contacted at participating centers before any Interviewers also briefly examined the oral cavity to deter-
cancer treatment. Topographic locations included the lip (ex- mine the number of missing teeth and the presence of visible
cluding the external lip), base of the tongue, other parts of the lesions. The association of oral cancer with some of the risk
tongue, gum, floor of the mouth, palate, other parts of the mouth, factors has been previously reported for some centers
tonsil, and oropharynx. Cancers of the oropharynx and tonsil (3– 6,15,16).
were combined as oropharynx, and cancers at other sites were Before initiating the study in each center, extensive discus-
grouped as oral cavity (see below). Patients with second primary sions with the local investigators were conducted, and question-
cancers were not eligible, and only carcinomas (including six naires were modified to adapt them to the local culture or habits
adenocarcinomas) were included according to the local pathol- (e.g., so that they asked about toombak dipping in Sudan, betel
ogist’s diagnosis. All cancers of the salivary gland and five quid chewing in India, and cigar smoking in Cuba). However,
lymphomas and one sarcoma of the other oral sites were the core questionnaire was maintained to allow pooling of the
excluded. main data. During the course of the study, the local investigators
In most centers, one hospital control subject per case patient closely supervised the interview and specimen collection pro-
was selected and frequency-matched on center, sex, and 5-year cesses. In addition, the IARC investigators visited study sites
age group. Control subjects were ineligible if they were mentally before initiation of the study and periodically visited the main
or physically unable to give consent, had a history of cancer of centers to ensure adherence to the common international
the oral cavity or oropharynx, or had a diagnosis associated with protocol.
the exposures of interest. For example, potential control subjects After administering the interview, information on tumor–
with tobacco- and/or alcohol-related diseases (such as cancers of node–metastasis classification and histology was obtained from
the lung, larynx, esophagus, bladder, kidney, liver, or pancreas; case patients. This classification was converted into stage ac-
chronic lung diseases; coronary heart diseases; venous throm- cording to the American Joint Committee on Cancer (AJCC)
bosis; hepatitis; or cirrhosis) were not eligible. Patients with Cancer Staging Manual (17). Superficial scrapes of the mucosa
cancers of the anogenital tract, skin, or unknown primary site were carried out with soft toothbrushes, by performing 5–10

Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003 ARTICLES 1773
gentle strokes in predefined areas as follows: right buccal mu- positive, subsequent typing was performed with EIA oligonu-
cosa (from high to low position), left buccal mucosa (from high cleotide probes individually. In this study, no sample with a
to low position), right side of the tongue, dorsal side of the negative EIA result had a positive hybridization result.
tongue, left side of the tongue, and inside of the upper and lower HPV testing of exfoliated cells was performed in 45.2% of
lip, in addition to the gentle brushing of the tumor in case case patients and 40.0% of control subjects from some of the
patients. Cells from brushes were suspended in tubes containing centers (Table 1). The coordinating group decided not to con-
phosphate-buffered saline (PBS). Participants subsequently gar- tinue testing exfoliated cell specimens and to focus on biopsy
gled with saline, and the resulting suspension was added to the material when we observed the lack of association between HPV
same tube. Cells were centrifuged twice and frozen at ⫺70 °C DNA detected in cells and HPV DNA in biopsy specimens (see
until shipment to Lyon (France) and then to Amsterdam (The below).
Netherlands) for HPV testing.
Attempts were made to obtain a biopsy specimen from each Detection of Antibodies Against HPV16 L1
case patient from a non-necrotic area of the tumor before any
cancer treatment, although priority was given to biopsy exami- Preparative purification of HPV16 virus-like particles.
nation required for diagnostic purposes. Biopsy specimens were For production of virus-like particles (VLPs), Trichoplusia ni
placed in liquid nitrogen or in freezers at ⫺70 °C or, in some (High Five) cells (Invitrogen, Carlsbad, CA) were infected with
areas, kept on ice or at 4 °C until later, always within 8 hours of HPV16 L1/L2 recombinant baculovirus, clone 114/K (a gift
collection. Specimens were shipped on dry ice to Lyon, where from John T. Schiller, National Cancer Institute, Bethesda, MD)
they were stored at ⫺70 °C until shipment to Amsterdam for and grown as adherent cultures in tissue culture plates. VLPs

Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015

HPV testing. were purified from cell lysates by density gradient ultracentrif-
A 10-mL heparinized blood sample was obtained, and ali- ugation, POROS 50 HS cation-exchange chromatography (Per-
quots of plasma were prepared and frozen until shipment on dry Septive Biosystems, Framingham, MA), and heparin–Sepharose
ice to Lyon. In Lyon, they were stored at ⫺20 °C until transfer chromatography (Amersham Pharmacia Biotech, Piscataway,
to collaborating laboratories in Heidelberg, Germany, and Bal- NJ), as described previously (22).
timore, MD. HPV 16 VLP-based enzyme-linked immunosorbent assay.
Plasma was tested for antibodies to HPV16 VLP in an enzyme-
DNA Analysis linked immunosorbent assay (ELISA), as described previously
(22). Briefly, 50 ng of VLP protein in PBS (pH 7.2) was
For polymerase chain reaction (PCR), suspensions of exfoli- incubated in wells of PolySorp microtiter plates (Nunc, Naper-
ated cells and biopsy specimens were subjected to freezing– ville, IL) overnight at 4 °C and then blocked for 3 hours at room
thawing and to boiling, as described (18). Snap-frozen biopsy temperature with 10% Superblock (Pierce, Rockford, IL) in PBS
specimens were serially sectioned (10 –15 sections) on a cryo- containing 0.05% Tween 20 (blocking buffer). Blocking buffer
stat. The first and last sections were stained with hematoxylin– was replaced with PBS, and plates were stored at ⫺20 °C until
eosin for histologic assessment of the presence of carcinoma. use. Before use and following each incubation step, the plates
Intermediate sections were used for crude DNA extraction, as were washed four times with PBS containing 0.05% Tween 20
described (19). ␤-Globin PCR analysis was performed (20) to (Sigma-Aldrich, St. Louis, MO) in an automatic plate washer
confirm the presence of human DNA in specimens; specimens (Skanwasher 300; Skatron, Lier, Norway). Plasma samples were
with negative results were considered to not have valid PCR diluted 1 : 100 in blocking buffer, 100 ␮L of this diluent was
results and were excluded from subsequent analyses. Histologic added to duplicate wells, and the microtiter plates were incu-
assessment was possible in 89.3% (811) of the 908 biopsy bated for 1 hour at 37 °C. Antigen-bound immunoglobulin was
specimens with valid PCR results. Of those 811 specimens, a detected with peroxidase-conjugated goat antibodies against hu-
total of 22.6% (183 specimens) were not adequate for histologic man immunoglobulin G (IgG; Zymed, San Francisco, CA) and
assessment (83 specimens) or apparently had no neoplastic cells diluted 1 : 4000 in a solution of 10% Superblock, 2.5% polyeth-
(100 specimens). However, our estimates of detection of HPV ylene glycol (molecular weight ⫽ 20 000; Sigma), and 0.5%
DNA in biopsy specimens did not vary according to the apparent Igepal CA-630 (Sigma) in PBS. After 30 minutes at 37 °C, color
presence or absence of cancer in the specimen. Because all development was initiated by the addition of ABTS peroxidase
subjects had histologically confirmed cancers, we decided to solution (Kirkegaard and Perry, Gaithersburg, MD). The reac-
include them all in the analyses. tion was stopped after 20 minutes by adding 1% sodium dodecyl
For HPV DNA detection, the general primer (GP)-mediated sulfate, and the optical density (OD) was measured at 405 nm,
PCR enzyme immunoassay (GP5⫹/biotinylated GP6⫹ PCR– with a reference wavelength of 490 nm, in an automated micro-
EIA) was performed (19). PCR products were subsequently titer plate reader (Molecular Devices, Menlo Park, CA). The cut
characterized in the following two ways: 1) EIA with two point for seropositivity was determined from the reactivity of
mixtures of oligonucleotide probes specific for 14 high-risk plasma samples from 108 women who were self-reported virgins
HPV types (HPV16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, from Costa Rica. The mean and standard deviation (SD) of
and 68) and six low-risk HPV types (HPV6, 11, 40, 42, 43, and optical density values for the control subjects were calculated,
44) and 2) agarose gel electrophoresis followed by low- and values greater than the mean plus three standard deviations
stringency Southern blot hybridization with a probe mixture were excluded. The analysis was repeated on the remaining
containing general primer PCR products specific for HPV6, 11, samples until no further optical density values could be excluded
16, 18, 31, and 33 (21). Southern blot analysis was performed by this criterion. After four rounds of analysis, with the exclu-
under low-stringency conditions to allow the detection of HPV sion of six samples, the cutoff point was defined as five standard
types not included in the EIA probe mixtures. When the EIA was deviations above the mean of this distribution (0.187 U).

1774 ARTICLES Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003
 Table 1. Distribution of study participants by country and availability of various markers of human papillomavirus (HPV) infection*

No. recruited Biopsy specimens HPV serology Cells

Valid
No. PCR E6/E7 Valid PCR
% No. PCR- results, results, L1 results, No. No. results
Country participation Control OC OP obtained tested No. (%) No. (%) No. (%) obtained PCR-tested No. (%)

Case patients

Italy 92.0 77 55 106 101 89 (67.4) 129 (97.7) 129 (97.7) 131 46 42 (31.8)
Spain 76.5 287 72 265 243 216 (60.2) 333 (92.8) 331 (92.2) 338 145 140 (39.0)
Northern Ireland 85.0 60 10 38 35 32 (45.7) 61 (87.1) 61 (87.1) 69 29 29 (41.4)
Poland 96.0 113 8 113 102 90 (74.4) 101 (83.5) 102 (84.3) 121 — —
India 93.0 547 35 438 398 274 (47.1) 572 (98.3) 572 (98.3) 565 460 329 (56.5)
Cuba 99.0 150 47 174 154 104 (52.8) 190 (96.4) 185 (93.9) 191 31 29 (14.7)
Canada 82.6 43 14 57 40 35 (61.4) 49 (86.0) 43 (75.4) 57 — —
Australia 96.0 26 11 37 36 24 (64.9) 33 (89.2) 20 (54.1) 36 — —
Sudan NA 112 3 95 89 44 (38.3) 94 (81.7) 94 (81.7) 102 45 32 (27.8)
Total 88.7 1415 255 1323 1198 908 (54.4) 1562 (93.5) 1537 (92.0) 1610 756 601 (36.0)
Control subjects

Italy 96.0 148 146 (98.6) 146 (98.6) 148 51 47 (31.8)

Spain 91.0 375 303 (80.8) 301 (80.3) 367 117 114 (30.4)

Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015

Northern Ireland 49.0 50 50 (100.0) 50 (100.0) 50 15 15 (30.0)
Poland 93.0 124 112 (90.3) 104 (83.9) 123 — —
India 90.0 582 564 (96.9) 563 (96.7) 576 410 364 (62.5)
Cuba 81.0 200 196 (98.0) 191 (95.5) 196 23 23 (11.5)
Canada 82.2 42 42 (100.0) 41 (97.6) 42 — —
Australia 94.0 60 38 (63.3) 6 (10.0) 59 — —
Sudan NA 151 130 (86.1) 125 (82.8) 98 52 50 (33.1)
Total 87.3 1732 1581 (91.3) 1527 (88.2) 1659 668 613 (35.4)

*OC ⫽ cancer of the oral cavity; OP ⫽ cancer of the oropharynx; PCR ⫽ polymerase chain reaction; — ⫽ zero; NA ⫽ not ascertained.

Determination of Antibodies Against HPV16 E6 and E7 duplicate on different plates. A mean specific reactivity of
greater than 0.08 optical density units and a variation of the
Plasma antibodies against HPV16 E6 and E7 proteins were duplicate values of greater than 40% were set as cut points for
detected with an ELISA that used the glutathione S-transferase repeat analysis, again in duplicates. Median optical density val-
(GST) capture method with bacterially expressed full-length E6 ues were used for further analysis to reduce the impact of
or E7 fused to GST as the antigens, as described in detail extreme outliers.
previously (23). A preparatory experiment (Pawlita M: unpub- Cutoff determination for serum groups from individuals who
lished data) directly compared serum and plasma samples with did not have cancer from Germany, Tanzania, and Colombia
and without heat inactivation for 30 minutes at 56 °C from 20 (calculated as mean ⫹ 3 SDs, excluding positive outliers)
patients with cervical carcinoma and 20 healthy control subjects yielded cutoff values for HPV16 E6 and E7 between 0.06 and
and found that heat inactivation and the use of plasma instead of 0.09 optical density units. To reduce the influence of borderline
serum did not alter ELISA reactivity with HPV16 E6 and E7 positive sera, a stringent cutoff point of 0.16 optical density units
proteins or increase unspecific background. was arbitrarily defined for both HPV16 E6 and E7 ELISAs
In brief, 96-well Polysorb plastic plates (Nunc, Roskilde, before analysis of the data.
Denmark) were coated with glutathione-casein. After blocking
with unmodified casein (0.2% unmodified casein in PBS con-
taining 0.05% Tween 20), plates were incubated with 100 ␮L of Statistical Analysis
cleared lysate (diluted in casein-blocking buffer to a total lysate
protein concentration of 0.25 ␮g/␮L) from Escherichia coli Cancers of the oropharynx and tonsil were combined as
overexpressing the antigen as a GST fusion protein or GST alone cancers of the oropharynx, and cancers at other sites were
for background determination. Human plasma samples were combined as cancers of the oral cavity. When multiple topog-
heat inactivated and diluted 1 : 50 in blocking buffer containing raphies were reported, tonsil and oropharynx were classified as
total lysate protein from E. coli overexpressing GST without E6 oropharynx regardless of whether other locations were men-
or E7 sequences at 0.25 ␮g/␮L. Bound human antibodies were tioned. Because of a report (24) that combined sites in the
detected by donkey anti-human IgG (heavy and light chain) Waldeyer’s ring (oropharynx, tonsil, and base of the tongue), we
polyclonal antibody, which recognizes all classes of human considered adding base of the tongue to the oropharynx group,
immunoglobulins, conjugated to horseradish peroxidase (Di- but the HPV markers in our specimens from patients with cancer
anova, Hamburg, Germany). The absorbance from wells with of the base of the tongue were more similar to those from oral
GST alone defined the background reactivity of a human plasma cavity cancers than to those from oropharynx cancers, and
and was subtracted from the absorbance of wells with the therefore we kept our original classification.
GST-E6 or GST-E7 proteins to calculate the specific reactivity Odds ratios (ORs) and 95% confidence intervals (CIs) were
of a sample with the respective antigen. Interassay variation computed for all centers combined by using unconditional mul-
ranged from 0.12 to 0.23. Each plasma sample was tested in tivariable logistic regression models. All models included terms

Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003 ARTICLES 1775
for age, sex, country, smoking tobacco, paan chewing, and cigarettes smoked per day. A clear dose–response relation-
drinking alcohol, as appropriate. ship between risk for cancer and the number of years of
For case patients and control subjects with available serologic smoking was also found (data not shown). Similarly, number
biomarker data, the odds ratios were calculated as estimates of of alcoholic drinks per day (Table 2) and duration of alcohol
disease risk. For HPV DNA in biopsy specimens, we estimated drinking (data not shown) were associated with a dose-
the odds ratios of HPV detection in cancers of the oropharynx dependent increase in risk. Paan chewing, which could be
compared with cancers of the oral cavity. All models included evaluated only in India, was associated with a strong dose-
terms for study design variables (i.e., age group, country, and dependent increase in the risk for cancer of the oral cavity.
sex) and, when appropriate, smoking tobacco, chewing tobacco, We did not observe an association between sexual behavior
and drinking alcohol. The combined effect of HPV and smoking indicators, such as lifetime number of sexual partners and the
was assessed as a departure from a multiplicative logistic re- practice of oral sex, and overall risk for cancer of the oral
gression model. The method used to assess departure from a cavity or oropharynx.
multiplicative model was the inclusion in the logistic regression The prevalence of the different HPV markers (HPV DNA in
models of an interaction term. Because of the different method biopsy specimens from cancer patients, HPV DNA in cells, and
used to select control subjects in India and the large number of antibodies against HPV16 L1, E6, and E7 from case patients and
subjects recruited in that country, we repeated the main analyses control subjects) by topographic site of the primary tumor is
excluding the Indian centers and obtained similar results for all shown in Fig. 1. The tonsil was the site with the highest prev-
the analyses presented in the “Results” section. alence of all markers, followed in most instances by the oro-
pharynx. Control subjects had among the lowest positivities for

Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015

RESULTS most markers, except HPV DNA in exfoliated cells, which was
somewhat more common in control subjects than in case patients
A total of 1670 case patients and 1732 control subjects were with tumors of several oral cavity sites.
recruited (Table 1). Of the case patients, 1415 had cancers of the The association of various HPV markers with risk for cancer
oral cavity and 255 had cancers of the oropharynx. Participation of the oral cavity and oropharynx, determined in a multivariable
among case patients ranged from 76.5% in Spain to 99.0% in model, is shown in Table 3. HPV DNA was detected in biopsy
Cuba, for an overall participation rate of 88.7%; participation specimens of 3.9% (95% CI ⫽ 2.5% to 5.3%) of cancers of the
among control subjects ranged from 49.0% in Northern Ireland oral cavity (30 of 766 specimens with valid PCR results) and in
to 96.0% in Italy, for an overall participation rate of 87.3%. The biopsy specimens of 18.3% (95% CI ⫽ 12.0% to 24.7%) of
proportion of the total case patients recruited in each center who cancers of the oropharynx (26 of 142 specimens with valid PCR
had cancers of the oropharynx varied between 2.6% in Sudan results) (case– case comparison for cancers of the oropharynx/
and 41.7% in Italy. This difference did not reflect the actual cancers of the oral cavity: OR ⫽ 4.9, 95% CI ⫽ 2.6 to 9.1).
proportion occurring in each location but did reflect the different HPV16 was the only HPV type found in biopsies of 89.3% of
referral patterns of the participating hospitals. HPV DNA–positive case patients, and HPV16 was present with
Biopsy specimens were available for 79.2% (1323) of case HPV18 in an additional 5.4%, for a total of 94.7% HPV16
patients, of which 10% were not tested for HPV DNA because among positive case patients (data not shown). One case patient
they contained only necrotic tissue. Valid PCR biopsy results had HPV18 alone, one had HPV33 and HPV35, and one had
(␤-globin–positive) were available from 908 case patients HPV35 alone.
(54.4% of those recruited). When case patients with valid PCR In multivariable models, after adjustment for age, sex,
results were compared with case patients with missing results by country, drinking alcohol, smoking tobacco, and chewing
using a multivariable logistic regression model, no statistically tobacco, as appropriate, detection of HPV DNA in tumors
significant differences were found in age, sex, smoking, alcohol was not statistically significantly different by age, sex, coun-
consumption and, in India, paan chewing habits (data not try, or alcohol use (data not shown). However, HPV DNA
shown). However, case patients without valid PCR results were was detected less frequently in biopsy specimens from ex-
more likely to be from India, Cuba, Sudan, and Australia—the smokers (OR for detection of HPV DNA ⫽ 0.6, 95% CI ⫽ 0.2
areas with the most challenging storage and transportation to 1.5) and current smokers (OR ⫽ 0.4, 95% CI ⫽ 0.2 to 0.9)
conditions. than in biopsy specimens from nonsmokers. In India, HPV
HPV16 E6 and E7 serology results were available for 93.5% DNA was detected less frequently in tumor specimens from
of case patients and 91.3% of control subjects. HPV16 L1 VLP tobacco chewers (OR for HPV detection ⫽ 0.5, 95% CI ⫽ 0.1
serology results were available for 92.0% of case patients and to 2.0) than in those from non-chewers. Case patients report-
88.2% of control subjects. Exfoliated cells were obtained from ing more than one sexual partner in their lifetime were more
more than 95% of case patients and control subjects, but testing likely to have HPV DNA in their tumors than those reporting
was limited to 43.6% of these subjects, as discussed above. only one sexual partner (OR for HPV detection ⫽ 2.4, 95%
Cancer stage was ascertained for 1558 (93.3%) of participating CI ⫽ 1.0 to 5.7), as were those reporting a history of oral sex
case patients and was distributed as follows: stage I, 14.7%; compared with those without such a history (OR for HPV
stage II, 18.6%; stage III, 29.1%; and stage IV, 37.4%. detection ⫽ 3.2, 95% CI ⫽ 1.5 to 6.4). The association of
The distribution of case patients and control subjects by HPV DNA with sexual behavior was similar for cancers of
selected risk factors and odds ratios associated with selected the oral cavity and oropharynx, although some of the esti-
variables is shown in Table 2. As expected, smoking was a mates were not statistically significant (data not shown).
strong risk factor for cancer of the oral cavity and, most The prevalence of HPV DNA detected in exfoliated cells was
notably, cancer of the oropharynx, with a clear dose–response 4.7% (95% CI ⫽ 2.9% to 6.5%) for patients with cancer of the
relationship between cancer risk and increasing number of oral cavity, 8.9% (95% CI ⫽ 3.0% to 14.8%) for patients with

1776 ARTICLES Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003
Table 2. Characteristics of case patients and control subjects and risks associated with tobacco smoking, paan chewing, alcohol drinking, and sexual behavior*

Case patients
No. of Cancer of the oral cavity Cancer of the oropharynx
control
Characteristic subjects (%) No. (%) OR† (95% CI) No. (%) OR† (95% CI)

Sex
Male 1078 (62.2) 880 (62.2) 214 (83.9)
Female 654 (37.8) 535 (37.8) 41 (16.1)

Age, y
ⱕ44 342 (19.7) 177 (12.5) 26 (10.2)
45–54 417 (24.1) 319 (22.5) 59 (23.1)
55–64 485 (28.0) 422 (29.8) 90 (35.3)
ⱖ65 488 (28.2) 497 (35.1) 80 (31.4)

No. of cigarettes per day‡

Never 891 (51.8) 542 (38.5) 1.0 (referent) 23 (9.3) 1.0 (referent)
1–5 106 (6.2) 71 (5.0) 1.4 (0.9 to 2.0) 8 (3.2) 2.4 (1.0 to 5.7)
6–10 161 (9.4) 121 (8.6) 1.9 (1.4 to 2.6) 25 (10.1) 4.9 (2.6 to 9.2)
11–15 96 (5.6) 93 (6.6) 2.9 (2.0 to 4.1) 25 (10.1) 7.0 (3.6 to 13.4)
16–20 186 (10.8) 208 (14.8) 3.7 (2.8 to 5.0) 46 (18.6) 6.9 (3.8 to 12.5)
21–30 157 (9.1) 207 (14.7) 4.5 (3.3 to 6.1) 62 (25.1) 13.2 (7.4 to 23.6)
⬎30 124 (7.2) 167 (11.9) 4.7 (3.4 to 6.6) 58 (23.5) 15.6 (8.5 to 28.8)

Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015

No. of alcoholic drinks per day§
Never drinker 590 (43.0) 388 (35.3) 1.0 (referent) 38 (15.9) 1.0 (referent)
ⱕ4 605 (44.1) 438 (39.8) 1.3 (1.0 to 1.6) 87 (36.4) 1.2 (0.7 to 1.9)
5–6 54 (3.9) 79 (7.2) 2.7 (1.7 to 4.2) 24 (10.0) 2.8 (1.4 to 5.6)
7–10 58 (4.2) 87 (7.9) 2.9 (1.9 to 4.5) 41 (17.2) 4.8 (2.5 to 9.2)
11–15 31 (2.3) 51 (4.6) 3.6 (2.1 to 6.1) 25 (10.5) 6.1 (2.9 to 12.9)
⬎15 33 (2.4) 57 (5.2) 3.3 (2.0 to 5.5) 24 (10.0) 6.0 (2.9 to 12.3)

No. of paan chewing units per day㛳

Never chewer 483 (84.0) 139 (25.7) 1.0 (referent) 27 (81.8) 1.0 (referent)
1–5 65 (11.3) 242 (44.8) 14.5 (10.1 to 20.7) 3 (9.1) 0.6 (0.2 to 2.1)
6–10 19 (3.3) 116 (21.5) 25.8 (15.1 to 44.4) 2 (6.1) 2.0 (0.4 to 10.0)
⬎10 8 (1.4) 43 (8.0) 21.7 (9.7 to 48.5) 1 (3.0) 4.5 (0.5 to 43.7)

Lifetime No. of sexual partners¶

0–1 936 (60.9) 807 (63.0) 1.0 (referent) 82 (35.7) 1.0 (referent)
2–5 256 (16.6) 235 (18.3) 0.9 (0.7 to 1.2) 54 (23.5) 1.1 (0.7 to 1.7)
6–10 119 (7.7) 88 (6.9) 0.7 (0.5 to 1.0) 28 (12.2) 1.0 (0.6 to 1.8)
11–20 88 (5.7) 75 (5.9) 0.8 (0.5 to 1.2) 26 (11.3) 1.1 (0.6 to 2.0)
21–50 90 (5.9) 44 (3.4) 0.5 (0.3 to 0.7) 23 (10.0) 1.2 (0.7 to 2.2)
51–100 27 (1.8) 11 (0.9) 0.4 (0.2 to 0.8) 7 (3.0) 1.0 (0.4 to 2.5)
⬎100 22 (1.4) 21 (1.6) 0.8 (0.4 to 1.6) 10 (4.3) 1.6 (0.7 to 3.9)

Frequency of oral sex#

No oral sex 1063 (75.7) 881 (78.0) 1.0 (referent) 141 (66.2) 1.0 (referent)
Occasionally 216 (15.4) 143 (12.7) 0.9 (0.7 to 1.1) 46 (21.6) 0.8 (0.5 to 1.2)
Often 67 (4.8) 81 (7.2) 1.5 (1.0 to 2.1) 14 (6.6) 0.8 (0.4 to 1.5)
Most times 58 (4.1) 25 (2.2) 0.6 (0.4 to 1.1) 12 (5.6) 0.8 (0.4 to 1.7)

*OR ⫽ odds ratio; CI ⫽ confidence interval.

†ORs adjusted for age, sex, center, smoking tobacco, chewing tobacco, and drinking alcohol, as appropriate.
‡Unknowns excluded (11 control subjects, 14 case patients).
§Excludes subjects from Bangalore, India, and Sudan from whom information on amount of alcohol consumed was not obtained. Unknowns excluded (10 control
subjects, 17 case patients).
㛳Restricted to Indian subjects. Unknowns excluded (seven control subjects, nine case patients).
¶Excludes subjects from Northern Ireland from whom information on sexual behavior was not obtained and unknowns (144 control subjects and 89 case patients).
#Excludes subjects from Northern Ireland from whom information on sexual behavior was not obtained and other unknowns (278 control subjects and 257 case
patients).

cancer of the oropharynx, and 6.9% (95% CI ⫽ 4.9% to 8.9%) the oral cavity (OR ⫽ 1.5, 95% CI ⫽ 1.1 to 2.1), and 13.4% of
for control subjects (confidence intervals for prevalence not case patients with cancers of the oropharynx (OR ⫽ 3.5, 95% CI
shown in tables). No statistically significant association between ⫽ 2.1 to 5.9). A statistically significant association between
HPV DNA and risk for either type of cancer was noted in the antibodies against HPV16 L1 and risk was also found in case–
case– control comparisons. In the case– case comparison, the case comparisons (OR ⫽ 1.9, 95% CI ⫽ 1.2 to 3.0) for cancer
odds ratio for HPV DNA positivity in cells was 2.0 (95% CI ⫽ of the oropharynx compared with cancer of the oral cavity.
0.7 to 5.1) for cancers of the oropharynx compared with that for Antibodies against HPV16 E6 were detected in 1.1% of
cancer of the oral cavity. control subjects, 2.6% of cancers of the oral cavity (OR ⫽ 2.6,
Antibodies against HPV16 L1 were detected in plasma from 95% CI ⫽ 1.4 to 5.0), and 9.9% of cancers of the oropharynx
6.0% of control subjects, 8.9% of case patients with cancers of (OR ⫽ 9.9, 95% CI ⫽ 4.7 to 20.7). A statistically significant

Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003 ARTICLES 1777
 Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015
Fig. 1. Prevalence of human papillomavirus (HPV) markers among control subjects limited for HPV DNA in cells. A) Prevalence of HPV DNA detection in biopsy
and by cancer subsite. CON ⫽ control subjects; BOT ⫽ base of the tongue; TON specimens by cancer subsite (specimens not available from control subjects). B)
⫽ other parts of the tongue; GUM ⫽ gum; FLO ⫽ floor of the mouth; PAL ⫽ Prevalence of HPV DNA detection in exfoliated cells among control subjects and by
palate; MOU ⫽ mouth; TNS ⫽ tonsil; ORO ⫽ oropharynx. Bars indicate 95% cancer subsite. C) Prevalence of antibodies against HPV16 L1 virus-like particles
confidence intervals. Estimates presented for each of the markers are based on all among control subjects and by cancer subsite. D) Prevalence of antibodies against
subjects tested for that marker. The number of subjects tested was particularly HPV16 E6 or E7 among control subjects and by cancer subsite.

association between HPV16 E6 and risk was also found in imens]). A strong association between detection of both HPV16
case– case comparisons (OR ⫽ 4.4, 95% CI ⫽ 2.3 to 8.2) for E6 and E7 antibodies and risk for cancer of the oropharynx was
cancer of the oropharynx compared with cancer of the oral found (OR ⫽ 67.1, 95% CI ⫽ 12.9 to 348.2). Accordingly, a
cavity, similar to that for HPV DNA in biopsy specimens. strong association was detected in the case– case comparison for
Similar patterns were observed for antibodies against HPV16 E7 cancer of the oropharynx compared with cancer of the oral
(Table 3). Thus, antibodies against either HPV16 E6 or E7 or cavity (OR ⫽ 13.5, 95% CI ⫽ 4.4 to 41.1). These associations
both were associated with the risk for cancers of the oral cavity were similar in both sexes, in different regions, and in two age
(OR ⫽ 2.9, 95% CI ⫽ 1.7 to 4.8) and the oropharynx (OR ⫽ 9.2, strata (⬍60 years old and ⱖ60 years old).
95% CI ⫽ 4.8 to 17.7). HPV DNA was detected at a higher level in stages III–IV
Among control subjects, we found no statistically significant cancers of the oropharynx (21.5%) than in stages 0 –II (7.4%) (P
association between HPV immunologic biomarkers (antibodies ⫽ .09). Antibodies against HPV16 E6 or E7 were also more
against HPV16 L1, E6, or E7) and age, sex, smoking, drinking, commonly detected in specimens from more advanced stage
or sexual behavior (data not shown). Prevalence of HPV16 L1 cancers of the oropharynx than in specimens from earlier stage
antibodies among control subjects varied between countries cancers of the oropharynx (14.7% of the 184 case patients at
from 0% in Ireland, Poland, and Canada to 10.5% in Cuba; stages III–IV versus 2.4% for the 41 case patients at stages 0 –II;
antibodies against HPV16 E6 or E7 were found infrequently in P ⫽ .03), but antibodies against HPV16 E6 or E7 did not vary
the same countries where HPV16 L1 antibodies were infrequent by stage of cancer of the oral cavity (data not shown). Antibod-
and were most frequently found among control subjects in ies against HPV16 L1 did not vary with the stage of either cancer
Australia (10.5%, n ⫽ 60). examined in this study.
Antibodies against both HPV16 E6 and E7 were rarely de- We next analyzed how frequently antibodies against HPV16
tected in control subjects (0.1% of control subjects or 7.7% of L1, E6, and E7 were detected in case patients with biopsy
those with detectable antibodies against HPV16 E6 or E7) and in specimens in which HPV DNA was or was not detected (Table
case patients with cancers of the oral cavity (0.5% [9.8% of 4). Antibodies against HPV16 L1, E6, and E7 were more com-
positive specimens]) but were more common in case patients mon in case patients whose biopsy specimens were positive for
with cancers of the oropharynx (6.6% [55.2% of positive spec- HPV DNA than in those whose biopsy specimens were negative

1778 ARTICLES Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003
 Table 3. Risk associated with selected human papillomavirus (HPV) markers and topographic sites*

Cancer of the oral cavity: Cancer of the oropharynx: Cancer of the oropharynx vs.
case patients vs. case patients vs. cancer of the oral cavity in
control subjects control subjects case patients
Control subjects,
HPV marker No. (%) No. (%) OR (95% CI) No. (%) OR (95% CI) OR (95% CI)

HPV DNA in biopsy specimens

Negative 736 (96.1) 116 (81.7)
Positive 30 (3.9) 26 (18.3) 4.9 (2.6 to 9.1)

HPV DNA in cells

Negative 571 (93.1) 487 (95.3) 1.0 (referent) 82 (91.1) 1.0 (referent) 1.0 (referent)
Positive 42 (6.9) 24 (4.7) 0.6 (0.3 to 1.1) 8 (8.9) 1.0 (0.4 to 2.5) 2.0 (0.7 to 5.1)

HPV16 L1 antibodies
Negative 1436 (94.0) 1184 (91.1) 1.0 (referent) 206 (86.6) 1.0 (referent) 1.0 (referent)
Positive 91 (6.0) 115 (8.9) 1.5 (1.1 to 2.1) 32 (13.4) 3.5 (2.1 to 5.9) 1.9 (1.2 to 3.0)

HPV16 E6 antibodies
Negative 1563 (98.9) 1285 (97.4) 1.0 (referent) 219 (90.1) 1.0 (referent) 1.0 (referent)
Positive 18 (1.1) 34 (2.6) 2.6 (1.4 to 5.0) 24 (9.9) 9.9 (4.7 to 20.7) 4.4 (2.3 to 8.2)

HPV16 E7 antibodies

Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015

Negative 1571 (99.4) 1286 (97.5) 1.0 (referent) 222 (91.4) 1.0 (referent) 1.0 (referent)
Positive 10 (0.6) 33 (2.5) 3.4 (1.6 to 7.3) 21 (8.6) 19.0 (7.5 to 47.8) 4.1 (2.1 to 8.1)

HPV16 E6 or E7 antibodies
Negative 1555 (98.4) 1258 (95.4) 1.0 (referent) 214 (88.1) 1.0 (referent) 1.0 (referent)
One positive 24 (1.5) 55 (4.2) 2.7 (1.6 to 4.7) 13 (5.3) 4.5 (2.0 to 10.1) 1.9 (0.9 to 3.7)
Either or both positive 26 (1.6) 61 (4.6) 2.9 (1.7 to 4.8) 29 (11.9) 9.2 (4.8 to 17.7) 3.3 (1.9 to 5.6)
Both positive 2 (0.1) 6 (0.5) 4.3 (0.8 to 23.2) 16 (6.6) 67.1 (12.9 to 348.2) 13.5 (4.4 to 41.1)

*Models adjusted for country, sex, age group, smoking tobacco, drinking alcohol, and paan chewing. Subjects with unknown values were excluded. Values differ
from those expected because categories are not mutually exclusive. OR ⫽ odds ratio; CI ⫽ confidence interval.

for HPV DNA. This association was particularly strong for who had both measurements (␬ ⫽ 0.059, 95% CI ⫽ ⫺0.035 to
cancers of the oropharynx, in which HPV16 L1 antibodies were 0.281; Table 5). Ninety percent of the patients with biopsy
present in 52% and HPV16 E6 or E7 antibodies were present in specimens that were positive for HPV DNA had exfoliated cells
65.4% of case patients with biopsies positive for HPV DNA. In in which HPV DNA was not detected.
case patients with cancer of the oral cavity, there was very little Associations between cancer and the combined presence of
correlation between detection of HPV16 DNA in biopsies and HPV16 L1 VLP antibodies and a history of smoking and/or
detection of HPV16 E6 or E7 antibodies (␬ ⫽ 0.080, 95% CI ⫽ chewing tobacco are shown in Table 6. When compared with
0.001 to 0.226). For cancer of the oropharynx, the correlation never smokers and/or never chewers who were negative for
was much stronger (␬ ⫽ 0.6, 95% CI ⫽ 0.4 to 0.8). Correspond- HPV16 L1 VLP, smokers who were negative for HPV16 L1
ing sensitivities (using DNA in biopsy specimens as the gold VLP (OR ⫽ 6.6, 95% CI ⫽ 5.3 to 8.2) and, to a lesser extent,
standard) were 14% (95% CI ⫽ 4.7% to 31.9%) for oral cavity never smokers who were positive for HPV16 L1 VLP (OR ⫽
and 64% (95% CI ⫽ 46.5% to 77.1%) for oropharynx, with 1.3, 95% CI ⫽ 0.7 to 2.3), and smokers who were positive for
specificities close to 95% for both. HPV16 L1 VLP (OR ⫽ 11.4, 95% CI ⫽ 7.4 to 17.6) had an
Little correlation was observed between the detection of HPV increased risk for cancer of the oral cavity (P for departure from
DNA in exfoliated cells and the detection of HPV DNA in multiplicative model ⫽ .461). When compared with never
biopsy specimens for the limited number of subjects (n ⫽ 349) smokers who were negative for HPV16 L1 VLP, smokers who

Table 4. Prevalence of immunoglobulin G (IgG) antibodies against HPV16 E6 or E7 and HPV16 L1 by HPV DNA status among case patients*

HPV DNA–negative HPV DNA–positive HPV DNA unknown

Oral cavity, Oropharynx, Oral cavity, Oropharynx, Oral cavity, Oropharynx,
Antibody No. (%) No. (%) No. (%) No. (%) No. (%) No. (%)

HPV16 L1
Negative 617 (89.9) 100 (92.6) 22 (73.3) 12 (48.0) 545 (93.5) 94 (89.5)
Positive 69 (10.1) 8 (7.4) 8 (26.7) 13 (52.0) 38 (6.5) 11 (10.5)
P* .39 .05 .15

HPV16 E6 or E7
Negative 672 (96.1) 107 (96.4) 26 (86.7) 9 (34.6) 560 (94.9) 98 (92.5)
Either or both 27 (3.9) 4 (3.6) 4 (13.3) 17 (65.4) 30 (5.1) 8 (7.5)
positive
P* .89 ⬍.001 .30

*Two-sided ␹2 test.

Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003 ARTICLES 1779
Table 5. Agreement between human papillomavirus (HPV) DNA detection in never smokers who were positive for HPV16 E6 or E7 (OR ⫽
biopsies and cells among cancer case patients* 64.5, 95% CI ⫽ 18.3 to 226.7), and smokers who were positive
HPV DNA in cells
for HPV16 E6 or E7 (OR ⫽ 56.2, 95% CI ⫽ 22.5 to 140.4) had
an increased risk for cancer of the oropharynx. The odds ratio for
HPV DNA No. No. Total
in biopsy negative positive No.
HPV-positive smokers was again much lower than that expected
under a multiplicative model (P for departure from multiplica-
No. negative 314 15 329 tive model ⫽ .001), in a pattern typical of an additive effect.
No. positive 18 2 20
Total No. 332 17 349
DISCUSSION
*␬ ⫽ 0.059 (95% confidence interval [CI] ⫽ ⫺0.035 to 0.281). Sensitivity ⫽
10% (95% CI ⫽ 1.9% to 29.5%). Specificity ⫽ 95.4% (95% CI ⫽ 94.9% to
This large case– control study evaluated the association be-
96.6%).
tween five markers of HPV infection and cancers of the oral
cavity and the oropharynx (HPV DNA in biopsy specimens,
were negative for HPV16 L1 VLP (OR ⫽ 9.2, 95% CI ⫽ 4.9 to HPV DNA in exfoliated cells, and antibodies against HPV16 L1,
17.1), never smokers who were positive for HPV16 L1 VLP (OR E6, and E7). We recruited more than 1600 case patients and
⫽ 6.7, 95% CI ⫽ 2.3 to 19.1), and smokers who were positive 1700 control subjects in nine countries and used similar proto-
for HPV16 L1 VLP (OR ⫽ 26.6, 95% CI ⫽ 11.9 to 59.4) had an cols for data and specimen collection and centralized HPV
increased risk for cancers of the oropharynx (P for departure testing. Our results indicate that HPV appears to play a definite
from multiplicative model ⫽ .181). etiologic role in a substantial fraction of cancers of the orophar-

Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015

Associations between the risk for cancer and the combined ynx and possibly in a small subgroup of cancers of the oral
presence of HPV16 E6 or E7 antibodies and a history of tobacco cavity. Nonsmokers are more likely than smokers to have HPV-
smoking and/or paan chewing are also shown in Table 6. When related tumors.
compared with never smokers/never chewers who were negative HPV DNA was detected in tumor biopsy specimens from
for HPV16 E6 and E7, smokers/chewers who were negative for 18.3% of case patients with cancer of the oropharynx and from
HPV16 E6 and E7 (OR ⫽ 6.7, 95% CI ⫽ 5.4 to 8.4), never 3.9% of patients with cancer of the oral cavity, which is an
smokers/never chewers who were positive for HPV16 E6 or E7 adjusted fivefold increase in the odds of HPV detection in case
(OR ⫽ 6.7, 95% CI ⫽ 2.6 to 17.3), and smokers who were patients with cancer of the oropharynx compared with that in
positive for HPV16 E6 or E7 (OR ⫽ 13.0, 95% CI ⫽ 7.2 to 23.5) case patients with cancer of the oral cavity. Among the topo-
had an increased risk for cancer of the oral cavity. The odds ratio graphic sites examined, HPV was most frequently detected in
for cancer of the oral cavity for HPV-positive smokers was much cancer of the tonsil (24.7% of specimens). These prevalences are
lower than that expected under a multiplicative model (P for consistent with, albeit somewhat lower than, those in previous
departure from multiplicative model ⫽ .03), in a typical pattern reports [for review, see (2,25)]. Similar to the results of other
for an additive effect. When compared with never smokers who studies (9 –11), the vast majority (95%) of HPV-positive case
were negative for HPV16 E6 and E7, smokers who were nega- patients in this study had HPV16, the most common type in
tive for HPV16 E6 and E7 (OR ⫽ 11.2, 95% CI ⫽ 5.9 to 21.4), genital cancers. The exclusive detection of HPV DNA by PCR

Table 6. Risk associated with smoking or chewing tobacco combined with detection of antibodies against human papillomavirus 16 (HPV16) L1 virus-like
particles (VLPs) or combined with detection of HPV16 E6 or E7 antibodies*

Tobacco smoking or paan chewing

Never Ever
No. of case No. of case
patients/No. patients/No.
of control of control
OR (95% CI) subjects OR (95% CI) subjects

HPV16 L1 VLP antibodies

Cancer of the oral cavity: case patients vs. control subjects
Negative 1.0 (referent) 176/672 6.6 (5.3 to 8.2) 1002/761
Positive 1.3 (0.7 to 2.3) 18/53 11.4 (7.4 to 17.6) 95/38
P for departure from multiplicative model ⫽ .461
Cancer of the oropharynx: case patients vs. control subjects
Negative 1.0 (referent) 12/672 9.2 (4.9 to 17.1) 194/761
Positive 6.7 (2.3 to 19.1) 6/53 26.6 (11.9 to 59.4) 26/38
P for departure from multiplicative model ⫽ .181
HPV16 E6 or E7 positivity
Cancer of the oral cavity: case patients vs. control subjects
Both negative 1.0 (referent) 187/740 6.7 (5.4 to 8.4) 1063/812
Either or both positive 6.7 (2.6 to 17.3) 12/8 13.0 (7.2 to 23.5) 49/18
P for departure from multiplicative model ⫽ .030
Cancer of the oropharynx: case patients vs. control subjects
Both negative 1.0 (referent) 11/740 11.2 (5.9 to 21.4) 203/812
Either or both positive 64.5 (18.3 to 226.7) 7/8 56.2 (22.5 to 140.4) 22/18
P for departure from multiplicative model ⫽ .001

*Models adjusted for country, age, sex, and alcohol consumption status. OR ⫽ odds ratio; CI ⫽ confidence interval.

1780 ARTICLES Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003
may, however, lead to overestimation of the number of case was some degree of adjustment for smoking by the use of
patients in which the virus is etiologically involved, as suggested cotinine levels in that study (13), the authors could not account
by the lower proportion of case patients in whom HPV16 E6 for other confounding factors. Schwartz et al. (11) also investi-
and/or E7 antibodies could also be detected. gated the association of antibodies against L1 with oral cancer
As reported by Gillison et al. (9), we detected HPV DNA and detected a twofold increased risk for all tumors and an
statistically significantly less often among tobacco smokers almost sevenfold increased risk for HPV16-positive cancers.
and/or chewers than among nonsmokers and/or non-chewers. When interpreting associations between the risk for cancer and
HPV was detected more commonly in biopsy specimens from the presence of these antibodies, it should be noted that they
cancer patients with more than one sexual partner and from those reflect infection in all susceptible mucosal sites, which may
who practiced oral sex than in biopsy specimens from those with introduce additional sources of confounding. Moreover, not all
fewer than two partners or who did not engage in oral sex, individuals exposed to HPV seroconvert or maintain detectable
suggesting the possibility of sexual transmission. Schwartz et al. antibody levels over time.
(11) also found an association between sexual behavior and risk HPV16 E6 and E7 antibodies are generally considered mark-
for HPV16-positive cancers for both sexes combined. However, ers of invasive HPV16-transformed tumors, possibly generated
much remains to be learned about the transmission of HPV in after antigen exposure, after development of a tumor vascular
normal subjects and the natural history of HPV infection in the bed, or after necrosis has occurred (26,27). A recent prospective
oral cavity (14). study (28) conducted to evaluate the predictive value of E6 and
A relatively high percentage (24.2%) of case patients who E7 antibodies as markers of early cervical cancer demonstrated
were tested by PCR had invalid results because of the absence of that only 7% of women who developed cervical cancer in the

Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015

adequate material (i.e., they were ␤-globin–negative). This prob- 0.5–20 years after blood was drawn had detectable levels of E6
lem was more common in case patients from India, Sudan, Cuba, or E7 antibodies and had a modest increased risk for later
and Australia, probably as a consequence of difficulties with developing cervical cancer compared with women without such
storing and shipping specimens. However, case patients with antibodies. However, some degree of expression of these onco-
valid results and without such results did not differ substantially proteins is probably present also in HPV infections and dysplas-
in their risk factor profile. tic lesions. In fact, a recent study (29) reported relatively high
HPV DNA in exfoliated cells was not detected more fre- prevalence of these antibodies in patients with evidence of HPV
quently in case patients than in control subjects. The poor
infection but not cervical cancer. It can be argued that if these
association between HPV DNA test results from exfoliated cells
antibodies are markers of invasion, their use as markers of
and those from biopsy specimens probably indicates that the
exposure in control subjects may overestimate the risk. How-
HPV status of exfoliated cells does not accurately reflect that of
ever, such antibodies can be detected in a few control subjects,
tumors. HPV was not detected in exfoliated cells from 90% of
and they can be helpful in identifying HPV-related cancers in
the case patients with HPV-positive biopsy specimens. This
tumor sites with heterogeneous etiologies. In our study, these
result is in concordance with those of Schwartz et al. (11),
antibodies were strongly associated with cancer of the orophar-
although their specimens were collected after treatment. In con-
ynx, with odds ratios near 10 for antibodies against HPV16 E6,
trast, Smith et al. (10) detected a threefold increased risk for oral
cavity and oropharyngeal cancer associated with the detection of 20 for antibodies against HPV16 E7, and almost 70 for the
HPV DNA in exfoliated cells before treatment. Thus, the appro- combination of the two. Conversely, risk for cancer of the oral
priate marker of HPV exposure measurable in case patients and cavity was increased about threefold when HPV16 E6 or E7
control subjects remains elusive, particularly given the biologic antibodies were detected, and it was not much higher when both
diversity of individual topographic sites in the oral cavity (14). were detected.
Using several different HPV markers (HPV DNA in biopsy A high prevalence of anti-HPV16 E6 and E7 antibodies is
specimens, HPV DNA in exfoliated cells, and antibodies against observed in advanced cervical cancer (30). In our study, 65% of
HPV16 L1, E6, and E7), we found a strong association between case patients with HPV16 DNA–positive cancers of the orophar-
HPV and cancer of the oropharynx and also some association ynx had antibodies against HPV16 E6 or E7, making these
between HPV and cancer of the oral cavity. The latter associa- proteins potentially useful and minimally invasive markers of
tion could be partly explained by misclassification of the tumor HPV-related tumors when appropriate cytologic or histologic
site if some HPV-positive cancers of the oropharynx were clas- specimens are not available. The lower antibody prevalence in
sified erroneously as cancers of the oral cavity, but it could also HPV DNA–positive cancers of the oral cavity than in cancers of
indicate that a small fraction of oral cavity lesions are the oropharynx might indicate a noncausal association between
HPV-related. HPV and cancer of the oral cavity, or it may reflect lower viral
Although antibodies against HPV16 L1 are relatively insen- oncogene expression, less invasive growth, or as yet unidentified
sitive markers of cumulative exposure to the virus, such anti- site-dependent differences in antibody response to the HPV
bodies were associated with a 1.5-fold increased risk for cancer oncoproteins.
of the oral cavity and a 3.5-fold increased risk for cancer of the Given that the different HPV markers are probably measuring
oropharynx, with a 2-fold increased risk for cancer of the oro- different aspects of HPV infection and that each has different
pharynx compared with that of the oral cavity. In a recent sampling and technical limitations, none of these risk estimates
prospective study (13), a twofold increased risk was observed can ultimately quantify the association between HPV and cancer
for head and neck tumors in subjects with detectable antibodies of the oral cavity or oropharynx. The strength of our findings is
against HPV16 L1 or L2 proteins and, similar to our findings, the consistency in the direction of the associations with all the
risk was highest for cancer of the tonsil and oropharynx with a markers, which were measured independently by investigators
smaller increased risk for cancer of the tongue. Although there masked to case– control status, topography, and other variables.

Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003 ARTICLES 1781
 We investigated the combined effect of tobacco use and HPV Lyon, France; Eduardo Franco, McGill University, Montreal, Quebec,
infection by using antibodies against HPV16 L1 VLPs and Canada; Parviz Ghadirian, Université de Montréal, Montreal; Carmen
antibodies against HPV16 E6 and E7. For the latter, the risks Martinez, Escuela Andaluza de Salud Publica, Granada, Spain; Angel
appeared to be additive, indicating the absence of synergism Rollón, Jose Ramón Armas, Hospital Virgen Macarena, Seville, Spain;
Philip Lamey, Ruth Leathem, Queen’s University Belfast, Northern
between tobacco use and HPV. The exclusion of case patients
Ireland, United Kingdom; Juan Lence, Rosa Maria Ortiz, Teresa Cruz,
from India only strengthened the additivity of the effects of María de los Angeles Ríos, Instituto Nacional de Oncologia y Radio-
tobacco and HPV (data not shown). Schwartz et al. (11) detected biologia, Havana, Cuba; Maria Jesús Quintana, Miquel Quer, Hospital
a synergistic (i.e., multiplicative) effect of smoking and HPV, as de Sant Pau, Barcelona, Spain; Antoni Monner, Amparo Juan, Marta
measured by antibodies against HPV16 L1 VLPs. In our study, Carrera, Ciutat Sanitària i Universitària de Bellvitge, Bellvitge, Spain;
the combined effects of HPV16 L1 VLP antibodies and smoking Gina Albero, Institut Català d’Oncologia, Barcelona, Spain; Paul van
did not show a statistically significant departure from the mul- der Valk, Vrije Universiteit Medical Center, Amsterdam, The Nether-
tiplicative model. However, small numbers may have hampered lands; Kamal H. Mohamed, Toombak and Smoking Research Center,
our ability to clarify the type of interaction, because we could Khartoum, Sudan; Juan Ramon Delgado, Adoración Martínez, Hospital
not restrict the analysis to heavy smokers. The presence of Virgen de las Nieves, Granada, Spain; Peter Sehr, Deutsches Krebsfor-
additive rather than multiplicative risks between HPV and smok- schungszentrum, Heidelberg, Germany; Janusz Piekarczyk, Danuta
Samolczyk-Wanyura, Agnieszka Pilarska, Pawel Pilarski, 2nd, Maxil-
ing/chewing tobacco use would suggest that these factors oper-
lofacial Surgery Clinic, Medical Academy, and Witold Zatonski, Can-
ate, in part, at the same step of multistage carcinogenesis in the cer Center, Warsaw, Poland; Christopher O’Brien, Sydney Head and
oral cavity and oropharynx (e.g., p53 inactivation). Still, HPV Neck Cancer Institute, Royal Prince Alfred Hospital, Sydney, Australia;
infection appears to contribute to an increased risk for cancer of Silvana Pilotti, Istituto Nazionale per la Cura e lo Studio dei Tumori,

Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015

the oral cavity and oropharynx also among tobacco smokers and Milan, Italy; Rudrapatna Jayshree Subramanyam, Sakaleshpur Veer-
chewers. appaiah Kumaraswamy, Geetashree Mukherjee, Kidwai Memorial In-
In addition to mounting epidemiologic evidence, extensive stitute of Oncology, Bangalore, India.
laboratory evidence (14) supports the association between HPV
and a subset of cancers of the oropharynx. HPV16 E6 protein REFERENCES
inactivates p53 protein, and HPV-associated tumors of the cer- (1) Bray F, Ferlay J, Parkin DM, Pisani P. Globocan 2000: cancer incidence,
vix and other anogenital sites [for review, see (31)] do not mortality and prevalence worldwide [CD-ROM]. Lyon (France): Interna-
usually have TP53 mutations. Tumors at those locations that tional Agency for Research on Cancer, 2001.
contain HPV DNA contain TP53 mutations less frequently than (2) Franceschi S, Bidoli E, Herrero R, Muñoz N. Comparison of cancers of the
HPV-negative tumors, and tumors that express HPV16 E6 or E7 oral cavity and pharynx worldwide: etiological clues. Oral Oncol 2000;36;
usually do not have a mutated TP53 gene (32,33). HPV- 106 –15.
(3) Castellsagué X, Quintana MJ, Martínez MC, Nieto A, Sánchez MJ,
associated cancers of the oral cavity and oropharynx are more
Amparo J, et al. The role of type of tobacco and type of alcoholic beverage
likely to have basaloid histologic features (9,33), an aspect we in oral carcinogenensis. Int J Cancer. In press 2003.
did not evaluate. Viral E6 and/or E7 transcripts and/or viral (4) Garrote LF, Herrero R, Ortiz Reyes RM, Vaccarella S, Anta JL, Ferbeye L,
integration have been detected more frequently in tumors con- et al. Risk factors for cancer of the oral cavity and oro-pharynx in Cuba.
taining HPV16 or HPV33 DNA (34,35), and both in situ hy- Br J Cancer 2001;85:46 –54.
bridization studies and microdissection studies have detected (5) Sanchez MJ, Martinez C, Nieto A, Castellsagué X, Quintana MJ, Bosch
viral DNA or RNA in the tumor but not in the stroma (36). In FX, et al. Oral and oropharyngeal cancer in Spain: influence of dietary
patterns. Eur J Cancer Prev 2003;12:49 –56.
addition, a cell line with integrated HPV16 was established from (6) Talamini R, Vacarella S, Barbone F, Tavani A, La Vecchia C, Herrero R,
an HPV16-containing oral carcinoma (37). Thus, the laboratory et al. Oral hygiene, dentition, sexual habits and risk of oral cancer. Br J
and epidemiologic data indicate that HPV may be a causal factor Cancer 2000;83:1238 – 42.
in a subset of oral and oropharyngeal carcinomas and may (7) Bosch FX, Lorincz A, Muñoz N, Meijer CJ, Shah KV. The causal relation
operate through the same mechanisms described in the patho- between HPV and cervical cancer. J Clin Pathol 2002;55:244 – 65.
genesis of all cervical carcinomas. (8) Shah KV. Do human papillomavirus infections cause oral cancer? J Natl
Cancer Inst 1998;90:1585– 6.
A parallel among oral, vulvar, vaginal, penile, and anal can-
(9) Gillison ML, Koch WM, Capone RB, Spafford M, Westra WH, Wu L, et
cers is apparent in that only a fraction of tumors at these sites al. Evidence for a causal association between human papillomavirus and a
appear to be related to HPV (8). Strong risk factors other than subset of head and neck cancers. J Natl Cancer Inst 2000;92:709 –20.
HPV exist for cancers of the oral cavity and oropharynx; there- (10) Smith EM, Hoffman HT, Summersgill KS, Kirchner HL, Turek LP, Hau-
fore, the relative contribution of HPV would probably be larger gen TH. Human papillomavirus and risk of oral cancer. Laryngoscope
in regions where exposures such as tobacco smoking and chew- 1998;108:1098 –103.
ing are less common, although cancers of the oral cavity and (11) 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 HPV
oropharynx would be rare. The mechanism of transmission of
infection. J Natl Cancer Inst 1998;90:1626 –36.
HPV to the oral cavity warrants further investigation. The pros- (12) Franceschi S, Muñoz N, Snijders PJ, Walboomers WW. Human papillo-
pect of HPV vaccine development (38) offers hope for preven- mavirus and cancers of the upper aerodigestive tract: a review of epidemi-
tion of cervical and anogenital cancers and possibly also for a ological and experimental evidence. Cancer Epidemiol Biomarkers Prev
substantial number of cancers of the oropharynx and oral cavity. 1996;5:567–75.
(13) Mork J, Lie AK, Glattre E, Hallmans G, Jellum E, Koskela P, et al. Human
papillomavirus infection as a risk factor for squamous-cell carcinoma of the
APPENDIX head and neck. N Engl J Med 2001;344:1125–31.
(14) Herrero R. Chapter 7: Human papillomavirus and cancer of the upper
In addition to the authors of this article, the following investigators aerodigestive tract. J Natl Cancer Inst Monogr 2003;(31):47–51.
are part of the IARC Multicenter Oral Cancer Study Group and they (15) Lissowska J, Pilarska A, Pilarski P, Samolczyk-Wanyura D, Piekarczyk J,
participated in this study: Rengaswamy Sankaranarayanam, IARC, Bardin-Mikollajczak A, et al. Smoking, alcohol, diet, dentition and sexual

1782 ARTICLES Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003
 practices in the epidemiology of oral cancer in Poland. Eur J Cancer Prev (30) Zumbach K, Hoffman M, Kahn T, Bosch F, Gottschlich S, Gorogh T, et al.
2003;12:25–33. Antibodies against oncoproteins E6 and E7 of HPV types 16 and 18 in
(16) Balaram P, Shridar H, Rajkumar T, Vaccarella S, Herrero R, Nandakumar patients with head and neck squamous cell carcinoma. Int J Cancer 2000;
A, et al. Oral cancer in Southern India: the influence of smoking, drinking, 85:815– 8.
paan chewing and oral hygiene. Int J Cancer 2002;98:440 –5. (31) Tomassino M, Accardi R, Caldeira S, Dong W, Malanchi I, Smet A, Zehbe
(17) Fleming ID, Cooper JS, Henson DE, Hutter RV, Kennedy BJ Murphy GP, I. The role of p53 in cervical carcinogenesis. Hum Mutat 2003;21:307–12.
et al., editors. AJCC cancer staging manual. 5th ed. Philadelphia (PA): (32) Wiest T, Schwarz E, Enders C, Flechtenmacher C, Boxch FX. Involvement
Lippincott-Raven; 1997. of intact HPV16 E6/E7 gene expression in head and neck cancers with
(18) Jacobs MV, Snijders PJ, van den Brule AJ, Helmerhorst TJ, Meijer CJ, unaltered p53 status and perturbed pRB cell cycle control. Oncogene
Walboomers JM. A general primer GP5⫹/GP6(⫹)-mediated PCR-enzyme 2002;21:1510 –7.
immunoassay method for rapid detection of 14 high-risk and 6 low-risk (33) Wilczynski SP, Lin BT, Xie Y, Paz IB. Detection of human papillomavirus
human papillomavirus genotypes in cervical scrapings. J Clin Microbiol
DNA and oncoprotein overexpression are associated with distinct morpho-
1997;35:791–5.
logical patterns of tonsillar squamous cell carcinoma. Am J Pathol 1998;
(19) van Houten VM, Snijders PJ, van den Brekel MW, Kummer JA, Meijer CJ,
152:145–56.
van Leeuwen B, et al. Biological evidence that human papillomaviruses are
(34) Snijders PJ, Cromme FV, van den Brule AJ, Schrijnemakers HF, Snow GB,
etiologically involved in a subgroup of head and neck squamous cell
Meijer CJ, et al. Prevalence and expression of human papillomaviruses in
carcinomas. Int J Cancer 2001;93:232–5.
(20) de Roda Husman AM, Snijders PJ, Stel HV, van den Brule AJ, Meijer CJ, tonsillar carcinomas, indicating a possible viral etiology. Int J Cancer
Walboomers JM. Processing of long-stored archival cervical smears for 1992;51:845–50.
human papillomavirus detection by the polymerase chain reaction. Br J (35) Snijders PJ, Meijer CJ, van den Brule AJ, Schrijnemakers HF, Snow GB,
Cancer 1995;72:412–7. Walboomers JM. Human papillomavirus (HPV) type 16 and 33 E6/E7
(21) van den Brule AJ, Meijer CJ, Bakels V, Kenemans P, Walboomers JM. region transcripts in tonsillar carcinomas can originate from integrated and

Downloaded from http://jnci.oxfordjournals.org/ at Kainan University on February 11, 2015

Rapid detection of human papillomavirus in cervical scrapes by combined episomal HPV DNA. J Gen Virol 1992;73(Pt 8):2059 – 66.
general primer-mediated and type-specific polymerase chain reaction. (36) Klussmann JP, Weissenborn SJ, Wieland U, Dries V, Kolligs J, Jungehu-
J Clin Microbiol 1990;28:2739 – 43. elsing M, et al. Prevalence, distribution, and viral load of human papillo-
(22) Viscidi RP, Ahdieh-Grant L, Clayman B, Fox K, Massad LS, Cu-Uvin S, mavirus 16 DNA in tonsillar carcinomas. Cancer 2001;92:2875– 84.
et al. Serum immunoglobulin G response to human papillomavirus type 16 (37) Steenbergen RD, Hermsen MA, Walboomers JM, Joenje H, Arwert F,
virus-like particles in human immunodeficiency virus (HIV)-positive and Meijer CJ, et al. Integrated human papillomavirus type 16 and loss of
risk-matched HIV-negative women. J Infect Dis 2003;187:194 –205. heterozygosity at 11q22 and 18q21 in an oral carcinoma and its derivative
(23) Sehr P, Zumbach K, Pawlita M. A generic capture ELISA for recombinant cell line. Cancer Res 1995;55:5465–71.
proteins fused to glutathione S-transferase: validation for HPV serology. (38) Koutsky LA, Ault KA, Wheeler CM, Brown DR, Barr E, Alvarez FB, et al.
J Immunol Methods 2001;253:153– 62. A controlled trial of a human papillomavirus type 16 vaccine. N Engl
(24) Paz IB, Cook N, Odom-Maryon T, Xie Y, Wilczynski SP. Human papil- J Med 2002;347:1645–51.
lomavirus (HPV) in head and neck cancer. An assocaition of HPV 16 with
squamous cell carcinoma of Waldeyer’s tonsillar ring. Cancer 1997;79:
595– 604.
(25) Miller CS, Johnstone BM. Human papillomavirus as a risk factor for oral
NOTES
squamous cell carcinoma: a meta-analysis. Oral Surg Oral Med Oral Pathol
Supported by grant S06 96202489 05F02 from Europe Against Cancer (to the
Oral Radiol Endod 2001;91:622–35.
International Agency for Research on Cancer, Lyon, France [N. Muñoz and R.
(26) Meschede W, Zumbach K, Braspenning J, Scheffner M, Benitez-Bribiesca
Herrero]) and grants FIS 97/0662, FIS 01/1236, and BAE 01/5013 (to the Institut
L, Luande J, et al. Antibodies against early proteins of human papilloma-
viruses as diagnostic markers for invasive cervical cancer. J Clin Microbiol Català d’Oncologia, Barcelona, Spain [X. Castellsagué and F. Xavier Bosch])
1998;36:475– 80. from “Fondo de Investigaciones Sanitarias” (FIS), Madrid, Spain. Funding was
(27) Stanley M. Antibody reactivity to HPV E6 and E7 oncoproteins and early also provided by the International Union Against Cancer (UICC) Yamagiwa-
diagnosis of invasive cervical cancer. Am J Obstet Gynecol 2003;188:3– 4. Yoshida Memorial International Cancer Study (to X. Castellsagué), the National
(28) Lehtinen M, Pawlita M, Zumbach K, Lie K, Hakama M, Jellum E, et al. Cancer Institute of Canada (to McGill University, Montreal, Quebec, Canada [E.
Evaluation of antibody response to HPV early proteins in women in whom Franco]), the Italian Association for Research on Cancer (to the Centro di
cervical cancer developed one to 20 years later. Am J Obstet Gynecol Riferimento Oncologico di Aviano, Aviano, Italy [S. Franceschi]), and the Pan
2003;188:49 –55. American Health Organization (to the Instituto Nacional de Oncologia y Radio-
(29) Rosales R, Lopez-Contreras M, Cortes RR. Antibodies against HPV type biologia, Havana, Cuba [L. Fernández]).
16 and 18 E2, E6 and E7 proteins in sera: correlation with presence of Manuscript received February 4, 2003; revised September 23, 2003; accepted
papillomavirus DNA. J Med Virol 2001;65:736 – 44. September 29, 2003.

Journal of the National Cancer Institute, Vol. 95, No. 23, December 3, 2003 ARTICLES 1783