244 Full PDF

Downloaded from http://jcp.bmj.com/ on September 8, 2015 - Published by group.bmj.
com
244
REVIEW
The causal relation between human papillomavirus and

cervical cancer
F X Bosch, A Lorincz, N Muoz, C J L M Meijer, K V Shah
.............................................................................................................................
J Clin Pathol 2002;55:244265
The causal role of human papillomavirus infections in INTRODUCTION

cervical cancer has been documented beyond A major discovery in human cancer aetiology
has been the recognition that cervical cancer is a
reasonable doubt. The association is present in virtually rare consequence of an infection by some
all cervical cancer cases worldwide. It is the right time mucosatropic types of HPV. In public health
for medical societies and public health regulators to terms, this finding is equally important as the
discovery of the association between cigarette
consider this evidence and to define its preventive and smoking and lung cancer, or between chronic
clinical implications. A comprehensive review of key infections with hepatitis B virus (HBV) or hepa-
studies and results is presented. titis C virus and the risk of liver cancer. Moreover,
as in the HBV disease model, intense efforts are
..........................................................................
currently going into the development and testing
of vaccines that may prevent the relevant HPV
SUMMARY infections, and presumably, cervical cancer.
During the 1990s, epidemiological studies, sup- By the year 2000, the epidemiological evidence
ported by molecular technology, provided evidence included a large and consistent body of studies
on the causal role of some human papillomavirus indicating, beyond any reasonable doubt, strong
(HPV) infections in the development of cervical and specific associations relating HPV infections
cancer. This association has been evaluated under to cervical cancer. The observations have been
all proposed sets of causality criteria and endorsed reported from all countries where investigations
by the scientific community and major review have taken place. Studies include prevalence sur-
institutes. The finding is universally consistent, veys, natural history investigations, casecontrol
and to date there are no documented alternative studies and, more recently, a randomised inter-
hypotheses for the aetiology of cervical cancer. vention trial. Natural history and follow up stud-
ies have clearly shown that HPV infection
HPV has been proposed as the first ever identi-
preceded the development of cervical cancer by
fied, necessary cause of a human cancer. In several years and confirmed that sexual transmis-
practical terms, the concept of a necessary cause sion is the predominant mode of HPV acquisition.
implies that cervical cancer does not and will not These studies satisfied, in biological terms, the
develop in the absence of the persistent presence long known clinical and epidemiological observa-
of HPV DNA. tions that cervical cancer displayed the profile of a
Cervical cancer is still the second most com- sexually transmitted disease (STD). Casecontrol
mon cancer in women worldwide, although it is a studies, case series, and prevalence surveys have
theoretically preventable disease. unequivocally shown that HPV DNA can be
In developed parts of the world, and in popula- detected in adequate specimens of cervical cancer
in 90100% of cases, compared with a prevalence
tions where cytology based programmes are
of 520% in cervical specimens from women
established, it would be beneficial to add HPV identified as suitable epidemiological controls.
testing to the screening protocol. HPV testing was The association has been recognised as causal
shown by several studies, including one ran- in nature by several international review parties
domised trial, to be of help in solving the since the early 1990s, and the claim has been
ambiguous cases generated by cytology reading. made that this is the first necessary cause of a
In populations where cytology programmes are human cancer ever identified.
See end of article for
authors affiliations either not in place or are not efficient, HPV testing
....................... should now be considered and evaluated as an .................................................
alternative test for primary screening.
Correspondence to: Abbreviations: AF, attributable fraction; ASCUS,
Dr F X Bosch, Prevention of exposure to high risk HPV types atypical squamous cells of undetermined significance; CI,
Epidemiology and Cancer by vaccination may prove to be the most efficient confidence interval; CIN, cervical intraepithelial neoplasia;
Registry Unit (SERC), and logistically feasible preventive intervention EBV, Epstein-Barr virus; GP, general primer; HCV, hepatitis
Catalan Institute of C virus; HIV, human immunodeficiency virus; HPV, human
Oncology (ICO), Hospital for cervical cancer.
papillomavirus; HSIL, high grade squamous intraepithelial
Duran i Reynals, Gran Via At this stage of development, regulatory agen- lesion; HSV-2; type 2 herpes simplex virus; IARC,
km 2,7, LHospitalet de cies are requested to evaluate the scientific International Agency for Research on Cancer; IBSCC,
Llobregat, 08907
evidence and weigh its implications in relation to international biological study on cervical cancer; OC, oral
Barcelona, Spain;
x.bosch@ico.scs.es costs, public health investments, and policy. This contraceptive; OR, odds ratio; ORF, open reading frame;
Pap, Papanicolaou; PF, protective fraction; PCR,
is a subjective evaluation that could be guided by polymerase chain reaction; RB, retinoblastoma; RR,
Accepted for publication
22 January 2002 a careful description of the most relevant studies relative risk; STD, sexually transmitted disease; Th1, T
....................... and findings. helper cell type 1
www.jclinpath.com
Downloaded from http://jcp.bmj.com/ on September 8, 2015 - Published by group.bmj.com
HPV and cervical cancer 245
The implications of the recognition that, in the absence of

Number of HPV related publications
viral DNA, cervical cancer does not develop, are of considerable 800
practical importance. On the one hand, the concept of risk Publications on HPV and cervical cancer
700
groups comes into focus. High risk women can now be sharply Abstracts at the HPV Annual Conferences
600
redefined as the group of persistent HPV carriers. Operatively,
500
this represents substantial progress from previous versions of
the high risk group that identified women by their exposure to 400
a constellation of ill defined factors (low socioeconomic status, 300
high number of sexual partners, smoking, use of oral contra- 200
ceptives, history of STDs, and any combination of the above).
100
Most of these factors are now viewed either as surrogates of
HPV exposure or as relevant cofactors given the presence of 0
82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 99 00
HPV DNA. On the other hand, if indeed HPV is a necessary Year
cause of cervical cancer, the implication is that specific
preventive practices targeting some putative non-HPV related Figure 2 Scientific publications on human papillomavirus (HPV)
cervical cancer cases are no longer justified. Finally, technol- identified by Medline and the number of research abstracts
ogy is now available to screen HPV DNA positive women in the presented at the annual Papillomavirus International Conferences.
general population. Therefore, the final consideration on the
nature of the association between HPV and cervical cancer is were predominantly used and an estimate of the periods in
of considerable public health relevance. Research at the popu- which the key types of studies were initiated.
lation level has largely accomplished its task by providing an Figure 2 displays the approximate number of scientific
exhaustive body of evidence. It is now time for public health papers identified by Medline searches on HPV and on HPV and
institutions to evaluate these achievements, consider the costs cervical cancer, and the number of research abstracts
and benefits involved, and apply this knowledge to their presented at the major annual papillomavirus conferences
guidelines, recommendations, and policy. (http://www.ipvsoc.org). The 1980s generated a rapidly
increasing number of publications on HPV DNA prevalence in
HISTORICAL OVERVIEW cervical cancer and reports on validation of the available
Research in relation to the aetiology of cervical cancer has detection methods. The 1990s produced the key results of
made substantial progress in the past two decades, both in casecontrol and cohort studies, and beginning in the late
scientific and operational terms. For decades, the epidemio- 1990s there was an increasing number of publications on the
logical profile of women with cervical cancer was recognised clinical uses of HPV testing in screening and triage.
as suggestive of a sexually transmitted process, and several
infectious agents were proposed over the years including CAUSALITY CRITERIA IN HUMAN CANCER
syphilis, gonorrhea, and type 2 herpes simplex virus (HSV-2). RESEARCH
The development of technology to test for the presence of Epidemiological studies are essential to establish the associ-
HPV DNA in cellular specimens in the early 1980s and the ation between risk factors and cancer and to qualify the nature
multidisciplinary collaboration within the field made possible of the association. Traditionally, these include case series,
the establishment of a definite aetiological role for HPV in casecontrol studies, cohort studies, and intervention studies.
cervical cancer. Evidence is also accumulating for HPV Comparisons of exposure between patients with cervical
involvement in a considerable proportion of cancers of the cancer and their relevant controls were initially established
vulva, vagina, anal canal, perianal skin, and penis. The associ- using questionnaires. Most studies conducted before the
ation of HPV with cervical cancer has provided the back- availability of HPV DNA detection systems identified as key
ground and the justification for improving screening pro- risk factors several variables related to the sexual behaviour of
grammes and for developing HPV vaccines. the women and of their sexual partners. The most frequently
Figure 1 is a schematic view of the time scale of this reported risk factors included the number of sexual partners,
dynamic process. It includes an indication of the results an early age at first intercourse, or any previous STD.1
obtained as technology evolved in sensitivity expressed as the Once the relevant biomarkers were validated, in this case
per cent of cervical cancer cases that were found to contain the presence of HPV DNA in exfoliated cervical cells, it became
viral DNA. The figure also indicates the types of HPV tests that possible to advance over questionnaire based studies and
establish biologically sound comparisons between patients
and controls. In epidemiological terms, these comparisons
HPV DNA in would analyse cervical cells from women with cervical cancer
3060% 75% 95% 99%
cervical cancer and from otherwise comparable women without cervical can-
cer (casecontrol studies), or from cohorts of women tested
SH for viral DNA (cohort studies). To characterise the link
HPV-DNA tests
FISH, TS-PCR HC I HC II/GP-PCR
between HPV and invasive cervical cancer, casecontrol stud-
HPV in triage ies proved to be the key study design and the only ones ethi-
Cohort cally acceptable in human populations. Typically in such a
Types of studies Case control study and at the time of fieldwork, controls are selected to
Prevalence HPV in screening match the age distribution of the cases and, as much as possi-
ble, the general characteristics of the cases (place of residence,
HPV in cases HPV vaccines Phase IIII
Trials socioeconomic status, health plan, etc). To characterise the
association, all study participants are requested to comply
Time 1980 1990 2000 with a questionnaire to assess their individual exposure to any
known or suspected risk factor for the disease. The
Figure 1 Evolution of epidemiological research on human information is then used to estimate the odds ratios (ORs) of
papillomavirus (HPV) and cervical cancer in the past two decades.
FISH, filter in situ hybridisation; GP-PCR, general primer PCR; HC
disease related to any given exposure. Multivariate analyses
III, hybrid capture first and second generation; PCR, polymerase have the ability to compare (through statistical adjustment)
chain reaction; SH, Southern blot hybridisation; TS-PCR, type specific strictly equivalent groups of women in relation to any of the
PCR. exposures of interest. The adjusted differences (ratios) in the
www.jclinpath.com
246 Bosch, Lorincz, Muoz, et al
prevalence in HPV markers between cases and controls are was discussed was largely based on antibody detection and the
then obtained after having eliminated the effects of any other studies involved were seroepidemiological surveys and case
differences in exposure. Likewise, comparisons of cases and control studies. Antibody measurements were the methods of
controls in relation to other variables of interest will provide choice for the assessment of exposure. Evans proposed a uni-
estimates of the relevance of other factors (oral contraceptives fied scheme for causation that included most of the criteria
(OCs) or smoking) and identify the variables that merely mentioned in table 1.14 In 1976, Rothman15 introduced the
reflect HPV exposure (surrogate variables). concepts of necessary and sufficient causes. This model is
When the technology to detect HPV DNA in samples of useful to accommodate the growing evidence of the multifac-
DNA extracted from exfoliated cervical cells became available, torial origin of human cancer in many instances. Finally, sev-
it was relatively easy to show that most of the sexual eral authors have defined criteria to evaluate the findings of
behaviour variables were in fact surrogate measures of HPV molecular technology that provided the basis of the studies of
exposure, reflecting the predominant pathway of acquisition HPV and cervical cancer.16 17
of HPV. As methods became more sensitive, the parameters Because of its wider acceptance, we will discuss in detail the
that merely expressed the probability of HPV (or any other criteria proposed by Hill, and its version adopted by the Inter-
STD) infection, such as number of sexual partners, became national Agency for Research on Cancer (IARC) monograph
statistically irrelevant.25 programme, in addition to the model on necessary and suffi-
Causality in public health requires a judgment based on sci- cient causes proposed by Rothman in 1995.18
entific evidence from human and experimental (animal) In brief, the criteria proposed by Hill8 as summarised by
observations. As such, only the latter may benefit from the Rothman19 include the following:
most stringent criteria of causality; that is, the repeated Hill suggested that the following aspects of an association
induction of the disease by exposure to the relevant agent(s) should be considered when attempting to distinguish causal
compared with the spontaneous occurrence of the same from non-causal associations: (1) strength, (2) consistency,
disease in unexposed and yet comparable groups of animals. (3) specificity, (4) temporality, (5) biological gradient, (6)
All causal associations of human cancers have been recognised plausibility, (7) coherence, (8) experimental evidence, and (9)
based on educated judgment of the results of epidemiological analogy.
studies at the level already available for HPV and cervical can-
cer. Final proof can only be confirmed by intervention Strength
(preventive) trials, in which a reduction of the disease burden By strength of association, Hill means the magnitude of the
(incidence or mortality) is observed following the introduc- ratio of incidence rates. Hills argument is essentially that
tion of a preventive practice in strictly controlled conditions. strong associations are more likely to be causal than weak
These studies typically include as controls populations to associations because if they were the result of confounding or
whom the existing standard of preventive care is being some other bias, the biasing association would have to be even
offered. stronger and would therefore presumably be evident. Weak
Table 1 displays some of the criteria that have been proposed associations, on the other hand, are more likely to be
to evaluate the nature of the associations encountered by epi- explained by undetected biases. Nevertheless, the fact that an
demiological studies. This is particularly relevant when association is weak does not rule out a causal connection.
causality is being proposed because, as a consequence, preven-
tive or clinical recommendations are made. Consistency
In addition to the criteria listed in table 1, some additional Consistency refers to the repeated observation of an associ-
contributions might be worth discussing. In 1976, Evans ation in different populations under different circumstances.
reviewed the history of the causality criteria in infectious dis-
ease models and adapted the early postulates of Henle-Koch Specificity
to both the viral origin of acute diseases and to the relation The criterion of specificity requires that a cause should lead to
between viral infections and cancer.14 The human models that a single effect, not multiple effects. However, causes of a given
inspired most of the latter included two examples: Epstein- effect cannot be expected to be without other effects on any
Barr virus (EBV) infections and Burkitts lymphoma, and logical grounds. In fact, everyday experience teaches us
HSV-2 viral infections and cervical cancer. The technology that repeatedly that single events may have many effects.
Table 1 Epidemiological considerations important for causal inference

Surgeon General (1964)9 MacMahon and Pugh
Lilienfeld (1959)6 Sartwell (1960)7 Hill (1965)8 and Susser (1973)10 (1970)11 IARC (1995)12
Magnitude of effect Strength of association Strength of association Strength/magnitude of Strength/magnitude of Strength of association
association association
Consistency Replication Consistency Consistency Consistency Consistency
Specificity Specificity Specificity
Temporality Temporality Temporality Temporality Temporality
Dose response Dose response Biological gradient Dose response Dose response Dose response
Biological mechanism Biological plausibility Biological mechanisms Mechanisms
Biological Biological coherence Biological coherence Consonance with existing
reasonableness knowledge
Experimentation Experimental Experimental Experimental
evidence (intervention) (intervention)
Analogy
Exclusion of alternative
explanations
Adapted from Weed 1995.13
www.jclinpath.com
Temporality of which five involve biological agents such as HPV.12 In its

Temporality refers to the necessity that the cause should pre- preamble, the monograph programme establishes guidelines
cede the effect in time. The temporality of an association, is a to qualify an epidemiological observation as causal, and also
sine qua non: if the cause does not precede the effect that is defines rules to be followed when human data suggest lack of
indisputable evidence that the association is not causal. carcinogenicity potential. These criteria are useful to challenge
any aetiological hypothesis when the epidemiological studies
Biological gradient are inconsistent or when only weak associations are reported.
Biological gradient refers to the presence of a doseresponse Finally, another useful way of examining the nature of an
curve. If the response is taken as an epidemiological measure association was provided by a model system that proposed
of effect, measured as a function of comparative disease inci- that any given disease would occur as a consequence of
dence, then this condition will ordinarily be met. human exposure to a sufficient cause.18 A sufficient cause is
described, in its simplest model, as the concurrence in a given
Plausibility individual of a constellation of factors (called the components
Plausibility refers to the biological plausibility of the hypoth- of the sufficient cause), following which the disease will
esis, an important concern but one that may be difficult to develop. Each given disease will have its own sufficient cause
judge or sets of sufficient causes (lung cancer may have a sufficient
cause that involves cigarette smoking, but another sufficient
Coherence cause that does not include smoking, such as intense radon
Taken from the Surgeon Generals report on Smoking and exposure in non-smokers). According to the model, a
Heath (1964)9: The term coherence implies that a cause and necessary cause is described as a component of a sufficient
effect interpretation for an association does not conflict with cause that is part of all the sufficient causes described. To pre-
what is known of the natural history and biology of the vent disease it is not necessary to identify all the components
disease. of a sufficient cause, or to remove them all: it is sufficient to
remove one component from each sufficient cause, that is to
Experimental evidence remove, if it exists, the necessary cause.
Such evidence is seldom available for human populations. In
human data, the experimental criterion takes the form of pre-
ventive interventions and explores whether there is evidence COMPLIANCE OF THE CAUSALITY CRITERIA IN THE
that a reduction in exposure to the agent is associated with a HPV AND CERVICAL CANCER MODEL
reduction in risk. In the following sections, we will review several studies that
have provided evidence of the association between HPV and
Analogy
cervical cancer, using the criteria outlined in table 1. For pur-
The insight derived from analogy seems to be handicapped by
poses of clarity, we shall concentrate the discussion on the cri-
the inventive imagination of scientists, who can find analogies
teria that have proved to be of greater value in the evaluation
everywhere. Nevertheless, the simple analogies that Hill
of human carcinogens and on the studies that focused on
offersif one drug can cause birth defects, perhaps another
invasive cervical cancer.
can alsocould conceivably enhance the credibility that an
association is causal.
Strength of the association
As is evident, these nine aspects of epidemiological evidence This criterion is usually discussed by examining the magni-
offered by Hill to judge whether an association is causal are tude of the relative risk (RR), or the OR, which is the estimate
saddled with reservations and exceptions; some may be wrong of the RR in casecontrol studies. We shall use as the primary
(specificity) or occasionally irrelevant (experimental evidence example the results of the IARC multicentre casecontrol
and perhaps analogy). Hill admitted that: none of my nine study on invasive cervical cancer, as presented at international
viewpoints can bring indisputable evidence for or against the scientific meetings, and either published or at different stages
cause and effect hypothesis and none (except temporality) can of preparation for publication. In brief, this project included
be required as a sine qua non. nine casecontrol studies in different parts of the world,
The IARC in its monograph programme largely adopted the mostly in high risk countries. HPV DNA testing was done in
causality criteria proposed by Hill and established rules to two central research laboratories using the MYO9/1120 and the
decide on the carcinogenicity of a given exposure, particularly general primer (GP) GP5+/6+21 22 polymerase chain reaction
when human data are scarce and must be combined with (PCR) testing systems. The published results have reported
experimental data. However, the final qualification of the car- ORs for cervical cancer in the range of 50 to 100 fold for HPV
cinogenicity of any given substance being evaluated is taken DNA. ORs for specific associations (such as HPV-16 and
by vote of the external (non-IARC) participants. squamous cell cancer and HPV-18 and cervical adenocarcino-
The monograph programme and its criteria has been mas) range between 100 and 900. These estimates lead to cal-
reviewed and accepted by most scientists in the field of human culations of attributable fractions (AF) for the entire study
carcinogenesis. To date, 77 monographs have been published, greater than 95%.23
Table 2 Size of the IARC multicentre casecontrol study and human papillomavirus
(HPV) DNA prevalence
Cervical cancer
Controls Adeno and mixed Squamous
No. countries N HPV DNA (% +ve) N HPV DNA (% +ve) N HPV DNA (% +ve)
9* 2491 13.4 2365 90.7

6 1466 15.4 141 91.9 2280 96.6
*Brazil, Morocco, Paraguay, the Philippines, Thailand, Peru, Mali, Spain, and Colombia.
Brazil, Morocco, Paraguay, the Philippines, Thailand, and Peru.
www.jclinpath.com
these very high ORs reflect the risk in relation to existing HPV
Controls
DNA in cervical cells (HPV DNA point prevalence), not in
Cases
relation to ever being infected with HPV (cumulative
lifetime exposure). Furthermore, if HPV shedding was
Spain intermittent among controls, their corresponding HPV preva-
lence would have been underestimated, resulting in an infla-
Colombia tion of the ORs observed. It is usually interpreted that the HPV
DNA point prevalence at advanced age (over 40 years of age)
Brazil reflects viral persistency. However, much research is still
Morocco devoted to defining viral persistency and its prognosis
accurately, a crucial definition for the clarification of the uses
Paraguay of HPV testing in screening and patient management.31
Most of the discussion in the text uses HPV DNA as a
The Philippines generic marker that includes any positive result for several
HPV types. It is now possible to provide estimates of the RR for
Thailand at least 10 different HPV types showing that there are no sig-
Peru
nificant differences in the risk of cervical cancer in relation to
the HPV types most commonly found in these lesions. The
Total preliminary results of the IARC multicentre casecontrol were
pooled and summarised by Muoz et al in 2000,32 at the HPV
0 20 40 60 80 100 2000 Papillomavirus Conference (www.hpv2000.com). These
analyses indicated that for squamous cell carcinomas, the age
Figure 3 Prevalence of human papillomavirus (HPV) DNA in cases
and controls in the IARC multicentre casecontrol study.2430
and centre adjusted OR was 83.3 (95% CI, 54.9 to 105.3). The
prevalence of the four most common HPV types and their ORs
among 1545 cases with single infections were: HPV-16, 59%
(OR = 182); HPV-18, 12% (OR = 231); HPV-45, 4.8%
Table 3 Odds ratio for the association of human
(OR = 148); and HPV-31, 3.7% (OR = 71.5). Other less
papillomavirus (HPV) DNA and cervical cancer in the
common HPV types showing equally high ORs were: HPV-33,
IARC multicentre casecontrol study: preliminary
OR = 77.6; HPV-35, OR = 34.8; HPV-51, OR = 42.7; HPV-52,
data23 32 33
OR = 145.7; HPV-58, OR = 78.9; and HPV-59, OR = 347.3.
HPV DNA The most common types among cases were also the most
prevalence (%) common types among HPV positive control women: HPV-16,
No.
studies Controls Cases OR (95% CI) 30.3%; HPV-18, 8.2%; HPV-31, 4.8%; and HPV-45, 3.9%. These
findings indicate that in addition to HPV-16 and HPV-18, HPV
Squamous 9 13.4 90.7 83.3 (54.9 to 105.3)
types 31, 33, 35, 45, 51, 52, 58, and 59 should be considered as
Adeno and
mixed 6 15.4 91.9 68.7 (36.2 to 130.5) human carcinogens.
The HPV type distribution in the population and in patients
CI, confidence interval; IARC, Agency for Research on Cancer OR, with cervical cancer shows a seemingly modest geographical
odds ratio. variability that has not been fully described (J Kornegay, per-
sonal communication, 2001).3436 The description and the
implications of such variability for HPV testing and HPV vac-
Table 2 shows the size of the multicentre casecontrol study cination are to be determined.
and the prevalence of HPV DNA in each relevant group. Figure The results of the multicentre study are consistent with
3 displays the HPV DNA prevalence in eight countries in cer- findings from other countries that have generated recent data
vical cancer cases and controls. It is noteworthy that the first on invasive cervical cancer and preinvasive disease in Costa
two studies conducted in Spain and Colombia (fig 3) used Rica,37 Thailand,38 Norway,39 Denmark,40 and virtually all other
early versions of the MYO9/11 PCR system that identified HPV countries in which these studies have been conducted.
DNA in approximately 75% of the cases. The rest of the stud- Multiple HPV types were detected in the multicentric study
ies were analysed using the GP5+/6+ PCR system and its on average in 7.3% of the cases and 1.9% of the controls, and
modifications, which resulted in an almost 20% increase in the did not show a significantly increased risk (OR = 54.5; 95%
HPV DNA detection rate. CI, 35.5 to 83.6) over women positive for only one HPV type
Table 3 shows the corresponding estimates of the RR (OR (OR = 86.6; 95% CI, 68.2 to 110).
and 95% confidence interval (CI)). Results are presented The proportion of multiple types in a given specimen varies
separately for squamous cell carcinomas and adenocarcino- across studies and particularly in relation to the HPV detection
mas of the cervix. Given the casecontrol design of the study, method used. Table 4 provides an indication of the proportion
Table 4 Prevalence of multiple human papillomavirus (HPV) types in patients with

cervical cancer and women without cervical cancer
Cases Non-cases
% Of all % Of all the % Of all

Ref specimens HPV +ve specimens
32 IARC multicentre 420% 10% 13%

37 Rural Costa Rica 32% 38% 4%
35 Rural Mozambique 41% 15%
41 Imprisoned women, Spain 71% 20%
42 HIV +ve, USA 42%
HIV -ve, USA 16%
IARC, International Agency for Research on Cancer.
www.jclinpath.com
of specimens from cases and from the general population that

25
showed multiple types. The table suggests that populations at Eluf-Neto et al (1994)
50
high risk of cervical cancer and with high rates of human Asato et al (1994)
51
Shen et al (1993)
immunodeficiency virus (HIV) positivity tend to show higher Arends et al (1993)
52
proportions of multiple types than do populations not belong- Das et al (1992)

53
54
ing to these risk groups. Longitudinal studies have suggested Ter Meulen et al (1992)
55
that the one time, cross sectional detection of type specific Muoz et al (1992)
56
Peng et al (1991)
HPV may underestimate the cumulative lifetime diversity of Griffin et al (1990)
57
exposure to HPV.31 However, in all studies of invasive

0.1 1 5 10 50 100 500 1000
carcinoma, the risk linked to multiple HPV types does not vary
significantly from the risk linked to single HPV types. Odds ratio
The similarity in the prognostic value of detection of any of Figure 5 Odds ratios (OR) and 95% confidence intervals for
the 10 high risk HPV types, in addition to any combination of associations found in casecontrol studies using PCR methods
them, clearly indicates that group testing for high risk HPVs between human papillomavirus 16 (HPV-16) (or its nearest surrogate)
and invasive cervical cancers. *The OR estimate is owing to the
would be sufficient in the context of clinical and screening
absence of HPV positive controls. Adapted from IARC monograph
protocols. 64, 1995.12
Figure 4 shows, for comparison purposes, some estimates of
the strength of associations between environmental factors
and human cancer that were recognised as causal in nature by Consistency
epidemiological studies and subsequently proved in human There is a striking consistency between the results of the
populations by intervention studies. The figure includes risk multicentre casecontrol study and over 50 other studies con-
(RRs or ORs) as the measurement of the strength of the asso- ducted in other countries, under different protocols and HPV
ciations and AFs representing the proportion of disease that is DNA testing systems. Figures 5, 6, and 7 summarise the results
attributable to (caused by) the exposure. Below the reference of studies that compared the prevalence of HPV DNA in
line the risk column displays its reverse estimate as a less than patients with cervical cancer and controls. Some of the studies
one (protective) OR or RR and the protective fractions (PF%) used the prevalence of HPV-16 DNA to calculate ORs and some
in the right hand column show results that have already been reported results for HPV DNA (all types combined). Some
achieved in disease reduction after specific exposure reduction studies focused on invasive cervical cancer, whereas others
interventions. used preinvasive lesions as the definition of cases. When indi-
cated, separate analyses are presented for squamous cell carci-
nomas and for adenocarcinomas. Studies that have compared
risk factors for cervical intraepithelial neoplasia stage 3 (CIN
Strength of association. Evaluation 3) and invasive cancer have not reported any significant
differences in their associations with HPV or with their epide-
The association between HPV DNA in cervical specimens miological profile.38 49
and cervical cancer is one of the strongest ever observed Apart from confirming the high ORs shown in figs 5 and 6,
for a human cancer. HPV-16 accounts for almost 50% of fig 7 also demonstrates the consistency of results between
the types identified in cervical cancer. The cancer risk for
any one of at least 10 HPV types or for any combination
of HPV types does not differ significantly. 58
Marin et al (1994)
59
Lrincz et al (1992)
60
Kanetsky et al (1992)
61
Kadish et al (1992)
62
Mandelblatt et al (1992)
63
Saito et al (1992)
RR/OR AF% Ohta et al (1991)
64
65
Wei et al (1991)
66
Si et al (1991)
67
>500 HPV-DNA-18 and cervical 99 Chang et al (1990)
68
adenocarcinoma in the Philippines Zhang et al (1990)
69
Yokota et al (1990)
HR-HPV-DNA and cervical cancer in Costa Rica 80 Colgan et al (1989)
70
71
HPV-DNA and cervical cancer in Bangkok 90 Czegledy et al (1989)
72
100 Reeves et al (1989)
HBsAg and liver cancer in Taiwan 60 Donnan et al (1989)
73
74
50 HBsAg and liver cancer in Greece 60 Zhang et al (1988)
75
Wilczynski et al (1988)
20 HCV and liver cancer in Italy 40 76
Fuchs et al (1988)
77
10 Cigarette smoking and lung cancer 80 Hsieh et al (1988)
78
Choo et al (1988)
79
PF% Kulski et al (1987)
80
1 Baseline reference Zhang et al (1987)
81
Lrincz et al (1987)
82
Schneider et al (1987)
0.1 Smoking cessation before middle age in 90 Meanwell et al (1987)
83
the UK and lung cancer 84
McCance et al (1985)
0.6 HBV vaccination of adults in Korea 40
and liver cancer 0.1 1 5 10 50 100 500 1000
0.1 HBV vaccination of newborns in Taiwan 90 Odds ratio
and liver cancer
Figure 6 Odds ratios (OR) and 95% confidence intervals for
Figure 4 Selected examples of the strength of the associations (RR/ associations found in casecontrol studies using non-PCR methods
OR) between risk factors and human cancer; estimates of the between human papillomavirus 16 (HPV-16) (or its nearest surrogate)
attributable fraction (AF%) and of the protective fraction (PF%). Refs: and invasive cervical cancers. *The OR estimate is owing to the
the Philippines,28 Costa Rica,37 Bangkok,5 Taiwan,43 Greece,44 Italy,45 absence of HPV positive controls. Adapted from IARC monograph
UK,46 Korea,47 and Taiwan.48 64, 1995.12
www.jclinpath.com
squamous cell carcinomas and adenocarcinomas, the consist- (2) Adenocarcinomas and adenosquamous cell carcinomas
ency of findings between preinvasive disease and invasive are more closely related to HPV-18 and its phylogenetically
cancer, and the consistency of findings between risk estimates related family (HPV types 39, 45, and 59) than are squamous
for HPV DNA (all types considered) and risk estimates cell carcinomas, which in turn are closely linked to HPV-16
restricted to high risk types. and its phylogenetically related family (HPV types 31, 35, and
52).34 87 The reasons for such specificity are unknown.
(3) Cancers of the vulva and vagina are closely related to HPV-
Consistency. Evaluation 16. Approximately 4050% of vulvar cancer shows HPV DNA,
and in several series HPV-16 is by far the predominant type in
The association between HPV DNA in cervical specimens more than 80% of cases.8890
and cervical cancer is consistent in a large number of (4) Cancer of the tonsil is closely related to HPV-16, whereas
investigations in different countries and populations. There other cancers of the oral cavity show inconsistent and lower
are no published studies with observations challenging the prevalences of HPV DNA.9194
central hypothesis on causality. (5) Skin cancers related to the epidermodysplasia verruci-
formis condition are related to a restricted number of derma-
totrophic HPV types. These are also recovered from basal cell
Specificity carcinomas and squamous cell carcinomas of the skin in
Specificity, as defined by Hill, tended to be relegated to a sec- immunosuppressed and immunocompetent individuals.95
ondary level for cancer causality evaluation once it became (6) Other associations, reported in a small number of cases,
clear that carcinogenic exposures are usually complex (for seem to occur with some specificity. For example HPV-16 and
example, cigarette smoke) and can induce cancer in different cancers of the conjunctiva96 and HPV-16 and cancers of the
organs and even cancers of different histological profile in the ungueal bed.
same organ. (7) Studies on HPV variants (variation within HPV types at
In the case of HPV, the complexity of the association is being the single nucleotide level) are beginning to unveil risk
unveiled. The HPV family includes over 100 HPV types, of differences.9799 The geographical distribution of HPV variants
which 3040 are mucosatropic and at least 15 types have been and its relevance for HPV testing and for vaccine development
clearly linked to cervical cancer. In addition, some of these are still uncertain.
types are also related to other cancers of the genital tract (vul- (8) HPV has been excluded as a likely cause or even as a risk
var cancer, vaginal cancer, and cancers of the anal canal, peri- factor for other human cancers. A large number of investiga-
anal skin, and the penis) and perhaps to cancers of other tions (largely unpublished) have not provided support to the
organs (such as oropharyngeal and skin cancer). hypothesis of the involvement of these viruses in the causation
To examine the association of HPV and human cancer in of cancers of the endometrium, ovary, prostate, or other sites
light of the specificity criteria, we shall widen the original (reviewed by Shah and Howley16 and Syrjnen and Syrjnen100).
scope (one exposure/one disease) to verify whether a more
complex model involving multiple HPV types and several can-
cer sites seems to occur with frequencies suggesting a consist-
ent departure from a random model. Specificity. Evaluation
(1) About 15 HPV types are involved in over 95% of the cervi-
cal cancer cases. HPV-16 and HPV-18 are the most common The association of type specific HPV DNA and cervical
types identified and represent 50% and 10%, respectively, of cancer is significantly different from random. Systematic
the viral types involved in invasive cancer. Figure 8 shows the patterns of HPV type and cervical cancer histology suggest
cumulative prevalence of five HPV types in cervical carcinomas a fair degree of specificity. Patterns are also observed
by histological type in 2400 cases included in the multicentre when the scope of HPV and cancer expands to include the
casecontrol study. It clearly shows that these five HPV types full spectrum of HPV types and the large number of addi-
comprise 8095% of the viral types identified in carcinomas. tional cancer sites that have been investigated.
85 Squamous
Josefsson et al (2000)
86
Ylitalo et al (2000)
2 HPV + smears 16 + 18 + 45 73.1
5
Thomas et al (2001)
16 + 18 + 45 + 31 76.9
16 + 18 + 45 + 31 + 33 80.3
37
Herrero et al (2000)
32
Muoz et al (2000)
33
Bosch et al (2000) Adenocarcinomas
0.1 Reference 5 10 50 100 500 2000

1000 4500 16 + 18 + 45 89.7
OR
16 + 18 + 45 + 59 92.7
Carcinoma in situ, HPV-16
Squamous cell carcinoma 16 + 18 + 45 + 59 + 33 94.1
Adenocarcinoma
OR HPV any type
OR high risk HPV 0 20 40 60 80 100
95% confidence interval (%)
Figure 7 Odds ratios (OR) and 95% confidence intervals for Figure 8 Cumulative prevalence of human papillomavirus (HPV)
associations found in casecontrol studies after the year 2000. HPV, types in cervical cancer. Taken from the IARC multicentre
human papillomavirus. casecontrol study; preliminary data.23
www.jclinpath.com
In conclusion, although the specificity criteria can be viewed Follow up studies

as of secondary applicability, the global picture indicates that For cervical cancer, compliance with the temporality criteria
HPV types are not randomly associated with human cancer. A has been established by numerous cohort studies that
fair degree of specificity is consistently reported, even if the monitored women from cytological normalcy to the stage of
complexities of the type specific viral properties and of the high grade cervical intraepithelial neoplasia (high grade
organ/cell susceptibility have not been fully disclosed. squamous intraepithelial lesions (HSIL) or CIN 2/3). Monitor-
ing of women to invasive disease is not acceptable on ethical
Temporality grounds and thus that information is not available.
Of the criteria outlined by Hill and repeatedly endorsed by the Repeated sampling of women being followed for viral
IARC monograph programme and other bodies, the demon- persistence and cervical abnormalities has shown that the
stration that exposure has occurred before the diagnosis is median duration of the infections is around eight months for
considered a sine qua non condition for a risk factor and for high risk HPV types, compared with 4.8 months for the low
establishing causality. Five groups of studies have contributed risk HPV types. In two unrelated studies, the time estimates
data relevant to the temporality criterion. were fairly consistent. In one study in a high risk population in
Brazil, the mean duration of HPV detection was 13.5 months
Descriptive data for high risk HPV types and 8.2 months for the non-oncogenic
Cross sectional studies have repeatedly reported that subclini- types. HPV-16 tended to persist longer than the average for
cal HPV infections are highly prevalent in young individuals, high risk types other than HPV-16.109 The results were remark-
whereas invasive cervical cancer typically develops in the third ably similar in a student population in the USA and in the
decade and later (fig 9). The cross sectional prevalence of HPV UK.31 110 The self limiting course of most HPV infections is
DNA decreases spontaneously to a background level of 28% consistent with the cross sectional profile displayed in fig 9.
in most populations in groups that are 40 years old and above. However, the currently observed time intervals may still suffer
In countries where intensive screening of young women takes from imprecision in the estimates of time at first exposure,
place, part of the HPV prevalence reduction could be attribut- from the variability in the endpoint definition, and from cen-
able to aggressive treatment of HPV related cervical lesions. soring as a result of treatment of the early lesions.
Women who remain chronic HPV carriers are currently Follow up studies of women with and without cervical
described as the true high risk group for cervical cancer. In abnormalities have indicated that the continuous presence of
some populations, a second mode of HPV DNA prevalence has HR-HPV is necessary for the development, maintenance, and
been observed for older women (50 years and above), with progression of progressive CIN disease.110114 A substantial
uncertain relevance in relation to the risk of cervical fraction (1530%) of women with HR-HPV DNA who are
cancer.36 37 101 In all settings investigated, the point prevalence cytomorphologically normal at recruitment will develop CIN 2
of HPV DNA in the young age groups is strongly related to the or CIN 3 within the subsequent four year interval.111 115 116 Con-
sexual behaviour patterns that are dominant in each versely, among women found to be HR-HPV DNA negative and
population.102107 cytologically identified as either atypical squamous cells of
These population studies provide support for the concept undetermined significance (ASCUS) or borderline or mild
that HPV infections precede the development of cervical can- dysplasia, CIN 2/3 is unlikely to develop during a follow up of
cer by some decades. In fact, from most cancer registries, two years, and their cytology is likely to return to normal.117 118
including the USA based registries, it is well established that Women found positive for low risk HPVs rarely become
the age specific incidence of cervical cancer has a rising trend persistent carriers and their probability of progression to CIN
in the age interval 2040, and shows a plateau or continues to 2/3 is extremely low.117 119
increase smoothly after that age. Only occasionally do cases of As ongoing cohorts expand their follow up time, more pre-
invasive disease occur at earlier ages. Figure 9 shows the age cise estimates are being provided on the predictive value of
specific, cross sectional prevalence of high risk HPV DNA in a viral persistence as defined by repeated measurements of viral
screening programme in the Netherlands, and the corre- types and variants. One such cohort in Sao Paulo has shown
sponding age specific incidence rates of cervical cancer in that that the incidence of cervical lesions in women who were HPV
country. The distributions shown in fig 9 are highly reproduc- negative twice was 0.73/1000 women months. The corre-
ible in studies in other settings in high and low risk sponding incidence among women with repeated HPV-16 or
countries.3 24 106 108 However, the age specific incidence rates of HPV-18 positive results was 8.68, a 12 fold increased incidence.
invasive cervical cancer are strongly influenced by the local The OR for HPV persistence among women who were twice
impact of screening programmes in each country.3 24 106 108 HPV positive for the same oncogenic types was OR = 41.2
(95% CI, 10.7 to 158.3).120 Retrospective assessment of HPV
status using archival smears from cases of cervical cancer and
20 ASIR
controls has provided evidence that HPV DNA preceded the
development of invasive disease, and showed its value in sig-
HR HPV %
nalling false negatives smears.117 An interesting observation
15 from the same group suggests that the clearance of HR-HPV in
otherwise established cytological lesions is a marker associ-
ated with the regression of CIN lesions.118 121 Finally, the
10 persistence of HPV DNA after treatment for CIN 2/3 is an
accurate predictor of relapse, and is at least as sensitive as
repeated vaginal cytology.122
5 These results are useful in defining the clinical role of HPV
testing. However, most observations on preinvasive disease have
limitations for making inferences on cervical cancer causality.
0 This is because even in controlled settings, observations are not
2024 2529 3034 3539 4044 4549 5054 5559 6064 allowed to continue beyond the stage of HSIL/CIN 3 or
Figure 9 Age specific prevalence (%) of high rish (HR) human carcinoma in situ.
papillomavirus (HPV) DNA in 3700 women entering a screening
programme and age specific incidence rate (x105) (ASIR) of cervical Retrospective cohorts
cancer in the Netherlands. Adapted from Jacobs et al and Parkin et A particularly interesting approach to conducting follow up
al.106 108 studies of invasive cancer (as opposed to studies of CIN 3)
www.jclinpath.com
without ethical and time constraints is provided by so called or the urethra of her husband or sexual partner.104 147149 More
nested casecontrol studies. These are studies initiated sev- recently, it has been possible to confirm that male circumci-
eral years in the past that assembled and stored large banks of sion protected men from being HPV carriers and their wives
biological specimens from healthy individuals. Linkage from developing cervical cancer.150 These observations con-
studies can then identify cases of cervical cancer (or any other firmed, in terms of HPV infections, observations made over a
condition) that have occurred in the interval and the original century ago151 and a scientific hypothesis formulated almost
specimens can then be analysed for the presence of HPV 30 years ago that male sexual behaviour is a central determi-
biomarkers. HPV DNA prevalence can then be compared with nant of the incidence of cervical cancer.152 153
the corresponding prevalence in specimens of epidemiologi- In conclusion, the natural history studies of HPV infections
cally sound controls (individuals from the same cohort who satisfy in biological terms most of the observations that were
did not develop the condition under otherwise equivalent historically linked to cervical cancer. In the past two decades,
exposures). These studies have documented the existence of the cervical cancer puzzle has become a coherent description
HPV exposure years before the development of the disease, that includes the identification of HPV as the sexually
thus reproducing the conditions of a longitudinal study. With transmitted aetiological agent and the characterisation of the
this approach, a RR estimate of 16.4 (95% CI, 4.4 to 75.1) was major determinants of HPV acquisition.154
seen for invasive cervical cancer in Sweden using DNA
extracted from stored Papanicolaou (Pap) smears123 and a RR
of 32 (95% CI, 6.8 to 153) was seen in the Netherlands.117 In a Temporality. Evaluation
similar study design, an OR of 2.4 (95% CI, 1.6 to 3.7) was
obtained using serological markers of HPV exposure.124 HPV infections precede cervical precancerous lesions and
cervical cancer by a substantial number of years. The epi-
Preventive interventions demiology and the dynamics of HPV infection in
Since the late 1980s, multiple studies have evaluated HPV test- populations satisfy previous observations that related
ing as an adjunct to cytology in screening programmes. These cervical cancer to a sexually transmitted disease.
have considered HPV testing either as a triage test in cases of
mild abnormalities125127 or as a primary screening test.128130 It is
not the purpose of this paper to review this literature and excel-
lent summaries are being regularly produced and updated (see Biological gradient
later). In brief, triage studies have shown that HPV testing is This refers to the presence of a doseresponse curve indicating
more sensitive than repeated cytology in identifying underlying that the magnitude of the exposure is related to the risk of dis-
high grade lesions in women with ASCUS.114 119 121 131 132 Studies ease. This requirement, largely supported by chemically induced
that reflect primary screening conditions (in the absence of fully models of carcinogenesis, is difficult to apply in models of
randomised trials) have shown that the sensitivity of HPV tests viruses and cancer. For HPV DNA, it is difficult to measure viral
is higher than standard cytology in detecting high grade lesions, load in relation to the DNA of the cancer cells in the specimen,
whereas the specificity is age dependent. HPV tests show lower although early studies tended to show a correlation between
specificity than cytology in younger women, accounting for the HPV DNA amount and disease status.127 Some recent publica-
bulk of transient infections, whereas in older women (ages tions have provided relevant evidence using real time PCR
3035 and above) specificities tend to be similar for both methods. A study that used a nested casecontrol design found
tests.107 133 134 that cases consistently had higher viral loads for HPV-16 than
In terms of causality assessment, these studies showed that it controls, and that high viral loads could be detected up to 13
is possible to predict the concurrent presence of neoplastic dis- years before the diagnosis of cervical cancer.155 Women with
ease (usually HSIL, CIN 23, or severe dyskaryosis), or the risk high viral loads for HPV-16 had a 30 fold greater risk of
of developing it, by means of HPV DNA detection. This property developing cervical cancer than did HPV negative women. This
of the HPV test offers an indirect measurement of the strength also applied to women under the age of 25. A related paper
of the association and of the temporal sequence of the events. using the same population showed that the 20% of the popula-
tion with the highest viral loads for HPV-16 had a 60 fold higher
Determinants of HPV infection risk of developing carcinoma in situ when compared with HPV
Epidemiological studies investigating risk factors for HPV negative women.85 Of importance for clinical and screening
infection clearly and consistently have shown that the key purposes, another study confirmed that high viral loads
determinants among women are the number of sexual predicted cervical lesions and, more interestingly, that the
partners, the age at which sexual intercourse was initiated, reduction of viral load or clearance of viral DNA in repeated
and the likelihood that each of her sexual partners was an visits predicted regression of CIN lesions to normalcy.156 These
HPV carrier.103 105 135141 These are lifelong behavioural traits, studies suggest that measuring viral load, at least of HPV-16,
thus clearly preceding the development of cervical cancer. may distinguish between clinically relevant infections and those
The role of men as possible vectors of HPV was measured in that are unlikely to progress. However, in contrast to the above
the early epidemiological studies by questionnaires that asked results one large prospective study in Portland USA, using
about the sexual behaviour of the husbands or sexual partners quantitative hybrid capture, did not find viral load to be a deter-
of patients with cervical cancer and controls. In addition, more minant of risk of future CIN 3 (A Lorincz et al, unpublished data,
recent studies had the ability to measure HPV DNA in exfoli- 2002). More research is needed to validate these methods and
ated cells from the penile shaft, the coronal sulcus, and the the results need to be extended and confirmed in clinical
distal urethra.142146 studies.157
These and other studies consistently showed that the risk of
cervical cancer for a given woman can be predicted by the
sexual behaviour of her husband as much as her own sexual Biological gradient. Evaluation
behaviour. In populations where female monogamy is
dominant, the population of female sex workers plays an The risk of cervical cancer may be related to estimates of
important role in the maintenance and transmission of HPV viral load. The technology to estimate viral load is being
infections. Moreover, the probability that a woman is an HPV developed and compliance with the biological gradient
carrier and her risk of developing cervical cancer have been requirement needs to be further validated.
shown to be related to the presence of HPV DNA in the penis
www.jclinpath.com
The pattern of transcription changes as the lesion increases in

HSIL Cancer
severity. All open reading frames (ORFs) are expressed in early
Normal cell lesions but the expression of ORFs E4 and E5 is not found in
LSIL HSIL Cancer many invasive cancers.165 (7) The E6 and E7 ORFs contain the
transforming ability of HPV. These are always intact and are
HPV Viral Neoplastic consistently expressed in cervical cancer cell lines, in cells
infection persistence progression transformed by HPV, and in HPV associated cancer tissue.
Cell mediated They are transcribed at higher levels in high grade lesions than
immunity in low grade lesions.165 168 (8) In most cell lines and in many
Viral Viral DNA Genetic HPV associated cancers, the HPV DNA is integrated into the
clearance integration instability cellular DNA. HPV-18 is nearly always integrated, whereas
HPV-16 can be found episomally or in the integrated
High amounts
of E6, E7 form.76 169 170
Defects in
antigen In reviewing work on the molecular genetics of cervical car-
Type specific
immunity presentation cinoma, Lazo indicated different mechanisms of cancer induc-
p53 RB tion. The effects of E6 and E7 on host regulatory proteins can
be considered to be HPV related mechanisms. An additional
Figure 10 Mechanisms of human papillomavirus (HPV)
effect could be expected from the consequences of viral
carcinogenesis. HSIL, high grade squamous intraepithelial lesion;
LSIL, low grade squamous intraepithelial lesion; RB, retinoblastoma integration and the specific impact on the integration sites.
gene. The third mechanism, which may or may not be related to
HPV, is the accumulation of the cellular genetic damage
needed for tumour development. The existence of this mech-
Biological plausibility and coherence
anism is strongly suggested by the observations of recurrent
The mechanisms by which HPV induces cancer in humans
losses of heterozygosity and by recurrent amplifications in a
and the molecular genetics of the process are being
large fraction of cervical carcinomas.160 The role of non-
investigated intensively and excellent reviews are readily avail-
identified tumour suppressor genes is also suggested by
able.12 16 17 158161 These investigations provide additional evidence
experiments showing that the tumorigenicity of HeLa cells
on the causality of the association by describing viral and host
could be suppressed by fusion with normal fibroblasts or
interactions leading to cell transformation and malignancy. Of
keratinocytes, and that the tumorigenicity of SiHa cells was
the criteria outlined in table 1, both the biological plausibility
suppressed by the introduction of chromosome 11 via micro-
and understanding of the mechanisms are in rapid expansion
cell transfer technology.171173 Similarly, the immortality of Hela
as a consequence of developments in molecular methods and
and Sitta cells was suppressed by the introduction of chromo-
technology.
somes 3, 4, and 6.174 175
Figure 10 shows in a schematic manner some of the major
Although a review of the field is far from the purposes of
components of the transition from HPV infection to cervical
this discussion, it seems quite clear that the biology of cervical
cancer. Whereas transient infections are largely subclinical,
cancer in relation to HPV has become a paradigm of viral
progression is closely related to the persistence of viral DNA.
mediated oncogenesis. The work being regularly published has
This process goes frequently with viral disruption in the E1/E2
clearly shown that the viral DNA detected by epidemiological
regions and integration into the cellular DNA. E2 disruption
studies is not a passenger infection of the cancerous tissue, but
releases the viral promoters of E6 and E7 and increases the
expression of these transforming genes. The E6 and E7 viral a biologically meaningful association.
proteins are capable of selectively degrading the p53 and
retinoblastoma gene (RB) products, respectively, leading to
Biological plausibility and coherence. Evaluation
inactivation of two important cellular negative regulatory
proteins.
Some characteristics that provide support for the role of The association of HPV DNA in cervical specimens and
HPV in the induction of cervical cancer were recently cervical cancer is plausible and coherent with previous
outlined.17 Accordingly, the causal nature of this association is knowledge. This includes in vitro experiments, animal
indicated by: (1) the regular presence of HPV DNA in the neo- experiments, and observations in humans. Novel criteria
plastic cells of tumour biopsy specimens; (2) the demonstra- of causality are being proposed and tested as molecular
tion of viral oncogene expression (E6 and E7) in tumour technology develops and is introduced into epidemiologi-
material (but not in stromal cells); (3) the transforming prop- cal research protocols.
erties of these genes (E6 and E7); (4) the requirement for E6
and E7 expression to maintain the malignant phenotype of
cervical carcinoma cell lines; (5) interaction of viral oncopro- Biological mechanisms of HPV carcinogenesis
teins with growth regulating host cell proteins; and (6) epide- In previous decades, our understanding of cancer pathways
miological studies pointing at these HPV infections as the was rudimentary and often incorrect. In the face of such
major risk factors for cervical cancer development. uncertainty, arguments based on assumptions of molecular
In their review, Shah and Howley16 provided references to biology were not particularly convincing. However, with the
some of the key experiments that exemplify most of the large body of work now available it is possible to develop a
requirements indicated by zur Hausen,17 namely: (1) The reasonable understanding of the ways in which cancer may
genomes of HPV-16 and HPV-18 are capable of immortalising develop and ways in which HPV infection can drive the proc-
human keratinocytes in cell culture, whereas the DNA of the ess. Thus, we can say with some confidence that it is plausible
low risk HPV types (6/11) do not.162 (2) In raft cultures, the for HPV to cause cervical cancer and, furthermore, we can
oncogenes of the high risk HPV types induce morphological describe with reasonable clarity the general steps by which
changes that closely resemble preinvasive cervical HPV may do so. Of course there is a lot of detail still to be
lesions.163 164 (3) In HPV associated lesions, the viral genome is revealed, but we are well on our way to a factual basis for
present in every cell and is always transcriptionally active.165 understanding carcinogenesis rather than the guesswork and
(4) The viral genome is present in the original tumour and in rudimentary models of just two decades ago.
metastases.166 (5) Most of the cell lines established from cervi- What follows is a description of selected information that
cal cancer contain either HPV-16 or HPV-18 genomes.167 (6) may illuminate salient aspects of the natural history of HPV
www.jclinpath.com
and reasons why a mostly benign infectious process some- in the liberation of E2F transcription factors, which play key
times results in malignancy. It must be understood by readers roles in promoting host cell and viral DNA synthesis. E7 also
that the pathways discussed are based on extensive experi- binds and activates cyclin complexes, such as p33cyclin
mentation in biopsied human tissues, in tissue culture, and dependent kinase 2,192 which control progression through the
other kinds of molecular biology systems. However, many cell cycle. E6 protein can overcome the p53 protective control
details of pathway modifications and aberrant pathway effects pathways,178 which are important in preventing the genetic
are speculativethey have not been shown to occur inside the damage that may lead to cancer.
relevant precancerous and cancerous tissues of living hosts. HPV genomes attach to host chromatin via the E2 protein
However, despite the many holes and inconsistencies, the and replicate at a steady state, once for each cell division.186188
models are still quite compelling and cohesive in facts. In It has been speculated that a benefit of this tethered theta
future, we expect to see these molecular models being tested mode of replication is that the loss of HPV DNA from cells by
in human subjects. non-disjunction is minimised and the presence of low
Essentially all HPV types produce warty lesions but only amounts of HPV DNA in cells is less likely to be detected by
high risk types promote the development of cervical cancer to intracellular interference mechanisms that could trigger
any appreciable extent. Such differences between HPV types apoptosis. As cells differentiate and move to the surface there
may seem surprising, given the high DNA and structural is a normal differentiation and maturation process that leads
similarities. However, a large functional divergence caused by to pyknotic condensed cells that slough from the tissue. How-
small genetic changes is the norm in many biological systems. ever, in virally infected tissues there is activation of unsched-
Variations in carcinogenic potential among HPVs are princi- uled DNA replication in some spinous cells, accompanied by a
pally governed by the E6 and E7 proteins; specifically by the switch in viral DNA replication to the rolling circle mode,
capacities of these proteins to interact with and alter or which leads to the production of viral progeny.179182 This reac-
destroy key cell cycle regulatory molecules.176178 tivation of DNA synthesis can be detected by the presence of
The progress and outcome of an HPV infection depend on punctate proliferating cell nuclear antigen tissue staining (a
HPV type, anatomical location, and the nature and timing of protein with a key role in DNA replication) and the presence
local cellular and tissue influences.179183 Virions access basal of HPV virions in a subset of upper layer cells.
and parabasal cells in areas of erosion and viral DNA enters HPV E7 proteins of both low and high risk types have an
the cell nuclei. Establishment is tied to the tissue proliferative ability to promote unscheduled DNA replication in spinous
activity of epithelial cells and, in the case of extensive tissue cells.179182 193 It is believed that the extent to which E7 stimu-
repair, the viral infection can become widely disseminated. lates cells, and the tissue location at which such stimulation
Persistence in keratinocytes is variable and related to viral occurs, is important to malignant progression. Spinous cells
type.109 112 184 Finally, integration of viral DNA may occur, respond to E7 by the production of a cyclin kinase inhibitor,
resulting in lifetime persistence of certain viral genes in the p21cip1, translated from sequestered RNA. In basal and
cell. In the cervix, detectable infection by low risk HPV types is parabasal cells existing mRNA for p21cip1 is not available
of relatively short duration, whereas infection by most high and the protein is typically produced from new transcripts
risk types lasts longer. On occasion, such infections may stimulated by p53; however, if p53 is inactivated by E6 the
become persistent and last years or even decades; it is in these p21cip1 cannot be made. Spinous cells thus have a control
cases that the risk of cancer is increased. advantage lacking in basal cells. Interestingly, high amounts
The establishment of HPV infection can be modulated by a of E7 can bind and block the activity of p21cip1. The relative
competent and primed immune system. In vitro experiments amounts of E7 and p21cip1 are believed to determine
have revealed an inverse association between the degree of whether cells re-enter S phase and replicate viral DNA or
cervical neoplasia and interleukin 2 production by peripheral whether cells block viral production. The inspection of tissues
blood mononuclear cells in response to HPV-16 E6 and E7 reveals a mutually exclusive set of spinous cells with high
peptides.184 185 Women with CIN 3 or cancer appear to have a amounts of either E7 or p21cip1.179182 Cells in which E7 over-
decreased ability to mount a T helper cell type 1 (Th1) medi- comes the p21cip1 block can become koilocytes and produce
ated immune response to HPV E6/E7, compared with women viral particles. This balance can explain the patchy expression
with CIN 1 or HPV infected women without lesions.184 185 It is of the HPV effect in infected tissues.177 194 A key function of the
possible that a Th1 mediated cellular immune response could E6 oncoprotein is the destruction of p53, a protein that is
play a role in host immunological control of HPV infection and activated upon phosphorylation via DNA damage sensing
that lack of such an appropriate response may predispose to proteins. Activated p53 stops the cell cycle in the G phase as a
the progression of cervical disease. result of direct stimulation of p21cip1 by this molecule. Alter-
If HPV enters immature metaplastic basal stem cells that natively, in the case of major DNA damage or high amounts of
are actively dividing the infection can become widely viral replication, p53 may activate an apoptotic pathway.195 196
dispersed and persistent. In contrast, if infection occurs only E7 also interferes with alternative non-p53 dependent apop-
in the parabasal transit amplifying cells the infection may totic pathways. Thus, in the case of E6 mediated destruction
become transient or quasi-persistent.179182 The size, histologi- of p53, cells are unable to prevent the accumulation of genetic
cal grade, and duration of lesions can depend on the number mutations. Cells have other defensive homeostasis mecha-
and types of cells that become infected by HPV. In either tran- nisms, but E6 and E7 have counter functions that can lift the
sient or persistent infection there may be periodic viral blocks and direct cells to enter S phase. Therefore, it appears
genome amplification, depending on the activity of the that the development of malignancy is a consequence of an
infected daughter cells, which can lead to variable detection of aberrant hostvirus interaction. A potentially important
the lesion by HPV DNA or Pap tests. event in this process is the aberrant regulation of E6/E7
There is an important difference in hostvirus interactions expression.182 187 197 198 In low grade CIN lesions, E6/E7 expres-
of carcinogenic HPV types and low risk HPV typesthe sion is mainly found in differentiating spinous cells that have
former have activities that more strongly interfere with a set of withdrawn from the cell cycle. In high grade CIN lesions and
host cell cycle control mechanisms. It is therefore useful to cervical carcinomas, strong E6/E7 expression is seen in the
consider the effects of carcinogenic HPV types. HPV initially proliferating cell compartments.
replicates to reach about 2550 genomes/cell.179182 186188 The HPV DNA is frequently integrated into the host genomes in
process by which this occurs is tied to the activities of four cancers in such a way that the E2 repressor protein is inactive
multifunctional viral proteins E1, E2, E6, and E7.176178 189191 One and allows overexpression of E6 and E7.170 199 200 In cases where
key activity of E7 is to overcome the pRB tumour suppressor HPV integration is not detected, other mutations can be
block.176 Binding of E7 to pRB and its related members results shown in the E2 protein or in repressor functions, such as
www.jclinpath.com
YYI sites,201203 which appear to allow continuous expression of

Experimental evidence. Evaluation
the E6 and E7 oncoproteins. Yet another way in which E6 and
E7 could be overexpressed in proliferating cells is by the gen-
eration of chimaeric HPV mRNAs encoding the E6 and E7 Experimental evidence shows that species specific papillo-
proteins that have host sequences at their 3 termini. Such maviruses induce papillomas and cancers in the suscepti-
RNAs are frequently more stable and allow more protein to be ble host.
synthesised.191
In persistent HPV lesions, viral genomes in the basal cells
continue to stimulate the cells to ignore the DNA damage that
may be accumulating.204 205 Cell stimulation by E6 and E7 of high Analogy
risk carcinogenic HPV types produces clones with an extended Analogy is no longer viewed as an essential criterion for cau-
life span that have passed a point called mortality 1 or M1, sality assessment. As proposed by Hill, analogy would allow
although the cells are still not immortal.198 An important step in several considerations in the HPV and cervical cancer model
immortalisation is related to telomeres. Normally, telomeres that could be mentioned:
shorten every cell generation and once they reach a critical (1) If other DNA viruses (such as HBV) can induce cancer in
length the cells die. Telomere length is maintained by telomer- humans, HPV could too.
ase, which in combination with a capping function, can stabilise (2) If animal PVs can induce papillomas and carcinomas in
and even lengthen telomeres, allowing cells to continue several animal models, HPV could do the same in humans.
dividing. E6 can activate telomerase and additional cell
(3) If viral products that interfere with p53 and RB in animal
mutation(s) can then stabilise the telomeres and allow cells to
models, such as the simian virus 40 large tumour antigen
pass a second stage called mortality 2 or M2. It is not known
(SV40 tag), induce cancers in animal species, HPV could do so
how many additional independent mutations are needed to
in humans.
transform immortalised cells fully to malignancy.191 One set of
mutations allows the cell to break through the basement mem-
brane by eliciting a set of novel proteases. Another mutation(s)
Analogy. Evaluation
allows cells to move in the dermis.206 Undoubtedly, metastatic
cells have accumulated many additional mutations that allow
them to create their own microenvironment for survival in for- The HPV and cervical cancer model is analogous to many
eign parts of the body. other examples of PV induced papillomas and carcinomas
and cancers caused by other viruses.
Biological mechanisms of HPV carcinogensis. Exclusion of alternative explanations

Evaluation In the early studies of HPV and cervical cancer, and currently
in most studies, a fraction of cases are labelled as HPV negative
and investigated under the hypothesis that HPV negative cases
The natural history of HPV infection and its relation to can- are a true biological entity.210 211 The proportion of such cases
cer development is being described by molecular technol- tends to be higher in studies of preinvasive neoplasia.212 213 As
ogy. These investigations indicate that the induction of a consequence, there was some uncertainty in the interpret-
cancer by HPV is mediated by viral interference with ation of the results until the early 1990s.2 214 215 In many stud-
essential regulatory mechanisms of cellular growth, DNA ies, HPV negative cases were frequently compared with HPV
repair, and immunological escape. The alternative hypoth- positive cases in relation to their epidemiological profile. In
esis of HPV being an opportunistic passenger in the broad terms, it was clear that HPV negative cases retained
tumoral tissue is no longer tenable. the same traits as the rest of the cases (similar age, high
number of sexual partners, young age at first sexual
intercourse, long term use of contraceptives, high parity, etc).
These results strongly suggested that the apparently HPV
Experimental evidence
Intervention studies in human populations would require the Exclusion of alternative explanations. Evaluation
demonstration of a reduction in cervical cancer incidence and
mortality following a reduction in the incidence of HPV infec- Alternative (non-HPV related) hypotheses to explain a frac-
tion in the underlying population. Prophylactic vaccines are tion of cervical cancer are not being proposed. The
being developed and vaccination trials are in advanced hypothesis that a fraction of 510% of cases may occur in
planning phases to evaluate such expected effects. the absence of HPV should be retained for research
Review of the animal models for HPV related lesions, purposes. Public health recommendations targeting a
including cancer, is out of the scope of this report. In brief putative proportion of HPV negative cervical cancer cases
however, there are animal models of PV infections that are not supported by current results and are not justified.
induce warts and carcinoma in the skin, mucosa, and the
digestive tract. The bovine PV (BPV) induces fibropapillomas
in the skin and carcinoma of the digestive tract in animals
exposed to bracken fern in their diets. Cottontail rabbit PV Overall evaluation
(CRPV) induces papillomas that could evolve into carcinomas
either spontaneously or following the application of various Systematic review of the causality criteria strongly
promoters. Canine oral PV (COPV) causes oral papillomas indicates that the association of HPV and cervical cancer
that do not undergo malignant transformation. Several of is causal in nature. The association is very strong, consist-
these models are actively being used in the preparation of ent, specific, and universal. HPV infection precedes prein-
vaccines (for a review see Saveria Campo207 and Tindle).208 vasive disease and the evidence for biological plausibility
Many other papillomaviruses have been identified in animal of the association is persuasive beyond reasonable doubt.
models.209
www.jclinpath.com
negative cases were also related to an STD pattern; however, by the IARC in 1991. The characteristics of the protocol, test-
none of the known sexually transmitted agents that had occa- ing system, and some of the key results have been described in
sionally been associated with cervical cancer is able to satisfy the country specific publications.3 2430 218 219 The pooled analy-
the causality criteria outlined in table 1. In the past decade ses for each risk factor will support their conclusions on
there has not been a hypothesis supported by sound analysis restricted to HPV positive cases and controls. For this
epidemiological or biological data indicating that the aetiology discussion, we will summarise the key preliminary findings.
of cervical cancer could be independent of HPV.
It should be retained as a scientific and research option that Oral contraceptives
the existence of some non-HPV related cervical cancer cases is The IARC multicentre study included 1768 cases and 262 con-
plausible. The grounds for such a statement rely on the trols that were positive for HPV DNA. OC use ever was asso-
following facts: (1) Epithelial cells are capable of developing ciated with a significant increase in risk (OR = 1.47; 95% CI,
into cancer cells and cancer growths in all human tissues 1.02 to 2.12). Use of OCs for less than five years was not related
regardless of a known, viral or non-viral, cause. Thus, cells in to cervical cancer (OR = 0.77; 95% CI, 0.46 to 1.29). The risk
the human cervix might have this capability too. (2) Cell genes increased for five to nine years of use (OR = 2.72; 1.36 to 5.46)
that are involved in HPV related carcinogenesis should be able and for more than 10 years (OR = 4.48; 2.24 to 9.36). The
to generate spontaneous or induced mutations leading to can- conclusion of this analysis indicates that the use of OCs for
cer in the absence of HPV. Available evidence suggests that this five or more years is a cofactor that increases up to fourfold the
event is rare within the life expectation of the human popula- risk of cervical cancer among women who are carriers of HPV
tion. (3) Relatively few cases of cervical cancer in very old DNA. These results provide a consistent summary of the pre-
women have been investigated. It is likely that the non-HPV vious publications of the individual studies included in the
related cancers currently occur very rarely and probably clus- multicentre project.220
ter in very old women. (4) Non-epithelial cancers do occur in The evidence for an association of cervical cancer with the
the cervix at a low frequency. use of oral or other hormonal contraceptives is not entirely
consistent. Several studies that investigated HPV positive
women found no associations or only weak associations in
ALTERNATIVE MODELS OF CERVICAL CANCER subgroup analyses.4 5 217 221 222 These apparently conflicting
AETIOLOGY results may reflect the increased cytological surveillance of
This section will discuss in some detail the role of factors other women who are taking OCs in developed countries and the
than HPV in cervical cancer and evaluate them under use of different case definitions (from ASCUS up to HSIL/CIN
Rothmans model of necessary and sufficient causality. 3 as opposed to cervical cancer) in cohort studies.
The IARC monograph programme reviewed the evidence
Other risk factors for cervical cancer for OCs and concluded that, for the associations with cervical
Most of the sexual behaviour parameters that were linked to cancer, biases related to sexual behaviour, screening, and other
cervical cancer in the past are being re-evaluated in studies that factors could not be ruled out as possible alternative explana-
considered the strong influence of the presence of HPV biomar- tions. One of the difficulties encountered in the conduct and
kers. Soon after the introduction of HPV testing into research analysis of epidemiological studies is the proper assessment,
protocols, it became clear that some of the key risk factors that through questionnaires, of the correlates of HPV exposure
reflected sexual behaviour, such as the number of sexual (sexual behaviour) and screening history. However, OCs were
partners, merely reflected the probability of HPV exposure. evaluated as human carcinogens (class 1) by the monograph
Other factors, such as the estimates of age at first exposure (as because of the evidence linked to liver cancer.223
indicated by age at first sexual intercourse or at first marriage) Because of the potential public health importance of
are still under evaluation. In addition, several environmental confirming an interaction between the long term use of OCs
factors that were historically related to cervical cancer are and HPV infections in the development of cervical cancer,
currently being assessed. These include hormonal factors (use efforts are now being devoted to confirm the results in differ-
of OCs and multiparity) other STDs (HSV-2, Chlamydia ent populations.
trachomatis, and occasionally other STDs), cigarette smoking,
and dietary factors. A special consideration should be given to Parity
exposure to HIV and to other situations of immunosuppression, In the IARC multicentre study, HPV positive women who
which will not be dealt with in detail in this discussion. reported seven or more full term pregnancies had a fourfold
Investigation of the role of such factors, once the central role increased risk of cervical cancer compared with similar HPV
of HPV was recognised, generated an analytical situation with positive women who were nulliparous (OR, 3.8; 95% CI, 2.7 to
few precedents in cancer epidemiology. This derived from the 5.5). There was still a twofold increased risk when women
observation that HPV DNA was almost always present in speci- reporting seven or more pregnancies were compared with HPV
mens from cervical cancer. Moreover, in one large study it was positive women who reported one to two full term
shown that the HPV DNA negative cases that were identified pregnancies.224
were largely false negative cases, in which HPV DNA was Similar results were obtained in Costa Rica,225 Thailand,5
initially undetected because of specimen inadequacy, a relative and among women with preinvasive disease in the Portland
predominance of integrated HPV-18, or technological study.4 In Denmark and in the Manchester cohort study, two
inadequacy.216 Therefore, it was thought inappropriate to allow populations with low parity,217 226 the effects were less visible.
in the statistical analyses a comparison group of HPV negative In another high parity countryHondurasthe effects of
cases even if, in analytical terms, this group existed as a small parity were not significant among the HPV positive cases and
fraction (between 5% and 10% of the cases) in most studies. controls.227
It soon became a standard procedure in the reports of case
control studies to include a separate analysis, or to restrict the Smoking
analyses to HPV positive cases and controls, to assess the con- Smoking has been related to cervical cancer since the late
tribution of additional factors to the risk of cervical 1970s, based upon the correlations seen between cervical can-
cancer.3 4 40 217 Given that, few studies were large enough to cer incidence and the incidence of other tobacco related
generate a sufficient number of HPV positive controls. In rela- cancers.228 The 1986 IARC monograph on smoking considered
tion to invasive cervical cancer, the leading project was the that the evidence available for cervical cancer was insufficient
multicentre casecontrol study on cervical cancer undertaken to rule out confounding with sexual behaviour traits, strongly
www.jclinpath.com
related to both smoking and cervical cancer.229 An extensive period.233 234 The prognosis of the EBV infection in relation to
review of the relation between smoking and cervical cancer Burkitts lymphoma is related to age at first infection and
was published in 1998, including eight cohort and 44 probably to the concurrent exposure to malaria.235 Similarly,
casecontrol studies. The report concluded that the associ- most studies have shown that the risk of cervical cancer is
ation was largely consistent in studies that adjusted for HPV related to age at first sexual intercourse. More recent studies
DNA or restricted analyses to HPV positive women. The mag- that included HPV measurements have shown age at first
nitude of the risk for current smokers was of the order of one intercourse to be a surrogate measure of age at first HPV
to threefold; the ORs tended to be higher in more advanced exposure. However, definite evidence that cervical cancer pro-
preinvasive neoplasia, and in several studies a doseresponse gression is linked to age at first HPV exposure has not been
relation with the amount of tobacco consumed was seen.230 A provided. It has been proposed that the developing cervix (at
recent review of the evidence conducted by the Surgeon Gen- the age of peri-menarche) or the healing cervix (as a
eral in the USA retained the hypothesis that a causal consequence of deliveries, cervical trauma, or any other STD
association between cigarette smoking and cervical cancer infection) are high risk situations for an HPV infection to
was plausible. However, the report indicated that the extent to reach the basal layer and establish a persistent infection.
which cigarette smoking could be considered independent of Some STDs, including HIV,12 42 have repeatedly been found
HPV could not be definitively assessed.231 to be associated with cervical cancer. The multicentre study
Restricting the review to studies that evaluated HPV exposed found a twofold increased risk for the presence of antibodies
women, the preliminary pooled results of the IARC multicentre to C trachomatis (OR = 2.1; 95% CI, 1.1 to 4.0).236 Antibodies to
study found that ever smoking was associated with a twofold, HSV-2218 236 and non-specific inflammatory changes237 have also
significantly increased risk of cervical cancer, with a significant been related to modest increases in risk among HPV positive
doseresponse (M Plummer et al, unpublished, 2002). These women. Other environmental risk factors under evaluation
findings are consistent with those found for current v never are socioeconomic status238 and nutritional factors.239
smoking among HPV positive women in the Costa Rica study
(OR = 2.3),225 the Portland study (OR = 2.7 for CIN 23),4 the Limitations of the evidence
Copenhagen study (OR = 1.9),232 and the Manchester study One limitation of the available studies and of the summary
(OR = 2.2).217 These recent studies are providing growing presented here is the crudeness of information available on
evidence on the carcinogenic effect of cigarette smoking in variables that may modulate the effect. This makes it difficult
women with HPV infection. However, the mechanisms by which to explain in finer detail the geographical variation in cervical
cigarette smoking may affect cervical cancer (such as a direct cancer incidence and the variability in risk estimates reported
effect of the tobacco metabolites, or an indirect effect related to in different populations. For example, the inconsistencies in
tobacco induced immunosupression or to reduced dietary anti- finding associations of cervical cancer with the use of OCs may
oxidants) remain elusive and further studies are warranted.230 be explained by factors related to the intensity, duration, or the
The general considerations in relation to any of the cofactors chemical composition of the exposure. Factors that are largely
considered and the carcinogenic effects of HPV are threefold. country specific, such as the variability in time since
First, in the analyses restricted to HPV positive women, the widespread introduction, availability of combined versus
magnitude of the RR estimates are moderate (range 2 to 6 for sequential products, oestrogen/progesterone doses, etc, could
extreme categories of exposure) for any of the variables when be important determinants; yet these would only partially
compared with the risk estimates for HPV (ranges 50 to > 500). account for the associations observed.
Second, the results in the literature concerning these factors A second limitation is that many studies did not consider host
remain inconsistent, again in contrast to the striking consist- factors in relation to HPV. Human major histocompatibility
ency of the associations found for HPV DNA. Third, the complex (HLA) types and p53 polymorphisms are being actively
interpretation of the risk estimates for cofactors should consider investigated because of the indication that they may play a role
that, as far as we understand them today, these are conditional in the natural history of HPV infections.158 240 The putative effect
effects that apply if, and perhaps only if, HPV DNA is present in of such individual susceptibility factors is not adjusted for in the
the specimen. Therefore, there is no convincing evidence that risk analyses of the environmental factors. However, at this
any of the three factors discussed are truly independent risk stage, there is limited information to substantiate an effect of
factor for cervical cancer. Some authors view these results (in these host factors as independent of HPV.241244
conjunction with the risk analyses of HPV exposure) as indica- There is a clear need to pursue research in understanding
tive of a promoter effect of OC use, smoking, or multiparity from the factors that determine whether a woman with an HPV
HPV infections to HPV related neoplasia. infection will clear the infection or become a persistent carrier.
Furthermore, studies are still needed to explore whether
Other factors additional factors play a role in determining neoplastic
The role of additional factors is being actively investigated. progression and how best to use them in screening and patient
Age at exposure has been shown to be a strong determinant of management.
the prognosis of a carcinogenic viral infection in relation to
cancer development. For example, HBV induced liver cancer is Likely components of sufficient causes of cervical cancer
closely related to age at infection, with the strongest risk The purpose of this section is to try to substantiate the
linked to infections occurring within the perinatal components of plausible sufficient causes of cervical cancer
Table 5 Risk factors for cervical cancer among human papillomavirus positive
women*
Risk factor Risk exposure Ref
HPV DNA in cervial exfoliates Positive for high risk types Negative
Use of oral contraceptives 5 or more years of use Never
Smoking Ever Never
Parity 5 or more pregnancies None or 12
*References adapted from the results of the IARC multicentre study.23
www.jclinpath.com
Table 6 Prevalence of four risk factors in 1490 cases of invasive cervical cancer
included in the IARC multicentre casecontrol study12 and approximately 800 cases
included in the IBSCC tested for human papillomavirus (HPV) DNA in 1995 and
1999
Study 1 Multicentric Study 2 IBSCC (1995) Study 3 IBSCC (1999)
Presence of risk factor N % N % N %
HPV 341 22.9 186 23.1 186 23.1

HPV + OC 54 3.6 25 3.1 25 3.1
HPV + Smk 115 7.7 96 11.9 978 12.0
HPV + Preg 587 39.4 302 37.5 322 40.0
HPV + OC + Smk 20 1.3 25 3.1 28 3.5
HPV + OC + Preg 72 4.8 14 1.7 16 2.0
HPV + Smk + Preg 223 15.0 82 10.2 88 10.9
HPV + OC + Smk + Preg 21 1.4 7 0.9 7 0.9
OC 2 0.1 0 0 0 0
OC + Smk 0 0 3 0.4 0 0
OC + Preg 0 0 2 0.2 0 0
OC + Smk + Preg 0 0 0 0 0 0
Smk 8 0.5 1 0.1 0 0

Smk + Preg 9 0.6 6 0.7 0 0
Preg 15 1.0 21 2.61 1 0.1
None of the four 23 1.5 25 3.1 2 0.2
Total number of cases 1490 100 806* 100 806* 100
*The total number in the column is not 806 because of missing values in some of the variables. References:
study 1, Muoz et al (2000)32; study 2, Bosch et al (1995)34; study 3, Walboomers et al (1999).216
HPV, HPV DNA positive; OC, duration of use of hormonal contraceptives >6 years; Preg, >5 pregnancies
(live or still born); Smk, smoker (ever). IARC, International Agency for Research on Cancer; IBSCC,
international biological study on cervical cancer.
Table 7 Theoretical models of sufficient causes for cervical cancer and

environmental risk factors
Sufficient cause Sufficient cause Sufficient cause Sufficient cause
Model 1 Model 2 Model 3 Model 4
HPV DNA in cervical HPV DNA in cervical exfoliates and any High parity None of the four
exfoliates other combination of factors Smoking factors
HPV, human papillomavirus.
according to Rothmans scheme,15 and based upon data from Using these distributions as a realistic guide, some theoreti-
large series of cases examined with reliable HPV DNA testing cally sufficient causal models could be proposed as plausible
systems. In the literature on cervical cancer, the independent for cervical cancer (table 7).
risk factors that have been repeatedly identified and at least Each combination of factors as described could theoretically
partially confirmed by some of the most recent studies could lead to cervical cancer and, under each model, the occurrence
be simplified to a dichotomous categorisation (table 5). of the disease may be further modulated by the presence or
Table 6 shows the distribution of approximately 2500 cases absence of the additional factors. These may determine
of invasive cervical cancer from over 25 countries classified parameters such as the age of onset, the aggressiveness, or
according to their exposure to every possible combination of other characteristics of tumour behaviour.
the four factors considered above. Approximately 1500 of Some of the categories in table 6 are compatible with the
these cases were included in the IARC multicentre study hypothesis that a fraction of cases of cervical cancer (47%)
(table 6, study 1). Two of the study sites (Colombia and Spain) are independent of HPV and even that some 12% are
have been removed from the series because HPV testing was independent of the four environmental risk factors considered
in table 6, either alone or in combination. To explain these
performed with the early MY09/11 PCR system. The remaining
groups, two alternatives could be considered.
study sites were investigated using the GP5+/6+ PCR, which
The first alternative is that some cervical cancer cases truly
has higher sensitivity for some HPV types (see also HPV-DNA
occur unrelated to any of the combinations of the four factors
prevalence by study site in fig 3). Table 6 also includes an considered. Some previous analyses that have compared the
equivalent distribution for 800 cases included in the inter- epidemiological profile of HPV positive and apparently HPV
national biological study on cervical cancer (IBSCC), analysed negative women with cervical cancer did not find significant
with a sequential combination of both techniques. Study 2 in differences in their sexual or reproductive behaviour. These
table 6 presents the distribution of the first testing exercise types of analyses were reported at the time that HPV DNA
with the MY09/11system34 and study 3 presents the distribu- detection technology was of limited sensitivity and 3040% of
tion of the same cases when the second testing round with cervical cancer cases were reported as HPV negative.3
GP5+/6+ and CPI/II for the HPV negative cases was Therefore, the HPV unrelated hypothesis would argue in
completed.216 favour of a role for other STDs. Current results of some major
www.jclinpath.com
studies found small and still inconsistent effects from

Likely components of sufficient causes of cervical
previous exposures to C trachomatis236 245 246 and/or HSV-2 infec-
cancer. Evaluation
tions (J Smith et al, unpublished data, 2002). However, these
effects are mostly seen in the presence of HPV DNA.
Conversely, the distributions presented in table 6 may still HPV is a necessary, non-sufficient cause of cervical cancer.
suffer from misclassification of any of the variables consid- In perhaps a quarter of a large series of cervical cancer
ered, notably of HPV. A particularly relevant exercise was cases, HPV was the only environmental risk factor
undertaken to verify technical HPV misclassification with the identified. It is likely that the additional factors that
cases included in the IBSCC study (studies 2 and 3 of table 6). combine with HPV to complete a sufficient cause in such
After the key results were published reporting an HPV DNA cases are of endogenous rather than of environmental
prevalence of 93.0%,34 the apparently HPV negative cases were nature.
further investigated. The second level evaluation included: (1)
A comparison of the serological and epidemiological profiles
of the HPV DNA negatives compared with the HPV DNA RECOGNITION OF THE CAUSAL LINK BETWEEN
positives. (2) The pathology of the specimen was reviewed. HPV DNA AND CERVICAL CANCER
Furthermore, serial cuts from the paraffin wax blocks were The evidence for a qualification of the association between
used to ensure that the specimen used for HPV DNA testing cervical cancer and HPV DNA as causal and the justification
included neoplastic tissue. (3) A different technology was for intervention strategies based upon this conclusion were
used for HPV DNA testing using shorter primers in the E7 debated in the late 1980s and early 1990s. At the time, the HPV
region and two sets of consensus primers (CPI/II and testing systems suitable for use in large epidemiological stud-
GP5+/6+). In brief, the results of the study indicated that ies had limited sensitivity and specificity. The results
most of the HPV negative cases were in fact HPV positive, and generated were difficult to interpret and skepticism was
that the HPV types newly identified were the common types, overtly expressed in the international
with an over-representation of HPV-18.216 The distribution of literature.2 55 72 214 215 226 247 248 The discussion initiated at the time
risk factors in study 3 of table 6 thus represents the conclusion to decide on the right time to make public health decisions is
of the large IARC effort in HPV and cervical cancer. Based an ongoing process with occasional spikes.249 250 Some of the
upon the final results of this study and the considerable body landmarks that may be used as examples are the scientific
of evidence accumulated from other studies at the time, it was publications coordinated by the IARC that involved a large
proposed that HPV DNA is a necessary cause of cervical cancer number of scientists dedicated to HPV research.
and that HPV DNA belongs to (is a component of) all reason- In 1989, one IARC scientific publication still retained the
ably conceived environmental sufficient causes. hypothesis that HPV DNA could be a passenger and that
The key consideration from this discussion, however, is that superinfection of the neoplastic tissue could not be safely
ruled out as an alternative explanation.251 In 1992, a follow up
with the current degree of understanding it would be difficult
publication again reviewed the evidence and concluded that
to justify an aetiological model that would not include HPV
previous incoherent epidemiological results could be ex-
DNA for most cervical cancer cases occurring worldwide. On
plained by measurement errors in HPV detection. The review
these grounds, HPV fulfils the criteria of a necessary cause of
concluded that bias, chance, or confounding in the epidemio-
cervical cancer.
logical studies could be safely ruled out and that the
The second qualifier of the evaluation in Rothmans scheme
association could be considered causal in nature.252 In 1995, a
refers to the sufficient criteria. From natural history studies, it
formal IARC monograph was convened to review available
is clear that a large fraction of HPV infections occur as a self evidence for HPV and cervical cancer and concluded that
limiting subclinical process with little impact on morbidity. HPV-16 and HPV-18 were carcinogenic to humans and the two
The exceptions are the fraction of infected women in whom types were classed as group 1 carcinogens.12
HPV persistence is established, and in whom the risk of cervi- The conclusion of the 1995 monograph was subsequently
cal neoplasia increases significantly. Because of the large pool endorsed by the scientific community at large, which uses the
of infections that will not progress to cancer, it is usually IARCs reference widely as the standard for causality. Other
accepted that HPV is not a sufficient cause of disease (it is relevant bodies have undertaken similar exercises and have
noteworthy that no accepted cause of cancer has ever been consistently reached similar conclusions. These include the
considered as sufficient). The implication has been that other Centers for Disease Control, the National Cancer Institute
factors that modulate the evolution of the relevant HPV infec- (National Institutes of Health), the American Cancer Society
tion are intervening. However, specific evidence on the nature (Cervical Cancer Resource Center), the American College of
of these factors is elusive. Obstetricians and Gynecologists, the US Department of
From the data in tables 6 and 7 it is clear that in a fair Health and Human Services, the Program for Appropriate
number of cases HPV is the only environmental factor identi- Technology in Health, the World Health Organisation, the
fied of the four evaluated. This accounts for 23% (341 of 1490) European Research Organisation on Genital Infection and
of the cases in the multicentre study and 23% (186 of 806) of Neoplasia, and the Health Technology Assessment of the UK,
the cases in the IBSCC study. It is thus plausible that in among others.
perhaps one quarter of cases, the additional factors that In addition to the formal reviews by institutional groups,
modulate the host response to HPV are not the environmental individual scientists and peer reviewed journals have repeat-
factors usually linked to cervical cancer. It is probable that edly expressed their understanding on the causal nature of the
some of the host response mechanisms to the virus are the association between HPV and cervical
determinants of the prognosis of a given infection to either cancer.4 17 37 119 133 141 183 216 217 225 253273
spontaneous resolution or to viral persistence and neoplastic
progression. From evidence to decisions
The identification and natural history of biomarkers related Decision making in public health is a process that uses scien-
to the immune response and host susceptibility to HPV is a tific evidence in addition to judgment and socioeconomic con-
subject under active research. Likewise, the criteria of causal- siderations. The turning point at which public health decisions
ity outlined in table 1 may require a revision to integrate crite- are taken and enforced as opposed to recommending
ria for host responses in addition to the canopy of additional research is often elusive. The role of scientists in
environmental risk factors. influencing such decisions is also debatable (for an extensive
www.jclinpath.com
review see the special edition of the American Journal of Epide- ACKNOWLEDGEMENTS
miology274). This work has been partially supported by research grant FIS 01/1237
Rothmans textbook describes a rather extreme dichotomy and a non-restricted collaborative research agreement between the
between scientists and public health policy makers: Institut Catal dOncologia (ICO) in Barcelona, Spain and Digene
Corporation in Gaithersburg, Maryland, USA.
Lanes (1985)275 has proposed that causal inference is not
Advanced versions of this text have been reviewed by a number of
part of science at all, but lies strictly in the domain of public researchers in the field of the viral aetiology of cervical cancer. Their
policy. According to this view, since all scientific theories could comments and contributions are deeply appreciated: LA Brinton,
be wrong, policy makers should weigh the consequences of National Cancer Institute, Bethesda, USA; TR Broker, University of
actions under various theories. Scientists should inform policy Alabama, Birmingham, USA; RD Burk, Albert Einstein College of
makers about scientific theories, and leave the choice of a Medicine, New York, USA; X Castellsagu, Catalan Institute of Oncol-
theory and an action to policy makers. Not many public health ogy, Barcelona, Spain; J Cuzick, Imperial Cancer Research Foundation,
scientists are inclined toward such a strict separation between London, UK; S de Sanjos, Catalan Institute of Oncology, Barcelona,
Spain; M Del Vecchio, Digene Corporation, Gaithesburg, USA; ELF
science and policy, but as a working philosophy, it has the Franco, McGill University, Montreal, Canada; S Franceschi, Inter-
advantage of not putting scientists in the awkward position of national Agency for Research on Cancer, Lyon, France, I Frazer, Center
being advocates for a particular theory.276 Indeed, history for Immunology and Cancer Research, Brisbane, Australia; R Herrero,
shows that skepticism is preferable in science.19 Proyecto Epidemiolgico Guanacaste, San Jose, Costa Rica; S Krger
The field of cervical cancer prevention is now in an active Kjaer, Danish Cancer Society, Copenhagen, Denmark; CJN Lacey,
phase of evaluation of novel technologies, to increase the effi- Imperial College School of Medicine, London, UK; DR Lowy, National
ciency of screening in developed populations and to offer real- Institutes of Health, Bethesda, USA; V Moreno, Catalan Institute of
Oncology, Barcelona, Spain; A Muoz, John Hopkins School of
istic options to populations that do not benefit from cytology
Hygiene and Public Health, Baltimore, USA; M Plummer, Inter-
based organised programmes. The recognition that HPV is the national Agency for Reasearch on Cancer, Lyon, France; WC Reeves,
causal agent of cervical cancer worldwide is a landmark Centers for Disease Control and Prevention, Atlanta, USA; M
achievement that sets the grounds for qualitative progress. Schiffman, National Cancer Institute, Bethesda, USA; JT Schiller,
Current efforts in cervical cancer prevention in relation to HPV National Cancer Institute, Bethesda, USA; H-R Shin, National Cancer
are focusing first on the evaluation of protocols for HPV test- Centre Research Institute, Kyonggi, Korea; J Smith, International
ing in the context of established screening programmes, Agency for Research on Cancer, Lyon, France; PJF Snijders, Vrije Uni-
secondly on developing HPV vaccines for the prevention and versiteit Medical Center, Amsterdam, The Netherlands; BM Steinberg,
Long Island Jewish Hillside Medical Center, New York, USA; MH
treatment of HPV infections, and thirdly on the development Stoler, University of Virginia, Charlottesville, USA; D Thomas, Fred
of HPV treatments. Hutchinson Cancer Research Center, Seattle, USA; H Vainio,
International Agency for Research on Cancer, Lyon, France; AJ van
den Brule, Free University Hospital, Amsterdam, The Netherlands; LL
CONCLUSION Villa, Ludwig Institute for Cancer Research, Sao Paulo, Brazil; TC
At the current level of knowledge, the causal role of persistent Wright, Columbia University, New York, USA. Many other researchers
HPV infections in the development of cervical cancer and its have contributed to understanding cervical cancer aetiology and to
this review with their work, presentations at meetings, and coffee
precursors has been proved beyond reasonable doubt. A large break discussions. We are grateful to them and hope this review is of
number of human studies have been conducted following use to anyone involved in the task of cervical cancer prevention. Many
standard criteria that reasonably excluded bias, chance, and thanks to M Diaz and E Guin for statistical support and to C Rajo and
confounding in the estimation of the risk. The evidence avail- M Gonzalez, for secretarial help.
able at this time is consistent with the established criteria of
causality. The association of HPV with cervical cancer has been .....................
proposed as the first ever described necessary cause of a Authors affiliations
human cancer. F X Bosch, Institut Catal dOncologia, Servei dEpidemiologia i Registre
At the turning point following a major discovery in the ori- del Cncer, Gran Via Km 2.7 s/n 08907 LHospitalet de Llobregat,
gin of cervical cancer, it is the responsibility of the public 08907 Barcelona, Spain
A Lorincz, Digene Corporation, 1201 Clopper Road, Gaithersburg, MD
health administrators to judge when the scientific evidence is 20878, USA
sufficient to take action and promote changes in medical N Muoz, International Agency for Research on Cancer (IARC), Unit of
practice. Field and Intervention Studies, 150 Cours Albert-Thomas, F-69372 Lyon,
Selected quotes from Sir Bradford Hill, subsequently Cedex 08, France
C J L M Meijer, Vrije Universiteit Medical Center, Department of
endorsed by MacMahon and Pugh,11 Weed ,10 and Rothman,19 Pathology, De Boelelaan 1117, PO Box 70571007, 1081 HV
may illustrate the current situation in relation to the adoption Amsterdam, The Netherlands
of HPV testing as a relevant strategy for the prevention and K V Shah, The John Hopkins School of Public Health, Department of
management of cervical neoplasia. Molecular Microbiology and Immunology, 615 North Wolfe Street,
Baltimore, MD 21205, USA
Finally, like Hill before them, they (MacMahon and Pugh)
believed that proof of cause was elusive, and in the face of This paper is dedicated to the memory of the late Jan M M Walboomers.
controversy, the crucial decision was one involving the point at
which prudence should reign and action should be taken, REFERENCES
rather than waiting for more evidence. 1 Peto R, zur Hausen H, eds. Viral origin of cervical cancer. Bambury
The failure of some researchers to recognise the theoretical report no. 21, New York: Cold Spring Harbor CSH Press, 1986.
2 Franco EL. The sexually transmitted disease model for cervical cancer:
impossibility of proving the causal nature of an association incoherent epidemiologic findings and the role of misclassification of
has led to fruitless debates pitting skeptics who await such human papillomavirus infection. Epidemiology 1991;2:98106.
proof against scientists who are persuaded to make an 3 Bosch FX, Muoz N, de Sanjos S, et al. Risk factors for cervical cancer
in Colombia and Spain. Int J Cancer 1992;52:75058.
inference on the basis of existing evidence. 4 Schiffman MH, Bauer HM, Hoover RN, et al. Epidemiologic evidence
The responsibility of scientists for making causal judgments showing that human papillomavirus infection causes most cervical
was Hills final emphasis in his discussion of causation: intraepithelial neoplasia. J Natl Cancer Inst 1993;85:95864.
5 Thomas DB, Ray RM, Koetsawang A, et al. Human papillomaviruses
All scientific work is incompletewhether it be observa- and cervical cancer in Bangkok. I. Risk factors for invasive cervical
tional or experimental. All scientific work is liable to be upset carcinomas with human papillomavirus types 16 and 18 DNA. Am J
or modified by advancing knowledge. That does not confer Epidemiol 2001;153:72331.
upon us a freedom to ignore the knowledge we already have, 6 Lilienfeld AM. On the methodology of investigations of etiologic factors
in chronic diseasesome comments. J Chronic Dis 1959;10:416.
or to postpone the action that it appears to demand at a given 7 Sartwell PE. On the methodology of investigations of etiologic factors in
time. chronic diseasefurther comments. J Chronic Dis 1960;11:613.
www.jclinpath.com
8 Hill AB. The environment and disease: association or causation? Proc R 41 de Sanjose S, Valls I, Caadas MP, et al. Infeccin por los virus del
Soc Med 1965;58:295300. papiloma humano y de la inmunodeficiencia humana como factores de
9 Surgeon Generals Advisory Committee on Smoking and Health. riesgo para el cncer de cuello uterino en mujeres reclusas. Med Clin
DHEW publication no. (PHS) 1103. MD, USA: US Public Health Service, (Barc) 2000;115:814.
1964. 42 Palefsky JM, Minkoff H, Kalish LA, et al. Cervicovaginal human
10 Susser M. Causal thinking in the health sciences. New York: Oxford papillomavirus infection in human immunodeficiency virus-1 (HIV)-positive
University Press, 1973. and high-risk HIV-negative women. J Natl Cancer Inst 1999;91:22636.
11 MacMahon B, Pugh TF. Concepts of causes. In: Epidemiology: principles 43 Beasley RP, Lin CH, Hwang L-Y, et al. Hepatocellular carcinoma and
and methods. Boston: Little, Brown, 1970:1729. hepatitis B virus. A prospective study of 22 707 men in Taiwan. Lancet
12 IARC. Human papillomaviruses. IARC monographs on the evaluation of 1981;21:112931.
carcinogenic risks to humans, Vol. 64. Lyon: IARC, 1995. 44 Kuper HE, Tzonou A, Kaklamari E, et al. Hepatitis B and viruses in the
13 Weed DL. Causal and preventive inference. In: Greenwald P, Kramer etiology of hepatocellular carcinoma; a study in Greece using
BS, Weed DL, eds. Cancer prevention and control. New York: M Dekker, third-generation assays: Cancer Causes Control 2000;11:1715.
1995:285302. 45 Donato F, Tagger A, Chiesa R, et al. Hepatitis B and C virus infection,
14 Evans AS. Causation and disease: the Henle-Koch postulates revisited. alcohol drinking, and hepatocellular carcinoma: a case-control in study
Yale J Biol Med 1976;49:17595. Italy. Hepatology 1997;26:57984.
15 Rothman KJ. Causes. Am J Epidemiol 1976;104:58792. 46 Peto R, Darby S, Deo H, et al. Smoking, smoking cessation, and lung
16 Shah KV, Howley PM. Papillomaviruses. In: Knipe DM, Howley PM, eds. cancer in the UK since 1950: combination of national statistics with two
Fields virology. Philadelphia: Lippincott-Raven, 1996:2077109. case-control studies. BMJ 2000;321:3239.
17 zur Hausen H. Papillomavirus causing cancer: evasion from host-cell 47 Lee M-S, Kim D-H, Kim H, et al. Hepatitis B vaccination and reduced risk
control in early events in carcinogenesis. J Natl Cancer Inst of primary liver cancer among male adults: a cohort study in Korea. Int J
2000;92:69098. Epidemiol 1998;27:31619.
18 Rothman KJ. Reviews and commentary. Causes. Am J Epidemiol 48 Chang M-H, Chen CH-J, Lai M-S, et al. Universal hepatitis B vaccination
1995;104:905. in Taiwan and the incidence of hepatocellular carcinoma in children. N
19 Rothman KJ. Causal inference in epidemiology. In: Rothman KJ. Modern Engl J Med 1997;336:1855907.
epidemiology. Boston, Toronto: Little, Brown and Company, 1986:723. 49 Moreno V, Muoz N, Bosch FX, et al. Risk factors for progression of
20 Gravitt PE, Manos MM. Polymerase chain reaction-based methods for cervical intraepithelial neoplasm grade III to invasive cervical cancer.
the detection of human papillomavirus DNA, Vol. 119. Lyon: IARC Cancer Epidemiol Biomarkers Prev 1995;4:45967.
Scientific Publications, 1992. 50 Asato T, Nakajima Y, Nagamine M, et al. Correlation between the
21 van den Brule AJC, Meijer CJLM, Bakels V, et al. Rapid human progression of cervical dysplasia and the prevalence of human
papillomavirus detection in cervical scrapes by combined general primer papillomavirus. J Infect Dis 1994;169:9401.
mediated and type-specific polymerase chain reaction. J Clin Microbiol 51 Shen C-Y, Ho M-S, Chang S-F, et al. High rate of concurrent genital
1991;28:273943 infections with human cytomegalovirus and human papillomaviruses in
22 de Roda Husman AM, Walboomers JMM, van den Brule AJC, et al. cervical cancer patients. J Infect Dis 1993;168:44952.
The use of general primers GP5 and GP6 elongated at their 3 ends with 52 Arends MJ, Donaldson YK, Duval E, et al. Human papillomavirus type
adjacent highly conserved sequences improves human papillomavirus 18 associates with more advanced cervical neoplasia than human
detection by polymerase chain reaction. J Gen Virol 1995;76:105762. papillomavirus type 16. Hum Pathol 1993;24:4327.
23 Bosch FX, Rohan T, Schneider A, et al. Papillomavirus research update: 53 Das BC, Sharma JK, Gopalkrishna V, et al. A high frequency of human
highlights of the Barcelona HPV 2000 international papillomavirus papillomavirus DNA sequences in cervical carcinomas of Indian women
conference. J Clin Pathol 2001;54:16375. as revealed by Southern blot hybridization and polymerase chain
24 Muoz N, Bosch FX, de Sanjos SM, et al. The causal link between reaction. J Med Virol 1992;36:23945.
human papillomavirus and invasive cervical cancer: a population-based 54 Ter Meulen J, Eberhardt HC, Luande J, et al. Human papillomavirus
case-control study in Colombia and Spain. Int J Cancer 1992;52:7439.
(HPV) infection, HIV infection and cervical cancer in Tanzania, East
25 Eluf-Neto J, Booth M, Muoz N, et al. Human papillomavirus and
Africa. Int J Cancer 1992;51:51521.
invasive cervical cancer in Brazil. Br J Cancer 1994;69:11419.
55 Muoz N, Bosch FX. HPV and cervical neoplasia: review of case-control
26 Chaouki N, Bosch FX, Muoz N, et al. The viral origin of cervical
and cohort studies. In: Muoz N, Bosch FX, Shah KV, et al, eds. The
cancer in Rabat, Morocco. Int J Cancer 1998;75:54654.
epidemiology of cervical cancer and human papillomavirus. Lyon: IARC
27 Roln PA, Smith JS, Muoz N, et al. Human papillomavirus infection
Scientific Publications No. 119, 1992:2516.
and invasive cervical cancer in Paraguay. Int J Cancer
56 Peng HQ, Liu Sl, Mann V, et al. Human papillomavirus types 16 and
2000;85:48691.
33, herpes simplex virus type 2 and other risk factors for cervical cancer
28 Ngelangel C, Muoz N, Bosch FX, et al. Causes of cervical cancer in
the Philippines: a case-control study. J Natl Cancer Inst 1998;90:439. in Sichuan Province, China. Int J Cancer 1991;47:71116.
29 Chichareon S, Herrero R, Muoz N, et al. Risk factors for cervical 57 Griffin NR, Bevan IS, Lewis FA, et al. Demonstration of multiple HPV
cancer in Thailand: a case control study. J Natl Cancer Inst types in normal cervix and in cervical squamous cell carcinoma using the
1998;90:507. polymerase chain reaction on paraffin wax embedded material. J Clin
30 Santos C, Munoz N, Klug S, et al. HPV types and cofactors causing Pathol 1990;43:526.
cervical cancer in Peru. Br J Cancer 2001;85:96671. 58 Marin J, Ursic Vrscaj M, Erzen M. Detection of human papillomaviruses
31 Woodman CB, Collins S, Winter H, et al. Natural history of cervical (HPV-16,18) in cervical smears by in situ hybridization. Israel J Med Sci
human papillomavirus infection in young women: a longitudinal cohort 1994;30:44850.
study. Lancet 2001;357:18316. 59 Lrincz AT, Reid R, Jenson AB, et al. Human papillomavirus infection of
32 Muoz N, Bosch FX, Chichareon S, et al. A multinational case-control the cervix: relative risk associations of 15 common anogenital types.
study on the risk of cervical cancer linked to 25 HPV types: which are the Obstet Gynecol 1992;79:32837.
high-risk types? In: Castellsagu X, Bosch FX, de Sanjos S, et al, eds. 60 Kanetsky PA, Mandelblatt J, Richart R, et al. Risk factors for cervical
18th International Papillomavirus Conferenceprogram and abstracts cancer in a black, elderly population: preliminary findings. Ethn Dis
book. Barcelona: Thau SL, 2000:125 (available on line: 1992;2:33742.
http:/www.hpv2000.com). 61 Kadish AS, Hagan RJ, Ritter DB, et al. Biologic characteristics of specific
33 Bosch FX, Muoz N, Chichareon S, et al. HPV and cervical human papillomavirus types predicted from morphology of cervical
adenocarcinoma: an IARC based multicentric case-control study. In: lesions. Hum Pathol 1992;23:12629.
Castellsagu X, Bosch FX, de Sanjos S, et al, eds. 18th International 62 Mandelblatt J, Richart R, Thomas L, et al. Is human papillomavirus
Papillomavirus Conferenceprogram and abstracts book. Barcelona: associated with cervical neoplasia in elderly? Gynecol Oncol
Thau SL, 2000:131 (available on line: http:/www.hpv2000.com). 1992;46:612.
34 Bosch FX, Manos MM, Muoz N, et al. Prevalence of human 63 Saito J, Fukuda T, Nakatani H, et al. Detection of human papillomavirus
papillomavirus cervical cancer: a worldwide perspective. J Natl Cancer DNA in the normal cervices of Japanese women by the dot-blot (Vira
Inst 1995;87:796802. PapTM) method. Asia-Oceania J Obstet Gynaecol 1992;18:2837.
35 Castellsagu X, Menndez C, Loscertales M-P, et al. Human 64 Ohta M, Casanova H, Mizuno K, et al. Epidemiologic background and
papillomavirus genotypes in rural Mozambique. Lancet changes in patients infected with human papilloma virus. Nippon Sanka
2001;358:142930. Fujinka Gakkai Zasshi 1991;43:47984.
36 Herrero R, Muoz N, Lazcano E, et al. HPV International prevalence 65 Wei Z, Yazhou S, Shunqian J, et al. The association between cervical
surveys in general populations. In: Castellsagu X, Bosch FX, de Sanjos carcinoma and human papilloma virus (HPV) in Xiangyuan county. Chin
S, et al, eds. 18th International Papillomavirus Conferenceprogram Med Sci J 1991;6:747.
and abstracts book. Barcelona: Thau SL, 2000:126 (available on line: 66 Si JY, Lee K, Han R, et al. A research for the relationship between human
http:/www.hpv2000.com). papillomavirus and human uterine cervical carcinoma. I. The
37 Herrero R, Hildesheim A, Bratti C, et al. Population-based study of identification of viral genome and subgenomic consequences in biopsies
human papillomavirus infection and cervical neoplasia in rural Costa of Chinese patients. J Cancer Res Clin Oncol 1991;117:4549.
Rica. J Natl Cancer Inst 2000;92:46474. 67 Chang F, Syrjnen S, Shen Q, et al. Detection of human papillomavirus
38 Thomas DB, Quin Q, Kuypers J, et al. Human papillomaviruses and (HPV) in genital warts and carcinomas by DNA in situ hybridization in
cervical cancer in Bangkok. II. Risk factors for in situ and invasive Chinese patients. Cytopathology 1990;1:97103.
squamous cell cervical carcinomas. Am J Epidemiol 2001;153:7328. 68 Zhang GA. A study on the relationship between human papilloma virus
39 Olsen AO, Gjoen K, Sauer T, et al. Human papillomavirus in cervical (HPV) and cervical cancer. Chinese Journal of Epidemiology
intraepithelial neoplasia grade II-III: a population-based case-control 1990;11:313.
study. Int J Cancer 1995;61:31215. 69 Yokota H, Yoshikawa H, Shiromizu K, et al. Detection of human
40 Kjaer SK, Van den Brule AJC, Boch JE, et al. Human papillomavirusthe papillomavirus types 6/11, 16 and 18 in exfoliated cells from the uterine
most significant risk determinant of cervical intraepithelial neoplasia. Int J cervices of japanese women with and without lesions. Jpn J Cancer Res
Cancer 1996;65:6016. 1990;81:896901.
www.jclinpath.com
70 Colgan TJ, Percy ME, Suri M, et al. Human papillomavirus infection of 100 Syrjnen K, Syrjnen S. HPV infections in other sites and lesions. In:
morphologically normal cervical epithelium adjacent to squamous Infections in human pathology. UK: Wiley 1999:44556.
dysplasia and invasive carcinoma. Hum Pathol 1989;20:31619. 101 Lazcano-Ponce E, Herrero R, Muoz N, et al. Epidemiology of HPV
71 Czegldy J, Gergely L, Herndi Z, et al. Detection of human infection among Mexican women with normal cervical cytology. Int J
papillomavirus deoxyribonucleic acid in the female genital tract. Med Cancer 2001;91:19.
Microbiol Immunol (Berl) 1989;178:30914. 102 Melkert PW, Hopman E, van den Brule AJ, et al. Prevalence of HPV in
72 Reeves WC, Brinton LA, Garcia M, et al. Human papillomavirus cytomorphologically normal cervical smears, as determined by the
infection and cervical cancer in Latin America. N Engl J Med. polymerase chain reaction, is age-dependent. Int J Cancer
1989;320:143741. 1993;53:91923.
73 Donnan SPB, Wong FWS, Ho SC, et al. Reproductive and sexual risk 103 Bauer HM, Hildesheim A, Schiffman MH, et al. Determinants of genital
factors and human papilloma virus infection in cervical cancer among human papillomavirus infection in low-risk women in Portland, Oregon.
Hong Kong. Chinese International Journal of Epidemiology Sex Transm Dis 1993;20:2748.
1989;18:326. 104 Bosch FX, Muoz N, de Sanjos S, et al. Importance of human
74 Zhang WH, Coppleson M, Rose BR, et al. Papillomavirus and cervical papillomavirus endemicity in the incidence of cervical cancer: an
cancer; a clinical and laboratory study. J Med Virol 1988;26:16374. extension of the hypothesis on sexual behavior. Cancer Epidemiol
75 Wilczynski SP, Walker J, Liao S-Y, et al. Adenocarcinoma of the cervix Biomarkers Prev 1994;3:3759.
associated with human papillomavirus. Cancer 1988;62:13316. 105 Kjaer SK, Van der Brule AJC, Boch JE, et al. Determinants for genital
76 Fuchs PG, Girardi F. Pfister human papillomavirus DNA in normal, human papillomavirus (HPV) infection in 1000 randomly chosen young
metaplastic, preneoplastic and epithelia of the cervix uteri. Int J Cancer Danish women with normal Pap smear: are there different risk profiles for
1988;41:415. oncogenic and nononcogenic HPV types? Cancer Epidemiol Biomarkers
77 Hsieh CY, Lee S-C, Huang S-C. Presence and expression of human Prev 1997;6:799805.
papillomavirus types 16 and 18 DNA sequences in cervical carcinomas. 106 Jacobs MV, Walboomers JMM, Snijders PJF, et al. Distribution of 37
Asia-Oceania J Obstet Gynaecol 1988;14:8795. mucosotropic HPV types in women with cytologically normal cervical
78 Choo K-B, Shen H-D, Leung W-Y, et al. A distinct difference in the smears: the age-related patterns for high-risk and low-risk types. Int J
prevalence of papillomavirus infection in cytologically normal and Cancer 2000;87:2217.
neoplastic cells of the uterine cervix. Chinese Med J (Taipei) 107 Schneider A, Hoyer H, Lotz B, et al. Screening for high-grade cervical
1988;42:16. intra-epithelial neoplasia and cancer by testing for high-risk HPV, routine
79 Kulski JK, Howard MJ, Pixley. ECDNA sequences of human cytology or colposcopy. Int J Cancer 2000;89:52934.
papillomavirus types 11, 16 or 18 in invasive cervical carcinoma of 108 Parkin DM, Whelan SL, Ferlay J, et al. Cancer incidence in five
Western Australian women. Immunol Cell Biol 1987;65:7784. continents. Lyon: IARC Scientific Publications, Vol. VII, No 143, 1997.
80 Zhang WH, Wu AR, Li JX, et al. Relation between human papillomavirus 109 Franco EL, Villa LL, Sobrinho JP, et al. Epidemiology of acquisition and
(HPV) and cancer of uterine cervix. Chinese Journal of Oncology clearance of cervical human papillomavirus infection in women from a
1987;9:4335. high-risk area for cervical cancer. J Infect Dis 1999;180:141523.
81 Lrincz AT, Temple GF, Kurman RJ, et al. Oncogenic association of 110 Ho GYF, Biermal R, Beardsley L, et al. Natural history of cervicovaginal
specific human papillomavirus types with cervical neoplasia. J Natl papillomavirus infection in young women. N Engl J Med
Cancer Inst 1987;79:6717. 1998;338:42328.
82 Schneider A , de Villiers E-M, Schneider V. Multifocal squamous 111 Koutsky LA, Holmes KK, Critchlow CW, et al. A cohort study of the risk
neoplasia of the female genital tract: significance of human of cervical intraepithelial neoplasia grade 2 or 3 in relation to
papillomavirus infection of the vagina after hysterectomy. Obstet Gynecol papillomavirus infection. N Engl J Med 1992;327:12728.
1987;70:2948. 112 Ho GYF, Burk RD, Klein S, et al. Persistent genital human papillomavirus
83 Meanwell CA, Cox MF, Blackledge G, et al. HPV 16 DNA in normal infection as a risk factor for persistent cervical dysplasia. J Natl Cancer
and malignant cervical epithelium: implications for the aetiology and Inst 1995;87:136571.
behaviour of cervical neoplasia. Lancet 1987;1:7037. 113 Remmink AJ, Walboomers JM, Helmerhorst TJ, et al. The presence of
84 McCance DJ, Campion MJ, Clarkson PK, et al. Prevalence of human persistent high-risk HPV genotypes in dysplastic cervical lesions is
papillomavirus type 16 DNA sequences in cervical intraepithelial associated with progressive disease: natural history up to 36 months. Int J
neoplasia and invasive carcinoma of the cervix. Br J Obstet Gynaecol Cancer 1995;61:30611.
1985;92:11015. 114 Nobbenhuis MAE, Walboomers JMM, Helmerhorst TJM, et al. Relation
85 Josefsson AM, Agnetha M, Magnusson PKE, et al. Viral load of human of human papillomavirus status to cervical lesions and consequences for
papilloma virus 16 as a determinant for development of cervical cervical-cancer screening: a prospective study. Lancet 1999;354:205.
carcinoma in situ: a nested case-control study. Lancet 2000;355:2189 115 Rozendaal L, Walboomers JM, van der Linden JC, et al. PCR-based
93. high-risk HPV test in cervical cancer screening gives objective risk
86 Ylitalo N, Josefsson A, Melbye M, et al. A prospective study showing assessment of women with cytomorphologically normal cervical smears.
long-term infection with human papillomavirus 16 before the development Int J Cancer 1996;68:7669.
of cervical carcinoma in situ. Cancer Res 2000;60:602732. 116 Rozendaal L, Westerga J, van der Linden JC, et al. PCR based high risk
87 Lacey JV, Frisch M, Brinton LA, et al. Association between smoking and HPV testing is superior to neural network based screening for predicting
adenocarcinomas and squamous cell carcinomas of the uterine cervix incident CIN III in women with normal cytology and borderline changes.
(United States). Cancer Causes Control 2001;12:15361. J Clin Pathol 2000;53:60611.
88 van Beurden M, ten Kate FWJ, Smits HL, et al. Multifocal vulvar 117 Zielinski GD, Snijders PJF, Rozendaal L, et al. HPV presence precedes
intraepithelial neoplasia grade III and multicentric lower genital tract abnormal cytology in women developing cervical cancer and signals
neoplasia is associated with transcriptionally active human false negative smears. Br J Cancer 2001;85:398404.
papillomavirus. Cancer 1995;75:287984. 118 Nobbenhuis MAE, Helmerhorst THJM, van den Brule AJC, et al.
89 van Beurden M, ten Kate FWJ, Tjong-A-Hunt SP, et al. Human Cytological regression and clearance of high-risk human papillomavirus
papillomavirus DNA in multicentric vulvar intraepithelial neoplasia. Int J in women with abnormal cervical smear. Lancet 2001;358:17823.
Gynecol Pathol 1998;17:1216. 119 Manos MM, Kinney WK, Hurley LB, et al. Identifying women with
90 Pinto AP, Signorello LB, Crum CP, et al. Squamous cell carcinoma of the cervical neoplasia: using human papillomavirus DNA testing for
vulva in brazil: prognostic importance of host and viral variables. equivocal Papanicolaou results. JAMA 1999;281:160510.
Gynecol Oncol 1999;14:617. 120 Schlecht NF, Kulaga S, Robitaille J, et al. Persistent human
91 Brandsman JL, Abramson AL. Association of papillomavirus with papillomavirus infection as a predictor of cervical intraepithelial
cancers of the head and neck. Arch Otolaryngol Head Neck Surg neoplasia. JAMA 2001;286:310614.
1989;115:6215. 121 Zielinski GD, Snijders PJF, Rozendaal L, et al. High-risk HPV testing in
92 Snijders PJF, Cromme FV, van der Brule AJC, et al. Prevalence and women with borderline and mild dyskaryosis: long-term follow-up data
expression of human papillomavirus in tonsillar carcinomas, indicating a and clinical relevance. J Clin Pathol 2001;193:18.
possible viral etiology. Int J Cancer 1992;51:84550. 122 Nobbenhuis MA, Meijer CJ, van den Brule AJ, et al. Addition of
93 Cruz IBF, Snijders PJF, Steenbergen RDM, et al. Age-dependence of high-risk HPV testing improves the current guidelines on follow-up after
human papillomavirus DNA presence in oral squamous cell carcinomas. treatment for cervical intraepithelial neoplasia. Br J Cancer
Eur J Cancer Oral Oncol 1996;32B:5562. 2001;84:796801.
94 Frisch M, Biggar RJ. Aetiological parallel between tonsillar and 123 Wallin K-L, Wiklund F, Angstrm T, et al. Type-specific persistence of
anogenital squamous-cell carcinomas: Lancet 1999;354:14425. human papillomavirus DNA before the development of invasive cervical
95 Harwood CA, Surentheran T, McGregor JM, et al. Human cancer. N Engl J Med 1999;341:16338.
papillomavirus infection and non-melanoma skin cancer in 124 Dillner J, Lehtinen M, Bjorge T, et al. Prospective seroepidemiologic
immunosuppressed and immunocompetent individuals. J Med Virol study of human papillomavirus infection as a risk factor for invasive
2000;61:28997. cervical cancer. J Natl Cancer Inst 1997;89:12939.
96 Newton RA. review of the aetiology of squamous cell carcinoma of the 125 Campion MJ, McCance DJ, Cuzick J, et al. Progressive potential of mild
conjunctiva. Br J Cancer 1996;74:151113. cervical atypia; prospective cytological, colposcopic, and virological
97 Xi LF, Koutsky LA, Galloway DA, et al. Genomic variation of human study. Lancet 1986;2:23740.
papillomavirus type 16 and risk for high grade cervical intraepithelial 126 Cuzick J, Terry G, Ho L, et al. Human papillomavirus type 16 DNA in
neoplasia. J Natl Cancer Inst 1997;89:796802. cervical smears as predictor of high-grade cervical cancer. Lancet
98 Villa LL, Sichero L, Rahal P, et al. Molecular variants of human 1992;339:95960.
papillomavirus types 16 and 18 preferentially associated with cervical 127 Cuzick J, Terry G, Ho L, et al. Type-specific human papillomavirus DNA
neoplasia. J Gen Virol 2000;81:295968. in abnormal smears as a predictor of high-grade cervical intraepithelial
99 Hildesheim A, Schiffman M, Bromley C, et al. Human papillomavirus neoplasia. Br J Cancer 1994;69:16771.
type 16 variants and risk of cervical cancer. J Natl Cancer Inst 128 Cuzick J, Szarewski A, Terry G, et al. Human papillomavirus testing in
2001;93:31518. primary cervical screening. Lancet 1995;345:15336.
www.jclinpath.com
129 Cuzick J, Sasieni P, Davies P, et al. A systematic review of the role of 159 Tommasino M. Papillomaviruses in human cancer: the role of E6 and
human papillomavirus (HPV) testing within a cervical screening E7 oncoproteins. Austin, Texas: Landes Bioscience, 1997.
programme: summary and conclusions. Br J Cancer 2000;83:5615. 160 Lazo PA. The molecular genetics of cervical carcinoma. Br J Cancer
130 Ratnam S, Franco EL, Ferenczy A. Human papillomavirus testing for 1999;80:200818.
primary screening of cervical cancer precursors. Cancer Epidemiol 161 McGlennen RC. Human papillomavirus oncogenesis. Clin Lab Med
Biomarkers Prev 2000;9:94551. 2000;20:383404.
131 Solomon D, Schiffman M, Tarone R. Comparison of three management 162 Woodworth CD, Doniger J, DiPaolo JA. Immortalization of human
strategies for patients with atypical squamous cells of undetermined foreskin keratinocytes by various human papillomavirus DNAs
significance: baseline results from a randomized trial. J Natl Cancer Inst corresponds to their association with cervical carcinoma. J Virol
2001;93:2939. 1989;63:15964.
132 Clavel C, Masure M, Bory JP, et al. Hybrid capture II-based human 163 Meyers C, Laimins LA. In vitro systems for the study and propagation of
papillomavirus detection, a sensitive test to detect in routine high-grade human papillomaviruses. Curr Top Microbiol Immunol
cervical lesions: a preliminary study on 1518 women. Br J Cancer 1994;186:199215.
1999;80:130611. 164 Steenbergen RDM, Walboomers JMM, Meijer CJLM, et al. Transition of
133 Schiffman MH, Herrero R, Hildesheim A, et al. HPV DNA testing in human papillomavirus type 16 and 18 transfected human foreskin
cervical cancer screening. JAMA 2000;283:8793. keratinocytes towards immortality: activation of telomerase and allele
134 Wright TC, Jr, Denny L, Kuhn L, et al. HPV DNA testing of self-collected losses at 3p, 10p, 11q and/or 18q. Oncogene 1996;13:124957.
vaginal samples compared with cytologic screening to detect cervical 165 Stoler MH, Rhodes CR, Whitbeck A, et al. Human papillomavirus type
cancer. JAMA 2000;283:816. 16 and 18 gene expression in cervical neoplasias. Hum Pathol
135 Hildesheim A, Gravitt P, Schiffman MH, et al. Determinants of genital 1992;23:11728 .
human papillomavirus infection in low-risk women in Washington, DC. 166 Lancaster WD, Castellano C, Santos C, et al. Human papillomavirus
Sex Transm Dis 1993;20:27984. deoxyribonucleic acid in cervical carcinoma from primary and metastatic
136 Wheeler CM, Parmenter CA, Hunt WC, et al. Determinants of genital sites. Am J Obstet Gynecol 1986;154:11519.
human papillomavirus infection among cytologically normal women 167 Schwarz E, Freese UK, Gissmann L, et al. Structure and transcription of
attending the University of New Mexico student health center. Sex human papillomavirus sequences in cervical carcinoma cells. Nature
Transm Dis 1993;20:2869. 1985;314:11114.
137 Franco EL, Villa LL, Ruiz A, et al. Transmission of cervical human 168 Durst M, Glitz D, Schneider A, et al. Human papillomavirus type 16
papillomavirus infection by sexual activity: differences between low- and (HPV 16) gene expression and DNA replication in cervical neoplasia:
high-risk oncogenic types. J Infect Dis 1995;172:75663. analysis by in situ hybridization. Virology 1992;189:13240.
138 Muoz N, Kato I, Bosch FX, et al. Risk factors for HPV DNA detection in 169 Durst M, Kleinheinz A, Hotz M, et al. The physical state of human
middle-aged women. Sex Transm Dis 1996;23:50410. papillomavirus type 16 DNA in benign and malignant genital tumours. J
139 Sillins I, Kallings I, Dillner J. Correlates of the spread of human Gen Virol 1985;66:151522.
papillomavirus infection. Cancer Epidemiol Biomarkers Prev 170 Cullen AP, Reid R, Campion M, et al. Analysis of the physical state of
2000;9:9539. different human papillomavirus DNAs in intraepithelial and invasive
140 Rousseau MC, Franco EL, Villa LL, et al. A cumulative case-control study cervical neoplasm. J Virol 1991;65:60612.
of risk factor profiles for oncogenic and non-oncogenic cervical human 171 Stanbridge EJ. Suppression of malignancy in human cells. Nature
papillomavirus infections. Cancer Epidemiol Biomarkers Prev 1976;260:1720.
2000;9:46976. 172 Saxon PJ, Srivatsan ES, Stanbridge EJ. Introduction of human
141 Kjaer SK, Chackerian B, Van den Brule AJC, et al. High-risk human chromosome 11 via microcell transfer controls tumorigenic expression of
papillomavirus is sexually transmitted: evidence from a follow-up study of HeLa cells. EMBO J 1986;5:34616.
virgins starting sexual activity (intercourse). Cancer Epidemiol Biomarkers 173 Koi M, Morita H, Yamada H, et al. Normal human chromosome 11
Prev 2001;10:1016. suppresses tumorigenicity of human cervical tumor cell line SiHa. Mol
142 Barrasso R, de Brux J, Croissant O, et al. High prevalence of Carcinog 1989;2:1221.
papillomavirus-associated penile intraepithelial neoplasia in sexual 174 Backsch C, Wagenbach N, Nonn M, et al. Microcell-mediated transfer
partners of women with cervical intraepithelial neoplasia. N Engl J Med of chromosome 4 into HeLa cells suppresses telomerase activity. Genes
1987;317:91623. Chromosomes Cancer 2001;31:1968.
143 Kjaer SK, de Villiers EM, Dahl C, et al. Case-control study of risk factors 175 Steenbergen RDM, Kramer D, Meijer CJLM, et al. Telomerase
for cervical neoplasia in Denmark. I: Role of the male factor in women suppression by chromosome 6 in a human papillomavirus type
with one lifetime sexual partner. Int J Cancer 1991;48:3944. 16-immortalized keratinocyte cell line and in a cervical cancer cell line. J
144 Bergman A, Nalick R. Prevalence of human papillomavirus infection in Natl Cancer Inst 2001;93:86572.
men: comparison of the partners of infected and uninfected women. J 176 Dyson N, Howley PM, Munger K, et al. The human papilloma virus-16
Reprod Med 1992;37:71012. E7 oncoprotein is able to bind to the retinoblastoma gene product.
145 Bosch FX, Castellsagu X, Muoz N, et al. Male sexual behavior and Science 1989;243:9346.
human papillomavirus DNA: key risk factors for cervical cancer in Spain. 177 Lrincz AT. Human papillomaviruses. In: Lennette EH, Smith TF, eds.
J Natl Cancer Inst 1996;88:10607. Laboratory diagnosis of viral infections. New York: Marcel Dekker,
146 Castellsagu X, Ghaffari A, Daniel RW, et al. Prevalence of penile 1999:63563.
human papillomavirus DNA in husbands of women with and without 178 Werness BA, Levine AJ, Howley PM. Association of human
cervical neoplasia: a study in Spain and Colombia. J Infect Dis papillomavirus types 16 and 18 E6 proteins with p53. Science
1997;176:35361. 1990;248:769.
147 de Sanjos S, Palacio V, Tafur L, et al. Prostitution, HIV, and cervical 179 Cheng S, Schmidt-Grimminger D-C, Murant T, et al.
neoplasia: a survey in Spain and Colombia. Cancer Epidemiol Differentiation-dependent up-regulation of the human papillomavirus E7
Biomarkers Prev 1993;2:5315. gene reactivates cellular DNA replication in suprabasal differentiated
148 Thomas DB, Ray RM, Kuypers J, et al. Human papillomaviruses and keratinocytes. Genes Dev 1995;9:233549.
cervical cancer in Bangkok. III. The role of husbands and commercial sex 180 Jian Y, Schmidt-Grimminger D-C, Chien W-M, et al. Post-transcriptional
workers. Am J Epidemiol 2001;153:7408. induction of p21cip1 protein by human papillomavirus E7 inhibits
149 Juarez-Figueroa LA, Wheeler CM, Uribe-Salas FJ, et al. Human unscheduled DNA synthesis reactivated in differentiated keratinocytes.
papillomavirus: a highly prevalent sexually transmitted disease agent Oncogene 1998;17:202738.
among female sex workers from Mexico City. Sex Transm Dis 181 Schmidt-Grimminger D-C, Wu X, Jian Y, et al. Post-transcriptional
2001;28:12530. induction of p21cip1 protein in condylomata and dysplasias is inversely
150 Castellsagu X, Bosch FX, Muoz N, et al. Male circumcision, penile related to human papillomavirus activities. Am J Pathol
human papillomavirus (HPV) infection and cervical cancer. N Engl J Med 1998;152:101524.
[In press.] 182 Steenbergen RDM, Parker JN, Isern S, et al. Viral E6-E7 transcription in
151 Rigoni-Stern. Fatti statistici relativi alle malattie cancerose. Giorn Prog the basal layer of organotypic cultures without apparent p21cip1 protein
Patol Terap 1842;2:50717. precedes immortalization of HPV 16 and HPV 18 transfected human
152 Beral V. Cancer of the cervix: a sexually transmitted infection? Lancet keratinocytes. J Virol 1998;72:74957.
1974;i:103740. 183 Stoler MH. Human papillomaviruses and cervical neoplasia: a model for
153 Skegg DC, Corwin PA, Paul C, et al. Importance of the male factor in carcinogenesis. Int J Gynecol Pathol 2000;19:1628.
cancer of the cervix. Lancet 1982;2:5813. 184 Hildesheim A, Schiffman MH, Tsukui T, et al. Immune activation in
154 Schiffman MH. New epidemiology of human papillomavirus infection cervical neoplasia: cross-sectional association between plasma soluble
and cervical neoplasia. J Natl Cancer Inst 1995;87:20. interleukin 2 receptor levels and disease. Cancer Epidemiol Biomarkers
155 Ylitalo N, Sorensen P, Josefsson AM, et al. Consistent high viral load of Prev 1997;6:80713.
human papillomavirus 16 and risk of cervical carcinoma in situ: a nested 185 Tsukui T, Hildesheim A, Schiffman MH et al. Interleukin 2 production in
case-control study. Lancet 2000;355:21948. vitro by peripheral lymphocytes in response to human
156 van Duin M, Snijders PJF, Schrijnemakers HFJ, et al. Human papillomavirus-derived peptides: correlation with cervical pathology.
papillomavirus 16 load in normal and abnormal cervical scrapes: an Cancer Res 1996;56:369774.
indicator of CIN II/III and viral clearance. Int J Cancer [In press.] 186 Chow LT, Broker TR. Papillomavirus DNA replication. Intervirology
157 Cubie H, Seagar A, McGoogan E, et al. Rapid real time PCR to 1994;37:1508.
distinguish between high risk human papillomavirus types 16 and 18. 187 Dollard SC, Wilson JL, Demeter LM, et al. Production of human
Mol Pathol 2001;54:249. papillomavirus and modulation of the infectious program in epithelial raft
158 Stern P, Duggan-Keen M. MHC expression in the natural history of cultures. Genes Dev 1992;6:113142.
cervical cancer In: Stern PL, Stanley MA, eds. Human papillomaviruses 188 Flores ER, Lambert PF. Evidence for a switch in the mode of human
and cervical cancer. Biology and immunology. New York: Oxford papillomavirus type 16 DNA replication during the viral life cycle. J Virol
University Press, 1994:16276. 1997;71:716779.
www.jclinpath.com
189 Munger K, Scheffner M, Huibregtse JM, et al. Interactions of HPV E6 220 Moreno V, Bosch FX, Muoz N, et al. Oral contraceptives and cervical
and E7 oncoproteins with tumour suppressor gene products. In: Levine cancer: pooled analysis of the IARC multicentric case-control study.
AJ, ed. Tumour suppressor genes, the cell cycle and cancer. New York: Lancet [In press.]
Cold Spring Harbor Laboratory Press, 1992:197217. 221 Kjaer SK, van den Brule AJC, Svare EI, et al. Different risk factor patterns
190 Turek LP, Smith EM. The genetic program of genital human for high-grade and low-grade intraepithelial lesions on the cervix among
papillomaviruses in infection and cancer. Obstet Gynecol Clin North Am HPV-positive and HPV-negative young women. Int J Cancer
1996;23:73558. 1998;76:6139.
191 zur Hausen H. Roots and perspectives of contemporary papillomavirus 222 Lacey JV, Brinton LA, Abbas FM, et al. Oral contraceptives as risk
research. J Cancer Res Clin Oncol 1996;122:313. factors for cervical adenocarcinomas and squamous cell carcinomas.
192 Tommasino M, Adamczewski JP, Carlotti F, et al. HPV16 E7 protein Cancer Epidemiol Biomarkers Prev 1999;8:107985.
associates with the protein kinase p33CDK2 and cyclin A. Oncogene 223 IARC. Hormonal contraception and post-menopausal hormonal therapy.
1993;8:195202. IARC monographs on the evaluation of carcinogenic risks to humans,
193 Demeter LM, Stoler MH, Broker TR, et al. Induction of proliferating cell Vol. 72. Lyon: IARC, 1999.
nuclear antigen in differentiated keratinocytes of human 224 Muoz N, Franceschi S, Bosch FX, et al. The influence of reproductive
papillomavirus-infected lesions. Hum Pathol 1994;25:3438. and menstrual factors on cervical cancer in HPV-positive women. Lancet
194 Stoler MH, Wolinsky SM, Whitbeck A, et al. Differentiation-linked [In press.]
human papillomavirus types 6 and 11 transcription in genital 225 Hildesheim A, Herrero R, Castle PE, et al. HPV co-factors related to the
condylomata revealed by in situ hybridization with message-specific RNA development of cervical cancer: results from a population-based study in
probes. Virology 1989;172:33140. Costa Rica. Br J Cancer 2001;84:121926.
195 Arends MJ, Wyllie AH, Bird CC. Human papillomavirus type 18 is 226 Kjaer SK, de Villiers EM, Haugaard BJ, et al. Human papillomavirus,
associated with less apoptosis in fibroblast tumours than human herpes simplex virus and cervical cancer incidence in Greenland and
papillomavirus type 16. Br J Cancer 1995;72:6469. Denmark. A population-based cross-sectional study. Int J Cancer
196 Jones DL, Munger K. Analysis of the p53-mediated G1 growth arrest 1988;41:51824.
pathway in cells expressing the human papillomavirus type 16 E7 227 Ferrera A, Velema JP, Figueroa M, et al. Co-factors related to the causal
oncoprotein. J Virol 1997;71:290512. relationship between human papillomavirus and invasive cervical cancer
197 McCance DJ, Kopan R, Fuchs E, et al. Human papillomavirus type 16 in Honduras. Int J Epidemiol 2000;29:81725.
alters human epithelial cell differentiation in vitro. Proc Natl Acad Sci U S 228 Winkelstein W, Jr. Smoking and cancer of the uterine cervix:
A 1988;85:716973. hypothesis. Am J Epidemiol 1977;106:2579.
198 McDougall JK, Klingelhutz AJ. Telomerase and the HPV oncogenes: 229 IARC. Tobacco smoking. Lyon: IARC monographs on the evaluation of
more than a terminal relationship. Papillomavirus Report 1999;10:814. carcinogenic risks to humans, Vol. 38. Lyon: IARC, 1986.
199 Jeon S, Allen-Hoffmann BL, Lambert PF. Integration of human 230 Szarewski A, Cuzick J. Smoking and cervical neoplasia: a review of the
papillomavirus type 16 into the human genome correlates with a selective evidence. J Epidemiology and Biostatistics 1998;3:22956.
growth advantage of cells. J Virol 1995;69:298997. 231 Surgeon Generals Advisory Committee on Women and Smoking.
200 zur Hausen H. Papillomavirus infectionsa major cause of human Public Health Service, 2001. (Available online: http:/www.cdc.gov/
cancers. Biochim Biophys Acta 1996;1288:F5578. tobacco.)
201 Bauknecht T, See RH, Shi Y. A novel C/EBP beta-YY1 complex controls 232 Kjaer SK. Risk factors for cervical neoplasia in Denmark. APMIS
the cell-type-specific activity of the human papillomavirus type 18 1998;80:141.
upstream regulatory region. J Virol 1996;70:7695. 233 Beasley RP, Trepo C, Stevens CE, et al. The E antigen and vertical
202 Dong X-P, Stubenrauch F, Beyer-Finkler E, et al. Prevalence of deletions transmission of hepatitis B surface antigen. Am J Epidemiol
of YY1-binding sites in episomal HPV 16 DNA from cervical cancers. Int J 1977;105:948.
Cancer 1994;58:8038. 234 Coursaget P, Yvonnet B, Chotard J, et al. Age and sex-related study of
203 May M, Dong X-P, Beyer-Finkler E, Stubenrauch F, et al. The E6/E7 hepatitis B virus chronic carrier state in infants from an endemic area
promoter of extrachromosomal HPV16 DNA in cervical cancers escapes (Senegal). J Med Virol 1987;22:15.
from cellular repression by mutation of target sequences for YY1. EMBO J 235 Evans AS, Niederman JC. Epstein-Barr virus. In: Evans AS ed. Viral
1994;13:14606. infections of humans, 3rd ed. New York: Plenum, 1989:26592.
204 Kessis TD, Slebos RJ, Nelson WG, et al. Human papillomavirus 16 E6 236 Smith JS, Muoz N, Herrero R, et al. Evidence for Chlamydia
expression disrupts the p53-mediated cellular response to DNA damage. trachomatis as an HPV cofactor in the etiology of invasive cervical cancer
Proc Natl Acad Sci U S A 1993;90:398892. in Brazil and the Philippines. J Infect Dis 2002;185:32431.
205 Scheffner M, Munger K, Byrne JC, et al. The state of the p53 and 237 Castle PE, Hillier SL, Rabe LK, et al. An association of cervical
retinoblastoma genes in human cervical carcinoma cell lines. Proc Natl inflammation with high-grade cervical neoplasia in women infected with
Acad Sci U S A 1991;88:55237. oncogenic human papillomavirus (HPV). Cancer Epidemiol Biomarkers
206 Liotta LA, Kohn EC The microenvironment of the tumour-host interface. Prev 2001;10:10217.
Nature 2001;411:3759. 238 de Sanjose S, Bosch FX, Muoz N, et al. Social differences in sexual
207 Saveria Campo M. Towards vaccines against papillomavirus. In: Stern behaviour and cervical cancer. In: Kogevinas M, Pearce N, Susser M, et
PL, Stanley MA, eds. Human papillomaviruses and cervical cancer. al, eds. Social inequalities and cancer. Lyon: IARC Scientific Publications,
Biology and immunology. New York: Oxford University Press, 1997 (No. 138):30917.
1994:17782. 239 Potischman L, Brinton LA. Nutrition and cervical neoplasia. Cancer
208 Tindle RW. Vaccines for human papillomavirus infection and anogenital Causes Control 1996;7:11326.
disease. Texas: RG Landes Company, 1999. 240 Storey A, Thomas M, Kalita A, et al. Role of p53 polymorphism in the
209 de Villiers EM. Taxonomic classification of papillomavirus. development of human papillomavirus-associated cancer. Nature
Papillomavirus Report 2001;12:5763. 1998;393:22934.
210 Riou G, Favre M, Jeannel D, et al. Association between poor prognosis 241 Makni H, Franco E, Kaiano J, et al. p53 polymorphism in codon 72 and
in early-stage invasive cervical carcinomas and non-detection of HPV risk of HPV-induced cervical cancer: effect of interlaboratory variation. Int
DNA. Lancet 1990;335:11714. J Cancer 2000;87:52833.
211 Viladiu P, Bosch FX, Castellsague X, et al. Human papillomavirus DNA 242 Madeleine MM, Shera K, Schwartz SM, et al. The p53 Arg72Pro
and antibodies to human papillomaviruses 16 E2, L2, and E7 peptides polymorphism, human papillomavirus, and invasive squamous cell
as predictors of survival in patients with squamous cell cervical cancer. J cervical cancer. Cancer Epidemiol Biomarkers Prev 2000;9:2257.
Clin Oncol 1997;15:61019. 243 Rezza G, Giuliani M, Garbuglia AR, et al. Lack of association between
212 Tabrizi SN, Fairley ChK, Chen Sh, et al. Epidemiological characteristics p53 codon-72 polymorphism and squamous intraepithelial lesions in
of women with high grade CIN who do and do not have human women with, or at risk for, human immunodeficiency virus and/or human
papillomavirus. Br J Obstet Gynaecol 1999;106:2527. papillomavirus infections. Cancer Epidemiol Biomarkers Prev
213 Burger MPM, Hollema H, Pieters WJLM, et al. Epidemiological evidence 2001;10:5656.
of cervical intraepithelial neoplasia without the presence of human 244 Zehbe I, Voglino G, Wilander E, et al. p53 codon 72 polymorphisms
papillomavirus. Br J Cancer 1996;73:8316. and various human papillomavirus 16 E6 genotypes are risk factors for
214 Muoz N, Bosch FX, Kaldor JM. Does human papillomavirus cause cervical cancer development. Cancer Res 2001;61:60811.
cervical cancer? The state of the epidemiological evidence. Br J Cancer 245 Koskela P, Antilla T, Biorge T, et al. Chlamydia trachomatis infection as
1988;57:15. a risk factor for invasive cervical cancer. Int J Cancer 2000;85:359.
215 Schiffman MH, Schatzkin A. Test reliability is critically important to 246 van den Brule AJC, Kjaer SK, Munk C, et al. Development of HPV
molecular epidemiology: an example from studies of human induced cervical intraepithelial neoplasias: infection with Chlamdya
papillomavirus infection and cervical neoplasia. Cancer Res trachomatis as a possible cofactor. In: Castellsagu X, Bosch FX, de
1994;54(suppl 7):1944s7s. Sanjos S, et al, eds. 18th International Papillomavirus
216 Walboomers JM, Jacobs MV, Manos MM, et al. Human papillomavirus Conferenceprogram and abstracts book. Barcelona: Thau SL,
is a necessary cause of invasive cervical cancer worldwide. J Pathol 2000:130 (available on line: http:/www.hpv2000.com).
1999;189:129. 247 Villa LL, Franco EL. Epidemiologic correlates of cervical neoplasia and
217 Deacon JM, Evans CD, Yule R, et al. Sexual behaviour and smoking as risk of human papillomavirus infection in asymptomatic women in Brazil.
determinants of cervical HPV infection and of CIN 3 among those J Natl Cancer Inst 1989;81:33240.
infected: a case-control study nested within the Manchester cohort. Br J 248 Guerrero E, Daniel RW, Bosch FX, et al. Comparison of ViraPap,
Cancer 2000;88:156572. Southern hybridization, and polymerase chain reaction methods for
218 de Sanjose S, Muoz N, Bosch FX, et al. Sexually transmitted agents human papillomavirus identification in an epidemiological investigation
and cervical neoplasia in Colombia and Spain. Int J Cancer of cervical cancer. J Clin Microbiol 1992;33:2229.
1994;56:35863. 249 zur Hausen H. Cervical carcinoma and human papillomavirus: on the
219 Bayo S, Bosch FX, de Sanjos S, et al. Risk factors of invasive cervical road to preventing a major human cancer. J Natl Cancer Inst
cancer in Mali. Int J Cancer [In press.] 2001;93:2523.
www.jclinpath.com
250 Bosch FX, Muoz N, de Sanjos S, et al. Re: cervical carcinoma and 264 Lehtinen M, Luukkaala T, Wallin K-L, et al. Human papillomavirus
human papillomavirus: on the road to preventing a major human cancer. infection, risk for subsequent development of cervical neoplasia and
J Natl Cancer Inst 2001;93:134950. associated population attributable fraction. J Clin Virol
251 IARC. Human papillomavirus and cervical cancer. Muoz N, Bosch FX, 2001;22:11724.
Jensen OM, eds. Lyon: IARC Scientific Publications, Vol. 94, 1989. 265 Anttila T, Saiku P, Koskela P, et al. Serotypes of Chlamydia trachomatis
252 IARC. The epidemiology of cervical cancer and human papillomavirus. and risk for development of cervical squamous cell carcinoma. JAMA
Muoz N, Bosch FX, Shah KV, et al, eds. Lyon: IARC Scientific 2001;285:4751.
Publications, Vol. 119, 1992. 266 Andersson S, Rylander E, Larsson B, et al. The role of human
253 Meijer CJLM, Brule AJC van den, Snijders PJF, et al. Detection of human
papillomavirus in cervical adenocarcinoma carcinogenesis. Eur J Cancer
papilloma virus in cervical scrapes by the polymerase chain reaction in
2001;37:24650.
relation to cytology: possible implications for cervical cancer screening.
In: Muoz N, Bosch FX, Shah KV, Meheus A, eds. The epidemiology of 267 Clavel C, Masure M, Bory J-P, et al. Human papillomavirus testing in
cervical cancer and human papillomavirus. Lyon: IARC, 1992:27181. primary screening for the detection of high-grade cervical lesions: a study
254 Cuzick J, Beverley E, Ho L, et al. HPV testing in primary screening of of 7932 women. Br J Cancer 2001;84:161623.
older women. Br J Cancer 1999;81:5548. 268 Goldie SJ, Kuhn L, Denny L, et al. Policy analysis of cervical cancer
255 Coutle F, Gravitt P, Richardson H, et al. Nonisotopic detection and screening strategies in low-resource settings. Clinical benefits and
typing of human papillomavirus DNA in genital samples by the line blot cost-effectiveness. JAMA 2001;285:310715.
assay. J Clin Microbiol 1999;37:18527. 269 Schwartz SM, Daling JR, Shera KA, et al. Human papillomavirus and
256 Sellors JW, Mahony JB, Kaczorowski J, et al. Prevalence and predictors prognosis of invasive cervical cancer: a population-based study. J Clin
of human papillomavirus infection in women in Ontario, Canada. Can Oncol 2001;19:190615.
Med Assoc J 2000;163:5038. 270 Liaw K-L, Hildesheim A, Burk RD, et al. A prospective study of human
257 Meijer CJLM, Walboomers JMM. Cervical cytology after 2000: where papillomavirus (HPV) type 16 DNA detection by polymerase chain
to go? J Clin Pathol 2000;53:413. reaction and its association with acquisition and persistence of other HPV
258 The Atypical Squamous Cell of Undetermined Significance/ types. J Infect Dis 2001;183:815.
Low-grade Squamous Intraepithelial Lesions Triage Study (ALTS) 271 Gravitt PE, Lacey JV, jr, Brinton LA, et al. Evaluation of self-collected
Group. Human papillomavirus testing for triage of women with cytologic cervicovaginal cell samples for human papillomavirus testing by
evidence of low-grade squamous intraepithelial lesions: baseline data polymerase chain reaction. Cancer Epidemiol Biomarkers Prev
from a randomized trial. J Natl Cancer Inst 2000;92:397402. 2001;10:95100.
259 Follen M, Richards-Kortum R. Emerging technologies and cervical
272 Harro CD, Pang Y-YS, Roden RBS, et al. Safety and immunogenicity trial
cancer. J Natl Cancer Inst 2000;92:36365.
in adult volunteers of a human papillomavirus 16 L1 virus-like particle
260 Sasieni PD. Human papillomavirus screening and cervical cancer
vaccine. J Natl Cancer Inst 2001;93:28492.
prevention. J Am Med Wom Assoc 2000;55:21619.
261 Stevenson , Hudson LC, Burns JE, et al. Inverse relationship between the 273 Peyton CL, Gravitt PE, Hunt WC, et al. Determinants of genital human
expression of the human papillomavirus type 16 transcription factor E2 papillomavirus detection in a US population. J Infect Dis
and virus DNA copy number during the progression of cervical 2001;183:155464.
intraepithelial neoplasia. J Gen Virol 2000;81:182532. 274 Multiple authors. From epidemiology to policy: a symposium on the
262 Carter JJ, Koutsky LA, Hughes JP, et al. Comparison of human translation of epidemiologic evidence into public health policy. Am J
papillomavirus types 16, 18 and 6 capsid antibody responses following Epidemiol (Special Issue) 2001;154.
incident infection. J Infect Dis 2000;181:191119. 275 Lanes S. Causal inference is not a matter of science. Am J Epidemiol
263 Franco EL, Duarte-Franco E, Ferenczy A. Cervical cancer: epidemiology, 1985;122:550.
prevention and the role of human papillomavirus infection. Can Med 276 Rothman KJ, Poole C. Science and policy making. Am J Public Health
Assoc J 2001;164:101725. 1985;75:3401.
www.jclinpath.com
The causal relation between human

papillomavirus and cervical cancer
F X Bosch, A Lorincz, N Muoz, C J L M Meijer and K V Shah
J Clin Pathol 2002 55: 244-265

doi:
Updated information and services can be found at:

http://jcp.bmj.com/content/55/4/244
These include:
References This article cites 221 articles, 83 of which you can access for free at:
http://jcp.bmj.com/content/55/4/244#BIBL
Email alerting Receive free email alerts when new articles cite this article. Sign up in the
service box at the top right corner of the online article.
Notes
To request permissions go to:

http://group.bmj.com/group/rights-licensing/permissions
To order reprints go to:

http://journals.bmj.com/cgi/reprintform
To subscribe to BMJ go to:

http://group.bmj.com/subscribe/

244 Full PDF

Uploaded by

Copyright:

Available Formats

244 Full PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

244 Full PDF

Uploaded by

Copyright:

Available Formats

Downloaded from http://jcp.bmj.com/ on September 8, 2015 - Published by group.bmj.

The causal relation between human papillomavirus and

J Clin Pathol 2002;55:244265

The causal role of human papillomavirus infections in INTRODUCTION

HPV and cervical cancer 245

The implications of the recognition that, in the absence of

246 Bosch, Lorincz, Muoz, et al

Table 1 Epidemiological considerations important for causal inference

Adapted from Weed 1995.13

HPV and cervical cancer 247

Temporality of which five involve biological agents such as HPV.12 In its

Controls Adeno and mixed Squamous

9* 2491 13.4 2365 90.7

248 Bosch, Lorincz, Muoz, et al

Table 4 Prevalence of multiple human papillomavirus (HPV) types in patients with

% Of all % Of all the % Of all

32 IARC multicentre 420% 10% 13%

IARC, International Agency for Research on Cancer.

HPV and cervical cancer 249

of specimens from cases and from the general population that

proportions of multiple types than do populations not belong- Das et al (1992)

exposure to HPV.31 However, in all studies of invasive

250 Bosch, Lorincz, Muoz, et al

0.1 Reference 5 10 50 100 500 2000

HPV and cervical cancer 251

In conclusion, although the specificity criteria can be viewed Follow up studies

252 Bosch, Lorincz, Muoz, et al

HPV and cervical cancer 253

The pattern of transcription changes as the lesion increases in

254 Bosch, Lorincz, Muoz, et al

HPV and cervical cancer 255

YYI sites,201203 which appear to allow continuous expression of

Biological mechanisms of HPV carcinogensis. Exclusion of alternative explanations

256 Bosch, Lorincz, Muoz, et al

HPV and cervical cancer 257

*References adapted from the results of the IARC multicentre study.23

258 Bosch, Lorincz, Muoz, et al

Presence of risk factor N % N % N %

HPV 341 22.9 186 23.1 186 23.1

Smk 8 0.5 1 0.1 0 0

Preg 15 1.0 21 2.61 1 0.1

None of the four 23 1.5 25 3.1 2 0.2

Total number of cases 1490 100 806* 100 806* 100

Table 7 Theoretical models of sufficient causes for cervical cancer and

HPV, human papillomavirus.

HPV and cervical cancer 259

studies found small and still inconsistent effects from

260 Bosch, Lorincz, Muoz, et al

HPV and cervical cancer 261

262 Bosch, Lorincz, Muoz, et al

HPV and cervical cancer 263

264 Bosch, Lorincz, Muoz, et al

HPV and cervical cancer 265

The causal relation between human

J Clin Pathol 2002 55: 244-265

Updated information and services can be found at:

To request permissions go to:

To order reprints go to:

To subscribe to BMJ go to: