Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Chapter 1: Human Papillomavirus and Cervical Cancer—Burden and Assessment of Causality

Chapter 1: Human Papillomavirus and Cervical Cancer—Burden and Assessment of Causality Abstract Cervical cancer remains the second most common cancer in women worldwide and the most frequent in developing countries. Pre-neoplasic cervical lesions represent an additional burden in countries where screening is widespread. The human papillomavirus (HPV) prevalence and type distribution in normal smears and in cancer specimens are being described and show relatively small international variation. State-of-the-art detection techniques have unequivocally shown that HPV-DNA can be detected in 95% to 100% of adequate specimens of cervical cancer, supporting the claim that HPV is the necessary cause. The odds ratios for cervical cancer related to a cross sectional detection of HPV-DNA range from 50 to several hundred in all studies. The risk for any of 15 high-risk types is not statistically different from the risk reported for HPV16. The estimates of the attributable fraction range from 90% to 98%. Additional work should be done in providing information on incidence of cervical cancer and on HPV infection in areas where the disease is common. Theoretical work including modeling of the incidence could be of potential use in the evaluation of the existing and novel preventive strategies. Research is currently being conducted on the mechanisms of HPV carcinogenesis. These include the determinants of the systemic and cellular immune response to the viral infection, the interaction between the host and the virus and the relevance of the different strains and variants of the HPV viral types. Technology developments in this area suitable for epidemiological studies are needed. In most developing countries, cervical cancer is the leading female malignancy and a common cause of death among middle-aged women. In developed populations with good screening options, invasive cervical cancer is a relatively rare condition, whereas its precursors and the equivocal cytological results represent a major health burden. Research at the population level has provided an exhaustive body of evidence on the viral etiology of cervical cancer, and on the human papillomavirus (HPV) types involved. Most of the progress made by epidemiological studies has been based on DNA technology to characterize the presence of the human papillomavirus DNA (HPV-DNA) in cervical specimens. Investigations have demonstrated that a range from 2% to more than 20% of the world’s female population have detectable HPV-DNA in their cervix at any time. Further, it has been established that, under certain conditions, these infections persist and are able to induce CIN2/3 and cervical cancer. Currently, the viral DNA can be detected in virtually all cases of invasive cancer. High parity, smoking, and long-term use of oral contraceptives have been established as risk cofactors for cervical cancer among women with persistent infections with HPV. Other sexually transmitted diseases such as Chlamydia trachomatis, herpes simplex virus type 2, and human immunodeficiency virus as well as some poorly known dietary factors are likely intervening factors. Continued research should focus on producing better estimates of the burden of disease linked to papillomaviruses. In the future, cancer registries should play an important role by monitoring cervical cancer incidence in populations vaccinated against HPV. Additional investigations should further explore the natural history and the cofactors that modulate the risk of cancer among HPV infected women. Burden of HPV Infection, Cervical Neoplasia, and Cervical Cancer Definite progress in understanding the etiology of cervical cancer has been achieved, and some types of HPV have been established as the central cause of cervical cancer worldwide (1). Further, technology is available to detect the presence of viral DNA in large volumes of cervical specimens, and a substantial number of epidemiological investigations are providing estimates on the prevalence and distribution of the HPV infections in human populations. Serological measurements of the type-specific antibody levels against HPV capsid antigens combined with DNA studies have provided evidence that most of the sexually active women (up to 70% in some populations in the United States) have been exposed to HPV at any point in time. Seroconversion following HPV exposure is, however, not universal; and with current detection methods, a range between 40 and 50% of HPV-DNA 16 positive cervical cancer cases do not show HPV16 antibodies. The natural history and the determinants of the immune response to HPV are still poorly understood, and antibody measurements have had limited use, thus far, in epidemiological studies. Prevalence of HPV-DNA in Cervical Specimens from the General Population Table 1(2–27) shows a selection of studies that investigated the prevalence of HPV-DNA in women from the general population in different regions in the world. Studies are included in the table for the following reasons: 1) they represent in as much as possible the underlying populations, although not necessarily of the country or the corresponding world’s region; 2) they used polymerase chain reaction (PCR) or Hybrid Capture 2 (HC2) technology to assess the presence of the viral DNA; and 3) they include a wide age range, a study size of at least 200 individuals, and had a well described methodological section in their published account. The table shows two prevalence estimates, one for all women included in the study, and one for women ages 30 years and above. The latter is generally considered an approximation of the prevalence of persistent carriers. This interpretation recognizes that most HPV infections are acquired soon after sexual initiation and are also cleared at an early age. Women who are HPV-DNA positive at ages 30 and above (allowing for certain intercountry variation) may represent failure of viral clearance and select the group of women at a relative higher risk of cervical cancer. The results displayed in Table 1 should be viewed as an approximation to the burden of the viral infection estimated by the HPV-DNA detection rate in cervical scrapes at a point in time. Limitations to current data include the variability introduced by the selection of the population, the sampling methods, and by the variability of the techniques used for HPV detection, specifically when nonstandardized PCR methods are used locally. In countries with intensive screening among young women, aggressive treatments of early or ambiguous cytology lesions (most of which are likely to be HPV-related) may have resulted in a reduction of the estimated HPV-DNA prevalence in more advanced age groups. In a few epidemiological studies, the exact same methods in the field and in centralized laboratories have been used to investigate populations of contrasting incidence of cervical cancer (i.e., Colombia and Spain, North and South Vietnam, International Agency for Research on Cancer (IARC’s) multicentric case control studies, different surveys in the United Kingdom). These studies have consistently shown strong geographical correlations between HPV-DNA prevalence and cervical cancer incidence. HPV Prevalence in Cervical Cancer Cases and HPV-type Distribution in Cases and Controls State-of-the art amplification techniques used in clinical and epidemiological studies have shown unequivocally that in adequate specimens of cervical cancer, HPV-DNA can be detected in 90% to 100% of the cases in any population where investigations have taken place. This compares with a prevalence of 5% to 20% from cervical specimens of women without cervical cancer, from the same populations (i.e., prevalence surveys or control groups in case control studies). Detailed investigations of the few cervical cancer specimens that appear as HPV-DNA negatives on routine testing strongly suggest that these are largely false negatives (28,29). As a consequence, the claim has been made that this is the first necessary cause of a human cancer ever identified, and it provides a strong rationale for the use of HPV tests in screening programs and for the development of HPV vaccines. Of the more than 35 HPV types found in the genital tract, HPV16 accounts for 50% to 60% of the cervical cancer cases in most countries, followed by HPV18 (10%–12%) and HPVs 31 and 45 (4%–5% each). Fig. 1 shows the cumulative distribution of the most common types in close to 5000 specimens included in the IARC’s multicenter cervical cancer studies. Among cases, HPVs 16, 18, 45, 31, and 33 accounted for 80% of the type distribution in squamous cell carcinomas and HPVs 16, 18, 45, 59, and 33 accounted for 94% of the type distribution in adenocarcinomas. To complete the additional 6% to 20%, the required number of HPV types increases to 20–23 types, each one of which would contribute between 3% and 0.1% of the types. These results should be viewed as indicative, because some of the testing systems used may be relatively less sensitive to the rarer types, notably in the presence of multiple infections in the same specimen. In most studies, HPV18 predominates in adenocarcinomas in absolute or in relative terms, however, the reasons for such specificity are still unknown. In the same studies, HPVs 16, 18, 45, 31, and 33 accounted for 46% of the prevalence observed in HPV-DNA positive women with normal cervical smears. These distributions are generally consistent worldwide (28,30), although several reports using highly sensitive HPV-DNA detection and typing systems are finding additional geographical variability (2,31–33), and for several regions in the world the information is either lacking or rather scarce. Discussion points/recommendations are as follows: 1) It would be important to have updates of the international literature on the prevalence and type distribution of HPV-DNA in the general population and in cervical cancer cases. These estimates should consider the various parameters that affect detection, notably the testing system and the biological specimens used. Such reviews may have implications for HPV vaccine composition. 2) It would be desirable to conduct additional surveys designed to cover the gaps in the maps, notably by including selected areas in Africa and Asia and in populations heavily exposed to the human immunodeficiency virus (HIV), malnutrition, and malaria. Assessment of the prevalence of viral variants of the most common types remains to be done. Methodological exercises have to be conducted to determine the proper HPV-DNA typing methods to be used, particularly if multiple vital types are common. 3) It is still worth replicating the findings of investigations on putative HPV-negative cervical cancer cases using, for example, microdissection technology and in situ hybridization methods to confirm that the 5–15% HPV-negative cases that are currently being reported from field or clinical studies are generally false-negatives. Incidence and Mortality of Cervical Cancer Table 2 shows estimates of the incidence of invasive cervical cancer as extracted from systematic sources, and compiled by IARC for the years 1985, 1990, and 2000 (34–36). The data included in the table have been calculated from incidence rates, mortality vital statistics, and census records. For some under-investigated populations, other ancillary sources and country-to-country extrapolations are occasionally used. Thus, the reliability of the figures in the table may differ by region. Table 2 shows the impact of the adjustment methods on the perception of risk by geographical regions. For example, crude rates are similar when the world is broadly divided into developing and developed regions; however, age-adjusted rates using the age structure of an artificial world standard population tend to decrease the estimates for developed (older) populations and to increase the estimates for less developed (younger) populations. Comparison of columns 2 to 4 (Table 2) shows one of the limitations of current estimates. When estimates for 1985 and 2000 are related, there appears to be a slight reduction in the total number of cases in developed countries and an increase in developing regions, notably in Africa, The Caribbean, and Central and South America. Estimates for 1990 are lower, strongly driven by an inconsistent reduction in the estimates for the different regions in Asia. Some authors have also reported substantial reductions in the mortality from cervical cancer in China in the interval 1970–1994 (37). However, artifacts due to differences in the diagnostic and registration methods cannot be excluded, notably when a few incidence registries are used to produce estimates for large populations. From the same source, the number of cervical cancer deaths in the world has been estimated as 233 400, which is close to half of the incident cases. The 5-year prevalence of cervical cancer has been estimated at 1.4 million cases (36). The estimated total number of cases of cervical cancer as presented in Table 2 is being largely accepted as a background reference. It is also often used for the promotion of preventive interventions in both developed and developing regions. However, to fully estimate the impact of cervical cancer, the limitations of current estimates have to be recognized and it is important to reach a comprehensive interpretation of the different parameters reflecting disease burden such as the four (number of cases, crude rates, age adjusted rates, and age-specific rates) presented in Table 2. In broad terms, it should be recognized that cervical cancer is a strong determinant of women’s cancer incidence and mortality in developing countries and that current figures may still underestimate its impact due to under diagnosis and under registration. Developed countries that have introduced massive screening have successfully shifted the focus of attention toward pre-invasive neoplasm. However, in large developed regions with unevenly implemented screening programs and an aging population, like in Europe, the total number of cases of cervical cancer diagnosed (some 65 000 cases) is still at a similar level as the one estimated for Africa (some 67 000 cases) and not far from the estimates for Central and South America combined (some 70 000 cases). Discussion points/recommendations are as follows: Cancer registries should continue to monitor the incidence of cervical cancer and to allow the evaluation of preventive strategies implemented in their populations. Updated estimates on the burden of HPV related cancers may have to expand to include other cancer sites for which an etiological role of HPV has been demonstrated (see also chapters 7, 8, and 9). Time Trends in Cervical Cancer Incidence by Histology Individual cancer registries provide the information required to estimate time trends in incidence in their populations. In countries with organized screening, decreases in rates for cervical cancer have been reported consistently, whereas increases in the incidence of cervical adenocarcinomas have been reported in several populations (38–40). Cervical adenocarcinomas currently represent 10% to 12% of cervical cancer and are an increasing concern of cytology-based screening programs. It is likely that these upward trends are the results of major changes in sexual behavior initiated in the 1950s and 1960s in their populations and to the failures of screening programs in the subsequent decades. It is known that cytology-based screening has a relatively lower sensitivity for the early glandular lesions than for the squamous-cell cancer precursors. Alternative explanations could consider the possibility that the underlying prevalence of the HPV types more related to cervical adenocarcinoma is changing over time (i.e., that the trends in prevalence of HPV18 are increasing). In addition, changes in the exposure of the populations to relevant cofactors should be considered. Discussion point/recommendation follows: It would be of interest to verify if there are any underlying time trends in the type distribution of HPV in cervical cancer. Studies of decades-old archival cervical cells or tissues are warranted. Age-Specific Incidence of Cervical Cancer In most developed countries, cervical cancer incidence shows a unique profile across age groups in that rates increase sharply after ages 25–30 years to reach a plateau after ages 45–50 years. With the availability of results from cancer registries in developing countries, it has been possible to recognize that, in the absence of screening, cervical cancer tends to follow a linear relationship with age, which is the pattern of the majority of the epithelial tumors. The shape of the age-specific incidence curve seems to be highly dependent on the screening efforts. Fig. 2 A shows an example from the United Kingdom and Brazil (41). Both countries show similar rates of invasive cervical cancer to age 30 years, suggesting similar levels of exposure to HPV, but rapidly separate apart in more advanced age groups, probably as a consequence of the organized screening program in the United Kingdom. The arithmetic scale has been chosen on this occasion for better visualization of differences. Figure 2 B(41) shows significant differences in the age-specific incidence rates in northern America across ethnic groups. Data on white populations from the Surveillance, Epidemiology and End Results Program (SEER1) and Canadian registries show that incidence rises steeply between ages 20 and 35 and stays stable in the middle-aged groups, with some slight increase after age 55 years. In contrast, the incidence increases steadily throughout lifetime for Blacks, Hispanic, and Chinese (with an unexplained and short-lasting dip between age groups 65–75 years), reaching two to three times the incidence among whites in the most advanced age groups. Migrant populations from high- to low-risk countries may have an important impact on the incidence of cervical cancer, notably if they remain outside of the screening system. 1 Editor’s note: SEER is a set of geographically defined, population-based, central cancer registries in the United States, operated by local nonprofit organizations under contract to the National Cancer Institute (NCI). Registry data are submitted electronically without personal identifiers to the NCI on a biannual basis, and the NCI makes the data available to the public for scientific research. In some countries, a second mode in the incidence of cervical cancer among women in the age groups over 55–60 years is observed. Fig. 2 C shows these distributions for some selected cancer registries in Europe. The interpretation of the second mode remains uncertain. Decrease in screening coverage or lower sensitivity of cytology in the old age groups are plausible explanations but immunosuppression or acquisition of new HPV infections during middle age cannot be ruled out. In the age groups 35 years and above, the presence of HPV-DNA in cervical cells probably reflects a combination of long-term persistent infections acquired in the younger age groups and newly acquired late infections. New infections should be largely related to changes in sexual behavior of both sexes. It can be hypothesized that, after a certain latency period, the presence of a recently acquired viral DNA in the middle-aged group, with or without the concurrent effect of cofactors, is capable of generating a second increase in the incidence of CIN3 and of invasive cancer (see also a discussion on the “Incidence of CIN3” section below). The predicted height of the second mode is modulated (attenuated) by both screening in these age groups and competing causes of death. Discussion points/recommendations are as follows: 1) Cohort studies should continue to provide data on sexual behavior during middle age and above age groups, notably among males. 2) Analyses of cervical cancer incidence by birth cohorts in different populations should be encouraged. These could relate current incidence of cervical cancer to past population-adopted changes in sexual behavior and to major migrant changes. Incidence of CIN3 There is limited information on the incidence of CIN3 lesions. However, in terms of burden of cervical pathology and the requirements imposed to the health systems in developed countries, it is likely that CIN3 is one of the key diagnoses to consider. Figure 3 shows estimates of the incidence of carcinoma in situ/CIN3 and of invasive cancer as provided by some registries. The figure shows very high rates of CIN3 in the young age groups (i.e., >400 per 100 000 in the 20 to 30 years old in the East Anglia Cancer Registry in the United Kingdom). Occasionally, a second mode in the 50 years and above age groups is observed. It is still uncertain which proportion of these CIN3 cases are deemed to progress to cervical cancer as compared with representing transient morphological changes. The population-based ratio of CIN3-to-invasive cancer seems a parameter of relevance that may provide indirect estimates of the amount of over diagnosis and overtreatment of pre-invasive lesions brought about by screening programs. For example, Fig. 3 A shows that in Finland, a country with a long-term, centralized, and well-established screening program, there are 1.6 cases of CIN3/in situ carcinoma per case of invasive cancer diagnosed. In contrast, the equivalent ratios for East Anglia and Iceland are 13.0 and 16.2, respectively (Fig. 3 B and C). It is uncertain if, in the absence of screening, all these CIN3/in situ cases would have progressed to invasive cancer. These observations may also reflect the variability in diagnosis and treatment of pre-invasive disease across populations. Discussion points/recommendations are as follows: It would be of interest to stimulate additional work on the incidence of pre-invasive lesions, including additional collection of data on the incidence of CIN3/carcinoma in situ and analysis of the available data from cancer registries. These analyses could relate incidence to the age-specific prevalence of HPV-DNA and to the corresponding incidence of invasive cancer. Analysis of the ratios of CIN3/cervical cancer considering other clinical parameters (i.e., standardization of the CIN3 diagnosis and pathology results from conization specimens) may allow estimates of the fraction of the CIN3 cases that are true precursors of cervical cancer in the context of specific screening practices. Newly organized screening programs or screening trials in developing populations (i.e., for HPV testing or direct visual inspection methods) should be able to estimate ratios of CIN3 to invasive cervical cancer in populations in the absence of preventive interventions. Analysis of the age at occurrence of CIN3 and of the ages at sexual initiation may provide new insights into the latency period for HPV and cancer. Modeling Geographic Variation in Cervical Cancer Incidence as a Basis for Classifying Populations in Relation to Their Risk of Cervical Cancer Maps of cervical cancer incidence based on age-adjusted rates tend to indicate that countries at high-risk cluster in sub-Saharan Africa, Latin America, and some parts of Asia. Identification of key environmental factors that may modify the risk of HPV exposure and the incidence of cervical cancer offers new and attractive opportunities to generate models of cervical cancer incidence. These estimates would allow a fresh look at the classification of populations in relation to their risk independently of the strong impact of screening. At a population level, high incidence rates of cervical cancer are observed and can be predicted for populations with high HPV-DNA background prevalence, paired with (and consequent to) an early age at sexual initiation, high number of partners of both men and women, and an important frequency of sexual contacts with prostitutes. Typically, these populations also have high parity and high frequency of exposure to other risk factors (e.g., oral contraceptives, smoking, other sexually transmitted diseases, poor hygiene, and cervical inflammation). Over and above their exposure levels, high-risk populations are dramatically underserved by screening programs or practices. On the reverse side, low-risk populations show either low levels of exposure to these factors and/or belong to communities with centralized and quality-controlled screening activity. As expected, countries and populations include unequal combinations of these risk and protection factors. Developed countries share many high-risk traits, largely compensated by important screening efforts and decreasing parity. The ratio of incident cases of CIN3/invasive cervical cancer tends to be high (i.e., 10/1). Examples of such countries could be the United States or some of the Nordic countries in Europe. Developing countries under strict religious (behavioral) influence show some low-risk traits (regulated sexual behavior, low prevalence of exposure to cofactors) and some high-risk traits such as high parity and absence of screening programs. These countries tend to show low incidence rates of cervical cancer, although the number of registries in these populations is also limited. Examples of this model would be the Muslim countries in the Middle East or Asia. Other developing countries without strict religious (behavioral) regulations accumulate high-risk traits not compensated by screening efforts. These situations result in the highest incidence rates of invasive disease recorded. Examples of this could be some populations in Central and Latin America. The combination of these factors into a model of disease incidence could be used to predict incidence of cervical cancer by birth cohorts when properly placed into population-specific time scales. This information would provide the rationale for the continuous support and development of screening programs. The models could also be used to estimate indirectly the efficiency of screening programs in developed countries, by comparing the predicted and the observed incidence rates. An ecological project along these lines has recently reported a strong correlation between predominant religion and cervical cancer in developing countries (42). Discussion points/recommendations are as follows: The acquisition of novel data on HPV prevalence and its determinants in a variety of regions covered by cancer registries encourages systematic analysis leading to the construction of statistical models of cervical cancer incidence. Migrant studies and HPV variant distributions could be of value. Causality Assessment: HPV and Other Risk Factors for Cervical Cancer The nature of the association between HPV and cervical cancer has been exhaustively investigated and since the early 1990s, and all academic reviews have consistently concluded that the evidence fulfills most of the established criteria of causality (1). The mechanisms by which HPV induces cancer in humans and the molecular genetics of the process are being intensively investigated and excellent reviews are available (43). To evaluate the role of additional factors once it was recognized that the role of HPV-DNA was a necessary factor it became a standard procedure to restrict the analysis to women who were HPV-DNA-positive (thus at risk of ever developing cervical cancer) in addition to the standard multivariate methods. Table 3(1,44–48) summarizes the established risk factors of cervical cancer in women who are carriers of HPV-DNA, and Table 4(49–51) presents the factors that are significantly related to cervical cancer in HPV-DNA positive women, but for which conclusive evidence is less established. As before, most of the conclusions and references quoted are results of the IARC’s multicenter study, which are consistent with the updated literature. Current Risk Estimates for HPV-DNA and HPV Type-Specific Estimates Current estimates of the odds ratio for HPV-DNA and cervical cancer are among the highest observed in human cancer. The final results of the IARC multicenter study reports an estimate of invasive cancer for HPV-DNA of OR = 158 (95%CI = 113.4 to 220.6) (44). Corresponding risk estimates for HPV16 was OR = 434.5 (95% CI = 278.2 to 678.7) and for HPV18 OR = 248.1 (95%CI = 138.1 to 445.8). The size of the project has allowed the generation of estimates of the associations between individual HPV types and cervical cancer. According to these results, any of at least 15 high-risk types, namely HPVs 16, 18, 45, 31, 52, 33, 58, 35, 59, 51, 56, 39, 68, 73, and 82 (IS39 / W13B, MM4), convey a statistically similar risk of developing cervical cancer. HPVs 26, 53, and 66 probably are also high-risk types; however, the limited number of observations with these viral types precluded a firmer conclusion. Nonsignificant differences between point estimates may reflect the overwhelming predominance of some of the types (i.e., HPVs 16 and 18) over other types. This IARC study also confirmed the strong correspondence between the epidemiological risk classification of HPV types and the phylogenetic classification which is based on comparisons of DNA sequences across types (44). Relative risks generated by cohort studies with limited HPV observations and short-term follow-up tend to be smaller than odds ratios generated by case–control studies. These studies are informative of the putative pre-invasive stages (low-grade squamous intra-epithelial lesions [LSIL] and high-grade squamous intra-epithelial lesions [HSIL]) but are difficult to interpret in terms of causality of cervical cancer. This is due to censoring of the women at highest risk because of treatment. Moreover, if in fact a fraction of the CIN3 cases (and most importantly if HSIL is the diagnostic endpoint) are not true cancer precursors, misclassification of the diagnosis may partly explain the lower risk estimates in cohort studies. Long-term cohorts that can reliably establish persistency of the same viral types have clearly established the time sequence between HPV exposure and cervical neoplasia and tend to report risk estimates at the level of the case–control studies referred above. It is plausible that novel DNA detection methods result in higher prevalence of HPV-DNA in women with normal cervical cytology. Some early work in Costa Rica using a different polymerases in the PCR reaction has resulted in an increased estimate of the HPV-DNA prevalence in controls (52). If current underdetection is a general issue, future studies would generate lower risk estimates; however, it is unlikely that any significant modification of the attributable fraction would occur as a result because of the direct relevance of the prevalence of HPV-DNA in cases in the calculations of the attributable fraction. In large prevalence surveys of HPV types in invasive cancer “low-risk HPV types” (e.g., HPV 6 or 11) are anecdotally found as the only type (44). Some cohort studies have likewise described that HPV 6 or 11 are predictors of both LSIL and HSIL (53–54). Experimental studies have occasionally shown that the E6 and E7 genes of low-risk types are capable of interfering with p53 and Rb at a low efficiency level [reviewed in (43)]. These results, although of little clinical relevance, are intriguing in that they indicate that, under certain circumstances, even low-risk types may be able to induce neoplastic changes and cancer. The proportion of multiple types in a given cervical cancer specimen varies across studies and particularly in relation to the HPV detection method used. Populations at high-risk of cervical cancer and populations with high rates of HIV tend to show higher proportions of multiple types as compared wtih populations not belonging to these risk groups (55). In the IARC multicenter study there was no statistically significant increased risk for women with multiple-type infections (OR = 115.6; 95% CI = 69.3 to 192.8) over women positive for only one HPV type (OR = 172.5; 95% CI = 122.2 to 243.6) (44). The finding is consistent in other settings (8), giving support to the notion that each HPV infection acts independently. However, some underdetection of multiple types may have occurred as a consequence of the HPV detection systems used. Research Issues on HPV and Cervical Cancer: Viral Factors Viral factors under investigation include the study of HPV variants, the measurement of HPV viral load, and viral integration. Limited information is available on HPV16 variants, suggesting an increased risk for some of the non-European prototypes (56). Viral load has been proposed as a useful marker in cohort studies with conflicting results (8,57–59). Part of the discrepancies can be related to the endpoints used in cohort studies (usually lower or equal to HSIL). Technology to characterize viral load has not been fully validated. Viral integration into cellular DNA was proposed as a marker of progression to cervical cancer. Integration is rarely seen in pre-invasive disease and invasive cancer shows either episomal or integrated viral DNA. It is still uncertain if integration in stages of HSIL predicts progression to cervical cancer. Research Issues on HPV and Cervical Cancer: Host Factors The host response and, possibly, genetically determined susceptibility remain the less understood of the key variables in the carcinogenic process. For example, experimental studies reporting an increase in risk among carriers of certain polymorphisms suggested that there is a susceptibility factor to HPV in p53 (60). However, these results were only occasionally confirmed by epidemiological data (61). Little has been done in trying to describe familial aggregation of cervical cancer, notably to disentangle the effects of an inherited trait from the effects of shared environment, behavior, and attitudes. One such attempt attributed one-third of the variance in cervical cancer incidence to shared genetic makeup. Discussion points/recommendations are as follows: It is uncertain if we need additional case control studies to investigate the basic HPV association with cervical cancer. New studies should focus on the exploration of additional associations with putative environmental cofactors (e.g., dietary factors) among HPV-DNA positive women, novel viral markers, and markers of viral and host interactions. Epidemiologists would benefit from storing biological materials from studies to be able to test novel hypothesis in well-characterized study populations. Amongst others, it would be desirable to see studies that would include the following: 1) Research into mechanisms of HPV carcinogenesis. Explore in greater detail the occasions in which low-risk HPV types are capable of inducing neoplastic transformation. 2) Consider issues of specificity of viral types (and of variants) to induce either squamous cell or adenocarcinoma in the cervix and of type specific biological advantages that could explain the predominance of HPV16 and to some extent of HPV18 over all other mucosal HPV types. 3) Generate more data on multiple types by determining the viral type implicated in a cancer in which several types are identified (i.e., by microdissecting the tissue and/or with carefully conducted in situ hybridizations) 4) Provide additional data on the risk linked to HPV variants. It would be necessary to develop technology that allows typing and variant detection in large numbers and generate worldwide results in the same manner as it has been done for the HPV type distributions. 5) Standardize technology for measuring viral load and viral integration in routine cohort studies. 6) Characterize new biomarkers of interaction between HPV-DNA and cellular DNA that would signal latency, persistency, or progression to neoplasm. 7) Characterize biomarkers of genetic susceptibility to be used in population studies and pursue research into the familial aggregation of cervical cancer. Table 1. Prevalence of cervical oncogenic HPV-DNA in selected samples of the general population*       HPV-DNA    Region  Testing method  Study population  Age range of study, y  (%+)  Age range above 30†  (%+)  Reference  *HC2 = Hybrid Capture 2; PCR = polymerase chain reaction. The HC2 assay was used in 14 samples without PCR amplification.  †As published or estimated by authors.  Africa      Eastern          Mozambique  PCR/HC2  Survey  14–61  40.0  31–61  30.5  (2)          Zimbabwe  HC2  Routine screening  25–55  42.8  —  —  (3)      Northern          Morocco  PCR  Hospital case-control  18–70  21.6  35–70  21.6  (4)      Southern          South Africa (Black)  HC2  Routine screening  —  —  35–65  21.3  (5)      Western          Senegal  PCR  Hospital case-control  —  —  35–83  43.7  (6)  America      Central          Mexico  PCR  Population-based  <25–>65  14.5  35–>65  14.1  (7)          Costa Rica  PCR  Routine screening  18–94  16.0  —  —  (8)      South          Colombia  PCR  Survey  13–85  14.9  35–85  8.4  (9)          Argentina  PCR  Survey  15–>55  16.6  35–>55  13.9  (10)      Northern          Canada  PCR  Routine screening  15–49  13.3  35–49  9.0  (11)          United States  PCR  Routine screening  16–77  22.5  —  —  (12)  Europe      Eastern          Russian Federation  PCR  Routine screening  15–45  29.0  31–45  25.5  (13)      Northern          United Kingdom  PCR  Routine screening  —  —  34–70  5.9  (14)          Denmark  PCR  Survey  20–29  18.0  —  —  (15)      Southern          Spain  PCR  Population-based  14–75  3.0  35–75  2.0  (15a)          Italy  PCR  Population-based  25–70  8.6  35–70  7.4  (16)          Greece  PCR  Routine screening  20–55  36.2  —  —  (17)      Western          The Netherlands  PCR  Population-based  15–69  4.6  30–69  4.3  (18)          Germany  PCR  Routine screening  18–70  7.8  36–70  4.9  (19)          France  HC2  Population-based  15–76  15.3  31–76  12.3  (20)  Asia      Europe          Japan  PCR  Routine screening  —  —  30–78  7.0  (21)          Korea, DR  PCR  Survey  20–74  10.4  35–74  9.6  (22)          Taiwan  PCR  Routine screening  —  —  30–64  9.2  (23)          China  HC2  Survey  —  —  35–45  18.0  (24)      South Eastern          Vietnam, North  PCR  Survey  15–69  2.0  35–69  1.9  (25)          Vietnam, South  PCR  Survey  15–69  10.9  35–69  8.0  (25)          Thailand  PCR  Population-based  15–>65  6.3  35–>65  5.0  (26)      Southcentral          India (Madras)  PCR  Hospital case-control  23–76  27.7  35–76  28.5  (27)        HPV-DNA    Region  Testing method  Study population  Age range of study, y  (%+)  Age range above 30†  (%+)  Reference  *HC2 = Hybrid Capture 2; PCR = polymerase chain reaction. The HC2 assay was used in 14 samples without PCR amplification.  †As published or estimated by authors.  Africa      Eastern          Mozambique  PCR/HC2  Survey  14–61  40.0  31–61  30.5  (2)          Zimbabwe  HC2  Routine screening  25–55  42.8  —  —  (3)      Northern          Morocco  PCR  Hospital case-control  18–70  21.6  35–70  21.6  (4)      Southern          South Africa (Black)  HC2  Routine screening  —  —  35–65  21.3  (5)      Western          Senegal  PCR  Hospital case-control  —  —  35–83  43.7  (6)  America      Central          Mexico  PCR  Population-based  <25–>65  14.5  35–>65  14.1  (7)          Costa Rica  PCR  Routine screening  18–94  16.0  —  —  (8)      South          Colombia  PCR  Survey  13–85  14.9  35–85  8.4  (9)          Argentina  PCR  Survey  15–>55  16.6  35–>55  13.9  (10)      Northern          Canada  PCR  Routine screening  15–49  13.3  35–49  9.0  (11)          United States  PCR  Routine screening  16–77  22.5  —  —  (12)  Europe      Eastern          Russian Federation  PCR  Routine screening  15–45  29.0  31–45  25.5  (13)      Northern          United Kingdom  PCR  Routine screening  —  —  34–70  5.9  (14)          Denmark  PCR  Survey  20–29  18.0  —  —  (15)      Southern          Spain  PCR  Population-based  14–75  3.0  35–75  2.0  (15a)          Italy  PCR  Population-based  25–70  8.6  35–70  7.4  (16)          Greece  PCR  Routine screening  20–55  36.2  —  —  (17)      Western          The Netherlands  PCR  Population-based  15–69  4.6  30–69  4.3  (18)          Germany  PCR  Routine screening  18–70  7.8  36–70  4.9  (19)          France  HC2  Population-based  15–76  15.3  31–76  12.3  (20)  Asia      Europe          Japan  PCR  Routine screening  —  —  30–78  7.0  (21)          Korea, DR  PCR  Survey  20–74  10.4  35–74  9.6  (22)          Taiwan  PCR  Routine screening  —  —  30–64  9.2  (23)          China  HC2  Survey  —  —  35–45  18.0  (24)      South Eastern          Vietnam, North  PCR  Survey  15–69  2.0  35–69  1.9  (25)          Vietnam, South  PCR  Survey  15–69  10.9  35–69  8.0  (25)          Thailand  PCR  Population-based  15–>65  6.3  35–>65  5.0  (26)      Southcentral          India (Madras)  PCR  Hospital case-control  23–76  27.7  35–76  28.5  (27)  View Large Table 2. Estimated incidence of invasive cervical cancer in the world*         Incidence (36)  Age-specific incidence, y  Region  No. of cases 1985 (34)  No. of cases 1990 (35)  No. of cases 2000 (36)  CR  ASRW  15–44  45–54  55–64  65+  *CR = crude rates per 100 000; ASRW = age-standardized rates per 100 000. World standard population. All Rates correspond to the most recent compilation of cancer incidence.  †Melanesia, Micronesia and Polynesia.  World  437 300  371 200  470 606  15.7  16.1  9.5  44.9  51.8  41.9      More developed  93 700  83 300  91 451  15.0  11.3  11.9  22.4  23.8  26.3      Less developed  343 600  287 900  379 153  15.8  18.7  9.0  53.6  65.0  53.8  Africa  51 500  52 500  67 078  17.1  27.3  11.0  71.5  100.5  95.4      Eastern  21 800  21 500  30 206  24.4  44.3  16.1  114.8  174.4  153.9      Middle  6600  5700  6947  14.4  25.1  8.5  54.0  73.3  137.4      Northern  6200  5200  10 479  12.2  16.8  6.2  49.0  68.5  45.9      Southern  6600  6500  5541  23.2  30.3  15.5  67.8  98.5  118.2      Western  10 300  13 600  13 903  12.5  20.3  9.5  57.4  70.6  60.3  America  68 000  74 800  92 136  22.0  21  15.1  55.2  57.8  55.0      Caribbean  3000  5000  6670  34.8  35.8  17.7  82.7  102.1  155.6      Central  13 700  17 700  21 596  31.7  40.3  22.5  111.7  109.9  136.1      South  35 300  36 900  49 025  28.1  30.9  16.8  85.5  90.2  101.4      United States and Canada  16 000  15 200  14 845  9.5  7.9  9.0  15.4  16.8  14.2  Europe  67 000  58 200  64 928  17.2  13  14.1  26.3  26.5  28.1      Eastern  40 100  27 500  35 482  21.9  16.8  17.8  34.5  34.9  36.6      Northern  6300  7600  6049  12.6  9.8  12.0  17.6  16.7  20.2      Southern  8700  9900  10 116  13.7  10.2  10.5  20.8  23.7  20.9      Western  11 900  13 200  13 282  14.2  10.4  11.3  20.5  19.6  25.1  Asia  249 000  183 400  245 670  13.6  14.9  7.2  44.0  52.8  39.6  Asia excluding China  170 800  158 700  212 297  18.0  21.1  10.3  63.0  75.0  53.5      Eastern  94 200  42 500  51 266  7.1  6.4  2.6  18.4  18.9  25.4          China  78 200  24 700  33 373  5.4  5.2  1.7  17.2  16.1  19.1          Japan  9400  8500  11 681  18.1  11.1  8.8  21.2  25.6  42.2          Other  6600  9300  6212  16.2  15.3  10.3  33.6  41.9  55.0      Southeastern  42 500  30 900  39 648  15.3  18.3  9.1  59.0  58.2  45.9      Southcentral  109 500  107 000  151 297  20.9  26.5  11.9  79.2  100.8  65.6      Western  2800  3000  3458  3.8  4.8  2.6  13.1  15.3  14.1  Oceania  1800  2100  2156  14.2  12.6  12.3  27.4  28.2  29.0      Pacific Islands†  500  800  1078  29.1  40.3  23.1  91.2  107.1  167.8      Australia/N. Zealand  1300  1300  1077  9.4  7.7  8.6  15.5  14.8  16.6          Incidence (36)  Age-specific incidence, y  Region  No. of cases 1985 (34)  No. of cases 1990 (35)  No. of cases 2000 (36)  CR  ASRW  15–44  45–54  55–64  65+  *CR = crude rates per 100 000; ASRW = age-standardized rates per 100 000. World standard population. All Rates correspond to the most recent compilation of cancer incidence.  †Melanesia, Micronesia and Polynesia.  World  437 300  371 200  470 606  15.7  16.1  9.5  44.9  51.8  41.9      More developed  93 700  83 300  91 451  15.0  11.3  11.9  22.4  23.8  26.3      Less developed  343 600  287 900  379 153  15.8  18.7  9.0  53.6  65.0  53.8  Africa  51 500  52 500  67 078  17.1  27.3  11.0  71.5  100.5  95.4      Eastern  21 800  21 500  30 206  24.4  44.3  16.1  114.8  174.4  153.9      Middle  6600  5700  6947  14.4  25.1  8.5  54.0  73.3  137.4      Northern  6200  5200  10 479  12.2  16.8  6.2  49.0  68.5  45.9      Southern  6600  6500  5541  23.2  30.3  15.5  67.8  98.5  118.2      Western  10 300  13 600  13 903  12.5  20.3  9.5  57.4  70.6  60.3  America  68 000  74 800  92 136  22.0  21  15.1  55.2  57.8  55.0      Caribbean  3000  5000  6670  34.8  35.8  17.7  82.7  102.1  155.6      Central  13 700  17 700  21 596  31.7  40.3  22.5  111.7  109.9  136.1      South  35 300  36 900  49 025  28.1  30.9  16.8  85.5  90.2  101.4      United States and Canada  16 000  15 200  14 845  9.5  7.9  9.0  15.4  16.8  14.2  Europe  67 000  58 200  64 928  17.2  13  14.1  26.3  26.5  28.1      Eastern  40 100  27 500  35 482  21.9  16.8  17.8  34.5  34.9  36.6      Northern  6300  7600  6049  12.6  9.8  12.0  17.6  16.7  20.2      Southern  8700  9900  10 116  13.7  10.2  10.5  20.8  23.7  20.9      Western  11 900  13 200  13 282  14.2  10.4  11.3  20.5  19.6  25.1  Asia  249 000  183 400  245 670  13.6  14.9  7.2  44.0  52.8  39.6  Asia excluding China  170 800  158 700  212 297  18.0  21.1  10.3  63.0  75.0  53.5      Eastern  94 200  42 500  51 266  7.1  6.4  2.6  18.4  18.9  25.4          China  78 200  24 700  33 373  5.4  5.2  1.7  17.2  16.1  19.1          Japan  9400  8500  11 681  18.1  11.1  8.8  21.2  25.6  42.2          Other  6600  9300  6212  16.2  15.3  10.3  33.6  41.9  55.0      Southeastern  42 500  30 900  39 648  15.3  18.3  9.1  59.0  58.2  45.9      Southcentral  109 500  107 000  151 297  20.9  26.5  11.9  79.2  100.8  65.6      Western  2800  3000  3458  3.8  4.8  2.6  13.1  15.3  14.1  Oceania  1800  2100  2156  14.2  12.6  12.3  27.4  28.2  29.0      Pacific Islands†  500  800  1078  29.1  40.3  23.1  91.2  107.1  167.8      Australia/N. Zealand  1300  1300  1077  9.4  7.7  8.6  15.5  14.8  16.6  View Large Table 3. Etiology of cervical cancer: established factors among HPV–DNA-positive women Factors  HPV  Use of exogenous hormones  Smoking  Parity  Sexual factors of the partner  Qualifiers  Persistent presence of any of at least 15 HPV–DNA types in cervical cells  Long-term use of hormonal contraception (5+ yr)  Dose response unproven
 Likely role of cancer promoter  High parity (five or more pregnancies)  (Probably related to the risk of HPV infection) 
 Partner with 6+ other sexual partners in some countries 
 Uncircumcised partner with 6+ other sexual partners  Research topics  Viral load 
 Multiple HPV types 
 Viral variants 
 HPV types and cancer histology 
 Predominance of HPV16 
 Factors and biomarkers for latency, persistency and progression  Hormonal composition 
 Hormonal replacement therapy 
 Tamoxifen 
 Duration of effect after cessation of use 
 Mechanisms of cancer promotion  Mechanisms of cancer promotion  Mechanisms of cancer promotion  Natural history of HPV infections in males  Epidemiological references from IARC studies  (1) 
 (44)  (45)  (46)  (47)  (48)  Factors  HPV  Use of exogenous hormones  Smoking  Parity  Sexual factors of the partner  Qualifiers  Persistent presence of any of at least 15 HPV–DNA types in cervical cells  Long-term use of hormonal contraception (5+ yr)  Dose response unproven
 Likely role of cancer promoter  High parity (five or more pregnancies)  (Probably related to the risk of HPV infection) 
 Partner with 6+ other sexual partners in some countries 
 Uncircumcised partner with 6+ other sexual partners  Research topics  Viral load 
 Multiple HPV types 
 Viral variants 
 HPV types and cancer histology 
 Predominance of HPV16 
 Factors and biomarkers for latency, persistency and progression  Hormonal composition 
 Hormonal replacement therapy 
 Tamoxifen 
 Duration of effect after cessation of use 
 Mechanisms of cancer promotion  Mechanisms of cancer promotion  Mechanisms of cancer promotion  Natural history of HPV infections in males  Epidemiological references from IARC studies  (1) 
 (44)  (45)  (46)  (47)  (48)  View Large Table 4. Etiology of cervical cancer: possible factors among HPV–DNA-positive women requiring further research Factors  Other STDs  Other environmental factors  Host Factors host and viral interactions  STDs = sexually transmitted diseases.  Research topics  C. Trachomatis HSV-2  Low Socio-economic status Dietary factors  Early age at infection 
 Genetic background (HLA, p53 . . .) 
 Susceptibility of the transitional zone 
 Viral integration 
 Immune response to HPV and determinants of viral clearance 
 Nonspecific cervical inflammation  Epidemiological references from IARC studies  (49)
 (50)  (51)    Factors  Other STDs  Other environmental factors  Host Factors host and viral interactions  STDs = sexually transmitted diseases.  Research topics  C. Trachomatis HSV-2  Low Socio-economic status Dietary factors  Early age at infection 
 Genetic background (HLA, p53 . . .) 
 Susceptibility of the transitional zone 
 Viral integration 
 Immune response to HPV and determinants of viral clearance 
 Nonspecific cervical inflammation  Epidemiological references from IARC studies  (49)
 (50)  (51)    View Large Fig. 1. View largeDownload slide Cumulative prevalence of the most common HPV types in cervical cancer cases by histology and in women with normal cytology (source: the International Agency for Research on Cancer multicenter control studies). Fig. 1. View largeDownload slide Cumulative prevalence of the most common HPV types in cervical cancer cases by histology and in women with normal cytology (source: the International Agency for Research on Cancer multicenter control studies). Fig. 2. View largeDownload slide Age-specific incidence rates of cervical cancer in selected countries in Latin America, Europe and in the United States by ethnic groups. Fig. 2. View largeDownload slide Age-specific incidence rates of cervical cancer in selected countries in Latin America, Europe and in the United States by ethnic groups. Fig. 3. View largeDownload slide Age specific incidence rate of CIN 3/carcinoma in situ and of invasive cervical cancer in selected populations. Ratio: Number of cases of CIN 3/in situ carcinoma of the cervix over the number of cases of invasive cancer diagnosed in the same population over the same time period across all age groups. *Data provided as a courtesy of the Finnish Cancer Registry, East Anglian Cancer Registry, and the Iceland Cancer Registry. Fig. 3. View largeDownload slide Age specific incidence rate of CIN 3/carcinoma in situ and of invasive cervical cancer in selected populations. Ratio: Number of cases of CIN 3/in situ carcinoma of the cervix over the number of cases of invasive cancer diagnosed in the same population over the same time period across all age groups. *Data provided as a courtesy of the Finnish Cancer Registry, East Anglian Cancer Registry, and the Iceland Cancer Registry. We thank Mireia Diaz for the extraction and preparation of data on cancer incidence, Mireia Diaz and Dr. Roberto Martinez-Vimbert for the literature review on HPV-DNA prevalence worldwide, and Cristina Rajo and Elisabet Luquín for secretarial and editorial support. This chapter was partially supported by a Research Agreement between the Epidemiology and Cancer Registration Unit (SERC) and the International Agency for Research on Cancer (IARC) (CRA FIS/01/04). Partial support was granted by the Fondo de Investigaciones Sanitarias of the Spanish Government (FIS 01/1237). Dr. Bosch received a Contract for Services offered by the National Cancer Institute (NCI). Dr. Martinez-Vimbert received a grant from Generalitat de Catalunya (CIRIT 2001FI00068). The authors are indebted to the Icelandic Cancer Registry, the Finnish Cancer Registry, and the East Anglian Cancer Registry, who provided data on the incidence of cervical cancer precursors. References 1 Bosch FX, Lorincz A, Munoz N, Meijer CJ, Shah KV. The causal relation between human papillomavirus and cervical cancer. J Clin Pathol  2002; 55: 244–65. Google Scholar 2 Castellsagué X, Menéndez C, Loscertales MP, Kornegay JR, dos Santos F, Gómez-Olivé FX, et al. Human papillomavirus genotypes in rural Mozambique. Lancet  2001; 358: 1429–30. Google Scholar 3 Blumenthal PD, Gaffikin L, Chirenje ZM, McGrath J, Womack S, Shah K. Adjunctive testing for cervical cancer in low resource settings with visual inspection, HPV, and the Pap smear. Int J Gynaecol Obstet  2001; 72: 47–53. Google Scholar 4 Chaouki N, Bosch FX, Muñoz N, Meijer CJLM, El-Gueddari B, El-Ghazi A, et al. The viral origin of cervical cancer in Rabat, Morocco. Int J Cancer  1998; 75: 546–54. Google Scholar 5 Wright Jr TC, Denny L, Kuhn L, Pollack A, Lorincz A. HPV DNA testing of self-collected vaginal samples compared with cytologic screening to detect cervical cancer. J Am Med Assoc  2000; 282: 81–86. Google Scholar 6 Lin P, Koutsky LA, Critchlow CW, Apple RJ, Hawes SE, Hughes JP, et al. HLA class II Dr-DQ and increased risk of cervical cancer among Senegalese women. Cancer Epidemiol Biomark Prev  2001; 10: 1037–45. Google Scholar 7 Lazcano Ponce E, Herrero R, Muñoz N, Cruz A, Shah KV, Alonso P, et al. Epidemiology on HPV infection among Mexican women with normal cervical cytology. Int J Cancer  2001; 91: 412–20. Google Scholar 8 Herrero R, Hildesheim A, Bratti C, Sherman ME, Hutchinson M, Morales J, et al. Population-based study of human papillomavirus infection and cervical neoplasia in rural Costa Rica. J Natl Cancer Inst  2000; 92: 464–74. Google Scholar 9 Molano M, Posso H, Weiderpass E, Van den Brule AJC, Ronderos M, Franceschi S, et al. Prevalence and determinants of HPV infection among Colombian women with normal cytology. Br J Cancer  2002; 87: 324–33. Google Scholar 10 Matos E, Loria D, Amestoy G, Herrera L, Prince MA, Moreno J, et al. Prevalence of human papillomavirus (HPV) infection among women in Concordia, Argentina: a population-based study. Sex Transm Dis . In press. 2003. Google Scholar 11 Sellors JW, Mahony JB, Kaczorowski J, Lytwyn A, Bangura H, Chong S, et al. Prevalence and predictors of human papillomavirus infection in women in Ontario, Canada. CMAJ  2000; 163: 503–8. Google Scholar 12 Cope JU, Hildesheim A, Schiffman MH, Manos MM, Lorincz AT, Burk RD, et al. Comparison of the hybrid capture tube test and PCR for detection of human papillomavirus DNA in cervical specimens. J Clin Microbiol  1997; 35: 2262–5. Google Scholar 13 Alexandrova YN, Lyshchov AA, Safronnikova NR, Imyanitov EN, Hanson KP. Features of HPV infection among the healthy attendants of gynecological practice in St. Petersburg, Russia. Cancer Lett  1999; 145: 43–8. Google Scholar 14 Cuzick J, Beverley E, Ho L, Terry G, Sapper H, Mielzynska I, et al. HPV testing in primary screening of older women. Br J Cancer  1999; 81: 554–58. Google Scholar 15 Kjaer SK, Chackerian B, van der Brule AJC, Svare EI, Paull G, Walboomers JM, et al. High-risk human papillomavirus is sexually transmitted: evidence from a follow-up study of virgins starting sexual activity (intercourse). Cancer Epidemiol Biomarkers Prev  2001; 10: 101–6. Google Scholar 15a de Sanjose S, Almizall R, Lloveras B, Font R, Diaz M, Muñoz N, et al. Cervical human papillomavirus infection in the female population in Barcelona, Spain. Sex Transm Dis . In press. 2003. Google Scholar 16 Ronco G, Ghisetti V, Snijders PJF, Gillio-Tos A, Segnan N, Franceschi S. Age dependence of HPV prevalence and its determinants in Turin, Italy. In: Institut Pasteur, editor. 20th International Papillomavirus Conference. Program and Abstracts book. (www.hpv2002.com), Paris: Colloquium; 2002. p. 232. Google Scholar 17 Agorastos T, Bontis J, Lambropoulos AF, Constantinidis TC, Nasioutziki M, Tagou C, et al. Epidemiology of human papillomavirus infection in Greek asymptomatic women. Eur J Cancer Prev  1995; 4: 159–67. Google Scholar 18 Jacobs MV, Walboomers JMM, Snijders PJF, Voorhorst FJ, Verheijen RHM, Fransen-Daalmeijer N, et al. Distribution of 37 mucosotropic HPV types in women with cytologically normal cervical smears: the age-related patterns for high-risk and low types. Int J Cancer  2000; 87: 221–7. Google Scholar 19 Schneider A, Hoyer H, Lotz B, Leistritza S, Kühne-Heid R, Nindl I, et al. Screening for high-grade cervical intra-epithelial neoplasia and cancer by testing for high-risk HPV, routine cytology or colposcopy. Int J Cancer  2000; 89: 529–34. Google Scholar 20 Clavel C, Masure M, Bory JP, Putaud I, Mangeonjean C, Lorenzato M, et al. Human papillomavirus testing in primary screening for the detection of high-grade cervical lesions: a study of 7932 women. Br J Cancer  2001; 84: 1616–23. Google Scholar 21 Zheng PS, Iwasaka T, Zhang ZM, Pater A, Sugimori H. Telomerase activity in Papanicolaou smear-negative exfoliated cervical cells and its association with lesions and oncogenic human papillomaviruses. Gynecol Oncol  2000; 77: 394–8. Google Scholar 22 Shin HR, Lee DH, Herrero R, Smith JS, Vaccarella S, Hong SH, et al. Prevalence of human papillomavirus infection in women in Busan, South Korea. Int J Cancer  2003; 103: 413–21. Google Scholar 23 Liaw K, Hsing AW, Chen ChJ, Schiffman MH, Zhang TY, Hsieh CY, et al. Human papilloma virus and cervical neoplasia: a case-control study in Taiwan. Int J Cancer  1995; 62: 565–71. Google Scholar 24 Belinson J, Qiao YL, Pretorius R, Zhang WH, Elson P, Li L, et al. Shanxi Province Cervical Cancer Screening Study: a cross-sectional comparative trial of multiple techniques to detect cervical neoplasia. Gynaecol Oncol  2001; 83: 439–44. Google Scholar 25 Anh PT, Hieu NT, Herrero R, Vaccarella S, Smith JS, Thuy NT, et al. Human papillomavirus infection among women in South and North Vietnam. Int J Cancer  2003; 104: 213–20. Google Scholar 26 Sukvirach S, Smith JS, Tunsakul S, Muñoz N, Kesararat W, Opasatian O, et al. Population-based human papillomavirus prevalence in Lampang and Songkla, Thailand. J Infect Dis . In press. 2003. Google Scholar 27 Rajkumar T, Franceschi S, Vaccarella S, Gajalakshmi V, Sharmila A, Snijders PJF, et al. The role of paan chewing and dietary habits in cervical carcinoma in Chennai, India. Br J Cancer . In press. 2003. Google Scholar 28 Bosch FX, Manos M, Muñoz N, Sherman M, Jansen A, Peto J, et al. Prevalence of human papillomavirus in cervical cancer: a worldwide perspective. J Natl Cancer Inst  1995; 87: 796–802. Google Scholar 29 Walboomers JMM, Jacobs MV, Manos MM, Bosch FX, Kummer JA, Shah KV, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol  1999; 189: 12–19. Google Scholar 30 Clifford GM, Smith JS, Plummer M, Muñoz N, Franceschi S. Human papillomavirus types in invasive cervical cancer worldwide: a meta-analysis. Br J Cancer  2003; 88: 63–73. Google Scholar 31 Giuliano AR, Papenfuss M, Abrahamsen M, Denman C, de Zapien JG, Henze JL, et al. Human papillomavirus infection at the United States-Mexico border: implications for cervical cancer prevention and control. Cancer Epidemiol Biomarkers Prev  2001; 10: 1129–36. Google Scholar 32 Chan PK, Li WH, Chan MY, Ma WL, Cheung JL, Cheng AF. High prevalence of human papillomavirus type 58 in Chinese women with cervical cancer and precancerous lesions. J Med Virol  1999; 59: 232–38. Google Scholar 33 Huang S, Afonina I, Miller BA, Beckmann AM. Human papillomavirus types 52 and 58 are prevalent in cervical cancers from Chinese women. Int J Cancer  1997; 70: 408–11. Google Scholar 34 Parkin DM, Pisani P, Ferlay J. Estimates of the worldwide incidence of eighteen major cancers in 1985. Int J Cancer  1993; 54: 594–606. Google Scholar 35 Parkin DM, Pisani P, Ferlay J. Estimates of the worldwide incidence of 25 majors cancers in 1990. Int J Cancer  1999; 80: 827–41. Google Scholar 36 Ferlay J, Bray F., Pisani P., Parkin D.M., editors. Globocan 2000: Cancer incidence, mortality and prevalence worldwide, version 1.0. IARC CancerBase No. 5. Lyon (France): International Agency for Research on Cancer (IARC Press); 2001. Google Scholar 37 Li H, Jin S, Xu H, Thomas DB. The decline in the mortality rates of cervical cancer and a plausible explanation in Shandong, China. Int J Epidemiol  2000; 29: 398–404. Google Scholar 38 Vizcaino AP, Moreno V, Bosch FX, Muñoz N, Barros-Dios XM, Parkin DM. International trends in the incidence of cervical cancer: I. Adenocarcinoma and adenosquamous cell carcinomas. Int J Cancer  1998; 75: 536–45. Google Scholar 39 Vizcaino AP, Moreno V, Bosch FX, Muñoz N, Barros-Dios XM, Borras J, et al. International trends in incidence of cervical cancer: II. Squamous-cell carcinoma. Int J Cancer  2000; 86: 429–35. Google Scholar 40 Sasieni P, Adams J. Changing rates of adenocarcinoma and adenosquamous carcinoma of the cervix in England. Lancet  2001; 357: 1490–1493. Google Scholar 41 Parkin DM, Whelan SL, Ferlay J, Raymond L, Young J, editors. Cancer incidence in five continents, vol. VII. Lyon (France): International Agency for Research on Cancer (IARC Press); 1997. Google Scholar 42 Drain PK, Holmes KK, Hughes JP, Koutsky LA. Determinants of cervical cancer rates in developing countries. Int J Cancer  2002; 100: 199–205. Google Scholar 43 Shah KV, Howley PM. Papillomaviruses. In: Fields BN, Knipe DM, Howley PM, et al., editors. Fields virology, Philadelphia: Lippincott-Raven; 1996. p. 2077–109. Google Scholar 44 Muñoz N, Bosch FX, de Sanjosé S, Herrero R, Castellsagué X, Shah KV, et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. New Engl J Med  2003; 348: 518–27. Google Scholar 45 Moreno V, Bosch FX, Munoz N, Meijer CJ, Shah KV, Walboomers JM, et al. Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control study. Lancet  2002; 359: 1085–92. Google Scholar 46 IARC Press. IARC Monographs on the evaluation of carcinogenic risks to humans, Volume 83. Tobacco smoke and involuntary smoking. Lyon, (France): IARC Press. In press. 2003. Google Scholar 47 Muñoz N, Franceschi S, Bosetti C, Moreno V, Herrero R, Smith JS, et al. Role of parity and human papillomavirus in cervical cancer: the IARC multicentric case-control study. Lancet  2002; 359: 1093–102. Google Scholar 48 Castellsague X, Bosch FX, Munoz N, Meijer CJ, Shah KV, de Sanjose S, et al. Male circumcision, penile human papillomavirus infection, and cervical cancer in female partners. N Engl J Med  2002; 346: 1105–12. Google Scholar 49 Smith JS, Muñoz N, Herrero R, Eluf-Neto J, Ngelangel C, Franceschi S, et al. Evidence for Chlamydia trachomatis as a human papillomavirus cofactor in the etiology of invasive cervical cancer in Brazil and the Philippines. J Infect Dis  2002; 185: 324–31. Google Scholar 50 Smith JS, Herrero R, Bosetti C, Muñoz N, Bosch FX, Eluf-Neto J, et al. Herpes simplex virus-2 as a human papillomavirus cofactor in the etiology of invasive cervical cancer. J Natl Cancer Inst  2002; 94: 1604–13. Google Scholar 51 de Sanjosé S, Bosch FX, Muñoz N, Shah K. Social differences in sexual behaviour and cervical cancer. In: Kogevinas M, Pearce N, Susser M, Boffetta P, editors. Social inequalities and cancer. Lyon (France): International Agency for Research on Cancer (IARC Press); 1997. p. 309–18. Google Scholar 52 Castle PE, Schiffman M, Gravitt PE, Kendall H, Fishman S, Dong H, et al. Comparisons of HPV DNA detection by MY09/11 PCR methods. J Med Virol  2002; 68: 417–23. Google Scholar 53 Liaw KL, Glass AG, Manos MM, Greer CE, Scott DR, Sherman M, et al. Detection of human papillomavirus DNA in cytologically normal women and subsequent cervical squamous intraepithelial lesions. J Natl Cancer Inst  1999; 91: 954–60. Google Scholar 54 Woodman CB, Collins S, Winter H, Bailey A, Ellis J, Prior P, et al. Natural history of cervical human papillomavirus infection in young women: a longitudinal cohort study. Lancet  2001; 357: 1831–36. Google Scholar 55 Palefsky JM, Minkoff H, Kalish LA, Levine A, Sacks HS, García P, et al. Cervicovaginal human papillomavirus infection in human immunodeficiency virus-1 (HIV)-positive and high-risk HIV-negative women. J Natl Cancer Inst  1999; 91: 226–36. Google Scholar 56 Xi LF, Carter JJ, Galloway DA, Kuypers J, Hughes JP, Lee SK, et al. Acquisition and natural history of human papillomavirus type 16 variant infection among a cohort of female university students. Cancer Epidemiol Biomark Prev  2002; 11: 343–51. Google Scholar 57 Sherman ME, Schiffman M, Cox JT. Effects of age and human papilloma viral load on colposcopy triage: data from the randomized Atypical Squamous Cells of Undetermined Significance/Low-Grade Squamous Intraepithelial Lesion Triage Study (ALTS). J Natl Cancer Inst  2002; 94: 102–7. Google Scholar 58 Josefsson AM, Magnusson PK, Ylitalo N, Sorensen P, Qwarforth-Tubbin P, Andersen PK, et al. Viral load of human papilloma virus 16 as a determinant for development of cervical carcinoma in situ: a nested case-control study. Lancet  2000; 355: 2189–93. Google Scholar 59 Ylitalo N, Sorensen P, Josefsson AM, Magnusson PKE, Andersen PK, Pontén J, et al. Consistent high viral load of human papillomavirus 16 and risk of cervical carcinoma in situ: a nested case-control study. Lancet  2000; 355: 2194–98. Google Scholar 60 Storey A, Thomas M, Kalita A, Harwood C, Gardiol D, Mantovani F, et al. Role of a p53 polymorphism in the development of human papillomavirus-associated cancer. Nature  1998; 393: 229–34. Google Scholar 61 Makni H, Franco EL, Kaiano J, Villa LL, Labrecque S, Dudley R, et al. P53 polymorphism in codon 72 and risk of human papillomavirus-induced cervical cancer: effect of inter-laboratory variation. Int J Cancer  2000; 87: 528–33. Google Scholar © Oxford University Press http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png JNCI Monographs Oxford University Press

Chapter 1: Human Papillomavirus and Cervical Cancer—Burden and Assessment of Causality

Bosch, F. Xavier; de Sanjosé, Silvia
JNCI Monographs , Volume 2003 (31) – Jun 1, 2003
Download PDF
11 pages

Loading next page...
 
/lp/oxford-university-press/chapter-1-human-papillomavirus-and-cervical-cancer-burden-and-ru5nqZX8jr

References (64)

Publisher
Oxford University Press
Copyright
© Oxford University Press
ISSN
1052-6773
eISSN
1745-6614
DOI
10.1093/oxfordjournals.jncimonographs.a003479
Publisher site
See Article on Publisher Site

Abstract

Abstract Cervical cancer remains the second most common cancer in women worldwide and the most frequent in developing countries. Pre-neoplasic cervical lesions represent an additional burden in countries where screening is widespread. The human papillomavirus (HPV) prevalence and type distribution in normal smears and in cancer specimens are being described and show relatively small international variation. State-of-the-art detection techniques have unequivocally shown that HPV-DNA can be detected in 95% to 100% of adequate specimens of cervical cancer, supporting the claim that HPV is the necessary cause. The odds ratios for cervical cancer related to a cross sectional detection of HPV-DNA range from 50 to several hundred in all studies. The risk for any of 15 high-risk types is not statistically different from the risk reported for HPV16. The estimates of the attributable fraction range from 90% to 98%. Additional work should be done in providing information on incidence of cervical cancer and on HPV infection in areas where the disease is common. Theoretical work including modeling of the incidence could be of potential use in the evaluation of the existing and novel preventive strategies. Research is currently being conducted on the mechanisms of HPV carcinogenesis. These include the determinants of the systemic and cellular immune response to the viral infection, the interaction between the host and the virus and the relevance of the different strains and variants of the HPV viral types. Technology developments in this area suitable for epidemiological studies are needed. In most developing countries, cervical cancer is the leading female malignancy and a common cause of death among middle-aged women. In developed populations with good screening options, invasive cervical cancer is a relatively rare condition, whereas its precursors and the equivocal cytological results represent a major health burden. Research at the population level has provided an exhaustive body of evidence on the viral etiology of cervical cancer, and on the human papillomavirus (HPV) types involved. Most of the progress made by epidemiological studies has been based on DNA technology to characterize the presence of the human papillomavirus DNA (HPV-DNA) in cervical specimens. Investigations have demonstrated that a range from 2% to more than 20% of the world’s female population have detectable HPV-DNA in their cervix at any time. Further, it has been established that, under certain conditions, these infections persist and are able to induce CIN2/3 and cervical cancer. Currently, the viral DNA can be detected in virtually all cases of invasive cancer. High parity, smoking, and long-term use of oral contraceptives have been established as risk cofactors for cervical cancer among women with persistent infections with HPV. Other sexually transmitted diseases such as Chlamydia trachomatis, herpes simplex virus type 2, and human immunodeficiency virus as well as some poorly known dietary factors are likely intervening factors. Continued research should focus on producing better estimates of the burden of disease linked to papillomaviruses. In the future, cancer registries should play an important role by monitoring cervical cancer incidence in populations vaccinated against HPV. Additional investigations should further explore the natural history and the cofactors that modulate the risk of cancer among HPV infected women. Burden of HPV Infection, Cervical Neoplasia, and Cervical Cancer Definite progress in understanding the etiology of cervical cancer has been achieved, and some types of HPV have been established as the central cause of cervical cancer worldwide (1). Further, technology is available to detect the presence of viral DNA in large volumes of cervical specimens, and a substantial number of epidemiological investigations are providing estimates on the prevalence and distribution of the HPV infections in human populations. Serological measurements of the type-specific antibody levels against HPV capsid antigens combined with DNA studies have provided evidence that most of the sexually active women (up to 70% in some populations in the United States) have been exposed to HPV at any point in time. Seroconversion following HPV exposure is, however, not universal; and with current detection methods, a range between 40 and 50% of HPV-DNA 16 positive cervical cancer cases do not show HPV16 antibodies. The natural history and the determinants of the immune response to HPV are still poorly understood, and antibody measurements have had limited use, thus far, in epidemiological studies. Prevalence of HPV-DNA in Cervical Specimens from the General Population Table 1(2–27) shows a selection of studies that investigated the prevalence of HPV-DNA in women from the general population in different regions in the world. Studies are included in the table for the following reasons: 1) they represent in as much as possible the underlying populations, although not necessarily of the country or the corresponding world’s region; 2) they used polymerase chain reaction (PCR) or Hybrid Capture 2 (HC2) technology to assess the presence of the viral DNA; and 3) they include a wide age range, a study size of at least 200 individuals, and had a well described methodological section in their published account. The table shows two prevalence estimates, one for all women included in the study, and one for women ages 30 years and above. The latter is generally considered an approximation of the prevalence of persistent carriers. This interpretation recognizes that most HPV infections are acquired soon after sexual initiation and are also cleared at an early age. Women who are HPV-DNA positive at ages 30 and above (allowing for certain intercountry variation) may represent failure of viral clearance and select the group of women at a relative higher risk of cervical cancer. The results displayed in Table 1 should be viewed as an approximation to the burden of the viral infection estimated by the HPV-DNA detection rate in cervical scrapes at a point in time. Limitations to current data include the variability introduced by the selection of the population, the sampling methods, and by the variability of the techniques used for HPV detection, specifically when nonstandardized PCR methods are used locally. In countries with intensive screening among young women, aggressive treatments of early or ambiguous cytology lesions (most of which are likely to be HPV-related) may have resulted in a reduction of the estimated HPV-DNA prevalence in more advanced age groups. In a few epidemiological studies, the exact same methods in the field and in centralized laboratories have been used to investigate populations of contrasting incidence of cervical cancer (i.e., Colombia and Spain, North and South Vietnam, International Agency for Research on Cancer (IARC’s) multicentric case control studies, different surveys in the United Kingdom). These studies have consistently shown strong geographical correlations between HPV-DNA prevalence and cervical cancer incidence. HPV Prevalence in Cervical Cancer Cases and HPV-type Distribution in Cases and Controls State-of-the art amplification techniques used in clinical and epidemiological studies have shown unequivocally that in adequate specimens of cervical cancer, HPV-DNA can be detected in 90% to 100% of the cases in any population where investigations have taken place. This compares with a prevalence of 5% to 20% from cervical specimens of women without cervical cancer, from the same populations (i.e., prevalence surveys or control groups in case control studies). Detailed investigations of the few cervical cancer specimens that appear as HPV-DNA negatives on routine testing strongly suggest that these are largely false negatives (28,29). As a consequence, the claim has been made that this is the first necessary cause of a human cancer ever identified, and it provides a strong rationale for the use of HPV tests in screening programs and for the development of HPV vaccines. Of the more than 35 HPV types found in the genital tract, HPV16 accounts for 50% to 60% of the cervical cancer cases in most countries, followed by HPV18 (10%–12%) and HPVs 31 and 45 (4%–5% each). Fig. 1 shows the cumulative distribution of the most common types in close to 5000 specimens included in the IARC’s multicenter cervical cancer studies. Among cases, HPVs 16, 18, 45, 31, and 33 accounted for 80% of the type distribution in squamous cell carcinomas and HPVs 16, 18, 45, 59, and 33 accounted for 94% of the type distribution in adenocarcinomas. To complete the additional 6% to 20%, the required number of HPV types increases to 20–23 types, each one of which would contribute between 3% and 0.1% of the types. These results should be viewed as indicative, because some of the testing systems used may be relatively less sensitive to the rarer types, notably in the presence of multiple infections in the same specimen. In most studies, HPV18 predominates in adenocarcinomas in absolute or in relative terms, however, the reasons for such specificity are still unknown. In the same studies, HPVs 16, 18, 45, 31, and 33 accounted for 46% of the prevalence observed in HPV-DNA positive women with normal cervical smears. These distributions are generally consistent worldwide (28,30), although several reports using highly sensitive HPV-DNA detection and typing systems are finding additional geographical variability (2,31–33), and for several regions in the world the information is either lacking or rather scarce. Discussion points/recommendations are as follows: 1) It would be important to have updates of the international literature on the prevalence and type distribution of HPV-DNA in the general population and in cervical cancer cases. These estimates should consider the various parameters that affect detection, notably the testing system and the biological specimens used. Such reviews may have implications for HPV vaccine composition. 2) It would be desirable to conduct additional surveys designed to cover the gaps in the maps, notably by including selected areas in Africa and Asia and in populations heavily exposed to the human immunodeficiency virus (HIV), malnutrition, and malaria. Assessment of the prevalence of viral variants of the most common types remains to be done. Methodological exercises have to be conducted to determine the proper HPV-DNA typing methods to be used, particularly if multiple vital types are common. 3) It is still worth replicating the findings of investigations on putative HPV-negative cervical cancer cases using, for example, microdissection technology and in situ hybridization methods to confirm that the 5–15% HPV-negative cases that are currently being reported from field or clinical studies are generally false-negatives. Incidence and Mortality of Cervical Cancer Table 2 shows estimates of the incidence of invasive cervical cancer as extracted from systematic sources, and compiled by IARC for the years 1985, 1990, and 2000 (34–36). The data included in the table have been calculated from incidence rates, mortality vital statistics, and census records. For some under-investigated populations, other ancillary sources and country-to-country extrapolations are occasionally used. Thus, the reliability of the figures in the table may differ by region. Table 2 shows the impact of the adjustment methods on the perception of risk by geographical regions. For example, crude rates are similar when the world is broadly divided into developing and developed regions; however, age-adjusted rates using the age structure of an artificial world standard population tend to decrease the estimates for developed (older) populations and to increase the estimates for less developed (younger) populations. Comparison of columns 2 to 4 (Table 2) shows one of the limitations of current estimates. When estimates for 1985 and 2000 are related, there appears to be a slight reduction in the total number of cases in developed countries and an increase in developing regions, notably in Africa, The Caribbean, and Central and South America. Estimates for 1990 are lower, strongly driven by an inconsistent reduction in the estimates for the different regions in Asia. Some authors have also reported substantial reductions in the mortality from cervical cancer in China in the interval 1970–1994 (37). However, artifacts due to differences in the diagnostic and registration methods cannot be excluded, notably when a few incidence registries are used to produce estimates for large populations. From the same source, the number of cervical cancer deaths in the world has been estimated as 233 400, which is close to half of the incident cases. The 5-year prevalence of cervical cancer has been estimated at 1.4 million cases (36). The estimated total number of cases of cervical cancer as presented in Table 2 is being largely accepted as a background reference. It is also often used for the promotion of preventive interventions in both developed and developing regions. However, to fully estimate the impact of cervical cancer, the limitations of current estimates have to be recognized and it is important to reach a comprehensive interpretation of the different parameters reflecting disease burden such as the four (number of cases, crude rates, age adjusted rates, and age-specific rates) presented in Table 2. In broad terms, it should be recognized that cervical cancer is a strong determinant of women’s cancer incidence and mortality in developing countries and that current figures may still underestimate its impact due to under diagnosis and under registration. Developed countries that have introduced massive screening have successfully shifted the focus of attention toward pre-invasive neoplasm. However, in large developed regions with unevenly implemented screening programs and an aging population, like in Europe, the total number of cases of cervical cancer diagnosed (some 65 000 cases) is still at a similar level as the one estimated for Africa (some 67 000 cases) and not far from the estimates for Central and South America combined (some 70 000 cases). Discussion points/recommendations are as follows: Cancer registries should continue to monitor the incidence of cervical cancer and to allow the evaluation of preventive strategies implemented in their populations. Updated estimates on the burden of HPV related cancers may have to expand to include other cancer sites for which an etiological role of HPV has been demonstrated (see also chapters 7, 8, and 9). Time Trends in Cervical Cancer Incidence by Histology Individual cancer registries provide the information required to estimate time trends in incidence in their populations. In countries with organized screening, decreases in rates for cervical cancer have been reported consistently, whereas increases in the incidence of cervical adenocarcinomas have been reported in several populations (38–40). Cervical adenocarcinomas currently represent 10% to 12% of cervical cancer and are an increasing concern of cytology-based screening programs. It is likely that these upward trends are the results of major changes in sexual behavior initiated in the 1950s and 1960s in their populations and to the failures of screening programs in the subsequent decades. It is known that cytology-based screening has a relatively lower sensitivity for the early glandular lesions than for the squamous-cell cancer precursors. Alternative explanations could consider the possibility that the underlying prevalence of the HPV types more related to cervical adenocarcinoma is changing over time (i.e., that the trends in prevalence of HPV18 are increasing). In addition, changes in the exposure of the populations to relevant cofactors should be considered. Discussion point/recommendation follows: It would be of interest to verify if there are any underlying time trends in the type distribution of HPV in cervical cancer. Studies of decades-old archival cervical cells or tissues are warranted. Age-Specific Incidence of Cervical Cancer In most developed countries, cervical cancer incidence shows a unique profile across age groups in that rates increase sharply after ages 25–30 years to reach a plateau after ages 45–50 years. With the availability of results from cancer registries in developing countries, it has been possible to recognize that, in the absence of screening, cervical cancer tends to follow a linear relationship with age, which is the pattern of the majority of the epithelial tumors. The shape of the age-specific incidence curve seems to be highly dependent on the screening efforts. Fig. 2 A shows an example from the United Kingdom and Brazil (41). Both countries show similar rates of invasive cervical cancer to age 30 years, suggesting similar levels of exposure to HPV, but rapidly separate apart in more advanced age groups, probably as a consequence of the organized screening program in the United Kingdom. The arithmetic scale has been chosen on this occasion for better visualization of differences. Figure 2 B(41) shows significant differences in the age-specific incidence rates in northern America across ethnic groups. Data on white populations from the Surveillance, Epidemiology and End Results Program (SEER1) and Canadian registries show that incidence rises steeply between ages 20 and 35 and stays stable in the middle-aged groups, with some slight increase after age 55 years. In contrast, the incidence increases steadily throughout lifetime for Blacks, Hispanic, and Chinese (with an unexplained and short-lasting dip between age groups 65–75 years), reaching two to three times the incidence among whites in the most advanced age groups. Migrant populations from high- to low-risk countries may have an important impact on the incidence of cervical cancer, notably if they remain outside of the screening system. 1 Editor’s note: SEER is a set of geographically defined, population-based, central cancer registries in the United States, operated by local nonprofit organizations under contract to the National Cancer Institute (NCI). Registry data are submitted electronically without personal identifiers to the NCI on a biannual basis, and the NCI makes the data available to the public for scientific research. In some countries, a second mode in the incidence of cervical cancer among women in the age groups over 55–60 years is observed. Fig. 2 C shows these distributions for some selected cancer registries in Europe. The interpretation of the second mode remains uncertain. Decrease in screening coverage or lower sensitivity of cytology in the old age groups are plausible explanations but immunosuppression or acquisition of new HPV infections during middle age cannot be ruled out. In the age groups 35 years and above, the presence of HPV-DNA in cervical cells probably reflects a combination of long-term persistent infections acquired in the younger age groups and newly acquired late infections. New infections should be largely related to changes in sexual behavior of both sexes. It can be hypothesized that, after a certain latency period, the presence of a recently acquired viral DNA in the middle-aged group, with or without the concurrent effect of cofactors, is capable of generating a second increase in the incidence of CIN3 and of invasive cancer (see also a discussion on the “Incidence of CIN3” section below). The predicted height of the second mode is modulated (attenuated) by both screening in these age groups and competing causes of death. Discussion points/recommendations are as follows: 1) Cohort studies should continue to provide data on sexual behavior during middle age and above age groups, notably among males. 2) Analyses of cervical cancer incidence by birth cohorts in different populations should be encouraged. These could relate current incidence of cervical cancer to past population-adopted changes in sexual behavior and to major migrant changes. Incidence of CIN3 There is limited information on the incidence of CIN3 lesions. However, in terms of burden of cervical pathology and the requirements imposed to the health systems in developed countries, it is likely that CIN3 is one of the key diagnoses to consider. Figure 3 shows estimates of the incidence of carcinoma in situ/CIN3 and of invasive cancer as provided by some registries. The figure shows very high rates of CIN3 in the young age groups (i.e., >400 per 100 000 in the 20 to 30 years old in the East Anglia Cancer Registry in the United Kingdom). Occasionally, a second mode in the 50 years and above age groups is observed. It is still uncertain which proportion of these CIN3 cases are deemed to progress to cervical cancer as compared with representing transient morphological changes. The population-based ratio of CIN3-to-invasive cancer seems a parameter of relevance that may provide indirect estimates of the amount of over diagnosis and overtreatment of pre-invasive lesions brought about by screening programs. For example, Fig. 3 A shows that in Finland, a country with a long-term, centralized, and well-established screening program, there are 1.6 cases of CIN3/in situ carcinoma per case of invasive cancer diagnosed. In contrast, the equivalent ratios for East Anglia and Iceland are 13.0 and 16.2, respectively (Fig. 3 B and C). It is uncertain if, in the absence of screening, all these CIN3/in situ cases would have progressed to invasive cancer. These observations may also reflect the variability in diagnosis and treatment of pre-invasive disease across populations. Discussion points/recommendations are as follows: It would be of interest to stimulate additional work on the incidence of pre-invasive lesions, including additional collection of data on the incidence of CIN3/carcinoma in situ and analysis of the available data from cancer registries. These analyses could relate incidence to the age-specific prevalence of HPV-DNA and to the corresponding incidence of invasive cancer. Analysis of the ratios of CIN3/cervical cancer considering other clinical parameters (i.e., standardization of the CIN3 diagnosis and pathology results from conization specimens) may allow estimates of the fraction of the CIN3 cases that are true precursors of cervical cancer in the context of specific screening practices. Newly organized screening programs or screening trials in developing populations (i.e., for HPV testing or direct visual inspection methods) should be able to estimate ratios of CIN3 to invasive cervical cancer in populations in the absence of preventive interventions. Analysis of the age at occurrence of CIN3 and of the ages at sexual initiation may provide new insights into the latency period for HPV and cancer. Modeling Geographic Variation in Cervical Cancer Incidence as a Basis for Classifying Populations in Relation to Their Risk of Cervical Cancer Maps of cervical cancer incidence based on age-adjusted rates tend to indicate that countries at high-risk cluster in sub-Saharan Africa, Latin America, and some parts of Asia. Identification of key environmental factors that may modify the risk of HPV exposure and the incidence of cervical cancer offers new and attractive opportunities to generate models of cervical cancer incidence. These estimates would allow a fresh look at the classification of populations in relation to their risk independently of the strong impact of screening. At a population level, high incidence rates of cervical cancer are observed and can be predicted for populations with high HPV-DNA background prevalence, paired with (and consequent to) an early age at sexual initiation, high number of partners of both men and women, and an important frequency of sexual contacts with prostitutes. Typically, these populations also have high parity and high frequency of exposure to other risk factors (e.g., oral contraceptives, smoking, other sexually transmitted diseases, poor hygiene, and cervical inflammation). Over and above their exposure levels, high-risk populations are dramatically underserved by screening programs or practices. On the reverse side, low-risk populations show either low levels of exposure to these factors and/or belong to communities with centralized and quality-controlled screening activity. As expected, countries and populations include unequal combinations of these risk and protection factors. Developed countries share many high-risk traits, largely compensated by important screening efforts and decreasing parity. The ratio of incident cases of CIN3/invasive cervical cancer tends to be high (i.e., 10/1). Examples of such countries could be the United States or some of the Nordic countries in Europe. Developing countries under strict religious (behavioral) influence show some low-risk traits (regulated sexual behavior, low prevalence of exposure to cofactors) and some high-risk traits such as high parity and absence of screening programs. These countries tend to show low incidence rates of cervical cancer, although the number of registries in these populations is also limited. Examples of this model would be the Muslim countries in the Middle East or Asia. Other developing countries without strict religious (behavioral) regulations accumulate high-risk traits not compensated by screening efforts. These situations result in the highest incidence rates of invasive disease recorded. Examples of this could be some populations in Central and Latin America. The combination of these factors into a model of disease incidence could be used to predict incidence of cervical cancer by birth cohorts when properly placed into population-specific time scales. This information would provide the rationale for the continuous support and development of screening programs. The models could also be used to estimate indirectly the efficiency of screening programs in developed countries, by comparing the predicted and the observed incidence rates. An ecological project along these lines has recently reported a strong correlation between predominant religion and cervical cancer in developing countries (42). Discussion points/recommendations are as follows: The acquisition of novel data on HPV prevalence and its determinants in a variety of regions covered by cancer registries encourages systematic analysis leading to the construction of statistical models of cervical cancer incidence. Migrant studies and HPV variant distributions could be of value. Causality Assessment: HPV and Other Risk Factors for Cervical Cancer The nature of the association between HPV and cervical cancer has been exhaustively investigated and since the early 1990s, and all academic reviews have consistently concluded that the evidence fulfills most of the established criteria of causality (1). The mechanisms by which HPV induces cancer in humans and the molecular genetics of the process are being intensively investigated and excellent reviews are available (43). To evaluate the role of additional factors once it was recognized that the role of HPV-DNA was a necessary factor it became a standard procedure to restrict the analysis to women who were HPV-DNA-positive (thus at risk of ever developing cervical cancer) in addition to the standard multivariate methods. Table 3(1,44–48) summarizes the established risk factors of cervical cancer in women who are carriers of HPV-DNA, and Table 4(49–51) presents the factors that are significantly related to cervical cancer in HPV-DNA positive women, but for which conclusive evidence is less established. As before, most of the conclusions and references quoted are results of the IARC’s multicenter study, which are consistent with the updated literature. Current Risk Estimates for HPV-DNA and HPV Type-Specific Estimates Current estimates of the odds ratio for HPV-DNA and cervical cancer are among the highest observed in human cancer. The final results of the IARC multicenter study reports an estimate of invasive cancer for HPV-DNA of OR = 158 (95%CI = 113.4 to 220.6) (44). Corresponding risk estimates for HPV16 was OR = 434.5 (95% CI = 278.2 to 678.7) and for HPV18 OR = 248.1 (95%CI = 138.1 to 445.8). The size of the project has allowed the generation of estimates of the associations between individual HPV types and cervical cancer. According to these results, any of at least 15 high-risk types, namely HPVs 16, 18, 45, 31, 52, 33, 58, 35, 59, 51, 56, 39, 68, 73, and 82 (IS39 / W13B, MM4), convey a statistically similar risk of developing cervical cancer. HPVs 26, 53, and 66 probably are also high-risk types; however, the limited number of observations with these viral types precluded a firmer conclusion. Nonsignificant differences between point estimates may reflect the overwhelming predominance of some of the types (i.e., HPVs 16 and 18) over other types. This IARC study also confirmed the strong correspondence between the epidemiological risk classification of HPV types and the phylogenetic classification which is based on comparisons of DNA sequences across types (44). Relative risks generated by cohort studies with limited HPV observations and short-term follow-up tend to be smaller than odds ratios generated by case–control studies. These studies are informative of the putative pre-invasive stages (low-grade squamous intra-epithelial lesions [LSIL] and high-grade squamous intra-epithelial lesions [HSIL]) but are difficult to interpret in terms of causality of cervical cancer. This is due to censoring of the women at highest risk because of treatment. Moreover, if in fact a fraction of the CIN3 cases (and most importantly if HSIL is the diagnostic endpoint) are not true cancer precursors, misclassification of the diagnosis may partly explain the lower risk estimates in cohort studies. Long-term cohorts that can reliably establish persistency of the same viral types have clearly established the time sequence between HPV exposure and cervical neoplasia and tend to report risk estimates at the level of the case–control studies referred above. It is plausible that novel DNA detection methods result in higher prevalence of HPV-DNA in women with normal cervical cytology. Some early work in Costa Rica using a different polymerases in the PCR reaction has resulted in an increased estimate of the HPV-DNA prevalence in controls (52). If current underdetection is a general issue, future studies would generate lower risk estimates; however, it is unlikely that any significant modification of the attributable fraction would occur as a result because of the direct relevance of the prevalence of HPV-DNA in cases in the calculations of the attributable fraction. In large prevalence surveys of HPV types in invasive cancer “low-risk HPV types” (e.g., HPV 6 or 11) are anecdotally found as the only type (44). Some cohort studies have likewise described that HPV 6 or 11 are predictors of both LSIL and HSIL (53–54). Experimental studies have occasionally shown that the E6 and E7 genes of low-risk types are capable of interfering with p53 and Rb at a low efficiency level [reviewed in (43)]. These results, although of little clinical relevance, are intriguing in that they indicate that, under certain circumstances, even low-risk types may be able to induce neoplastic changes and cancer. The proportion of multiple types in a given cervical cancer specimen varies across studies and particularly in relation to the HPV detection method used. Populations at high-risk of cervical cancer and populations with high rates of HIV tend to show higher proportions of multiple types as compared wtih populations not belonging to these risk groups (55). In the IARC multicenter study there was no statistically significant increased risk for women with multiple-type infections (OR = 115.6; 95% CI = 69.3 to 192.8) over women positive for only one HPV type (OR = 172.5; 95% CI = 122.2 to 243.6) (44). The finding is consistent in other settings (8), giving support to the notion that each HPV infection acts independently. However, some underdetection of multiple types may have occurred as a consequence of the HPV detection systems used. Research Issues on HPV and Cervical Cancer: Viral Factors Viral factors under investigation include the study of HPV variants, the measurement of HPV viral load, and viral integration. Limited information is available on HPV16 variants, suggesting an increased risk for some of the non-European prototypes (56). Viral load has been proposed as a useful marker in cohort studies with conflicting results (8,57–59). Part of the discrepancies can be related to the endpoints used in cohort studies (usually lower or equal to HSIL). Technology to characterize viral load has not been fully validated. Viral integration into cellular DNA was proposed as a marker of progression to cervical cancer. Integration is rarely seen in pre-invasive disease and invasive cancer shows either episomal or integrated viral DNA. It is still uncertain if integration in stages of HSIL predicts progression to cervical cancer. Research Issues on HPV and Cervical Cancer: Host Factors The host response and, possibly, genetically determined susceptibility remain the less understood of the key variables in the carcinogenic process. For example, experimental studies reporting an increase in risk among carriers of certain polymorphisms suggested that there is a susceptibility factor to HPV in p53 (60). However, these results were only occasionally confirmed by epidemiological data (61). Little has been done in trying to describe familial aggregation of cervical cancer, notably to disentangle the effects of an inherited trait from the effects of shared environment, behavior, and attitudes. One such attempt attributed one-third of the variance in cervical cancer incidence to shared genetic makeup. Discussion points/recommendations are as follows: It is uncertain if we need additional case control studies to investigate the basic HPV association with cervical cancer. New studies should focus on the exploration of additional associations with putative environmental cofactors (e.g., dietary factors) among HPV-DNA positive women, novel viral markers, and markers of viral and host interactions. Epidemiologists would benefit from storing biological materials from studies to be able to test novel hypothesis in well-characterized study populations. Amongst others, it would be desirable to see studies that would include the following: 1) Research into mechanisms of HPV carcinogenesis. Explore in greater detail the occasions in which low-risk HPV types are capable of inducing neoplastic transformation. 2) Consider issues of specificity of viral types (and of variants) to induce either squamous cell or adenocarcinoma in the cervix and of type specific biological advantages that could explain the predominance of HPV16 and to some extent of HPV18 over all other mucosal HPV types. 3) Generate more data on multiple types by determining the viral type implicated in a cancer in which several types are identified (i.e., by microdissecting the tissue and/or with carefully conducted in situ hybridizations) 4) Provide additional data on the risk linked to HPV variants. It would be necessary to develop technology that allows typing and variant detection in large numbers and generate worldwide results in the same manner as it has been done for the HPV type distributions. 5) Standardize technology for measuring viral load and viral integration in routine cohort studies. 6) Characterize new biomarkers of interaction between HPV-DNA and cellular DNA that would signal latency, persistency, or progression to neoplasm. 7) Characterize biomarkers of genetic susceptibility to be used in population studies and pursue research into the familial aggregation of cervical cancer. Table 1. Prevalence of cervical oncogenic HPV-DNA in selected samples of the general population*       HPV-DNA    Region  Testing method  Study population  Age range of study, y  (%+)  Age range above 30†  (%+)  Reference  *HC2 = Hybrid Capture 2; PCR = polymerase chain reaction. The HC2 assay was used in 14 samples without PCR amplification.  †As published or estimated by authors.  Africa      Eastern          Mozambique  PCR/HC2  Survey  14–61  40.0  31–61  30.5  (2)          Zimbabwe  HC2  Routine screening  25–55  42.8  —  —  (3)      Northern          Morocco  PCR  Hospital case-control  18–70  21.6  35–70  21.6  (4)      Southern          South Africa (Black)  HC2  Routine screening  —  —  35–65  21.3  (5)      Western          Senegal  PCR  Hospital case-control  —  —  35–83  43.7  (6)  America      Central          Mexico  PCR  Population-based  <25–>65  14.5  35–>65  14.1  (7)          Costa Rica  PCR  Routine screening  18–94  16.0  —  —  (8)      South          Colombia  PCR  Survey  13–85  14.9  35–85  8.4  (9)          Argentina  PCR  Survey  15–>55  16.6  35–>55  13.9  (10)      Northern          Canada  PCR  Routine screening  15–49  13.3  35–49  9.0  (11)          United States  PCR  Routine screening  16–77  22.5  —  —  (12)  Europe      Eastern          Russian Federation  PCR  Routine screening  15–45  29.0  31–45  25.5  (13)      Northern          United Kingdom  PCR  Routine screening  —  —  34–70  5.9  (14)          Denmark  PCR  Survey  20–29  18.0  —  —  (15)      Southern          Spain  PCR  Population-based  14–75  3.0  35–75  2.0  (15a)          Italy  PCR  Population-based  25–70  8.6  35–70  7.4  (16)          Greece  PCR  Routine screening  20–55  36.2  —  —  (17)      Western          The Netherlands  PCR  Population-based  15–69  4.6  30–69  4.3  (18)          Germany  PCR  Routine screening  18–70  7.8  36–70  4.9  (19)          France  HC2  Population-based  15–76  15.3  31–76  12.3  (20)  Asia      Europe          Japan  PCR  Routine screening  —  —  30–78  7.0  (21)          Korea, DR  PCR  Survey  20–74  10.4  35–74  9.6  (22)          Taiwan  PCR  Routine screening  —  —  30–64  9.2  (23)          China  HC2  Survey  —  —  35–45  18.0  (24)      South Eastern          Vietnam, North  PCR  Survey  15–69  2.0  35–69  1.9  (25)          Vietnam, South  PCR  Survey  15–69  10.9  35–69  8.0  (25)          Thailand  PCR  Population-based  15–>65  6.3  35–>65  5.0  (26)      Southcentral          India (Madras)  PCR  Hospital case-control  23–76  27.7  35–76  28.5  (27)        HPV-DNA    Region  Testing method  Study population  Age range of study, y  (%+)  Age range above 30†  (%+)  Reference  *HC2 = Hybrid Capture 2; PCR = polymerase chain reaction. The HC2 assay was used in 14 samples without PCR amplification.  †As published or estimated by authors.  Africa      Eastern          Mozambique  PCR/HC2  Survey  14–61  40.0  31–61  30.5  (2)          Zimbabwe  HC2  Routine screening  25–55  42.8  —  —  (3)      Northern          Morocco  PCR  Hospital case-control  18–70  21.6  35–70  21.6  (4)      Southern          South Africa (Black)  HC2  Routine screening  —  —  35–65  21.3  (5)      Western          Senegal  PCR  Hospital case-control  —  —  35–83  43.7  (6)  America      Central          Mexico  PCR  Population-based  <25–>65  14.5  35–>65  14.1  (7)          Costa Rica  PCR  Routine screening  18–94  16.0  —  —  (8)      South          Colombia  PCR  Survey  13–85  14.9  35–85  8.4  (9)          Argentina  PCR  Survey  15–>55  16.6  35–>55  13.9  (10)      Northern          Canada  PCR  Routine screening  15–49  13.3  35–49  9.0  (11)          United States  PCR  Routine screening  16–77  22.5  —  —  (12)  Europe      Eastern          Russian Federation  PCR  Routine screening  15–45  29.0  31–45  25.5  (13)      Northern          United Kingdom  PCR  Routine screening  —  —  34–70  5.9  (14)          Denmark  PCR  Survey  20–29  18.0  —  —  (15)      Southern          Spain  PCR  Population-based  14–75  3.0  35–75  2.0  (15a)          Italy  PCR  Population-based  25–70  8.6  35–70  7.4  (16)          Greece  PCR  Routine screening  20–55  36.2  —  —  (17)      Western          The Netherlands  PCR  Population-based  15–69  4.6  30–69  4.3  (18)          Germany  PCR  Routine screening  18–70  7.8  36–70  4.9  (19)          France  HC2  Population-based  15–76  15.3  31–76  12.3  (20)  Asia      Europe          Japan  PCR  Routine screening  —  —  30–78  7.0  (21)          Korea, DR  PCR  Survey  20–74  10.4  35–74  9.6  (22)          Taiwan  PCR  Routine screening  —  —  30–64  9.2  (23)          China  HC2  Survey  —  —  35–45  18.0  (24)      South Eastern          Vietnam, North  PCR  Survey  15–69  2.0  35–69  1.9  (25)          Vietnam, South  PCR  Survey  15–69  10.9  35–69  8.0  (25)          Thailand  PCR  Population-based  15–>65  6.3  35–>65  5.0  (26)      Southcentral          India (Madras)  PCR  Hospital case-control  23–76  27.7  35–76  28.5  (27)  View Large Table 2. Estimated incidence of invasive cervical cancer in the world*         Incidence (36)  Age-specific incidence, y  Region  No. of cases 1985 (34)  No. of cases 1990 (35)  No. of cases 2000 (36)  CR  ASRW  15–44  45–54  55–64  65+  *CR = crude rates per 100 000; ASRW = age-standardized rates per 100 000. World standard population. All Rates correspond to the most recent compilation of cancer incidence.  †Melanesia, Micronesia and Polynesia.  World  437 300  371 200  470 606  15.7  16.1  9.5  44.9  51.8  41.9      More developed  93 700  83 300  91 451  15.0  11.3  11.9  22.4  23.8  26.3      Less developed  343 600  287 900  379 153  15.8  18.7  9.0  53.6  65.0  53.8  Africa  51 500  52 500  67 078  17.1  27.3  11.0  71.5  100.5  95.4      Eastern  21 800  21 500  30 206  24.4  44.3  16.1  114.8  174.4  153.9      Middle  6600  5700  6947  14.4  25.1  8.5  54.0  73.3  137.4      Northern  6200  5200  10 479  12.2  16.8  6.2  49.0  68.5  45.9      Southern  6600  6500  5541  23.2  30.3  15.5  67.8  98.5  118.2      Western  10 300  13 600  13 903  12.5  20.3  9.5  57.4  70.6  60.3  America  68 000  74 800  92 136  22.0  21  15.1  55.2  57.8  55.0      Caribbean  3000  5000  6670  34.8  35.8  17.7  82.7  102.1  155.6      Central  13 700  17 700  21 596  31.7  40.3  22.5  111.7  109.9  136.1      South  35 300  36 900  49 025  28.1  30.9  16.8  85.5  90.2  101.4      United States and Canada  16 000  15 200  14 845  9.5  7.9  9.0  15.4  16.8  14.2  Europe  67 000  58 200  64 928  17.2  13  14.1  26.3  26.5  28.1      Eastern  40 100  27 500  35 482  21.9  16.8  17.8  34.5  34.9  36.6      Northern  6300  7600  6049  12.6  9.8  12.0  17.6  16.7  20.2      Southern  8700  9900  10 116  13.7  10.2  10.5  20.8  23.7  20.9      Western  11 900  13 200  13 282  14.2  10.4  11.3  20.5  19.6  25.1  Asia  249 000  183 400  245 670  13.6  14.9  7.2  44.0  52.8  39.6  Asia excluding China  170 800  158 700  212 297  18.0  21.1  10.3  63.0  75.0  53.5      Eastern  94 200  42 500  51 266  7.1  6.4  2.6  18.4  18.9  25.4          China  78 200  24 700  33 373  5.4  5.2  1.7  17.2  16.1  19.1          Japan  9400  8500  11 681  18.1  11.1  8.8  21.2  25.6  42.2          Other  6600  9300  6212  16.2  15.3  10.3  33.6  41.9  55.0      Southeastern  42 500  30 900  39 648  15.3  18.3  9.1  59.0  58.2  45.9      Southcentral  109 500  107 000  151 297  20.9  26.5  11.9  79.2  100.8  65.6      Western  2800  3000  3458  3.8  4.8  2.6  13.1  15.3  14.1  Oceania  1800  2100  2156  14.2  12.6  12.3  27.4  28.2  29.0      Pacific Islands†  500  800  1078  29.1  40.3  23.1  91.2  107.1  167.8      Australia/N. Zealand  1300  1300  1077  9.4  7.7  8.6  15.5  14.8  16.6          Incidence (36)  Age-specific incidence, y  Region  No. of cases 1985 (34)  No. of cases 1990 (35)  No. of cases 2000 (36)  CR  ASRW  15–44  45–54  55–64  65+  *CR = crude rates per 100 000; ASRW = age-standardized rates per 100 000. World standard population. All Rates correspond to the most recent compilation of cancer incidence.  †Melanesia, Micronesia and Polynesia.  World  437 300  371 200  470 606  15.7  16.1  9.5  44.9  51.8  41.9      More developed  93 700  83 300  91 451  15.0  11.3  11.9  22.4  23.8  26.3      Less developed  343 600  287 900  379 153  15.8  18.7  9.0  53.6  65.0  53.8  Africa  51 500  52 500  67 078  17.1  27.3  11.0  71.5  100.5  95.4      Eastern  21 800  21 500  30 206  24.4  44.3  16.1  114.8  174.4  153.9      Middle  6600  5700  6947  14.4  25.1  8.5  54.0  73.3  137.4      Northern  6200  5200  10 479  12.2  16.8  6.2  49.0  68.5  45.9      Southern  6600  6500  5541  23.2  30.3  15.5  67.8  98.5  118.2      Western  10 300  13 600  13 903  12.5  20.3  9.5  57.4  70.6  60.3  America  68 000  74 800  92 136  22.0  21  15.1  55.2  57.8  55.0      Caribbean  3000  5000  6670  34.8  35.8  17.7  82.7  102.1  155.6      Central  13 700  17 700  21 596  31.7  40.3  22.5  111.7  109.9  136.1      South  35 300  36 900  49 025  28.1  30.9  16.8  85.5  90.2  101.4      United States and Canada  16 000  15 200  14 845  9.5  7.9  9.0  15.4  16.8  14.2  Europe  67 000  58 200  64 928  17.2  13  14.1  26.3  26.5  28.1      Eastern  40 100  27 500  35 482  21.9  16.8  17.8  34.5  34.9  36.6      Northern  6300  7600  6049  12.6  9.8  12.0  17.6  16.7  20.2      Southern  8700  9900  10 116  13.7  10.2  10.5  20.8  23.7  20.9      Western  11 900  13 200  13 282  14.2  10.4  11.3  20.5  19.6  25.1  Asia  249 000  183 400  245 670  13.6  14.9  7.2  44.0  52.8  39.6  Asia excluding China  170 800  158 700  212 297  18.0  21.1  10.3  63.0  75.0  53.5      Eastern  94 200  42 500  51 266  7.1  6.4  2.6  18.4  18.9  25.4          China  78 200  24 700  33 373  5.4  5.2  1.7  17.2  16.1  19.1          Japan  9400  8500  11 681  18.1  11.1  8.8  21.2  25.6  42.2          Other  6600  9300  6212  16.2  15.3  10.3  33.6  41.9  55.0      Southeastern  42 500  30 900  39 648  15.3  18.3  9.1  59.0  58.2  45.9      Southcentral  109 500  107 000  151 297  20.9  26.5  11.9  79.2  100.8  65.6      Western  2800  3000  3458  3.8  4.8  2.6  13.1  15.3  14.1  Oceania  1800  2100  2156  14.2  12.6  12.3  27.4  28.2  29.0      Pacific Islands†  500  800  1078  29.1  40.3  23.1  91.2  107.1  167.8      Australia/N. Zealand  1300  1300  1077  9.4  7.7  8.6  15.5  14.8  16.6  View Large Table 3. Etiology of cervical cancer: established factors among HPV–DNA-positive women Factors  HPV  Use of exogenous hormones  Smoking  Parity  Sexual factors of the partner  Qualifiers  Persistent presence of any of at least 15 HPV–DNA types in cervical cells  Long-term use of hormonal contraception (5+ yr)  Dose response unproven
 Likely role of cancer promoter  High parity (five or more pregnancies)  (Probably related to the risk of HPV infection) 
 Partner with 6+ other sexual partners in some countries 
 Uncircumcised partner with 6+ other sexual partners  Research topics  Viral load 
 Multiple HPV types 
 Viral variants 
 HPV types and cancer histology 
 Predominance of HPV16 
 Factors and biomarkers for latency, persistency and progression  Hormonal composition 
 Hormonal replacement therapy 
 Tamoxifen 
 Duration of effect after cessation of use 
 Mechanisms of cancer promotion  Mechanisms of cancer promotion  Mechanisms of cancer promotion  Natural history of HPV infections in males  Epidemiological references from IARC studies  (1) 
 (44)  (45)  (46)  (47)  (48)  Factors  HPV  Use of exogenous hormones  Smoking  Parity  Sexual factors of the partner  Qualifiers  Persistent presence of any of at least 15 HPV–DNA types in cervical cells  Long-term use of hormonal contraception (5+ yr)  Dose response unproven
 Likely role of cancer promoter  High parity (five or more pregnancies)  (Probably related to the risk of HPV infection) 
 Partner with 6+ other sexual partners in some countries 
 Uncircumcised partner with 6+ other sexual partners  Research topics  Viral load 
 Multiple HPV types 
 Viral variants 
 HPV types and cancer histology 
 Predominance of HPV16 
 Factors and biomarkers for latency, persistency and progression  Hormonal composition 
 Hormonal replacement therapy 
 Tamoxifen 
 Duration of effect after cessation of use 
 Mechanisms of cancer promotion  Mechanisms of cancer promotion  Mechanisms of cancer promotion  Natural history of HPV infections in males  Epidemiological references from IARC studies  (1) 
 (44)  (45)  (46)  (47)  (48)  View Large Table 4. Etiology of cervical cancer: possible factors among HPV–DNA-positive women requiring further research Factors  Other STDs  Other environmental factors  Host Factors host and viral interactions  STDs = sexually transmitted diseases.  Research topics  C. Trachomatis HSV-2  Low Socio-economic status Dietary factors  Early age at infection 
 Genetic background (HLA, p53 . . .) 
 Susceptibility of the transitional zone 
 Viral integration 
 Immune response to HPV and determinants of viral clearance 
 Nonspecific cervical inflammation  Epidemiological references from IARC studies  (49)
 (50)  (51)    Factors  Other STDs  Other environmental factors  Host Factors host and viral interactions  STDs = sexually transmitted diseases.  Research topics  C. Trachomatis HSV-2  Low Socio-economic status Dietary factors  Early age at infection 
 Genetic background (HLA, p53 . . .) 
 Susceptibility of the transitional zone 
 Viral integration 
 Immune response to HPV and determinants of viral clearance 
 Nonspecific cervical inflammation  Epidemiological references from IARC studies  (49)
 (50)  (51)    View Large Fig. 1. View largeDownload slide Cumulative prevalence of the most common HPV types in cervical cancer cases by histology and in women with normal cytology (source: the International Agency for Research on Cancer multicenter control studies). Fig. 1. View largeDownload slide Cumulative prevalence of the most common HPV types in cervical cancer cases by histology and in women with normal cytology (source: the International Agency for Research on Cancer multicenter control studies). Fig. 2. View largeDownload slide Age-specific incidence rates of cervical cancer in selected countries in Latin America, Europe and in the United States by ethnic groups. Fig. 2. View largeDownload slide Age-specific incidence rates of cervical cancer in selected countries in Latin America, Europe and in the United States by ethnic groups. Fig. 3. View largeDownload slide Age specific incidence rate of CIN 3/carcinoma in situ and of invasive cervical cancer in selected populations. Ratio: Number of cases of CIN 3/in situ carcinoma of the cervix over the number of cases of invasive cancer diagnosed in the same population over the same time period across all age groups. *Data provided as a courtesy of the Finnish Cancer Registry, East Anglian Cancer Registry, and the Iceland Cancer Registry. Fig. 3. View largeDownload slide Age specific incidence rate of CIN 3/carcinoma in situ and of invasive cervical cancer in selected populations. Ratio: Number of cases of CIN 3/in situ carcinoma of the cervix over the number of cases of invasive cancer diagnosed in the same population over the same time period across all age groups. *Data provided as a courtesy of the Finnish Cancer Registry, East Anglian Cancer Registry, and the Iceland Cancer Registry. We thank Mireia Diaz for the extraction and preparation of data on cancer incidence, Mireia Diaz and Dr. Roberto Martinez-Vimbert for the literature review on HPV-DNA prevalence worldwide, and Cristina Rajo and Elisabet Luquín for secretarial and editorial support. This chapter was partially supported by a Research Agreement between the Epidemiology and Cancer Registration Unit (SERC) and the International Agency for Research on Cancer (IARC) (CRA FIS/01/04). Partial support was granted by the Fondo de Investigaciones Sanitarias of the Spanish Government (FIS 01/1237). Dr. Bosch received a Contract for Services offered by the National Cancer Institute (NCI). Dr. Martinez-Vimbert received a grant from Generalitat de Catalunya (CIRIT 2001FI00068). The authors are indebted to the Icelandic Cancer Registry, the Finnish Cancer Registry, and the East Anglian Cancer Registry, who provided data on the incidence of cervical cancer precursors. References 1 Bosch FX, Lorincz A, Munoz N, Meijer CJ, Shah KV. The causal relation between human papillomavirus and cervical cancer. J Clin Pathol  2002; 55: 244–65. Google Scholar 2 Castellsagué X, Menéndez C, Loscertales MP, Kornegay JR, dos Santos F, Gómez-Olivé FX, et al. Human papillomavirus genotypes in rural Mozambique. Lancet  2001; 358: 1429–30. Google Scholar 3 Blumenthal PD, Gaffikin L, Chirenje ZM, McGrath J, Womack S, Shah K. Adjunctive testing for cervical cancer in low resource settings with visual inspection, HPV, and the Pap smear. Int J Gynaecol Obstet  2001; 72: 47–53. Google Scholar 4 Chaouki N, Bosch FX, Muñoz N, Meijer CJLM, El-Gueddari B, El-Ghazi A, et al. The viral origin of cervical cancer in Rabat, Morocco. Int J Cancer  1998; 75: 546–54. Google Scholar 5 Wright Jr TC, Denny L, Kuhn L, Pollack A, Lorincz A. HPV DNA testing of self-collected vaginal samples compared with cytologic screening to detect cervical cancer. J Am Med Assoc  2000; 282: 81–86. Google Scholar 6 Lin P, Koutsky LA, Critchlow CW, Apple RJ, Hawes SE, Hughes JP, et al. HLA class II Dr-DQ and increased risk of cervical cancer among Senegalese women. Cancer Epidemiol Biomark Prev  2001; 10: 1037–45. Google Scholar 7 Lazcano Ponce E, Herrero R, Muñoz N, Cruz A, Shah KV, Alonso P, et al. Epidemiology on HPV infection among Mexican women with normal cervical cytology. Int J Cancer  2001; 91: 412–20. Google Scholar 8 Herrero R, Hildesheim A, Bratti C, Sherman ME, Hutchinson M, Morales J, et al. Population-based study of human papillomavirus infection and cervical neoplasia in rural Costa Rica. J Natl Cancer Inst  2000; 92: 464–74. Google Scholar 9 Molano M, Posso H, Weiderpass E, Van den Brule AJC, Ronderos M, Franceschi S, et al. Prevalence and determinants of HPV infection among Colombian women with normal cytology. Br J Cancer  2002; 87: 324–33. Google Scholar 10 Matos E, Loria D, Amestoy G, Herrera L, Prince MA, Moreno J, et al. Prevalence of human papillomavirus (HPV) infection among women in Concordia, Argentina: a population-based study. Sex Transm Dis . In press. 2003. Google Scholar 11 Sellors JW, Mahony JB, Kaczorowski J, Lytwyn A, Bangura H, Chong S, et al. Prevalence and predictors of human papillomavirus infection in women in Ontario, Canada. CMAJ  2000; 163: 503–8. Google Scholar 12 Cope JU, Hildesheim A, Schiffman MH, Manos MM, Lorincz AT, Burk RD, et al. Comparison of the hybrid capture tube test and PCR for detection of human papillomavirus DNA in cervical specimens. J Clin Microbiol  1997; 35: 2262–5. Google Scholar 13 Alexandrova YN, Lyshchov AA, Safronnikova NR, Imyanitov EN, Hanson KP. Features of HPV infection among the healthy attendants of gynecological practice in St. Petersburg, Russia. Cancer Lett  1999; 145: 43–8. Google Scholar 14 Cuzick J, Beverley E, Ho L, Terry G, Sapper H, Mielzynska I, et al. HPV testing in primary screening of older women. Br J Cancer  1999; 81: 554–58. Google Scholar 15 Kjaer SK, Chackerian B, van der Brule AJC, Svare EI, Paull G, Walboomers JM, et al. High-risk human papillomavirus is sexually transmitted: evidence from a follow-up study of virgins starting sexual activity (intercourse). Cancer Epidemiol Biomarkers Prev  2001; 10: 101–6. Google Scholar 15a de Sanjose S, Almizall R, Lloveras B, Font R, Diaz M, Muñoz N, et al. Cervical human papillomavirus infection in the female population in Barcelona, Spain. Sex Transm Dis . In press. 2003. Google Scholar 16 Ronco G, Ghisetti V, Snijders PJF, Gillio-Tos A, Segnan N, Franceschi S. Age dependence of HPV prevalence and its determinants in Turin, Italy. In: Institut Pasteur, editor. 20th International Papillomavirus Conference. Program and Abstracts book. (www.hpv2002.com), Paris: Colloquium; 2002. p. 232. Google Scholar 17 Agorastos T, Bontis J, Lambropoulos AF, Constantinidis TC, Nasioutziki M, Tagou C, et al. Epidemiology of human papillomavirus infection in Greek asymptomatic women. Eur J Cancer Prev  1995; 4: 159–67. Google Scholar 18 Jacobs MV, Walboomers JMM, Snijders PJF, Voorhorst FJ, Verheijen RHM, Fransen-Daalmeijer N, et al. Distribution of 37 mucosotropic HPV types in women with cytologically normal cervical smears: the age-related patterns for high-risk and low types. Int J Cancer  2000; 87: 221–7. Google Scholar 19 Schneider A, Hoyer H, Lotz B, Leistritza S, Kühne-Heid R, Nindl I, et al. Screening for high-grade cervical intra-epithelial neoplasia and cancer by testing for high-risk HPV, routine cytology or colposcopy. Int J Cancer  2000; 89: 529–34. Google Scholar 20 Clavel C, Masure M, Bory JP, Putaud I, Mangeonjean C, Lorenzato M, et al. Human papillomavirus testing in primary screening for the detection of high-grade cervical lesions: a study of 7932 women. Br J Cancer  2001; 84: 1616–23. Google Scholar 21 Zheng PS, Iwasaka T, Zhang ZM, Pater A, Sugimori H. Telomerase activity in Papanicolaou smear-negative exfoliated cervical cells and its association with lesions and oncogenic human papillomaviruses. Gynecol Oncol  2000; 77: 394–8. Google Scholar 22 Shin HR, Lee DH, Herrero R, Smith JS, Vaccarella S, Hong SH, et al. Prevalence of human papillomavirus infection in women in Busan, South Korea. Int J Cancer  2003; 103: 413–21. Google Scholar 23 Liaw K, Hsing AW, Chen ChJ, Schiffman MH, Zhang TY, Hsieh CY, et al. Human papilloma virus and cervical neoplasia: a case-control study in Taiwan. Int J Cancer  1995; 62: 565–71. Google Scholar 24 Belinson J, Qiao YL, Pretorius R, Zhang WH, Elson P, Li L, et al. Shanxi Province Cervical Cancer Screening Study: a cross-sectional comparative trial of multiple techniques to detect cervical neoplasia. Gynaecol Oncol  2001; 83: 439–44. Google Scholar 25 Anh PT, Hieu NT, Herrero R, Vaccarella S, Smith JS, Thuy NT, et al. Human papillomavirus infection among women in South and North Vietnam. Int J Cancer  2003; 104: 213–20. Google Scholar 26 Sukvirach S, Smith JS, Tunsakul S, Muñoz N, Kesararat W, Opasatian O, et al. Population-based human papillomavirus prevalence in Lampang and Songkla, Thailand. J Infect Dis . In press. 2003. Google Scholar 27 Rajkumar T, Franceschi S, Vaccarella S, Gajalakshmi V, Sharmila A, Snijders PJF, et al. The role of paan chewing and dietary habits in cervical carcinoma in Chennai, India. Br J Cancer . In press. 2003. Google Scholar 28 Bosch FX, Manos M, Muñoz N, Sherman M, Jansen A, Peto J, et al. Prevalence of human papillomavirus in cervical cancer: a worldwide perspective. J Natl Cancer Inst  1995; 87: 796–802. Google Scholar 29 Walboomers JMM, Jacobs MV, Manos MM, Bosch FX, Kummer JA, Shah KV, et al. Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol  1999; 189: 12–19. Google Scholar 30 Clifford GM, Smith JS, Plummer M, Muñoz N, Franceschi S. Human papillomavirus types in invasive cervical cancer worldwide: a meta-analysis. Br J Cancer  2003; 88: 63–73. Google Scholar 31 Giuliano AR, Papenfuss M, Abrahamsen M, Denman C, de Zapien JG, Henze JL, et al. Human papillomavirus infection at the United States-Mexico border: implications for cervical cancer prevention and control. Cancer Epidemiol Biomarkers Prev  2001; 10: 1129–36. Google Scholar 32 Chan PK, Li WH, Chan MY, Ma WL, Cheung JL, Cheng AF. High prevalence of human papillomavirus type 58 in Chinese women with cervical cancer and precancerous lesions. J Med Virol  1999; 59: 232–38. Google Scholar 33 Huang S, Afonina I, Miller BA, Beckmann AM. Human papillomavirus types 52 and 58 are prevalent in cervical cancers from Chinese women. Int J Cancer  1997; 70: 408–11. Google Scholar 34 Parkin DM, Pisani P, Ferlay J. Estimates of the worldwide incidence of eighteen major cancers in 1985. Int J Cancer  1993; 54: 594–606. Google Scholar 35 Parkin DM, Pisani P, Ferlay J. Estimates of the worldwide incidence of 25 majors cancers in 1990. Int J Cancer  1999; 80: 827–41. Google Scholar 36 Ferlay J, Bray F., Pisani P., Parkin D.M., editors. Globocan 2000: Cancer incidence, mortality and prevalence worldwide, version 1.0. IARC CancerBase No. 5. Lyon (France): International Agency for Research on Cancer (IARC Press); 2001. Google Scholar 37 Li H, Jin S, Xu H, Thomas DB. The decline in the mortality rates of cervical cancer and a plausible explanation in Shandong, China. Int J Epidemiol  2000; 29: 398–404. Google Scholar 38 Vizcaino AP, Moreno V, Bosch FX, Muñoz N, Barros-Dios XM, Parkin DM. International trends in the incidence of cervical cancer: I. Adenocarcinoma and adenosquamous cell carcinomas. Int J Cancer  1998; 75: 536–45. Google Scholar 39 Vizcaino AP, Moreno V, Bosch FX, Muñoz N, Barros-Dios XM, Borras J, et al. International trends in incidence of cervical cancer: II. Squamous-cell carcinoma. Int J Cancer  2000; 86: 429–35. Google Scholar 40 Sasieni P, Adams J. Changing rates of adenocarcinoma and adenosquamous carcinoma of the cervix in England. Lancet  2001; 357: 1490–1493. Google Scholar 41 Parkin DM, Whelan SL, Ferlay J, Raymond L, Young J, editors. Cancer incidence in five continents, vol. VII. Lyon (France): International Agency for Research on Cancer (IARC Press); 1997. Google Scholar 42 Drain PK, Holmes KK, Hughes JP, Koutsky LA. Determinants of cervical cancer rates in developing countries. Int J Cancer  2002; 100: 199–205. Google Scholar 43 Shah KV, Howley PM. Papillomaviruses. In: Fields BN, Knipe DM, Howley PM, et al., editors. Fields virology, Philadelphia: Lippincott-Raven; 1996. p. 2077–109. Google Scholar 44 Muñoz N, Bosch FX, de Sanjosé S, Herrero R, Castellsagué X, Shah KV, et al. Epidemiologic classification of human papillomavirus types associated with cervical cancer. New Engl J Med  2003; 348: 518–27. Google Scholar 45 Moreno V, Bosch FX, Munoz N, Meijer CJ, Shah KV, Walboomers JM, et al. Effect of oral contraceptives on risk of cervical cancer in women with human papillomavirus infection: the IARC multicentric case-control study. Lancet  2002; 359: 1085–92. Google Scholar 46 IARC Press. IARC Monographs on the evaluation of carcinogenic risks to humans, Volume 83. Tobacco smoke and involuntary smoking. Lyon, (France): IARC Press. In press. 2003. Google Scholar 47 Muñoz N, Franceschi S, Bosetti C, Moreno V, Herrero R, Smith JS, et al. Role of parity and human papillomavirus in cervical cancer: the IARC multicentric case-control study. Lancet  2002; 359: 1093–102. Google Scholar 48 Castellsague X, Bosch FX, Munoz N, Meijer CJ, Shah KV, de Sanjose S, et al. Male circumcision, penile human papillomavirus infection, and cervical cancer in female partners. N Engl J Med  2002; 346: 1105–12. Google Scholar 49 Smith JS, Muñoz N, Herrero R, Eluf-Neto J, Ngelangel C, Franceschi S, et al. Evidence for Chlamydia trachomatis as a human papillomavirus cofactor in the etiology of invasive cervical cancer in Brazil and the Philippines. J Infect Dis  2002; 185: 324–31. Google Scholar 50 Smith JS, Herrero R, Bosetti C, Muñoz N, Bosch FX, Eluf-Neto J, et al. Herpes simplex virus-2 as a human papillomavirus cofactor in the etiology of invasive cervical cancer. J Natl Cancer Inst  2002; 94: 1604–13. Google Scholar 51 de Sanjosé S, Bosch FX, Muñoz N, Shah K. Social differences in sexual behaviour and cervical cancer. In: Kogevinas M, Pearce N, Susser M, Boffetta P, editors. Social inequalities and cancer. Lyon (France): International Agency for Research on Cancer (IARC Press); 1997. p. 309–18. Google Scholar 52 Castle PE, Schiffman M, Gravitt PE, Kendall H, Fishman S, Dong H, et al. Comparisons of HPV DNA detection by MY09/11 PCR methods. J Med Virol  2002; 68: 417–23. Google Scholar 53 Liaw KL, Glass AG, Manos MM, Greer CE, Scott DR, Sherman M, et al. Detection of human papillomavirus DNA in cytologically normal women and subsequent cervical squamous intraepithelial lesions. J Natl Cancer Inst  1999; 91: 954–60. Google Scholar 54 Woodman CB, Collins S, Winter H, Bailey A, Ellis J, Prior P, et al. Natural history of cervical human papillomavirus infection in young women: a longitudinal cohort study. Lancet  2001; 357: 1831–36. Google Scholar 55 Palefsky JM, Minkoff H, Kalish LA, Levine A, Sacks HS, García P, et al. Cervicovaginal human papillomavirus infection in human immunodeficiency virus-1 (HIV)-positive and high-risk HIV-negative women. J Natl Cancer Inst  1999; 91: 226–36. Google Scholar 56 Xi LF, Carter JJ, Galloway DA, Kuypers J, Hughes JP, Lee SK, et al. Acquisition and natural history of human papillomavirus type 16 variant infection among a cohort of female university students. Cancer Epidemiol Biomark Prev  2002; 11: 343–51. Google Scholar 57 Sherman ME, Schiffman M, Cox JT. Effects of age and human papilloma viral load on colposcopy triage: data from the randomized Atypical Squamous Cells of Undetermined Significance/Low-Grade Squamous Intraepithelial Lesion Triage Study (ALTS). J Natl Cancer Inst  2002; 94: 102–7. Google Scholar 58 Josefsson AM, Magnusson PK, Ylitalo N, Sorensen P, Qwarforth-Tubbin P, Andersen PK, et al. Viral load of human papilloma virus 16 as a determinant for development of cervical carcinoma in situ: a nested case-control study. Lancet  2000; 355: 2189–93. Google Scholar 59 Ylitalo N, Sorensen P, Josefsson AM, Magnusson PKE, Andersen PK, Pontén J, et al. Consistent high viral load of human papillomavirus 16 and risk of cervical carcinoma in situ: a nested case-control study. Lancet  2000; 355: 2194–98. Google Scholar 60 Storey A, Thomas M, Kalita A, Harwood C, Gardiol D, Mantovani F, et al. Role of a p53 polymorphism in the development of human papillomavirus-associated cancer. Nature  1998; 393: 229–34. Google Scholar 61 Makni H, Franco EL, Kaiano J, Villa LL, Labrecque S, Dudley R, et al. P53 polymorphism in codon 72 and risk of human papillomavirus-induced cervical cancer: effect of inter-laboratory variation. Int J Cancer  2000; 87: 528–33. Google Scholar © Oxford University Press

Journal

JNCI Monographs – Oxford University Press

Published: Jun 1, 2003

There are no references for this article.