Human papillomavirus in cervical screening and ... - Clinical Science

Clinical Science (2006) 110, 543–552 (Printed in Great Britain) doi:10.1042/CS20050230

R

E

V

I

E

W

Human papillomavirus in cervical screening and vaccination Emma J. CROSBIE and Henry C. KITCHENER Academic Unit of Obstetrics and Gynaecology, University of Manchester, St Mary’s Hospital, Whitworth Park, Manchester M13 0JH, U.K.

A

B

S

T

R

A

C

T

Recent decades have witnessed a reduction in the incidence of cervical cancer in countries where screening programmes have achieved broad coverage. The recognized importance of high-risk HPV (human papillomavirus) infection in the aetiology of cervical cancer may introduce a role for HPV DNA testing in cervical screening programmes. Positive HPV DNA tests indicate women at risk of cervical cancer with greater sensitivity, but reduced specificity, compared with exfoliative cytology. Combining HPV testing with cytology may be useful in the triage of minor cytological abnormalities into those requiring referral to colposcopy (HPV positive) compared with those who can be safely managed by cytological surveillance (HPV negative). With its high sensitivity and high-negative-predictive value, HPV testing may also be useful for predicting treatment failure, since residual disease is very unlikely in the event of a negative HPV test. Ultimately, prevention is better than cure, and the advent of HPV prophylactic vaccines may obviate the need for population-based cervical screening programmes in the future. A multivalent vaccine administered to adolescents prior to the onset of sexual activity and boosted at regular intervals throughout their sexually active life may provide protection against type-specific HPV infection, malignant precursors and invasive cervical disease. Several large randomized placebo-controlled trials have been conducted with promising results. For those generations of women already exposed to high-risk HPV infection, therapeutic vaccines may offer advantages over conventional treatment, although much work still needs to be done.

INTRODUCTION Cancer of the cervix is almost unique as a malignant disease. It has a well-characterized long-lasting pre-invasive phase which invariably precedes clinical cancer. The cervix is accessible for screening, and malignant disease can be simply and effectively treated in the preinvasive phase. It is therefore an ideal disease in which to employ screening as a strategy for secondary prevention. The means of screening for the past 50 years has been exfoliative cytology, the so called Pap smear, after its inventor George Papanicolaou. The advent of colposcopy in the 1970s provided a means of visualizing the cervix

under magnification and obtaining a directed biopsy. It also facilitated directed therapy such as laser and, more recently, diathermy loop excision, both of which can be used as out-patient or office procedures. These developments have enabled the detection and treatment of cancer precursors, leading to spectacular reductions in the incidence of cervical cancer in countries where broad screening coverage has been achieved. In the U.K., for example, deaths from cervical cancer have been cut by 50 % since 1988 [1] when a national call/recall programme was introduced, and the trend continues downward. So is the battle against cervical cancer won? Emphatically not in developing countries where the

Key words: cervical cancer, cervical intraepithelial neoplasia, cervical screening, human papillomavirus (HPV), prophylactic vaccination, therapeutic vaccination. Abbreviations: CIN, cervical intraepithelial neoplasia; CTL, cytotoxic T-lymphocyte; HPV, human papillomavirus; VLP, virus-like particle. Correspondence: Professor Henry C. Kitchener (email [email protected]).

C

2006 The Biochemical Society

543

544

E. J. Crosbie and H. C. Kitchener

Figure 1 Schematic representation of the genomic organization of HPV 16

The 7.9 kb genome of HPV 16 can be divided into three functional regions: the early region (E), which encodes six early proteins E1 to E7; the late region (L), encoding the viral capsid proteins (L1 and L2); and a long control region (LCR). resources to run screening programmes and facilities to treat malignant precursors are not available. Perversely, it is these countries which bear the greatest burden of disease: annually around 400 000 cases worldwide. This results in pain and misery for thousands of women who die prematurely. In countries where screening programmes are in place, cancers still occur, some in unscreened women and some in screened women. It has been estimated that cervical screening in its present form will prevent approx. 70 % of cancer deaths [2]. In the 1980s the epidemiological evidence linking cervical carcinogenesis with a sexually transmitted agent was confirmed, with HPV (human papillomavirus) DNA being identified in cervical cancers and pre-cancers. The papillomaviruses are a group of more than 120 different double-stranded DNA viruses that infect epithelial cells [3]. Papillomaviruses are classified as cutaneous or mucosal depending on the type of epithelium they infect. Mucosal types are further sub-divided according to their oncogenicity. Low-risk types (e.g. HPV6 and 11) cause benign warts, whereas high-risk types (e.g. HPV16 and 18) are associated with malignant disease of the female lower genital tract. The life cycle of the papillomavirus is absolutely dependent upon the complete differentiation programme of the keratinocyte [4]. HPV infects cells in the basal layer of the epithelium, but produces virus only in the superficial terminally differentiated cells at the culmination of a maturation process that takes weeks to complete. The papillomavirus genome encodes six early (E) functional proteins and two late (L) structural proteins (Figure 1). C


The early proteins E1 and E2 are required for viral replication [5]. E6 and E7, the viral oncoproteins, exert their effects via interaction with cellular tumour-suppressor gene products p53 and Rb (retinoblastoma) respectively. Together they create a cellular environment conducive to proliferation in which the normal checks on cellcycle control are disrupted. This allows occasional mitotic events to occur and errors to go unchecked, contributing to the malignant transformation of the cell [6] (for a comprehensive review on the molecular biology of HPV and cervical cancer, see [6a]). It is now accepted that HPV is a necessary cause of cervical cancer [7]. It has been shown that the continued expression of HPV is necessary to maintain the malignant phenotype [8]. In addition, persistent HPV infection over 2 years with the oncotype 16 is associated with a very high relative risk of acquiring CIN (cervical intraepithelial neoplasia) 3, the true precursor of cervical cancer [9,10]. HPV infection is thought to occur in the majority of women at some time during their sexually active life. It is extremely common in the late teens and early twenties. The prevalence then falls steadily from approx. 40 % to 5 % by the age of 50 years [11]. This is generally thought to be due to immunological clearance of the virus from the cervix, with reduced opportunities for re-infection as women get older. It is persistent infection that results in a very high relative risk of developing CIN. Cervical cancer occurs in 1 in 10 000 women per year in the U.K. Pre-cancers are ten times more common and HPV infections 1000 times more common, being present in 15 % of screened women. In some respects cervical cancer can be viewed as a rare complication of a very common infection. This causal link has two crucial implications. The first is that testing for HPV could have real benefits in cervical screening programmes, and the second is that vaccine development could produce both preventative and therapeutic strategies, affording huge benefit to the health of women worldwide. This article will review firstly the potential for HPV testing in screening and, secondly, HPV vaccines in the context of cervical cancer prevention and treatment.

HPV IN TESTING PRIMARY CERVICAL SCREENING There are a number of problems in using exfoliative cytology in cervical screening. The most important is its limited sensitivity [12]. Another difficulty is that it does not yield a dichotomous result: normal or abnormal. It identifies a spectrum of abnormality and, in order to maximize sensitivity, the low-grade abnormalities need to be acted upon. This results in reduced specificity requiring women to undergo unnecessary investigations. Could HPV testing improve this situation? Because of the necessity of HPV infection in cervical carcinogenesis, it should be a highly sensitive screening

Human papillomavirus in cervical screening and vaccination

test and this has been borne out in a number of studies [13–16]. Conversely, of course, the same studies indicate that the specificity of a positive HPV test, particularly in young women, is poorer than a Pap smear. Although its superior sensitivity over cytology has been shown convincingly, optimal strategies for using HPV testing in routine screening have not yet been worked out. One possibility would be to screen first for HPV and then to retest all positives with cytology. This would help overcome the specificity issue, but it is not clear whether the limited sensitivity of cytology would still result in falsenegative tests. The optimal way to manage women who were HPV positive/cytology negative has not been resolved and there are risks of lingering anxiety in affected women and unnecessary intervention, as most will have no underlying disease. Clearly, the epidemiological evidence points to repeat testing, but optimal intervals for this are not clear. In order to understand how HPV testing could be best employed in screening, a number of trials are currently in progress in The Netherlands, Sweden, Italy, U.K. and Canada. These involve different randomizations, but are designed to assess the added impact of HPV testing. Within the next 2 years a great deal of data from these trials will become available which will address some key questions. (i) What is the duration of protection of a negative HPV result? (ii) Is a cytology negative/HPV negative result associated with a longer lasting negative predictive value than HPV negative alone? (iii) Does intervening for HPV-positive tests reduce the risk for developing high-grade CIN compared with cytology screening? (iv) How should a cytology negative/HPV-positive woman be managed? It does appear that primary HPV testing could be both effective and cost effective, but its widespread introduction into a healthcare system used to providing cytology and a population of women accustomed to the ‘smear test’ will be challenging. Another aspect of HPV testing is its connotation of sexual transmission, which is troubling for a significant proportion of women. Qualitative research has identified a number of issues which concern women, including a feeling of stigma and also the implications of the infection within a stable sexual relationship [17]. If HPV testing were to become widely used there would need to be an effective programme of education to explain clearly to women what a HPV test means in terms of cervical screening in order that it is perceived in a positive light. Other than primary screening, there are two other settings where HPV has been demonstrated to be effective: (i) to triage low-grade cytological abnormalities in order to select for colposcopic diagnosis; and (ii) to act as a ‘test of cure’ following treatment. In both of these examples, HPV testing can provide a straightforward guide to management. In the case of triage, the value of HPV testing lies in its ability to in-

crease the positive predictive value over the smear result alone, i.e. to add specificity to the process, increasing the likelihood that there is an underlying high-grade CIN lesion. At the same time, a negative HPV result carries a very high negative predictive value, which means the likelihood of an underlying high-grade lesion is extremely low, excluding the need for colposcopy [18]. In this way, HPV testing can increase diagnostic efficiency. It will require additional colposcopies compared with a strategy of repeated smears, but it shortens the time to diagnosis, which many women prefer, and it would reduce the risk of default from cytological surveillance [19]. In the case of post-treatment testing, the high sensitivity and high negative predictive value means that residual disease is very unlikely if the HPV test is negative [20]. This provides a strategy for returning treated women to call/recall after say 12 months compared with repeated annual cytology for 5–10 years. Conversely, a positive result would require continued surveillance and, if persistent, repeat colposcopy even in the presence of normal cytology as there may be residual disease. Neither of these uses for HPV testing is NHS policy yet, but demonstration pilot studies have been undertaken and the potential for implementation is being assessed.

THE HPV TEST So what about the test itself? HPV detection has employed a number of methods over the years mostly PCRbased and conducted in research laboratories. Clinical testing, however, requires a quality assured kit test that uses a methodology suitable for mass testing and yields highly reproducible results. The first commercially available kit was called Hybrid CaptureTM (Digene). This technique relies on a nucleic acid hybridization reaction, which reads as a chemoluminescent signal. It does not identify individual HPV types, but uses a probe mix of 13 high-risk types. This system has been widely used and, although there is probably some degree of cross-reaction with low-risk types, it has generally performed well. Another commercially developed test is the Microwell Plate System (Roche) which, unlike Hybrid CaptureTM , requires PCR to amplify the DNA. Claims have been made that this test is more sensitive, and this may or may not be clinically relevant. One key aspect of HPV testing in the context of cervical screening is its ability to be performed in liquidbased cytology systems. This means that, if a smear is taken and put into a liquid base instead of the conventional smear on a glass slide, the woman does not need to be recalled if a HPV test is required, as the liquid-based residue can be used. Both of these commercial kits contain a ‘cocktail’ of probes against a subject of the so-called high-risk HPV C


545

546


types, i.e. the 13 types most frequently associated with cervical cancer; the commonest high-risk types being 16, 18, 31, 33 and 45. By contrast, types 6 and 11 are associated with genital warts and non-neoplastic lesions. In addition to these DNA detection systems, an RNA detection system has been developed (PretectTM ; Norchip) which detects the E6/E7 in RNA. It is claimed that this may detect more relevant ‘infection’ by identifying persistent expression of the HPV oncogenes, but whether it performs better in a clinical setting is unproven. The above systems do not identify individual HPV types, and typing may become relevant in screening because a small subset of types, e.g. 16 and 18, may be dominantly associated with a higher risk of underlying CIN compared with certain other high-risk types, which may be relatively more common in women with normal cytology. One HPV typing system which should become commercially available is the Linear Array System (Roche). HPV will undoubtedly find its place in cervical screening, initially for triage of low-grade abnormalities and in follow-up after treatment of CIN. How effectively and cost effectively it can be used in primary screening is still uncertain. It could be used in conjunction with cytology as a double test, but this would appear unnecessarily expensive unless it can be shown that there is added value in testing cytology negative/HPV negative compared with HPV negative alone. More likely, it could be used as a primary ‘stand alone’ test with another test being used to increase the specificity if the HPV test was positive. At the moment, this would be liquid-based cytology, but other cell-cycle and molecular markers are being developed which could provide a simpler testing platform than cytology. Having reviewed the potential for HPV as a diagnostic test, we will now turn to its role as a prophylactic vaccine, which could result in a strategy of primary prevention for cervical cancer. Non-cervical intraepithelial neoplasia, e.g. vulval and anal, requires improved therapies and, although effective therapeutic vaccines are not yet available, some early candidates have shown promise.

RATIONALE FOR VACCINE DEVELOPMENT As described above, the central role of high-risk HPV infection in cervical carcinogenesis is well established [3]. Genital HPV infections are common in young sexually active women and, although most are cleared spontaneously without ever causing dysplasia [18], persistent high-risk HPV infections are implicated in progressive cervical disease [19]. Such persistent infections are likely to result from the combined effects of viral strategies that allow HPV to avoid immune detection on the one hand, combined with ineffective host immunological responses that fail to clear the virus on the other. In accordance with this theory, immunocompromised women have a higher C


incidence of HPV-associated warts, CIN and cervical cancer [20]. Where HPV infections persist and lead to dysplastic change in the cervix, there is still considerable opportunity for natural immune responses to eradicate them; indeed, most untreated CIN lesions undergo spontaneous regression with time [21]. Thus cervical cancer can be considered a late and rare complication of highrisk HPV infection [3].

NATURAL IMMUNE RESPONSES IN HPV-ASSOCIATED DISEASE Precisely why some infections persist and others undergo natural clearance has been the focus of considerable research effort over the past two decades. Unfortunately, these types of study are fraught with methodological difficulties. Measuring HPV-specific immune responses requires prior knowledge of the type(s) of immunity that may be important and different immunological assays often produce conflicting results. Neutralizing antibodies are likely to be important in preventing viral infection, but as HPV is an intracellular virus and thus protected from interference by host antibodies throughout its life cycle, it is unlikely that antibodies alone will be sufficient to clear an established HPV infection. Rather, it is believed that Tcell recognition and the subsequent elimination of HPVinfected cells via antigen-specific cellular immunity is the key to disease resolution [22]. An understanding of what is happening locally at the cervix may be more relevant than what can be measured systemically, but the process of taking a biopsy to allow infiltrating immune cells to be typed, quantified and their specificities determined may well alter the natural course of events. Despite these difficulties, several longitudinal studies have compared naturally occurring immune responses in women infected with high-risk HPV in whom lesions persist with those in whom lesions clear spontaneously. Although healthy controls frequently show evidence of HPV16 E6-specific T-cell responses [23–25], an absence of this immunity has been reported in women with persistent HPV16 infections [26] and cervical carcinoma [25,27]. By contrast, HPV16 E7-specific T-cell responses have been detected at low levels in the systemic circulation of patients with persistent HPV16-associated dysplastic and neoplastic lesions [27–30], but less frequently in healthy controls. E7-specific T-cells have also been found in the draining lymph nodes and tumours of cervical cancer patients [31]. Most studies have concentrated on measuring HPV16 E6- and/or E7-specific immune responses, since these antigens are continuously expressed throughout the full spectrum of HPV-associated disease development [32]. More recently, E2-specific T-cell responses have been reported in patients and healthy volunteers alike, although the relationship between the measured immunity and disease outcome is unclear [25,33,34]. Taking all of these results together, it appears


that the specific type(s) of cellular immunity that an individual mounts in response to an HPV infection may influence its outcome. The precise combination of particular HPV-specific CTL (cytotoxic T-lymphocyte) and T-helper responses, together with the appropriate cytokine milieu [25], may be critically important in determining whether an individual infection persists or is cleared. Antibodies to the major papillomavirus capsid protein L1 are seen concomitantly with or within a few months following acquisition of HPV16 DNA in up to 70 % of those infected [35,36]. Antibody titres appear to be correlated with antigen exposure, with higher levels found in women with persistent HPV infections and high-grade CIN [37]. In vitro studies indicate that these capsid antibodies block the interaction between infectious virions and their epithelial receptor, providing protection against repeat infection with the same papillomavirus [38]. Recent evidence in humans, however, suggests that naturally induced HPV-specific capsid antibodies may not necessarily protect against subsequent infections with homologous or genetically related HPV types [39] and some authors have suggested that T-cell immunity may also play a critical role in preventing repeat infections with the same or similar HPV types.

VACCINE DEVELOPMENT TO PREVENT AND/OR TREAT HPV-ASSOCIATED DISEASE Clearly, a detailed understanding of the precise immune mechanisms that are important in the natural control of HPV infections is necessary for the optimization of vaccine design. While this knowledge accumulates, two main types of vaccine have been proposed for the control of HPV-associated disease: prophylactic vaccines, to prevent HPV infection, and therapeutic vaccines, to treat established infection and HPV-associated disease.

Prophylactic HPV vaccines Prophylactic vaccines are designed to induce high titres of virus-neutralizing antibodies in the female lower genital tract that are capable of preventing infection [40]. The binding of antibody to intact papillomaviruses is thought to block the interaction with their receptor on the cervical epithelium, facilitating opsonization and degradation of the virions by macrophages [38]. A major breakthrough in prophylactic vaccine design followed the discovery that the L1 capsid protein has the intrinsic ability to self assemble into VLPs (virus-like particles) when expressed in the absence of other papillomavirus gene products [41]. The resulting VLPs are morphologically identical with native virions, but are not in themselves infectious because they lack the viral genome. High titres of papillomavirus-type-specific neutralizing antibodies and subsequent protection against high-dose experimental viral challenge have been seen in animals immunized

with these VLP vaccines [42]. Further encouragement has been provided by early phase clinical trials which have established that the vaccines are both well tolerated and highly immunogenic, eliciting antibody titres at least 40 times higher than those seen after natural infection [43].

Prophylactic HPV16 VLP vaccine Following on from this success, three large randomized controlled efficacy trials have now established proofof-principle for HPV prophylactic vaccines [44–46] (Table 1). In the first of these studies, Koutsky et al. [44] randomly assigned over 1500 young women to receive either three doses of an HPV16 L1 VLP vaccine or placebo. During an average follow-up period of 17 months, all 41 cases of persistent HPV16 infection (defined as two or more HPV16 DNA-positive samples at consecutive visits 4 or more months apart) and all nine cases of HPV16-positive CIN (five CIN 1 lesions and four CIN 2 lesions) occurred in the placebo group. There were no cases of HPV16-positive CIN 3 lessions or invasive cervical cancer detected in either group during the study [44]. Although the ultimate aim of prophylactic vaccination is to protect against cervical cancer, this end point is difficult to study for obvious ethical reasons, and persistent HPV infection is therefore regarded as an appropriate surrogate clinical end point. Although the results of this study are extremely promising, they do serve to highlight some of the challenges that remain in the implementation of HPV prophylactic vaccination strategies. Of the original 2392 women recruited into the trial, over 500 were excluded from the primary efficacy analysis because they were already HPV16 seropositive or HPV16 DNA positive at enrolment. A population-based prophylactic vaccination programme would therefore need to be aimed at women much younger than the 16–23-year olds studied here if it were to significantly impact on HPV16 acquisition rates. Vaccinating girls prior to the onset of sexual activity, possibly as young as 10–13 years of age, will require sensitive education programmes aimed at adolescents and their families, as well as parental consent. Another important point raised by this study [44] is that protection against CIN offered by prophylactic vaccination appears to be type-specific. All nine cases of HPV16-associated CIN occurred in the placebo group, but a further 22 cases of CIN associated with other HPV types were seen in each of the vaccinated and placebo groups. This finding emphasizes the importance of vaccinating against more than one HPV type. Over 50 % of cervical cancers are associated with HPV16 infection, but the remaining proportion are associated with any of a whole range of high-risk HPV types, the most common of which (types 18, 31, 33 and 45) account for up to a further 30 % of cervical cancer cases worldwide [47]. C


547

548


Table 1 Proof-of-principle HPV prophylactic vaccine trials

CI, confidence intervals. References Koutsky et al. (2002) [44]

Harper et al. (2004) [45]

Villa et al. (2005) [46]

Vaccine manufacturer Vaccine type

Merck Research Laboratories Monovalent HPV16 L1 VLP

GlaxoSmithKline Biologicals Bivalent HPV16 and 18 L1 VLP

Expression system Concentration

Saccharomyces cerevisiae (yeast) 40 µg of HPV16

Adjuvant

Amorphous aluminium hydroxyphosphate sulphate 0.5 ml, intramuscular 0, 2 and 6 months 768 vaccinees and 765 placebo U.S.A. 16–23 years HPV16 seronegative and HPV16 DNA negative at enrolment

Baculovirus (insect cells) 20 µg of HPV16 and 20 µg of HPV18 AS04

Merck Research Laboratories Quadrivalent HPV6, 11, 16 and 18 L1 VLP Saccharomyces cerevisiae (yeast) 20 µg of HPV6, 40 µg of HPV11, 40 µg of HPV16 and 20 µg of HPV18 Amorphous aluminium hydroxyphosphate sulphate 0.5 ml, intramuscular 0, 2 and 6 months 276 vaccinees and 275 placebo U.S.A., Europe and Brazil 16–23 years HPV6, 11, 16 and 18 seronegative at enrolment and HPV6, 11, 16 and 18 DNA negative throughout vaccination regimen History of abnormal cervical cytology or five lifetime male sex partners

Dose and administration Dosing schedule Trial population Trial site Age of participants Inclusion criteria

Exclusion criteria

Duration of follow-up Safety profile Vaccine immunogenicity Antibody titres (measured at month 7) compared with natural infection Efficacy in preventing transient infections Efficacy in preventing persistent infections Efficacy in preventing CIN

History of abnormal cervical cytology or > five lifetime male sex partners Up to 48 months; median follow-up, 17.4 months Well tolerated, few minor adverse events; no serious adverse events 100 % seroconversion 60-fold greater

Well tolerated, few minor adverse events; no serious adverse events 100 % seroconversion 100-fold (HPV16) and 80-fold (HPV18) greater respectively

Up to 36 months

91 % (95 % CI, 80–97)

92 % (95 % CI, 65–98)

100 % (95 % CI, 90–100)

100 % (95 % CI, 77–100)

89 % (95 % CI, 70–97)

100 % (ten cases of HPV16+ CIN in placebo group; none in vaccinated group)

100 % (six cases of HPV16 or 18+ CIN in placebo group, none in vaccinated group)

100 % (three cases of HPV6, 11, 16 or 18+ CIN in placebo group; none in vaccinated group)

It is unclear whether increasing the breadth of genotype coverage for prophylactic HPV vaccines will impact on the strength of the resulting immune response against each of the different HPV types. In the second proofof-principle efficacy trial, a bivalent HPV16 and 18 VLP vaccine was compared with placebo in over 1000 women aged between 15 and 25 years over a 27 month follow-up period [45]. All vaccinated women produced HPV16- and 18-specific antibodies at levels much higher than those seen following natural infection. The vaccine gave 100 % protection against persistent HPV16 and 18 2006 The Biochemical Society

History of abnormal cervical cytology/CIN or > six lifetime male sex partners Up to 27 months

Well tolerated, few minor adverse events; no serious adverse events 100 % seroconversion 10-fold (HPV6), 7-fold (HPV11), 100-fold (HPV16) and 20-fold (HPV 18) greater respectively Not reported

Prophylactic bivalent HPV16 and 18 VLP vaccine

C

0.5 ml, intramuscular 0, 1 and 6 months 560 vaccinees and 553 placebo U.S.A., Canada and Brazil 15–25 years HPV16 and 18 seronegative and high-risk HPV DNA negative (14 types) at enrolment

infections and 93 % protection against HPV16- or 18associated cytological abnormalities. Unfortunately, the trial was not powered to estimate efficacy for histopathologically confirmed CIN, with only six cases of CIN (all positive for HPV16 DNA) reported in the placebo group and one (positive for HPV51) in the vaccinated group. As expected, the incidence of HPV18 infection was very low in this study, but by analysing the intention-to-treat cohort, where sufficient numbers of events occurred, the authors were able to conclude that the vaccine is effective in preventing both transient and persistent HPV18 infections. Whether it additionally protects against HPV18-associated CIN remains unclear.


Prophylactic quadrivalent HPV6, 11, 16 and 18 VLP vaccine The most recent prophylactic vaccine efficacy trial compared with placebo a quadrivalent VLP vaccine comprising HPV types 6, 11, 16 and 18 in more than 500 women between the ages of 16 and 23 years [46]. Over the 35-month follow-up period, persistent infection associated with HPV6, 11, 16 or 18 decreased by 89 % in participants who received at least one dose of vaccine compared with those who received placebo. The study was not powered to assess vaccine efficacy for disease end points and only six women developed genital warts or type-specific CIN (all in the placebo group) during follow-up. The vaccine proved to be highly immunogenic, with all patients showing evidence of HPV6-, 11-, 16- and 18-specific antibodies following vaccination. By month 36, however, only 76 % of the vaccinated cohort had significant detectable antibody responses to HPV18. Seropositivity for HPV6, 11 and 16 appeared to last longer. The longevity of protection provided by HPV prophylactic vaccination is clearly an important issue. Since it is neither practical nor desirable to revaccinate frequently, a vaccine that offers an extended period of protection against HPV-associated disease is essential. The quadrivalent VLP vaccine appears to protect not only against high-risk HPV types 16 and 18, but also against low-risk HPV6 and 11. These low-risk genotypes cause 90 % of anogenital warts, a benign but costly condition to treat associated with significant psychological morbidity. If HPV prophylactic vaccination strategies are to be effective at reducing the incidence of cervical cancer, vaccinating males may be critical for the induction of herd immunity, particularly since coverage of female vaccinees is unlikely to reach 100 % [48]. Although high-risk HPV types are uncommonly associated with malignant disease of the anogenital tract in males, it is envisaged that offering the additional benefit of protection against anogenital warts may provide further incentive for vaccination. The value of including men in HPV prophylactic vaccination strategies is still completely unknown, however, since there is currently no evidence that vaccination can prevent infection in men or reduce exposure to women. Overall, the results of these efficacy trials are very encouraging. Implementation of universal HPV prophylactic vaccination requires a detailed understanding of the disease incidence, screening programmes and treatment options available in any given setting, and it is clearly not possible to address all these issues in any one trial. Using a mathematical simulation model, Goldie et al. [49] predicted that a prophylactic vaccine that protects against 70 % of persistent HPV16/18 infections should substantially reduce the incidence of HPV16/ 18-associated CIN and cervical cancer, even in a setting of established cytological screening. Clearly the benefit of such a vaccine to developing countries, where the incidence of cervical cancer is high and diagnostic and

treatment services are poor, should not be underestimated. The logistics of delivering an HPV prophylactic vaccine to the developing world are daunting, not least because the target population of adolescent females is not currently reached by any global health care programme. The vaccine is expensive to produce and purchase, requires a cold chain and must be administered in repeated injected doses. As a result, the financial implications of universal HPV prophylactic vaccination are not insignificant.

THERAPEUTIC HPV VACCINES HPV prophylactic vaccination may not prevent HPVassociated disease in women who have already been exposed to the virus. For these women, a therapeutic vaccine, capable of eradicating established HPV infection, may offer significant advantages over conventional treatment for cancer of the cervix and its precursor lesions. Induction of HPV-specific immunity could allow the selective destruction of HPV-infected cells with minimal consequent damage to normal tissues. Longlasting immunological memory could eliminate recurrence or relapse following treatment and protect against the emergence of HPV-associated lesions elsewhere in the female lower genital tract. Vaccination may also offer renewed hope for those women with advanced disease in whom standard treatment has failed. Therapeutic vaccines aim to induce immunological mechanisms that are capable of recognizing and killing HPV-infected cells. To this end, most therapeutic vaccines have concentrated on eliciting CTL responses against HPV-specific viral antigens. The viral antigens of choice have traditionally been E6 and E7, since their continued expression throughout tumorigenesis is well established [32]. More recently, however, vaccines targeted against the E2 protein have been studied [50].

Antigen delivery An important consideration is how best to deliver the viral antigen(s) to the immune system. In a patient with established HPV-associated disease, it is clear that any existing immune responses have so far been ineffective at clearing the infection. Candidate therapeutic vaccines have used a whole range of antigen-delivery systems to generate sufficiently strong immune responses to eradicate infection.

Peptides and recombinant proteins Peptide- and protein-based therapeutic vaccines are not in themselves immunogenic, but they are cheap to produce and generate predictable immune responses that are relatively straightforward to measure. When formulated with appropriate adjuvants, such as heat-shock proteins or ISCOMATRIX, protein- and peptide-based vaccines exhibit enhanced immunogenicity, but the measured C


549

550


immunity does not always appear to correlate with clinical responsiveness [51,52]. An alternative approach is to load autologous dendritic cells ex vivo with tumour lysate proteins, thus harnessing the natural T-cell activation capabilities of professional antigen-presenting cells [53,54]. This technique has yielded impressive clinical outcome data but, since vaccines must be prepared on an individual basis, it is an extremely expensive and labour-intensive approach.

DNA vaccines DNA vaccines are stable and cheap to manufacture, but do not induce potent anti-tumour immune responses without manipulation. When modified to encode specific adjuvants, such as cytokine genes or molecules involved in antigen processing and presentation, vaccine immunogenicity has been reported but with no reliable association with clinical outcome [55,56]. It should be borne in mind that DNA vaccines are potentially oncogenic and full-length HPV genes must be genetically modified to prevent tumorigenesis at the site of vaccination.

Recombinant viruses Recombinant viral vectors are attractive candidate vaccines for the treatment of HPV-associated disease. An MVA E2 recombinant virus vaccine has recently been tested in 36 patients with CIN yielding promising immunological and clinical results [50]. Another vector, known as the vaccinia virus, induces potent antibody and cell-mediated immune responses, and was successfully used in the immunization programme that resulted in the worldwide eradication of smallpox. A recombinant vaccinia virus, genetically modified to encode HPV16 and 18 E6 and E7 (TA-HPV), is to date the most extensively studied HPV therapeutic vaccine. Clinical trials in patients with advanced and early stage cervical carcinoma [57,58], as well as those with high-grade lower genital tract neoplasia [59–61], have established safety, immunogenicity and clinical responsiveness in a proportion of vaccinated patients. Unfortunately, measured HPV-specific immune responses do not appear to correlate strongly with clinical outcome following vaccination.

Prime-boost vaccination strategies Preclinical studies have indicated that a prime-boost vaccination strategy employing heterologous vaccines may offer advantages over the use of either agent alone in the treatment of HPV-associated disease [62]. A fusion protein of HPV16 L2E6E7 (TA-CIN) has recently been used in combination with TA-HPV in women with high-grade lower genital tract neoplasia [63]. The vaccination schedule demonstrated safety, immunogenicity and clinical responsiveness in a proportion of the patients [63]. C


CHALLENGES IN THERAPEUTIC HPV VACCINE DESIGN Several candidate therapeutic vaccines for HPV-associated disease have now been tested in early phase clinical trials. Vaccine safety has been confirmed, but immunogenicity and clinical efficacy have often been difficult to establish. In particular, there appears to be no consistent relationship between measured HPV-specific immune responses and clinical outcome following vaccination. Correlating HPV-specific immunity with clinical responsiveness is difficult if we are unsure what immune mechanisms are important and how best to measure them. Papillomaviruses have co-evolved with mankind over millions of years and owe much of their success to their ability to frustrate the natural immunity of the host. If the virus has persisted unnoticed in the infected tissue for long enough to cause neoplastic change, it is clear that a therapeutic vaccine must be extremely immunogenic to override these sophisticated mechanisms of immune escape. Moreover, as disease progresses to cancer, there is an increasing likelihood of systemic immunosuppression, and this will impact significantly on the expected success of therapeutic vaccination. Tumour cells often exhibit MHC class I down-regulation, which restricts the ability of the immune system to recognize and destroy abnormal cells [64]. Thus it may be more effective to focus therapeutic vaccines at women with persistent HPV infection or low-grade CIN, rather than those with advanced disease. Despite continued focussed research, therapeutic HPV vaccination may still be several years from realization. Ultimately, a vaccine that combines prophylactic and therapeutic modalities may provide the best solution for HPV-associated disease. A chimaeric HPV16 VLP– E7 vaccine has been constructed by Greenstone et al. [65], which induces E7-specific T-cell responses as well as virus-neutralizing antibodies in the murine model. If effective in humans, this type of vaccine may allow women who are na¨ıve to the virus, as well as those who have already been exposed, to benefit from reduced rates of cervical cancer.

CONCLUSIONS HPV testing has yet to find its full role in cervical screening, but it threatens the pre-eminence of exfoliative cytology as the primary screening test. Testing for HPV is simpler than cytology, but its lack of specificity will require a second test and possibly repeat testing if the second test is negative. Population testing using HPV will require public education programmes and training for healthcare professionals. Prophylactic vaccination promises a great deal, but in developed countries with effective screening programmes the introduction of vaccines in pre-adolescents would require continuation of


screening for a generation of unvaccinated females. Even vaccinated women may require a degree of screening and this will require careful planning. If a programme of vaccination for all females, at least, were to be introduced it would need to be demonstrably cost effective and would require broad acceptance by society. If it could be implemented in developing countries it could be of enormous benefit and save hundreds of thousands of lives.

REFERENCES 1 Cancer Research UK CancerStats (http:// info.cancerresearchuk.org/cancerstats/types/cervix/ mortality) 2 Sasieni, P., Adams, J. and Cuzick, J. (2003) Benefit of cervical screening at different ages: evidence from the UK audit of screening histories. Br. J. Cancer 89, 88–93 3 Chan, S. Y., Bernard, H. U., Ong, C. K. et al. (1992) Phylogenetic analysis of 48 papillomavirus types and 28 subtypes and variants: a showcase for the molecular evolution of DNA viruses. J. Virol. 66, 5714–5725 4 Stanley, M. A. (1994) Virus-keratinocyte interactions in the infectious cycle. In Human Papillomaviruses and Cervical Cancer (Stern, P. L. and Stanley, M. A., eds.), pp. 116–131, Oxford University Press, Oxford 5 Sedman, J. and Stenlund, A. (1995) Co-operative interaction between the initiator E1 and the transcriptional activator E2 is required for replicator specific DNA replication of bovine papillomavirus in vivo and in vitro. EMBO J. 14, 6218–6228 6 McDougall, J. K. (1996) HPV immortalisation and transformation. In Papillomavirus Reviews: Current Research on Papillomaviruses (Lacey, C., ed.), pp. 47–53, Leeds University Press, Leeds 6a Doorbar, J. (2006) Molecular biology of human papillomavirus infection and cervical cancer. Clin. Sci. 110, 525–541 7 Walboomers, J. M., Jacobs, M. V., Manos, M. M. et al. (1999) Human papillomavirus is a necessary cause of invasive cervical cancer worldwide. J. Pathol. 189, 12–19 8 Nobbenhuis, M. A., Walboomers, J. M., Helmershorst, T. J. et al. (1999) Relation of human papillomavirus status to cervical lesions and consequences for cervical-cancer screening: a prospective study. Lancet 324, 20–25 9 Kjaer, S. K., Van den Brule, A. J., Paull, G. et al. (2002) Type specific persistence of high risk human papillomavirus (HPV) as indicator of high grade cervical squamous intraepithelial lesions in young women: population based prospective follow up study. Br. Med. J. 325, 572 10 Kitchener, H., Almonte, M., Wheeler, P. et al. (2005) The ARTISTIC Trial of HPV testing in primary cervical screening: methods and baseline data, 22nd International Papillomavirus Conference, Vancouver, C-07 11 Clavel, C., Masure, M., Bory, J. P. et al. (2001) Human papillomavirus testing in primary screening for the detection of high-grade cervical lesions: a study of 7932 women. Br. J. Cancer 84, 1616–1623 12 Cuzick, J., Szarewski, A., Terry, G. et al. (1995) Human papillomavirus testing in primary cervical screening. Lancet 345, 1533–1536 13 Schiffman, M., Herrero, R., Hildesheim, A. et al. (2000) DNA testing in cervical cancer screening: results from women in a high-risk province of Costa Rica. JAMA, J. Am. Med. Assoc. 283, 87–93 14 Belinson, J. L., Qiao, Y. L., Pretorius, R. G. et al. (2003) Shanzi Province cervical cancer screening study II: self-sampling for high-risk human papillomavirus compared to direct sampling for human papillomavirus and liquid based cervical cytology. Int. J. Gynecol. Cancer 13, 819–826

15 Waller, J., McCaffery, K. J., Forrest, S. and Wardle, J. (2004) Human papillomavirus and cervical cancer: issues for biobehavioral and psychosocial research. Ann. Behav. Med. 27, 68–79 16 Cuzick, J., Szarewski, A., Cubie, H. et al. (2003) Management of women who test positive for high-risk types of human papillomavirus: the HART study. Lancet 362, 1871–1876 17 Flannelly, G., Anderston, D., Kitchener, H. C. et al. (1994) Management of women with mild and moderate cervical dyskaryosis. Br. Med. J. 308, 1399–1403 18 Ho, G. Y., Bierman, R., Beardsley, L. et al. (1998) Natural history of cervicovaginal papillomavirus infection in young women. N. Eng. J. Med. 338, 423–428 19 Remmink, A. J., Walboomers, J. M., Helmerhorst, T. J. et al. (1995) The presence of persistent high-risk HPV genotypes in dysplastic cervical lesions is associated with progressive disease: natural history up to 36 months. Int. J. Cancer 61, 306–311 20 Ferenczy, A., Coutlee, F., Franco, E. and Hankins, C. (2003) Human papillomavirus and HIV coinfection and the risk of neoplasias of the lower genital tract: a review of recent developments. Can. Med. Assoc. J. 169, 431–434 21 Ponten, J., Adami, H. O., Friberg, L. G. et al. (1995) HPV and cervical cancer. Int. J. Cancer 63, 317 22 Man, S. (1998) Human cellular immune responses against human papillomaviruses in cervical neoplasia. Expert Rev. Mol. Med. 1, 1–19 23 Kadish, A. S., Ho, G. Y., Burk, R. D. et al. (1997) Lymphoproliferative responses to human papillomavirus (HPV) type 16 proteins E6 and E7: outcome of HPV infection and associated neoplasia. J. Natl. Cancer Inst. 89, 1285–1293 24 Welters, M. J., de Jong, A., van den Eeden, S. J. et al. (2003) Frequent display of human papillomavirus type 16 E6-specific memory t-Helper cells in the healthy population as witness of previous viral encounter. Cancer Res. 63, 636–641 25 de Jong, A., van Poelgeest, M. I., van der Hulst, J. M. et al. (2004) Human papillomavirus type 16-positive cervical cancer is associated with impaired CD4+ T-cell immunity against early antigens E2 and E6. Cancer Res. 64, 5449–5455 26 Nakagawa, M., Stites, D. P., Patel, S. et al. (2000) Persistence of human papillomavirus type 16 infection is associated with lack of cytotoxic T lymphocyte response to the E6 antigens. J. Infect. Dis. 182, 595–598 27 Bontkes, H. J., de Gruijl, T. D., van den Muysenberg, A. J. et al. (2000) Human papillomavirus type 16 E6/E7-specific cytotoxic T lymphocytes in women with cervical neoplasia. Int. J. Cancer 88, 92–98 28 de Gruijl, T. D., Bontkes, H. J., Walboomers, J. M. et al. (1998) Differential T helper cell responses to human papillomavirus type 16 E7 related to viral clearance or persistence in patients with cervical neoplasia: a longitudinal study. Cancer Res. 58, 1700–1706 29 van der Burg, S. H., Ressing, M. E., Kwappenberg, K. M. et al. (2001) Natural T-helper immunity against human papillomavirus type 16 (HPV16) E7-derived peptide epitopes in patients with HPV16-positive cervical lesions: identification of 3 human leukocyte antigen class II-restricted epitopes. Int. J. Cancer 91, 612–618 30 Youde, S. J., Dunbar, P. R., Evans, E. M. et al. (2000) Use of fluorogenic histocompatibility leukocyte antigenA∗ 0201/HPV 16 E7 peptide complexes to isolate rare human cytotoxic T-lymphocyte-recognizing endogenous human papillomavirus antigens. Cancer Res. 60, 365–371 31 Evans, E.-M., Man, S., Evans, A. S. et al. (1997) Infiltration of cervical cancer tissue with human papillomavirusspecific cytotoxic T lymphocytes. Cancer Res. 57, 2943–2950 32 Stanley, M. A. (2001) Human papillomavirus and cervical carcinogenesis. Best Pract. Res. Clin. Obstet. Gynaecol. 15, 663–676 33 de Jong, A., van der Burg, S. H., Kwappenberg, K. M. et al. (2002) Frequent detection of human papillomavirus 16 E2-specific T-helper immunity in healthy subjects. Cancer Res. 62, 472–479

C


551

552


34 Davidson, E. J., Sehr, P., Faulkner, R. L. et al. (2003) Human papillomavirus type 16 E2- and L1-specific serological and T-cell responses in women with vulval intraepithelial neoplasia. J. Gen. Virol. 84, 2089–2097 35 Carter, J. J., Koutsky, L. A., Wipf, G. C. et al. (1996) The natural history of human papillomavirus type 16 capsid antibodies among a cohort of university women. J. Infect. Dis. 174, 927–936 36 Wikstrom, A., Vandoornum, G. J. J., Quint, W. G. V. et al. (1995) Identification of human papillomavirus seroconversions. J. Gen. Virol. 76, 529–539 37 Ho, G. Y., Studentsov, Y. Y., Bierman, R. et al. (2004) Natural history of human papillomavirus type 16 virus-like particle antibodies in young women. Cancer Epidemiol. Biomarkers Prev. 13, 110–116 38 Garland, S. M. (2002) Human papillomavirus update with a particular focus on cervical disease. Pathology 34, 213–224 39 Viscidi, R. P., Schiffman, M., Hildesheim, A. et al. (2004) Seroreactivity to human papillomavirus (HPV) types 16, 18, or 31 and risk of subsequent HPV infection: results from a population-based study in Costa Rica. Cancer Epidemiol. Biomarkers Prev. 13, 324–327 40 Schiller, J. T. and Lowy, D. R. (2001) Papillomavirus-like particle based vaccines: cervical cancer and beyond. Expert Opin. Biol. Ther. 1, 571–581 41 Kirnbauer, R., Booy, F., Cheng, N. et al. (1992) Papillomavirus L1 major capsid protein self-assembles into virus-like particles that are highly immunogenic. Proc. Natl. Acad. Sci. U.S.A. 89, 12180–12184 42 Campo, M. S. (2002) Animal models of papillomavirus pathogenesis. Virus Res. 89, 249–261 43 Harro, C., Pang, Y., Roden, R. et al. (2001) Safety and immunogenicity trial in adult volunteers of a human papillomavirus 16 L1 virus-like particle vaccine. J. Natl. Cancer Inst. 93, 284–292 44 Koutsky, L. A., Ault, K. A., Wheeler, C. M. et al. (2002) A controlled trial of a human papillomavirus type 16 vaccine. N. Engl. J. Med. 347, 1645–1651 45 Harper, D. M., Franco, E. L., Wheeler, C. et al. (2004) Efficacy of a bivalent L1 virus-like particle vaccine in prevention of infection with human papillomavirus types 16 and 18 in young women: a randomised controlled trial. Lancet 364, 1757–1765 46 Villa, L. L., Costa, R. L. R., Petta, C. A. et al. (2005) Prophylactic quadrivalent human papillomavirus (types 6, 11, 16, and 18) L1 virus-like particle vaccine in young women: a randomised double-blind placebo-controlled multicentre phase II efficacy trial. Lancet Oncol. 6, 271–278 47 Bosch, F. X., Manos, M. M., Munoz, N. et al. (1995) Prevalence of human papillomavirus in cervical cancer: a worldwide perspective. International biological study on cervical cancer (IBSCC) Study Group. J. Natl. Cancer Inst. 87, 796–802 48 Hughes, J. P., Garnett, G. P. and Koutsky, L. (2002) The theoretical population-level impact of a prophylactic human papillomavirus vaccine. Epidemiology 13, 631–639 49 Goldie, S. J., Kohli, M., Grima, D. et al. (2004) Projected clinical benefits and cost-effectiveness of a human papillomavirus 16/18 vaccine. J. Natl. Cancer Inst. 96, 604–615 50 Corona Gutierrez, C. M., Tinoco, A., Navarro, T. et al. (2004) Therapeutic vaccination with MVA E2 can eliminate precancerous lesions (CIN 1, CIN 2, and CIN 3) associated with infection by oncogenic human papillomavirus. Hum. Gene Ther. 15, 421–431

51 Goldstone, S. E., Palefsky, J. M., Winnett, M. T. et al. (2002) Activity of HspE7, a novel immunotherapy, in patients with anogenital warts. Dis. Colon Rectum 45, 502–507 52 Frazer, I. H., Quinn, M., Nicklin, J. L. et al. (2004) Phase 1 study of HPV16-specific immunotherapy with E6E7 fusion protein and ISCOMATRIX adjuvant in women with cervical intraepithelial neoplasia. Vaccine 23, 172–181 53 Adams, M., Navabi, H., Jasani, B. et al. (2003) Dendritic cell (DC) based therapy for cervical cancer: use of DC pulsed with tumour lysate and matured with a novel synthetic clinically non-toxic double stranded RNA analogue poly [I]:poly [C12 U] (Ampligen R). Vaccine 21, 787–790 54 Santin, A. D., Bellone, S., Gokden, M. et al. (2002) Vaccination with HPV-18 E7-pulsed dendritic cells in a patient with metastatic cervical cancer. N. Eng. J. Med. 346, 1752–1753 55 Klencke, B., Matijevic, M., Urban, R. G. et al. (2002) Encapsulated plasmid DNA treatment for human papillomavirus 16-associated anal dysplasia: a Phase I study of ZYC101. Clin. Cancer Res. 8, 1028–1037 56 Garcia, F., Petry, K. U., Muderspach, L. et al. (2004) ZYC101a for treatment of high-grade cervical intraepithelial neoplasia: a randomized controlled trial. Obstet. Gynecol. 103, 317–326 57 Borysiewicz, L. K., Fiander, A., Nimako, M. et al. (1996) A recombinant vaccinia virus encoding human papillomavirus types 16 and 18, E6 and E7 proteins as immunotherapy for cervical cancer. Lancet 347, 1523–1527 58 Kaufmann, A. M., Stern, P. L., Rankin, E. M. et al. (2002) Safety and immunogenicity of TA-HPV, a recombinant vaccinia virus expressing modified human papillomavirus (HPV)-16 and HPV-18 E6 and E7 genes, in women with progressive cervical cancer. Clin. Cancer Res. 8, 3676–3685 59 Davidson, E. J., Boswell, C. M., Sehr, P. et al. (2003) Immunological and clinical responses in women with vulval intraepithelial neoplasia vaccinated with a vaccinia virus encoding human papillomavirus 16/18 oncoproteins. Cancer Res. 63, 6032–6041 60 Baldwin, P. J., van der Burg, S. H., Boswell, C. M. et al. (2003) Vaccinia-expressed human papillomavirus 16 and 18 e6 and e7 as a therapeutic vaccination for vulval and vaginal intraepithelial neoplasia. Clin. Cancer Res. 9, 5205–5213 61 Smith, K. L., Tristram, A., Gallagher, K. M. et al. (2005) Epitope specificity and longevity of a vaccine-induced human T cell response against HPV18. Int. Immunol. 17, 167–176 62 Schneider, J., Gilbert, S. C., Hannan, C. M. et al. (1999) Induction of CD8+ T cells using heterologous prime-boost immunisation strategies. Immunol. Rev. 170, 29–38 63 Smyth, L. J., Van Poelgeest, M. I., Davidson, E. J. et al. (2004) Immunological responses in women with human papillomavirus type 16 (HPV-16)-associated anogenital intraepithelial neoplasia induced by heterologous prime-boost HPV-16 oncogene vaccination. Clin. Cancer Res. 10, 2954–2961 64 Keating, P. J., Cromme, F. V., Duggan-Keen, M. et al. (1995) Frequency of down-regulation of individual HLA-A and -B alleles in cervical carcinomas in relation to TAP-1 expression. Br. J. Cancer 72, 405–411 65 Greenstone, H. L., Nieland, J. D., deVisser, K. E. et al. (1998) Chimeric papillomavirus virus-like particles elicit anti-tumour immunity against the E7 oncoprotein in an HPV16 tumour model. Proc. Natl. Acad. Sci. U.S.A. 95, 1800–1805 Received 26 July 2005/25 October 2005; accepted 8 November 2005 Published on the Internet 11 April 2006, doi:10.1042/CS20050230

C
