Abstract
There has been considerable progress in the development of a prophylactic human papillomavirus vaccine in the past 10 years, since the discovery of human papillomavirus virus-like particles. Licensure of the human papillomavirus vaccine is probably not far away. This would make it the first licensed vaccine against a common sexually transmitted infection. Although hepatitis B is a sexually transmitted infection for which there is an effective prophylactic vaccine, it is often not perceived as such by individuals taking the vaccine. Preclinical studies have already produced attractive vaccine candidates and recent clinical trials have yielded strikingly promising results. The candidate vaccines are generally well tolerated, induce high titers of serum antibodies to the human papillomavirus types and effectively prevent acquisition of infection and early clinical disease caused by common human papillomavirus types.
Investigators in Germany first suggested a role for high-risk human papillomavirus (HPV) types in the etiology of cervical cancer 30 years ago [1]. It took almost 10 years until HPV type 16 was discovered in cervical cancer tissue [2]. It took another 10 years to show that past infection with HPV-16 increases the risk of subsequent development of invasive cervical cancer [3], and yet another decade to show that the seven most prevalent HPV types cause 87% of all cervical cancers [4,5]. Many other high-risk HPV types have subsequently been identified and linked to cervical cancer [6]. Sexually transmitted HPV infections are much more prevalent than sexually transmitted chlamydia or herpes simplex virus infections. All of these sexually transmitted infections (STIs) cause subclinical or asymptomatic infections that are difficult to control by traditional prevention methods. More than 100 types of HPVs have been identified, of which at least a third have been detected in the anogenital tract. HPVs are broadly grouped into low-risk or nononcogenic types (such as types 6 and 11), which are associated with benign anogenital warts, and high-risk, or oncogenic HPV types (such as types 16, 18, 31, 33, 35, 45, 52, and 58), which are associated with cervical cancer [6].
A small proportion of HPV infections caused by high-risk HPV types persist and cause cervical cytologic atypia, which can progress through cervical intraepithelial neoplasia to invasive cervical cancer [7]. HPV is the leading cause of invasive cervical cancer worldwide. Combining data from more than 20 countries, the worldwide HPV prevalence of cervical carcinomas was found to be 99.7% [4,5]. The presence of HPV DNA in virtually all cervical cancers implies the highest worldwide attributable fraction so far reported for a specific cause of any major human cancer.
Studies now suggest that a highly efficacious prophylactic vaccine against HPV infection will be available soon [8–10]. Vaccines have been a cost-effective way to prevent viral diseases. Candidate vaccines for HPV consist of virus-like particles (VLPs) generated by recombinant expression of the major capsid protein, L1, in an expression vector system. The L1 VLP is an empty capsid (i.e., it contains no DNA) that has the correct conformation and seems to have identical morphology as, and contains the major neutralizing epitopes of, the native virion. Thus, the future for a HPV vaccine looks bright.
Prevalence
The lifetime risk for HPV infection is extremely high. Most individuals are exposed to HPV shortly after their sexual debut [11]. The cumulative rate of cervical HPV infection is approximately 50% 3 years the first sexual encounter. The authors' recent prevalence study in a student health clinic population among first-year students showed that 33% were HPV DNA positive [12]. Of those positive, 85% were positive for the high-risk HPV types. The high attack rate can be explained by risk-taking behavior, which has increased among young individuals. For instance, a survey of 14-year-old adolescent women in Finland found that 18% have already had sexual intercourse, and this increases to 59% by the age of 16 [13]. Similarly, since the 1970s, the lifetime number of sex partners has tripled and the age at which sexual intercourse first takes place has decreased significantly. Similar data has recently been reported from the UK [14].
Natural history
It is important to emphasize that most HPV infections are transient and that the overall prevalence among women decreases with increasing age. Up to 80% of HPV infections spontaneously regress within 2 years [15,16]. Risk for cervical neoplasia is associated with persistent infection with high-risk HPV types [17]. It is becoming increasingly clear that cytologic atypia consistent with a low-grade squamous intraepithelial lesion (SIL) is also a transient manifestation of HPV infection in young women that will only rarely progress to high-grade SILs. In one recent study the probability of regression was 61% after 12 months and 91% after 36 months [18].
Disease burden
The global estimate of sexually transmitted HPV infections is more than that of sexually transmitted chlamydial infections. HPV disease burden is not limited to cervical cancer alone. A substantial proportion of other anogenital cancers including anal, vulvar, vaginal and penile cancer are also strongly associated with high-risk HPV types [19,20]. A smaller proportion of oropharyngeal and laryngeal carcinomas have been linked to HPV [21].
Globally, cervical cancer is the second most common cancer in women with approximately half a million new cases every year. Approximately 80% of all cases occur in developing countries [22]. The highest incidence rates are now observed in sub-Saharan Africa, Melanesia, Latin America and the Caribbean, South-Central Asia and South-East Asia [22]. Although cytologic mass screening programs have been very effective in the prevention of cervical cancer in developed countries, the mortality from this cancer is still extremely high in many parts of the world. Finland probably has the most effective cervical cancer prevention program, based on the mass screening which started in 1963. Since then, the rate of cervical cancer has decreased by 80%. However, the incidence of cervical cancer has rapidly increased among young women during the past 10 years, which is alarming [23]. This is in line with an increasing attack rate of HPV-16 infection based on seroepidemiologic studies [24,25]. Thus, the increasing incidence of cervical cancer among young women is not totally unexpected.
HPV VLP vaccine
A major advance has been the success of the HPV vaccine. A HPV vaccine raises the prospect of the prevention of cervical cancer globally within the foreseeable future. The major breakthrough in prophylactic vaccine development was the discovery that L1 can self-assemble into VLPs when expressed in mammalian cells, insect cells, yeast or even bacteria [26–28]. The creation of VLP vaccines has been a rapid breakthrough. VLPs mimic the true structure of the virion and induce a striking antibody response after vaccination. VLPs seem to work by eliciting high titers of neutralizing serum antibody, but whether this is the actual mechanism by which the vaccine protects against infection is as yet unknown. Nevertheless, concentrations of serum antibody to L1 and the persistence of the antibody response might be crucial for measuring the amount of protection. Passive transfer of immune serum in animal model studies clearly demonstrates that antibodies are at least sufficient to induce protection from an experimental challenge. VLP vaccines may only provide type-specific protection, and it may be that the number of HPV types in the vaccine needs to be increased to prevent the majority of cervical cancers. Whether such a polyvalent vaccine would result in immunologic equivalence, so that each VLP component induces a protective antibody response, is unclear thus far.
Early vaccination trials
Phase I and II trials of a HPV VLP vaccine have demonstrated that the vaccine is highly immunogenic and induces high titers of neutralizing antibodies, 50- to 100-times higher than in natural HPV infection [29]. The vaccine is also well tolerated. The local or systemic adverse effects are not significantly higher than those with placebo. Dose-escalation studies have demonstrated that the three-dose schedule, at either 0, 1 and 6 months or 0, 2, 6 months, is most effective. Practically all vaccinated individuals seroconvert. Phase IIb clinical trials have shown that the VLP vaccine is highly effective against HPV infections. As shown in Table 1, the vaccine efficacy was 100% against persistent HPV-16 or persistent HPV-16/18 infection. The vaccine efficacy was 91–92% against any HPV-16 or HPV-16/18 infection [8,9]. It is important to point out that the presence of HPV DNA was detected in these studies. This could represent incident infection, but could also present the possibility of contamination from a partner or reactivation of latent infection. The cytologic end points used by Harper and colleagues represent the clinical manifestations of infections with the oncogenic HPVs. It is encouraging that the bivalent vaccine also protects against these cytologic abnormalities and cervical intraepithelial neoplasia [9]. However, long-term passive follow-up of cohorts of vaccinated and nonvaccinated patients based on population-based cancer registries is needed to prove that HPV vaccination ultimately protects against invasive cervical cancer [30,31].
Summary of early results in HPV vaccination trials.
HPV: Human papillomavirus; VLP: Virus-like particles.
Koutsky, Ault, Wheeler et al.: HPV 16 VLP vaccine p < 001 [8].
Harper, Franco, Wheeler et al.: Bivalent HPV VLP vaccine (HPV 16/ 18); p < .001 [9].
Villa and colleagues presented the results of a double-blind, placebo-controlled efficacy trial of a quadrivalent (HPV types 6, 11, 16 and 18), aluminum-adjuvant vaccine in young women negative for HPV [10]. In this study, concentrations of serum antibody to L1 in those assigned the vaccine were much the same for types 6, 11 and 18, but antibody concentrations against type 16 were up to 1 log higher at both 7 and 36 months after vaccination. As in previous reports, peak antibody concentrations were much higher in vaccinated individuals than in seropositive nonvaccinated individuals, and these concentrations remained higher 36 months after vaccination. Despite the difference in antibody concentrations, patients were protected against persistent infection with HPV-16 and −18, confirming data from previous trials [8,9].
Phase III vaccination trials
Phase III clinical efficacy trials for the HPV vaccine are already underway. Multinational multicenter vaccination trials targeting 11,500–18,000 individuals to be followed for 4 years were initiated in 2002–2004, either with the quadrivalent HPV VLP vaccine (HPV-6, −11, −16 and −18) by Merck and Co., Inc., or the bivalent HPV VLP vaccine (HPV-16, −18) by GlaxoSmithKline Biologicals. The two companies are also using different adjuvants in their vaccines. In Finland, adolescent women aged 16–17 years have been enrolled, which has been extremely successful, with a total of 7000 adolescents enrolled by the end of June 2005 [31]. Both the vaccine cohorts and the unvaccinated reference cohorts are population based and will be followed for 4 years. The objectives of the Phase III efficacy trials are to determine whether the vaccine reduces the incidence of persistent HPV infection with each of the HPV types included in the vaccine, and whether the vaccine prevents the development of high-grade cervical intraepithelial neoplasia.
Unanswered questions
While the early phase clinical trials have been very encouraging, several important research questions remain unanswered. Selected questions have been listed in Box 1.
A HPV vaccine may have potential problems [32,33]. One is so-called HPV type replacement, which involves the resurgence of HPV types not included in the preventive vaccine. This remains a potential risk, although it has yet to be proven. In addition, a STI vaccine might lead to increased risk-taking behavior among adolescents. Finally, the vaccine-induced protection probably wanes over time and booster vaccinations may be needed.
The role of herd immunity in determining the effect of vaccines against HPV infection has recently been emphasized [33]. The effectiveness of vaccination depends both on HPV prevalence in the target population and vaccine coverage [30,33]. The effectiveness of the vaccination on HPV prevalence depends on whether it is effective in decreasing transmission among women with prevalent infection at the time of vaccination. If vaccine coverage is low, vaccination of both sexes would be more effective. In addition, vaccination should be administered prior to the start of sexual activity before a risk of exposure to HPV exists. To maximize the reduction in HPV cases, the VLPs of more HPV types should probably be introduced into the vaccine cocktail, matched to local distributions of HPV types. The presence of multiple types of HPV raises some fundamental questions about the population biology of HPV infection and the answers to these questions will have profound implications for the use of a vaccine. A vaccine that would generate cross-immunity against other high-risk HPV types would be of great advantage. However, in settings of limited resources, one could argue that it is more important to maximize the number of women receiving a HPV-16/18 vaccine than to maximize the additional types of HPV in the vaccine.
Selected research questions
What is the duration of the human papillomavirus (HPV) virus-like particles (VLPs) antibody response?
Is the protection induced by the VLP vaccine type-specific only?
Is a multivalent VLP vaccine required?
Does serum immnoglobulin (Ig)G antibody alone protect against cervical HPV infection?
Does the vaccine induce a genital tract secretory IgA antibody response?
Is exudated serum IgG enough to protect against HPV infection when epithelial trauma occurs?
Is parenteral intramuscular vaccination more effective than mucosal application of the vaccine?
How many HPV VLP types should be included in the vaccine to prevent at least 80% of cervical cancers worldwide?
Should both men and women be vaccinated?
Specific antibody levels found in the cervix during the menstrual cycle of women vaccinated with the HPV-16 VLP vaccine demonstrate that the cervical antibody titers vary during the menstrual cycle [34]. Antibody titers are highest during the proliferative phase, decreasing by approximately tenfold around ovulation and increasing again during the luteal phase. The decrease in antibody titers around ovulation raises the possibility that the HPV-16 VLP vaccine might be less effective during the periovulatory phase among women not taking oral contraceptives. This observation suggests that endogenous sex hormones are responsible for regulating antibody concentration in the female genital tract. Therefore, further studies are needed to find out whether ovulating women who are not taking oral contraceptives are more susceptible to genital HPV infection than women taking oral contraceptives.
Although neutralizing antibodies have an important role in the prevention of initial HPV infection, cellular immune response to HPV may also have an important role in viral clearance. Another Phase II trial has evaluated the cell-mediated immune responses to HPV-16 L1 VLP vaccine in peripheral blood mononuclear cells [35]. This study suggests that the HPV-16 vaccine induces not only B-cell responses but also T-cell responses detectable by proliferation of both CD4+ and CD8+ T-cells and in vitro production of both T-helper1-and T-helper2-type cytokines. This cell-mediated immune response may be an important second line of defense against HPV when it has already passed the first line of defense provided by neutralizing antibodies.
Other vaccine-specific issues
Vaccine implementation issues for developing countries are of paramount importance. Up to 80% of the global burden of cervical cancer occurs in developing countries where papilloma (Pap) screening programs are not available or have been largely ineffective. Ongoing Phase III trials of prophylactic HPV vaccines are extremely complex and expensive. Similar trials are difficult to accomplish in developing countries. However, it is important to start simpler trials designed to demonstrate the effectiveness of the HPV vaccine in field conditions in developing countries. Undoubtedly, it will take a great effort from international bodies, manufacturers and specific countries to ultimately make the vaccine available in the developing world, which suffers the greatest burden of mortality from cervical cancer. An important related issue is the appropriate coordination of strategies and development of algorithms for co-existing primary prevention programs, for instance, vaccination, and secondary prevention programs, for example, HPV DNA screening or Pap screening, or both, in different countries. The optimal combination of preventive vaccination and different screening activities should be defined in Phase IV trials in communities where vaccine(s) will be available. Furthermore, the effect of so-called once in a lifetime HPV DNA screening among women over 30 years of age needs to be evaluated in combination with conventional cytologic screening programs and HPV vaccination. The recently published interim guidance for the use of HPV DNA testing as an adjunct to cervical cytology for screening will be helpful in this respect [36]. It is clear that screening programs will also play a significant role in the vaccine era. It is highly unlikely that Pap screening programs will become redundant in the foreseeable future unless the vaccine is highly effective and coverage is high [30].
Executive summary
There has been considerable progress in the development of a prophylactic human papillomavirus (HPV) vaccine since the discovery of HPV virus-like particles (VLPs).
Phase I and II trials of a HPV VLP vaccine have demonstrated that the vaccine is highly immunogenic and well tolerated.
Phase IIb clinical trials have shown that the vaccine is 100% effective against persistent HPV-16 or HPV-16/18 infection and 91–92% effective against any HPV-16 or HPV-16/18 infection.
The vaccine also protects against cytologic abnormalities and cervical intraepithelial neoplasia associated with HPV-16/18.
Large Phase III clinical efficacy trials of prophylactic HPV vaccines are already underway.
Phase III vaccination trials can only provide an answer to the question of whether the vaccine protects against HPV infection or high-grade cervical intraepithelial neoplasia (a surrogate marker for cervical cancer).
Therefore, these trials should be extended to Phase IV trials based on cancer registries in countries with effective cervical cancer screening programs.
HPV vaccine efficacy will ultimately be tested in developing countries where morbidity and mortality from cervical cancer is highest.
Vaccine implementation in developing countries remains a major public health challenge to international bodies, manufacturers and specific countries.
Universal HPV vaccination depends not only on vaccine efficacy but also on vaccine coverage.
Second generation HPV vaccines acting both therapeutically and prophylactically are currently under development.
Modeling studies suggest that vaccination of girls against high-risk HPV is cost effective even when vaccine efficacy is low [37]. Even though gains in life expectancy remain modest at the individual level, population benefits will be substantial. Thus, the HPV vaccine will be cost effective when compared with many other generally accepted health interventions, since it would remove a substantial burden from the healthcare system by preventing morbidity and mortality caused by cervical cancer and reducing the frequency of abnormal Pap smears. At present, healthcare workers expend considerable energy in their efforts to prevent cervical cancer. A vaccine that prevents persistent HPV-16/18 infection will reduce the incidence of cervical cancer associated with HPV-16/18 even in settings with cytologic screening programs. A vaccination program that permits a later age of screening initiation and a less frequent screening interval is likely to be a highly cost-effective use of healthcare resources [38].
HPV vaccine acceptability has been studied among parents of 10–15-year-old adolescents, among college students and among young women in general. All published studies suggest that vaccine acceptability is extremely high and that the vaccine would be well received among young adults [39–41]. These results are in agreement with our recent experience with enrolment of 16–17-year-old adolescent women in Phase III vaccination trials. It is unlikely that poor public attitude concerning HPV vaccines will be an obstacle.
Only prophylactic HPV vaccines are currently undergoing clinical evaluation in Phase III trials. These vaccines might meet the needs of developed countries reasonably well. However, they have fundamental weaknesses for widespread distribution in developing countries. VLP-based vaccines are expensive to make, expensive to distribute and protection will almost certainly be type-specific, which would not be expected to protect against cervical cancers caused by other high-risk HPV types. These VLP vaccines are not expected to induce regression of established HPV disease. Therefore, the public health benefit of VLP vaccines will be substantially delayed. So-called second-generation HPV vaccines are currently under development. These vaccines should be inexpensive to manufacture and distribute and should protect against all oncogenic types, acting both therapeutically and prophylactically. However, there are still many obstacles to the successful development of such second-generation HPV vaccines.
Conclusion
Since the ongoing Phase III vaccination trials can only provide an answer to the question of whether the vaccine protects against HPV infection or high-grade cervical inraepithelial neoplasia (a surrogate marker for cervical cancer), these trials should be extended to Phase IV trials in order to answer the question of whether the vaccine protects against cervical cancer [30]. HPV vaccination also shows great promise for the control of cervical cancer in developing countries that cannot be reached by cytologic mass screening programs or HPV DNA testing. An advance such as an effective vaccine against HPV would be a major breakthrough. Universal HPV vaccination might be most effective when implemented in young individuals who are likely to be HPV-negative. This could be the beginning of the end for cervical cancer.
Future perspective
Although prophylactic HPV VLP vaccines are extremely promising, questions regarding true vaccine efficacy can only be answered based on Phase IV trials using cancer registries in countries with effective cervical cancer screening programs. However, HPV vaccine efficacy will ultimately be tested in developing countries where morbidity and mortality from cervical cancer is highest. Vaccine implementation in developing countries remains a major public health challenge. Prophylactic and therapeutic second-generation HPV vaccines are already under development. Cervical cancer screening programs will also play a significant role in cancer prevention in the vaccine era.