After decades of epidemiologic and laboratory research to identify the sexually
transmitted pathogen associated with cervical cancer, most experts agree that infection
with certain types of human papillomavirus (HPV) causes cervical cancer. Over 30 different
types of HPV are sexually transmitted and infect the genital area. Eighty percent of
cervical cancers contain HPV types 16, 18, 31, or 45. Genital HPV types 6 and 11 causes
genital warts and low-grade Pap smear abnormalities. Estimated prevalence of genital HPV
infection is 45 million, and estimated incidence is 5,000,000 new cases yearly. At least
50% of sexually active men and women will acquire genital HPV infection. Most infected
persons are totally unaware they are infected and potentially infectious to a sex partner.
The causal association between persistent genital HPV infection and cervical cancer is
similar to that between cigarette smoking and lung cancer. Genital HPV infections are not
curable, and little information is available on effectiveness of prevention strategies
including abstinence, male and female condoms, and microbicides. Vaccine development is
currently underway and human trials are in progress. New tests for HPV may allow for
improved detection of cervical cancer and streamlining of costs associated with evaluation
of abnormal Pap smears when used as an adjunct to Pap smear for triage of women with
abnormal Pap smears, or for primary screening in certain populations of women. Before we
can formulate effective public health strategies to prevent genital HPV infection and
cervical cancer, we must fill critical gaps in our knowledge about feasibility of HPV
testing for cervical cancer screening, appropriate counseling messages for infected
persons, prevention of genital HPV infection, burden of infection, trends in prevalence of
HPV and sequelae, vaccine marketing strategies, and awareness of HPV among health care
providers and the general public.
HUMAN PAPILLOMAVIRUS INFECTION AND CANCER OF THE CERVIX:
WHAT DO WE KNOW AND WHAT ARE THE IMPLICATIONS?
I am Dr. Ronald O. Valdiserri, Deputy Director of the National Center for HIV, STD, and
TB Prevention at the Centers for Disease Control and Prevention (CDC). Thank you for the
opportunity to present what we know about the relationship between human papillomavirus
(HPV) infection and cancer of the cervix which was one of the most common cancers among
women in this country prior to the introduction of Pap smear screening and remains one of
the most common cancers worldwide. For more than a century, there have been suspicions
that cancer of the cervix is caused by an infectious agent and behaves like a sexually
transmitted disease (STD). For example, epidemiologic studies have consistently shown that
cervical cancer is rare in virgins but much more common in women who are sexually active
and at risk for other STDsespecially so in
women who became sexually active at a young age, who have multiple sexual partners, or who
have sexual contact with a man who has had multiple partners.
Over the past 50 years, there have been many studies attempting to assess whether a
particular infection -- such as gonorrhea, syphilis, chlamydia, or genital herpes -- was
the sexually transmitted agent that led to cervical cancer. Many of these studies cast
suspicion on one or more of these infections, but the results remained inconclusive until
the 1980s when, using newly developed laboratory techniques, evidence began to point to
another, less well understood, STD: the human papillomavirus or HPV. Prior to that time,
HPV was known to cause non-sexually transmitted warts at body sites such as the hands or
feet, as well as sexually transmitted warts around the genitals. But because warts were
rarely found on the cervix, it was thought unlikely that HPV could be playing a role in
causing cervical cancer. The inability to recognize cervical HPV infection was in large
part due to the problem that, unlike most other STD organisms, there was and still is no
way to culture HPV in the laboratory.
The development of laboratory tests for detection of HPV/DNA, the genetic material of
the virus, helped to overcome this problem and dramatically increased our estimate of just
how frequent HPV infection of the cervix and other genital sites actually occurs. It is
now estimated that approximately 5,000,000 new cases of genital HPV infection occur in the
each year, making it the most common of all of the STDs. It is further
estimated that at least 50 percent of sexually active men and women will acquire genital
HPV infection at some point and that as many as 45,000,000 Americans may already be
infected. As with many STDs, most of these infections are asymptomatic, so that the
majority of those with genital HPV are unaware of their infection -- further contributing
to its spread. The economic burden resulting from these millions of infections has not
been clearly determined, but is likely quite large. One recent estimate by the Institute
of Medicine was over $3 billion per year, more than that for any other STD apart from HIV
The HPV DNA tests have revealed that there are many different strains or types of HPV;
more than 80 types have been identified. Approximately 30 of these are found primarily in
the genital area and are considered "genital HPV". While some of these 30 types are considered "low-risk,"
primarily causing genital warts and low-grade Pap smear abnormalities, approximately 10 of
these types are considered "high-risk"for cancer in that they are found in approximately
95 percent of all tissue specimens from cervical cancer. Large epidemiologic studies
comparing women with cervical cancer to those without it have shown that, even when
controlling for other factors that might make cervical cancer more likely, being infected
with one of these high-risk HPV types increases the risk of cervical cancer by at least
30-fold, a level similar to or higher than the risk of lung cancer from smoking. In
addition to these human studies, laboratory experiments provide additional support that
HPV causes cervical cancer, by showing that when inoculated into cell culture systems, HPV
causes the cells to grow in an "out-of-control", cancer-like fashion and that these out-of-control
cells can then cause cancer when injected into mice. Thus, while definitively proving that
an infectious agent causes a disease can be quite difficult, based on a large number of
studies, there is now widespread consensus among cancer researchers that high-risk types
of genital HPV clearly play a causative role in the development of cervical cancer, and
probably other types of anogenital cancer, such as cancer of the penis and anus.
Having HPV seems to be "necessary"for developing cervical cancer, although having the
infection alone is not "sufficient"to produce cancer, and other co-factors such as
smoking, an abnormal immune system, and other infections may be important as well. The
role of the immune system has been most clearly demonstrated in patients with HIV
infection in whom very high rates of HPV infection occur and in whom both cervical and
anal cancer appear to be increased. Although a large proportion of sexually active women
will become infected with genital HPV, the majority of these infections become
undetectable over time without specific treatment or the development of complications.
Only those women whose infection persist are at risk for developing cancer, and it has
been estimated that approximately 5-10 percent of women with high-risk types of HPV
infection will develop cervical cancer. Pap smear screening programs and early treatment
reduces this percentage even further.
The recognition that this important cancer is caused by a highly prevalent STD has
important implications for public health. The first strategy to consider is that of
primary prevention, namely, preventing cancer by preventing infection. Unfortunately, the
traditional STD control strategy of preventing transmission by identifying infected
persons and then treating them and their partners in order to prevent transmission to
other partners currently has limited value for viral STDs such as HPV because existing
therapies do not cure infection. The therapies available for both genital warts and
cervical HPV infection will eradicate the tissue abnormality, but probably do not
eliminate the infection entirely. Abstinence should be effective for preventing HPV
infections, since the large majority are sexually transmitted. However, other approaches
to prevent HPV infection are also promising. Latex condoms can be expected to be
protective if they cover the genital skin that is infected and if they are used
consistently and correctly. Several studies have shown condoms to provide some protection
against cervical cancer, and the more recently developed female condom has promise as a
physical barrier in the prevention of viral STDs because of its greater surface area.
Microbicides, chemicals that inhibit microbial growth and could potentially function as
"chemical barriers"also have potential benefit. Some of these agents
currently under investigation have been shown to inactivate genital HPV in the laboratory.
Advantages of microbicides include both the possibility of inhibiting multiple STDs --
such as HPV and HIV -- with one agent, and providing a protective strategy under the
control of the woman, in contrast to male condoms.
The most promising primary prevention strategy would be the development of an HPV
vaccine. There are several animal models in which papillomavirus infections specific to
the particular animal can be effectively prevented by immunization, which has created
great optimism that vaccines against HPV might be beneficial in humans as well.
Several small studies are now underway in humans to determine whether the experimental
HPV vaccines are sufficiently safe and effective at producing an immune response to
warrant larger, more definitive studies. Because of the relatively large number of
high-risk HPV-types believed to cause cervical cancer, effective vaccines will have to
contain multiple types of HPV to achieve high levels of benefit, which increases the
complexity and length of time it will take to develop and test them. Such preventive
vaccines would ideally be given prior to the onset of sexual activity probably in early
adolescence since most people who contract genital HPV infection do so within the first
several years of sexual activity. Because the peak incidence of cervical cancer are
between 35 to 55 years of age, it would likely be at least 20 years after the initiation
of vaccine programs before we would see reductions in cancer rates. However, effective
vaccines would also reduce the rate of pre-cancerous Pap smear abnormalities, known as
dysplasia. Considering that the evaluation and treatment of dysplasia is among the most
expensive aspects of the current cervical cancer prevention efforts, reductions would most
likely occur much earlier in cost as well as the avoidance of anxiety that often
accompanies the diagnosis of an incurable STD or pre-cancerous changes on a Pap smear.
Our current strategy is to prevent cancer in those who already have HPV infection. In
essence, this is what Pap smear screening is directed towardthe early detection of pre-cancerous changes caused
by HPV infection which can be evaluated and treated to prevent their progression. With the
knowledge that HPV infection causes cancer, it may be possible to use HPV/DNA tests as an
adjunct to the Pap smear to improve its accuracy. A single Pap smear does not identify all
women who have serious abnormalities, so serial Pap smear screening is the current
standard of care. Studies are underway now to find out if using HPV/DNA tests, along with
the Pap smear, will increase the test sensitivity (in other words, the likelihood of
identifying women with abnormal Pap smears). If these tests work well enough, they might
not only prevent women with treatable problems from being missed, they might also allow
Pap smears to be done less frequently than annually in most women, thereby reducing costs
of screening. Furthermore, because samples for HPV testing are easier to collect than Pap
smear samples, they may permit the development of self-collected swab kits for women,
which, by avoiding the need for a full gynecologic exam, might be more convenient for many
women and could encourage many more women to get tested for HPV. Such self-collected
testing also facilitate development of outreach efforts, where field workers go into
non-clinic locations to do testing, similar to approaches that have been used for
community-based programs to address high blood pressure, high cholesterol, tuberculosis,
and even STDs like chlamydia.
An even more immediate use of HPV tests for secondary prevention is their use to triage
women with low-grade Pap smear abnormalities. Currently, the large majority of women in
the United States
with abnormal Pap smears have early changes that have a very low risk of
progression to cancer, and yet, to be sure an important problem isn't missed, these women usually need to come back for
several follow-up examinations, creating tremendous anxiety and expense. Several studies
have reported that providing HPV testing for these women can help determine who is likely
to have a more serious problem. If these reports can be confirmed by larger studies now
underway, they may permit a more cost-effective approach to this very common problem.
Important work remains to be done before these strategies will be ready for widespread
implementation. CDC is currently involved in a number of applied research and service
activities to improve prevention of genital HPV infection and cervical cancer. Among these
- studies of the epidemiology and natural history of HPV infection and cervical cancer.
- studies to better define approaches to clinical use of HPV tests.
- studies to assess HPV-related complications in patients with HIV infection.
- studies to determine mechanisms by which HPV causes cervical cancer.
- development and assessment of improved HPV tests.
- implementation of a pilot national population-based serosurveillance study to more
accurately assess the extent of genital HPV infection.
- support of health care provider training programs regarding both cervical cancer and
genital HPV infection.
- development of clinical practice guidelines for genital HPV infection.
- education of the general public through the CDC National STD Hotline
- implementation of the National Breast and Cervical Cancer Early Detection Program that
provides access to cancer screening and follow-up for underserved women.
- development of the National Program of Cancer Registries that will enhance surveillance
of cervical and other HPV-related cancers.
Currently, serious gaps in our knowledge preclude the formulation of more effective
prevention strategies for genital HPV infection and cervical cancer:
- If the ongoing studies to assess use of HPV tests for triage of women with low-grade Pap
smear abnormalities find this to be a helpful strategy, we must determine if this approach
works equally well in all groups of women. For example, because younger women have much
higher background rates of HPV infection than do older women, HPV testing may be too
non-specific (i.e. likely to test positive when no serious abnormality really exists) to
be helpful in the younger group, and could turn out to be a "double-edged sword",
creating more anxiety and costs than it saves.
- Studies to assess the use of HPV tests as an adjunct to Pap smear screening will also
need to demonstrate which groups of women (such as younger vs older) get the most benefit
from this extra test.
- As "self-test" kits are developed, program evaluations will be
necessary to find out how best to distribute them and encourage their use. Any use of such
HPV tests will require the development of approaches both to counsel women who suddenly
discover that they have a cancer-associated STD, and to evaluate their sexual partners.
HPV Vaccine Development and Use
- The development of effective HPV vaccines would be enhanced by collection of additional
surveillance data on the prevalence of different types of HPV infection in different
groups of men and women, both to determine exactly which types of HPV a final vaccine
should contain and to track early benefit of vaccines once they are licensed and widely
- There is virtually no experience in "marketing" vaccines for prevention of STDs and cancer to the
general public or to health care providers. Yet for HPV vaccines to achieve their promise,
their use will need to be as widespread in the population as is the virus. Sexually active
persons in all socioeconomic groups are at risk for HPV infection; thus, immunization of
all persons who will potentially be sexually active in the future would likely be the most
effective prevention approach. Behavioral and social marketing research to explore this
issue will be important and such research may also have benefit for other STD vaccines,
including those for HIV.
- To the extent that effective HPV vaccines are developed and utilized and Pap smear
abnormalities prevented, approaches used in Pap smear screening programs will also likely
evolve, since criteria for what constitutes a suspicious smear may change as certain types
of HPV infection are prevented.
Assessment of Non-Vaccine Strategies for Primary Prevention
- Pending the availability of effective vaccines, a better understanding of how well other
primary prevention strategies may work is important. Understandably, one of the major
concerns of patients diagnosed with genital HPV infection is how to prevent it from being
transmitted to sexual partners, an issue that will only increase if clinical use of HPV
testing becomes more widespread. To this end, better information is needed to determine
how long someone with genital HPV is contagious to a sexual partner and which prevention
strategies work best to prevent transmission.
Programs to Assess Burden of Infection
- Monitoring systems to provide information about rates of various types of Pap smear
abnormalities and of type-specific genital HPV infections in targeted populations will be
important in planning and evaluating vaccine programs, as well as in tracking the
distribution of HPV infection in the population. Such studies may be particularly useful
in clarifying rates and types of infections in men about which far less is known than for
infections in women.
- Economic assessments of the costs resulting from HPV infection are limited and not
available for all populations. Furthermore, existing analyses address only direct medical
costs (the costs of actually providing care), and there is virtually no information on
indirect costs (those resulting from lost productivity or premature death of someone with
a medical problem) or intangible costs (such as anxiety and distress in personal
relationships). Such information is critical in determining the potential public health
and societal benefit of various prevention programs.
Programs to Increase Public and Health Care Provider Awareness
- While better understanding of the prevalence of HPV and its relationship to cancer will
support better prevention efforts, messages to educate the general public about HPV will
need to be clearly crafted to avoid undue anxiety, competition with other public health
prevention messages, and the possibility because of the stigma associated with STD and
undermining Pap smear screening programs.
- The issues around HPV are complex ones for health care providers who must convey
messages that are both accurate and helpful to patients with concerns, often in
time-constrained clinical settings. In addition, because genital HPV infection is a minor
health problem for the vast majority of infected people, proper education and counseling
may be as important as treatment. More cost-effective means to convey this information is
an important priority.
In April, 1999, CDC and the American Cancer Society will convene a pivotal meeting of
national and international experts, including our NIH colleagues, to review possible
prevention strategies and prevention research needs for genital HPV infection and its
complications. The goal of this meeting is to develop priorities for a linked programmatic
and research agenda for CDC and other public health agencies.