Preventing Cervical Cancer Deaths: How to save 4 million lives in ten years

July 20, 2007

by Sue J. Goldie

 Sue Goldie
Sue J. Goldie, M.D., M.P.H. is a professor in the Department of Health Policy and Management at the Harvard School of Public Health. Trained as a physician, decision scientist, and public health researcher, she is best known for bringing together a wide variety of disciplinary approaches to address critical global health challenges. She is the recipient of a MacArthur award for “genius and creativity" in applying the tools of decision science to combat major public health problems.  An accomplished scientist, Dr. Goldie has published more than 100 original research papers and is the director of the Program in Health Decision Science at the Harvard School of Public Health.  She has served on several national and international committees at the World Health Organization, Institute of Medicine, National Institutes of Health, and the Centers for Disease Control and Prevention. She completed her medical training and residency at Yale New Haven Hospital, Yale University School of Medicine, and graduate work at the Harvard School of Public Health and Harvard University.  


Among the most tragic public health failures of the last decade are the preventable deaths of young women in developing countries from maternal mortality and cervical cancer.  

What do maternal mortality and cervical cancer have in common? First, efforts to address them have been among the greatest public health successes in wealthier countries. Second, both affect women in their most productive years. The death of these women has a tremendous negative impact on the wellbeing of children, family, and community. Third, despite formidable data limitations, ample evidence shows the effectiveness of several interventions that can reduce these causes of death. (1,2) Fourth, as reported in the Disease Control Priorities in Developing Countries (DCP2), interventions that are cost-effective exist to reduce mortality from either of these burdens. (3)

We are now facing unprecedented opportunities to prevent these unnecessary and tragic deaths. In fact, recent concerted efforts have been made to assemble, synthesize, and interpret the available data with an eye towards actionable steps, and to comprehensively reflect on what has worked and what has not. Moreover, researchers, public health scientists, and policymakers are beginning to engage with the distinct purpose of agreeing on the most promising strategic approaches to eradicating preventable deaths in women. 

In the case of cervical cancer, the following are among many factors altering the landscape of how we think about cervical cancer prevention:

• First, we know far more now than we did 20 years ago about the role of high-risk types of human papillomavirus (HPV) in the development of cervical cancer.

• Second, studies have shown HPV DNA testing to be more sensitive than a single cytology screening test in detecting precancerous lesions in older women. This finding has resulted in spirited dialogue about new strategies for cervical cancer screening.

• Finally, clinical trials of vaccines against two high-risk HPV types (HPV-16 and HPV-18) have shown a high degree of efficacy at preventing infection and precancerous changes from HPV types 16 and 18 in women not previously infected with these types.

Is Cervical Cancer Really a Problem in Developing Countries?

Cervical cancer is the single biggest cause of years of life lost from cancer in the developing world, affecting women at a younger age than those with almost any other cancer. (4) Nearly half a million women develop cervical cancer each year, the vast majority of whom are diagnosed in late stages with a low likelihood of long-term survival, and for whom treatment is rarely available. (5) The incidence is generally higher in the developing countries of Latin America (age-standardized incidence rates (ASR) 33.5 per 100,000) and the Caribbean (ASR 33.5), sub-Saharan Africa (ASR 31.0), and South-Central Asia (ASR 26.5) and Southeast Asia (ASR 18.3). (6)

Often referred to as "high risk,” more than 13 of the over 100 known genetically different types (genotypes) of HPV can cause cancer of the cervix and are also associated with other cancers of the genitals and anus, as well as cancers of the head and neck. (7,8) Eight of the most common HPV genotypes1 account for 90 percent of cervical cancer cases. Two of these cancer-causing HPV types, 16 and 18, account for between 65 percent and 72 percent of cervical cancer in developing countries, and a slightly higher fraction, up to 77 percent, in developed countries.

Unlike most cancers, cervical cancer is preventable through screening to detect and treat precancerous lesions. A conventional screening program, based on the cytological examination2 of cervical smears, can require up to three visits: an initial screening visit, colposcopic3 evaluation of abnormalities, and treatment. In countries that have been able to achieve broad cervical cancer screening coverage using cytology at frequent intervals, deaths have decreased considerably. However, in the vast majority of resource-poor settings such screening programs have proven difficult to implement and sustain due to a lack of human, technical, and monetary resources, and often inadequate health infrastructure. Additionally, the requirement for multiple visits, together with the need to screen at frequent intervals, has made it impossible to implement and sustain widespread organized screening in most poor countries.

Screening and Treatment Strategies

Over the last few years, an increasing number of studies have provided support for the potential benefits and cost-effectiveness of innovative cervical cancer strategies for low-income countries. (1,9,10) These include focusing screening efforts on women between 30 and 45 years of age, minimizing loss to follow-up by delivering screening and treatment in as few visits as possible, and considering a range of screening test options in addition to cytology, such as HPV DNA testing and visual inspection methods.

Which screening test an individual country should use (cytology, HPV testing, visual screening) depends on how easily the test can be implemented in order to tighten the linkage between screening and treatment; the degree to which testing might be affected by limited human resources, laboratory capacity, and general infrastructure; and the nature of associated costs. For example, screening costs associated with different approaches vary by country due to differences in the cost of labor and non-tradable goods and the relative proportion of direct medical, time, and transportation costs. For a country with an existing effective cytology-based screening program, it would be reasonable to focus efforts on improving quality and coverage.

However, in those developing countries where cytology screening has not been feasible, where resources are severely limited, and where there is minimal laboratory infrastructure, visual screening methods may be the most feasible option. On the other hand, for countries, or settings within countries, with some capacity for centralized processing of laboratory tests, HPV DNA testing may be a promising option. As newer HPV DNA tests become available, that are less costly and able to be processed rapidly in a clinic setting, the cost-effectiveness of HPV DNA testing will likely become even more attractive than already reported. (10)

Although there is still uncertainty in many assumptions (e.g., ability to achieve high coverage rates in adult women), clear evidence gaps (e.g., long-term effectiveness of cryosurgery in screening and treatment strategies), and limited cost data on scaling-up interventions,  there is general consensus that clear promising options exist for cost-effective activities for early disease detection with the potential to save lives. Of note, the uncertainties listed here are far less severe than those that existed on the effectiveness of screening when it was introduced in the United States.

The Promise of New Vaccines

Despite the enthusiasm for new and innovative screening strategies, the majority of media attention of late has been focused on reports documenting the high efficacy of the HPV vaccines. (11-14) There are two relevant HPV vaccines that are designed to prevent infection and disease with HPV high-risk types 16 and 18.4 One of these also protects against low-risk genotypes 6 and 11, and is currently licensed in the U.S. and has been approved for use in more than 70 countries. The other vaccine was recently licensed in Australia. The number of countries that will approve both of these vaccines is likely to grow dramatically in the coming years.

In women who show no evidence of past or current infection with the HPV genotypes included in the vaccine, the vaccines have been shown to prevent more than 90 percent of type-specific HPV infection. Most recently, outcomes from two large, ongoing, randomized, placebo-controlled trials of the quadrivalent vaccine on rates of precancerous cell growth showed strongly positive results in preventing illness associated with vaccine-specific types of HPV in women who were HPV-free at the time of vaccination. (12,13) The data confirmed a much lower success rate among those women already exposed to HPV, showing clearly that vaccinating at an age before females are exposed to HPV would have the greatest impact. 

Although there is already evidence of protection for at least five years after vaccination, monitoring will be necessary to determine the long-term durability of protection and the need for possible booster immunization. Also recently reported, the bivalent vaccine has shown 90 percent efficacy against high-grade cervical lesions associated with HPV-16 or -18 and partial cross-protection against new infections with genotypes 31 and 45, two genotypes closely related to 16 and 18. Longer follow-up will be required to establish the degree and durability of cross-strain protection. (14) Data are also not yet available on the safety and efficacy of HPV vaccines in Africa, or in populations with high HIV prevalence.(11)

Is an HPV Vaccine Realistic for Developing Countries?

The price of the vaccine will be a major determinant of both the cost-effectiveness and the affordability of a vaccination program. The composite costs of administration, wastage, and impact on the cold chain (i.e., a temperature-controlled supply and storage chain to maintain proper vaccine temperatures required to preserve potency) will also be important. While these latter costs can be estimated based on experience with childhood immunization, the programmatic and delivery costs associated with reaching an adolescent population are not yet known. (15) Of particular relevance to an HPV vaccination strategy will be the costs associated with alternative modes of delivery and of achieving incremental increases in coverage rates. Although school-based interventions are potentially a means of delivery, where school enrollment rates of girls are low, community outreach efforts to reach girls outside of school will be necessary. (16)

Mathematical models are increasingly being used to combine the information that is available now (e.g., epidemiology, sexual behavior, vaccine efficacy, economic) with the evidence gaps that remain (i.e., what we do not yet know), in order to assist policymakers in making informed decisions in the short-term. (15,17)

These modeling exercises predict that the overall benefit to the population from vaccinating girls in different regions of the world will depend on the age when they are vaccinated, on the epidemiology of type-specific HPV in the given population, on the coverage rates, and in the setting of incomplete coverage, the indirect protection to unvaccinated individuals via reduction of HPV 16 or 18 transmission in the community. Even considering the uncertainties, based on what we know now, in resource-poor settings without existing screening programs, the clinical benefits of even a partially effective HPV vaccine are likely to be substantial compared with the status quo. (11,17)  Of course, the long-term benefits will depend on the durability of vaccine-induced protection, any positive effects from the vaccine on other HPV types (i.e., cross-protection), and on the role of non-vaccine targeted HPV types in a vaccinated population. 

An analysis conducted to explore cervical cancer control options in Brazil show that at a per dose cost of $5 (excluding wastage, administration, and vaccine support costs), the cost-effectiveness ratio associated with adolescent vaccination combined with screening three times per lifetime is less than the country-specific GDP, a suggested threshold for a very cost-effective intervention.   However, a per dose cost approximating the price in the U.S., vaccination would not be cost-effective in Brazil compared to screening three times per lifetime. (17,18) For countries with GDPs per capita less than $1,000, the per-dose cost may need to be as low as $1-$2 to be both cost-effective and affordable. Manufacturers have declared their willingness to set tiered prices for countries of different economic settings but negotiations with procurement agencies will likely occur only after pre-qualification of the vaccine by the World Health Organization (WHO).

Ultimately, the real-world effectiveness of HPV vaccination will require new ways to facilitate adolescent access to the health system or creative means of reaching them outside the system. The potential future introduction of HPV vaccines with a successful adolescent immunization program creates opportunities for strengthening health systems through the establishment of mechanisms for vaccine delivery and monitoring of impact. If successful, this could facilitate the delivery of other adolescent health services, and serve as potential vaccination opportunities for tetanus, measles, rubella, meningococcal, typhoid or even, ultimately, vaccines for HIV. (16)

When trying to prioritize disease-specific programs to improve health, decisionmakers consider:
• The magnitude of disease burden relative to other competing priorities;
• The availability of effective interventions;
• The capacity to deliver effective interventions;
• The affordability and cost-effectiveness of strategies to implement the interventions;
• The feasibility and costs of scaling up the service; and
• The health infrastructure and general capacity to support the service.

In the context of decisions about cervical cancer screening and HPV vaccination, decisionmakers in individual countries will need to formally consider the above-mentioned factors, the coverage rates achievable in their settings, and the context for vaccination and screening separately. Because these interventions are applicable to such different age groups, require money that may come from different sources, are subject to different constraints and bottlenecks, and depend to different degrees on existing health infrastructure, the feasibility of implementing programs that are able to cover a large segment of the eligible population may vary greatly between countries and within countries.  Also, as emphasized in the Disease Control Priorities in Developing Countries (DCP2), (19) because of the interaction between cost-effectiveness, disease burden, and available resources, cost-effectiveness alone provides only part of the overall information needed to inform decisions about adoption. Additional criteria such as affordability, distributional effects and equity, and cultural preferences are equally influential and important to consider.  Decisionmakers may need to give careful thought to the relative likelihood of cultural acceptability as well as political will and public sector support for these different approaches to cervical cancer prevention.

Time for Action

Already, 99 percent of maternal deaths occur in developing countries. Without a drastic change in available preventive measures, more than 90 percent of cervical cancer deaths will also occur in developing countries in the next 15 years. Although millions of women living in poor countries have died of invasive cervical cancer in over just two decades, with the availability of new options for cervical cancer screening, and an effective, safe preventive vaccine, there is real hope for reducing these numbers.

But there are constraints. Achieving broad vaccine coverage of young adolescents, negotiating tiered pricing to a few dollars per dose, and securing expedient vaccine financing will be challenging. Achieving broad screening coverage of women between ages 35 and 45, training the technical workforce and providers to deliver screening and provide treatment, and improving the quality of care for cancer treatment and palliative care will be no less challenging.  Both will require political will, global cooperation and commitment of stakeholders, and champions who will keep momentum moving forward.

In the meantime, there is, however, a substantial risk of stalled movement forward and paralysis in the face of limited health care budgets.  The promising results coming from the clinical trials on HPV vaccines could invite a wait-and-see attitude concerning cervical cancer prevention via new screening approaches. On the other hand, enthusiasm for investment in HPV 16/18 vaccination could be hampered by decisionmakers faced with other new vaccines competing for the same public health dollars, rationalizing that this disease can be prevented by means of early disease detection with screening. Yet a delay of more than five years in bringing HPV vaccination and new screening strategies to developing countries means another 2 million women will die from a preventable disease. Careful deliberation, but not inaction, is needed to address the decisions countries face now.


1 The eight genotypes are 16, 18, 31, 33, 35, 45, 52, and 58.
2 Cytology is the study of cell biology and is used here to describe the examination of cells from the cervix to look for abnormal cell growth.
3 A colposcope is a magnifying instrument designed to facilitate visual inspection of the vagina and cervix.
4 The two HPV prophylactic vaccines include a vaccine that targets 4 HPV types - 6/11/16/18 (Gardasil®; Merck & Co., Inc., Whitehouse Station, New Jersey) and a vaccine that targets 2 HPV types - 16/18 (Cervarix; GlaxoSmithKline, Uxbridge, Middlesex, United Kingdom).

Sue Goldie’s work is supported in part by Bill & Melinda Gates Foundation grant 30505, “Development and Evaluation of HPV Vaccines and Related Diagnostics: Global Policy Model.”


(1) Brown M, S.J. Goldie, G. Draisma, J. Harford, and J. Lipscomb. 2006. “Health Service Interventions for Cancer Control in Developing Countries.” In Disease Control Priorities in Developing Countries, 2nd ed., ed. D.T. Jamison, A.R. Measham, J.B. Breman et al., 569-90. New York: Oxford University Press.

(2) Graham W.J., J. Cairns, S. Bhattacharya, H.W. Colin, C.H.W. Bullough, Z. Quayyum, and K. Rogo. 2006. “Maternal and Perinatal Conditions.” In Disease Control Priorities in Developing Countries, 2nd ed., ed. D.T. Jamison, A.R. Measham, J.B. Breman et al., 35-86. New York: Oxford University Press.

(3) Laxminarayan R., J. Chow, and S.A. Shahid-Salles. 2006. “Intervention Cost-Effectiveness: Overview of Main Messages.” In Disease Control Priorities in Developing Countries, 2nd ed., ed. D.T. Jamison, A.R. Measham, J.B. Breman et al., 35-86. New York: Oxford University Press.

(4) Yang B.H., F.I. Bray, D.M. Parkin, J.W. Sellors, and Z.F. Zhang. “Cervical Cancer as a Priority for Prevention in Different World Regions: An Evaluation Using Years of Life Lost.” International Journal of Cancer. 2004; 109(3): 418-24.

(5) Parkin D.M. and F. Bray. “The Burden of HPV-related Cancers.” Vaccine. 2006; 24 (Suppl 3) S11–S25.

(6) Ferlay J., F. Bray, P. Pisani, and D.M. Parkin. 2004. GLOBOCAN 2002 Cancer Incidence. Mortality and Prevalence Worldwide. IARC CancerBase no. 5 version 2.0. Lyon: IARC Press.

(7) Munoz N., F.X. Bosch, S. de Sanjose et al. “Epidemiologic Classification of Human Papillomavirus Types Associated With Cervical Cancer.” New England Journal of Medicine. 2003; 348: 518-27.

(8) Clifford G., S. Franceschi, M. Diaz, N. Munoz, and L.L. Villa. “HPV Type-distribution in Women With and Without Cervical Neoplastic Diseases.” Vaccine. 2006; 24 (Suppl 3): S26–S34.

(9) Sankaranarayanan R., L. Gaffikin, M. Jacob, J. Sellors, and S. Robles. “A Critical Assessment of Screening Methods for Cervical Neoplasia.” International Journal of Gynaecological Obstetrics. 2005; 89 (Suppl. 2): S4-S12.

(10) Goldie S.J., L. Gaffikin, J.D. Goldhaber-Fiebert, A. Gordilla, C. Levin, C. Mahe, and   T. Wright. “Cost-effectiveness of Cervical Cancer Screening in Peru, India, Kenya, Thailand and South Africa.” New England Journal of Medicine. 2005; 353: 2158-2168.

(11) Initiative for Vaccine Research, Department of Immunization, Vaccines and Biologicals, World Health Organization. 2007. Human Papillomavirus and HPV Vaccines: Technical Information for Policy-makers and Health Professionals. Geneva: World Health Organization.

(12) Garland S.M., M. Hernandez-Avila, and C.M. Wheeler et al. “Quadrivalent Vaccine Against the Human Papillomavirus to Prevent Anogenital Diseases.” New England Journal of Medicine. 2007; 356: 1928-43.

(13) The FUTURE II Study Group. “Quadrivalent Vaccine Against Human Papillomavirus to Prevent High-Grade Cervical Lesions.” New England Journal of Medicine. 2007; 356: 1915-27.

(14) Paavonen J., D. Jenkins, and F.X. Bosch et al. “Efficacy of a Prophylactic Adjuvanted Bivalent L1 Virus-like-particle Vaccine Against Infection With Human Papillomavirus Types 16 and 18 in Young Women: An Interim Analysis of a Phase III Double-Blind, Randomised Controlled Trial.” The Lancet. 2007; 369: 2161-70.

(15) Goldie S.J., J.D. Goldhaber-Fiebert, and G. Garnett. “Public Health Policy for Cervical Cancer Prevention: Role of Decision Science, Economic Evaluation, and Mathematical Modeling. Towards a New Paradigm in Cervical Cancer Prevention.” Vaccine. 2006: 24; S155-S163.

(16) Agosti J.M. and S.J. Goldie. “Introducing HPV Vaccine in Developing Countries — Key Challenges and Issues.” New England Journal of Medicine. 2007; 356: 1908-10.

(17) Garnett G., J.J. Kim, K. French, and S.J. Goldie. “Modelling the Impact of HPV Vaccines on Cervical Cancer and Screening Programmes.” Vaccine. 2006: 24; S178-S186.

(18) Goldie S.J., J.J. Kim, K. Kobus, J.D. Goldhaber-Fiebert, J. Salomon, M.K.H. O’Shea, F.X. Bosch, S. De Sanjose, and E.L. Franco. “Cost-Effectiveness of HPV 16, 18 Vaccination in Brazil.” Vaccine. (in press)

(19) Musgrove P. and J. Fox-Rushby. “Cost-Effectiveness Analysis for Priority Setting.” In Disease Control Priorities in Developing Countries, 2nd ed., ed. D.T. Jamison, A.R. Measham, J.B. Breman et al., 271-83. New York: Oxford University Press.




Sign up to receive periodic email updates from DCPP