INCTR's Evolving Strategy
Part I - Cancer in Developing Countries

by Ian Magrath

Public education in a subsistence farming region in Kenya accessible only on foot. See "Partner Profile" for more information.

The World Health Organization (WHO) recently projected that by approximately 2010, cancer would overtake ischemic heart disease as the leading cause of death in the world. Between 2005 (when some 7.6 million people died from cancer, accounting for 13% of global deaths)and 2015, it is anticipated that 84 million people will die of cancer. In 2005, approximately 70% of cancer deaths occurred in low and middle income countries where, although cancer has a lower incidence, survival rates are much lower, largely because of delays in diagnosis leading to presentation with advanced disease. Many poor patients receive affordable (by their standards) or available treatment rather than optimal treatment, and those with little chance of benefiting from cancer treatment, or without financial support, are not infrequently sent home to die - without even the comfort of palliative care. Many patients (uncounted) never reach a center capable of providing treatment. This catastrophe will soon become a crisis, since the global cancer burden is increasing rapidly in the developing countries where populations continue to expand and communicable diseases are better controlled, resulting in longer life spans. Unfortunately, death from infection is all too often replaced by death from smoking, which, along with an unhealthy diet (lacking in fruits and vegetables with a high overall calorie intake) coupled to a sedentary life style, predisposes to a number of non-communicable diseases. Tobacco and diet, together, account for up to 60% of cancer in high income countries. These risk factors, which are, in theory, avoidable, will take an increasing toll on the health of the emerging middle class, particularly in those countries which are rapidly developing, such as India and China. The increase in cancer deaths will not be small; the International Agency for Research on Cancer (IARC) predicts that by 2030 there will be 27 million new cases and 17 million cancer deaths per year - an extra 10 million deaths compared to 2005.

The lower incidence of cancer in developing countries (Figure 1), even using age-adjusted rates, is largely due to the much smaller impact in these countries, to date, of tobacco and diet/exercise. The tobacco epidemic is not yet at its height and fruits and vegetables generally comprise a higher fraction of the overall lower caloric intake in low income populations - who lead anything but sedentary lifestyles. But even in developing countries, the fraction of overweight people is steadily increasing (although malnutrition also remains a problem). The increased risk of cancer with age is, to a large degree, a function of prolonged exposure to risk factors. Unfortunately, any reduction in cancer incidence resulting from better control of chronic infections will almost certainly be overwhelmed by the consequences of failure to preempt the effects of smoking and dietary factors.

In less developed countries, the fact that a higher fraction of patients die from cancer demonstrates that their attempts to control cancer are much less effective - which is hardly surprising, given the remarkable disparities in resources between the lowest and highest income countries. The poorest populations, particularly those in African find themselves trapped in a series of vicious cycles (Figure 2) from which escape is difficult, but essential if their socioeconomic status and health care is to improve.

Figure 1. Incidence of cancer in more and less developed countries - Globocan estimates of crude (actual) rates and rates adjusted to the age structure of the world population. The difference in crude and age-specific rates will narrow as populations age. The differences in incidence and mortality rates provide an indication of the efficacy of therapy.

Resource Limitations in Developing Countries

According to the World Bank, approximately 1.3 billion people live on less than a dollar a day, and almost half the world's population lives on less than 2 billion dollars a day. While these numbers are somewhat arbitrary, as is the definition of "poverty," they clearly indicate that the bulk of humanity is extremely poor. Poverty at the level of individuals is reflected in the poverty of governments, resulting in major negative impacts upon education and health care. At the turn of the millennium, for example, a billion people were unable to read or sign their name. According to UNICEF, some 11 million children die each year as a direct consequence of poverty, and the United Nations Development Program (Human Development Report, 2006) states that a million children a year die for want of clean water and adequate sanitation. Poor health (malnutrition, anemia and common infectious diseases) also causes frequent interruptions in education, and reduce the ability to learn. Many children never go to school (which is rarely free), and the lack of even a basic education (Figure 3) condemns most to a life of perpetual poverty and potential catastrophe; in this setting, natural disasters or inter-ethnic violence precipitate serious added health problems.

Difficult decisions must be made by governments regarding priorities for their limited budgets. Health expenditure is usually just a few percent of total government spending - in absolute terms, sometimes hundreds of times less than is spent per capita in high income countries (Figure 4). Health workforces are correspondingly small and unable to cope with the burden of disease. The WHO reports that sub-Saharan Africa, with 11% of the world's population and 25% of the global burden of disease, accounts for less than 1% of the global health expenditure. In contrast, the Americas, with 14% of the world's population and 10% of the global burden of disease, account for more than 50% of the global health expenditure. Comparisons of the numbers of doctors and nurses per capita in selected countries are shown in Figure 5. Some populations in low income countries are considerably worse off than even these figures would suggest since the inadequate workforces are also mal-distributed (rural regions are particularly poorly provided for) and health services poorly structured and managed. Some countries have fewer nurses than doctors such that nursing tasks, to the extent possible, must be performed by family members, or not at all - an unthinkable situation in high income countries.

Figure 2. Vicious cycles that feed off each other - poverty, education, knowledge, evidence, access to care, prevention, early detection and treatment outcome.

Too many patients for the available health providers results in no, or limited time for continuing education, and minimal infrastructure for the collection of data relevant to developing an effective strategy for cancer control and measuring the effects of interventions. Although there are cancer registries in developing countries, much of the available data is institutional rather than population based and of variable quality. Poorly organized health services and inadequate training of health workers often lead to delays in diagnosis, mis-diagnosis, loss of clinical samples, inadequate investigations, poor recordkeeping and transfer of relevant information to referral centers, poor supportive care and limited or no patient follow up. Clinicians often lack the information needed to determine appropriate therapy, which may, in any case, be unavailable, and treatment is not infrequently abandoned prior to completion. Poor communication among specialists hinders effective combination of therapeutic modalities and lost opportunities for sharing experiences or discussing difficult cases. Professional education often leaves much to be desired. This gloomy picture varies markedly from country to country and institution to institution. Some centers provide the best quality services possible to their patients with the basic resources available. Some countries in the upper middle income category provide patient care and professional education that, at least in the best centers, can be equal to or close to the level of high income countries.

Resources for each of the three major modalities of cancer therapy are severely limited in low and lower middle income countries. Surgeons skilled in cancer surgery are in short supply and, like all other resources, more likely to be available in urban rather than rural regions. Radiation therapy, which evolved early in the 20th century in Europe, has still not spread to all countries in the world, in part because of the capital cost of equipment, but also because of the lack of radiation oncologists and medical physicists. According to the International Atomic Energy Agency, half of the world's countries have 85% of all radiation therapy machines in the world - leaving 15% for the other half. Although the lower incidence of cancer leads to a lower requirement of radiotherapy machines per capita, there is still clearly an insufficient number to provide for the needs of many countries; Barton et al., (Lancet Oncology, 2006,7;584) have estimated that at the end of the 1990s there was a need for 842 megavoltage machines in Africa, 4936 in 12 low and middle income countries in the Asian Pacific region and 1530 in 23 selected Latin-American countries. Limited maintenance and outdated cobalt sources compound the problem.

Figure 3. Mean years of schooling in selected countries between 1991 and 2001. Source, World Development Indicators, 2005.

Chemotherapy and hormonal therapy is available pratically everywhere, although many countries purchase only "essential" cytotoxic drugs based on the WHO Essential Drugs list, and deficiencies in procurement procedures often lead to intermittent supplies. Cost is also a major issue since much or all of the care, particularly drug costs, must frequently be paid for "out-of-pocket." Even generic drugs manufactured in developing countries, such as India, and available at much lower cost, may be beyond the means of many families, particularly when import duties and other taxes are added to the drug costs. In some developing countries there are no medical oncologists and/or pediatric oncologists, and few have adequate numbers of specialists.

These multiple deficiencies are compounded by the temptation to emigrate in search of improved professional and financial rewards. Training fellowships given by high income countries that may have been established with the intent of improving the workforces of low income countries can significantly exacerbate this problem, but professionals are also actively recruited. Similarly, an increasing number of trained persons move partly or entirely into the for-profit health sector - leaving the poorer elements of the population with even less access to care. Some would argue that countries should simply train more doctors, nurses or other health professionals than they themselves require, but this apparently simple solution presupposes a sufficient number of institutions of higher education, of teachers, and of young people qualified to receive higher education - i.e., a higher level of socioeconomic development than currently exists. Although emigration rates vary markedly from one country to another, and statistics can be difficult to compile for various reasons, it would appear, according to the Center for Global Development, that at least 40% of African born physicians work outside their country of birth. The migrants are more likely to come from countries with better health systems, such that as development proceeds, the situation will get worse before it gets better.

Figure 4. Health Expenditure per capita (international dollars) in selected countries. Source, World Development Indicators, 2005.

Cancer Control Strategies

Treatment has always been the primary approach to cancer control - in developing countries and in the world. This is hardly surprising since the most immediate need is to care for sick patients. In developing countries particularly, the imbalance between the cancer burden and the health workforce leaves little time for therapists to think about epidemiology or public health, nor are most trained to do so. Even late presentation is generally not seen as something they have a responsibility to address. At the same time, those who deal with population health in the poorest countries continue to be concerned almost exclusively with infectious diseases and nutrition, rather than non-communicable diseases. Where cancer control is contemplated, the lack of resources leads many to suggest that palliative care and primary prevention of cancer, i.e., the avoidance or reduction of exposure to environmental risk factors, should be the highest priorities, rather than treatment, which already demands more resources than are available. But palliative care, however necessary, will not reduce mortality and thus provides no hope for the future, while prevention, although more appealing and potentially less expensive than treatment, is dependent upon sufficient knowledge of causal, or at least, predisposing factors (which is not available for many cancers), and the political will to initiate necessary actions. It also requires decades for its effects to be felt.

Not only are all aspects of cancer control important, but they need to be much more closely linked than has hitherto been the case.

Figure 5. Doctors and Nurses per 1000 people in selected countries. Source, World Health Report, 2006.

Prevention

The first reports of cancers associated with exposure to environmental agents were published by Sir John Hill (snuff-taking and cancer of the nose) and Sir Percival Pott (scrotal cancer in chimney sweeps) in 1761 and 1775, respectively. Pott recommended baths to prevent the latter, but this advice, although heeded by Scandinavians with great success, was largely ignored in Britain until 1892! Progress in identifying additional causal factors was slow, and predominantly recognized in the context of occupational exposures. In the course of the latter half of the 20th century, a great deal of information has been collected regarding chemical carcinogens and infectious agents that predispose to cancer. Definitive evidence of the causal association of smoking with lung cancer was published in 1950, by which time the incidence of lung cancer in men in more developed countries had increased markedly as a result of the dramatic rise in smoking rates throughout the first half of the century (see www.deathsfromsmoking.net), although the peak incidence, in the UK, for example, was not reached until about 1990. Prevention is not, unfortunately, always as easy as taking a bath. Even the administration of vaccines is associated with cost, logistic issues, and may be opposed, for various reasons, by some sectors of the community. Much more difficult to deal with than occupational exposures, or cancers associated with chronic infectious diseases, are those for which preventive measures may entail a significant change in lifestyle; or even, in the case of smoking, the single most important cause of cancer, overcoming a powerful addiction or the psychological pressure to smoke exerted on targeted segments of the population by the transnational tobacco companies (these days, often covert). Prevention, no less than treatment, is closely linked with business and politics and is rendered more difficult to achieve because one sector's gain is another sector's loss.

Preventive interventions are largely carried out, or advocated for, by a different set of persons or organizations than those involved in treatment. Yet primary health providers have a critically important role to play. Unfortunately, basic medical and nursing training in most countries includes a minimum of information (or none) on the importance of promoting a more healthy lifestyle in their patients. Public educational campaigns via the media can also be immensely valuable, and ideally, information about cancer and its prevention should be incorporated into primary and secondary education and addressed by a variety of associations at the community level.

Treatment

The fact that primary prevention is not possible for all cancers should not detract from the high priority it deserves since, to the degree to which it is effective, the outcome is an eventual reduction in incidence, and a reduced need for the resources required for treatment and palliative care. Prevention, however, can never be the sole approach to cancer control, and should not be the sole focus of national cancer control committees, no matter how limited resources may be, since such a policy would be tantamount to turning one's back on the cancer deaths anticipated in the next several decades. At the very least, consideration should be given to improving the efficiency of existing treatment programs particularly with regard to potentially curable cancers. Estimating the potential benefits of treatment and, particularly, cure rates, is difficult, but the most recent 5-year relative survival rates reported in the Eurocare-4 study and by the US SEER program indicate that on average, some 50-64% of all cancer patients, in Europe and the USA, are alive at 5 years (Verdecchia et al., Lancet Oncology, 2007,8;784). A high fraction of patients can be cured when the cancer remains localized and a similarly high fraction of pediatric cancers, testicular cancer and a number of lymphoid neoplasms are curable - sometimes even when advanced. Unfortunately, the fraction of cancers that are curable is significantly lower in developing countries because of late diagnosis and limited resources for investigations and treatment, but the paucity of data coupled to the enormous variations in disease patterns and access to care in different populations or in different regions, even within the same country, make estimates hazardous. Some idea, at least, is provided by a comparison of mortality and incidence rates (Figure 1).

Like risk factors which can be relevant to several cancers, treatment approaches, at least in terms of general principles and the treatment modalities used (loco-regional therapy, including surgery and radiation therapy and systemic therapy - i.e,. chemotherapy agents, hormones or novel agents such as monoclonal antibodies) also overlap, in that they are applied to many different cancers. There are, however, over 100 types of cancer, and each stage of cancer (a shorthand notation of the size, degree of loco-regional invasion and the presence of distant spread) needs an appropriately tailored treatment approach, often requiring the combined expertise of a number of specialists, as well as skilled nursing, pharmaceutical and technical support. Therapeutic decisions are dependent upon expert pathological examination and radiological or other types of "imaging" studies, or, in some cases, the detection of a serum marker (such as carcinoembryonic antigen). In developing countries, many of the investigations used in high income countries to determine the stage of disease are either not available, or must be restricted to particular uses because of limited availability, or high cost. Such tests, however, are probably overused in high income countries and careful consideration should be given as to how much each contributes to therapeutic decisions in various cancer types such that their use is kept to a minimum.

Effective cancer control requires effective collaboration.

In some cases the biological characteristics of a cancer may be a critical factor in treatment decisions (e.g., the expression of hormone or growth factor receptors in breast cancer) and will become increasingly important in an age of targeted therapy. However, the benefits of new treatment approaches vary markedly, and may sometimes be very small, although their cost is invariably high such that most are completely beyond the ability of poorer patients or governments to pay. Even in resource-rich countries, there is considerable debate about the high cost of new systemic therapies.

Early Detection

The role of the primary care provider in cancer treatment is, in some respects, as important as that of the cancer specialist, since the best chance for cure is when the possibility of cancer is considered soon after symptoms arise, such that the diagnosis is established and treatment initiated at an early point in the evolution of the disease. Approaches to early diagnosis include education of the public, in order that care is sought at the earliest time, and of the primary care providers, to ensure that cancer is considered when warranted and appropriate diagnostic steps are taken.

In some cases, cancer can be detected in asymptomatic patients by screening - when, for example, the primary anatomical location of the cancer is readily accessible (e.g., skin, breast, mouth, cervix). Techniques include direct vision, sometimes aided by vital stains (e.g., acetic acid or iodine in cervical cancer), or special tests such as cytology. Early cancers at less readily accessible sites may be detected endoscopically, or by radiology or other imaging procedures. Tests for minor bleeding not recognized by the patient, or for a chemical marker in the blood, are also used to detect asymptomatic cancers. Such tests vary markedly in terms of the skill and time spent in their performance by the individual or team of health professionals involved, the equipment required, the sensitivity and specificity of the test and, of course, the cost. All of these, in conjunction with logistical issues such as whether the test is offered "opportunistically" or by recruitment, are relevant to coverage of the target population and hence, to the impact on mortality rates.

The digital era could change the standard equations for early detection, since any visual technique can be digitalized - both at the level of the procedure and any subsequently required histopathological or cytological examination. Images can then be transmitted electronically for reading by off-site experts. Digitalization can save time, improve data management and compensate for local deficiencies in human resources. It also permits distance training and allows the introduction of efficient quality control. Cost benefit ratios may depend upon many factors, including the cost of equipment, image transmission, salaries and fees for technicians and readers as well as the beneficial impact on treatment costs and outcome that is achieved by earlier diagnosis.

A potentially negative aspect of screening is that lesions may be detected that are not yet invasive cancer, and may or may not ever become so. Such lesions, however, cannot be ignored. When treatment is simple, inexpensive and with few or no side effects (such as cryotherapy for cervical cancer), this may be of minimal concern, but when surgery and/or radiation therapy are necessary (e.g., in non-invasive breast or prostate lesions), the added cost, unnecessary, worrisome and potentially harmful interventions must also be taken into consideration in calculating the cost-benefit ratio.

Disciplinary and Cultural Divides

Cancer control is often thought of as referring to prevention and early diagnosis and primarily involving epidemiologists and public health specialists. However, effective public education requires the active participation of care givers, who also play a critical role in early detection. Moreover, early detection is pointless unless treatment can be immediately instituted. Treatment (including palliative care) is an essential element of cancer control in its own right - and the only way of controlling some cancers. Effective cancer control will entail collaboration among institutions and organizations at both national and international levels that have quite different cultures and often different goals, rendering effective communication and concerted action more difficult. Such divides must be overcome, since each element of the community has an important role to play. Supranational organizations provide information, guidance and support to national governments which are responsible for the creation of relevant legislation, determining and regulating the structure of the health care system, and promoting (or endorsing) and supporting the creation and implementation of a national cancer control strategy. Academic institutions and cancer centers offer education and training for health professionals, while simultaneously providing health care and conducting the research needed to create and expand the evidence on which effective action depends. Industry is ultimately responsible for the manufacture of all the equipment and drugs needed for cancer control, and has an important role in the development of novel approaches and new technologies for diagnosis and treatment. Finally, civil society provides advocacy, funding, and information and may play an active role in professional and public education as well as participating directly in cancer control activities. As such, it plays an important role in helping to provide a connecting matrix among the range of involved institutions, and in bringing distant disasters into the purview of those who are able to help - striving to create, in the process, a political climate of collective solidarity.

In part two of this message, some of the ways in which INCTR works, or plans to work, in creating the networks necessary for an integrated approach to cancer control will be discussed.