The Evidence Base for Cancer Control in Developing Countries: What is to be done?

In July 1945 an article entitled “As We May Think” appeared in the journal Atlantic Monthly. Written by Dr Vannevar Bush (the US wartime Director of the Office of Scientific Research and Development) the article dealt not with the arbitrary power of military science to shape the future of civilizations, but a far older problem: how to retain, and benefit from, the accumulation of knowledge. Even allowing for his insider knowledge of the early development of ‘big brain’ computers to which his post made him privy, Bush’s predictions, were as visionary as those proffered to earlier generations by Jules Verne and H.G. Wells. Specialized electronic databasesaccessed through personal desk top computers, the development of the Internet and the World Wide Web are all outlined in his article. “Wholly new forms of encyclopaedias will appear, ready-made with a mesh of associative trails running through them… The physician, puzzled by its patient’s reactions, strikes the trail established in studying an earlier similar case, and runs rapidly through analogous case histories, with side references to the classics for the pertinent anatomy and histology.” The article is ‘a good read’ and contains a useful maxim: “A record, if it is to be useful to Science, must be continuously extended, it must be stored, and above all, it must be consulted.”

Sixty years on, how much of Bush’s vision has been achieved in the struggle to establish effective cancer control in developing countries? Access to most of the information hardware and infrastructure that Bush predicted has become commonplace in the developed countries but remains a relative rarity within the resource-poor populations of developing countries - a privilege of the well-placed few. More crucially, we lack data and the means to collect data. We still do not have a comprehensive picture of cancer in developing countries, and which health care interventions have proved to be effective.

Data begets evidence; evidence supports knowledge. Knowledge of the distribution of particular types of cancer, and the efficacy and cost-effectiveness of various interventions in each cancer type, is critically important. There are still not enough cancer registries in developing countries, and some of those that do exist currently produce suboptimal data. There has been even less effort expended in the proper evaluation of the effects of cancer control interventions in developing countries; i.e., the preparation of systematic reviews of studies relevant to cancer prevention, screening, diagnosis, treatment and supportive and palliative care.

Effective cancer control requires a solid foundation of evidence on which to base the selection of priorities and the choice of interventions. At present, more than half of incident cases and 70% of global cancer deaths occur in the less-developed regions of the world. By 2020, the WHO predicts that there will be at least 16 million new cases of cancer per year, 70% in developing countries. Time is not on our side. In March 2006 volumes I-VIII of Cancer Incidence in Five Continents were made available via the Internet by the International Agency for Cancer Research. (http://www-dep.iarc.fr) Volume IX is in preparation and is due to be published around late 2007. Despite the acknowledged deficiencies in cancer registry coverage, these must perforce be our source documents for prioritizing which cancers require the most urgent control, along with the ever present need for tobacco control.

Once these priorities have been determined, the evidence from all previous relevant research conducted on the local populations should be identified, collected and made as accessible as possible. Placing the focus on the evidence for the effectiveness of interventions ‘on the ground’ in the developing countries is important. Approaches to cancer control in developing countries need to be adapted to specific national needs and resources. The applicability of the much larger, and generally better known, body of evidence derived from research conducted in the developed countries is likely to vary. High- and low-income countries differ with respect to the patterns of cancer, genetics, environments, socioeconomic status, literacy rates and behavior of populations as well as with respect to the resources they can command (human, physical and financial).

Topic     Bangladesh     Bhutan     India     Maldive Is.     Nepal   Pakistan     Sri Lanka   TOTAL     Prevention     0     0     0     0     0   0     0   0     Risk     0     0     5     0     0   0     0   5     HPV related     0     0     23     0     0   0     0   23     HIV related     0     0     4     0     0   0     0   4     Diagnosis     0     0     11     0     0   0     0   11     Screening     0     0     24     0     4   3     0   31     Treatment     0     0     39     0     1   0     0   40     Secondary cancers     0     0     8     0     0   0     0   8     Psychosocial     0     0     1     0     0   0     0   1     Palliative care     0     0     0     0     0   0     0   0     Other     1     0     95     0     0   4     0   100     Totals     1     0     210     0     5   7     0   223   Table 1. Reports of cancer research relevant to cervical cancer originating from South Central Asia indexed in Medline and Embase databases between January 2001 and February 2006.

Identifying and collecting evidence is not straightforward. A preliminary survey conducted by the Cochrane Cancer Network has shown that the evidence originating from developing countries in the various fields of cancer control appears limited and is difficult to access. Searches of the Medline and Embase databases for reports relevant to uterine cervical cancer in South Central Asia published between January 2001 and February 2006 identified just 223 reports. Manual searches of conference proceedings (ESMO 2002/2004; ECCO 2003/2005, ASCO 2003-2005) and electronic searches of the INDMED database identified a further 40 reports (18%) not indexed in either Medline or Embase. The importance of identifying relevant conference proceedings in order to reduce the effects of publication bias been acknowledged. What is less well recognised is that electronic copies of conference proceedings generally do not allow search by addresses, making manual searches for abstracts originating from developing countries the only viable option. Neither are the major databases consistent in the way in which they index the countries of origin in their address fields.

New standards are being set for the registration of human medical research with the construction of the International Clinical Trials Registry Platform. The establishment of an accessible register of all relevant studies conducted in developing countries, identified by systematic searches of databases and non-indexed journals, the grey literature (conference proceedings, dissertations and industry reports) and sources of unpublished trials provide the foundation stone of the evidence base for global cancer control. This data source – which should include all types of research (translational research, epidemiological studies, as well as controlled clinical trials) will facilitate the preparation of systematic reviews of the effects and effectiveness of interventions, which should be used by NGOs (acting in alignment with national governments), health professionals and hospitals to guide improvement in clinical practice and to inform the cancer control research agenda in developing countries.

Systematic reviews: Establish a reliable and comprehensive knowledge base Identify gaps in research knowledge Prevent duplication of research & waste of resources Identify methodological difficulties and enable us to benefit from the experience of previous investigators Identify adverse effects of interventions Protect patients from unnecessary or inappropriate research

Box 1. Benefits of systematic reviews.

What is to be done?

In order to establish a comprehensive evidence base for cancer control in developing countries therefore, in addition to providing more and better data coverage by cancer registries, we need to build two new resources: (1) A register of studies and trials conducted in, or relevant to high-incidence cancers and tobacco control in developing countries (2) A database of systematic reviews of effectiveness of cancer control interventions in developing countries.

The benefits of systematic reviews are outlined in Box 1. Once developed, the register of trials and the database of reviews can used to generate a third resource: an evidence-based atlas of individuals and institutions located in developing countries contributing to cancer research, listing their areas of study and the methodology used. Linked up with the country profile database being developed by the International Union Against Cancer (UICC), this useful by-product will enable NGOs and national governments to identify the appropriate centers of research effort and, equally important, those developing countries where no cancer control research appears yet to have been done.

The identification of the source data – the primary research conducted in the developing countries – and the preparation of the systematic reviews of effectiveness will require the recruitment and training of content experts throughout the developing world. This knowledge transfer is a sensible strategic investment. As Bush’s maxim indicates, the continuous extension of the scientific record through the ongoing collection of data is essential. Updating the cancer registry data and the reviews of effectiveness of interventions will enable decision-makers to measure the success (or otherwise) of cancer control programs and to identify trends in cancer incidence and mortality which may influence changes in cancer control priorities or methodologies. A maintained body of research evidence is required in order for improved or more cost-effective methodologies to be developed.

In conclusion, the comprehensive collection of cancer registry data, the register of studies and trials, the database of systematic reviews of effectiveness, the cadre of content experts trained in preparing reliable analyses are all essential components in the building of a reliable evidence base for cancer control in developing countries. Clinical investigators have an ethical requirement to ensure that the outcomes of their clinical research are made as accessible as possible. All relevant sources of evidence, especially systematic reviews, should be carefully considered prior to undertaking new research or formulating health care policy. Clinicians and health care professionals in developing countries should be offered the skills and training to enable them to prepare and maintain systematic reviews which are both reliable and relevant to the circumstances in their own countries.

Mark Lodge
Cochrane Cancer Network
Headington, UK