E-DRUG: NEJM on cancer treatment in low-income countries
------------------------------------------------
David J. Kerr, M.D., D.Sc., and Rachel Midgley, M.D. Can We Treat Cancer for a Dollar a Day? Guidelines for Low-Income Countries NEJM August 25, 2010.
In 2002, approximately 7.6 million people worldwide died of cancer.
These deaths accounted for 13% of all deaths worldwide in that year and,
perhaps surprisingly, exceeded the number of deaths from HIV-AIDS,
tuberculosis, and malaria combined (which totaled approximately 5.6
million). The World Health Organization (WHO) has estimated that if
current trends continue, the global cancer burden will increase from 10
million new cases per year in 2000 to 16 million by 2020. Remarkably,
70% of these cases will be in the developing world, where the number
will grow from 5.2 million annually to 8.8 million by 2020.1
Currently, a cancer diagnosis in the developing world is likely to mean
a painful and distressing death. Although there is growing awareness of
the magnitude of the increasing cancer problem in low-income countries,
the challenges of producing comprehensive national cancer plans are
substantial; not least of these is the need for financial
prioritization.2 The estimated amount spent on health care through
Britain's National Health Service in 2008 was approximately $3,000 per
capita - considerably more than that spent for health care for citizens
of Kenya, about $8.30 per capita annually.
In high-income countries, we are becoming accustomed to the public
debate surrounding the licensing of expensive new anticancer agents,
many of which are associated with marginal clinical benefits.
Government-sponsored health care systems may deny patients access to
such medications, judging them cost-ineffective - moves that tend to
prompt patient advocates to exert political pressure on the funding
bodies, often engaging the media in an effort to get the decisions
reversed. Britain's National Institute for Health and Clinical
Excellence (www.nice.org.uk), for instance, has ruled against the use of
a number of recently licensed, expensive anticancer agents such as
bevacizumab. This debate must appear rather odd to oncologists who
practice in low-income countries, where the simplest generic treatment
is often beyond the reach of the average citizen. And yet those
oncologists have access to the latest online clinical guidelines
promulgating 'optimal' therapy and are no doubt invited by the
pharmaceutical industry to international meetings to hear the latest
trial results. Moreover, every country will always have an elite that
will demand the 'best care that money can buy'.Although the 'top
end' of the therapeutic spectrum is well covered by myriad guidelines
(e.g., those of the National Comprehensive Cancer Network,
www.nccn.org), very little has been done to develop evidence-based
guidelines for cost-effective cancer treatment or recommended
therapeutic algorithms for use in poorly funded health care systems.
The WHO has established a database on the cost-effectiveness of more
than 700 health-related interventions, using a standard measure,
disability-adjusted life-years (DALYs),3 which permits the comparison of
interventions within a particular field (e.g., the cost for treating
early-stage breast cancer is $78 per DALY with surgery and radiotherapy
vs. $4,986 per DALY with systemic chemotherapy for metastatic disease)4
as well as interventions across specialties (e.g., cataract extraction
and lens implantation costs $89 per DALY). These sorts of data allow
health and finance ministers to focus investment on interventions that
provide a cost-effective benefit and to conclude, for example, that it
is rational to detect and treat early breast cancer.
Economists have methods of charting the relative benefits of any new
therapy by comparing both its costs and its efficacy with those of
existing treatments - by means of the so-called cost-effectiveness
plane. Such an approach could be modified to assess the
cost-effectiveness of previously established chemotherapy regimens - for
example, for breast cancer, the combination of oral cyclophosphamide,
methotrexate, and fluorouracil versus single-agent intravenous
doxorubicin or paclitaxel; or for Burkitt's lymphoma, single-agent
cyclophosphamide versus complex combination therapy. Preference could be
given to the agents on the WHO's list of 17 'essential drugs for
cancer therapy', most of which have generic equivalents that offer the
possibility of less expensive treatment. Plots of efficacy versus cost
could be used in various ways to define acceptable benefits for optimal
resource use. For example, a national health authority could first fix a
budget and then decide whether the associated health improvement
warranted the cost or, if there was a clear point of inflection in the
curve suggesting marginal benefits for any additional cost, could use
this information to define a cost-effective regimen.
We believe that the first step forward is to establish an
interdisciplinary task force that could develop cancer-treatment
guidelines for low-income countries. Such a task force would need to
include experts in oncology, local and international health economics,
and clinical pharmacology who would consider the impact of
pharmacogenetics, pragmatic definitions of acceptable toxicity levels,
nutritional status, and intercurrent illnesses; where data are lacking,
this group could recommend the types of clinical trials and studies that
need to be performed to fill the gap. At the same time, we should not
stop attempting to solve specific problems through 'lateral
thinking': for instance, the Africa Oxford Cancer Foundation
(www.afrox.org) is currently planning a trial of nurse-dispensed
tamoxifen for Ghanaian women 50 years of age or older who, at informal
breast-examination clinics held in village churches, are discovered to
have breast lumps that are clinically adjudged to be cancer. Women who
can afford the cost of formal investigation (a minority) will be
referred to one of Ghana's cancer centers (each of which serves a
population of about 10 million), and the remaining women will receive
(free of charge) a 2-to-3-month course of tamoxifen, which might be
expected to stabilize or reduce the tumor burden in about 40 to 50% of
the patients. Tamoxifen will continue to be dispensed in the community
every 3 months by the nurses to patients with stable or responsive
disease. This highly utilitarian approach will lead to the overtreatment
of some women with breast lumps who may not have cancer, but tamoxifen
is a very safe drug that can at least offer some chance of palliative
benefit to women who attend the clinic for informal breast screening but
who cannot afford to obtain formal health care.
Does this mean that cancer medicine in low-income countries will always
be rooted in the past? Not necessarily, for we have seen some innovative
and successful projects in which pharmaceutical companies have provided
new drugs free of charge to poorer countries - for example, the Glivec
International Patient Assistance Program for patients with chronic
myeloid leukemia.5 More could be made of these open-access projects,
especially if they are sufficiently flexible to answer key research
questions about tolerability, efficacy, and pharmacokinetics, to help
refine treatment regimens that would be feasible for treatment of people
in developing countries who have cancer.
So will we be able to treat cancer for a dollar a day? At one level,
the answer might be yes, but we need to develop an evidence base that is
appropriate to the capacity of available health funds. This, we believe,
will require a multidisciplinary approach through international
alliances, such as the Breast Health Global Initiative (www.bhgi.info),
that bring individuals, health organizations, government agencies, and
corporations together in a concerted effort to improve cancer care in
resource-poor countries.
Disclosure forms provided by the
authors are available with the full
text of this article at NEJM.org.
Source Information
From the Sidra Medical and Research Center, Doha, Qatar (D.J.K.); and
the Department of Clinical Pharmacology, University of Oxford,
Headington, Oxford, United Kingdom (R.M.).
References
1. Lingwood R, Boyle P, Milburn A, et al. The challenge of cancer control
in Africa. Nat Rev Cancer 2008;8:398-403
2. Kerr F, Kerr DJ. Do we bear any moral responsibility for improving
cancer care in Africa? Ann Oncol 2006;17:1730-1731
3. Tan-Torres ET, Baltussen R, Adam T, et al. Making choices in health:
WHO guide to cost-effectiveness analysis. Geneva: World Health
Organization, 2003.
4. Groot MT, Baltussen R, Uyl-de Groot C, Anderson BA, Hortobagyi GN.
Costs and health effects of breast cancer interventions in
epidemiologically different regions of Africa, North America, and Asia.
Breast J 2006;12:Suppl 1:S81-S90
5. CanTreat International. Scaling up cancer diagnosis and treatment in
developing countries: what can we learn from the HIV/AIDS epidemic? Ann
Oncol 2010;21:680-682
------------------------------------------
Dr Valeria Frighi
University Dept. of Psychiatry
Neurosciences Building
Warneford Hospital
Oxford
OX3 7JX
UK
Tel. -44 -1865 -223779
07974920013
Fax -44 -1865 251076
Valeria.Frighi@psych.ox.ac.uk