Stopping the Use of Arsenic

Jean-François Corty

Human African trypanosomiasis (more commonly known as sleeping sickness) was a frequent scourge of the colonial period, but virtually disappeared in the years following independence. It made a comeback in the closing decades of the twentieth century, however, and now affects between 50,000 and 150,000 people a year, and threatens nearly 50 million. Thanks to improved surveillance since the early 1990s, the post-colonial epidemic now appears under control. Where before there were high-prevalence foci, there are now numerous minor foci scattered throughout Africa’s tsetse fly belt.

Aside from its unique pathophysiology, what makes this disease so serious is the insidious and potentially devastating dynamic of outbreaks (Cattand, 2001, p. 315). These occur primarily in the Democratic Republic of Congo (DRC), Angola, and southern Sudan, although the Congo Republic, Ivory Coast, Tanzania, and the Central African Republic also report some cases. With no existing vaccine and a prevalence that can exceed 50% in places, it’s not rare for trypanosomiasis to be the most common, or second most common, cause of death in certain communities, ahead of HIV/AIDS (Courtin, 2008).

MSF began tackling the disease in the 1980s. Now, two decades later, the organization treats as many as 20% to 30% of all sleeping sickness cases—46,200 patients in 2006 alone. Thanks to its experience in a number of high-prevalence contexts, MSF understands the challenges of managing sleeping sickness, and is working to bring innovation in three areas. The first is treatment, where existing drugs are hard to handle, often very toxic, resistance-prone, and prohibitively expensive—in the rare instances they are available. The second is diagnostic tools, which are virtually non-existent. The third is the overall operational organization of care. As we will see, MSF has been a driving force in designing disease-control policies, as well as diagnostic and therapeutic techniques. All of these efforts aim to respond to the urgent need for care.

Sleeping Sickness Wakes Up: Neglected Populations and Neglected Diseases

Sleeping sickness is a vector-borne parasitic disease that exists exclusively in sub-Saharan Africa, with the vast majority of cases in rural areas. The parasite, a protozoan of the genus Trypanosoma (brucei gambiense), is transmitted to man by the bite of a tsetse fly previously infected by biting human or other animal carriers of the pathogen. There is a second form, Trypanosoma brucei rhodesiense, but it has low prevalence in humans, progresses more rapidly, and is easier to diagnose.

The disease is characterized by an indolent, progressive course in two stages. During the first, or hemolymphatic, stage, besides posterior cervical lymphadenopathy, the clinical signs are non-specific: recurrent fever, pain, fatigue, itching. The second, or meningoencephalitic, stage, affects the central nervous system, causing pain and severe psychiatric and neurological problems. Without treatment, the disease is invariably fatal after a period of wasting and coma.

By the 1980s and 1990s—nearly a century after the first efforts to control the disease—the practical problems had changed very little. Diagnosis was complicated, and treatments such as suramin and pentamidine were only effective in stage 1 of the disease. There were only two drugs approved for stage 2: melarsoprol, which is extremely toxic, and eflornithine, which is extremely expensive.

In 1978 some research progress was made, including the creation of the Card Agglutination Test for Trypanosomiasis (CATT), while capillary tube centrifugation was already available.Capillary tube centrifugation is a laboratory technique in which a small cylinder, filled by capillary action with a biological fluid, is placed into a machine that uses centrifugal force to separate sediment to be used for laboratory measurements. Yet only two new drugs have been discovered in the past fifty years: eflornithine and nifurtimox. Due primarily to a lack of political will, innovations could barely get past the invention stage. “While a ‘white elephant’ for which no one has any use might be a great invention in terms of technological achievement, it may never become an innovation if there is no demand for it. … It’s clear that a policy to stimulate technology is therefore not the same thing as a policy to develop innovation” (Blondel, 2002, p. 135). There were few actors in the field to take advantage of these developments, and the World Health Organization (WHO) had only two specialists working on the issue at the time.

In response to the resurgence of sleeping sickness, the Thirtysixth World Health Assembly adopted resolution WHA36.31 in 1983, urging the WHO to increase controls, and a sleeping sickness prevention and control program was created the following year.

In 1992 the WHO proposed an “Initiative for Central Africa” aimed at coordinating and strengthening sleeping sickness control in the region. The project, a joint effort by the WHO, the Food and Agriculture Organization, the International Atomic Energy Agency (IAEA),The IAEA participates in vector control by using radiation to sterilize male tsetse flies. and the Organization of African Unity (OAU), was finally launched in 1995. Support from the French and Belgian governments, and later the European Union, helped fund a program to coordinate sleeping sickness control activities in Central and West Africa. The program increased the participation of national officials and intensified collaboration with a number of institutions, including the Organisation de Coordination et de Coopération pour la lutte contre les Grandes Endémies, based in Ivory Coast; the Organisation de Coordination pour la lutte contre les Endémies en Afrique Centrale, based in Cameroon; MSF; the Fonds Médical Tropical; and France’s Institut de Recherche pour le Développement. In 1997 efforts were bolstered by the adoption of resolution WHA50.36 by the Fiftieth World Health Assembly.

Despite these developments, however, management of the disease remained sketchy, which proved that the initiatives were unsuitable. As the number of actors grew, the campaigns lost coherence and long-term vision, and there was a lack of political and organizational foresight. Some believed that the focus should have been on preventing the disease from being forgotten or neglected (Jannin, 2000). Because sleeping sickness sufferers were of little economic interest, the disease was still not a priority. The combination of a neglected disease and a neglected population turned out to be particularly devastating in the affected regions (Jannin, Simarro, Louis, 2003).

In 1999 the WHO launched an international campaign to promote the control of sleeping sickness. The same year, at MSF’s urging, a group of experts was formed under United Nations coordination. It issued two recommendations. First, it stressed the importance of investing in research for new drugs to treat the disease. Second, because the short-and mediumterm results from this type of research are unpredictable, the experts recommended that combinations of existing drugsIn 2000 Epicentre and MSF-France developed the first protocol for a combination therapy study in Uganda.

also be studied. This was when public-private partnerships were formed, and pharmaceutical firms such as Aventis, Bayer, and Bristol-Myers Squibb (BMS) joined the WHO in the fight against sleeping sickness.

In July 2000 the Pan-African Trypanosomiasis and Tsetse Eradication Campaign (PATTEC) was launched at the OAU’s Summit of Heads of State and Government in Lomé. PATTEC concentrated its efforts on preparing a major anti-vector campaign, in combination with an effort to improve disease surveillance and treatment. It also worked to encourage the development of new alternatives to resistance-prone drugs.

MSF co-founded and launched the Drugs for Neglected Diseases Initiative (DNDi) in 2003, an independent, non-profit drug research and development foundation. Sleeping sickness was one of its top priorities.For more on DNDi missions, see the study by C. Vidal and J. Pinel in this volume.

DNDi has focused its efforts on identifying promising new drugs, and hopes to begin clinical trials for one of them—fexinidazole. Since 2005, it has also been part of the Nifurtimox Eflornithine Combination Therapy (NECT) study, launched by Epicentre and MSF-Holland in 2003.

DNDi also set up an R&D network with academic research groups at Japan’s Kitasato Institute, to develop two new clinical trial–ready drug candidates. Sleeping sickness control programs and researchers from endemic countries are also involved in these efforts (DNDi, November 2007).

From One Empirical Approach to Another: The Return of Dr. Jamot

MSF opened its first trypanosomiasis program in 1986, in the Moyo district of northern Uganda, and the teams faced the problem of how to manage the disease. This meant applying practices introduced by military doctors in the first half of the twentieth century.

The heart of Eugène Jamot’sMilitary doctor Eugène Jamot (1879–1937) developed a large-scale method for treating trypanosomiasis (Milleliri, 2004). He helped create the Service General Autonome de la Maladie du Sommeil in French West Africa and in Togo. strategy was mobile medicine and mass treatment. He increased the number of men and equipment in his teams, and sent them into the bush to screen for the disease and treat sufferers (Louis, Simarro, Lucas, 2002, p. 334). These principles, which would live on after him, involved systematic testing of all populations and a standardized regimen of diagnosis and immediate treatment. In order to succeed, Jamot developed a vertical strategy (devoted specifically to sleeping sickness) in high-prevalence areas.

When launching its sleeping sickness activities in Uganda in 1986, MSF found itself at a loss; with a lack of data, tools, and general interest, the teams were in an almost amateur situation. EpicentreEpicentre is an MSF satellite organization specializing in intervention epidemiology research.hadn’t yet been created, and it would be another nine years (1995) before MSF-Belgium would start a similar project. Having to overcome an unwillingness (both inside and outside MSF) to challenge outdated practices only made matters worse. Initially, the disease control strategy consisted in identifying large foci of infection that were priority zones for intervention. A mobile approach was only adopted later, thanks to information from local actors as foci were discovered. So micro-epidemiology supplanted macro-epidemiology, and the practice of targeting according to previous endemic episodes was dropped. The field teams drew a map of the disease and identified high-risk areas. In the early 1980s, some Ugandans had taken refuge in tsetse fly–infested parts of Sudan. They subsequently resettled in Uganda, in regions that were also infested by the vector. Once maps were drawn up using data on tsetse fly prevalence in these regions (obtained from vector control agencies), disease prevalence in the Sudanese migration zones, and the origin of patients, a more rational targeting of populations became possible.

On the operational level, in addition to the vertical approach, the teams set up a system of standardized patient cards that enabled them to monitor individual patients more closely and therefore analyze actions. Later, in 1994, Epicentre developed the Epitryp electronic database, which automated mapping and medical monitoring. This tool is still being used by MSF in the field. Monitoring entire patient cohorts allowed measuring the efficacy of treatment for the first time, and confirmed what practitioners already suspected regarding the dangers of melarsoprol, at a time when an alternative like eflornithine was already known. Practitioners also noted inconsistencies between their clinical observations in Uganda and the standard disease descriptions, and challenged the criteria for determining the disease stage (Dumas and Bouteille, 2002). Their clinical and biological data were used to create the first reliable screening and diagnosis algorithms, later adopted by other medical actors (Paquet, Castilla, Mbulamberi, et al., 1995).

Laboratory diagnosis at that time was pitiful and only the erythrocyte sedimentation rate was used. At the Antwerp Institute of Tropical Medicine, Magnus, Vervoort, and Van Meirvenne had developed a new test in 1978, the CATT. It had not yet been tested in vivo so an application to validate the tool was needed. In 1986–87, MSF embarked on a study of the test’s sensitivity and specificity. Epicentre then worked on the problem of how to interpret the results in previously treated patients with persistent antibodies (Paquet, Ancelle, Gastellu-Etchegorry, et al., 1992). Even now, the CATT does not have reliable sensitivity or specificity, and the accompanying microscopic examination is both laborious and insensitive, hence the large number of undetected cases. Moreover, the lack of adequate infrastructure makes the tests—which use molecular markers—difficult to use in rural endemic areas.

Determining the stages of the disease also remains an issue. The process involves analyzing cerebrospinal fluid in order to choose the appropriate treatment. This requires performing lumbar puncture, an invasive procedure that further complicates the diagnostic process. Better, simpler, more robust tests capable of determining the stage of the disease would revolutionize sleeping sickness control, making mass screening a more realistic objective.Laboratory diagnosis is still a major issue. Thanks to a donation from the Gates Foundation, the WHO and the Foundation for Innovative New Diagnostics have created a partnership to develop and assess new tests that are simpler, more accurate, and less expensive than existing ones.

The effectiveness in treating large endemic foci and a larger number of dispersed low-prevalence areas changed the epidemiological landscape, requiring new approaches that were less restrictive in terms of resources, and easier to implement.

In 2004, MSF-Holland launched a pilot strategy in the Congo Republic, nicknamed “Hit and Run,” based on observations: first, the limited number of programs using the eflornithinebased protocol for first-line treatment; and second, the fairly general reduction in prevalence in the region. Neither conventional programs nor active screening made sense anymore. What was needed was diagnosis and treatment at peripheral health posts, and rapid intervention in zones where cases were occurring. Using a horizontal approach integrated at the primary care level, these peripheral posts would provide passive screening and treatment. If several cases were detected from the same region, suggesting elevated prevalence in a localized area, a mobile team was sent to do active screening and on-the-spot treatment. The goal was to eliminate the disease at the source, even if this meant not fully eradicating it. The positive results obtained with this strategy led MSF to create and distribute a standard protocol for its field programs in endemic areas.

Monotherapy: Unorthodox Alliances

I n the early twentieth century, the first sleeping sickness treatment protocols used an arsenic derivative called atoxyl. A compound from the same family, melarsoprol, is still being used today. After tryparsamide was synthesized in the early 1920s, and the number of treatment failures grew, various arsenic-based combination therapies were tested. At the same time, non-arsenic-based medications such as pentamidine were developed in the 1940s and adopted for use in mass prophylaxis campaigns (Ollivier, Legros, 2001). There was a lot of inertia and very little originality in the development of new and effective drugs in the 1980s. Melarsoprol had devastating side effects, and eflornithine—a very promising drug synthesized in 1979—was still little known, in limited production, expensive, and required significant human and financial resources for poor countries. Nifurtimox, used for the first time in 1980, was not yet being considered for treatment of sleeping sickness because phase 2 and phase 3 clinical trials were needed to confirm its effects.Nifurtimox is a drug prescribed for treatment of Chagas disease. It has been used off-label for treating sleeping sickness, particularly in Zaire and Sudan in the 1980s. Cheap and easy to use, promising in combination, it appears to be better tolerated than melarsoprol, despite some side effects, which include anorexia, weight loss, and neurological effects. A donation of nifurtimox for use in treatment of sleeping sickness was obtained through a 2004 agreement between Bayer and the WHO. This medication was intended for clinical trials and for compassionate use, usually off-label, as an alternative to eflornithine in melarsoprol failure.

Proving the Efficacy or Inefficacy of Existing Drugs

Studies on trypanosome’s melarsoprol resistance were the starting point for MSF’s recent innovations for treating stage 2 of the disease.

IIn 1949, Friedheim developed the organic arsenical melarsoprol (or Mel B), an atoxyl derivative whose advantage is that it lacks tryparsamide’s potential for optic nerve toxicity. As the standard treatment for stage 2, it helps—at least partially— with the problem of arsenic resistance, but at the cost of deadly arsenic-related encephalopathies (5% to 10%). It also has to be administered under direct medical supervision in a hospital. Intravenous administration and toxicity therefore limit its use in everyday practice, particularly by mobile treatment teams. MSF quickly discovered the limits of using this standard drug in the field. A follow-up analysis in 1994 on the seven thousand patients included in the program since 1986 in Uganda added to fears about melarsoprol use.Epicentre set up a permanent base in Uganda to help MSF projects with their research and innovation efforts. In 1992 and 1993 there were two outbreaks of post-treatment reactive encephalopathy at the sleeping sickness treatment center in Adjumani, during which more than 10% of patients treated with melarsoprol developed arsenical encephalopathy. Retrospective studies helped identify exogenous risk factors—in particular, concurrent use of an anti-parasitic agent (thiabendazole) and poor general health at the start of treatment (Ancelle et al., 1994). A few years later, Epicentre published the first study showing an increased failure rate with melarsoprol in the Omugo program in northern Uganda. While the expected rate was usually between 3% and 9%, the study revealed that nearly 30% of the 428 patients treated from 1995 to 1997 were in treatment failure (Legros, Fournier, Gastellu-Etchegorry, et al., 1999a; Legros et al., 1999b).

In paediatrics, a retrospective case study of children under six years old treated in an MSF sleeping sickness treatment center in southern Sudan between 2000 and 2002 showed that clinical signs alone were not adequate for determining the stage of the disease. Laboratory staging was therefore essential to avoid the unnecessary use of melarsoprol, with its adverse effects (Eperon et al., 2007). In addition, melarsoprol was just as toxic in this age group, making first-line use of eflornithine absolutely necessary in both children and adults.

Ultimately, melarsoprol resistance—brought to light in 1999—was a major argument for prioritizing research for stage 2 sleeping sickness treatments. The scientific community and pharmaceutical companies failed to heed the call, however, so MSF began looking for alternatives. This included making better use of drugs already on the market—hence the interest in eflornithine.

Also known as DFMO or Ornidyl, eflornithine is an antimitotic agent approved for the treatment of cancer. Its antiprotozoal properties in late-stage sleeping sickness were confirmed by Bacchi and his team in the 1980s (Bacchi, Nathan, Hutner, et al., 1980; McCann, Bacchi, Clarkson, et al., 1981). Van Nieuwenhove put it to the test in southern Sudan, where about twenty sleeping sickness patients—eighteen of them in stage 2 and sixteen of them arsenical-resistant—received oral eflornithine monotherapy (Van Nieuwenhove, Schechter, Declercq, et al., 1985). Soon after the study’s publication in 1990, the drug was approved for this indication by the US Food and Drug Administration.

Eflornithine’s advantage lay in its relative lack of toxicity compared to melarsoprol and its efficacy in arsenical-resistant patients. It was, however, very complex to use (four infusions a day for fourteen days) and too expensive for African populations, and so it went out of production in 1995.

In early 2000 there was little in scientific literature about eflornithine. The few studies being done showed a nearabsence of the encephalopathies observed with melarsoprol, but a substantial mortality rate of between 2% and 6%. A series of studies in Africa was therefore needed to confirm expectations for the drug. That year, MSF-France took over a trypanosomiasis program that had been started by MSF-Holland in Ibba, Sudan, in 1999. After just over a year, the epidemiological surveillance system in the remote, conflict-ridden area revealed worrying melarsoprol failure rates of around 30%. Because eflornithine was not available, MSF began using a new protocol combining melarsoprol and nifurtimox. In September 2001, however—thanks to a drug donation from Aventis and Sudanese government authorization and involvement—eflornithine was adopted for first-line, fourteen-day monotherapy. This illustrates the fundamental role played by the authorities in MSFintervention countries in the implementation, follow-up, and success of studies. Ibba became the first project in Africa to use eflornithine as the first-line treatment for stage 2 sleeping sickness—despite the objections of some within MSF, who felt there was not enough evidence to support using this protocol.

The results were positive: 1% mortality in the first cohort, then 1.2% in larger cohorts, with a drug that was well tolerated and effective (Priotto, 2008). At the same time, it appeared that the quality of nursing care played a crucial role in the treatment. Indeed, the results for the first Ibba cohort showed that mortality could be attributed, in part, to infections due to poor infusion techniques. Improvements in nursing care helped make the use of eflornithine safer.

MSF-Switzerland also had a program in southern Sudan, which started in 2000 in Kajo-Keji. The results from Ibba justified introducing eflornithine as the first-line treatment there in January 2003. The protocol change was a success. A retrospective comparison between the melarsoprol cohort (June 2000 to December 2002, n = 708) and the eflornithine cohort (January 2003 to December 2003, n = 251) showed a significant difference in mortality during treatment (3.5% for melarsoprol, versus 0.8% for eflornithine), and no difference in mortality or relapse rates at twelve months post-treatment.Several recommendations resulted from this study: first, that eflornithine be the drug of choice for stage 2 of the disease, not just in areas of high melarsoprol resistance; and second, that research on combination therapies be undertaken to minimize the emergence of resistance (Chappuis et al., 2005).The Republic of Congo also provided fertile ground for progress on sleeping sickness. A retrospective study using data collected on the Gamboma, Bouenza, and Mossaka programs (MSF-Holland) from 2001 to 2005 confirmed earlier results and lent strength to the idea that eflornithine was less toxic and more effective than melarsoprol. The authors then recommended eflornithine as the first-line treatment in the Republic of Congo.Making Eflornithine Available

In 1995, it was announced that eflornithine would no longer be produced, due to its prohibitive cost for African countries.

The company that made eflornithine—in the midst of a merger with Hoechst—was selling the anti-parasitic at cost price. Not making any profit, it granted licensing rights for the drug to the WHO, which would have to find another company to manufacture it. Though dozens of drug companies were contacted, none offered to do it at a low enough price. Only the original manufacturer—which had by then become Aventis— was willing to provide the WHO and MSF with a limited and irregular supply, until the inventory on hand was exhausted.

In the field, lots were delivered sparingly for clinical trials or for compassionate use in melarsoprol failure. In 1999, MSF received one of Aventis’ last ad hoc donations—ten thousand ampoules of eflornithine from the 1995 production (they had been used to treat relapses in Uganda). Jean Jannin of the WHO then confirmed that “the situation was becoming downright worrying. It was obvious that eflornithine would run out before the end of 2001. The situation for other treatments wasn’t much better. Whether intended for the early stage of the disease (pentamidine and suramin) or the late stage (melarsoprol and nifurtimox), all drugs were more or less in danger of running out.Bayer and Aventis weren’t sure whether they would continue production” (Libération, February 21, 2001).

MSF considered having the drug manufactured, but it was a complex process, beyond the scope of the organization’s expertise. As it turned out, production was rescued by an eflornithine-based depilatory cream marketed by BMS under the name Vaniqa. MSF and the Campaign for Access to Essential Medicines then joined forces to create a lobbying group, and approach Aventis and BMS, while pushing the WHO to shoulder the responsibility.

On May 3, 2001, after two years of talks, MSF, Aventis, Bayer, and other private partners established with the WHO a publicprivate partnership to fight sleeping sickness. Aventis signed a $25 million, five-year donation agreement (2001–2006). This collaboration agreement with the WHO had three components:

• Drug donation: the company pledged to produce and supply eflornithine, melarsoprol, and pentamidine free of charge.Sleeping Sickness. MSF Fact Sheet, MSF and the Campaign for Access to Essential Medicines, May 2004. Storage, distribution, and delivery of the donated drugs would be the responsibility of MSF-Logistique in Merignac, which would be managing the stock for all prescribers worldwide. All orders from national programs and NGOs would need prior WHO approval. When the agreement ended in 2006, Aventis said they wanted to transfer the technology to another pharmaceutical company, if one would agree to continue production (Prescrire, 2006).

• Surveillance and control: with agreed financial backing from Aventis, the WHO pledged to speed up disease control activities by supporting national programs, in particular for training medical personnel and routine screening in endemic zones.

• Research and development: financial support was also intended to help organize new research via the WHO’s Program for Research and Training in Tropical Diseases (TDR),Supported by the United Nations Children’s Fund, the United Nations Development Programme, the World Bank, and the WHO. with an emphasis on adapting treatments, developing new methods for making one of the existing products and, ultimately, developing new drugs.

In agreement with Aventis, BMS promised to finance a one-year supply (60,000 vials) of eflornithine’s active ingredient, and pledged $400,000 over two years to the WHO to support treatment efforts. In 2002, Bayer agreed to supply as much suramin as the WHO felt necessary, free of charge, for five years. At the same time, the company supported studies on the use of nifurtimox for treating sleeping sickness.

A WHO press release on the Aventis–WHO agreement cited MSF’s central role: “Both WHO and Aventis commend nongovernmental organizations, especially MSF, for the public awareness raised on sleeping sickness affecting poor African countries” (WHO, 2001). What made these alliances unique was the crossing of interests between, on the one hand, humanitarian organizations, concerned with the urgent need for treatment, and, on the other, companies, concerned about their corporate image. Within MSF, the strategy was a delicate balance between opportunism and risk-taking, not to mention the ambiguity of simultaneously challenging and collaborating with pharmaceutical firms.

In 2006, Sanofi Aventis and the WHO signed a new agreement extending the five-year agreement signed in 2001.The terms of the earlier agreement regarding the production and donation of drugs (pentamidine, melarsoprol, and eflornithine) were renewed, in addition to a $5 million donation. Again, there was an additional financial contribution of $20 million to fight the four most neglected tropical diseases (trypanosomiasis, leishmaniasis, Buruli ulcer, and Chagas disease).

Simplifying Treatment Delivery to Patients

In 2006—five years after free access, and before the WHO– Sanofi Aventis agreement was renewed—only 20% of patients were getting first-line eflornithine. These were primarily the patients treated by MSF. Simplified distribution and a free supply of the infusions and other solutions needed by national programs to implement the protocol were essential. The DFMO (eflornithine) kit was therefore created, and presented by MSF to the 28th Meeting of the International Scientific Council for Trypanosomiasis Research and Control (ISCTRC)The ISCTRC is a statutory body of the African Union; it meets every two years. in Addis Ababa in 2005. The WHO and MSF-Logistique worked together to create the kits, which were ready in 2007. The 2006 Sanofi– WHO agreement ensured that production and supply would be free (to the capitals of requesting countries).

Combination Therapies: Daring Clinical Trials

It is currently believed that it will be more than ten years before there is a new drug for treating stage 2 sleeping sickness. In the meantime, drug combinations, which are easier to administer and require smaller doses and a shorter course of treatment, are considered the best way to meet patient needs. They also provide relative protection against resistance.

As early as 1998, a WHO technical report judged it “essential that the possibility of drug combinations be thoroughly explored” (WHO, 1998) and, in 1999, the Human African Trypanosomiasis Treatment and Drug Resistance Network, to which MSF belonged, identified combination therapies, including nifurtimox, as one of its research priorities (WHO, 2001). It was also at this time that the internal MSF sleeping sickness working group was created. The group’s purpose was to highlight priorities in sleeping sickness treatment, centralize information coming from different field projects, and harmonize recommendations.

In early 2001 the first MSF patients were enrolled in the three bi-therapy trials at the Omugo site in Uganda. This was a comparative trial, in which patients were randomly allocated to three groups for the three possible combinations (nifurtimoxeflornithine, nifurtimox-melarsoprol, and eflornithine-melarsoprol). This first study, highly exploratory in design, proposed starting directly in phase 3, thereby bypassing the pre-clinical and clinical phases 1 and 2, which would have required an enormous amount of funding. The new protocol proposed four injections per day of eflornithine, but for only one week instead of two, and ten days of nifurtimox instead of thirty. The study began in 2001, but soon had to be stopped because two of the groups turned out to be toxic, with four deaths in the melarsoprol-nifurtimox group and one death in the melarsoprol-eflornithine group. There was also a significant drop in prevalence.The MSF project was already showing good results by the time the Epicentre study got underway! Not enough patients could be recruited to provide conclusive statistics for the otherwise encouraging nifurtimoxeflornithine group (Priotto et al., 2006).

To remedy this situation, the Nifurtimox-Eflornithine Case Series study was launched at MSF’s Yumbé site in Uganda in 2002.

The results of the two studies were unprecedented: a 100% cure rate for the nifurtimox-eflornithine combination, with no deaths or relapses at two years—but the sample size was still considered inadequate to confirm these conclusions.

The NECT study protocol was written by Epicentre in 2002 in collaboration with an international scientific committee. It proposed comparing nifurtimox-eflornithine to eflornithine alone. The aim was to prove that the combination of these two drugs was as safe and effective as eflornithine monotherapy and also easier to use. The study design was based on these constraints. It contained three important changes: the combination of eflornithine with nifurtimox, a reduction in the number of daily infusions from four to two while keeping the total daily dose the same, and a shorter duration of treatment, from fourteen days to seven. It was not yet known how the drugs would react in combination.

There was some opposition, at the WHO in particular, but also within MSF and DNDi. The study was delayed for a while for lack of nifurtimox, which the WHO refused to supply, citing, on one hand, the lack of knowledge about its effects (research on which dates back to the mid-twentieth century), and, on the other, the failure to respect biomedical ethics and to conform to usual clinical trial standards—a conflict reminiscent of the 1990s epistemic crisis in the research world (Dodier, Barbot, 2000). There is still some debate over the benefits and limitations of the Guidelines for Good Clinical Practice (ICH, 1996),GCPs are rules written and imposed by industrialized nations to prevent scientific fraud, improve documentation, and ensure respect for the Declaration of Helsinki (1964), which protects the rights of patients participating in clinical trials. which affect research in general, especially into tropical diseases in the developing world (White, 2006). As with the ACT UP model for AIDS in the 1980s, the urgency of finding answers justified taking risks and ignoring standard rules in the development of new treatments. Epicentre and MSF adapted their practices and GCPs to suit the intervention context, while preserving the basic ethical and methodological rules related to safety and patient consent.

The NECT study began in 2003 in a trypanosomiasis control program run by MSF-Holland at the Nkayi site in the Congo Republic, thanks to a specific nifurtimox donation by Bayer. Unfortunately, supply and authorization problems meant the program got off to a late start, and after more than a year and a half of recruitment efforts, the sample size was still too small to complete the study.

The study required a sample size of 280 patients, which implied a multi-center study. Epicentre decided to explore other operational sites, and came to an agreement with MSFBelgium, which in 2004 was preparing to open a project in the Democratic Republic of Congo (DRC). The two main opponents to these trials, DNDi and WHO–TDR, joined the study in 2005.

Without the experience in Nkayi, it is likely that the multicenter study would have been delayed for another four or five years because of the strict standard recommendations. MSF and Epicentre decided to risk using the NECT protocol, relying on the initial results to convince skeptics that the trials should progress much more quickly.

The study became multi-centric, with three major sponsors: MSF-Holland, DNDi, and the WHO’s TDR program. The TDRfunded sites later withdrew, however, and Epicentre’s analysis included only the results from the other four.

The study at the first center, launched at the Nkayi site in 2003 by MSF, has ended. The combination is better tolerated than eflornithine alone, and its non-inferiority has held so far (two relapses in each group). The hospitalization stages for the other three sites in the DRC ended in late 2006, and there will be analyses on patient follow-up. This trial is undoubtedly an innovation in the treatment of stage 2 sleeping sickness (Priotto et al., 2009). Compared to eflornithine alone, the eflornithinenifurtimox combination is easier to use, less expensive, and less toxic (it causes less neutropenia, thus reducing the risk of infection).

Since 2004 MSF has been treating 20% to 30% of all sleeping sickness cases using first-line eflornithine for stage 2 of the disease in all its programs. With the introduction of the DFMO (eflornithine) kits, there is a realistic hope that more cases in Africa will be treated using this protocol. MSF had made a crucial contribution in research and innovation by accurately and systematically documenting its experiences, and by publishing and presenting its results in international forums. MSF has institutional expertise and responsibility with regard to sleeping sickness, making it one of the foremost authorities on medical treatment of the disease. Vector control, on the other hand, is not one of its strong points.

Sleeping sickness prevalence remains high in the DRC, Angola, and southern Sudan. MSF, Epicentre, DNDi, and the WHO are collaborating in the fight against trypanosomiasis, working together to help health ministries in the affected countries draft protocols and other guidelines, as well as train care providers and technical staff. The difficulty MSF now faces lies with its operational choices. There is a need to consolidate knowledge, but the drop in prevalence and the replacement of large foci by scattered, smaller foci means that intervention by specialized MSF programs can no longer be justified.

A medium-term commitment, for example with a national or trans-national mobile team capable of responding to alerts from different endemic zones, is one possibility; in any case, the therapeutic innovations from the NECT trial will appear all the more pertinent when applied in the field. A massive effort will be needed to get this protocol recommended as the first-line therapy.

Bibliography

Ancelle, T., B. Barret, L. Flachet, A. Moren. 1994. “Étude de deux épidémies d’encéphalopathies arsénicales dans le traitement de la trypanosomiase, Uganda, 1992–1993.” Bulletin de la Société de Pathologie Exotique 87 (5): 341–346.

Bacchi, C. J., H. C. Nathan, S. H. Hutner, et al. 1980. “Polyamine metabolism: a potential therapeutic target in trypanosomes.” Science 210 (4467): 332–334.

Blondel, D. 2002. “Le rôle des scientifiques dans le processus d’innovation.” In N. Altrer, editor, Les logiques de l’innovation. Paris: La Découverte&Syros.

Cattand, P. 2001. “L’épidémiologie de la trypanosomiase humaine africaine: une histoire multifactorielle complexe.” Médecine Tropicale 61 (4-5): 313–322.

Chappuis, F., N. Udayraj, K. Steitenroth, A. Meussen, P.A. Bovieret. 2005. Eflornithine is Safer than Melarsoprol for the Treatment of Second-Stage Trypanosoma brucei gambiense Human African Trypanosomiasis.” Clinical Infectious Diseases 41 (5): 748–751.

Courtin, F., V. Jamonneau, G. Duvallet, A. Garcia, B. Coulibaly, J.P. Doumenge, G. Cuny, P. Solano. 2008. “Sleeping sickness in West Africa (1906–2006): changes in spatial repartition and lessons from the past.” Tropical Medicine & International Health 13 (3): 334–344.

Dodier, N., J. Barbot. 2000. “Le temps des tensions épistémiques: le développement des essais thérapeutiques dans le cadre du sida.” Revue française de Sociologie 41 (1): 79–118.

Drugs for Neglected Diseases Initiative. 2006. DNDi newsletter. No. 14. http://www.dndi.org/newsletters/n14/page1.htm.

Dumas, M., B. Bouteille. 2002. “La trypanosomose humaine africaine: propos sur le traitement actuel et les perspectives.” Bulletin de la Société de Pathologie Exotique 95 (5): 341–344.

Eperon, G., C. Schmid, L. Loutan, F. Chappuis. 2007. “Clinical presentation and treatment outcome of sleeping sickness in Sudanese pre-school children.” Acta Tropica 101 (1): 31–39.

Hamel, A. 2001. “Genèse de la Campagne pour l’Accès aux Médicaments Essentiels.” Humanitaire 3: 115–123.

International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use. 1996. ICH Harmonised Tripartite Guideline. Guideline for Good Clinical Practice. Current Step 4 version.

Jannin, J. 2000. “Actualités de la trypanosomiase humaine.” Médecine Tropicale 60 (2S): 56–57.

Jannin, J., P. P. Simarro, F. J. Louis. 2003. “Le concept de maladie négligée.” Médecine Tropicale 63 (3): 219–221.

Legros, D., C. Fournier, M. Gastellu-Etchegorry, F. Maiso, E. Szumilin. 1999a. “Échecs thérapeutiques du mélarsoprol parmi des patients traités au stade tardif de trypanosomose humaine africaine à T. b. gambiense en Ouganda.” Bulletin de la Société de Pathologie Exotique 92 (3): 171–172.

Legros, D., S. Evans, F. Maiso, J.C Enyaru, D. Mbulamberi. 1999b. “Risk factors for treatment failure after melarsoprol for Trypanosoma brucei gambiense trypanosomiasis in Uganda.” Transactions of the Royal Society of Tropical Medicine and Hygiene 93 (4): 439–442.

Louis, F. J., P. P. Simarro, P. Lucas. 2002. “Maladie du sommeil: cent ans d’évolution des stratégies de lutte.” Bulletin de la Société de Pathologie Exotique 95 (5): 331–336.

McCann, P.P., C.J. Bacchi, A.B. Clarkson, J.R Seed, H.C Nathan, B.O. Amole, S.H. Hunter, A. Sjoerdsma. 1981. “Further studies on difluoromethylornithine in African trypanosomes.” Medical Biology 59 (5–6): 434–40.

Milleliri, J.-M. 2004. “Jamot, cet inconnu.” Bulletin de la Société de Pathologie Exotique 97 (3): 213–222.

Olivier, G., D. Legros. 2001. “Trypanosomiase humaine africaine: historique de la thérapeutique et de ses échecs.” Tropical Medicine and International Health 6 (11): 855–863.

Paquet C., T. Ancelle, M. Gastellu-Etchegorry, J. Castilla, I. Harndt. 1992. “Persistence of antibodies to Trypanosoma brucei gambiense after treatment of human trypanosomiasis in Uganda.” The Lancet 340 (8813): 250.

Paquet C., J. Castilla, D. Mbulamberi, M.F. Beaulieu, M. GastelluEtchegorry, A. Moren. 1995. “La trypanosomiase à Trypanosoma brucei gambiense dans le foyer du Nord-Ouest de l’Ouganda. Bilan de 5 années de lutte (1987–1991).” Bulletin de la Société de Pathologie Exotique 88 (1): 38–41.

Prescrire. 2006. “Editorial. Maladie du sommeil: la cosmétologie au secours de la santé publique.” Prescrire 26 (269): 135–136.

Priotto G., C. Fogg, M. Balasegaram, O. Erphas, A. Louga, F. Checci, S. Ghabri, P. Piola. 2006. “Three Drug Combinations for Late-Stage Trypanosoma brucei gambiense Sleeping Sickness: A Randomised Clinical Trial in Uganda.” doi:10.1371/journal. pctr.0010039. PLoS Clinical Trials 1 (8): e39.

Priotto G., L. Pinoges, I. B. Fursa, B. Burke, N. Nicolay, G. Grillet, C. Hewison, M. Balasegaram. 2008. “Safety and effectiveness of first line eflornithine for Trypanosoma brucei gambiense sleeping sickness in Sudan: cohort study.” doi:10.1136/bmj.39485.592674. BE. British Medical Journal 336: 705–708,

Priotto G., S. Kasparian, W. Mutombo, D. Ngouama, S. Ghorashian. 2009. “Multicentre randomised non-inferiority trial of nifurtimox-eflornithine combination therapy for second-stage gambiense sleeping sickness.” The Lancet 374 (9683): 56–64.

Van Nieuwenhove S., P. J. Schechter, J. Declercq, G. Bone, J. Burke, A. Sjoerdsma. 1985. “Treatment of gambiense sleeping sickness in the Sudan with oral DFMO (DL-alpha-difluoromethylornithine), an inhibitor of ornithine decarboxylase; first field trial.” Transactions of the Royal Society of Tropical Medicine and Hygiene 79 (5): 692–698.

World Health Organization (WHO). 1998. Control and Surveillance of African Trypanosomiasis. Report of a WHO Expert Committee, Technical Report Series 881. Geneva: World Health Organization.

–––. 2001a. “World Health Organization and Aventis Announce a Major Initiative to Step Up Efforts Against Sleeping Sickness.” Press release. 3 May. Geneva.

–––. 2001b. HAT Treatment and Drug Resistance Network. Report of the Second and Third Steering Committee Meetings. 29 September 1999, Mombasa, Kenya; 28–30 May 2000, Bruges, Belgium, WHO/CDS/ CSR/EDC/2001.13.

White, N. J. 2006. “Editorial. Clinical trials in tropical diseases: a politically incorrect view.” Tropical Medicine & International Health 11 (10): 1483–1484.