Testing new drugs on the world’s poor
Many medicines developed for the rich are tested on poor people from developing countries. This is a failure in the allocation of scientific resources and, worse, a threat to the public health of the entire world.
By Sonia Shah
Read also by Sonia Shah: The Body Hunters: Testing New Drugs on the World's Poorest Patients
Never before have drug-makers lavished so much attention upon the world’s poor. Boehringer Ingelheim’s sparkling test labs in South Africa are outside the sprawling slums ringing Capetown. Novartis’ gleaming white facility, where researchers develop new drugs in India, nestles against the smog-blackened slums of Mumbai. In recent years Pfizer, GlaxoSmithKline (GSK) and AstraZeneca have also set up global clinical-trial hubs in India.
But they are not there to cure the impoverished sick who line up at their shiny research clinics. Drug companies have migrated to developing countries to carry out experiments. In 2006 more than half GSK’s drug trials took place outside western markets, mainly in “low cost” countries like Bulgaria, Zambia, Brazil and India to which tens of thousands of clinical trials have been outsourced (1).
The companies built these laboratories to produce medicines aimed at wealthy westerners with age-related conditions such as heart disease, arthritis, hypertension and osteoporosis. With the average American bringing home 10 prescription drugs every year, the United States is the biggest drug market in the world. The pharmaceutical industry has grown by 15% every year since 2000 and tripled its output of experimental drugs between 1970 and 1990. This is thanks in large part to changes in US drug regulations. In 1984 the US Food and Drug Administration (FDA) extended drugmaker’s patents on new drugs; in 1992 they started accepting payments from drug companies in return for quicker reviews of new drug applications; and in 1997 the agency gutted rules banning television ads for new drugs.
The more drugs Americans enjoy, the less willing they seem to enrol in the clinical trials required to develop new ones. Each new drug requires over 4,000 patients enrolled in clinical trials, which in turn means that over 100,000 people must be enticed into test clinics for initial screenings (2).
Why so many? It doesn’t take many test subjects to prove the effectiveness of, say, insulin for people in diabetic comas, because the drug’s effect is so dramatic. But it is far more difficult to prove the effectiveness of new drugs for heart disease, arthritis, hypertension and other chronic conditions: and, despite the industry’s best efforts, most new drugs aimed at these illnesses are only marginally effective. Some are hardly better than a placebo. “You usually have to struggle to find a difference” between treated and untreated patients, said one veteran clinical researcher.
The industry’s need for experimental subjects is therefore vast. Yet fewer than one in 20 Americans can be bothered to take part in clinical trials. To short-circuit this problem, drug-makers often run tests comparing their new drug’s effect against that of a placebo. Fewer subjects are needed for such trials and proving that a new drug works better than nothing is all that the FDA requires.
Pressing need for trials
The only trouble with placebo trials is that they require sufficient numbers of people who are willing to participate in an experiment where they may get no active treatment at all – an increasingly difficult task, especially in the drug-marinated West where it is difficult to find “treatment-naïve” patients (sick people too poor to get medical treatment) and where ethics demand the accustomed minimum standard of care. As a result, 80% of the drug industry’s clinical trials fail to meet their recruitment deadlines. Each day that a drug remains locked up in development, drug companies haemorrhage some $1m in lost sales, while their rivals beat them to market.
The sicknesses endemic in poor countries are a low priority when a $200m market is the bare minimum to elicit industry interest. But the populations of developing countries do not suffer only from malaria and tuberculosis. According to the World Health Organisation (WHO), 80% of deaths from chronic non-communicable diseases now occur in developing countries. Ninety percent of the world’s cases of type 2 diabetes occur in India and China. In some parts of Africa, one in five suffers from diabetes, and 20 million Africans suffer from hypertension (3).
According to the WHO, the public health implications of this phenomenon are “staggering and are already becoming apparent”. Patients being treated for these conditions inevitably suffer more complications than do those in the West. This presents an opportunity for industry trials. To prove that a heart drug works, for example, one needs to show that people who take the drug suffer fewer “events” (heart attacks and deaths) than those who do not. Trials in poor countries can therefore be completed much faster. As an executive from a clinical trials company noted during a conference presentation on the desirability of poor countries for clinical trials: “If you don’t have enough events, you’ll never finish your trial.”
“South Africa is a great country [for Aids],” another clinical trials company executive told me, because of the large number of HIV-infected patients who have yet to be treated with antiretroviral drugs. Treatment-naïve patients are at a great premium in clinical trials.
Clinical trials companies (also called contract research organisations or CROs) specialise in conducting clinical trials in foreign countries on behalf of the major pharmaceutical companies. Quintiles, Covance, Charles River Laboratories and PPD (see graph below) have offices and facilities in many developing countries. Quintiles, for example, has clinical sites in Chile, Mexico, Brazil, Bulgaria, Estonia, Romania, Croatia, Latvia, South Africa, India, Malaysia, the Philippines and Thailand.
In 25 years the number of therapeutic trials has multiplied by seven. Clinical trials companies are operating on every continent. The sector’s turnover is growing by nearly 20% a year and reached $15.4bn in 2006. Source: Thomson CenterWatch analysis of company reports, 2006.
The main attraction of poorer countries is rapidity. It can take months and even years to recruit sufficient numbers of test subjects in trials in western countries. In South Africa, Quintiles enrolled 3,000 patients for an experimental vaccine trial in nine days and 1,388 children for another trial in 12 days. In the US between 40% and 70% of enrolled subjects hem and haw and eventually drop out of clinical trials: in India clinical trials companies say they retain 99.5% of their enrolled subjects (4).
Covance says it can run trials at 25,000 separate medical sites in over a dozen countries. The trade press of the clinical trials industry brims with enthusiastic articles. “Ski where the snow is,” recommends one ad from a CRO. “Conduct clinical trials where the patients are.”
Win-win, for some
Defenders claim that this is a win-win situation. Clinical trials offer better care than local clinics where patients can wait for entire days to see overworked, under-resourced staff. Poor patients should consider themselves lucky – and their alacrity to participate suggests that they know it. Clinics and hospitals in poor countries get access to the latest technology and frequently capitalise on the new equipment supplied by drug companies for the trials. “We got some equipment,” said one clinical researcher in India, “and they didn’t ask for it back.”
Being a human guinea pig may be work that westerners no longer want to do, but that doesn’t rule it out as a good deal for the poor, who benefit from the best care and get paid for it. If factories can be relocated to take advantage of lower salaries or less rigorous environmental constraints, why not clinical trials? “They said [I was] taking advantage,” complained an industry researcher criticised for conducting trials in poor countries. “But without that trial, those children would be dead.” “I think it is usually good for people to be in clinical trials,” said FDA medical director Robert Temple. “Half of the people [get an active drug] and better care. The other half…[get] better care.”
And yet offering one’s body to science is not the same as a day job at the factory. Even a sweatshop job, in the usual course of things, will offer quantifiable benefits to the individual, however slim. The clinical trial can make no such promises. This uncertainty, of course, is part of the reason why an experiment is called for in the first place.
The cornerstone of ethical research on humans, as encoded in a myriad of documents (notably the Nuremberg Code, adopted while Nazi doctors were on trial in 1947, and the World Medical Association’s Declaration of Helsinki, agreed in 1964 and updated in 2004) is that research subjects should give informed and voluntary consent. The notion of coercion should include over-generous compensation packages. When Aids activists demanded that researchers guarantee lifelong HIV treatment for subjects who became infected in the course of experimental vaccine trials, researchers argued that such a requirement violated the principle of voluntary consent. The deal was just too good: even uninfected people might enrol just for the sake of the free drugs.
And yet a growing body of evidence suggests that experimental subjects in developing countries are not voluntarily consenting to be experimented upon. Bioethicists track the number of people who refuse to participate in or drop out of trials as a sort of after-the-fact indicator: by refusing to participate or dropping out, subjects show that they understand that their participation in trials is voluntary. Refusal and drop-out rates in western trials can run up to 40% or more. But many clinical investigators in developing countries stated that potential subjects never refused to participate in trials. The great speed of recruitment in these trials is another indication of this geographical discrepancy (5).
In some trials up to 80% of the subjects were unaware that they were free to leave – evidence of coercion that was used as a reason to conduct more trials. According to an article in the magazine Applied Clinical Trials, Russian subjects “don’t miss appointments, they take all their required pills… and only very rarely do they withdraw their consent…. Russian subjects do what their doctors tell them to do. What a phenomenon!” A report by the clinical trials industry consultancy CenterWatch similarly noted that “the Chinese are not that fully emancipated as in the US. They are more willing to be guinea pigs.”
Oversight is minimal. Data from overseas trials are accepted by American and European regulatory authorities, but neither requires that drug-makers alert them before they start their overseas trials. The sole requirement is that the Declaration of Helsinki or local rules, whichever afford more protection, are observed. If these trials fail – and 90% of drugs entering clinical trials do fail – the results simply vanish without a trace.
Lucrative income stream
Local ethics committees and regulators are supposed to ensure that subjects’ rights are protected: this would probably be fine if they were up to the task. Government officials in India focus on encouraging clinical trials, which they see as a lucrative income stream. Various officials are on record as saying they hope to increase the value of industry-sponsored trials in India from $70m a year to $1bn. The regulations have been relaxed to allow various exemptions from customs duties and taxation: phase 3 trials no longer have to be completed elsewhere before launching in India: experimental drugs no longer have to demonstrate any special value to India (according to Ken Getz of CenterWatch, who described to me how he was feted in India as if he were a head of state).
According to The Economic Times, the country’s leading business newspaper: “The opportunities are huge, the multinationals are eager, the Indian companies are willing. We have the skills, we have the people, and we have an advantage which China doesn’t and probably never will. Best of all, this is one sort of outsourcing which American workers aren’t likely to protest” (6).
In almost every area of medical practice and research in India, there is a glaring lack of regulation. Medical schools have been caught hiring fake teachers to fool inspectors, selling admissions and auctioning medical degrees. Six hundred inspectors struggle to contend with a pharmaceutical market that is saturated with ineffective or dangerous products. And, according to prominent drug-expert Chandra Gulhati, editor of the Monthly Index of Medical Specialties in India: “Even if an erring company is caught red-handed indulging in illegal activities, it is let off, for reasons best known to regulators, with a light warning.”
Not surprisingly, there has been a whole series of scandals. During the 1970s an unapproved antimalarial drug, quinacrine, was distributed to hundreds of thousands of illiterate women, triggering permanent sterilisation. In the 1980s an injectable contraceptive, that had been withdrawn from the market when it was found to cause tumours in rats, was tested on villagers who later said they had “no idea they were participating in a trial”.
In the late 1990s government researchers purposely withheld treatment from over 1,100 illiterate women with precancerous lesions on their cervices, in order to study the progression of the disease. These subjects (like those involved in the notorious Tuskegee study) (7) were later found to be uninformed and non-consenting. In 2001, in the state of Kerala, a researcher from Johns Hopkins university was caught testing an experimental cancer drug on patients before it had been proven safe in animals. In 2003 an experimental cancer drug was administered to over 400 women seeking to boost their fertility: the drug was toxic to embryos. None of these scandals, well-publicised in the press, have led to any legal protection of research subjects.
Unethically conducted clinical trials also undermine the legitimacy of western medicine among people in the developing world. To give just two examples: South Africa’s health minister condemned HIV medicines as poisons: and Nigerian religious officials rejected a polio vaccine as dangerous. The spectre of a boom in poorly regulated, secretive clinical trials inflames such reactions, which have public health implications for all of us.
This argument is rarely so baldly stated, but it is a common undercurrent. It may be true that the quality of care in clinical trials is often superior to regular care, and that clinicians involved in trials get access to the latest technology and tools and to desperately needed income that they can divert to caring for patients. But the data itself cannot be equated with medical progress, as anyone who has seen cutting-edge vaccines rotting in tropical warehouses can attest.
Basic justice demands that subjects in trials should get eventual access to the medicines whose approval they have helped to secure. Too often, new drugs developed via experiments on people in poor countries are not licensed for use in those countries, are priced prohibitively, or are unusable there because the drug isn’t relevant from a clinical perspective.
Such moves may close down some trials. But as the bioethicist Jonathan Moreno said, that’s part of the price we pay for recognising that there is a difference between a laboratory rat and a human being.
Original text in English
Sonia Shah is a Canadian journalist and the author of The Body Hunters: How the Drug Industry Tests its Products on the World’s Poorest Patients (New Press, New York, 2006)
(1) See Jean-Philippe Chippaux, “Pharmaceutical colonialism in Africa”, Le Monde diplomatique, English edition, August 2005.
(2) See for example Stan Bernard, “The Drug Drought: Primary causes, promising solutions”, Pharmaceutical Executive, November 2002.
(3) For a thorough study of the role of “nutritional transition” in the development of some diseases, see Benjamin Caballero and Barry M Popkin eds, The Nutrition Transition: Diet and Disease in the Developing World (Academic Press, London, 2002).
(4) Rapid and voluminous enrollment figures feature prominently on CRO’s websites: see, for example, www.quintiles.com
(5) See “Ethical and Policy Issues in International Research: Clinical Trials in Developing Countries”, National Bioethics Advisory Commission, April 2001.
(6) The Economic Times, Mumbai, 10 March 2004.
(7) During the US Public Health Service’s Tuskegee syphilis study, dozens of poor black men from rural Alabama were denied treatment for decades in order to observe the natural course of the disease. The resulting scandal led in 1974 to the first legal protections for research subjects in the US.
Neither informed nor consenting
By Sonia Shah
Since fewer than 1% of Indians have health insurance, most people pay upfront for care and a personal relationship with a physician is highly prized. The balance of power between doctor and patient is therefore different from the way it is in the West. Farhad Kapadia, a clinical investigator at Hinduja Hospital in Mumbai, says that only those who “have absolutely no choice” agree to forsake the judgment of their physician for a protocol randomly selected by a computer which would leave them ignorant about what is being administered to them. So those who take part in clinical trials are the poorest, living in a different world from socially powerful doctors.
US Food and Drug Administration officials claim to have confidence in the ability of trial subjects to protect themselves by giving or withholding consent. Dr Robert Temple, the FDA’s medical director, says questioning the ability of poor patients to exercise informed consent is patronising: “You can’t treat them as incapable of pursuing their own ends.” Illiteracy and poverty are not insurmountable barriers: scientific concepts in clinical research might be foreign but they are not incomprehensible.
Medical anthropologists have devised a simple way of verifying that consent in a trial is informed. Questionnaires handed to 33 Thai participants in an HIV vaccine trial revealed 30 of them were misinformed. All the subjects in a contraceptive study in Brazil were found to be misinformed. In an HIV transmission trial in Haiti, 80% were found to have misunderstood the basic outlines of the trial. “Informed consent is a joke,” an investigator told the National Bioethics Advisory Commission. Another said: “It is not possible to claim that a person who has never heard of a bacteria or virus is informed.” Another: “This concept of consent of the individual… doesn’t exist. People do what they are told.”
Since 2001 the FDA has been slowly trying to distance itself from the Declaration of Helsinki, because it deems the protections too stringent. The FDA refused to incorporate new restrictions on placebo trials in 2001, and it continues to oppose several paragraphs in the declaration. The zeitgeist was summed up last year, when the leading scientific advisory body in the US, the Institute of Medicine, recommended lifting bans on trials on prisoners, describing as “myopic” the concerns about informed consent that have prevented experiments on prisoners for decades.