Testing, Testing, Testing
issue 165 - November 1986
Testing, testing, testing
Roughly a quarter of a million Americans, both volunteers and paid,
are experimented on with new drugs every year. Millions of rabbits,
monkeys and beagle dogs have the same experience; all this in the
interest of testing for more, presumably better, drugs.
Is it worthwhile? Francesca Lyman reports.
AIDS was spreading like the plague last year as laboratory researchers feverishly tested a half-dozen experimental drugs to treat victims of the virus. One treatment for a deadly AIDS-related pneumonia did reach the market in 1985. Yet the vast majority of new drugs that were developed and tested by pharmaceutical companies that year, as in most years, will not be ground-breaking cures and medical advances but only add to an overstocked medicine chest of drugs. Too many drugs turn out to be just clones of old ones.
The facts speak for themselves. Out of more than 100 drugs approved each year by the US Food and Drug Administration (FDA), only a fraction represent major therapeutic advances'. For example, in 1984, there were 142 new drug applications approved, of which 22 were 'new chemical entities' - that is, used new chemical molecules and were not variations on existing drugs. Out of those 22 new chemicals (mostly antibiotics, antidepressants and agents for heart disease) only two were judged to be 'major advances' by the FDA and eight 'modest advances'. Most of the other 12 were the so called 'me too' drugs by which a company makes its own version of an already marketed drug (see Key Terms, page 6).
Each year the large, mostly multinational, drug firms come up with new molecular compounds which they must subject to clinical trial testing that usually takes several years. The companies do their own testing, although legislative reforms since the 1960s have strengthened FDA's power to oversee the process. But the price of arduous safety testing is high - it costs industry millions of dollars and the government further millions to regulate - and can mean considerable risks for its subjects, both animal and human. Of course there should be extensive testing for safety. But are most of these drugs, the 32 out of the 142, worth the effort? And how good a job does industry do on testing? And do most drugs serve useful needs?
Thanks to FDA's drug laws, one can no longer buy Dr Wizard's Elixir at the neighbourhood drug store. The agency's drug approval system, despite what critics say about current implementation of the law, is generally recognized as having set standards of medical rigor that are emulated worldwide. Companies normally try out a chemical on laboratory animals first to see whether or not it performs effectively - and later they move on to see about safe testing on humans.
Tests often show up side-effects - sometimes strange ones undreamed of by researchers. 'One antidepressant while being tested was found to cause colour-blindness,' recalls pharmacologist Lione, 'and an interferon drug made someone's eyelashes grow about six inches long.'
According FDA's estimates, by the time a company has submitted a new drug, claiming safety and efficacy, it 'has been studied in several hundred to several thousand patients'. FDA's job then is to decide if the benefits of the drugs outweigh the risks - since almost no new drug comes without side-effects.
That, in theory, is the way drugs are supposed to be tested and reviewed before they reach the market.
Unfortunately tests have failed to pick up serious and often fatal side-effects of drugs. By anyone's estimate, clinical trials of drugs are an inexact science. Only widespread use reveals all a drug's features. Testing on humans doesn't cover a wide pool of people - at most only a few thousand - to detect reliably many adverse reactions. Grim as it may sound, scientists today, according to FDA's Dr Temple from the Office of Drug Research and Review, learn more about side-effects in the first few years of the drug's marketing than from the testing. Someone even facetiously said that we should put a skull and crossbones on new drugs for the first year of marketing,' says Temple.
The government routinely inspects industry testing facilities, but it seems to depend for the most part on honour. 'Most of the trials are done by respected members of the medical community,' says Temple. 'Over the years,' he adds, 'the instances of fraudulent behaviour are very few.
That depends on whom you ask Safety testing, John Braithwaite recounts in his book Corporate Crime in the Pharmaceutical lndustry has been pervaded by negligence and fraud. In a chapter on testing, he writes, 'in the three years 1977 to 1980 the FDA claims to have discovered at least 62 doctors who had manipulated or downright falsified clinical data. Dr Ronald Smith, a psychiatrist, was hired by six pharmaceutical companies between 1971 and 1978, including Sandoz, Upjohn and Cyanamid, to test at least a dozen psychotropic drugs. An FDA scientist says, 'We learned from an office assistant. that the way the doctor got the pill count to come out correct was to count the correct number of pills the patient should have taken and then flush them down the toilet" According to Braithwaite, only 3 or 4 of the 60 patients that were listed had been tested.
Furthermore, frequent examples of companies fiddling with safety data are described by Braithwaite - these involve everything from case reports on fictitious subjects to false claims about a drug's effectiveness.
Have pharmaceutical companies cleaned up their act? The case of Merital may be the latest warning sign that the industry needs closer watching. Hoechst-Roussel Pharmaceuticals, one of the biggest West German pharma companies, first came to the FDA with its application for Merital (an antidepressant prescription drug) in the early 1970s. But FDA, after six years of agonizing over whether or not studies proved the drug's effectiveness, concluded in December 1979 that it couldn't be approved because the company hadn't produced enough tests. Out of some dozen studies, only three showed positive results.
Five years later, however, the agency approved the application after the company re-analyzed the same data and studies. A year after the drug was approved in the US people in Europe started dying of a seemingly rare disease called hemolytic anaemia.
A hearing held by the subcommittee on lntergovernmental Operations (of the House Government Operations Committee) uncovered that Hoechst withheld data on more than 50 cases of serious reactions to the drug in Europe, including some deaths. By the end of the year at least five people had died of hemolytic anaemia, a disease that causes the body's immune system to attack and destroy red blood cells. In January 1986, Hoechst withdrew the drug from the worldwide market under pressure from Britain, because more and more cases of hemolytic anemia began turning up. Now under investigation by FDA, the Hoechst case raises many questions about how a drug can pass muster - from the adequacy of industry testing to the faith FDA places in company evidence.
Should governments approve drugs that are not better and probably worse than what's already on the market? FDA's Dr Robert Temple doesn't worry that so many new drugs aren't star quality. 'Breakthroughs don't come very often,' he says. 'Usually a company hopes they will be coming up with a real advance but it doesn't turn out to be.'
A pharmaceutical industry spokesman offers a franker answer. 'What makes a company come up with a new drug?' asks Dr. Paul Kaufman of the American Pharmaceutical Manufacturers Association. 'It's felt to be needed. Why do they put out 16 valiums? Probably the real reason is they've gotta make a living.'
The result is an array of drugs with similar properties differing in their side-effects. But there's always the possibility that a drug's usefulness may grow. Ultimately faced with the option of forcing drug companies to return to the drawing boards if they've nothing new to offer, most medical professionals object. 'It's very difficult to do this,' says Frederick Wolff, a clinical pharmacologist. 'It's a free society. You do need competition in order to get things developed.' To conclude: we have to have testing for the drugs we need. The colossal waste is in testing on apparently pointless new compounds. That's the problem.
Francesca Lyman is editor of the Friends of the Earth (US) magazine, Not Man Apart.