On October 21, Drug Administration States announced new rules for accelerating access to treat life-threatening and severely debilitating conditions. Although the new rules, and other measures that have been announced since last year, are seen mainly as a response to the AIDS epidemic, the changes apply to all serious conditions. I am involved in shepherding a new type of anti-cancer treatment through regulatory agencies and I welcome FDA's initiatives. They are a significant step forward and will undoubtedly help prompt reexamination of regulatory approaches worldwide. However, I think it important to reflect on some of the ethical issues that bear on access to experimental drugs, and in particular the possibility that early access by current sufferers risks harm to future sufferers. Understanding drug development and regulation, as well as the general ethical principles involved, will help ensure that the effects of earlier access are beneficial.

The "catastrophic case" may seem quite straightforward from the point of view of ethics. On what ethical grounds could one refuse an experimental drug to a person who will almost certainly die if he does not receive it? One ethical principle on which to base a refusal is paternal ism. Here paternalism is roughly the view that the patient's conception of his or her own good should be overridden because the experimental drug has not yet been judged safe and effective according to the current canons of drug regulation. Paternalism has not enjoyed great favor in recent years. It certainly will not win the day here. Because the patient is facing near maximum risk and experimental drugs hold out the possibility of some benefit, denial of the experimental drug can hardly be considered in his best interest.

Does this conclusion mean that there are no ethical grounds for denying experimental drugs? No, I think at least one ethical principle does weigh against giving people even in life-threatening situations open access to experimental drugs. Note that I said "weigh:" I do not believe this principle-or any combination of other ethical principlescan provide a full justification for denying experimental drugs to people in such situations. Other considerations are relevant too.

The ethical principle I have in mind may be stated in this way: The desires of an individual can be overridden by a legitimately constituted authority when failure to do so is likely to undermine the well-being of society as a whole or large numbers of fellow citizens.

Perhaps the most obvious way that this principle bears on the case at hand is that, arguably, early access will make it difficult to recruit patients for clinical trials. Thus, the safety and efficacy of new drugs may never be scientifically demonstrated.

This point is less poignant for AIDS than for most other catastrophic conditions because AIDS support groups are sufficiently organized to ensure enough volunteers for clinical trials even if the drug in question is also available under less strict conditions.

Appreciation of the relevance, though not necessarily the persuasiveness, of the principle requires understanding drug development and regulation. What happens after a university researcher holds a press conference to announce to the world that she has discovered a chemical that significantly enhances the immune system or a drug that attacks cancer cells without harming normal tissue? Though such announcements are usually accompanied by cautions that a great deal more testing will be required to ensure that the material is actually effective and safe, probably few know what that disclaimer means in detail.

Almost certainly the drug will not have been tested in humans at the time of the announcement. Before it is administered to humans it will have to undergo a number of tests in animals or cell cultures to establish doses at which toxicity can be expected. These toxicological tests cost a great deal of money-from several hundreds of thousands of dollars to several million-so someone must be found to underwrite the cost.

That "someone" will almost certainly be a major pharmaceutical company. It will have to address a range of business questions, for example: Can it get a patent on the material or some other proprietary protection? After spending a minimum of $50 million and 5 years or so in getting the drug through the regulatory process, will the market for the product be large enough to justify the expense?

In addition to these business questions, the company will have to answer innumerable scientific and technological questions such as: Can the manufacturing process be replicated? Is the material stable? What is the best way to formulate the material so that it is effective in humans-as a powder or a liquid, taken by mouth or intravenously?

There are many steps leading from "discovery" at the university through toxicological testing to human trials. Then there are three phases of human trials: Phase I is usually done on at most a few dozen subjects and is designed to test the safety of the product. Phase II is done on roughly 100 subjects and tests the dose sizes and, somewhat, the efficacy. Phase III usually involves several hundred subjects and compares the safety and efficacy of the new drug with those currently in use.

What we see is a long (and costly) process beginning in basic science, continuing into a development phase, and ending in clinical trials. As this process unfolds more and more scientific information is gathered about the drug and its positive and negative effects. Clearly, the earlier access is permitted, the less is known about the drug; and the less known, the greater the possibility of undesirable consequences. The obvious danger is that faced by individuals, namely that very sick people will be harmed seriously or taken advantage of by unscrupulous quacks.

Less obvious and perhaps more dire from an ethical point of view is that releasing drugs early could cause harm to collectives or society as a whole by undermining the scientific assessment of drugs in general. After all, at some point drug companies will say to themselves, "Why are we spending billions of dollars a year on medical research and clinical trials when someone can come along and claim that seagull droppings cure cancer and get away with marketing it?" This is where my principle comes in:

The drug development system, though certainly not fault-free, has done a creditable job of providing a reasonably steady stream of drugs that we have scientific grounds for believing are safe and effective. Other things being equal, it is of course right to make drugs immediately available under catastrophic conditions. But things are not necessarily equal. Giving current sufferers early access to drugs may jeopardize a very much larger number of future sufferers.

I believe that this consequence must be considered, but I do not believe it is by itself conclusive. The new initiatives in the U.S. and some other jurisdictions are on the right track. But implementation of the new rules requires radical changes on the part of all those involved:

• Physicians administering experimental drugs outside of strict clinical trials will have to collect information.
  
• Regulatory bodies will have to ensure that that information can be used as data.
  
• Drug companies will have to be convinced to cooperate in making the drug available.
  
• Patients will have to ensure enough volunteers for full clinical trials and will have to forego some liability claims in return for earlier access.
  
• The public will have to understand the complexities of the issues. If these conditions are met, experimental drugs will be made more readily available to persons in catastrophic situations without harm to future sufferers.