fter investigating the death of a young, healthy volunteer last month in a research study at Johns Hopkins University, a federal oversight agency yesterday suspended almost all the university's federally financed medical research involving human subjects.
One reason for its action, the agency, the Office for Human Research Protections, said in a letter to Hopkins, was that the ethics committee that approved the study in question had failed to take proper precautions to protect its subjects. Suspending research is an unusual step.

Johns Hopkins receives more federal research money than any other university, $310 million last year. But a spokeswoman for the university said she did not know how many studies or how many subjects would be affected.

To resume the studies, the agency said, the university will have to address concerns raised in the letter, restructure its system for protecting human subjects and develop a plan to make sure its ethics committee and researchers are educated in the regulatory requirements for protecting human subjects. Hopkins said the agency agreed today to expedite its review of the university's response. It is not clear how long the process will take.

The volunteer, Ellen Roche, died on June 2, a month after she inhaled an unapproved drug as part of a research study to examine the causes of asthma. Her lungs were destroyed, apparently by the chemical she inhaled, hexamethonium. She was 24 and a technician at Johns Hopkins.
Hopkins officials reacted with outrage to the suspension of research, issuing a statement calling the action "unwarranted, unnecessary, paralyzing and precipitous" and "an extreme example of regulatory excess." They pointed out that the suspension could have "potentially devastating" effects, stopping studies that included cancer patients receiving experimental drugs.

The university said that tens of thousands of people had volunteered for its medical studies and that Ms. Roche was the only one who had died. And, the medical center said, it had recognized its deficiencies and had taken action to correct them.
The continuing examination of what went wrong at Hopkins has raised broader questions about how tightly the government should regulate academic research. The Food and Drug Administration says scientists should seek its approval for any study involving new or unusual uses of drugs, but the agency does not police the universities to make sure they comply. Some researchers have sought F.D.A. approval, but others, like the researcher at Hopkins, have not. Hopkins says the F.D.A. has not made its policy clear.

The F.D.A. said the researcher, Dr. Alkis Togias, should have sought approval from the F.D.A. before he began his study. That could have required voluminous work to assess the drug's safety and the study's design, but some medical experts say such a review could have prevented Ms. Roche's death.

"What we are talking about is a system of controls that works to protect human subjects," said Dr. David Lepay, who is a senior adviser for clinical science in the office of the F.D.A. commissioner. "Our regulations are fairly specific. It is not a question of how many subjects are in the study or whether the study is being used for purposes of marketing. What we are talking about is the study of an unapproved drug in a clinical investigation."
But the agency's policies have been unclear, some medical experts say, and academic scientists doing basic research have not felt bound by the same rules that govern researchers at drug companies.
At Covance Laboratories in Madison, Wis., a company that does drug research under contracts for universities and industry, the opinion was that basic research not involving drug development did not require the drug agency's approval.

"To do an experiment without the F.D.A.'s review is O.K. as long as it is based on strong scientific evidence and it has peer review," said Anthony Clemento, the company's vice president for regulatory affairs.
Drug companies had the impression that academics played by different rules.

"The F.D.A. has a double standard," said Dr. Bert Spilker, who is vice president for scientific and regulatory affairs at the Pharmaceutical Research and Manufacturers of America. "The standards for industry are much higher than those followed by academic investigators."
Applying to the F.D.A. could cost millions of dollars, Mr. Clemento said. The paperwork typically takes months and costs $50,000 to $60,000, he said. The application requires an extensive review of published papers that could uncover drug toxicities, and the agency could require exhaustive testing in animals before the drug is given to humans.

William Vodra, a former F.D.A. lawyer now working for the Washington law firm Arnold & Porter, said drug companies are used to getting agency approval and have entire staffs whose job it is to make sure the job gets done. Universities have no such resources, Mr. Vodra said, which could make it very hard to comply with F.D.A. regulations.

"If the F.D.A. holds their feet to the fire, I think it's going to be a major disruption to academic research," he said. "The universities don't have the infrastructure."
Hopkins officials said there was confusion about when to seek the agency's approval. In 1997, the head of the ethics board at the medical center, which approves research using human subjects, sent a letter describing a hypothetical situation involving the application or inhalation of naturally occurring substances, like the hormone bradykinin. Was F.D.A. approval necessary for such studies, he asked.
Two years later, the F.D.A. said that the question was complicated and that the agency was working on a reply. The university tried again to get an answer on April 19 of this year, but got no reply.
Dr. Lepay said the F.D.A. thought Hopkins was looking for information on studies that fall into a gray area, not ones involving unapproved drugs like hexamethonium.

In the meantime, Dr. Togias went ahead with his study, without applying to the F.D.A. After assessing hexamethonium's safety by searching the medical literature, he concluded that its main side effect was a possible transient lowering of blood pressure. The ethics committee approved his study, which was to involve 10 healthy volunteers.
He then bought hexamethonium, sterilized it and began giving it to his study subjects, in amounts as high as 1 gram each. Some doctors said that was an extraordinarily large amount to inhale.
The first person developed a cough and shortness of breath but got better in a week. Dr. Togias did not report those symptoms, deciding that the volunteer had caught a cold or that her symptoms were caused by the acidity of the hexamethonium solution. He altered the chemical to make it less acidic and gave it to a second volunteer, who had no adverse effects.
On May 4, he gave the chemical to Ms. Roche.
She began feeling ill the day after she took the drug. A few days later she was admitted to the hospital, the air sacs of her lungs collapsing. Her lungs failed and she died on June 2, after weeks in an intensive care unit, tethered to a respirator. A Hopkins committee investigating the death reported last week that the hexamethonium was probably what killed her. And it found reports in the medical literature that Dr. Togias had apparently missed, indicating that the chemical might injure the lungs.
If Dr. Togias had gone to the F.D.A. first, his study probably would not have been approved, some medical experts said.
Asked if he would approve a study that, like Dr. Togias's, gave volunteers a gram of hexamethonium, Dr. Lepay said, "I have a personal problems with that concept, but I don't want to go out and say what makes a decision on the part of our review division."
Could the F.D.A. even handle an onslaught of applications from academic researchers? "This is clearly something we are going to have to look into," Dr. Lepay said.
But, he added, the issue was discussed in the agency recently and "the unanimous opinion was that they universally encouraged" seeking agency approval.

Some who have worked at the F.D.A. said that was a change of attitude. In the past, they said, the agency has not been eager to take on the extra work of reviewing applications to do basic research.
Jerome Halperin, the president and chief executive of the Food and Drug Law Institute, a nonprofit organization in Washington, said that when he was at the F.D.A., in the late 1970's and early 1980's, officials decided that it simply was not worth their time to vigorously regulate basic research at universities. "We tried to clear out of the system things that did not provide useful information for drug development and did not raise the level of safety concern," Mr. Halperin said. "Most of the academic experiments were of much lower orders of risk."

And so, he said, when academic researchers inquired about getting F.D.A. approval for their studies with human subjects, "we almost discouraged them" from applying.
But the time, expense and trouble it would take for academics to submit their research proposals and for the F.D.A. to review them are no excuse to let them slip by, said Dr. Spilker of the pharmaceutical manufacturers group.

"You have to consider benefit-risk issues," he said. "The risk is far too great. That's why we have standards."
If someone had asked him about agreeing to inhale a gram of hexamethonium in a study that had not been approved by the F.D.A., Dr. Spilker said, he would have told them to refuse. "I would never want anyone to go into a study like that, never."