After almost a decade of reflection, University of Pennsylvania researcher James M. Wilson says problems with the gene-therapy experiment that killed an Arizona teenager were "absolutely unacceptable" and ultimately "my responsibility."

"I have tried to admit where I did wrong, where I fell short," Wilson, 53, said this week in his first media interview about Jesse Gelsinger's death in September 1999. "I have tremendous regrets about what happened. I feel absolutely awful about what it has done to the family, to this university, to the field."

Gelsinger's death set off years of government investigations, congressional hearings, and efforts to improve human-subject protections, while the much-hyped but unproven field of gene therapy shriveled and retrenched. In recent years, it has just begun to report some success.

Wilson is still a gene-therapy researcher at Penn, but his well-funded laboratory is focused on the basic test-tube and animal studies that he now says should have been done a decade ago. Recently, he has been giving cautionary lectures to colleagues at Penn and elsewhere on the "lessons" he learned from the tragedy that reshaped his career.

In February, he detailed those lessons in a commentary that he had agreed to write as part of the U.S. Department of Justice's 2005 legal settlement with him, his two primary co-researchers, and their institutions. That settlement slapped a $517,000 fine on Penn, along with research restrictions, but it did not require admissions of wrongdoing.

"I think it's a 6,000-word apology," Wilson said of his six-page commentary. It was published in Molecular Genetics and Metabolism because the relatively little-known journal agreed to give him all the space he wanted.

That apology was not unqualified. In carefully parsed, often convoluted sentences, Wilson contended, as he always has, that Gelsinger's fatal immune response was unforeseeable and that previous volunteers had no serious adverse reactions to the therapy. Federal investigators disagreed.

Gelsinger's father, Paul, who became a champion of stronger protection of human subjects, dismissed Wilson's piece.

"I don't see it as an apology at all," he said from his Tucson, Ariz., home. "The only thing he regrets is that Jesse died. But he has never accepted responsibility for that. There's never been an acknowledgment of wrongdoing." (Penn settled with Paul Gelsinger for an undisclosed sum within two months of the death.)

Colleagues, in contrast, commended Wilson.

Paul Lanken, a Penn professor of medicine and bioethics who invited Wilson to lecture to medical students in January, said: "He came across as a sincere person who has done a lot of reflection. He acknowledged his shortcomings."

Penn bioethicist Art Caplan said Wilson's self-critique was unusual, if not unheard of, in the staid forum of scientific literature.

"I've never seen anything like this - this sort of putting on a hair shirt and saying, 'Here's what I learned' and 'Don't go down a path that leads to problems I encountered.' "

Wilson's latest hard-earned lesson appears today in the influential journal Science. He urges stem-cell scientists not to rush prematurely into human testing - as he and other gene-therapy researchers did.

"The death of an 18-year-old, Jesse Gelsinger, in a gene-therapy clinical trial that I led . . . initiated a chain of events that seriously derailed the field," Wilson wrote.

Jesse Gelsinger was exuberant and fun-loving, despite a genetic defect that caused his rare liver-enzyme disorder. His own illness was controlled with diet and drugs, but his heart went out to parents of babies with a deadly form of the disease. His desire to help find a cure, his father said, prompted Jesse to volunteer for Wilson's trial.

At Penn, he was given trillions of transporter molecules - inactivated cold viruses - that carried a normal gene into his liver in hopes of replacing the defective gene.

But the transporters, called a "vector," triggered an immune reaction so devastating that Gelsinger's organs failed.

Wilson had much less contact with volunteers than the other primary researchers, Steven Raper of Penn and Mark Batshaw of Children's National Medical Center in Washington. However, Wilson was the high-profile head of Penn's gene-therapy institute, and he had applied to the Food and Drug Administration for permission to test the experimental treatment on humans.

When Gelsinger became the world's first gene-therapy victim, Wilson was at the eye of the regulatory and media storm.

His apparent lack of sympathy did not help. After the Justice Department settlement, for example, Wilson issued a short statement through Penn that said he would continue to devote himself to his lab research; he didn't mention Gelsinger.

This week Wilson said: "There isn't a day goes by that I don't think about it. It was humbling for me in many ways, including sort of a realistic reassessment of what we know and don't know - what we really know. It led to a dramatic change in the way I approach this field and focus our work."

Wilson's 90-person lab, which has $11 million in grants from federal and private sources, is focused on finding better, safer "vectors" to ferry therapeutic genes into cells. A handful of vectors existed in 1999. Since then, his lab has discovered more than 100 that it ships worldwide to researchers studying genetic diseases of many organs.

Human trials using some of the new vectors are now being planned. Wilson, who is banned from conducting human research until next year, said he would probably stay away from it.

Why did he think he was ready to expose humans to the little-understood risks a decade ago?

"I was naive," he said. "The technologies we had available to us at the time were inadequate."

In his commentary, Wilson pointed to miscommunication, overwork and lack of funding as factors - but not excuses - for lapses that occurred. He acknowledged that federal investigators were right about many of those lapses: toxic effects, monkey deaths, and Wilson's financial ties to a gene-therapy company were not disclosed to volunteers. Federal reporting rules were not strictly followed.

And, perhaps the most serious failing: Members of Wilson's team violated their own study protocol when they gave the treatment to Gelsinger. His blood ammonia level - a sign of liver malfunction - rose above the predetermined threshold before he got the treatment. But since it fell back into the safe zone, the team decided to proceed.

"It wasn't OK," said Wilson, who noted that he was not consulted at the time. "Something as critical as [eligibility criteria], there shouldn't be any ambiguity."

Wilson takes heart that Penn and other institutions have made "substantial reforms . . . in terms of oversight of human subject research."

Caplan, among others, does not.

"I don't think a lot has changed," Caplan said. "We still haven't figured out how to deal with informed consent. We hope conflicts of interest are well-managed, but revelations that come out almost weekly tell me it isn't. And deaths continue to occur."

On a personal note, Wilson, who is married with four children aged 14 to 23, said the fallout from the trial taught him to "back off."

"I've learned that it's more fun and more productive not to drive so hard," he said. "Give people space. They're happier. They have lives. You need to step away from all this. I do that, too."

Despite all Wilson has learned, his answer to a crucial question is equivocal: Would Jesse Gelsinger have lived if not for the problems in the trial?

"I don't know," he said.

Contact staff writer Marie McCullough at 215-854-2720 or