In the summer of 2003 Washington University in St. Louis sued one of its former researchers, prostate cancer specialist William Catalona, for ownership of tissue samples collected from thousands of his patients over the years. To some of those patients, Catalona was the sole intended recipient of the specimens, and the university's actions were inconsistent with the research to which they had consented. This spring a St. Louis judge ruled in favor of the university, raising questions about the adequacy of informed consent practices.

Catalona, developer of the PSA diagnostic test for prostate cancer, had left Washington University in 2003 after conflicts over access to the samples, which the university controlled. He then sent letters to 10,000 of his patients asking that they request their samples be transferred to his new employer, Northwestern University. Six thousand patients did as he suggested. When Washington University refused their requests, some asked that their tissues be withdrawn from the repository. In response, the university anonymized the samples, removing personal information that would identify them as belonging to the donors.

After the university filed suit, eight patients joined Catalona as co-defendants. They had donated their samples specifically to him, they argued, and retained ownership of them based on their right to withdraw from research. Some of the informed consent documents they had signed did mention a right to withdraw, which derives from regulations based on the Federal Policy for the Protection of Human Subjects, known as the Common Rule. Exempted from the rule are samples that are not readily traceable to a specific person. Washington University argued that because it had anonymized the samples of those requesting withdrawal, the specimens were no longer subject to the Common Rule.

In late March, Senior District Judge Stephen N. Limbaugh in Missouri ruled the patients had gifted their samples to the university. The right to withdraw simply means that donors can stop providing samples at any time, he found. Moreover, Limbaugh wrote in his opinion, "the integrity and utility of all biorepositories would be seriously threatened if [research participants] could move their samples from institution to institution any time they wanted." The American Association of Medical Colleges expressed a similar sentiment in a brief filed in support of the university. Imagine a unique library of donated books, any of which could be removed on a whim, says David Korn, the association's senior vice president for research. "The tissue's just damned important and should be minimally encumbered," he contends.

The ruling echoes two previous major cases: Moore v. Regents of the University of California (1990) and Greenberg et al. v. Miami Children's Hospital Research Institute, Inc., et al. (2003). Patients had sued researchers for patenting a cell line or gene isolated from samples without the donors' knowledge. In both cases, courts ruled that patients had no property rights to the tissues or what was derived from them.

To patient advocates, the policy implications of the latest ruling are more important than its legal correctness. "Maintaining public trust in research is critically important. What will affect medical research is if people think they're being taken advantage of," says Ellen Wright Clayton of Vanderbilt University, who testified on behalf of Catalona's patients. "I think we need a larger discussion about control over the use of stored tissue samples for research."

The degree of control that people want over their donated samples seems to be low in most cases, but not all. In one 2005 study, investigators found that 87.1 percent of tissue donors who filled out consent forms at the National Institutes of Health granted unlimited further use of their biological specimens, largely irrespective of the signers' age, race, residence or possibility of benefiting from research. Only 6.7 percent refused any ongoing use of their tissues, suggesting that a simple all-or-nothing consent form may be enough, the authors concluded. But if given the additional choice of being recontacted for approval of future research, a substantial minority, 26.2 percent, selected that option.

There are signs that research institutions perceive a need to maintain the public's trust. The nonprofit Coriell Institute for Medical Research in Camden, N.J., which houses the samples collected by the International HapMap Project, requires investigators to consult with the communities providing specimens if they reside in the U.S. Perhaps researchers are coming to believe that, just as with tissue, lost trust is hard to get back.