There's a lesson about one of the testosterone studies released this week that has nothing to do with testosterone: The study on how testosterone affects anemia was designed with an ethical lapse that nobody noticed until the study was complete.
That's surprising because it was designed and carried out by a couple of dozen of well-regarded scientists. Their protocols were reviewed by 12 university institutional review boards, whose job is to evaluate the ethics of an experiment. It was funded by the National Institutes of Health, and the trial was overseen by a watchdog data safety and monitoring board.
But all of those safety features fell short this time.
A reviewer at JAMA, the journal of the American Medical Association, noticed that participants who were diagnosed with anemia (low iron in the blood) at the outset of the study were not told of that fact.
"Abnormal results on this simple blood test could have been an early warning sign of a serious illness which, if diagnosed and treated early could have greatly improved patient outcomes," says Dr. Bernard Lo, a bioethicist who runs the Greenwall Foundation.
Lo says low iron levels could be an early sign of colorectal cancer, which is quite treatable if it's caught early. Lo noted that 126 of the 788 study participants had low iron in the blood, but nobody thought to inform them.
Scientists do not know whether this lack of information actually harmed any of the participants. "But my own feeling is if it happened once, despite all these safeguards, it could happen again" in another major study, Lo says. "It's a warning sign that there's a major problem in the system that needs to be addressed."
Lo and Dr. Deborah Grady at the University of California-San Francisco wrote an editorial about the ethical lapse in JAMA Internal Medicine that accompanied articles about anemia and other studies that showed some benefits and some downsides to prescribing testosterone supplements to older men with low levels.
Testosterone supplements have been controversial for some time, and the Food and Drug Administration increased its warnings about the potential cardiac side effects of taking the hormone in 2015.
Dr. Peter Snyder, a professor of medicine at the University of Pennsylvania who organized the anemia study, readily admits this was an oversight. He says the scientists and review boards had focused their attention on potential harms from the testosterone treatment.
"We put an enormous amount of time and effort into attempting to protect the participants [from the side effects of testosterone], and that's where our focus was," he says.
Testosterone can affect the prostate and the heart, so the study was designed to watch for those health risks.
Men with severe anemia were excluded from the trial to begin with, Snyder says. "Men who had a mild degree of anemia weren't informed because we just didn't think of that," he tells Shots. "It didn't occur to anybody until this review pointed it out."
Snyder says scientists followed up by sending a letter to all 788 participants, informing of them of their iron levels at the start of the trial as well as at the end.
He says since the trial has now ended, there is no plan to systematically study the men who had mild anemia. "At this point we have not been notified of any problem."
Ethical lapses like this are not often recognized or highlighted. Lo says this occasion provides an opportunity to trouble-shoot a system that should not fail like this.
Lo says the ethical lapse was not enough to prevent publication of the study.
ARI SHAPIRO, HOST:
We have an important postscript to that study you just heard about. A journal reviewer discovered a design flaw that raises an ethical issue. NPR's Richard Harris has more.
RICHARD HARRIS, BYLINE: One ethical rule for clinical studies is if you find a health issue among your participants, you should inform them so they can get it checked out. That didn't happen in this case. More than a hundred men in the study had suffered from mild anemia at the time they enrolled, and nobody thought to tell them.
BERNARD LO: Abnormal results on this simple blood test could have been an early warning sign of a serious illness.
HARRIS: Bioethicist Bernard Lo at The Greenwall Foundation says the anemia could indicate colorectal cancer, which is most treatable when caught early.
LO: Why wasn't this picked up until the astute journal noticed it after the study was done?
HARRIS: After all, the study had a dozen different institutional review boards that were supposed to look for ethical issues. Two dozen well-regarded scientists ran the experiment with funding from the National Institutes of Health. And to top it off, the study was overseen by a data and safety monitoring board, Dr. Lo notes in an editorial in JAMA Internal Medicine
LO: And the system didn't work. And so we're saying, there needs to be in-depth analysis of what happened here and how to prevent it from happening again.
PETER SNYDER: It didn't occur to anybody until this reviewer pointed it out.
HARRIS: Dr. Peter Snyder at the University of Pennsylvania was in charge of the study in question. Speaking via Skype, he said he and his colleagues had focused their concern on the potential risks of testosterone itself, which can promote prostate cancer and possibly heart disease.
SNYDER: And that's where our focus was.
HARRIS: As soon as the lapse came to his attention, he recognized it as a problem. The study officials sent a letter to all the participants, giving them their blood test results before and after the experiment. So far, he says he has not heard of anyone who was harmed. Richard Harris, NPR News. Transcript provided by NPR, Copyright NPR.