A new bombshell has been dropped in the scandal-plagued world of Alzheimer’s disease drug development. According to a New York Times report this week, Eisai and Eli Lilly, in the clinical trials for their respective Alzheimer’s drugs Leqembi and Kisunla, genetically tested participants to identify those with a higher risk of developing the disease, but didn’t disclose the results to patients—even though that genetic predisposition was also linked to a higher risk of brain injuries as a side effect of both drugs.
Between the two companies, more than 560 people joined the trials without being told they had a heightened risk of dangerous side effects such as brain bleeding or swelling, the Times reported, including 274 in the Eisai trial and 289 in Lilly’s.
Both companies had participants sign consent forms that included secrecy provisions around the results of the genetic risk testing, according to the Times.
Two high-risk participants in the Eisai trial died, and more than 100 others experienced brain injuries, most of them “mild and asymptomatic,” but some “serious and life-threatening,” per the report. In the Lilly study, researchers reportedly identified “dozens” of participants who experienced brain bleeding classified as “severe.”
The risk factor in question is a variant of the apolipoprotein e gene known as APOE4. Having one or two copies of APOE4 increases the risk of developing Alzheimer’s disease, as well as the risk of brain bleeding, a potential side effect of both Leqembi and Kisunla.
Not disclosing to patients that they were genetically predisposed to a higher risk of side effects is a questionable decision, Matthew Schrag, M.D., Ph.D., told Fierce Biotech in an interview. Schrag is a neuroscientist at Vanderbilt University Medical Center who assisted with the autopsy of one of the patients who died in Eisai’s trial and has contributed to investigations of misconduct in neurodegenerative disease research. His investigative work is separate from his job at Vanderbilt.
“It's important that when we accept patients’ trust in a clinical trial that we provide them with as much relevant information as is reasonably possible,” Schrag said. “I would expect these sponsors to err on the side of giving patients more information, especially when the risks of very serious side effects are as nontrivial as they are in these studies.”
In emailed statements to Fierce Biotech, both Eisai and Lilly said that they did inform patients that carrying copies of the APOE4 gene variant comes with a higher risk of side effects. However, neither told patients whether or not they actually had APOE4.
“The Clarity AD Informed Consent forms disclosed the known risks, including side effects and the increased risk of vasogenic edema in subjects with APOE4, and were approved by Institutional Review Boards and Ethics Committees in regions and countries where the study was conducted,” Eisai said in its statement.
In its own response, Lilly said it will disclose APOE4 status to investigators at the conclusion of its Trailblazer-Alz 2 study, who can then disclose it to participants who wish to know.
“Our advice is for participants to assume they have the higher risk,” Lilly said.
The scientific world's understanding of how APOE4 influences the risk of side effects from these drugs has increased over the last few years, Schrag said.
“The study sponsors are correct that there's probably some complexity to this issue,” he added. “Things that look crystal clear now might not have looked that way when the trial started and in the planning period.”
Still, he said, the risks posed by APOE4 have been well-known for some time.
Lilly chose not to disclose APOE4 status due to the potential bias that knowledge could introduce in participants, according to the company's statement. A 2014 study found that healthy adults who knew they had copies of APOE4 performed worse on memory tests than patients who had the variant but didn’t know about it—a bias that could have potentially influenced the trial’s findings.
“I think that's a silly argument, because the clinical trial is blinded,” Schrag said.
Patients didn’t know whether they were receiving the drug or a placebo, and since the trials were randomized, he added, individuals with APOE4 should have been evenly distributed across the trial’s arms. That would’ve meant that any bias stemming from patients knowing their APOE4 status could be controlled for in the analysis of the trial’s results, assuming the bias has a consistent effect across patients.
Following the Times’ story, nonprofit consumer advocacy group Public Citizen called on the FDA to investigate the conduct of the two trials and the actions of the institutional review boards that had oversight over them, saying that the principle of informed consent had been undermined.
“If subjects had known that they were at higher risk of brain injuries, they might have decided not to participate in the trials,” Robert Steinbrook, M.D., the director of Public Citizen’s Health Research Group, said in a statement. “This is information potential participants would want to know.”
Editor's note: This story was updated to include additional comments from Matthew Schrag about the issue of bias in patients who know their APOE4 status.