Alzheimers drug horror story from the NYT

Story continues

The secrecy provisions, which have not been previously reported, came to light as The Times investigated how the long, maddening and vastly expensive search for an effective Alzheimer’s treatment led to the testing and approval of Leqembi (and subsequently Kisunla). Reporters reviewed clinical trials, patient records and injury reports and interviewed researchers, neurologists, trial participants, families of Alzheimer’s patients, drug industry representatives and F.D.A. officials.

Leqembi and Kisunla seek to remove a misshapen protein called beta amyloid that forms plaque in the brains of patients with Alzheimer’s. To a large extent the drugs have succeeded, a remarkable scientific achievement.

Yet the drugs do not halt cognitive decline or reverse brain damage. Leqembi slows the decline for roughly five months, while Kisunla achieves a slightly longer delay. The evidence of their limited benefit has contributed to a growing realization that the dominant theory of Alzheimer’s — that sticky bands of amyloid trigger a cascade of toxic events leading to the disease — is at best incomplete and perhaps simply wrong.

At the same time, many Alzheimer’s experts worry that the new drugs’ risks have been neither fully appreciated nor understood, especially when set against their modest benefit.

“The people who are in charge of the clinical trial have not come to grips with the severity of the toxicity” of Leqembi, said Dr. Rudolph J. Castellani, a pathology professor at Northwestern’s Feinberg School of Medicine in Chicago. Dr. Castellani performed an autopsy on Jean Terrien, the other high-risk volunteer who died during the Leqembi trial.

In July, the European Union’s drug regulator recommended against approving Leqembi, co-marketed by Biogen. Last week, Australia’s regulator also declined to approve the drug. Both agencies said the drug’s temporary delay of cognitive decline did not outweigh the safety risks. In the United States, a similar conclusion was reached by the Institute for Clinical and Economic Review, a widely used independent group of analysts.

In addition, a new analysis by nine leading researchers suggests that patients taking Leqembi and an earlier anti-amyloid drug, Aduhelm, had a higher mortality rate than a similarly aged population of untreated Alzheimer’s patients in the United States.

Concern about brain injuries has run through years of amyloid drug trials. In 2010, the F.D.A. recommended tightening protocols to protect the most vulnerable subjects. But researchers pushed back, and when they instead argued for broadening eligibility, the federal regulator went along.

For the Leqembi study, the secrecy provision was approved by an institutional review board run by a private-equity-backed company, Advarra. Under federal law, such boards are tasked with ensuring that trial participants do not face unnecessary risks and are informed of the studies’ risks.

While the Leqembi trial was underway, Advarra published an online “tip sheet” calling informed consent “one of the central protections” for research subjects. When asked by The Times about Advarra’s approval of the secrecy provision in the Leqembi trial, a company spokesman said he was unable to provide answers.

Eisai canceled an interview and did not respond to repeated messages over several months seeking an explanation for its decision not to disclose the genetic findings.

Story continues

Genevieve Lane had come to the Villages in 2014, joining her lifelong Chicago friend Vicki Holmes. Together they had traveled the world and navigated marriages, mother-daughter conflicts and the specter of old age.

“I always said, ‘Genna, I’ll retire and find someplace in Florida, and you’ll just come live with me,’” Ms. Holmes recalled. “That was always the plan, probably, for 20 some years.”

Before her retirement, Ms. Lane had been a vice president of a transportation company. Now, she was slipping mentally. But Ms. Holmes gladly rescued her when she got lost, helped plan her daily activities and drove her to the market, hair dresser and nail salon

Then, in 2020, Ms. Lane saw a Charter ad seeking volunteers for a trial of an experimental Alzheimer’s drug. She grew excited. Maybe the trial could help her and others, especially her children, who might someday develop the disease. Maybe this drug could make her less frustrated, more capable of enjoying life.

As Charter’s ad said: “Hope Starts Here.” Charter couldn’t promise a cure, only the hope that came with the drug known as BAN2401.

Managing the Risk

Since the 1990s, Alzheimer’s researchers had focused on proving the “amyloid-cascade hypothesis” and finding a treatment that would attack the culprit protein. Yet in trial after trial, trying to remove amyloid had produced significant side effects.

The most consequential failure involved a drug called bapineuzumab, known as bapi. Wall Street analysts predicted annual sales reaching $13 billion; one medical journal proclaimed that the future of Alzheimer’s research “may hinge on bapineuzumab’s outcome.” Those hopes dimmed in 2008 when bapi was found to cause brain injuries with little or no cognitive improvement.

Bapi’s safety issues were on the minds of F.D.A. officials in 2010 when Alzheimer’s researchers gathered in Honolulu for their annual conference. Before long, a dispatch from the federal regulator sent tremors through the conference hall. To better protect the most vulnerable patients, the F.D.A. recommended that future trials exclude volunteers with a history of microhemorrhages, ruptures of small blood vessels in the brain. At the time, people with one or two prior microhemorrhages were accepted.

Fearing this change would unduly handicap efforts to study new amyloid drugs, an ad hoc group of prominent researchers at the conference — most with close ties to the pharmaceutical industry — planned a counteroffensive. The way forward, they reasoned, lay in managing the risk, not eliminating it.

“We need to be less risk-averse in Alzheimer’s disease,” Dr. Philip Scheltens, a member of the group, would later say. “We should carefully dose up until side effects tell us to hold off.”

In the end, the group actually recommended expanding eligibility for the very people the government had hoped to protect. Now, four microhemorrhages would be allowed. The F.D.A. acquiesced, earning praise for “exemplary” collaboration from the Honolulu meeting’s sponsor, the Alzheimer’s Association, an advocacy group that accepts industry financing and arranges major conferences.

The Honolulu group took another step: It rebranded the name of the brain injuries, in part to make them sound less scary. Instead of vasogenic edema and microhemorrhages, the condition would now be called by the operatic name ARIA, an acronym for amyloid-related imaging abnormality.

But if the new guidelines opened up the trial process, they did far less to bring to fruition the industry’s quest. Over time, this search consumed so much research money that it became “too big to fail,” said Dr. Perry, the journal editor and an Alzheimer’s researcher at the University of Texas at San Antonio.

The drug maker Biogen finally broke through in June 2021, when the F.D.A. granted accelerated approval for Aduhelm, the first drug to treat Alzheimer’s purported root cause.

It proved a Pyrrhic victory. As recounted in an editorial in JAMA, the influential medical journal, Aduhelm’s approval “generated significant backlash due to unclear evidence of the drug’s clinical efficacy,” severe adverse effects and an approval process that one congressional inquiry called “rife with irregularities.” (With shrinking sales, Biogen jettisoned the drug this January.)

Eisai, though, was pinning its hopes on Leqembi.

Story continues

Though typically mild and reversible, side effects such as headaches, convulsions and vision loss were sometimes so severe that patients required prolonged hospitalization and discontinuation of the therapy, according to adverse event reports sent by Eisai to the F.D.A.

Among those patients was a 70-year-old man who experienced progressive headaches, followed by a seizure, and was ultimately found to have had 61 microhemorrhages. Eisai confirmed the events were “related to study drug.” He left the study; his status was later updated to “not recovered.”

After her third dose of Leqembi, an 81-year-old woman experienced such significant brain swelling that she was withdrawn from the study. The symptoms were “related to study medication,” Eisai wrote. Over a year later, an update said she had “not recovered.” (The adverse event reports do not include APOE4 status.)

Much remains unknown about the outcomes of subjects who left the study.

Dr. Madhav Thambisetty, a neurologist and former senior investigator at the National Institute on Aging, faulted Eisai for not releasing patient-level trial data on the long-term impact of ARIA on cognition.

“I think that’s a glaring omission in my mind,” he said. “We don’t know what happened to those patients who developed serious symptoms due to ARIA,” except for two who were studied by French clinical-trial investigators. (Dr. Thambisetty recently took a job in the drug industry.)

Dr. Nicolas Villain, the French neurologist who treated the two patients, said in September that one was nonverbal and bedridden, and the other was incapacitated with severe dementia. Dr. Villain said he believed the drug’s black box warning was too weak. “These terrible events showed us that it wasn’t enough,” he said.

Some neurologists fear that brain injuries will increase now that Leqembi is widely available outside the trial, where doctors are less familiar with the drug and monitoring may not be as rigorous.

Susan Aaron, a 74-year-old retired medical coder from the Bronx, started on Leqembi this past May, according to her longtime companion, Valerie Porter. Shortly after her third infusion, Ms. Aaron, who had two copies of the APOE4 gene, was found on her couch, unconscious and drooling. She never regained consciousness. An M.R.I. showed her brain was swollen and had at least seven new microhemorrhages.

The new retrospective analysis of patients taking Leqembi and Aduhelm said their mortality rate was three to four times higher than for Alzheimer’s patients not taking the drugs. While noting the limitations of their analysis because of incomplete data, the authors, including Dr. Schrag and Dr. Alberto J. Espay at the University of Cincinnati, concluded that their finding “not only deserves to be part of informed-consent discussions” but also warrants an expanded black box warning.

Little is understood about another potential risk of amyloid-lowering drugs — accelerated brain shrinkage.

Scott Ayton, a professor of neuroscience at the University of Melbourne, has studied the phenomenon. “The shocking results that came from our analysis,” he said in an interview, “is that these drugs, in every class that we looked at, did not preserve brain volume — they accelerate the apparent shrinkage.” He, too, criticized the drugmakers for failing to publish patient-level data to better understand that.

Brain atrophy comes naturally with aging, but it occurs faster in Alzheimer’s patients and faster yet in patients on amyloid-lowering drugs, according to neurologists

“All of these anti-amyloid drugs have brain shrinkage associated with them — nobody’s dealt with it,” said Dr. Perry, the journal editor. “There’s no indication shrinking the brain is good.”

The Quest for a Cure Continues

More than a century after Alzheimer’s discovery and decades into drug trials, the sobering reality is that scientists still can’t agree on what causes the disease, much less how to defeat it. They don’t know what role amyloid plays in the development of Alzheimer’s or whether it is Alzheimer’s that causes the development of amyloid. Or why someone can have amyloid but not Alzheimer’s, or Alzheimer’s but no amyloid.

What a growing number of scientists say, though, is that while there’s nothing wrong with continuing to study amyloid-lowering drugs, the time has come to expand the focus.

“While beta amyloid may play a role in Alzheimer’s disease, it’s not the central disease driver, and we need a more nuanced understanding of this disease if we’re going to be successful in really moving the ball,” Dr. Schrag said.

Researchers have begun exploring other avenues, including drugs to reduce inflammation, improve blood flow and protect neurons. They are also studying repurposing drugs already F.D.A.-approved for other diseases.

this section is out of the correct order

Patients Left in the Dark

The same year as the Honolulu conference, researchers from 19 drug, biotech and medical companies came together at a Phoenix airport hotel for a highly unusual meeting. Though fierce competitors, the scientists wanted to collaborate on strategies for Alzheimer’s research — specifically whether anti-amyloid drugs could prevent Alzheimer’s in people who were still cognitively normal, before the onset of decline.

“You really needed people who might progress to cognitive impairment or start to develop symptoms of the disease in a relatively short period of time,” recalled Jessica Langbaum, a senior director at the Banner institute, which hosted the meeting.

In other words, they needed trial subjects with a gene variant called APOE4 — people with a high genetic probability of developing Alzheimer’s. People with two copies of the gene variant constitute an estimated 2 to 3 percent of the general population and 15 to 20 percent of people with Alzheimer’s. Those with only one copy make up about half of Alzheimer’s patients.

One question in Phoenix was how, or even if, these trial subjects should be told their grim genetic profiles, according to a written, contemporaneous account of the meeting.

“Researchers unfortunately have an inherent conflict of interest,” said Dr. Robert Klitzman, director of the Masters of Bioethics Program at Columbia University. “They want people to be in their study, and there are researchers who feel, if I tell people the full facts and risks, they may not want to be in the study.”

A general agreement emerged from the meeting around the importance of transparency; participants would be informed. To cope with this news, they would first undergo genetic counseling.

Subsequently, two pharmaceutical companies, Novartis and Amgen, committed to work with Banner to test an experimental drug. The trial abruptly ended in 2019 after participants experienced a “worsening in some measures of cognitive function,” according to Novartis.

Eisai took a different approach on disclosure in its Leqembi study.

In its trial protocol, the company specified that it wanted participants already experiencing mild cognitive decline. “No less than 70 percent” would have the APOE4 gene. Carriers were known to face a higher risk of brain injuries, especially those with two copies.

Before joining the trial, all volunteers had to sign a consent form. The form said they would be tested for a genetic profile that meant a higher risk of bleeding abnormalities from the drug, including cerebral microhemorrhages and brain swelling. But the form stipulated that the test results were “for research purposes” and “will not be shared with you, any insurance company, your employer, your family or any other doctor who is treating you.”
In all, the trial enrolled 957 people with one copy of the risky gene and 274 with two copies.

Dr. Marwan Sabbagh, a neurologist who advised Eisai on Leqembi, called the disclosure of genetic information “study dependent.” “Every study decides to do it a little differently,” he said.

But Arthur Caplan, a leading bioethicist at the New York University Grossman School of Medicine, said trial participants should know the dangers they face. “It’s not even a matter of ethics, it’s a matter of common sense,” he said after The Times told him about the nondisclosure clause.

An F.D.A. spokesman, Jeremy Kahn, did not address The Times’s questions about the appropriateness of the nondisclosure provision, other than to say that the agency had reviewed the trial protocol and determined it was safe.

Asked why Advarra’s institutional review board had approved the decision to keep participants in the dark, a company spokeswoman, Mel Johnson, wrote, “I’m afraid at this time I’m not going to be able to get you answers on this one.” She declined to explain why.

Institutional review boards are an outgrowth of the National Research Act of 1974, passed in response to ethical violations in clinical trials. The boards are supposed to protect the rights and welfare of human research subjects.

Originally, university-based ethics boards reviewed most drug trials, but in recent years drug companies have found it more efficient to pay a single review board to oversee trials with multiple sites. Recognizing that a profit could be made from running these boards, private equity began buying them up.

In 2021, just two private-equity-controlled companies — Advarra and WCG — reviewed 92 percent of drug trials submitted to independent I.R.B.s, according to a Government Accountability Office report last year.

The report cited concerns by some in the industry that “private-equity-backed I.R.B.s are beholden to their clients” and as a result “may be more inclined to approve a protocol and do so expediently in order to satisfy a client.”

Ms. Lane was among those in the Leqembi trial who carried two copies of APOE4. She also possessed another risk factor, the one the F.D.A. worried about when it proposed tightening eligibility requirements in 2010.

She had suffered four previous microbleeds, increasing the likelihood of brain bleeding when taking Leqembi. Eisai’s doctors apparently could not identify them, but Dr. Schrag, who assisted with Ms. Lane’s autopsy, found them when examining her pretrial brain scans.

“I’m quite confident in our interpretation,” he said. “And we published those scans so that people could contest our counts if they wanted to, and nobody has.”

Unaware of her dual risks, Ms. Lane arrived at Charter Research two days before Christmas in 2020 and signed her consent form. In the months that followed, she received only a placebo. On July 25, 2022, she agreed to participate in a new phase of the trial, in which patients could choose to receive the drug.

On Aug. 8, she had her first infusion of Leqembi.

Black Box Warnings

Nearly two years after Leqembi’s approval, several major health care institutions, including Northwestern Medicine, the Beth Israel Deaconess Medical Center in Boston and the Department of Veterans Affairs, have chosen not to give the drug to anyone with two copies of APOE4.

The drug’s rollout has been hindered by its cost — $26,500 a year — its limited efficacy and the need for frequent, expensive M.R.I. scans. More than a third of U.S. neurologists do not recommend Leqembi to Alzheimer’s patients, according to a recent report by Spherix Global Insights, a market research firm.

The F.D.A. required that Eisai include a black box warning urging physicians to consider the drug’s potential risks. Eisai now advises that “testing for APOE4 status should be performed prior to initiation of treatment,” and that prescribers should discuss the risks of ARIA with patients.

At the Alzheimer’s Association, the chief science officer, Maria C. Carrillo, believes that for many patients, Leqembi is well worth the risks.

“I think it’s transformational,” Dr. Carrillo said. “It is not a cure. We understand that. And it has side effects. So it may not be for everyone. But for those that could benefit, it offers more time during the most critical stage where you’re still independent, you still have a lot of opportunity to enjoy time with family, baptisms, weddings, graduations.”

Several experts interviewed for this article, however, argued that the risks outweighed those benefits.

In the Leqembi trial, Eisai reported that 99 people, or 39.8 percent of those carrying two copies of APOE4, experienced brain bleeding; 86 people, or 34.5 percent of those with two copies, had brain swelling. Sixteen percent of patients with one copy experienced brain bleeds.