The Alzheimer’s Drug Approval Mess Has Left Millions Of People In Limbo

By Robert Langreth

Medicare won’t pay for Aduhelm, the FDA-approved Alzheimer’s treatment, and leading drugs aimed at pharma’s primary target have repeatedly fallen short.

Geri Taylor ran a large long-term care facility into her mid-60s, when she started becoming forgetful. One time, she was in the middle of running a staff meeting and lost her train of thought, couldn’t get it back, and one of her deputies had to take over. Another time, she got off at the wrong Manhattan subway stop and had no idea why she was there or where she was going. Incidents like these led her to retire earlier than she might have, but she put off seeing a neurologist for years, until one day she went into the bathroom and couldn’t recognize her own face in the mirror. Finally, in 2012, came the diagnosis she’d feared: mild cognitive impairment, likely due to Alzheimer’s disease.

In 2015, Taylor enrolled in a clinical trial for a drug meant to slow cognitive decline. Tests confirmed that her brain, like those of most Alzheimer’s patients, contained an unusual buildup of a protein called amyloid. For a large part of the past three decades, amyloid clumps, called plaques—a hallmark of the disease—have been the prime target for most Alzheimer’s researchers. Rid the brain of enough clumps or prevent their buildup, the consensus goes, and you could slow cognitive decline, perhaps indefinitely. This was the great hope of the trial Taylor joined for the compound Aduhelm. Made by Biogen Inc., Aduhelm was the latest in a long line of experimental anti-amyloid drugs that never quite seemed to work. But this one, the researchers said, was the biggest amyloid reducer yet, and it might open a whole new era of Alzheimer’s treatment, improving each patient’s quality of life for years or more.

Taylor’s husband, James, a retired IBM manager, drove her the two hours from their Manhattan apartment to Yale, the nearest trial site, for her monthly infusions. Even before she knew she was getting the Aduhelm, not the placebo, just being in the trial gave Taylor the kind of hope that was otherwise in short supply.

Seven years later, doctors can’t agree on whether the drug helped her or anyone else. Aduhelm’s two advanced drug trials yielded contradictory results. Nonetheless, last year the US Food and Drug Administration approved Biogen’s drug for widespread use, drawing a level of backlash rarely seen in medical circles. This spring, the federal Centers for Medicare & Medicaid Services (CMS) refused to cover Aduhelm under Medicare unless it was part of a clinical trial, arguing that the drug’s efficacy hasn’t been adequately demonstrated and that questions about its safety remain unanswered. It was the biggest turn yet in a bizarre saga that has repeatedly left Taylor and 6 million other Alzheimer’s sufferers in limbo, briefly raising their hopes only to dash them.

“It is very hard to have nothing,” Taylor says from her 21st-floor apartment on the Upper West Side. “This at least gave us a chance.” She’s sitting next to James on a green leather couch, sometimes holding his hand. She speaks slowly and hesitantly, in short sentences using simple words, and appears confused by some questions. Her husband does most of the talking and sometimes rephrases questions to help her understand. Taylor received the drug until this spring through another clinical trial, but she’s stopped taking it while she receives treatment for her recently diagnosed breast cancer. For now, the options for Aduhelm patients like her are unclear. A year’s supply of the drug costs $28,200 out of pocket, well beyond the reach of most people until further research can resolve the feds’ concerns.

For Alzheimer’s patients, their families, researchers, and the pharmaceutical industry, the bigger question is whether drugs targeting amyloid have been a multibillion-dollar dead end. Amyloid plaques definitely have something to do with the disease, and other trials are examining whether amyloid reduction can play a more marked role in preventing the disease in at-risk populations. But all we know for certain after decades of research is that if removing amyloid helps people with early Alzheimer’s at all, the benefits are modest, and the evidence that plaques are responsible for the average case of Alzheimer’s remains largely circumstantial. “If these wipe out amyloid and people are not much better,” says Mary Sano, who directs the Alzheimer’s Disease Research Center at Mount Sinai’s Icahn School of Medicine in New York City, “what do you do next?”

In 1906 the German psychiatrist Alois Alzheimer reported finding strange protein clumps and tangles in the brain of a 51-year-old woman who died with dementia. The plaques were found in the spaces between neurons and were later shown to consist of sticky fragments of amyloid. The tangles, found inside neurons, were later shown to be made up of a second protein called tau. For decades, scientists thought Alzheimer’s disease was an exceedingly rare problem found only in relatively young people. In the 1970s, however, pathology studies showed that it was common in elderly people with dementia. Alzheimer’s is thought to be the most common form of dementia, in which cognitive abilities decline enough to interfere with daily life. But all the various types of dementia may overlap, and dementia caused solely by Alzheimer’s may be less common than people assume.

In the 1980s scientists purified the amyloid protein in the plaques, allowing molecular biologists to study them in more detail. In the early ’90s geneticists found that patients with certain rare, early-onset forms of Alzheimer’s had mutations in a gene responsible for creating amyloid’s precursor protein. This strongly hinted that amyloid could directly cause Alzheimer’s. British geneticist John Hardy, who found the first of these genes, popularized the amyloid cascade hypothesis in a 1992 essay that suggested amyloid buildup led to a host of other problems, including brain cell death and the tau tangles.

The relationship, others pointed out, wasn’t clearly causal. Amyloid is more common in people with dementia, but hasn’t been conclusively shown to be the primary culprit. It has been found in the brains of many cognitively normal people, and tau buildup correlates even more closely with neurodegeneration. Still, the amyloid hypothesis found powerful supporters. In the US, Harvard Medical School neurologist Dennis Selkoe championed the idea that amyloid was the primary cause of Alzheimer’s. Selkoe, now 78, leads a 15-person lab on the 10th floor of a research building at Brigham & Women’s Hospital in Boston and has published more than 100 papers on amyloid.

From the beginning, drug trials targeting amyloid have been frustrating. One of the first was conducted by Elan Corp., which had acquired Selkoe’s own amyloid-focused biotech company. Elan abandoned its trial in 2002 after some patients developed brain inflammation. The experimental vaccine it was testing completely cleared amyloid from the brains of at least five patients, but they still died with dementia, according to independent follow-up studies. Elan also pioneered a second amyloid-removing antibody drug that was later developed by Pfizer Inc. and Johnson & Johnson; it failed in several large trials. An analysis published in 2021 combined data from 14 trials of anti-amyloid drugs and found no clear effect on cognitive functioning. The most recent setback came in June, when a study in Colombia showed that an amyloid-lowering antibody from Roche Holding AG failed to prevent decline in people with a rare gene mutation that causes early-onset Alzheimer’s.

Selkoe argues that the evidence backing amyloid as the culprit is as strong as ever and has merely been held back by corporate and scientific missteps, such as insufficient doses. “There are reasonable scientific explanations for why the trials failed,” he says. “They were incredibly ill-conceived and ill-executed.” His lab is still actively designing new amyloid antibodies, and he hasn’t wavered in his conviction that the drugs will eventually work. “The biological evidence keeps getting stronger every week,” he says.

Biogen’s Aduhelm (generic name: aducanumab) was supposed to put all the doubts to rest. The company’s initial trial clearly showed that the drug removed large amounts of amyloid from patients’ brains. After Biogen announced preliminary results at an investor conference in December 2014, Geri Taylor’s husband read a blurb about it, and started calling trial sites to see if any were still accepting people. Scans revealed that Geri had amyloid buildup in her brain, which qualified her for the trial. She also had two copies of an Alzheimer’s-associated gene called APOE4, which meant a higher risk of the disease for her son.

Although some people in her trial were assigned a placebo for the first year, everyone got various doses of the active drug after that. Taylor eventually learned she’d received the highest dose of the real drug from the start and never experienced any of the side effects, which can include headaches, dizziness, and brain swelling or brain bleeding. In the first several years she was on the drug, her cognition declined slowly, her husband says, but those losses added up. Eventually, she couldn’t remember numbers.

While Taylor’s Phase I trial continued, Biogen started two Phase III trials to determine whether the drug slowed cognitive decline, eventually enrolling more than 3,200 patients with early Alzheimer’s symptoms. Then in early 2019, Biogen made a fateful decision that stunned Alzheimer’s researchers. It halted its two big trials before they were complete, after a preliminary analysis merging data from both trials indicated the drug was unlikely to work.

Biogen wound down its other Aduhelm trials, and Taylor had to stop taking the drug. She first heard the news from a friend calling to ask about it and didn’t believe it at first. During that period, she developed increasing problems finding her words. “My language just went out,” she says. “I used to speak very well.”

After halting the two big trials based on a joint data analysis, Biogen decided to look at the numbers from each trial separately. By late 2019 one trial was still showing the drug had failed, but there appeared to be a small positive effect in the second. The company reversed course again and said it would apply for approval based on the positive results. In spring 2020 it began a new trial for patients who’d taken part in previous trials, including Taylor.

But Biogen’s early halt to both big trials threw their contradictory results further into question. It was like stopping two close baseball games in the seventh inning, then trying to predict who would have won if the games had been completed and also which game should matter more. Biogen noted that more patients had received the full dose of the drug for an extended period in the trial that worked.

The FDA sidestepped this efficacy debate last June by approving the drug based solely on its ability to lower amyloid concentration, declaring that this was reasonably likely to deliver a later benefit. Internally, the accelerated approval was thought of as a compromise—allowing patients to get the drug while conditioning broader approval on the completion of a third, more successful trial, says Yaning Wang, a former FDA official involved in the review.

After the approval, all hell broke loose. Three members of an FDA panel of advisers that had voted unanimously against the drug in November 2020 resigned in protest. Many hospital systems decided not to administer it, including Selkoe’s Brigham & Women’s Hospital. The inspector general of the US Department of Health and Human Services (HHS) opened a broad investigation of the FDA’s accelerated-approval procedures, including the events leading to Aduhelm’s clearance. Before it decided not to cover the drug, Medicare raised premiums sharply for 2022 partly to account for the possibility of Aduhelm-related costs.

Medicare covers virtually all FDA-approved drugs. Given the controversy, however, the government took the unusual step of commissioning a monthslong review of the need to cover Aduhelm. The Taylors did their best to help advocacy groups pushing for coverage, including meeting with Medicare officials over Zoom and rallying in front of HHS headquarters in Washington. “We have basically one thing, and to take that away, it is just wrong,” says Taylor. It didn’t work; the agency said the benefits of cognitive improvement weren’t proven and there were real risks of serious side effects. “Our foremost goal is to protect beneficiaries from potential harm from an intervention without known benefits in the Medicare population,” CMS Chief Medical Officer Lee Fleisher told reporters after the preliminary decision in January.

Medicare’s decision to restrict coverage to people in clinical trials, which was finalized in April, essentially killed Aduhelm as a commercial product. Biogen halted almost all marketing of the drug, withdrew its application for approval in Europe, and said it would look for a new chief executive officer. The company is providing Aduhelm free to people who started taking it prior to the Medicare decision, while others continue to get it through clinical trials. Anyone else who wants it would likely have to pay cash. Biogen reported a paltry $2.9 million in Aduhelm sales through the first six months of this year.

Some researchers who think the drug amounts to snake oil argue that it should be taken off the market entirely. “How many hundreds or thousands of people do we need to actually put through a trial to say it doesn’t work?” asks Rob Howard, a professor of old-age psychiatry at University College London. Harvard neuroscience professor Jie Shen, who’s pursuing other avenues for treatment including genetic malfunctions, says: “To this day, after trying so hard, there are no facts linking amyloid to neurodegeneration.”

The industry’s focus has started to shift, but there’s no leading candidate to replace the amyloid hypothesis. “It was very appealing as a silver bullet—if we just find a way to get rid of it, we are done,” says Ralph Nixon, a professor of psychiatry and cell biology at NYU Langone Health who’s been studying other potential causes. “The investment was so enormous that it crowded out any other idea.”

Beyond Alzheimer’s, there are many contributors to dementia. Others include ministrokes and vascular damage from diabetes and hypertension, as well as a misfolded protein called TDP-43 found in the brains of many elderly people. One emerging set of drug targets are the mysterious immune cells called microglia. Normally they help prune nerve cells and clear debris, including amyloid, but when they go awry, they may trigger damaging brain inflammation. Of the 90-plus Alzheimer’s risk genes that have been discovered, more than half implicate microglia or brain inflammation, says Rudolph Tanzi, a neurology professor at Harvard Medical School and Massachusetts General Hospital. If the inflammation is the true culprit, he says, then removing amyloid in someone already suffering from Alzheimer’s is like expecting statins to cure someone whose heart is already failing.

“If you say amyloid doesn’t matter in this disease, you are hiding your head in the sand,” Tanzi says. But for people who are already sick, “just stopping the cause of it that happened 30 years ago isn’t going to help.” He’s more optimistic about using anti-amyloid drugs earlier as a preventative measure, like statins for bad cholesterol, though he says safer and more convenient versions are prerequisites for mass adoption.

Tanzi says his lab has identified more than 100 possible compounds that might block brain inflammation. And numerous startups are now trying to devise drugs that counteract faulty microglia, including Alector Inc., which is testing with partner AbbVie Inc. an antibody drug that reprograms microglia, and Vigil Neuroscience Inc., which is working on a pill that could restore microglia’s healthy function. The Vigil drug hasn’t yet begun human tests.

The most prominent drugs in late-stage testing target amyloid, so success in the short term still mostly relies on the amyloid hypothesis being correct. Over the next year, results from trials of three more amyloid-lowering drugs—from Eli Lilly, Roche, and Eisai—are expected to roll in. All three are using relatively high doses of antibodies that scientists believe are more likely to show an effect. Shortly after the Medicare decision this spring, Biogen began a trial that aims to resolve questions over Aduhelm’s efficacy, but it won’t yield results for several years.

Eisai Co. is expected to release its results as early as September. It invented one of the first drugs to treat Alzheimer’s symptoms more than two decades ago, a pill called Aricept, and has been collaborating with Biogen on anti-amyloid drugs since 2014. “At some point we are going to bring forward a treatment to slow the disease,” says Ivan Cheung, global Alzheimer’s disease officer for Eisai. “Until that is done, we are not going anywhere.” Eisai’s antibody lecanemab, licensed from Swedish biotech BioArctic AB, targets the removal of a slightly different form of amyloid than previous medicines, one that many researchers suspect is more likely to be involved in nerve damage.

In Eisai’s Phase II study, lecanemab failed to meet the primary goal of slowing clear decline after 12 months. But after 18 months, those receiving the highest dose had declined at a slower rate than people on a placebo. It’s this high dose that the company is testing in a Phase III trial with 1,800 people. Eisai says it won’t halt the trial early as Biogen did. Analysts at Jefferies Group LLC give the trial a 35% probability of success.

Roche is now testing its drug, which previously failed to show an effect, at almost five times the original dose. Eli Lilly & Co., which plans to complete its latest trial next year, has had arguably the most promising results so far, slowing the rate of cognitive decline by almost one-third over 18 months in a 257-person study of its drug donanemab published last year. But that’s no silver bullet, either. “I don’t think amyloid-reducing drugs are going to stop disease progression,” said Daniel Skovronsky, Eli Lilly’s chief scientific and medical officer, at an investor conference in May. “I think we’re looking at a class that probably will have 20% slowing or something like that, which is a good start.” In an interview, Skovronsky calls Alzheimer’s the most challenging disease that Lilly works on and says that companies are “right on the verge” of making real progress.

If any of the trials clearly succeed, insurers including Medicare would be under tremendous pressure to cover the drug, even if the efficacy is modest. If they all fail, backers of the amyloid hypothesis will have few arguments left to justify further tests. In the meantime, Eisai and Lilly are seeking accelerated FDA approval for their products, but Medicare has ruled that it won’t broadly cover anti-amyloid drugs unless they can show strong enough evidence of cognitive benefit to get the FDA’s full—meaning nonaccelerated—approval.

Most drugs that enter human testing fail, especially those aimed at neurological diseases. What’s different about Alzheimer’s has been pharma’s tunnel vision. Clear results from this batch of trials could finally resolve the amyloid debate enough to start redirecting resources toward other potential treatments.

The Taylors are doing their best to push amyloid drugmakers onward. In June they flew to Indianapolis to meet with Lilly’s dementia team. They’ve also Zoomed with officials at Eisai and Roche. Geri knows that even if they work, the other drugs may arrive too late to be useful for her. She goes on walks with her husband and generally tries to stay positive. “You have got to work with what you have got,” she says. She hasn’t given up hope of eventually getting back on Aduhelm or a similar drug. “My only chance is the one I have been in.”

Share this