Bioethics Forum Essay

Don’t Give Symptom-Free People Alzheimer’s Drugs

A large government-funded study of solanezumab, an experimental Alzheimer’s drug that targets amyloid in the brain, has found no benefit in people with normal cognitive function who have elevated amyloid levels. The results of this long-term, definitive, randomized controlled trial should stay the alarming enthusiasts who have advocated drugging perfectly normal people who have abnormal amyloid levels. But, after decades of investment in unproven anti-amyloid treatments, it’s unlikely that more data will lead industry and industry-funded groups to reverse course.

The Alzheimer’s Association, in fact, recently proposed new diagnostic criteria that define Alzheimer’s “biologically,” using plasma biomarkers for amyloid plaque. Astoundingly, the association recommends that someone with abnormal amyloid biomarkers be diagnosed with Alzheimer’s – regardless of whether the person has memory loss or any other symptoms of Alzheimer’s.

The assumption is that people with plaque in their brains will eventually develop symptoms, and if they live their whole lives without ever developing cognitive issues then they just didn’t live long enough to experience their preprogrammed decline. This is not only wrong, but dangerous. Many people ages 65 and older have significant amounts of amyloid plaque, yet never have cognitive impairment. The connection between plaque and Alzheimer’s is unclear – while amyloid is a hallmark of Alzheimer’s, there is insufficient evidence to support the theory that it actually causes the disease.

The pharmaceutical industry would love to target cognitively normal people who have amyloid plaque because that’s a huge market–30%-to-50% of older individuals. Several pharmaceutical companies have new drugs that remove amyloid plaque and are meant to treat early stages of Alzheimer’s disease with cognitive impairment, and more of these drugs are coming down the pike. In 2021, the Food and Drug Administration gave accelerated approval to aducanemab (Aduhelm), and in July 2023, Eisai/Biogen’s lecanemab (Leqembi) received full approval. Medicare will cover lecanemab for eligible patients enrolled in a registry and major health systems are expected to offer the drug. The FDA is likely to decide whether to approve the latest anti-amyloid drug, Eli Lilly’s donanemab, by the year’s end. 

At the same time, blood tests that detect amyloid are now available, including one that can be purchased directly online by consumers. None of these tests have received FDA approval. While neither the testing companies nor advocacy groups recommend diagnosis by blood test alone at this point, they are already signaling their enthusiasm for simple blood tests they think will revolutionize the diagnostic process.

Until the blood tests came along, assessment of amyloid plaque was determined either by a PET scan or analysis of cerebrospinal fluid collected by lumbar puncture. PET scans are expensive and available only in metropolitan areas; lumbar punctures are invasive, uncomfortable and can cause severe headache and other side effects. None of these analyses is sufficient to diagnose Alzheimer’s.

The recent availability of both new drugs and new blood tests raise the question of whether the timing of the Alzheimer’s Association’s updated diagnostic guidelines is entirely coincidental. As one of the authors of the new guidelines explained, “it is completely feasible to diagnose the disease biologically at a mass scale” and “there’s something you can actually do about the disease.” In other words, access to blood tests will funnel more patients – and more perfectly normal people–to expensive amyloid-busting drugs.

People without symptoms of cognitive loss should not take these drugs, which have substantial risks, including death. It’s not even clear that patients who actually have Alzheimer’s disease should take these drugs, because they don’t improve symptoms. In clinical trials, patients in both treatment and placebo groups continued to decline, and the small difference between them, while statistically significant, was not clinically meaningful. In other words, neither patients nor families noticed any difference.

Anti-amyloid drugs cause serious harms, including brain bleeds and brain shrinkage. Three patients are known to have died in clinical trials for Leqembi and another three for donanemab, suggesting a rate of 1 to 2 deaths per 1,000 patients in the healthier-than-normal clinical trial population. Among the general population, who may have multiple illnesses and may be on multiple drugs, this death rate is likely to be higher.

In addition, anti-amyloid drugs, and the costs associated with treatment, are high—in the case of Leqembi, expected to run to $82,500 a year per patient. Because most patients receiving this drug are on Medicare, most of these costs will be borne by taxpayers.

Diagnosis by blood test makes sense if the goal is to identify many patients as soon as possible. Of course, drug companies want to expand the pool of people eligible for an early Alzheimer’s diagnosis, because the available anti-amyloid drugs are approved only for patients with early-stage disease. Currently, a diagnosis of Alzheimer’s involves multiple steps, including cognitive assessments, and there is a limited number of specialists with the requisite expertise to confirm the diagnosis. That’s the way it should be.

Industry and industry-funded advocacy groups have set their sights on streamlining the diagnosis process to maximize the number of people eligible for presymptomatic treatment. But if individuals with increased levels of a biomarker never experience cognitive impairment, then what, exactly, is being diagnosed or treated?

In the solanezumab trial, both the treated and placebo groups had the same rate of cognitive decline over 4.5 years. There is no s that any of the other drugs that target amyloid will do better in preventing cognitive decline in normal people. 

And it’s ethically questionable to treat asymptomatic patients solely based on a biomarker. Patients are, naturally, most concerned about actual symptoms, not how much of a biomarker is in their bodies. As an epidemiologist noted, removing a biomarker can “cure” a disease without improving a patient’s experience at all. And a diagnosis of Alzheimer’s can devastate both patients and families. Recommendations to label cognitively normal people as Alzheimer’s patients, and thus make them eligible for treatment with expensive, dangerous drugs, are unconscionable.

Spending billions of dollars–sales of Leqembi and  donanemab are projected to reach $5.5 billion globally by 2030 – to remove a biomarker without evidence that it is a cause of a disease is an unwise investment. Those dollars could be better spent on resources to improve known modifiable risk factors, including hypertension, hearing impairment, and diabetes, that account for around 40% of worldwide dementias. What’s worse than the misspent dollars, however, is the false hope patients and their families are being given when industry and its minions promote drugs that have shown no clinically meaningful benefit and have significant harms.

Adriane Fugh-Berman, MD, is a physician and professor of pharmacology and physiology and family medicine at Georgetown, where she directs PharmedOut, a research and education project that promotes rational prescribing. Judy Butler is a research fellow at PharmedOut. @Pharmed_Out