A new proprietary analysis of cutting-edge Alzheimer's drugs finds growing excitement around experimental therapies targeting a brain protein called tau, despite uncertainties about effectiveness and safety.

Why it matters: Existing Alzheimer's drugs have shown limited benefits and potentially serious side effects.

• The market for next-gen treatments could be worth as much as $70 billion within a decade, given the more than 7 million Alzheimer's patients in the U.S. and the aging population, the TD Cowen analysis found.
• But expensive new treatments could intensify affordability concerns as Americans live longer and there's more early recognition of Alzheimer's.

Where it stands: Researchers and clinicians are increasingly focusing on tau, a protein that helps neurons communicate in healthy brains but has been linked with diseases like Alzheimer's and frontotemporal dementia when it builds up to abnormal levels.

• In July, a tau-targeting therapy from Biogen was shown to slow cognitive decline in people in the early stages of Alzheimer's, with only mild or moderate side effects. The Alzheimer's Association said it was "the kind of signal that gives the field reason for optimism."
• 88% of physicians who TD Cowen surveyed said they're either likely or very likely to use an effective therapy that doesn't require as much monitoring for brain swelling or other side effects associated with existing drugs.
• "The physician community agrees this is a hot target," Joseph Thome, TD Cowen managing director and senior research analyst, told Axios, adding that slowing disease progression by 25%–30% is the bar for the new treatments.

By the numbers: Just over half of the Alzheimer's population (about 3.8 million people) have mild cognitive impairment, which makes them the likeliest to benefit from the new drugs.

• If three-quarters of that group opted in for a treatment costing $20,000 a year, the market could be worth $60 billion and grow to $70 billion by 2035, per the analysis.

Reality check: The hunt for Alzheimer's cures has been long and frustrating.

• There are 158 drugs in development, some of which aim to modify the disease, others to treat its symptoms.
• A new treatment that could slow cognitive decline without as many safety risks would be a meaningful improvement for patients.

The FDA hasn't yet approved any tau therapies. Researchers have made strides understanding the way tau can factor in brain diseases of lab animals like mice, but those findings may not translate to humans.

• There's also disagreement about how much tau can be removed safely without creating other neurological issues — and the most effective way to administer the drugs so that they cross the blood-brain barrier.
• The commercial success of any new drug will ultimately hinge on whether Medicare or commercial insurers decide to pay for it.

Between the lines: Billions of dollars have already been spent developing Alzheimer's therapies that target another brain protein called amyloid.

• Two of these drugs — Eisai and Biogen's Leqembi and Eli Lilly's Kisunla — have shown modest success slowing Alzheimer's progression in patients with mild cognitive impairment. But they come with safety risks like brain swelling, bleeding and other adverse events.
• Concerns about the affordability of anti-amyloid drugs largely didn't pan out because of low uptake, but a more effective or safer class of drugs would face the same cost issues.

Researchers say the future may lie in combination therapies that use new and old drugs, similar to the model used to treat some cancers.

• Patients could be treated early on with a drug that removes amyloid, followed by one that suppresses tau production — potentially alongside drugs that control inflammation.
• About 84% of physicians TD Cowen surveyed said they'd likely or very likely use tau therapy alongside an anti-amyloid treatment.

What we're watching: How the tau-targeting drugs perform in late-stage trials, and how much manufacturers like Biogen, Novartis and Alector invest to reach development milestones.

• "It might take several years for these therapies to pan out, but they have the potential to be really life-changing," Thome said.

AI could narrow the projected health care workforce shortage by more than a third through 2040, by automating documentation and handling more administrative tasks like billing, according to a new report from the health care education company Covista.

Why it matters: Shortages of nurses, physicians and other providers drive up wait times, increase the chances that a health problem will be missed and can shorten average life expectancy.

• That could result in $140 billion in avoidable health costs by 2040, per the report.
• But while AI may deliver new efficiencies and better serve patients, it also can help health systems bill more for the same level of care and drive up costs.

What they found: AI could close 37% of the projected 2040 workforce gap without replacing a single worker, by transcribing or summarizing office visit notes, drafting referral orders, pre-filling medication lists and handling other tasks, according to the Covista Healthcare Workforce Index.

• In the base case, use of the technology results in a 5% productivity gain among physicians and a 2.5% gain among nurses and allied health staff.
• Those percentages could more than double if AI becomes commonplace in imaging and other diagnostics, coordinating care, surgical assistance and drug dispensing.
• But AI's ability to catch health problems earlier and surface unmet care needs can also increase demand for services, potentially offsetting many of those productivity gains.

Reality check: AI adoption has been limited by tight budgets, skepticism from patients and clinicians and structural barriers like nurse-patient ratios.

• There's also a patchwork of AI tools for individual tasks that can overwhelm IT teams.

The bottom line: AI can do things like help direct care or summarize an appointment. But it may never get a patient's full trust or have the precise experience and intuition that a physician possesses.

• Fixing workforce shortages will also require breakthrough therapies, changes in care models and other innovations.
• "AI matters, but it isn't a fix on its own," the analysis states.