Alzheimer's disease (AD) is traditionally considered irreversible. However, a team of scientists led by Case Western Reserve University, University Hospitals and the Louis Stokes Medical Center in Cleveland, VA, has provided proof of principle for the therapeutic reversibility of advanced Alzheimer's disease. Using a variety of preclinical mouse models and analysis of human Alzheimer's disease brains, they showed that the brain's inability to maintain normal levels of a central cellular energy molecule, nicotinamide adenine dinucleotide (NAD+), is a major driver of Alzheimer's disease, and that maintaining proper NAD+ balance can prevent and even reverse the disease.
The severity of Alzheimer's disease correlates with dysregulation of NAD+ homeostasis. Image credit: Chaubey etc.., doi: 10.1016/j.xcrm.2025.102535.
AD, generally considered irreversible since its discovery more than a century ago, is the leading cause of dementia and is projected to affect more than 150 million people by 2050.
Current treatments targeting amyloid beta (Aβ) or clinical symptoms provide limited benefit to patients, highlighting the need for complementary and alternative treatments.
Notably, people carrying autosomal dominant AD mutations can remain asymptomatic for decades before clinical manifestations appear, and some people, known as nondemented people with Alzheimer's neuropathology, accumulate abundant amyloid plaques but remain cognitively intact.
These findings imply the existence of intrinsic brain resilience mechanisms that delay or counteract disease progression, suggesting the potential for maintaining or enhancing such processes to alter disease trajectory or enhance recovery from AD.
NAD+ homeostasis plays a central role in cellular resistance to oxidative stress, DNA damage, neuroinflammation, deterioration of the blood-brain barrier, impaired hippocampal neurogenesis, deficits in synaptic plasticity, and neurodegeneration.
In a new study, Case Western Reserve University professor Andrew Pieper and colleagues show that the decline in NAD+ is more severe in the brains of people with AD, and that the same phenomenon also occurs in mouse models of the disease.
Although AD is a unique human disease, it can be studied in the laboratory in mice that have been genetically modified to express genetic mutations known to cause AD in humans.
The researchers used two of these mouse models: one had multiple human mutations in amyloid processing; the other carried a human tau protein mutation.
Both strains of mice develop brain pathology reminiscent of AD, including deterioration of the blood-brain barrier, axonal degeneration, neuroinflammation, impaired hippocampal neurogenesis, decreased synaptic transmission, and widespread accumulation of oxidative damage.
These mice also develop characteristics of the severe cognitive impairment seen in people with AD.
Having found that brain NAD+ levels are dramatically reduced in both human and mouse AD, the scientists tested whether preventing loss of brain NAD+ balance before disease onset or restoring brain NAD+ balance after significant disease progression could prevent or reverse AD, respectively.
The study built on their previous work showing that restoring NAD+ balance in the brain leads to pathological and functional recovery after severe and long-lasting traumatic brain injury.
They restored NAD+ balance by administering a well-studied pharmacological agent known as P7C3-A20.
Remarkably, maintaining NAD+ balance not only protected mice from developing Alzheimer's disease, but delaying treatment in mice with advanced disease also allowed the brain to register underlying pathological events caused by disease-causing genetic mutations.
Moreover, cognitive function was fully restored in both strains of mice. This was accompanied by normalization of blood levels of phosphorylated tau-217, a recently approved clinical biomarker of AD in humans, providing evidence of disease reversal and highlighting an objective biomarker that could be used in future clinical trials for AD restoration.
“We were very excited and encouraged by our results,” said Professor Pieper.
“Restoring brain energy balance led to pathological and functional recovery in both strains of mice with late-stage Alzheimer's disease.”
“Observing this effect in two completely different animal models, each driven by different genetic causes, strengthens the new idea that recovery from advanced disease may be possible in people with AD when NAD+ balance in the brain is restored.”
The results encourage a paradigm shift in how researchers, clinicians, and patients can think about AD treatment in the future.
“The main takeaway is a message of hope: the effects of Alzheimer's disease will not necessarily be permanent,” Professor Pieper said.
“A damaged brain can, under some conditions, recover and regain function.”
“Through our study, we have demonstrated one way to achieve this goal with drugs in animal models and have also identified candidate proteins in the brain of people with Alzheimer's disease that may be relevant to the ability to reverse Alzheimer's disease,” said Dr. Kalyani Chaubey, a researcher at Case Western Reserve University and University Hospitals.
In animal models, currently available over-the-counter NAD+ precursors have been shown to increase cellular NAD+ to dangerously high levels, promoting cancer development.
However, the pharmacological approach in this study uses a pharmacological agent (P7C3-A20) that allows cells to maintain proper NAD+ balance under conditions of severe stress without increasing NAD+ to supraphysiological levels.
“This is an important factor when considering patient care, and clinicians should consider the possibility that therapeutic strategies aimed at restoring the brain's energy balance may offer a path to recovery,” Professor Pieper said.
conclusions appear in the magazine Cell reports, medicine.
_____
Kalyani Chaubey etc.. Pharmacological reversal of late-stage Alzheimer's disease in mice and identification of potential therapeutic nodes in the human brain. Cell reports, medicinepublished online December 22, 2025; doi: 10.1016/j.xcrm.2025.102535
