Lecanemab or Lithium for Alzheimer Disease Dementia? New Data Change Everything

Lecanemab is being heavily promoted for treatment of minimal cognitive impairment (MCI) and early Alzheimer disease (AD) dementia. But recent data support consideration of low-dose lithium for this role and for primary prevention of cognitive decline. Should lithium be offered as an option for patients who are concerned about AD?

This is the first of 2 articles addressing this question. The second will compare the benefits, risks, and costs of lecanemab and lithium. But first, here is a brief history of the lithium-for-dementia concept, followed by a summary of the key findings in the recent Nature article by Liviu Aron and colleagues (senior author Bruce Yankner).¹ Understanding these findings is essential for a clinical comparison of lithium and lecanemab.

A Brief History

That idea that lithium might help prevent dementia has a long history, but it gathered steam following the publication of a small randomized trial in patients with MCI, by Orestes Forlenza and colleagues.²

In the Forlenza study, ADAS-Cog scores for participants receiving lithium carbonate increased from 1.6 points (11.0 to 12.6 ; increasing values indicate worsening cognitive function), while scores for the placebo group increased by 3.2 points (P=0.03).

Plans for a study replicating the Forlenza trial were submitted to ClinicalTrials.gov in 2017. The results were recently reported, 8 years later.³ The primary outcome measure was the California Verbal Learning Test. Scores range from 0-16; decreasing indicates worsening. The lithium group’s scores decreased from 7.95 to 6.46 over 2 years, while those for placebo group declined from 7.90 to 5.10, a significant difference favoring lithium. Interestingly, and surely disappointing to the investigators, most of a large battery of secondary outcome measures were negative.

To date we have a substantial body of indirect data suggesting lithium may be able to counter cognitive decline,⁴ and 2 small-but-mostly-positive randomized trials in patients with MCI. For comparison with the following data, remember that those 2 randomized trials used lithium carbonate.

Brain Lithium Concentrations

Using human postmortem brain sections, Aron et al found that relative to people with no cognitive impairment, those with MCI had less lithium in their prefrontal cortices and those with AD had even less. They repeated the study in a different sample, finding the same result.

Lithium and Plaque Formation

Where did the lithium go? Recall that the pathology of AD is the aggregation of A-beta protein in plaques, and creation of neurofibrillary tangles involving the phosphorylated form of another protein, tau. Aron et al showed that in a strain of mice that develops this same pathology, including cognitive decline, lithium was sequestered in amyloid plaques as the mice aged. The more lithium in plaque, the less was available in the nonplaque space.

The researchers then showed that a lithium-deficient diet, which reduces nonplaque lithium, led to more plaque formation. This implies that nonplaque lithium inhibits plaque formation, creating a positive feedback loop: as lithium is sequestered in plaque, nonplaque lithium concentrations decrease; these lower nonplaque lithium levels allow more plaque formation, which increases lithium sequestration; and so on.

Giving the Mice Lithium Inhibits Plaque Formation

To test this presumption, the researchers put lithium in the water of the same strain of mice from early adulthood onward, while control mice received regular water. Since these are AD-prone mice, the controls developed a substantial plaque load as they aged. But the mice who received lithium all their adult lives had less than 10% of the control plaque load. Lithium also reduced phosphorylation of tau protein. These results demonstrate a presumptive mechanism by which lithium could prevent AD dementia: by inhibiting plaque formation and the phosphorylation of tau proteins.

Lithium Orotate vs Lithium Carbonate

Won’t the extra dietary lithium get sequestered in plaque just like the mouse’s own brain lithium? To address this potential problem, the researchers noted that lithium orotate is less electrically charged than lithium carbonate and speculated that this might help lithium stay in the nonplaque space. To test this, they used both forms in their lithium supplementation experiments. Their results are shown schematically in Figure 1.

Lithium orotate worked far better than lithium carbonate in preventing plaque development. Lithium carbonate was better than plain water, but not significantly so. Similar results were obtained for prevention of phosphorylation of tau proteins. (Might it have been orotate, not lithium, that was the active ingredient? The team controlled for this as well, giving some mice sodium orotate instead, which produced no benefit.)

Can Lithium Reverse Cognitive Decline?

Given the strength of the aforementioned results, Aron et al proceeded to test whether lithium might be able to undo the pathologic changes of AD when those changes are already underway. They waited to begin lithium supplementation until starting and continuing from 17 to 22 months of mouse age, roughly akin to 55 to 65 human years. In these AD-prone mice, plaque formation and memory impairment begin at adult maturity, 5 to 6 months of age, comparable with the “early onset” form of AD dementia.⁵

The point of controlling plaque and tangle development is to preserve cognitive function, particularly memory. Figure 2 shows a measure of mouse memory: time spent in a branch of a water maze where a reward had previously been found (one of several tests of mouse cognitive function reported, all of which show the same result).

Wild type mice are not AD prone. Their results indicate the performance one can expect from a healthy old-age mouse. The other 3 bars represent outcomes for AD-prone mice. Water is the control condition (no lithium added). As you can see, lithium orotate—given at an age when AD pathology is already extensive—almost completely restored cognitive function. Lithium carbonate is less effective.

Concluding Thoughts

Brain lithium prevents amyloid plaque formation and phosphorylation of tau proteins. In the process of AD dementia, lithium is sequestered in plaques, creating a positive feedback loop: more plaque, less lithium, leading to more plaque, and so on. Giving lithium orotate to young adult mice almost completely prevented plaque formation and tau phosphorylation. Starting lithium orotate after plaques and phosphorylated tau have already formed almost completely reversed the expected cognitive impairment. Lithium carbonate is far less effective. If all this were true in humans, lithium orotate would be an obvious treatment both to prevent AD dementia and to treat it once detected.

Of course, skeptics’ first response has been “these are mouse data.” Aron et al point out that lithium levels in human and mouse brains are comparable, supporting the relevance of mouse models for studying the biological effects of lithium. Skeptics, including a prominent neurologist following a national presentation on AD treatment, have said that we should wait for a randomized trial of lithium orotate in humans (personal communication, August 2025). But the recent lithium carbonate randomized trial took 8 years to mount and complete. What shall we suggest to patients and families for the next 8 years?

A healthy lifestyle—including a Mediterranean-like diet, regular physical activity, and avoidance of smoking, excessive alcohol, social isolation, sleep disorders, and hearing loss—is an important means of preserving cognitive function in people at risk of developing dementia.⁶

The subsequent article will compare lecanemab and lithium’s benefits, risks, and costs. With ApoE genotyping and the new pTau/amyloid blood test, patients and families need help now deciding between treatment alternatives. 

Dr Phelps is retiring from 30 years of treating complex mood disorders, and recently founded another website, DepressionEducation.org. He is the bipolar disorder section editor for Psychiatric Times® and the author of A Spectrum Approach to Mood Disorders for clinicians and Bipolar, Not So Much for patients and their families.

References

1. Aron L, Ngian ZK, Qiu C, et al. Lithium deficiency and the onset of Alzheimer’s disease. Nature. 2025;645(8081):712-721.

2. Forlenza OV, Radanovic M, Talib LL, Gattaz WF. Clinical and biological effects of long-term lithium treatment in older adults with amnestic mild cognitive impairment: randomised clinical trial. Br J Psychiatry. 2019;215(5):668-674.

3. Gildengers A. Lithium As a Treatment to Prevent Impairment of Cognition in Elders (LATTICE). Accessed November 12, 2025. https://clinicaltrials.gov/study/NCT03185208

4. De-Paula VJR, Radanovic M, Forlenza OV. Lithium and neuroprotection: a review of molecular targets and biological effects at subtherapeutic concentrations in preclinical models of Alzheimer's disease. Int J Bipolar Disord. 2025;13(1):16.

5. Fu Y, Rusznák Z, Kwok JB, et al. Age-dependent alterations of the hippocampal cell composition and proliferative potential in the hAβPPSwInd-J20 mouse. J Alzheimers Dis. 2014;41(4):1177-1192.

6. Lazar RM, Howard VJ, Kernan WN, et al. A primary care agenda for brain health: a scientific statement from the American Heart Association. Stroke. 2021;52(6):e295-e308.