Cautious optimism over lithium orotate as a treatment for Alzheimer’s disease

Alzheimer’s disease (AD) is, by far, the most common cause of dementia and ranks among the most deadly diseases in the modern world. Unfortunately, in contrast to many other deadly diseases such as heart disease, cancer, and type II diabetes, the last century has seen only modest, if any, strides in AD treatment and prevention.

It’s certainly not for lack of trying. An enormous volume of research has been devoted to the disease. Even as our understanding of AD clinical characteristics, pathogenesis, and risk factors has advanced over the decades, a silver bullet against this disease—akin to lipid- and blood pressure-lowering drugs for cardiovascular disease or GLP-1 receptor agonists for diabetes—has continued to elude us. But recently, researchers at Harvard Medical School generated substantial hype that such a silver bullet might in fact come in the form of a different metal—lithium. 

What did they find? And how does it fit within our broader understanding of the pathology and treatment of Alzheimer’s disease?

Why lithium?

Lithium, an alkali metal long used as a frontline treatment for bipolar disorder, is also naturally present in the brain, but the precise nature and extent of its functions are not entirely clear. In a study published last month, investigators Aron et al. sought to characterize these functions, particularly through the lens of how they might relate to the development and progression of AD.¹

The authors first identified lithium as an element of interest by examining levels of an array of metals in post-mortem brain samples from humans with AD, mild cognitive impairment (MCI), or normal cognition. Of the 27 metals they analyzed, only lithium was found to be significantly lower in the prefrontal cortex (an area of the brain that is particularly affected by AD pathology) of humans with MCI or AD relative to cognitively normal controls. Further, the authors found that in all MCI and AD brain samples, lithium was highly concentrated in amyloid-β (Aβ) plaques—a classical feature of AD pathology—whereas non-plaque areas showed significantly less lithium than controls, and lithium in these non-plaque regions correlated with performance on cognitive tests conducted prior to participants’ death (i.e., lower lithium levels were associated with worse cognitive performance).

Together, these results suggested that Aβ aggregation might be disrupting normal lithium distribution and utilization throughout the brain, potentially contributing to neural damage and dysfunction. Thus, the authors proceeded to conduct a series of experiments in mice to explore in depth the effects of lithium deficiency and replacement and the role that this metal might play in neurological health. In other words, is the observed effect a causal one?

Linking lithium and cognitive decline

Using both transgenic mouse models of AD (which spontaneously develop Aβ plaques) and aged wild-type mice, Aron et al. investigated the effects of lithium deficiency, which they induced through near-complete elimination of lithium from the mouse diet. This diet, which reduced lithium levels in the brain by approximately 50%, was found to result in over two-fold greater Aβ plaque deposition in AD mouse models compared to animals on a lithium-replete control diet, in addition to causing increased concentrations of pathogenic Aβ42 fragments in wild-type aged mice (which do not normally develop AD pathology). Lithium deficiency also resulted in elevated levels of phosphorylated tau protein—another typical feature of AD—in transgenic AD mice. Critically, these neuropathological features also corresponded to functional deficits, as both transgenic and wild-type mice deprived of lithium showed impairments in learning and memory tasks relative to animals on normal diets.

But perhaps the greatest cause for excitement came from Aron et al.’s experiments involving lithium supplementation. By testing 16 different lithium salts, the researchers identified one in particular—lithium orotate, which they abbreviated as LiO—that exhibited significantly lower affinity for Aβ plaques than others. This, in turn, meant that LiO was more capable of raising lithium concentrations specifically in non-plaque areas of the brain, without requiring extreme doses or raising lithium levels elsewhere in the body above normal physiological ranges. (By contrast, therapeutic doses of lithium, which are much higher, are restricted to a narrow range, as this metal becomes very toxic at high levels.) In other words, for equal doses of lithium, lithium orotate raises levels of bioavailable lithium throughout the brain more than the formulation typically used in clinical settings, lithium carbonate. Thus, the authors then sought to determine whether supplementation with LiO might block the development of Aβ pathology and cognitive deficits in AD mouse models and aging wild-type mice on an otherwise normal diet (i.e., not a lithium-deficient diet).

Remarkably, transgenic mice treated with a low dose of LiO through most of their adulthood (administered through drinking water for 7–9 months to achieve serum lithium levels comparable to untreated humans, in whom serum levels are approximately the same between those with and without AD) were almost completely free of Aβ plaques or aggregates of phosphorylated tau. By contrast, control animals treated with either water or an alternative lithium salt (lithium carbonate, the clinical standard for bipolar disorder) were found to have significant plaque and tau burdens, as expected for the mouse models used in this experiment. These between-group differences were also mirrored by results on cognitive function tests, in which transgenic animals treated with LiO performed comparably to cognitively normal wild-type animals on learning and memory tasks, while transgenic animals on control treatments exhibited typical deficits in these areas.

Finally, the investigators also assessed the impact of LiO treatment on aging wild-type animals, and results were similarly promising. While these animals do not spontaneously develop AD neuropathological features, the authors tested other readouts of brain aging and cognitive decline—finding, for instance, that LiO reduced markers of neuroinflammation and prevented age-related losses in synaptic connections in the hippocampus (a brain region that is vital for memory). Perhaps most tellingly, aging animals treated with LiO (at levels several orders of magnitude less than doses used for human bipolar disorder) also performed significantly better than untreated controls on learning and memory tasks, nearly matching the performance level of younger animals. These results indicate that LiO can largely block age-related deficits in cognitive function, even outside of the context of Aβ plaques and other AD pathology.

What this study adds… and what it doesn’t

Aron et al.’s work constitutes some of the most comprehensive mechanistic data to date on the relationship between lithium and Alzheimer’s pathology, as well as on the role of lithium in the brain more generally. For example, in addition to the experiments described above, the researchers characterized how low lithium levels altered neuroinflammatory phenotypes and gene expression patterns. They also established an enzyme called glycogen synthase kinase 3β (GSK3β) as a key mediator of the negative effects of lithium deficiency, demonstrating that administration of a GSK3β inhibitor blocked many of the cell- and tissue-level changes they had observed in lithium-deprived animals.

However, despite the implications of many popular press articles about the study,² the general idea that lithium might be an effective AD treatment certainly didn’t originate with Aron et al.’s findings. The neuroprotective potential of this metal, in terms of both neuropathology and cognitive performance, has been recognized for at least a quarter of a century. Several studies in animal models and cell culture from the 1990s and 2000s have established that lithium treatment can decrease Aβ and phosphorylated tau accumulation, at least in part through its capacity to act as an inhibitor of GSK3β.^3–5 Chronic lithium treatment has also been shown to enhance learning and memory both in cognitively normal rodents^6,7 and in various animal models of AD or other forms of cognitive impairment.^8–10

The volume of evidence even includes a handful of randomized trials in humans with MCI¹¹ or AD.^12,13 Though these trials have been relatively small in scale, they have shown significant improvements in cognitive scores among participants on lithium relative to participants on placebo treatments, with no apparent concerns with respect to safety or tolerability. (In AD trials, dosages ranged from 150 mg to 600 mg to achieve target serum concentrations of 0.2-0.6 mmol/L, while the MCI trial microdosed lithium at 300 μg/day.) Indeed, the largest such trial to date—comprising 80 adults with MCI randomized to lithium or placebo for a treatment period of two years—was recently completed, and results are expected to be published in this coming winter.¹⁴

A better formulation for future trials

Given this wealth of evidence and ongoing research, Aron et al.’s study starts to look far less revolutionary. But while it’s true that most of the findings praised in mainstream media have in fact been known for many years, this work does indeed substantially advance the field in a way that is perhaps less attention-grabbing—but nonetheless powerful. 

Investigations on the efficacy of lithium for AD or MCI treatment have thus far (including the recently completed MCI trial) focused on the standard formulation used for treating bipolar disorder: lithium carbonate. As mentioned above, these studies have shown statistically significant benefits for cognition and neuropathology with this treatment relative to placebo, but in absolute terms, the effects were fairly modest—enough to maintain research interest in lithium as a therapeutic strategy, but not enough to generate much excitement that it could be the silver bullet we’ve been waiting for.

But according to Aron et al., lithium carbonate barely scratches the surface of lithium’s therapeutic capacity, as animals on this treatment barely differed from water-treated mice in most outcomes tested. The investigators found that alternative lithium salts—particularly lithium orotate—are far more capable of avoiding sequestration in Aβ aggregates and correcting low bioavailability of lithium in the rest of the brain (while still maintaining total lithium concentrations in normal physiological ranges)—and are thus more effective in eliciting improvements in cognitive and neuropathological trajectories. 

If these results are eventually found to translate to humans, just think how the modest benefits observed with lithium carbonate might be magnified in future trials with lithium orotate. While research to date has focused on lithium carbonate and has struggled with low efficacy and high toxicity, we now see that we can potentially circumvent both of those problems simply by using a different salt (!!). In other words, Aron et al.’s work might someday prove to be revolutionary after all—not because it was the first to establish lithium as a potential therapy for AD or MCI, but because the researchers took the extra time to consider how to maximize the therapeutic potential of this simple metal. (However, it’s worth noting that the potential for LiO to reduce age-related cognitive decline—that is, in the absence of AD pathology—to an equal or greater extent than lithium carbonate is less convincing at present, as this study did not compare the two formulations for these outcomes.)

Keeping perspective on practical implications

These results suggest that simply by changing the specific lithium salt used for supplementation, we can significantly increase lithium’s efficacy as a treatment for AD. In light of these data—and given the very low cost of lithium salts—early-stage clinical trials with LiO are likely forthcoming. But in the meantime, we’d be remiss not to point out a few important limitations with regard to the practical applicability of these findings.

First, none of the experiments in this study (or in any clinical trials to date) were designed to assess whether lithium is effective in preventing AD or MCI. While this study did not conduct pre-treatment assessments, we can safely assume that AD pathology would have existed even at baseline, as earlier research has shown that the transgenic mice used in this study start developing AD neuropathology and cognitive deficits by adolescence—younger than the age at which lithium treatment was initiated in these experiments.^15,16 Thus, nothing in this study indicates that taking lithium pre-emptively—i.e., outside of the context of existing neurodegenerative disease—would provide any benefit for cognitive trajectories. (Although such a benefit is theoretically possible, we have no evidence of it at present.)

Indeed, even if we did have evidence for an effect on AD prevention, this wouldn’t mean that everyone should run out to the drug store and start taking lithium orotate supplements. Lithium—regardless of its salt formulation—is acutely toxic at high doses and can cause chronic toxicity, particularly to the kidneys, even at doses within typical therapeutic ranges. No form of lithium should ever be taken without oversight by a medical professional, and with respect to LiO in particular, we have very little safety data at present, and what little we have comes from studies in rodents, not humans. This also means that we have no insight on which patients (if any) are most likely to benefit, which side effects might exist, or what protocols to use with regard to reasonable dosages, frequency, or duration of treatment. In other words, it would be very premature to add lithium orotate to any plan for AD prevention or treatment in humans.

A cause for [cautious] optimism

Few diseases have defied searches for effective treatments to quite the same extent as Alzheimer’s disease, and it’s far too early to crown lithium orotate as the answer to our prayers. Many stumbling blocks exist on the road from animals to humans, particularly where AD is concerned, as AD mouse models—which involve genetic manipulations to produce pathological features that don’t occur naturally in mice—don’t precisely replicate the pathogenesis and progression of spontaneous AD in humans. Even if LiO does prove to be an effective human treatment, most experts agree that treating (let alone curing) this complex disease will likely require more than any single therapy can achieve on its own. Rather, it may demand a multipronged approach aimed at various clinical features and pathogenic mechanisms.

But in light of the long history of data suggesting lithium as a possible AD treatment, paired with the apparent improvements to this therapeutic strategy that Aron et al. demonstrated with lithium orotate, we have plenty of cause for cautious optimism. Indeed, the fact that many of the present study’s results had already been shown shouldn’t dampen our hopes over these findings—it should strengthen them. Lithium has long been on the radar as a potential AD therapy, but virtually every game-changing advance in medicine could never have been possible without many incremental steps, a number of false leads, and a few key “aha” moments along the way. Whether this present work will prove to be incremental, misguided, or “aha” remains to be proven more definitively, but the data from Aron et al. provide a very compelling rationale for further investigation of LiO as an AD treatment—potentially turning a risky therapy with ho-hum benefits into a highly efficacious therapy with a lower risk profile. I hope that researchers will answer the call for clinical trials in the near future—because in the case of AD, even moderately effective treatments would be a welcome breakthrough, “silver bullet” or not.

For a list of all previous weekly emails, click here. 

podcast | website | ama

References

1. Aron L, Ngian ZK, Qiu C, et al. Lithium deficiency and the onset of Alzheimer’s disease. Nature. Published online August 6, 2025:1-10. doi:10.1038/s41586-025-09335-x
2. Goodman B. Scientists say it may be possible to protect aging brains from Alzheimer’s with an old remedy — lithium. CNN. Published online August 7, 2025. Accessed September 9, 2025. https://www.cnn.com/2025/08/06/health/lithium-alzheimers-brain-study
3. Manji HK, Moore GJ, Chen G. Lithium at 50: have the neuroprotective effects of this unique cation been overlooked? Biol Psychiatry. 1999;46(7):929-940. doi:10.1016/s0006-3223(99)00165-1
4. Alvarez G, Muñoz-Montaño JR, Satrústegui J, Avila J, Bogónez E, Díaz-Nido J. Regulation of tau phosphorylation and protection against beta-amyloid-induced neurodegeneration by lithium. Possible implications for Alzheimer’s disease. Bipolar Disord. 2002;4(3):153-165. doi:10.1034/j.1399-5618.2002.01150.x
5. Phiel CJ, Wilson CA, Lee VMY, Klein PS. GSK-3alpha regulates production of Alzheimer’s disease amyloid-beta peptides. Nature. 2003;423(6938):435-439. doi:10.1038/nature01640
6. Nocjar C, Hammonds MD, Shim SS. Chronic lithium treatment magnifies learning in rats. Neuroscience. 2007;150(4):774-788. doi:10.1016/j.neuroscience.2007.09.063
7. Tsaltas E, Kontis D, Boulougouris V, et al. Enhancing effects of chronic lithium on memory in the rat. Behav Brain Res. 2007;177(1):51-60. doi:10.1016/j.bbr.2006.11.003
8. Nunes MA, Schöwe NM, Monteiro-Silva KC, et al. Chronic microdose lithium treatment prevented memory loss and neurohistopathological changes in a transgenic mouse model of Alzheimer’s disease. PLoS One. 2015;10(11):e0142267. doi:10.1371/journal.pone.0142267
9. Fiorentini A, Rosi MC, Grossi C, Luccarini I, Casamenti F. Lithium improves hippocampal neurogenesis, neuropathology and cognitive functions in APP mutant mice. PLoS One. 2010;5(12):e14382. doi:10.1371/journal.pone.0014382
10. Zhu ZF, Wang QG, Han BJ, William CP. Neuroprotective effect and cognitive outcome of chronic lithium on traumatic brain injury in mice. Brain Res Bull. 2010;83(5):272-277. doi:10.1016/j.brainresbull.2010.07.008
11. Forlenza OV, Diniz BS, Radanovic M, Santos FS, Talib LL, Gattaz WF. Disease-modifying properties of long-term lithium treatment for amnestic mild cognitive impairment: randomised controlled trial. Br J Psychiatry. 2011;198(5):351-356. doi:10.1192/bjp.bp.110.080044
12. Andrade Nunes M, Araujo Viel T, Sousa Buck H. Microdose lithium treatment stabilized cognitive impairment in patients with Alzheimer’s disease. Curr Alzheimer Res. 2013;10(1):104-107. doi:10.2174/156720513804871354
13. Devanand DP, Crocco E, Forester BP, et al. Low dose lithium treatment of behavioral complications in Alzheimer’s disease: Lit-AD randomized clinical trial. Am J Geriatr Psychiatry. 2022;30(1):32-42. doi:10.1016/j.jagp.2021.04.014
14. Gildengers AG, Ibrahim TS, Zeng X, et al. The LATTICE Study: Design of a pilot feasibility randomized controlled trial of lithium to delay cognitive decline in mild cognitive impairment. Alzheimers Dement (N Y). 2025;11(2):e70112. doi:10.1002/trc2.70112
15. Belfiore R, Rodin A, Ferreira E, et al. Temporal and regional progression of Alzheimer’s disease-like pathology in 3xTg-AD mice. Aging Cell. 2019;18(1):e12873. doi:10.1111/acel.12873
16. De Plano LM, Saitta A, Oddo S, Caccamo A. Navigating Alzheimer’s disease mouse models: Age-related pathology and cognitive deficits. Biomolecules. 2024;14(11). doi:10.3390/biom14111405