Is It Time to Repurpose Calcineurin Inhibitors for the Treatment of Cognitive Impairment and Dementia?

Calcineurin (CN) is a Ca2+ /calmodulin-dependent protein phosphatase best known for its critical role in regulating T-cell activation and the adaptive immune response. Indeed, CN is the primary target of the immunosuppressants, tacrolimus and cyclosporine, which are staple drugs for the prevention of solid organ transplant rejection. In T-cells, CN regulates immune signaling through activation of nuclear factor of activated T cells (NFAT) transcription factors and the production of key cytokine species. However, the biologic and therapeutic significance of CN isn’t limited to T-cells and immune/inflammatory signaling. CN is ubiquitously expressed across mammalian tissues, and is abundant in brain, especially in neurons. Though less abundant in glia, CN signaling becomes more prominent in astrocytes and microglia as these cells transition to reactive states. In neurons, CN shapes the structural and functional properties of synapses through actions on cytoskeletal elements, membrane receptors/channels, and membrane trafficking machinery [1]. In glia, CN regulates cytokine signaling and glutamate uptake through activation of NFATs and other transcription factors [2].

Due to exquisite sensitivity to fluctuating Ca2+ levels, CN is highly vulnerable to the Ca2+ dysregulation that arises in both neurons and glia during aging, injury, and/or disease [3–6]. Consequently, CN signaling is elevated at early stages of cognitive decline in humans [7, 8] and correlates with the accumulation of key AD biomarkers [7, 10]. Suppression of CN signaling using drugs or transgenic approaches ameliorates several key phenotypes associated with AD including neuronal degeneration [11], synapse loss/dysfunction [10, 12–15], neuroinflammation and excitotoxicity [14, 17], and impaired neurovascular coupling [18]. CN/NFAT inhibitors also extend lifespan and/or improve survival [15, 20]. Most of this work has been performed in rodents, but similar results have been found in higher mammalian models (i.e., aging canines), as well [21, 22]. Conversely, disease-like phenotypes commonly emerge in otherwise healthy brain tissues (e.g. synaptic depression, dendritic atrophy, glial reactivity/neuroinflammation, cognitive loss) when CN is forcibly hyperactivated in neurons [23, 24] and astrocytes [25–27] using transgenic approaches. Together, these observations suggest that aberrant CN signaling is an important pathophysiologic mechanism and promising drug target for the treatment of cognitive loss associated with AD, and possibly other neurodegenerative conditions.

Giulio Taglialatela and colleagues have been leaders in this research area and provided much of the initial evidence that tacrolimus improves neurologic function in amyloid mouse models [3, 29]. In a 2015 study [30], Taglialatela et al. asked the next logical question: do people who take FDA-approved calcineurin inhibitors (CNIs) show a lower incidence of dementia than individuals who do not use CNIs? To address this question, they turned to the medical records of more than 2,600 kidney transplant patients at the University of Texas Medical Branch who used tacrolimus daily to prevent organ transplant rejection. The major finding from this study was that kidney transplant patients, administered daily tacrolimus, were significantly less likely to present with dementia symptoms at follow-up physician visits, compared to age-matched cohorts in the general population who were not on tacrolimus therapy. Perhaps the most intriguing aspect of these observations is that the prevalence of dementia was reduced in tacrolimus cohorts across all age ranges, even though many of the factors that lead to kidney failure (and transplant) are also major risk factors for dementia and AD.

In the present study from Taglialatela and colleagues [31], the authors extend their earlier epidemiologic work in several important ways including: 1) A much larger and diverse sampling of tacrolimus-treated individuals across the entire United States (tens of thousands of individuals) versus a control cohort (i.e., not on CNI therapy); 2) Propensity score-matching to control for the impact of aging, sex, race, ethnicity, and multiple disease risk factors like hypertension, diabetes, depression, cerebrovascular pathologies, and traumatic brain injuries; and 3) comparisons to additional cohorts of people taking other common immunosuppressant drugs (i.e., cyclosporine and sirolimus). The evaluation of multiple immunosuppressants is important because of key similarities and differences in mechanism of action and brain penetrance. Tacrolimus and cyclosporine are similar in that both inhibit CN. However, tacrolimus inhibits CN through interactions with FK-506 binding proteins (FKBPs), while cyclosporine must bind to cyclophilins en route to CN inhibition. Perhaps more importantly, tacrolimus crosses the blood-brain barrier and accumulates in brain, while cyclosporine exhibits poor brain penetrance [32, 33]. This difference means that both drugs will inhibit peripheral CN activity (to cause immunosuppression), but only tacrolimus will directly affect brain CN signaling. Sirolimus is structurally similar to tacrolimus and binds to the same FKBP species, but the sirolimus/FKBP complex does not inhibit CN. Instead, sirolimus/FKBP inhibits the mammalian target of rapamycin (mToR), which has also been implicated in aging and aging-related diseases [34].

In support of the 2015 study, the results of Silva et al. 2023 [31] show a significantly reduced incidence of dementia in all three immunosuppressant cohorts, relative to control cohorts. Moreover, the tacrolimus cohort showed significantly reduced dementia prevalence relative to the cyclosporine cohort. Tacrolimus also showed a consistent trend for reduced dementia prevalence, relative to the sirolimus cohort, but these differences did not reach statistical significance. It’s worth noting that the relatively small size of the sirolimus cohort ( 3,000) resulted in statistically underpowered comparisons. Based on post hoc power analyses, the authors point out that a significant difference between the sirolimus and tacrolimus cohorts would have been detected if the n/cohort was increased to  34,000 people (more similar to the ns used for the tacrolimus versus control cohort comparisons).

Together, these results—as well as abundant findings from the animal literature—suggest that commercially available immunosuppressants could be repurposed for the treatment of dementia and neurodegenerative disorders. While peripheral immunosuppression alone may provide significant protection (as shown by the cyclosporine cohort), inhibition of central CN signaling (and possibly FKBP/mToR signaling) with tacrolimus seems to offer the best alternative. Indeed, peripheral actions of CNIs are more likely to be associated with adverse effects like nephrotoxicity, which is usually a limiting factor in treating organ transplant patients [35]. However, it is important to recognize that these adverse effects are clearly dose-dependent and that many individuals have safely used tacrolimus at low maintenance doses for decades without significant nephrotoxicity. Moreover, a recent longitudinal study on aging beagles from our group has shown that twice daily doses of tacrolimus, at sub-immunosuppression levels, causes no nephrotoxicity while significantly improving brain microstructural measures and preserving cognitive function over one-to-three years of treatment [21, 22]. These observations suggest that it is time to evaluate the anti-dementia effects of CN inhibitors and similar compounds in a rigorously controlled human trial. At the same time, mechanistic studies on the contributions of CN signaling to distinct and overlapping disease phenotypes in neurons, astrocytes, microglia, and pericytes will be useful for identifying additional CN-related targets (e.g., NFATs), which could, in turn, lead to the development of even more specific and possibly safer CN-inhibiting strategies.