Lithium Shows Promise as a Disease-Modifying Therapy for Parkinson's Disease

Parkinson's disease (PD) affects over 6 million people worldwide, with projections reaching 12 million by 2050. Despite decades of research, no approved therapy halts the underlying neurodegeneration—dopaminergic neuron loss in the substantia nigra and α-synuclein aggregate formation. This reality makes the search for disease-modifying treatments urgent. Lithium, a mood stabilizer approved for bipolar disorder since 1970, has emerged as a promising repurposing candidate due to its pleiotropic neuroprotective mechanisms that simultaneously target multiple facets of PD pathophysiology.

The review details six key mechanisms through which lithium may modify PD progression. First, lithium directly inhibits GSK-3β by competing with magnesium at ATP-binding sites, reducing pathological hyperphosphorylation of tau and α-synuclein and slowing fibrillization. Second, by inhibiting inositol monophosphatase, lithium triggers mTOR-independent autophagy, promoting lysosomal clearance of misfolded proteins and damaged mitochondria. Third, lithium attenuates microglial activation and suppresses pro-inflammatory cytokines (TNF-α, IL-1β, IL-6) via NF-κB and GSK-3β–dependent pathways, reducing neuroinflammatory damage. Fourth, lithium activates the NRF2–ARE antioxidant pathway, upregulating heme oxygenase-1 and superoxide dismutase to counter ROS generated from mitochondrial complex I impairment. Fifth, lithium blunts excitotoxic glutamatergic signaling by reducing NMDA receptor–mediated calcium entry and enhancing glutamate uptake. Sixth, in chronic MPTP mouse models, lithium increases methylation of the SNCA promoter, epigenetically suppressing α-synuclein gene expression and correlating with reduced dopaminergic neuron loss.

Preclinical evidence is extensive but heterogeneous. Dosing regimens vary from acute high-dose exposures (2–4 mEq/kg) to chronic low-dose regimens approximating human serum levels of 0.2–0.6 mmol/L. Neuroprotection is generally stronger when lithium is administered before toxin exposure rather than after symptom onset, raising questions about whether the drug's utility may be primarily preventive. Notably, at least one study (Yong et al.) found no dopaminergic neuroprotection in the 6-OHDA rat model despite confirmed GSK-3β inhibition, underscoring the inconsistency across models and the need for standardized protocols.

On the clinical side, a 2025 biomarker study found that PD patients achieving serum lithium levels of 0.21–0.56 mmol/L showed a median 12.8% reduction in serum neurofilament light chain (NfL), a marker of axonal degeneration (p = 0.0001). Low-dose lithium also reduced L-Dopa–induced dyskinesias by 40–50% in MPTP-lesioned mice via calpain-1 inhibition. These findings prompted the International Linked Clinical Trials programme to initiate a Phase 1b trial (NCT06099886) evaluating lithium aspartate in 35 early-stage PD patients. Alternative formulations—lithium orotate and lithium aspartate—are being explored to enhance CNS delivery at lower systemic exposures.

Key caveats remain. Higher-dose lithium (0.8–1.2 mmol/L) carries risks of cerebellar neurotoxicity and drug-induced parkinsonism, and a retrospective cohort study of 1,749 elderly lithium users found an 87% increased incidence of dopaminergic drug prescriptions (HR 1.87), suggesting potential misdiagnosis of lithium-induced tremor as PD. NfL has not been validated as a regulatory endpoint for PD. Dropout rates of 33% in medium-dose arms of trials threaten generalizability. Robust phase 3 data, long-term safety assessments, standardized dosing protocols, and identification of patient subgroups most likely to benefit are all critically needed before lithium can be recommended as a disease-modifying PD therapy.