Neuromodulation to Promote Upper Limb Recovery After Stroke: Clinical and Preclinical Insights Into Cerebellar Transcranial Direct Current Stimulation Combined With Robotic or Motor Training

Stroke is one of the leading causes of disability worldwide, and upper limb impairment is among its most disabling consequences, significantly affecting quality of life. Robotic-assisted therapy has demonstrated effectiveness in improving motor function even in the chronic phase, while non-invasive brain stimulation (NIBS), particularly transcranial direct current stimulation (tDCS), is used to modulate neuronal excitability and promote recovery. Although stimulation of the motor cortex is the most studied approach, not all patients respond effectively. Cerebellar stimulation has been proposed as an alternative target due to its connections with the motor cortex through the dentato-thalamo-cortical pathway and its role in motor learning. This double-blind randomized controlled pilot study aimed to evaluate the combined effects of cerebellar tDCS and robotic training on upper limb functional recovery and cognitive performance in patients with chronic supratentorial stroke. Twenty-seven patients were enrolled and randomly assigned to two groups: the experimental group (n = 14) received cerebellar tDCS (20 min/day, 2 mA, cathodal, 5 sessions/week for 2 weeks) combined with robotic upper limb rehabilitation using the MOTORE® device, while the control group (n=13) received sham stimulation with the same robotic training. Clinical assessments were performed at baseline (T0), after treatment (T1), and at one-month follow-up (T2), and included different functional and cognitive outcomes. Statistical analysis was conducted using a mixed-effects linear model. Primary outcomes were Fugl-Meyer Assessment-Upper Limb and Motricity Index; secondary outcomes included disability and spasticity‑related measures, cognitive assessments, as well as pre‑specified MOTORE® outputs including range of motion and strength. Significant improvements over time were observed in the primary outcomes (p < 0.05) regardless of group allocation, indicating a robust time effect associated with the rehabilitation program. No consistent additional benefit of cerebellar tDCS over sham stimulation was identified across any outcome measure, with isolated statistically significant effects showing no coherent stimulation specific pattern or consistent directionality. In parallel, two pilot experiments were conducted on wild-type mice to investigate cerebellar activation. In the first experiment, mice were divided into three groups: sham stimulation during rotarod training, cerebellar tDCS during rest, and cerebellar tDCS during rotarod activity. In the second experiment, sedated mice received sham stimulation or cerebellar tDCS with varied parameters (standard, increased duration, or increased intensity). After stimulation, brains were processed for c-Fos immunofluorescence analysis as a marker of neuronal activity. Results indicated modest activation of the dentate nucleus in awake mice receiving cerebellar tDCS, regardless of motor activity, while no significant activation was observed in sedated mice. In conclusion, robotic‑assisted upper limb rehabilitation resulted in significant motor improvements in patients with chronic stroke. The addition of cerebellar tDCS did not provide a clear advantage over robotic training alone. Preclinical findings showed modest cerebellar activation dependent on behavioral state, emphasizing the need for further studies to clarify underlying mechanisms, optimize tDCS stimulation parameters, and identify individual and task‑ or context‑dependent factors influencing responsiveness to cerebellar neuromodulation.