Anchoring neurostimulation on crossed cerebellar diaschisis for motor recovery in adults with hemiparesis

Given the unmet medical needs for stroke rehabilitation, neurotechnologies with innovative rationales and good designs hold promise for restoring motor function in patients worldwide. These features are of unique importance in developing a more physiologically based, individualized, precise therapy to improve motor function prognoses, even after a chronic stroke. Among the emerging neurotechnologies, deep brain stimulation (DBS) enables precise modulation of specific neural circuits to enhance motor recovery for neurological disorders such as stroke.¹ In a paper recently published in Nature Medicine, Baker et al. proposed a masterful DBS approach based on the therapeutic proposition of alleviating crossed cerebellar diaschisis (CCD) to address upper-extremity hemiparesis since ascending input from the dentato-thalamo-cortical (DTC) pathway can activate the ipsilesional motor cortex and beyond, including prefrontal and parietal areas. In this first-in-human study, the authors highlighted the potential of combining DBS electrodes inserted into the contralateral dentate nucleus (DN-DBS) with rehabilitation therapy as a novel approach with clinical significance for adults with hemiparesis 1–3 years after a middle cerebral artery infarction.²

The DN-DBS protocol is grounded on elegant anatomical and neurophysiological knowledge, which provides the foundation for applying DBS to the contralateral dentate nucleus. The DTC pathway comprises the dominant ascending fibers projecting from the cerebrum into the ipsilesional motor, prefrontal, and parietal regions. Excitatory input to the cerebellar hemisphere can be reduced after a middle cerebral artery ischemia due to disruption of the corticopontocerebellar pathway. Consequently, the decreased activation of the dentate nucleus lowers its output to the ipsilesional motor-related cortices, which was shown to be associated with reduced motor performance in patients after a stroke. Therefore, neuromodulation of the dentate nucleus may enhance cortical excitability to promote motor recovery in these patients. As reported in this study, the trial intervention was feasible and well tolerated, although adverse events occurred in all patients, and the recruitment rate was relatively low.

Driven by the CCD hypothesis, the scientific rationale of this neurostimulation configuration could benefit from reporting the extent of diaschisis within the pathway.³ Consequently, inspecting the associations between cortico-cerebellar connectivity and the participants' preservation of gross motor impairment and distal dexterity would be more convincing. Notably, the latter is a crucial determinant in assessing intervention response, as the post-hoc subgroup analysis indicates. Nonetheless, incorporating structural, functional, or neuroelectrophysiological integrity measures into the scheduled visits would be highly beneficial to substantiate the intended clinical rationale of DN-DBS anchored on the CCD hypothesis. The metabolic changes characterized by positron emission tomography-computed tomography offer exploratory insights into the activation profiles of the ipsilesional motor-associated cortical regions after DBS. Therefore, the authors suggested that these effects support the advantage of DTC fibers in transducing DBS to the perilesional cortical areas compared to epidural cortical stimulation. However, without evidence demonstrating CCD recovery, the exact mechanism by which DN-DBS works remains unknown. With rapid technological advancements, it may raise concerns about why non-invasive neurostimulation therapies or brain-computer interfaces targeting cerebral networks were not considered instead of painstakingly invasive intervention, which could be paradoxical.

Robust clinical data are essential to translate the DN-DBS concept into routine practice, and Baker et al. provide a first step. However, one may further argue the clinical implications of a joint neurostimulation and rehabilitation protocol. As noted in this paper, the specific effects of DN-DBS can hardly be distinguished from those of rehabilitation sessions because the trial design does not directly compare the two synergistic components. It adds to the existing concerns that conventional training can yield similar results in the chronic stroke stages.⁴ To partly mitigate this issue, the authors have attempted to account for potential confounders related to chronic physical deconditioning by conducting a “rehab-only” phase. Unfortunately, rehabilitation may have contributed more to the participants' motor improvement. Dividing the Fugl–Meyer Upper Extremity score change by time indicates that the participants were less responsive to the “DBS + rehab” phase. Therefore, further investigation and iterative optimization are still warranted because advanced neurotechnology should be built on solid clinical evidence of CCD recovery.

In summary, our discussion clarifies the importance of targeting neurostimulation of the DTC pathway to activate motor-related brain cortices using a multidisciplinary neuroengineering approach. We believe the above concerns do not entirely obscure the merits of this first-in-human trial since the novel DBS paradigm appears to feasibly and safely broaden the therapeutic window for patients after a chronic stroke. Baker et al. have been working to discover a candidate DBS neural pathway from basic research, precisely anchor neurostimulation on the contralateral dentate nucleus, and design a delicate trial to investigate its clinical importance. The translational research paradigm indicated by their extensive efforts can undoubtedly facilitate further advancement of relevant, innovative therapeutics in this field by translating them from bench to bedside.⁵

Ze-Jian Chen: Conceptualization, manuscript drafting and manuscript editing. Ming-Hui Gu: Manuscript editing and reviewing. Yong Chen, Xiao-Lin Huang: Manuscript drafting, manuscript reviewing, supervision and funding acquisition.

The authors declare no competing interests.

The ethics approval was not needed in this study.