Multiple sclerosis (MS) is a chronic inflammatory disease causing widespread degeneration of the central nervous system. The disease, with different features and progression according to the clinical phenotype [
1], gradually results in severe neurological deficits [
2] with complex, variable and unpredictable patterns of symptoms [
2] including different motor deficits [
1]. Locomotor disability and balance disorders affect approximately 75% of persons with MS, with altered coordination of posture and gait [
3,
4], mobility problems [
5,
6], reduced walking competency [
3] and increased risk of falling [
7]. In progressive MS, the high prevalence of motor disorders and gait disabilities, the negative impact on personal activities and quality of life (QoL), and the limited effects of specific medications [
8] make gait rehabilitation a crucial part of the management. The aim is to increase patients’ levels of activity and independence [
9] and their QoL, even independent of symptom regression [
10,
11]. Gait disabilities showed improvement following physical therapy [
4,
12‐
14] and low-to-moderate-intensity aerobic over-ground or treadmill training, which represents a useful option for rehabilitation, also in combination with body weight support [
15‐
17]. To this end, a robot-driven gait orthosis was recently developed, studied and considered a feasible and effective therapeutic option in MS subjects with severe walking disabilities. Robot-assisted gait training (RAGT) allows a more effective support of walking movements and imitation of a nearly normal gait pattern during treadmill training at a higher speed, with improvements in walking distance, velocity and knee extensor strength compared to conventional therapy [
18,
19]. Several studies have tested in samples of MS patients the effects of interventions, such as treadmill training [
4], bodyweight-supported training on a treadmill [
20,
21], RAGT [
18,
19,
22‐
26], or both treatments combined within a single session [
27], reporting small but positive effects on functional status [
4,
18‐
20,
23‐
26] or QoL [
10,
21]. Recently, improvements were reported in the 6-min Walking Test and in the balance domain after RAGT [
25] but not in gait speed measured by the 10-m Walk Test [
26]. Unfortunately these studies, using different devices and training protocols (12 to 15 sessions over 3–6 weeks), including heterogeneous MS subgroups with a limited number of subjects and a wide range of gait disabilities (Expanded Disability Status Scale, EDSS 3–7.5), failed to offer an exhaustive evidence of the superiority of RAGT over other specific gait trainings, so larger trials are necessary [
28]. Otherwise, this might be partially explained by the fact that the rehabilitation process in MS subjects is complex and person-specific [
28] as the response to treatments regarding neuronal plasticity is highly individual. Functional recovery in MS is achieved by repair of damage through remyelination, with resolution of inflammation and functional reorganization. Evidence from brain systems supports an adaptive role for neuroplastic changes in MS despite its widespread pathology. Specifically, it may limit the negative effects of MS on behavior [
29‐
32] and differs between patients and various disease types, with lower response according to patient age and disease duration [
33,
34]. Moreover, different rehabilitation treatments might switch on different adaptive response. High-intensity interventions might be more effective on neural reorganization and motor recovery involving synaptic transmission and formation of novel synapses, cortical reorganization and induction of neurogenesis limited to the site of injury or involving distant healthy brain regions [
35]. The effects of neuroplasticity-based technologies and interventions, virtually beneficial for functional recovery, have been poorly tested so far. Different tools, such as positron emission tomography and functional magnetic resonance imaging, could be appropriate to evaluate such recovery-related processes. Several studies have employed these techniques, revealing that in MS patients a decreased hemispheric lateralization [
32] and an increasingly bilateral activation, even for simple motor tasks involving higher-control sensorimotor areas, were observed [
36,
37]. Other noninvasive, reliable and less expensive measurements, such as transcranial magnetic stimulation and near-infrared spectroscopy (NIRS), could also be useful. In a NIRS-based study the coherence, considered a potentially useful marker in disorders with white matter damage or axonal loss, was found to be similar in MS subjects and controls in the resting phase, but significantly decreased during motor tasks [
38]. Moreover, relevant information to identify patterns of recovery in MS patients could be added by the measurement of molecular regulators of neuronal or vascular plasticity. These biomarkers derived from blood tests include circulating cell subsets and soluble factors measurable in plasma. Previous studies involving MS subjects showed that the
N-acetylaspartate concentration correlated with an increased lateralization index; neurotrophins that regulate neural plasticity, such as brain-derived neurotrophic factor (BDNF) [
39], were found in lower concentrations compared to healthy subjects [
40]. Furthermore, inflammatory cytokines, such as interleukin-1β, negatively interact with BDNF and amyloid-β has been observed in multifocal MS lesions [
41,
42]. Finally, growth factors participate in neural cell survival and tissue repair processes [
43‐
45] and, in particular, platelet-derived growth factor decreases with disease duration, being low in primary progressive MS patients [
46] while incomplete glucose oxidation by glycolysis and mitochondria results in increased oxidative stress that promotes lesion progression rather to repair [
47‐
49]. Also, disturbances of the hemostatic mechanisms, which are closely and reciprocally related to inflammation, are relevant for neurological disorders, in particular procoagulant factors and receptors, as well as main anticoagulant proteins endowed with anti-inflammatory activities, that exert cytoprotective effects and favor endothelial barrier stabilization, neurogenesis and angiogenesis [
50]. Lastly, besides soluble factors, growing evidence advances the concept that stem cells can modulate nervous system action as well as the dysregulation of inflammatory responses and immune self-tolerance has to be considered a key element in the autoreactive immune response in MS. To this end, circulating stem/progenitor cells capable of homing in on neovascularization sites [
51], and regulatory T-cells (Treg) have emerged as crucial players in the pathogenic scenario of autoimmune inflammation with a role in their modulation by pharmacological and rehabilitation therapy.