Exercise or active rehabilitation is a non-pharmacological approach increasingly used for people with Multiple Sclerosis (MS), in support of disease-modifying therapies (DMTs), with the aim of improving the quality of life and engagement in daily activities. Exercise improves several disease outcomes, like cardiovascular and neuromuscular functions and walking abilities. However, its disease modifying potential is poorly explored. Exercise might target two relevant disease hallmarks that are interconnected, such as the dysregulated immune system and the inflammatory synaptopathy. Exercise might act through the activation of the autonomic part of the vagus nerve, which is an important modulator of both the innate and adaptive immune system, through the so-called cholinergic anti-inflammatory pathway-CAP.

This study aims to address the effect of exercise in reducing peripheral inflammation that drives the synaptic pathology and neurodegeneration occurring in the brain of MS patients. Patients will undergo a therapeutic exercise program, consisting of 3 hours of treatment per day, 6 days/week for a total of 6 weeks. The treatment will include both passive and active therapeutic exercises targeted to restore or preserve muscular flexibility, motor coordination and ambulatory function. The day of recruitment (time 0) patients will undergo neurological and mood examination and blood withdrawal to analyze peripheral markers of immune function. Moreover, transcranial magnetic stimulation (TMS) will be used to measure synaptic transmission, while the heart rate variability (HRV) test will be performed to explore vagal function. The effect of exercise will be evaluated at the end of rehabilitation (after 6 weeks-time 1), on the above parameters. A follow up will be included (time 2, 8 weeks after the end of the treatment) to address long-term effects on neurologic and mood measurements as well as peripheral marker levels.

Clinical manifestations of Multiple Sclerosis (MS) indicate the involvement of motor, sensory, visual systems, cognition and emotion, as well as peripheral autonomic system (PAS). Disease modifying therapies (DMTs) are immunomodulatory drugs designed to dampen the immune reaction occurring in MS. Indeed, MS pathogenesis is supposed to rely on the break of immunological tolerance against myelin epitopes, which trigger an inflammatory cascade that leads to chronic inflammation, axonal loss and neurodegeneration. T cell population in MS presents several metabolic dysfunctions, such as glycolysis alterations that can be attenuated by DMTs. Studies of synaptic transmission conducted on both MS patients, via transcranial magnetic stimulation (TMS), and EAE mice, via electrophysiological recordings of single neurons, showed an early synaptopathy characterized by an impairment of glutamatergic and GABAergic transmissions. Such synaptopathy is independent of demyelination and caused by inflammation. Importantly, TMS cortical excitability measures positively correlate with disability in MS patients. Moreover, chimeric experiments obtained incubating MS T cells and murine brain slices, clearly indicate that T cells drive synaptic damage during MS, suggesting that interfering with T cell-neuron crosstalk could be a possible therapeutic target.

Due to the complexity and the heterogeneity of the disease course and the clinical symptoms, the search for the appropriate personalized treatment and the disease management remains a challenging issue. It is increasingly recognized that a multi-disciplinary approach in MS treatment, including non-pharmacological interventions is required to treat MS. Active-rehabilitation or exercise has been proven effective in the improvement of cardiovascular functions, aerobic capacity, muscular strength and ambulatory performance, while some data indicate that other outcomes, like balance and depression can be positively influenced by exercise. Symptoms of sympathovagal imbalance, like altered heart rate variability (HRV), previously shown to depend on inflammatory bulk in MS, may be positively modulated by exercise, which is known to regulate both the peripheral nervous system and the immune system. However, the mechanisms involved in exercise-beneficial effects as well as the impact of exercise on MS pathophysiological hallmarks, especially those regarding the immune-synaptic axis, are still poorly elucidated.

This longitudinal, interventional, non-pharmacological study is designed to enrol 44 MS patients and 30 healthy controls matched by gender and age to the MS group. The MS patient group will undergo a conventional 6-week rehabilitation program. Physical therapy will be performed for 6 days/week for 6 weeks and will consist of 3 hours of treatment per day. The rehabilitation program will be planned by a physician specialized in physical and rehabilitation medicine and will consist of both passive and active therapeutic exercises specifically aimed at restoring or maintaining muscular flexibility, range of motion, balance, coordination of movements, postural passages and transfers, and ambulation. According to the patient's disability status, different therapeutic exercises will be performed by qualified physiotherapists. Intensity of exercise will be tailored to the level of patient's disability. Furthermore, advanced robotic therapy such as Lokomat® exoskeleton (Hocoma AG, Volketswil, Switzerland), Biodex® Stability System (BSS, Biodex, Inc, Shirley, NY), G-EO System™ (Reha Technology AG, Olten, Svizzera) and Indego® Therapy (Parker USA), will be used to standardize rehabilitation treatment and obtain more objective indices of motor function and will be applied according to clinical indications. Three time-points (t) of evaluations are included in the study: t0 (before starting the rehabilitation period), t1 (soon after rehabilitation) and t2 (follow-up, after 8 weeks by the end of rehabilitation). Therapeutic efficacy will be evaluated at the end of the exercise program (t1) by repeating evaluations performed at t0, which include neurological and psychological assessments, together with measures of brain synaptic activity and vagal function and immune function. At t2, analysis will be limited to neurological and psychological assessments and immune function. Thus, blood samples will be collected at t0, t1 and t2 to study changes in immune function that might correlate with clinical parameters described as primary and secondary outcomes at the different time-points.

Statistical analysis will be performed by IBM SPSS Statistics 15.0. Data will be tested for normality distribution through the Kolmogorov-Smirnov test. Differences between pre- and post-values will be analyzed using parametric Student's t-test for matched pairs, or if necessary, nonparametric Wilcoxon signed-rank test for matched pairs. Changes in categorical variables will be assessed by McNemar test. Correlation analysis will be performed by calculating Pearson or Spearman coefficients as appropriate. Changes in categorical variables will be evaluated by the test McNemar. Data will be presented as the mean (standard deviation, sd) or median (25th- 75th percentile). The significance level is established at p<0.05.

Sample size calculation was performed according to the following criteria. Assuming that in MS patients the cytokine values in particular the TNF level after exercise therapy decrease in a manner similar to that showed in the study by Hedegaard et al (2008). Based on these results, calculating an average difference between pre and post exercise values of TNF equal to 1365.1 pg / ml (sd = 2570), d = 0.53, in order to appreciate a moderate effect with a statistical capacity of 95% and assuming a two-tailed a = 0.05 and applying a Wilcoxon rank test for paired values, the investigators estimate a total number of patients equal to 40. Analysis was performed with the G * POWER v3.1.9.2 program. Considering possible drop-outs, the investigators estimate to increase the number of patients recruited by one percentage equal to 10%, meaning 4 subjects. Moreover, using Power Analysis d=0.61, it has been calculated that the number of healthy volunteer subjects needed to be recruited for the study of the immunophenotype and secretoma will be 30 subjects per experimental group, in order to be able to refuse the null hypothesis that the two groups are equal with a test power of 95% and appreciate a difference of 1600.9 pg / ml between the means of the experimental groups (healthy control vs MS) (standard deviation equal to 2599), d = 0.61. The probability of Type I error associated with this test for this hypothesis is 5%.

Inclusion Criteria:

• Ability to provide written informed consent to the study;
• Diagnosis of MS definite according to 2010 revised McDonald's criteria (Polman et al., 2011);
• Age range 18-65 (included);
• EDSS range between 4,5 and 6,5 (included);
• Ability to participate to the study protocol.

Exclusion Criteria:

• Inability to provide written informed consent to the study;
• Altered blood count;
• Female with a positive pregnancy test at baseline or having active pregnancy plans in the following months after the beginning of the protocol;
• Contraindications to gadolinium (MRI);
• Contraindications to TMS;
• Patients with comorbidities for a neurological disease other than MS, included other neurodegenerative chronic diseases or chronic infections (i.e tuberculosis, infectious hepatitis, HIV/AIDS);
• Unstable medical condition or infections;
• Use of medications with increased risk of seizures (i.e. Fampridine, 4- Aminopyridine);
• Concomitant use of drugs that may alter synaptic transmission and plasticity (cannabinoids, L-dopa, antiepileptics, nicotine, baclofen, SSRI, botulinum toxin).