The goal of this multicentre, randomized and controlled cross-over trial is to evaluate the efficacy of a programme of Inspiratory Muscle Training in subjects with Late On-set Pompe Disease (LOPD). The main question is to:

• verify changes in Forced Vital Capacity, Postural Drop, Maximal Inspiratory Pressure, Maximal Expiratory Pressure, Peak expiratory cough pressure, Maximal Inspiratory Capacity, six- minute walk test and or 6-minute pegboard ring test.-
• measure changes in some questionnaries investigating dispnoea and quality of life (Short-Form 36, Individualized-Neuromuscular-Quality-of-Life, Maugeri-Respiratory-Failure 28, Borg scale, Dispnoea 12, Mulditimensional Dispnea Profile, modified Medical Research Council, Fatigue Severity Scale, Epsworth Scale, Visual Analogue Scale).

Measurement will take place at baseline and after one, three, four, six and twelve months.

Participants will undergo a specific treatment consisting of aerobic exercise and Inspiratory Muscle Training with Powerbreathe device or Air-Stacking.

Researchers will study if Powerbreathe device is more effective than Air-stacking maneuvres

This is an experimental study (Multicentre, randomized and controlled cross-over trial) with low risk impact. The literature supports us in determining that the risks of respiratory muscle fatigue, which will be carefully evaluated through standardized measures of dyspnea, are nonexistent1Patients will undergo respiratory function tests with evaluation of the following parameters: forced vital capacity (FVC %) sitting on their back with measurement of postural fall ie the percentage difference between FVC in sitting and supine position (%PD), maximum inspiratory pressure (MIP), maximum expiratory pressure (MEP), peak expiratory flow during coughing (PCEF), maximum insufflation capacity (MIC), 6-minute walk test (6MWT) for walking patients and six-minute pegboard and ring test (6PBRT) "Ring testing" for non-dombulants. Also participants will compile some questionnaries investigating dispnea, sleep and quality of life (Short Form-36, Individualized-Neuromuscular-Quality-of-Life, Maugeri-Respiratory-Failure 28, Borg scale, Dispnea-12, Muldimensional Dispnea Profile, modifier Medical Research Council, Fatigue Severity Scale, Epsworth Sleepness Scale and Visual Analogue Scale). The evaluation phase identifies zero time (T0). Subsequently, patients enrolled will be randomized to the first treatment (training of inspiratory muscles) or the second treatment (training of inspiratory muscles + air-stacking). Each training period of the respiratory muscles will last 2 months and will be followed by a wash out period of 1 month before moving on to the next arm treatment. The first month patients will perform only aerobic activity. The total treatment time will be 6 months. walking patients will perform an aerobic activity of their choice, according to individual preferences, involving as many muscles as possible (e.g. walking, pedaling, climbing stairs) for at least 30 minutes 3 v/week of sufficient intensity to determine an increase in heart rate (FC) 60% of the predicted FC max or a perception of effort between 4-6 of the modified Borg scale.

For patients who are not walking, there are activities with arms not supported (initially without load and then with small weights of ½ Kg, 1 Kg, 2 Kg) involving the main muscle groups of the upper limbs performed for at least 30 minutes 3 v/week and of such intensity as to generate a perception of effort between 4-6 of the modified Borg scale. The type of instrument used for the training of inspiratory muscles will be the Powerbreathe KHP2 which is a tool with variable load (Fig 1) and can be set according to the characteristics of individual patients as explained below. Intensity and mode of training of inspiratory muscles (8-10 repetitions for 4 sets/day with breaks of at least 2 minutes between one series and the other for 6 days/week) Initial load 30% MIP (the load must be progressively increased to the maximum tolerated level) Patients will be instructed to perform a complete inhalation and exhalation at each respiratory act (about 70% of their CVF).

The respiratory effort perceived by patients should ideally be between 4-6 of the modified Borg scale. Increase of 2 cmH2O with Borg < 4 or average inhaled volume < 70% of CVF. Air "stacking" maneuver in the respiratory shaft through the use of an Ambu balloon: the patient encases and retains a series of volumes of air until reaching his MIC (Maximum Inspiratory Capacity) for 10 maneuvers/ day for 6 days/ week.

Patients will be evaluated at the reference hospital centre and will perform the programme at home. They will be followed by physiotherapists specialized in respiratory physiotherapy and the intervention will be articulated with home visits alternated with reinforcement calls and monitoring in equivalent number in the two treatment groups. Follow-up measures will be performed at 1 month (T1), 3 months (T3), 4 months (T4), 6 months (T6) and 12 months (T12).

Patient diaries and records of functional data and load increases will be evaluated. A subject who performs at least 2/3 sessions per week of aerobic training for at least 5/6 months will be defined as adhering to the treatment; 5/6 sessions per week of training of breathing muscles with at least 3/4 series completed, 5/6 sessions per week of air stacking with 8/10 managements carried out.

Also will be measured: Number of phone calls/video calls made by patients in addition to those scheduled; Number of patients in drop out; Number of accesses to the prescribing center for any clinical urgencies; Number of unscheduled visits; Questionnaire of satisfaction to the use of the device. Randomization will be carried out through a derived Internet randomization software in order to ensure the concealment of assignments.

Twenty participants will be randomized (1:1) in blocks of 2 to the trial treatment or control arm. The randomization model will be defined by the study statistician. Based on the defined parameters, the module creates an assignment table model, which is used to structure the randomization table. The module also monitors the overall progress of the assignment and the allocation of randomized participants. The doctor, to collect the data, will use as a password protected Microsoft Excel database. This will happen after the pre-test evaluation and before the first treatment session. The randomization field will always be locked and unmodifiable both before and after the randomization of a participant. The randomization documentation will be stored electronically in Excel and will include the participant identifier, the assignment to the treatment or control arm and the date and time of randomization. This electronic register of randomization will be accessible only to the doctor (not blind). Within each study arm, categorical data will be presented as frequencies and percentages; continuous data will be presented as mean, median, standard deviation and range. Primary analysis of the study will be conducted on intention-to-treat (ITT) cases. Non-directional hypotheses (two-tailed tests) will be tested. The measurement of the primary outcome defined by changes in vital capacity, drop posture and maximum inspiratory pressure will be evaluated with alpha=0.05. The results of the secondary outcomes will be evaluated with and without adjustment for multiple comparisons using the Bonferroni test. In addition, the analyses will be conducted on the participants who completed the study. All data analysis will be performed with the statistical software Jamovi version 2.3.3. No intermediate analysis.It is expected from this study that the training of inspiratory muscles can contribute to an improvement in respiratory function and that this results in a better quality of life

Inclusion Criteria:

• Late On-set Pompe Disease diagnosis already in Enzyme Replacement Therapy (ERT) since 12 months
• Forced Vital Capacity <80% predicted or Forced Vital Capacity >80% but with Postural Drop >25-30%

Exclusion Criteria:

• Severe Cardiologic or Respiratory comorbidities
• Complete dependence on ventilation