Aim: To assess the efficacy of different frequencies of physical exercise on glycaemic control in adults with prediabetes. Methods: parallel, randomised, controlled, clinical trial will be carried out, with a total of 90 participants. Exercise modality that showed the best glycaemic control in first phase of GLYCEX study (NCT05612698) will be used. Participantds will be randomised in 3 groups: 1) frequency of 5 days/week, 2) frequency of 3 days/week and 3) frequency of 2 days/week. Data collection will be performed at baseline and after 15-weeks of follow up. Sociodemographic data, medication, comorbidity, blood biochemical parameters, blood pressure, anthropometric measurements, body composition, physical activity, sedentary lifestyle, diet, smoking, alcohol consumption, quality of life and sleep questionnaires will be collected. Physical activity, sedentary behaviour and sleep will be further determined with an accelerometer, and continuous glycaemia will be determined with a glycaemic monitor, both during seven days, in two time points. The main dependent variable will be the reduction of the mean amplitude of glycaemic excursions. The impact of the interventions on health will also be evaluated through gene expression analysis in peripheral blood cells. Discussion: The results of this study will contribute to better understanding of the response of glucose mechanisms to physical exercise in a population with prediabetes as well as improving physical exercise prescriptions for diabetes prevention. Increasing glycaemic control in people with prediabetes through physical exercise offers an opportunity to prevent diabetes and reduce associated comorbidities and health costs.

Principal Aim. The main aim of the preset study is to compare the efficacy of 3 different frequencies of physical exercise on MAGE in adults with prediabetes.

Secondary Aim: The secondary aims are to evaluate the effect of different physical exercise frequencies on the following variables: FPG, glycaemic variability measured continuously (24h), normalization of glycaemic values (reversal), BMI, body composition (percentage and grams of total fat mass and muscle mass; visceral adipose tissue (VAT) waist circumference, lipid profile, inflammation markers, blood pressure and sleep duration and quality; transcriptomic markers of metabolic health assessed in peripheral blood cells (PBC).

Participants Participants' identification and recruitment will be carried out at different primary care centres from Mallorca, Spain, that formally agree to participate in this phase, and through posters distributed at different points of interest: hospitals, universities, primary care cen-ters, patient associations, etc. Likewise, it will also be disseminated through social net-works, such as Twitter. Interested patients will receive the information sheet, sign the writ-ten informed consent and eligibility criteria will be verified.

Sample size and randomization The sample size has been calculated to detect significant differences in MAGE of at least 1.5 mg/dl and considering a standard deviation of 1.4 mg/dl in any of the three modalities and a dropout rate of 20%, so 30 subjects will be needed in each group, with a sample of 90 participants. Randomization will be done by permuted blocks of 8 in a 1:1:1 ratio and stratified by prognostic factors: gender, age and obesity using the open-source Oxford Minimization and Randomization (OxMaR) program.

Description of intervention

The intervention will be designed and implemented by a sports science professional. All groups will perform a minimum of 150 minutes of moderate physical exercise per week or a minimum of 75 minutes of vigorous exercise per week, always performed under the supervision of a sports science professional. During the intervention participants will progress to 300 minutes of moderate PA or 150 of vigorous PA. Prior to the start of the full intervention, participants will undergo a 3-week pre-intervention physical conditioning. Each intervention will last 12 weeks. Before starting the intervention program, participants will be referred to a physician for medical testing and clearance. During all physical exercise sessions, participants will wear a heart rate (HR) device that will relay their HR to a laptop, from which the sports scientist will be able to control the intensity of the session. The heart rate monitor model to be used is the Polar OH1, which has been validated in previous studies. Polar HR monitors have demonstrated a high level of agreement with the electrocardiogram and can therefore be used as a valid measure of HR in both laboratory and field studies to measure HR during moderate-to-vigorous physical activity (MVPA). Adherence to the intervention will be measured by attendance at the training sessions that make up the intervention. Different exercise frequencies per week will be compared using the modality that best controlled blood glucose in phase 1 (NCT05612698): 1) Five sessions/week; 2) three sessions/week and 3) two sessions/week. As mentioned above, all groups in both phases will perform a minimum of 150 minutes of moderate physical exercise per week or 75 minutes of vigorous exercise per week, always performed under the supervision of a sports science professional.

- Data collection and procedures: Data collection (visit -1 and 1) will be performed at the Health Research Institute of the Balearic Islands (IdISBa). Visit 0 and the intervention sessions will be held at the facilities of the University of the Balearic Islands.

• Visit -1 (V-1): Before inclusion in the study, potential participants will be scheduled for a first visit to verify inclusion and exclusion criteria. During this visit, informed consent and baseline data will be collected including sociodemographic data, medication and concomitant pathologies, tobacco and alcohol consumption, assessment of PA, SB, diet, quality of life, quality of sleep, anthropometric and body composition parameters, blood pressure, biochemical blood analysis and gene expression analysis in blood cells. Participants meeting all eligibility criteria will be randomly assigned, stratified by sex, age and obesity, to one of the three intervention groups or control group. The same procedure will be followed for both phases of the study. All participants will be asked to wear an accelerometer, a continuous blood glucose monitor and a daily activities e-diary (intake, sleep, SB and PA) for 14 days to collect accurate baseline PA and blood glucose data.
• Visit 0 (V0): Participants will be informed of the exercise modality or frequency to which they have been assigned and the days on which the physical exercise sessions will take place will be agreed upon. The accelerometer and the blood glucometer will be taken away in the subsample. During the visit, the participant will wear a heart rate monitor to record their resting heart rate.
• Visit 1 (V1): Before the follow-up visit (week 15 of intervention), the accelerometer and the continuous glucose monitoring biosensor will be placed for 14 consecutive days, 24h/d, and participants will complete the e-diary during these 14 days. At the end of follow-up (15 weeks), the accelerometer and continuous glucose monitoring biosensor will be re-moved. During this visit medication and concomitant pathologies, assessment of PA, SB, diet, tobacco and alcohol consumption, assessment of the quality of life and quality of sleep, anthropometric and body composition parameters, blood pressure, biochemical blood analysis, gene expression analysis in blood cells and adverse events will be collected for all participants.

Data collection

- Biological samples and laboratory procedures: At visits V-1, and V1, venous blood samples will be collected, after an overnight fast of ≥8 hours. Blood tests performed will include FPG, HbA1c, total cholesterol, high-density lipoprotein cholesterol (HDL-c), low-density lipoprotein cholesterol (LDL-c), triglycerides (TG), gamma-glutamyl transferase (GGT), aspartate aminotransferase (AST), platelets, leukocytes, high-sensitivity C-reactive protein (hsCRP), interleukin-1 (IL-1), interleukin-6 (IL-6), interleukin-8 (IL-8), advanced glycation end products (AGEs) adiponectin, leptin. Blood samples will be analysed centrally. The TG-glucose index (TG) will also be calculated. Additionally, the expression of relevant genes indicative of metabolic health previously identified [59, 60], as well as selected genes with a key role in the pathogenesis of T2D will be assessed by real-time qPCR in blood cells.

• Glycaemic variability The continuous blood glucose monitoring system (Dexcom G6, Dexcom Inc., USA) will be used to assess glycaemic variability for 14 consecutive days in the all sample. The sensor will be placed at V-1 and removed at V0. It will be placed again at the beginning of week 15 and removed at V1. The Dexcom G6 sensor will be inserted into the subcutaneous adipose tissue in the lower abdomen (below the umbilicus). To optimise the usefulness of continuous glycaemic monitoring, participants will be asked to keep an electronic diary (e-diary), where they will record the intakes, the timing of intakes and the time they go to sleep and wake up. Mean amplitude of glycaemic excursions and time in range (TIR) will be estimated from the 14-day sensor glucose profiles. Mean amplitude of glycaemic excursions will be calculated by taking the arithmetic mean of the increases or decreases in blood glucose (from nadirs to peaks or vice versa) when the rising and falling segments exceed the 1 standard desviation blood glucose value over a 24-hour measurement period.
• Accelerometer To quantify PA, SB and sleep, participants will wear a wrist accelerometer on the non-dominant arm for 14 days, 24 hours a day (V-1 and V1), and will complete a sleep log. The device (GENEActiv, ActivInsights Ltd., Kimbolton, UK) is a tri-axial accelerometer that allows the following information to be obtained: sedentary time, light PA, moderate PA, vigorous PA and sleep time. The raw data files will be processed with the R package (R Core Team, Vienna, Austria) using the open-source R package GGIR, version 1.2-5 (cran.rproject.org/web/packages/ggir/index.html), which has been validated against the self-calibrated functions.
• Diet At visits V-1 and V1, diet quality will be assessed using the validated 17-item MedDiet adherence questionnaire. Each item is scored 1 (adherence) or 0 (non-adherence), so the total score can range from 0 to 17, with 0 indicating no adherence and 17 indicating maximum adherence. Moreover, to determine the quality of the diet beyond adherence to the Mediterranean diet, a short food frequency questionnaire will be collected. In addition, other questions related to diet will be included in the sleep log, specifically the time of each intake and the food eaten.
• Physical activity and sedentary behaviour At visits V-1 and V1, the validated REGICOR PA questionnaire for adults, which measures PA levels over the previous 12 months and the validated SB Nurses' Health Study (NHS) questionnaire for the Spanish population, will be administered. The REGICOR PA questionnaire is a sensitive tool to detect moderate and vigorous PA changes in epidemiological studies. It includes a total of 12 questions divided into 3 categories: questions on the type and intensity of leisure-time physical exercise (walking, brisk walking, walking in the countryside, gardening and climbing stairs), occupational PA, and SB, including time spent sleeping. The average daily energy cost of PA is then calculated and expressed in metabolic equivalents (METs). The REGICOR PA questionnaire also collects data on the average monthly time spent in light (< 4 METs), moderate (4-5.5 METs) and vigorous (≥ 6 METs) PA. Finally, the num-ber of weekly hours of SB is also recorded. The number of hours spent on SB will be recorded using the NHS questionnaire. The types of activities recorded are watching TV, sitting in front of a computer/screen, sitting during transport as a driver or passenger and total sitting time. The frequency of activities is assessed as times in minutes (<30 minutes; between 30 and 60 minutes) or hours (between 1 and 9 or more) on an average day, both on weekdays and weekends separately.
• Anthropometric measurements Height, body weight, BMI, waist circumference (WC), and body composition (bioelectrical impedance; Tanita BC-418; Tanita Corp., Tokyo, Japan), will be obtained at visits V-1 and V1 (except for height, which will be measured only at V-1). All anthropometric measurements will be collected according to the recommendations of the International Standards for the Anthropometric assessment (ISAK). Furthermore, all measurements will be performed by well trained technicians or researchers to minimize coefficients of variations. Height will be measured to the mm with a book-foot stadiometer with the participants standing barefoot and with the head placed in the Frankfort plan; body weight will be measured to the nearest 0. 1 kg with a Tanita body fat analyser, which will also measure body composition by bioelectrical impedance; BMI will be calculated using the standard formula (weight (kg)/height (m²)); using a tape measure, WC will be taken at the narrowest point between the lower costal rib and the iliac crest. During height and weight measurements, participants shall adopt a relaxed standing position, with arms crossed over the chest, feet together on the floor and buttock muscles relaxed; both measurements will be taken twice, and the median value will be considered for statistical analyses.
• Blood pressure At visits V-1 and V1, consecutive blood pressure measurements will be taken in each arm and recorded. The arm with the highest median blood pressure shall be considered for statistical analysis and subsequent measurements during follow-up shall be taken from this same arm (V1). If the median blood pressures of the two arms are identical, subsequent measurements shall be taken from the non-dominant arm. Blood pressure measurements will be taken at the brachial artery after 5 minutes of rest in a seated position with a validated oscillometer (Omron M3). For statistical analyses, the mean of two measurements taken 2 minutes apart will be considered.
• Quality of life and sleep Quality of life and sleep quality will be assessed at V-1 and V1, using the following validated questionnaires: The European Quality of Life-5 Dimensions (EuroQol-5D) questionnaire is a widely used measure of self-reported health-related quality of life. The questionnaire is divided into two parts. The first comprises five domains: mobility, personal care, usual activities, pain/discomfort, and anxiety/depression. For each domain, participants can indicate the level of problem experienced on a three-point categorical response scale (no problem, some problems or severe problems). From the answers given a single summary score is created, which indicates the self-reported health status. The second part is a Visual Analogue Scale that captures the overall health perceived by the participant in a score ranging from 0 (worst imaginable health) to 100 (best imaginable health). The 6-item Medical Outcome Study Son Index (Bo-cado-Sleep) is the shortened version of the 12-item MIS-Sleep scale. It is a validated instrument used to assess the quality and quantity of self-reported sleep during the previous month. It uses a list of six items divided into three different domains: sleep disturbances, daytime sleepiness, sleep adequacy, and awakening with shortness of breath or headache. The items are scored on a six-point scale ranging from all the time to never. For interpretation, the direct scores are transformed into a scale from 0 to 100, with no cut-off points; the higher the score, the greater the intensity of the item assessed.

Inclusion Criteria:

• Adults aged 18-65 years
• with overweight or obesity (BMI >25 and <35 Kg/m2)
• inactive (<150 min PA/week)
• with prediabetes (fasting blood glucose 100-126 mg/dl)
• who have signed the informed consent will be included.

Exclusion Criteria:

• People with uncontrolled hypertension
• diagnosis of T2D or oral anti-diabetic prescription
• active cancer
• terminal illness
• cognitive impairment
• pregnancy
• cardiovascular disease
• inability to perform moderate-vigorous physical exercise for the next 3 months
• major surgery or hospital admission in the last 3 months
• haematological disease that interferes with HbA1c determination
• presence of any condition (medical, psychological, social or geographical) current or anticipated that limits participation in the study
• participation in another clinical trial will be excluded.