Precision Physical Exercise for Personalized Onco-Hematology.

The main aim of the PEPOH project is to characterise and predict clinical response to strength training in solid and/or hematological tumors in patients (under active therapy and/or during the therapy process) by an exhaustive and systematic integration of multi-omics datasets to determine signatures associated with treatment tolerance, disease progression and overall survival.

Several specific objectives derive from this main goal:

1. - Determine the immunomodulator effect of physical exercise by deciphering differential molecular profiles in peripheral blood by next-generation sequencing (RNA, SNPs, CHIP), proteomics and metabolomics, which provide novel insights for a comprehensive analysis of the effect of exercise on patients
2. - Evaluate the impact of a physical exercise programme in oncohematological patients on multidimensional indicators of their intrinsic capacity related to overall quality of life:
3. - Functional physical condition (muscle strength, walking speed, balance, and fatigue).
4. - Psychological well-being and mental health.
5. - Body composition and anthropometric variables.
6. - Sleep quality, overall quality of life, and cancer-related quality of life. All these objectives respond to three main priorities in current oncology:
7. to understand the physiological and molecular mechanisms underlying the therapeutic effects of exercise.
8. to generate robust evidence in highly vulnerable and underrepresented populations, such as hematologic cancer patients.
9. to establish the basis for incorporating strength training as a standardized adjuvant intervention within the clinical management of cancer. The added value of PEPOH lies in its translational and multidisciplinary approach, which combines basic science with clinical practice, molecular biology with exercise physiology, and multi-omic analysis with comprehensive patient assessment. Through this framework, the project aims to contribute to a paradigm shift in oncology care by integrating physical exercise as a personalized therapeutic tool capable of improving survival, well-being, and quality of life in people with cancer. Bearing in mind the multidisciplinary aspect of the PEPOH project, the molecular fingerprint -obtained by multi-omics characterization- of oncohematological patients under strength training will be an asset to provide the biological basis of its therapeutic use, as it is related to treatment tolerance and survival. Additionally, these molecular signatures could provide novel biological insights in inflammation regulation, energy metabolism, tumor immunity, muscle plasticity, aging and frailty. All these findings will allow the investigators to decipher mechanisms that explain the functional and emotional improvement observed after exercise. Moreover, the deep multi-omic characterization of a well-characterized and defined cohort of patients will provide novel datasets, which could be highly useful for further biomedical studies and a reference for the scientific community. From a clinical point of view, the systematic and accurate integration of molecular and clinical datasets will enable the development of predictive response models which help to evaluate the efficacy of exercise interventions to each patient and tumor, named precision exercise. In summary, the PEPOH project will provide translational evidence to support the incorporation of exercise as a personalized and cost-effective adjuvant intervention within the comprehensive approach to patients with cancer, contributing to the development of precision oncology. Study Design PEPOH is a randomized, controlled clinical trial. The study will be conducted at Salamanca University Hospital, with the technical and scientific support of FIBSAL. Population The trial will include five patient cohorts corresponding to the following diagnoses: lung cancer, colorectal cancer, breast cancer, chronic lymphocytic leukemia, and multiple myeloma. Each cohort will consist of 50 patients, resulting in a total sample of 250 patients. After a baseline assessment, patients will be randomly assigned (1:1) to one of two study groups: Supervised intervention group: a strength training programme conducted twice per week for 12 weeks under professional supervision, complemented by a home-based exercise programme. Control group: a home-based exercise programme following the same recommendations and guidelines as a supervised group but without supervised in-person sessions. Assessments will be performed at baseline and after the 12-week intervention, with an additional follow-up at 6 months to analyze the persistence of clinical and molecular effects in the medium term. Inclusion criteria:

• Adult patients (≥18 years), capable of signing the informed consent and willing to participate in the study.
• Performance status (ECOG) 0-2. Specific inclusion criteria are defined for each cohort involved in the study as follows: Lung cancer: Patients diagnosed with non-small cell lung carcinoma in metastatic stage, eligible for combined immunotherapy and chemotherapy, with active treatment. Colorectal cancer: Patients with stage II, III, or IV resected colorectal cancer undergoing adjuvant chemotherapy. Breast cancer: Female patients with histologically confirmed hormone receptor-positive/HER2-negative breast cancer, receiving active hormonal therapy (tamoxifen, aromatase inhibitor, or fulvestrant) in either adjuvant or stable metastatic phase. Chronic lymphocytic leukemia (CLL): Patients diagnosed with CLL who have not received and are not in need of pharmacological treatment. Multiple myeloma (MM): Patients with a confirmed new diagnosis of MM under active first line treatment and non-candidates to autologous transplantation. Exclusion criteria: Presence of unstable bone metastases or extensive bone involvement associated with a high risk of fracture. Uncontrolled cardiovascular, respiratory, musculoskeletal, or metabolic diseases that contraindicate physical exercise at the investigator's discretion. Explicit medical contraindication to exercise training. Documented poor adherence (<80%). Treatment discontinuation, disease progression, intolerance, or any other medical, personal, or logistical circumstance that, in the opinion of the investigators, may compromise participant safety or the integrity of the study. Sample size A two-factor ANOVA design with a between individual factor (control and intervention groups) and a within individual factor (pre-post and six-month follow-up) was chosen to evaluate the sample size of study. The total sample size required would be 42, considering an effect size of f=0.20, a significance level of 5%, a power of 80%, and a correlation between repeated measures of 0.5. It is estimated that there may be a 20% loss to follow-up between the two experimental groups, and therefore the total sample size is 50 individuals per cohort. Methodology Biological samples of study Peripheral blood samples (one 10 mL EDTA tube) will be collected at three time points: 1) before starting the programme; 2) after 12 weeks of intervention; 3) 6-month follow-up. Isolated peripheral blood mononuclear cells (PBMCs), serum and plasma, which will be properly preserved at IBSAL biobank under standardized conditions. Proteomics Characterization Quantitative proteomic analysis of plasma by PreOmics® ENRICHplus technology, followed by liquid chromatography coupled with ion mobility spectrometry (LC-IMS/MS), which will allow for reproducible and highly sensitive analysis from a minimum plasma volume (50µL). Quantitative proteomic analysis of the eluted peptides will be performed on an LC Evosep One system (Evosep Biosystems, Denmark), coupled to a MS timsTOF Pro 2 (Bruker Corporation, USA), using a separation method of 60 samples/day. A DIA-PASEF approach will be applied with a mass/charge range of 100 to 1700 and an ion mobility range of 0.85-1.30 1/K0, with a cycle time of 100ms. Immunophenotyping For immune population analyses (objective 1), investigators will use a panel including antibodies against CD45, CD3, CD4, CD8, CD5, CD7, TCR, CD27, CD45RA, CD127, CD25RA, CD56, HLA-DR, CD11c, CD123, CD33, CD14, and CD16 with a minimum acquisition of 500,000 events. This panel will allow the identification of different subpopulations of T lymphocytes, NK cells, dendritic cells, and monocytes. For B cell analysis, we will employ the LST screening strategy as described by EuroFlow. All the analyses will be carried out using a 8-color FACScanto cytometer and Infinicyt software. All samples will be processed at the General Flow Cytometry Service of the University of Salamanca (www.nucleus.usal.es) and at the Hematology Department. Metabolomics Profiling Plasma proteins will be precipitated with methanol, and glycerophospholipids extracted by SPE. After filtration, each solution will be adjusted to 600 μL with D₂O and transferred to 5 mm NMR tubes. NMR analyses will be performed at 298 K on a Bruker Avance Neo 400 MHz spectrometer with a cryoprobe. Parameters will be optimized for metabolite quantification using Chenomx NMR Suite 10 (Chenomx, Edmonton, Canada). Metabolite identification and quantification will be carried out with this software, which enables reliable deconvolution and concentration determination in complex spectra. Proton peak assignments will be further validated by comparing chemical shifts with the Human Metabolome Database (http://www.hmdb.ca). Transcriptomic and Genomic Analysis cfRNA and DNA Extraction Plasma cfRNA extraction will be performed using the miRNeasy Serum/Plasma Advanced Kit (Qiagen), optimized for low-yield and short RNA fragments. RNA concentration and purity will be determined fluorometrically using the Qubit RNA HS Assay (Invitrogen).Plasma ctDNA will be extracted using the QIAamp Circulating Nucleic Acid Kit (Qiagen, Hilden, Germany). Genomic DNA (gDNA) will be isolated from peripheral blood using the QIAamp DNA Blood Midi Kit (Qiagen, Germany) following standard protocols. Library Preparation and Sequencing cfRNA libraries will be generated using the Qiagen QIAseq cfRNA All-in-One Kit, optimized for low RNA input and short fragments. Sequencing will be performed on a NovaSeq X sequencer (Illumina, San Diego, CA, USA), producing 150bp paired-end reads (2×150bp) with an average depth greater than 50 million reads to maximize the detection of low-abundance transcripts. ctDNA and gDNA library preparation and sequencing will be conducted using the SureSelect V6 capture kit (Agilent, Santa Clara, CA, USA) and Illumina technology (NovaSeq 6000, 150PE 2×150bp), reaching an average sequencing depth of 18Gb/sample. Bioinformatic Analysis The quality of FASTQ files obtained from sequencing will be assessed with FastQC (Quality Control Tool for High Throughput Sequence Data). Raw reads will undergo quality control with FastQC and MultiQC, followed by adapter trimming and removal of low-quality reads using Trimmomatic. Alignment to the human reference genome (GRCh38) will be performed with STAR (v2.7), and gene abundance quantified using featureCounts (Subread package). Normalization and differential expression analysis will be carried out using DESeq2 (Bioconductor), applying adjusted p-values <0.05 as the significance threshold. Functional and pathway enrichment analyses will be conducted with clusterProfiler, GOplot, and GSEA. Differentially expressed genes will be validated by reverse transcription and quantitative real-time PCR. Genomic Analysis Paired-end reads will be aligned to the human genome (GRCh37) using BWA-MEM (v0.7.17). The resulting SAM files will be processed to improve the efficiency of downstream analyses, including duplicate removal using MarkDuplicatesSpark (GATK). Base quality score recalibration will be performed with BaseRecalibrator and ApplyBQSR (GATK). Somatic variant calling will be conducted with Mutect2 (GATK, paired mode), followed by filtering using FilterMutectCalls. Germline variants (SNVs and INDELs) will be identified with HaplotypeCaller and filtered using hard filters. Selected variants will be annotated with Funcotator, and variant pathogenicity assessed with Varank (v1.4.3). The frequency and distribution of detected SNVs and INDELs will be compared against healthy donor databases (NHLBI-ESP and gnomAD) using ANNOVAR. Multi-Omics Integration Omics data will be integrated using pipelines developed by the Bioinformatics Unit of IBSAL. Machine learning tools (Random Forest, SVM) and network analysis (Weighted Gene Co-expression Network Analysis, WGCNA) will be applied to identify common and tumor-type-specific molecular patterns. Multi-OMics-Factor Analysis (MOFA). The integration of multi-omic and clinical data will enable the development of predictive models of response to physical exercise. Clinical, Functional, and Psychological Assessments All clinical and functional outcomes will be collected at three time points: 1) before starting the programme, 2) after 12 weeks of intervention, 3) 6-month follow-up. In all cases, all individuals will be evaluated: Muscle strength (Dynamometry), body composition (Tanita BC-418), functional status (Short Physical Performance Battery), quality of life (EORTC QLQ-C30), Psychological well-being (Hospital anxiety and depression scale), sleep quality (Athens Insomnia scale), physical activity level (International Physical Activity Questionnaire). Intervention The intervention of both groups (strength training programme and a home-based exercise programme) is detailed in the published study protocols.