Already existing and coded tumor biological material and health-related patient data will be retrospectively collected from institutional biobanks and patients' charts or electronic medical records upon receipt of ethical approval. Each patient enrolled in the study will be assigned a unique identification numerical code upon registration in the study. The unique identification code will be used to record health-related data and to label biological samples. The coded biological material will be transferred to the coordinating center at the Institute of Oncology Research (IOR) in Bellinzona. Health-related data will be collected in the electronic case report form (eCRF) (OpenClinica). Data quality will be ensured by query generation.
Annotated baseline features will include the date of diagnosis, date of biopsy, age, gender, Eastern Cooperative Oncology Group Performance Status (ECOG PS), Ann Arbor stage, lactate dehydrogenase (LDH), number and location of extranodal sites, bone marrow involvement and percentage, peripheral blood involvement, number of nodal sites, B symptoms, lymph nodes larger than 7 cm, hemoglobin (Hb), platelets, lymphocytes, beta-2-microglobulin, albumin, infections (hepatitis C virus, Helicobacter pylori, Chlamydophila psittaci, Achromobacter xylosoxidans, Campylobacter jejuni), serum paraprotein presence and type.
Annotated follow-up features included the date of progression to a disease requiring treatment, type of first-line treatment, date of start of the first line treatment, date of progression after first line treatment, date of the second line treatment, type of second line treatment, date of transformation, date of death, cause of death, and date of last follow-up. Mutation analysis, immunoglobulin genes and T cell receptor rearrangement analysis, copy number aberration analysis, structural variant analysis, and deoxyribonucleic acid (DNA) methylation profile will be performed by next-generation sequencing of genomic DNA extracted from the biopsy. Gene expression will be assessed by next-generation sequencing of RNA extracted from the biopsy. Protein expression will be assessed by mass spectrometry of proteins extracted from the biopsy. AI-based computational pathology will be performed on digitized whole-slide images of the biopsy to extract quantitative histologic features.
Single-cell transcriptomic profiling will be used to resolve intratumoral heterogeneity and to define malignant and non-malignant cellular populations. After quality control, normalization, and batch correction, unsupervised clustering will be applied to identify transcriptionally distinct cell states. Malignant B-cell populations will be distinguished from reactive B cells using a combination of copy number inference, immunoglobulin expression patterns, and canonical marker genes. Differential expression and pathway enrichment analyses will be conducted to identify signaling programs associated with anatomical site, immune context, and disease features.
In details:
* Spatial transcriptomics and/or spatial proteomics will be employed to preserve tissue architecture and assess the spatial organization of tumor cells, immune infiltrates, and stromal components. Spatial domains and cellular neighborhoods will be identified using data-driven segmentation and neighborhood analysis approaches. Spatial co-localization and interaction analyses will be used to infer tumor-microenvironment crosstalk, immune exclusion or activation niches, and site-specific spatial patterns that may influence therapeutic response.
* Immune repertoire profiling (BCR/TCR sequencing) will be integrated, where available, to assess clonality, antigen-driven selection, and spatial distribution of immune clones. In malignant B cells, immunoglobulin features will be used to explore evidence of antigen dependence and ongoing selection, while TCR diversity and expansion will be evaluated in relation to immune niches and tumor proximity.
* Proteomic and/or epigenetic data, when available, will be incorporated to refine functional interpretation of transcriptional programs and to identify post-transcriptional or regulatory mechanisms underlying observed cellular states.
Cross-modal integration will be performed using established computational frameworks to generate unified cell state annotations and pathway activity scores. Comparative analyses across anatomical sites will be a central assessment. Conserved versus site-specific cellular states, immune compositions, and signaling pathways will be systematically evaluated to identify shared disease mechanisms as well as context-dependent features. Associations with clinical variables (e.g., site of origin, prior treatment, disease stage) will be explored in an exploratory, hypothesis-generating manner.
All analyses will follow reproducible computational workflows with stringent quality control, appropriate correction for technical confounders, and transparent reporting of limitations. The resulting datasets and analytical outputs will provide an integrated molecular and spatial reference framework for MALT lymphoma, supporting downstream biomarker discovery and future translational studies.
