特発性多中心性キャッスルマン病

Background/Objectives: Castleman disease is a rare cytokine-driven lymphoproliferative disorder in which lymph node histopathology provides key diagnostic information. Morphologic features are graded semiquantitatively and often show substantial interobserver variability. Methods: We developed an automated approach to grade six Castleman disease-associated histologic features on hematoxylin and eosin (H&E) whole slide images (WSIs) using an attention-based multiple instance learning (MIL) model built on a large pathology foundation model encoder. A multi-institution cohort comprised 544 lymph node WSIs, including 397 from cases with Castleman disease or Castleman-like histology and 147 from cases without suspicion for Castleman disease. These case ascertainment categories were not model prediction targets. Slides were assigned ordinal grades (0-3) for regressed germinal centers, follicular dendritic cell prominence, increased vascularity, hyperplastic germinal centers, plasmacytosis, and follicular twinning by hematopathologists. Model performance was assessed on a held-out evaluation set of 142 WSIs. Results: Across the six features, accuracy ranged from 0.52 to 0.68 (mean 0.60). Disagreements were predominantly minor: 96% of predictions were within one grade of the reference. Feature-specific tile contribution heatmaps showed qualitative spatial correspondence with selected plasma cell-rich, vessel-rich, and follicular regions but were not evaluated as quantitative feature localization maps or causal explanations. In a preliminary reader study, concordance with the reference varied widely among hematopathologists (Krippendorff's alpha 0.20-0.95, mean 0.50), with the model demonstrating concordance in the mid-range (0.56). Conclusions: These findings support the feasibility of automated grading of Castleman disease-associated histologic features for research standardization. The model does not classify Castleman disease, and its potential use as an input to diagnostic or clinical trial workflows requires evaluation in separately designed studies with adjudicated diagnostic labels and integrated clinical data.

Castleman disease (CD) represents a heterogeneous group of lymphoproliferative disorders, with the idiopathic multicentric CD (iMCD) subtype associated with thrombocytopenia, anasarca, fever, reticulin fibrosis/renal dysfunction, and organomegaly (TAFRO) syndrome (iMCD-TAFRO) posing significant research challenges because of its severity and tissue scarcity. This study used integrated bulk and spatial proteomics to characterize protein expression profiles in lymph node samples across clinical and pathological subtypes. Bulk proteomics revealed subtype-specific molecular signatures, including distinct protein profiles for hyaline vascular vs plasmacytic variants, systemic complement activation distinguishing iMCD from unicentric CD, and molecular evidence supporting the classification of iMCD-idiopathic plasmacytic lymphadenopathy as a distinct entity. Specifically, iMCD-TAFRO exhibited upregulation of angiogenesis drivers (PDGFRβ and NOTCH3), interferon signaling proteins (STAT1/ISG15), and fibrosis markers (COL3A1/LOXL1). Spatial proteomics, through laser capture microdissection of follicular and vascular niches, delineated compartmentalized pathogenesis in iMCD-TAFRO. Intrafollicular vessels showed enrichment of myofibroblast markers (ACTA2) and programmed cell death regulators (eg, GSDMD and TFRC), whereas interfollicular regions displayed TGF-β/SMAD3-mediated upregulation of LOXL1 linked to fibrosis. Follicular zones demonstrated complement activation (CFHR1/CFHR2) and M2 macrophage infiltration. Our integrated approach delineated the molecular heterogeneity of CD, revealed spatially resolved pathogenic drivers in iMCD-TAFRO, and provided insights into its unique vascular-fibroinflammatory pathology, offering a foundation for advancing precision diagnostics and targeted therapies.

OBJECTIVES: To characterise early treatment-associated changes in circulating immune mediators following sirolimus initiation in patients with tocilizumab-resistant idiopathic multicentric Castleman disease (iMCD). METHODS: As a predefined translational sub-analysis of a placebo-controlled exploratory trial, we performed longitudinal serum proteomic profiling using a RayBiotech L-1000 antibody array (∼1,000 proteins) in a sirolimus-treated patient with tocilizumab-resistant iMCD. Serum samples were obtained at baseline and at weeks 8, 12, and 16. Protein expression levels were normalised, log₂-transformed, and analysed for stepwise temporal changes. Pathway-level effects were assessed using gene set enrichment analysis (GSEA). RESULTS: Six immune mediators-CCR5, ICAM-1, RANTES, S100A8/A9, CD40 ligand, and OX40 ligand-showed consistent monotonic decreases from baseline through week 12. Heatmap visualisation of the top 80 dynamically regulated proteins confirmed a broader suppression of immune activation signatures. GSEA demonstrated significant negative enrichment of coagulation, complement, IFN-γ response, IL-6-JAK-STAT3 signalling, TNFα signalling via NF-κB, and MMP-driven extracellular matrix remodelling. Temporal analysis confirmed sustained downregulation of IFN-γ-inducible chemokines (CXCL9, CXCL10, CXCL11) and a broad panel of interferon-response mediators through weeks 12 and 16. CONCLUSIONS: Early and coordinated decreases in immune-inflammatory mediators were observed during sirolimus therapy in tocilizumab-resistant iMCD, extending beyond the IL-6 axis to encompass IFN-γ, TNF-α, and chemokine signalling. These findings are hypothesis-generating and provide a mechanistic rationale for mTOR inhibition as a potential disease-modifying strategy warranting evaluation in future controlled studies.

BACKGROUND: Idiopathic multicentric Castleman disease (iMCD) is a rare lymphoproliferative disorder that is associated with a broad range of symptoms, including constitutional, gastrointestinal, neuropsychiatric, dermatologic, respiratory, and hematologic or lymphoreticular problems. These broad symptoms can impact the daily lives of people living with iMCD, creating a high symptom burden. Despite this, no robust, disease-specific patient-reported outcome measure (PROM) for subjective iMCD symptom burden exists. This limits accurate symptom monitoring, impacts sensitive end point selection in clinical trials, and represents a regulatory gap in patient-centered evidence generation when evaluating iMCD treatments. OBJECTIVE: This protocol describes the international multistakeholder Idiopathic Multicentric Castleman Disease Symptom Burden Scale (ISBUS) project. The primary aim of this project is to develop and preliminarily evaluate the measurement properties of a novel PROM for capturing symptom burden (ie, perceived symptom frequency and impact on daily life) in people living with iMCD. The central objective is to produce a valid and robust PROM that can be used to assess iMCD symptom burden in research, clinical trials, and symptom monitoring in clinical practice. METHODS: The project has a mixed methods design, split into 4 sequential stages, with collaborative advisory input throughout. Stage 1 uses existing data and consultation with expert clinical and patient advisors to generate draft PROM content. Stage 2 uses qualitative cognitive debriefing interviews (n=10) to evaluate the content validity of the draft PROM content and enable meaningful revisions. Stage 3 involves a cross-sectional quantitative survey design in people living with iMCD (n≥50), allowing for psychometric analyses of structural validity (using classical test theory and/or Rasch methodology), construct validity (known-group validity and correlational methods), and internal consistency reliability (Cronbach α). Stage 4 uses a mixed methods design, including longitudinal quantitative survey evidence (n≥20) and qualitative interviews (n=10), triangulated to estimate preliminary meaningful change estimates for the new PROM. The project is international in scope, with primary data collection in 6 countries (ie, Australia, Brazil, Canada, New Zealand, the United Kingdom, and the United States). RESULTS: Funding for ISBUS began in June 2023 and is ongoing. As of March 2026, stage 3 had been completed, with 51 people living with iMCD completing the survey, and stage 4 was ongoing, with 32 follow-up surveys completed. Analyses for stages 1-3 were completed and are expected to be published in 2026, followed by stage 4 findings in 2027. CONCLUSIONS: The aim of the ISBUS project is to produce a novel symptom burden PROM codeveloped with patients with iMCD to measure what matters to patients with iMCD. It is anticipated that the new PROM will be used in iMCD research, as a secondary end point in trials, and as a clinical tool to monitor symptom burden in the management of iMCD. TRIAL REGISTRATION: ClinicalTrials.gov NCT05995834; https://clinicaltrials.gov/study/NCT05995834. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/96022.