---
title: "Clusterin expression and germinal center morphometry distinguish iMCD from select lymphadenopathies"
id: "frontiers-in-immunology-16-clusterin-expression-and-germinal-center-morphometry-help-distinguish"
canonical_url: "https://medichelpline.com/clinical-feed/frontiers-in-immunology-16-clusterin-expression-and-germinal-center-morphometry-help-distinguish"
content_type: "clinical_feed_article"
specialty: "Infectious Disease"
source_name: "Frontiers in Immunology"
source_url: "https://www.frontiersin.org/articles/10.3389/fimmu.2026.1872505"
published_at: "2026-09-17T00:00:00.000Z"
evidence_level: "Journal Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Clusterin expression and germinal center morphometry distinguish iMCD from select lymphadenopathies
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/frontiers-in-immunology-16-clusterin-expression-and-germinal-center-morphometry-help-distinguish
- **Specialty:** [Infectious Disease](https://medichelpline.com/clinical-feed/infectious-disease.md)
- **Primary Source:** Frontiers in Immunology
- **Source URL:** [Original Journal Publication](https://www.frontiersin.org/articles/10.3389/fimmu.2026.1872505)
- **Published At:** 2026-09-17T00:00:00.000Z
- **Evidence Rating:** Journal Feed
## Executive GIST (TL;DR)
- Idiopathic multicentric Castleman disease (**iMCD**) is a heterogeneous cytokine-driven lymphoproliferative disorder with clinical subtypes including **iMCD-TAFRO**, iMCD-IPL, and iMCD-NOS, and variable histopathology. Diagnosis is challenging because features overlap with autoimmune and neoplastic lymphadenopathies. - The authors performed bulk RNA sequencing and targeted gene expression (NanoString) on lymph node (LN) tissue from multiple iMCD clinical subtypes, sentinel controls, SLE, and DLBCL to identify distinguishing molecular features. - In the discovery cohort (iMCD-TAFRO vs sentinel controls) they found 249 upregulated and 42 downregulated genes; enriched pathways included angiogenesis, cell proliferation, and humoral and innate immune responses by GSEA. - Shared differentially expressed genes across all three iMCD clinical subtypes included SPP1, XBP1, PRDM1, and **CLU** (Clusterin). Unique DEGs were identified per subtype: PLA2G2A, MSR1, IL8 (iMCD-TAFRO); IL9, KIR3DL1, IL18RAP (iMCD-IPL); IL7, TLR8, CX3CR1 (iMCD-NOS). - **Clusterin (CLU)** was significantly upregulated in iMCD lymph nodes but not in related conditions; IHC localized elevated Clusterin to germinal center and mantle zone regions in iMCD samples. - The team developed a digital pathology algorithm to quantify DAB Clusterin staining and manually segmented germinal centers, mantle zones, and interfollicular regions for morphometric analysis. - A composite model combining Clusterin expression in lymph node compartments with germinal center features showed strong ability to discriminate iMCD from selected lymphadenopathies, suggesting Clusterin plus GC morphometry as a candidate biomarker. - The study used discovery (bulk RNAseq) and validation cohorts, NanoString targeted panels, cell-type deconvolution using CellTypist references and Bisque, and queried LINCS1000 to nominate potential reversing drug perturbations for iMCD-TAFRO. - Sample sizes reported include discovery sentinel controls (n=7), discovery iMCD-TAFRO (n=7), Nanostring cohorts encompassing iMCD-TAFRO (n=12), iMCD-IPL (n=3), iMCD-NOS (n=4), SLE (n=4), and DLBCL (n=5); validation cohorts included additional iMCD samples described in supplementary material. - All methods, cohorts, and analytical pipelines (Kallisto, DESeq2, GSEA, RUVSeq, limma, Bisque) were described in the source; specific numerical performance metrics for the composite classifier were not detailed in the provided text.
## Clinical Analysis & Structured Key Points
Frontiers | Clusterin expression and germinal center morphometry help distinguish idiopathic multicentric castleman disease from select lymphadenopathies ORIGINAL RESEARCH article Front. Immunol. , 17 September 2026 Sec. Autoinflammatory Disorders Volume 17 - 2026 | https://doi.org/10.3389/fimmu.2026.1872505 Published in Frontiers in Immunology Autoinflammatory Disorders 7 impact factor 11.3 citescore Editor & Reviewers Edited by M K Marshall Kadin Reviewed by M Y Motohisa Yamamoto R O Robert Ohgami Outline Figures and Tables Figure 1 View in article Figure 2 View in article Figure 3 View in article Figure 4 View in article Figure 5 View in article Table 1 Demographic and clinical characteristics of study cohorts. View in article Table 2 Significantly enriched hallmark gene pathways in the transcriptome of iMCD-TAFRO lymph node (FDR 1.5). View in article Table 3 Top 20 genes by log 2 fold change in the discovery iMCD-TAFRO cohort, with corresponding validation cohort statistics, adjusted p-values, and ranks in both cohorts. View in article ORIGINAL RESEARCH article Front. Immunol. , 17 September 2026 Sec. Autoinflammatory Disorders Volume 17 - 2026 | https://doi.org/10.3389/fimmu.2026.1872505 Clusterin expression and germinal center morphometry help distinguish idiopathic multicentric castleman disease from select lymphadenopathies M V Michael V. Gonzalez 1 K W Kaiwen Wang 1 J Z Joseph Zinski 1 M D Melanie D. Mumau 1 K S Katherine S. Forsyth 1 A H Abiola H. Irvine 1 J B Joshua Brandstadter 2 S G Stacy G. Guzman 3 L Z Lu Zhang 4 S K Sheila K. Pierson 1 B A Bridget Austin 1 +3 more D C David C. Fajgenbaum 1 * 1. Center for Cytokine Storm Treatment & Laboratory, Department of Medicine, University of Pennsylvania, Philadelphia, PA, United States 2. Division of Hematology/Oncology, Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia, Pennsylvania, PA, United States 3. Department of Biochemistry, Biophysics, and Chemical Biology, University of Pennsylvania Perelman School of Medicine, Philadelphia, PA, United States 4. Department of Hematology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China See more Article metrics View details Abstract Idiopathic multicentric Castleman disease (iMCD) is a heterogeneous cytokine storm disorder involving systemic inflammation, multicentric lymphadenopathy with characteristic histopathology, and life-threatening multiple organ dysfunction. Patients can present with symptoms ranging from thrombocytopenia, anasarca, fever/elevated C-reactive protein (CRP), reticulin myelofibrosis, renal dysfunction, and organomegaly (iMCD-TAFRO) to thrombocytosis, hypergammaglobulinemia, and plasmacytosis (iMCD-IPL), with patients not falling into either group (iMCD-NOS). The molecular mechanisms across the clinical subtypes have not been elucidated and new effective treatments are needed. Here, we performed bulk RNA sequencing and targeted gene expression quantification in lymph node tissue from multiple iMCD clinical subtypes, pathologically related disorders, and controls. We identified 249 upregulated and 42 downregulated genes in iMCD-TAFRO lymph node tissue, which were enriched in the following pathways: angiogenesis, cell proliferation, and various aspects of the humoral and innate immune response. The targeted gene expression analysis revealed shared differentially expressed genes (DEGs) across all three iMCD clinical subtypes, including SPP1 , XBP1 , PRDM1 , and CLU . We identified unique DEGs in each iMCD clinical subtype, including PLA2G2A, MSR1, IL8 (iMCD-TAFRO), IL9, KIR3DL1, IL18RAP (iMCD-IPL), and IL7, TLR8, and CX3CR1 (iMCD-NOS). Since Clusterin ( CLU ) was significantly upregulated in iMCD but not in related conditions, we performed immunohistochemistry and observed that Clusterin is elevated in the germinal center and mantle zone regions of iMCD patient lymph nodes. A composite model incorporating CLU expression in these lymph node compartments with germinal center features demonstrated strong performance in differentiating iMCD from selected lymphadenopathies. Together, this work identified pathways dysregulated in iMCD lymph nodes and suggests that Clusterin, particularly when integrated with germinal center features, could be a novel biomarker. Introduction Idiopathic multicentric Castleman Disease (iMCD) is a rare polyclonal lymphoproliferative disorder of unknown etiology with an annual incidence of approximately 1,500 individuals in the United States and a 25-35% 5-year mortality rate ( 1 – 3 ). iMCD is characterized by a wide range of clinical and pathological features, including multifocal lymphadenopathy with unique histopathologic features and cytokine-driven inflammation ( 4 ). Within iMCD, multiple clinical phenotypes exist ( 5 ). Patients presenting with thrombocytopenia, anasarca, fever/elevated C-reactive protein (CRP), reticulin myelofibrosis/renal dysfunction, and organomegaly have been defined as iMCD-TAFRO. iMCD with idiopathic plasmacytic lymphadenopathy (iMCD-IPL) is defined by thrombocytosis, hypergammaglobulinemia, and a milder clinical course. Patients who do not meet these clinical criteria are classified as iMCD–not otherwise specified (iMCD-NOS), which has milder clinical features more similar to iMCD-IPL than to iMCD-TAFRO. Patients can also be classified by histopathological subtype, including hyaline vascular/hypervascular, plasmacytic, and a mixed variant that incorporates features of both. However, the clinical implications of these histopathological subtypes remain unclear ( 6 ). The etiology of iMCD is not well understood, and its heterogeneous presentation suggests multiple possible pathophysiological mechanisms. Roughly half of iMCD patients respond to siltuximab, a monoclonal antibody targeting the pro-inflammatory cytokine interleukin-6 (IL-6 ( 7 , 8 ). Systemic inflammatory symptoms of iMCD can be observed in other inflammatory disorders and hematologic malignancies, complicating its diagnosis. However, gene expression differences in lymph node (LN) tissue between these clinically and pathologically related disorders have not been thoroughly explored, providing fertile ground to better identify more specific features of iMCD. Most discovery-based investigations of gene expression networks or proteomic signatures involved in iMCD have been done in circulating peripheral blood mononuclear cells (PBMCs) or in plasma/serum, but not in the LN ( 8 – 11 ). Nevertheless, important pathogenic insights have been made. In a large serum proteomics-based study involving 88 iMCD patients and several clinico-pathologically related disorders, we found that the PI3K-AKT-mTOR, angiogenesis, IL-6-JAK-STAT3, and CXCL13 pathways, as well as up-regulated cytokines, were among the most enriched in iMCD serum ( 3 , 8 , 9 ). In PBMCs, we found that interferon-stimulated genes (ISGs) were upregulated and strongly correlated with mTOR signaling in several immune cell types in iMCD patients ( 10 ); interestingly, mTOR inhibition is an effective treatment in some patients. More recently, Wing et al. ( 12 ) and Horna et al. ( 13 ) explored gene expression changes in MCD LN tissue compared to healthy controls. These studies identified plasma cell differentiation ( XBP1 ), follicular dendritic cell markers ( CXCL13 , CLU ), angiogenesis ( VEGF, VEGFR ), mTORC1 pathway genes, and complement cascade signatures that were significantly increased in MCD LN tissues ( 12 , 13 ). However, the sample size was limited in these studies, and HHV-8 status was not defined in Wing et al. Additionally, subtyping based on the clinical manifestations of MCD was not performed in either study. Herein, we investigate the LN transcriptome across multiple iMCD subtypes (iMCD-TAFRO, iMCD-IPL, iMCD-NOS) compared to both sentinel controls and clinico-pathologically overlapping autoimmune (systemic lupus erythematosus, SLE) and neoplastic (diffuse large B-cell lymphoma, DLBCL) conditions, and perform targeted studies of genes that distinguish these conditions. Although these subtypes of iMCD have been defined and described ( 14 , 15 ), the LN transcriptomic differences have not been explored. Materials and methods Patient samples and histologic review Discovery and validation cohorts We identified samples from 7 patients enrolled in the ACCELERATE natural history registry (NCT02817997) that met criteria for iMCD according to current international diagnostic criteria ( 16 ). Additionally, all samples/studies described were approved by the Institutional Review Board of the University of Pennsylvania (IRB # 824758). Based on clinical and laboratory data collected at the time of diagnosis, patients were categorized as iMCD-TAFRO (n=7) ( 17 ). Additionally, 7 LN samples resected from breast cancer patients negative for metastasis were chosen to represent controls, herein referred to as sentinel controls ( Table 1 ). Histologically negative sentinel lymph nodes were used as a control group because they represent non-diseased lymphoid tissue obtained through routine clinical care that lack the clinical and pathological features of iMCD and other lymphadenopathies under study. We also obtained LN samples from patients with DLBCL (n=5) from a University of Pennsylvania pathology core service. Table 1 Characteristic Discovery: sentinel controls (n=7) Discovery: iMCD-TAFRO (n=7) Nanostring: sentinel controls (n=8) Nanostring: iMCD-TAFRO (n=12*) Nanostring: iMCD-IPL (n=3) Nanostring: iMCD-NOS (n=4) Nanostring: SLE (n=4) Nanostring: DLBCL (n=5) Age, mean (SD) NA 49.1 (12.01) NA 44.7 (13.71) 40.9 (12.07) 44.4 (16.67) NA NA Sex (M:F) NA 2:5 NA 4:8 1:2 3:1 NA NA Location of tissue sampled, n (%) NA Axillary 4 (67%), Groin 1 (16.7%), Mediastinal 1 (16.7%), Not specified 1 (16.7%) NA Axillary 5 (41.7%), Inguinal 2 (16.7%), Cervical 1 (8.3%), Groin 1 (8.3%), Mediastinal 1 (8.3%) Cervical 2 (66%), Groin 1 (33%) Axillary 2 (50%), Abdominal 1 (25%), Cervical 1 (25%) NA NA Hyaline vascular, n (%) NA 5 (83.3%) NA 9 (75%) 1 (33%) 2 (50%) NA NA Plasmacytic or mixed, n (%) NA 1 (16.7%) NA 3 (25%) 2 (66%) 2 (50%) NA NA Not specified histology, n (%) NA 1 (16.7%) Demographic and clinical characteristics of study cohorts. NA, Not Available; SD, Standard Deviation. *7 iMCD-TAFRO samples in the Nanostring cohort overlap bulk RNAseq samples described in the Discovery cohort. In addition to the discovery cohort, we identified 22 additional patients enrolled in the ACCELERATE natural history registry. These patients were categorized into two subtype groups in the validation cohort: iMCD-TAFRO (n=15) and iMCD-NOS/IPL (n=12). The NOS/IPL subtypes were combined because the number of IPL samples was insufficient for comparison (n=1). We also included previously published iMCD samples and sentinel controls ( 15 ), which included 8 non-specific MCD and 19 sentinel controls. The composition of the validation cohort is summarized in Supplementary Table 1 . Targeted gene expression cohort Finally, our targeted gene expression cohort included multiple iMCD clinical subtypes and inflammatory disease comparators. Based on clinical and laboratory data collected at the time of diagnosis, patients were categorized into one of three subtypes: iMCD-TAFRO (n=12), iMCD-IPL (n=3), or iMCD-NOS (n=4) ( 18 ). In addition to the iMCD subtype samples, SLE (n=3) and DLBCL (n=5) samples were used as disease comparator groups ( Table 1 ). RNA integrity values were used to select high-quality RNA samples. Samples that overlap between the Discovery and Targeted Gene Expression Cohort can be found in Supplementary Table 2 . Sample preparation and gene expression quantification RNA extraction from FFPE unstained slides RNA was isolated from Formalin Fixed Paraffin Embedded (FFPE) tissue sections fixed on slides using the RNeasy DSP FFPE kit (Qiagen, Hilden, Germany). Briefly, sections 10 microns thick were scraped from 5 slides of each sample and submerged in Deparaffinization Solution (DSP) according to the protocol. During this process, tissue digestion using Proteinase K was also done. DNAseI-treated RNA was subsequently isolated using Qiagen’s proprietary spin column method, as per the protocol. RNA integrity was assessed using the TapeStation RNA ScreenTape (Agilent, Santa Clara, CA), and concentration was determined using the Qubit 2.0 Fluorometer (Thermo Fisher Scientific, Waltham, MA). Since FFPE RNA is expected to be highly degraded, a DV200 value, representing the percentage of RNA fragments greater than 200 bp, was determined for each sample. In the validation cohort, RNA sample preparation and short-read sequencing were performed as previously described ( 19 ). Preparation of bulk RNA sequencing samples Libraries for whole transcriptome RNA sequencing were prepared using the Stranded Total RNAseq with Ribo-zero Plus kit (Illumina, San Diego, CA) as per the manufacturer’s instructions, starting with an input of 100ng of RNA and 15 cycles of final PCR amplification. Library size was assessed using the 2100 Bioanalyzer and the High-Sensitivity DNA assay (Agilent, Santa Clara, CA). Concentration was determined using the Qubit Fluorometer 2.0. Next-generation sequencing with a paired-end 2x100 bp run length was performed on the NextSeq 2000 platform (Illumina, San Diego, CA). A minimum of 30M reads per sample was acquired for each sample. Preparation of Nanostring nCounter samples 100 ng of RNA fragments 200 bp and over were used for the NanoString nCounter assay (NanoString, Seattle, WA). Briefly, RNA was hybridized at 65 °C for 19 hours with proprietary fluorescently labeled reporter and capture probes to bind target genes. The hybridized mix of probes and target genes was washed, enriched, and isolated using the NanoString Prep station, and then scanned for fluorescent intensity using the NanoString Digital Analyzer. The nCounter Immunology panel (v2, Supplementary Table 3 ) was used to quantify 594 genes known to be involved in autoimmune/immune responses, along with 15 internal reference control genes to facilitate normalization. Informatic and gene expression analysis Bulk RNAseq analysis in discovery and validation cohorts RNA-seq reads were demultiplexed using bcl2fastq (Illumina, San Diego). Demultiplexed FASTQ files were aligned to GRCh38 using the Kallisto pseudo-aligner ( 20 ) using default settings. The generated BAM files were read into the R statistical computing environment, and gene counts were quantified using the GenomicAlignments package ( 21 ). Preliminary QC and differential expression analysis were performed using the R/Bioconductor package DESeq2 ( 22 ). Read counts were filtered to include only genes with a read count >=20 in greater than 33% of samples. In the discovery cohort, patients were grouped by phenotype (iMCD-TAFRO, iMCD-NOS/IPL, DLBCL, and sentinel control), and DEGs were considered for further investigation using a log 2 fold change (log 2 FC)>|1| and a Benjamini-Hochberg corrected P -value 1.5, adjusted- P < 0.05) were used to construct a gene signature for database entry, and standard settings were used to predict reversing drug perturbations. Predicted reversing drugs were grouped by mechanism of action (MOA) and were ranked by the number of significant individual drugs within that group. For each compound within each cell line, replicate P-values were collapsed into a single line-level value using the MODZ-style robust z-mean (p→z, Tukey biweight mean, z→p) previously employed ( 32 ). Line-level P-values for the same compound were then meta-analyzed across cell lines using Stouffer’s Z (equal weights), restricting to cell lines with P<0.05 to focus on lines showing significant reversal signal. Scores are reported as −log 10 P and used for ranking. We repeated the procedure after mapping compounds to MOA to obtain MOA-level scores. Results are presented in four contexts: all lines (“iMCD”), all lines restricted to immune cell lines only (“iMCD-Immune”), iMCD-TAFRO, and iMCD-TAFRO restricted to immune cell lines only (TAFRO-immune). Figures display the top hits by −log10P. Quantification of follicular clusterin expression and germinal center features for ROC-based discrimination of iMCD from non-CD controls To label Clusterin in LNs, we immunohistochemically stained for Clusterin using DAB (3,3′-diaminobenzidine) with a hematoxylin counterstain to mark nuclei. Samples overlapping between the previously described discovery and targeted NanoString cohorts and the IHC cohort are listed in Supplementary Table 2 . We then identified a DAB threshold to divide Clusterin into positive and negative regions. To analyze the Clusterin DAB and nuclear hematoxylin stains, we developed a Python-based algorithm quantify Clusterin staining intensity. We used the rgb2hed function from the Python scikit-image library to separate the DAB and hematoxylin channels. We quantified DAB stain in optical density (OD) units. We identified the LN from the background using a global Otsu threshold applied to the LN on the hematoxylin channel. This LN segmentation was used to measure the total LN area. Germinal centers (GCs) and mantle zones (MZs) were manually segmented to attain representative samples. GC, MZ, and interfollicular (IF) regions were manually segmented using the hematoxylin channel alone to blind users to the Clusterin stain. For each sample, the total number of
## Related Clinical Research

- [Obesity-associated inflammation and cancer: mechanisms linking adipose immune changes to tumor risk](https://medichelpline.com/clinical-feed/frontiers-in-immunology-6-obesity-associated-inflammation-and-cancer-molecular-mechanisms-evidence-gaps.md)
- [Probable CLIPPERS in a Patient with Cutaneous T‑Cell Lymphoma Remission: 5‑Year Follow‑Up (Report](https://medichelpline.com/clinical-feed/frontiers-in-immunology-11-case-report-probable-clippers-in-a-patient-with-cutaneous-t-cell-lymphoma-in.md)
- [Childbirth Rates Among Female Cancer Survivors in Taiwan: Nationwide Population Study](https://medichelpline.com/clinical-feed/plos-one-0-rates-of-childbirth-in-female-cancer-survivors-a-population-based-study-in.md)
- [Increased 1-Year Risk of Ischemic Stroke After Invasive Cancer Diagnosis: Nationwide Cohort Study](https://medichelpline.com/clinical-feed/pubmed-42706874.md) (DOI: 10.1002/cncr.70536)
- [TPGS-stabilized pyrimethamine micelles induce ROS-mediated apoptosis and slow tumour growth](https://medichelpline.com/clinical-feed/pubmed-42700206.md) (DOI: 10.1007/s40199-026-00656-4)

## Navigation
- [← Back to Infectious Disease Feed](https://medichelpline.com/clinical-feed/infectious-disease.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.