Proteomic analysis of pericardial effusions in hematopoietic stem cell transplant recipients

Hematopoietic stem cell transplant (HSCT) is an established treatment for malignant and non-malignant disorders in children and young adults. Pericardial effusion (PEF) is a potentially life-threatening complication of HSCT that occurs in up to 19% of pediatric patients and is associated with increased mortality due to cardiac tamponade.^1-3 In our clinical practice, most clinically significant PEF occur in conjunction with transplant-associated thrombotic microangiopathy (TA-TMA), though the exact mechanism remains unclear.⁴ Other potential risk factors for developing PEF after HSCT include myeloablative conditioning, graft-versus-host-disease (GVHD), GVHD prophylaxis, Epstein-Barr virus viremia, and abnormal pre-transplant cardiac function testing.^2-7 Knowledge of PEF mechanisms would guide the use of novel targeted therapies and, combined with diligent screening, may prevent the need for pericardial drain placement in some patients. Our objective was to characterize potential mechanistic pathways in the proteome of pericardial fluid from pediatric HSCT recipients with PEF. We obtained permission from our institutional review board to retrospectively review the clinical course of seven patients with TA-TMA and PEF with pericardial fluid samples stored in our biorepository. Transplantation demographics, details of TA-TMA diagnosis and management, and details of PEF management and clinical course were collected. Pericardial fluid from these seven HSCT recipients and four control autopsy pericardial fluid samples from available pediatric patients who died from non-cardiac causes were analyzed using an aptamer-based proteomics (SomaScan^® 11K Assay) platform. Due to institutional review board restrictions, we were unable to ascertain any other cause of death details in the non-HSCT controls. TA-TMA was prospectively diagnosed using laboratory and clinical diagnostic criteria previously published by Jodele et al.⁸ Patient demographics, transplant-related characteristics and the clinical courses of the seven patients with PEF are described in Table 1. All patients were diagnosed with TATMA, and the diagnosis was made at a median time of 21 days (range, 2-66 days) from HSCT. Cardiac assessments revealed predominantly sinus rhythm in pre-transplant electrocardiogram (ECG) results, with one instance of sinus tachycardia. Pre-transplant ECG findings included dilated aortic root, mildly thickened mitral valve, and a patent foramen ovale. Pericardial fluid sample collection with drain placement occurred between 4 to 224 days after HSCT following diagnosis by echocardiography. Eculizumab was given to all patients for TA-TMA-directed therapy. Six patients required placement of a pericardial drain and two of those six required placements of a second drain due to fluid re-accumulation. No PEF was classified as malignant on clinical pathology review.

We wanted to characterize potential mechanistic pathways in the proteome of pericardial fluid from HSCT recipients with PEF. Analysis of gene expression and pathway activation revealed significant findings when comparing the post-HSCT PEF cases with normal pericardial fluid collected at autopsy for non-cardiac deaths. A total of 1,271 differentially expressed proteins (DEP) were identified with a P value ≤0.05. The volcano plot in Figure 1 illustrates the log₂ fold change against the -log₁₀ P value, highlighting several genes with notable expression changes, including CFHR5, SAA1, LEPROT, APOB, FTMT, TNFRSF25, HSD3B2, M1AP and LUM, suggesting potential involvement in the underlying biological processes leading to PEF. Twenty-seven proteins met more stringent differential expression criteria with an adjusted P value (P_adj) ≤0.05 (Figure 1). A list of the first 100 DEP is shown in the Online Supplementary Table S1.

DEP were then entered in Qiagen Ingenuity Pathway Analysis (IPA) software using a P value cutoff of 0.05 and absolute log fold-change cutoff of 0.75 (N=785 proteins). Enriched pathways are shown in Figure 2A using an absolute z-score cutoff of 1.5. The top enriched pathway by P value was regulation of insulin-like growth factor transport and uptake by insulin-like growth factor binding proteins (IGFBP; P=2.2e-14, z-score=4.3). IGFBP have been previously described as diagnostic markers in malignant pleural and peritoneal effusions,⁹ though interestingly none of the effusions in our study were malignant.¹⁰ IGFBP have a recognized role in cell senescence which has been described in endothelial disorders and kidney disease.^11,12 Sensescence occurs following tissue injury and associated inflammation, therefore the IGFBP enrichment observed may support endothelial and/or leukocyte senescence as a contributing factor to PEF biology in HSCT recipients.

Other highly enriched pathways were related to inflammation, including the pathogen-induced cytokine storm signaling pathway, interleukin (IL)-17 signaling, IL-17A signaling in airway cells, IL-17A signaling in fibroblasts, acute-phase response signaling, and complement and coagulation cascades. IPA also generates leading upstream regulators of differentially expressed proteins and identified complement factor 5 (C5; P=3.2e-10, z-score=3.3), complement C5a receptor 1 (C5AR1; P=1.1e-5, z-score=3.3), IL-6 (P=5.9e-16, z-score=3.1) and IL-17RA (P=1.6e-7, z-score=2.9) as leading activated upstream regulators. Complement-related upstream regulator protein networks are shown in Figure 2B, C. The observed protein, pathway and upstream regulator changes highlight a novel and complex interplay of IL-17 and complement-related pathways in HSCT recipient PEF. Complement factors are known to affect the IL-17 axis, which makes it plausible that these complement and IL-17 findings are both true and potentially related.¹³

Table 1.Patient demographics and transplant characteristics.

Clinically, we have previously reported that PEF after HSCT frequently occurs in the setting of TA-TMA.¹⁴ Based on this, we focused our analysis on pathways relevant to TA-TMA, specifically complement and coagulation cascades, and identified IL-17 as a potential upstream mediator. This is biologically plausible, as IL-17 is known to promote endothelial activation, enhance vascular permeability, and prime the endothelium for pro-thrombotic states-hallmarks of TA-TMA. IL-17 also promotes endothelial cell senescence, which creates a mechanistic link to IGFBP dicussed above.¹⁵ Additionally, IL-17 has been shown to upregulate C3 expression and drive terminal complement activation in both endothelial and epithelial cells, supporting its role in complement-mediated endothelial injury.¹³ Consistent with this, we found that CFHR5, was one of the strongest differentially expressed proteins in our experiment. This finding aligns with the broader enrichment of complement and coagulation pathways observed in Figure 2, further implicating IL-17 driven complement dysregulation in the pathogenesis of PEF after HSCT. CFHR5 functions to protect the body from complement-mediated injury and mutations in this gene have been linked to CFHR5 nephropathy as well as TA-TMA.^16,17 The observed increase in CFHR5 may represent a host protective response to overactive complement activation in TA-TMA. Alternatively, this could indicate an underlying functional polymorphism in CFHR5 that may predispose these patients to TA-TMA and even PEF. Our prior study of complement gene polymorphisms in TA-TMA identified four TA-TMA patients with CFHR5 polymorphisms which was the second most commonly mutated gene observed.¹⁸ The association between IL-17, IGFBP, complement proteins and PEF therefore merits further study.

The data presented in this report are important for clinicians managing pediatric HSCT recipients, as PEF is a known complication with significant morbidity. All HSCT recipients in this study were diagnosed with TA-TMA and treated with a C5 inhibitor, therefore the effect of this therapy on PEF after HSCT is not testable in our study cohort. However, complement pathway enrichment was still strong in the pericardial fluid proteome despite C5 inhibition. Our novel observations involving IL-17 and IG-FBP pathways therefore merit further mechanistic study as potential targets of pharmacologic intervention outside of the complement system. IL-17 and IGFBP pathways may represent a unique mechanism of inflammation in the pericardial space of HSCT recipients that is linked to complement system activation. We acknowledge there are limitations to this study including the small sample size and use of normal pericardial fluid and not PEF from non-HSCT subjects as controls. While this limits the ability to differentiate the mechanism of HSCT PEF from non-HSCT PEF, we were still able to identify differentially expression proteins and enriched pathways in HSCT PEF. Pericardial fluid specimens from HSCT recipients without effusions were understandably not available.

Figure 1.Volcano plot of differentially expressed proteins in pericardial fluid from hematopoietic stem cell transplant recipients with pericardial effusions. This volcano plot displays the log₂ fold change (x-axis) versus -log₁₀ P value (y-axis) for all proteins analyzed. A total of 1,271 proteins were differentially expressed with P≤0.05. Red and green dots indicate significantly upregulated and downregulated proteins, respectively. Several proteins demonstrated marked differential expression highlighting potential mechanistic pathways. A subset of 27 differentially expressed proteins met stricter criteria with adjusted P values ≤0.05.

Figure 2.Pathway enrichment and upstream regulator analysis of differentially expressed proteins in pericardial fluid from hematopoietic stem cell transplant recipients with pericardial effusions. (A) Top enriched canonical pathways identified by Ingenuity Pathway Analysis from 785 proteins with P≤0.05 and absolute log₂ fold change ≥0.75. Pathways are ranked by -log(P value), and activity is predicted by z-score (orange: activated, blue: inhibited). The most significantly enriched pathway was “Regulation of insulin-like growth factor transport and uptake by insulin-like growth factor binding proteins,” followed by several inflammation-related pathways including interleukin (IL)-17 signaling, cytokine storm signaling, acute phase response, and complement cascade. (B, C) Predicted upstream regulator networks derived from Qiagen Ingenuity Pathway Analysis. Complement factors and cytokines were identified as leading upstream regulators, including C5, C5AR1, IL6, and IL-17RA. Network visualizations show predicted regulatory relationships and directionality of protein expression changes. These findings suggest a central role for complement and IL-17 pathways in the pathogenesis of pericardial effusions following hematopoietic stem cell transplant.

In conclusion, rapidly growing PEF can quickly lead to life-threatening complications and the need for invasive procedures. Our study is the first to shed light on the mechanisms of PEF in HSCT recipients and identified targetable pathways and proteins for future study and validation.

Footnotes

• Received September 3, 2025
• Accepted October 31, 2025

Correspondence

J. KOO - Jane.koo@cchmc.org

Disclosures

No conflicts of interest to disclose.

Contributions