Graph perceiver network for lung tumor and premalignant lesion stratification from histopathology

Gindra, R. H.; Zheng, Y.; Green, E. J.; Reid, M. E.; Mazzilli, S. A.; Merrick, D. T.; Burks, E. J.; Kolachalama, V. B.; Beane, J. E.

doi:10.1101/2023.10.16.562550

bioRxiv · 10.1101/2023.10.16.562550

Graph perceiver network for lung tumor and premalignant lesion stratification from histopathology

Gindra, R. H.·Zheng, Y.·Green, E. J.·Reid, M. E.·Mazzilli, S. A.·Merrick, D. T.·Burks, E. J.·Kolachalama, V. B.·Beane, J. E.

View original source ↗Export metadata ↓

Abstract

Bronchial premalignant lesions (PMLs) precede the development of invasive lung squamous carcinoma (LUSC), posing a significant challenge in distinguishing those likely to advance to LUSC from those that might regress without intervention. In this context, we present a novel computational approach, the Graph Perceiver Network (GRAPE-Net), leveraging hematoxylin and eosin (H&E) stained whole slide images (WSIs) to stratify endobronchial biopsies of PMLs across a spectrum from normal to tumor lung tissues. GRAPE-Net outperforms existing frameworks in classification accuracy predicting LUSC, lung adenocarcinoma (LUAD), and non-tumor (normal) lung tissue on The Cancer Genome Atlas (TCGA) and Clinical Proteomic Tumor Analysis Consortium (CPTAC) datasets containing lung resection tissues while efficiently generating pathologist-aligned, class-specific heatmaps. The network was further tested using endobronchial biopsies from two data cohorts, containing normal to carcinoma in situ histology, and it demonstrated a unique capability to differentiate carcinoma in situ lung squamous PMLs based on their progression status to invasive carcinoma. The network may have utility in stratifying PMLs for chemoprevention trials or more aggressive follow-up.

Source connections

Published article ↗

Explore related subjects

pathology Preprint

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Gindra, R. H., Zheng, Y., Green, E. J., Reid, M. E., Mazzilli, S. A., Merrick, D. T., Burks, E. J., Kolachalama, V. B., Beane, J. E.. 2023-10-19. Graph perceiver network for lung tumor and premalignant lesion stratification from histopathology. https://doi.org/10.1101/2023.10.16.562550

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

bioRxiv2026

NAE1-Dependent Protein Neddylation Preserves Endothelial Identity and Vascular Integrity

Background: Endothelial dysfunction is a central driver of cardiovascular and inflammatory diseases, yet the post-translational mechanisms that preserve endothelial homeostasis remain incompletely understood. Protein neddylation, the covalent conjugation of a ubiquitin-like modifier, regulates diverse cellular processes, yet its physiological role in the vascular endothelium remains unknown. This study investigated whether protein neddylation is required to preserve endothelial identity and vascular homeostasis. Methods: We generated tamoxifen-inducible endothelial-specific Nae1 knockout mice to inhibit neddylation and combined bulk RNA sequencing, single-cell and single-nucleus transcriptomics, quantitative proteomics, biochemical analyses, and gain- and loss-of-function approaches to define the role of endothelial neddylation in vascular homeostasis and inflammatory injury. Results: Endothelial-specific Nae1 deletion caused rapid mortality associated with vascular leakage, platelet accumulation, inflammation, and multi-organ injury. Multi-omics analyses demonstrated profound loss of endothelial identity, characterized by suppression of core endothelial programs and activation of inflammatory, procoagulant, and pyroptotic pathways. Single-cell analyses revealed progressive endothelial dysfunction culminating in depletion of the endothelial population and remodeling of the vascular niche. Mechanistically, endothelial neddylation deficiency activated gasdermin D (GSDMD)- and gasdermin E (GSDME)-dependent pyroptosis, whereas dual inhibition of GSDMD and GSDME markedly attenuated inflammatory transcriptomic remodeling, vascular injury, hepatocyte death, immune cell infiltration, and platelet accumulation. Translational analyses demonstrated reduced endothelial neddylation in experimental endotoxemia and decreased expression of neddylation pathway components in human atherosclerosis and COVID-19 datasets. Conversely, restoration of endothelial neddylation partially reversed inflammatory endothelial transcriptomic reprogramming in vivo. Conclusions: NAE1-dependent protein neddylation is an essential regulator of endothelial identity and vascular integrity. Loss of endothelial neddylation promotes gasdermin-dependent pyroptosis and thrombo-inflammatory vascular injury, whereas restoration of the neddylation pathway mitigates inflammatory endothelial dysfunction. These findings identify endothelial neddylation as a fundamental mechanism maintaining vascular homeostasis and a potential therapeutic target for cardiovascular and inflammatory diseases.

pathology↗

bioRxiv2026

Non-HLA antibodies worsen the histological phenotype and prognosis of antibody mediated rejection in kidney allografts

IntroductionAntibody-mediated rejection (AMR) remains a leading cause of kidney allograft failure, with both HLA and non-HLA antibodies implicated in its pathogenesis. The contribution of non-HLA antibodies (non-HLA Abs) to microvascular inflammation (MVI) and graft outcome, particularly in cases lacking donor-specific anti-HLA antibodies (HLA-DSAs), remains incompletely understood. MethodsWe analyzed 571 post-transplant serum samples from 326 patients with histological features of AMR (AMRh) and 164 stable controls. Non-HLA Abs were detected using the previously developed Non-HLA Antibody Detection Immunoassay (NHADIA), and associations were examined with histological lesions, AMRh persistence, and graft outcomes. Biopsies were scored according to Banff 2022 criteria, and patients were stratified by HLA-DSA and NHADIA status. ResultsNHADIA values were significantly higher in AMRh patients compared to controls (P=0.0001), regardless of HLA-DSA status. NHADIA values correlated with the severity of glomerulitis, peritubular capillaritis and global MVI scores. In AMRh patients with HLA-DSAs, non-HLA Abs remained independently associated with MVI severity. Follow-up biopsies revealed persistent AMR lesions in patients with both HLA-DSAs and non-HLA Abs. Allograft survival was lowest in double-positive patients, and NHADIA positivity independently predicted graft loss (HR=2.25, 95% CI: 1.03-4.92, P=0.042). Incorporating NHADIA into the Banff classification reclassified 61% of AMRh cases as "double-positive AMRh," and identified new subgroups with significant prognostic differences. ConclusionPost-transplant detection of non-HLA antibodies identifies a distinct subset of AMR with more severe histology and worse graft prognosis, particularly when coexisting with HLA-DSAs. Integrating non-HLA Ab testing into current diagnostic frameworks may refine AMR classification and improve risk stratification. TRANSLATIONAL STATEMENTThis study highlights the clinical relevance of non-HLA antibodies, identified using our innovative endothelial cell-based assay (NHADIA), in kidney transplant recipients. Their presence is associated with more severe antibody-mediated rejection (AMR) and poorer graft outcomes, even in the absence of donor-specific HLA antibodies. Incorporating non-HLA antibody detection into routine post-transplant evaluation may allow clinicians to better identify high-risk patients, including those previously classified as DSA-negative AMR. This expanded immunological profiling refines AMR diagnosis, improves risk stratification, and opens new avenues for personalized immunosuppressive strategies, ultimately enhancing long-term graft survival and patient care.

pathology↗

bioRxiv2026

Lrat-Cre Exhibits Widespread Expression Beyond Hepatic Stellate Cells Across Multiple Tissues

Hepatic stellate cells (HSCs) play a central role in liver fibrosis, shifting from quiescent vitamin A-storing cells to activated, myofibroblast-like cells that secrete collagen and other profibrotic factors1. HSCs have thus become a major focus in liver fibrosis research, and several Cre driver lines have been created to target HSCs in mice. However, early Cre lines had significant limitations. Glial fibrillary acidic protein (Gfap)-Cre labels only a subset of HSCs and also induces recombination in cholangiocytes2. Collagen type I alpha 1 (Col1a1)-Cre and alpha-smooth muscle actin (SMA)-Cre/CreERT2 primarily label activated myofibroblasts and broadly mark portal fibroblasts and vascular smooth muscle cells3,4. Platelet-derived growth factor receptor beta (Pdgfr{beta})-Cre reliably labels HSCs but also recombines pericytes and smooth muscle cells, limiting its specificity5. The introduction of lecithin-retinol acyltransferase (Lrat)-Cre marked a major advance, offering highly specific labeling of quiescent and activated HSCs and rapidly becoming the most widely used driver for HSC tracing and genetic perturbation2. However, the extrahepatic expression of Lrat-Cre remains incompletely understood. This is a critical limitation, given that liver biology is closely coordinated with other organs to maintain systemic metabolism. Addressing these gaps is essential for the accurate interpretation of HSC-specific genetic models in liver biology.

pathology↗