Objective: To investigate the clinicopathological and molecular features of atypical spindle cell/pleomorphic lipomatous tumor (ASPLT). Methods: A retrospective analysis was conducted on nine cases of ASPLT diagnosed at the Fourth Affiliated Hospital of Soochow University and the First Hospital of Jilin University from February 2023 to September 2025. The clinical, histopathologic, immunohistochemical, and molecular genetic features were analyzed, supplemented by a comprehensive literature review. Results: The cohort comprised seven males and two females with an age of 59.0 (54.0, 63.0) years. Tumor arose in the neck, buttock, abdominal wall, face, and axilla. Histologically, the tumors exhibited a variable mixture of atypical spindle cells, adipocytes, pleomorphic/multinucleated cells, and lipoblasts embedded within a collagenous-to-myxoid stroma. Distinctive ropy collagen bundles were identified in seven cases. Mitotic activity was negligible, and tumor necrosis or dedifferentiation was consistently absent. Immunohistochemically, all cases showed diffuse CD34 expression and loss of nuclear RB1 expression. The Ki-67 proliferative index was 1%-2%, and p53 immunostaining demonstrated a wild-type expression pattern in all eight tested cases. Fluorescence in situ hybridization confirmed RB1 gene deletion in all five tested cases. Conclusions: ASPLT is a rare adipocytic neoplasm characterized by a broad morphologic spectrum, distinctive genetic alterations, and an indolent clinical course. Due to substantial morphological overlap with other lipomatous tumors, diagnosis can be challenging. Accurate identification relies on meticulous microscopic evaluation combined with ancillary testing, specifically, CD34 positivity, immunohistochemical loss of RB1 expression, and/or molecular detection of RB1 gene deletion, to ensure correct classification and prevent overtreatment.

Objective: To investigate the molecular mechanisms by which secretory carrier membrane protein 2 (SCAMP2) contributes to colorectal cancer cell proliferation via regulating early growth response factor 1 (EGR1). Methods: Ten pairs of pathologically confirmed colorectal cancer tissues and adjacent non-tumor tissues were collected at the Affiliated Hospital of Qingdao University between January 2022 and March 2023. The expression of SCAMP2 was examined using Western blot and immunohistochemistry. SW480 and HT29 cell lines were transfected with small interfering RNA (siRNA), short hairpin RNA (shRNA) and SCAMP2/EGR1 overexpression plasmids to establish cell models with SCAMP2 or EGR1 overexpression and knockdown. Cell counting kit-8 (CCK8) assay, flow cytometry and nude mouse subcutaneous tumorigenesis assay were performed to evaluate the effects of SCAMP2 expression on the proliferation and apoptosis of colorectal cancer cells. Transcriptome sequencing was performed to identify early growth response factor 1 (EGR1) as the downstream regulatory factor of SCAMP2. Rescue experiments were used to confirm the interaction between SCAMP2 and EGR. Results: Western blot demonstrated that the expression level of SCAMP2 in 7 pairs of colorectal cancer tissues was lower than that in adjacent normal intestinal tissues. Immunohistochemistry revealed that SCAMP2 showed strongly positive staining (3+) in normal intestinal epithelium, but negative/weakly positive staining (1+) in colorectal cancer. Consistently, EGR1 exhibited strongly positive immunoreactivity (3+) in normal intestinal epithelium, but negative/weak positive staining (1+) in colorectal cancer. In vitro cellular experiments confirmed that overexpression of SCAMP2 or EGR1 significantly suppressed the proliferation of colorectal cancer cells (all P<0.05). Furthermore, the nude mouse xenograft tumor model showed that SCAMP2 overexpression remarkably inhibited tumor growth in vivo (P<0.05). Transcriptome sequencing showed that the mRNA expression level of EGR1 was significantly up-regulated in SCAMP2-overexpressing colorectal cancer cells (P<0.05). Rescue experiments illustrated that knockdown of EGR1 partially reversed the inhibitory effects on cell proliferation and pro-apoptotic effects induced by SCAMP2 overexpression. Conversely, overexpression of EGR1 partially abrogated the enhanced proliferation and reduced apoptosis triggered by SCAMP2 knockdown. Conclusions: SCAMP2 expression is downregulated in colorectal cancer. Low SCAMP2 expression may impair EGR1 signaling, thereby suppressing apoptosis and promoting colorectal cancer cell proliferation. The SCAMP2/EGR1 signaling axis plays a crucial role in colorectal cancer progression.

Objective: To investigate the associations of aberrant CD10 expression with clinicopathological features and prognosis of marginal zone lymphoma (MZL). Methods: A retrospective analysis was conducted on 13 cases of MZL with aberrant CD10 expression, diagnosed at the First Affiliated Hospital of Zhengzhou University, Zhengzhou, China between April 2015 and July 2024. Clinical features, morphological findings, immunohistochemical profiles, and molecular testing results were analyzed. The patients were also assessed via telephone follow-up. Results: There were 7 male and 6 female patients, aged 55.0 (51.5,70.0) years on average. Subtypes comprised 10 extranodal MZLs of mucosa-associated lymphoid tissue (EMZL), 2 splenic MZLs (SMZL), and 1 nodal MZL (NMZL). Histologically, tumor cells exhibited diffuse (10 cases) or nodular (3 cases) growth patterns. In all three nodular cases, the proportion of CD10-positive cells was 50% or higher. Under high-power magnification, most cases (12 of 13 cases) were dominated by proliferation of small lymphocyte-like cells. One case was characterized by a mixed proliferation of monocytoid B cells and small lymphocyte-like cells. Histological features included follicular colonization (8 cases) and lymphoepithelial lesions (2 gastric cases and 2 pulmonary cases). Plasma cell differentiation was observed in 10 cases (9 EMZLs and 1 NMZL) and scattered interstitial hemosiderin deposits were seen in 7 cases (5 EMZLs and 2 SMZLs). Immunophenotypically, tumor cells expressed CD20 (13/13 cases), CD79α (11/13 cases), and bcl-2 (12/13 cases), whereas CD3, bcl-6, and Cyclin D1 were all negative. All 13 cases were positive for CD10, with the proportion of positive cells ranging from 5% to 80%. Among them, 6 cases exhibited CD10 positivity in less than 50% of tumor cells, and 7 cases showed CD10 positivity in 50% or more of tumor cells. The Ki-67 proliferation index ranged from approximately 5% to 30%. Molecular testing revealed monoclonal immunoglobulin gene rearrangement in 2 cases, MALT-1 gene break in one EMZL case, and no bcl-2 gene break in 10 MZL cases. Follow-up durations ranged from 11 to 122 months. Favorable post-treatment outcomes were observed in all 13 patients. Conclusions: MZL with aberrant CD10 expression is a rare variant, demonstrating clinicopathological features similar to conventional MZL and a generally favorable prognosis. Its immunophenotype may mimic other CD10-positive small B-cell lymphomas, particularly follicular lymphoma. An accurate diagnosis appears to require integration of histologic, immunophenotypic, and molecular findings to avoid misdiagnosis.

Objective: To investigate the clinicopathological features, immunophenotype, and molecular characteristics of sweat gland carcinoma with neuroendocrine differentiation (SCAND). Methods: Three cases of SCAND diagnosed at the Department of Pathology, Zhongshan Hospital, Fudan University, Shanghai, China from June 2022 to June 2025 were collected. Hematoxylin-eosin (HE) staining and immunohistochemistry (EnVision method) were performed to characterize morphologic and immunophenotypic features. Next-generation sequencing (NGS) was utilized to explore their molecular pathological characteristics. Results: All three patients were male, aged 67, 63, and 48 years, respectively. The primary tumors presented as solitary nodules or masses located in the lower abdomen, anterior chest wall, or inguinal region, with a maximum diameter of 2.0, 1.3, and 1.8 cm, respectively. Histologically, the tumors were situated primarily within the dermis, invading the epidermis and subcutaneous tissue, with focal mucin production. The tumor cells were arranged in nests, cords, and sieve-like patterns, exhibiting mild to moderate atypia. The nuclei were round to oval with coarsely granular chromatin and relatively inconspicuous nucleoli, accompanied by eosinophilic cytoplasm. Mitotic figures were scarce. All cases were positive for CK7, GATA3, TRPS1, Ber-EP4, epithelial membrane antigen, estrogen receptor, progesterone receptor, and the neuroendocrine markers of synaptophysin and chromogranin A. The tumors were negative for p40 and HER2. The Ki-67 proliferation index ranged from 2% to 30%. Multiple class Ⅲ and Ⅳ variants were identified by NGS analysis in cases 1 and 2. In case 3, a class Ⅱ variant, specifically an SF3B1 missense mutation, was identified along with multiple class Ⅲ and Ⅳ variants. Additionally, multiple germline variants were detected in all three cases, while none of them were classified as pathogenic or likely pathogenic. The patients were followed up for 13, 37 and 155 months, respectively. Metastases were detected 6, 36, and 96 months after diagnosis. Case 2 showed local metastasis, while cases 1 and 3 developed multiple nodal and bone metastases. Conclusions: SCAND is a rare cutaneous adnexal neoplasm characterized by co-expression of sweat gland and neuroendocrine markers. Despite low-grade morphology, it follows a non-indolent clinical course warranting close clinical attention.

Polycythaemia vera (PV) may rarely evolve into acute erythroid leukaemia (AEL). We herein report a known case of JAK2 p. V617F mutated PV presenting with progressive hepatosplenomegaly, abdominothoracic lymphadenopathy, ascites and various haematological derangements. Rapid diagnostic work-up revealed extensive marrow, hepatic and ascitic fluid infiltration by many CD71+, LMO2+ and p53+ proerythroblasts. Next generation sequencing showed TP53, ASXL1 and TET2 gene mutations. A chemotherapeutic regimen of ruxolitinib, azacitidine and venetoclax was initiated. A comprehensive review of the English literature emphasises clinical, morphologic, immunophenotypic and genetic characteristics of AEL transformation in PV.

Canine mammary tumours (CMTs) are one of the most frequently observed malignancies in female dogs. Many studies have investigated epigenetic changes in CMT at the nuclear level, yet the role of mtDNA methylation, particularly in the D-loop region, remains largely unexplored. Given the significant importance of the D-loop region in mtDNA replication and transcription, and the previous identification of CpG enrichment there, this preliminary study aimed to assess apparent methylation signals in the mitochondrial D-loop region. We examined CpG, CHG and CHH methylation contexts in matched tumour and normal tissues from six German Shepherd dogs with histologically classified mammary carcinomas using enzymatic methyl-sequencing and high-throughput amplicon sequencing. Overall, CpG and CHH contexts exhibited greater variability between tumour and normal tissues than CHG, which remained relatively stable. Correlation analyses revealed a putative positive relationship between differences in CpG and CHH methylation (Spearman's r = 0.829, p = 0.042). This might suggest a coordinated epigenetic modulation of these two methylation patterns in the carcinogenesis process, yet due to the limited study group (n = 6) the hypothesis should be verified on a larger cohort. These findings demonstrate a plausible existence of intra-individual epigenetic divergence at the mitochondrial level. To our knowledge, this is the first study to assess the influence of mtDNA methylation on CMT, and it supports further investigation of mitochondrial epigenetics in canine oncology.

NLRC3 serves as a critical negative regulator of inflammatory signalling pathways and is implicated in the pathogenesis of multiple disorders, including inflammatory diseases and malignancies. However, the specific functions and underlying mechanisms of NLRC3 in hepatocellular carcinoma (HCC) development remain largely elusive. Here, we report that the expression of NLRC3 was downregulated in HCC tissues and was associated with poor prognosis. Mice deficient in NLRC3 were hypersusceptible to chemically induced hepatocarcinogenesis, and NLRC3 exerted a tumour-suppressive function during HCC development by inhibiting the TRAF6/MYC/NF-κB signalling cascade. Mechanistically, the NLRC3-HSP90 interaction competitively inhibited the binding of the HSP90 chaperone to TRAF6/MYC and impaired the K63-linked ubiquitination and stabilization of the TRAF6/MYC proteins. In the absence of NLRC3, TRAF6 and MYC reciprocally regulated the expression of the other and triggered a TRAF6-MYC positive feedback loop that accelerated the progression of HCC. Our study reveals a novel functional link between NLRC3 and MYC signalling and indicates that HSP90 is a potential target for hindering the progression of NLRC3-deficient HCC.

The application of Multiple Instance Learning (MIL) for classifying Whole Slide Images (WSIs) has gained extensive use in recent years, primarily due to the high cost and time consumption associated with pixel-level annotation of WSIs, which is challenging to accomplish. The advancements in MIL for WSIs have predominantly concentrated on two fronts: the development of superior feature extractors (for instance, utilizing self-supervised learning for training feature extractors) and the formulation of enhanced instance aggregation strategies. Regrettably, the majority of the most advanced approaches have neglected phenotypic variances among instances when employing attention mechanisms. To capitalize on the disparities between instance tissues, we have introduced a phenotypic self-distillation approach to MIL. Our framework is composed of three components: i) a self-supervised feature extractor based on contrastive learning and a phenotype extractor pre-trained on the Kather100K dataset, which automatically provides 9-class tissue phenotype labels (e.g., tumor epithelium, stroma, lymphocytes) without requiring manual annotation, ii) the incorporation of a self-distillation loss between the features of instances and their phenotypes to augment the informational content of both perspectives, and iii) the aggregation of MIL instances for the final MSI prediction. The efficacy of this framework was evaluated on two datasets: the TCGA-CRC dataset was used for training and internal testing with a fixed 70%/30% split, while the Tangshan People's Hospital cohort served as an independent external validation set. On the TCGA-CRC dataset (n = 360; 65 MSI-H, 295 MSS), our model achieved an AUC of 0.8846 and an accuracy of 0.84, using a fixed 70%/30% train-test split. On the Tangshan People's Hospital dataset (n = 472; 56 MSI-H, 426 MSS), the model attained an AUC of 0.7258 and an accuracy of 0.70.

The effect of genetic polymorphisms on survival outcomes in nasopharyngeal carcinoma (NPC) has not been fully elucidated. This study aimed to identify genetic variants associated with NPC survival using a genome-wide association study (GWAS) approach.

Fusobacterium nucleatum (Fn) drives chemotherapy resistance in colorectal cancer (CRC) by activating hypoxia-mimicking signaling pathways, yet the molecular mechanisms linking Fn to 5-fluorouracil (5-FU) resistance remain unclear. We investigated how Fn-induced hypoxia-related gene expression contributes to 5-FU resistance and identified a pharmacological strategy to overcome it.

Adoptive cell transfer of tumor infiltrating CD4+ T cells or T cells genetically modified to express antitumor CD4+ T cell receptors (TCRs) has shown clinical efficacy in patients with advanced solid tumors. We previously defined unique transcriptomic states of antitumor, neoantigen-specific CD8+ and CD4+ tumor infiltrating lymphocytes (TIL) in human cancer. However, optimal cell surface protein markers that identify and would enable viable cell isolation and selective enrichment of antitumor CD4+ T cells have yet to be identified. Here, we define transcriptomic and corresponding cell surface protein profiles of antitumor CD4+ TIL within metastatic colorectal tumors using high dimensional single cell transcriptomic and proteomic analysis. Comprehensive analysis of 45 antitumor CD4+ TCR clonotypes demonstrated uniquely high cell surface protein expression of ADGRG1, CD86, and CD57. In all samples containing tumor neoantigen-reactive CD4+ TIL clonotypes, ADGRG1- and CD86-based cell surface enrichment of known reactive TCR clonotypes were 11-fold and 9-fold higher than bulk CD4+ TIL, respectively. ADGRG1- and CD86-based cell surface enrichment identified >60 validated antitumor CD4+ TCR-clonotypes including tumor organoid-reactive TCRs, as well as neoantigen-specific TCRs targeting private tumor mutations and cancer driver mutations. While both ADGRG1 and CD86 enriched CD4+ TIL displayed an exhausted neoantigen-specific gene expression signature, CD86 also enriched for cytotoxic CD4+ TIL phenotypes. CD4+ TIL isolated based on ADGRG1 expression and expanded in vitro also demonstrated neoantigen reactivity in two samples. Thus, ADGRG1 and CD86 appear to be effective protein markers for isolating human tumor-specific CD4+ TCR clonotypes and studying antitumor CD4+ TIL.

Liver cancer is the fourth leading cause of global cancer mortality, with hepatocellular carcinoma (HCC) accounting for most primary liver cancers. The nucleolus, as the primary site of ribosome biogenesis, plays a crucial role in HCC development. Here, we report that the nucleolar protein KRR1 interacting protein (KRI1) is aberrantly overexpressed in HCC tissues and interacts with nucleophosmin 1 (NPM1) via phase separation to maintain nucleolar structure and ribosome biogenesis. Downregulation of KRI1 significantly impairs HCC cell proliferation and increases apoptosis, accompanied by disrupted nucleolar structure and ribosome biogenesis. Moreover, we show that KRI1 phase separation is enhanced by the phosphorylation of three serine residues (S94, S95, and S97) in its N-terminal intrinsically disordered region, and pyruvate kinase M2 (PKM2) is identified as the key kinase. PKM2 inhibitor shikonin markedly attenuates KRI1 phase separation and its interaction with NPM1, disrupts nucleolar structure and ribosome biogenesis, thereby inhibiting HCC progression. This study uncovers the crucial role of KRI1 in HCC progression and provides theoretical and experimental evidence for the development of nucleolar protein-targeted therapeutic strategies for HCC treatment.

Non-small cell lung cancer (NSCLC) exhibits stem-like characteristics that drive tumor aggressiveness and treatment resistance. The molecular chaperone Glucose-Regulated Protein 78 (GRP78) is substantially elevated in NSCLC compared to normal tissues and cell lines. In clinical samples, GRP78 protein levels correlated with advanced tumor stage and lymph node metastasis. Pharmacological inhibition of GRP78 using BAPTA-AM and GRP78-IN-3 significantly suppressed NSCLC cell viability while promoting apoptosis. The inhibitors effectively attenuated cell migration by downregulating matrix metalloproteinases and reversed epithelial-mesenchymal transition by modulating key cadherin expression profiles. Importantly, GRP78 inhibition substantially reduced cancer stem cell markers and impaired spheroid formation capability. Genetic knockdown of GRP78 in A549 and H1975 cells recapitulated the effects of pharmacological inhibition, reducing cell migration, reversing EMT, attenuating stemness, and suppressing β-catenin signaling. Conversely, GRP78 overexpression in A549 and H1975 cells induced opposite effects, promoting mesenchymal markers, enhancing stemness proteins, and increasing spheroid size. Mechanistic studies revealed that this anti-tumor activity operates through suppression of β-catenin signaling mediated by SIX1, as evidenced by SIX1-overexpression rescue experiments. These findings demonstrate that GRP78 acts as a key regulator of NSCLC stemness and metastatic potential by coordinating epithelial plasticity and stemness pathways, positioning it as a promising therapeutic target for NSCLC treatment.

Advances in spatial transcriptomics (ST) technologies enable spatially resolved gene expression profiling, yet most platforms remain limited to spot-level resolution, where each measurement aggregates signals from multiple cells and obscures cell-specific programs and microenvironmental interactions. Existing computational approaches either provide limited sub-spot refinement or rely on matched single-cell RNA sequencing references that are costly and difficult to obtain. Here, we present DeSpaST (Deconvoluting Spatial Transcriptomics Signals to Cell-Level Resolution Using Histology Images), a dynamic edge-conditioned graph convolutional network that deconvolves spot-level ST data into cell-level gene expression profiles using only paired histology images. DeSpaST extracts nucleus-level morphological features, constructs directed cellular interaction graphs, and integrates spatial and transcriptional information through message passing. Validated across four cancer datasets with orthogonal Xenium, immunofluorescence, ablation, and interslice evaluations, DeSpaST enhances spatial resolution, facilitates downstream cellular-level analyses, and provides deeper insights into tissue biology.