Explainable coronary artery disease prediction model based on AutoGluon from AutoML framework.

Objective: This study focuses on the innovative application of Automated Machine Learning (AutoML) technology in cardiovascular medicine to construct an explainable Coronary Artery Disease (CAD) prediction model to support the clinical diagnosis of CAD. Methods: This study utilizes a combined data set of five public data sets related to CAD. An ensemble model is constructed using the AutoML open-source framework AutoGluon to evaluate the feasibility of AutoML in constructing a disease prediction model in cardiovascular medicine. The performance of the ensemble model is compared against individual baseline models. Finally, the disease prediction ensemble model is explained using SHapley Additive exPlanations (SHAP). Results: The experimental results show that the AutoGluon-based ensemble model performs better than the individual baseline models in predicting CAD. It achieved an accuracy of 0.9167 and an AUC of 0.9562 in 4-fold cross-bagging. SHAP measures the importance of each feature to the prediction of the model and explains the prediction results of the model. Conclusion: This study demonstrates the feasibility and efficacy of AutoML technology in cardiovascular medicine and highlights its potential in disease prediction. AutoML reduces the barriers to model building and significantly improves prediction accuracy. Additionally, the integration of SHAP enhances model transparency and explainability, which is critical to ensuring model credibility and widespread adoption in cardiovascular medicine.

Building Block-Centric Approach to DNA-Encoded Library Design.

DNA-encoded library technology grants access to nearly infinite opportunities to explore the chemical structure space for drug discovery. Successful navigation depends on the design and synthesis of libraries with appropriate physicochemical properties (PCPs) and structural diversity while aligning with practical considerations. To this end, we analyze combinatorial library design constraints including the number of chemistry cycles, bond construction strategies, and building block (BB) class selection in pursuit of ideal library designs. We compare two-cycle library designs (amino acid + carboxylic acid, primary amine + carboxylic acid) in the context of PCPs and chemical space coverage, given different BB selection strategies and constraints. We find that broad availability of amines and acids is essential for enabling the widest exploration of chemical space. Surprisingly, cost is not a driving factor, and virtually, the same chemical space can be explored with "budget" BBs.

Immune cell patterns before and after neoadjuvant immune checkpoint blockade combined with chemoradiotherapy in locally advanced esophageal squamous cell carcinoma.

BACKGROUND: Neoadjuvant immune checkpoint blockade (ICB) combined with chemoradiotherapy offers high pathologic complete response (pCR) rate for patients with locally advanced esophageal squamous cell carcinomas (ESCC). But the dynamic tumor immune microenvironment modulated by such neoadjuvant therapy remains unclear. PATIENTS AND METHODS: A total of 41 patients with locally advanced ESCC were recruited. All patients received neoadjuvant toripalimab combined with concurrent chemoradiotherapy. Matched pre- and post-treatment tissues were obtained for fluorescent multiplex immunohistochemistry (mIHC) and IHC analyses. The densities and spatial distributions of immune cells were determined by HALO modules. The differences of immune cell patterns before and after neoadjuvant treatment were investigated. RESULTS: In the pre-treatment tissues, more stromal CD3 + FoxP3 + Tregs and CD86+/CD163 + macrophages were observed in patients with residual tumor existed in the resected lymph nodes (pN1), compared with patients with pCR. The majority of macrophages were distributed in close proximity to tumor nest in pN1 patients. In the post-treatment tissues, pCR patients had less CD86 + cell infiltration, whereas higher CD86 + cell density was significantly associated with higher tumor regression grades (TRG) in non-pCR patients. When comparing the paired pre- and post-treatment samples, heterogeneous therapy-associated immune cell patterns were found. Upon to the treatment, CD3 + T lymphocytes were slightly increased in pCR patients, but markedly decreased in non-pCR patients. In contrast, a noticeable increase and a less obvious decrease of CD86 + cell infiltration were respectively depicted in non-pCR and pCR patients. Furthermore, opposite trends of the treatment-induced alterations of CD8 + and CD15 + cell infiltrations were observed between pN0 and pN1 patients. CONCLUSIONS: Collectively, our data demonstrate a comprehensive picture of tumor immune landscape before and after neoadjuvant ICB combined with chemoradiotherapy in ESCC. The infiltration of CD86 + macrophage may serve as an unfavorable indicator for neoadjuvant toripalimab combined with chemoradiotherapy.

Pan-Cancer Proteomics Analysis Reveals Wiskott-Aldrich Syndrome Protein as a Potential Regulator of Programmed Death-Ligand 1.

The programmed death-ligand 1 (PD-L1) is a key mediator of immunosuppression in the tumor microenvironment. The expression of PD-L1 in cancer cells is useful for the clinical determination of an immune checkpoint blockade (ICB). However, the regulatory mechanism of the PD-L1 abundance remains incompletely understood. Here, we integrated the proteomics of 52 patients with solid tumors and examined immune cell infiltration to reveal PD-L1-related regulatory modules. Wiskott-Aldrich syndrome protein (WASP) was identified as a potential regulator of PD-L1 transcription. In two independent cohorts containing 164 cancer patients, WASP expression was significantly associated with PD-L1. High WASP expression contributed to immunosuppressive cell composition, including cells positive for immune checkpoints (PD1, CTLA4, TIGIT, and TIM3), FoxP3+ Treg cells, and CD163+ tumor-associated macrophages. Overexpression of WASP increased, whereas knockdown of WASP decreased the protein level of PD-L1 in cancer cells without alteration of PD-L1 protein stability. The WASP-mediated cell migration and invasion were markedly attenuated by the silence of PD-L1. Collectively, our data suggest that WASP is a potential regulator of PD-L1 and the WASP/PD-L1 axis is responsible for cell migration and an immunosuppressive microenvironment.

HnRNPR-mediated UPF3B mRNA splicing drives hepatocellular carcinoma metastasis.

INTRODUCTION: Abnormal alternative splicing (AS) contributes to aggressive intrahepatic invasion and metastatic spread, leading to the high lethality of hepatocellular carcinoma (HCC). OBJECTIVES: This study aims to investigate the functional implications of UPF3B-S (a truncated oncogenic splice variant) in HCC metastasis. METHODS: Basescope assay was performed to analyze the expression of UPF3B-S mRNA in tissues and cells. RNA immunoprecipitation, and in vitro and in vivo models were used to explore the role of UPF3B-S and the underlying mechanisms. RESULTS: We show that splicing factor HnRNPR binds to the pre-mRNA of UPF3B via its RRM2 domain to generate an exon 8 exclusion truncated splice variant UPF3B-S. High expression of UPF3B-S is correlated with tumor metastasis and unfavorable overall survival in patients with HCC. The knockdown of UPF3B-S markedly suppresses the invasive and migratory capacities of HCC cells in vitro and in vivo. Mechanistically, UPF3B-S protein targets the 3'-UTR of CDH1 mRNA to enhance the degradation of CDH1 mRNA, which results in the downregulation of E-cadherin and the activation of epithelial-mesenchymal transition. Overexpression of UPF3B-S enhances the dephosphorylation of LATS1 and the nuclear accumulation of YAP1 to trigger the Hippo signaling pathway. CONCLUSION: Our findings suggest that HnRNPR-induced UPF3B-S promotes HCC invasion and metastasis by exhausting CDH1 mRNA and modulating YAP1-Hippo signaling. UPF3B-S could potentially serve as a promising biomarker for the clinical management of invasive HCC.

End to end vision transformer architecture for brain stroke assessment based on multi-slice classification and localization using computed tomography.

BACKGROUND: Brain stroke is a leading cause of disability and death worldwide, and early diagnosis and treatment are critical to improving patient outcomes. Current stroke diagnosis methods are subjective and prone to errors, as radiologists rely on manual selection of the most important CT slice. This highlights the need for more accurate and reliable automated brain stroke diagnosis and localization methods to improve patient outcomes. PURPOSE: In this study, we aimed to enhance the vision transformer architecture for the multi-slice classification of CT scans of each patient into three categories, including Normal, Infarction, Hemorrhage, and patient-wise stroke localization, based on end-to-end vision transformer architecture. This framework can provide an automated, objective, and consistent approach to stroke diagnosis and localization, enabling personalized treatment plans based on the location and extent of the stroke. METHODS: We modified the Vision Transformer (ViT) in combination with neural network layers for the multi-slice classification of brain CT scans of each patient into normal, infarction, and hemorrhage classes. For stroke localization, we used the ViT architecture and convolutional neural network layers to detect stroke and localize it by bounding boxes for infarction and hemorrhage regions in a patient-wise manner based on multi slices. RESULTS: Our proposed framework achieved an overall accuracy of 87.51% in classifying brain CT scan slices and showed high precision in localizing the stroke patient-wise. Our results demonstrate the potential of our method for accurate and reliable stroke diagnosis and localization. CONCLUSION: Our study enhanced ViT architecture for automated stroke diagnosis and localization using brain CT scans, which could have significant implications for stroke management and treatment. The use of deep learning algorithms can provide a more objective and consistent approach to stroke diagnosis and potentially enable personalized treatment plans based on the location and extent of the stroke. Further studies are needed to validate our method on larger and more diverse datasets and to explore its clinical utility in real-world settings.

Building Block-Based Binding Predictions for DNA-Encoded Libraries.

DNA-encoded libraries (DELs) provide the means to make and screen millions of diverse compounds against a target of interest in a single experiment. However, despite producing large volumes of binding data at a relatively low cost, the DEL selection process is susceptible to noise, necessitating computational follow-up to increase signal-to-noise ratios. In this work, we present a set of informatics tools to employ data from prior DEL screen(s) to gain information about which building blocks are most likely to be productive when designing new DELs for the same target. We demonstrate that similar building blocks have similar probabilities of forming compounds that bind. We then build a model from the inference that the combined behavior of individual building blocks is predictive of whether an overall compound binds. We illustrate our approach on a set of three-cycle OpenDEL libraries screened against soluble epoxide hydrolase (sEH) and report performance of more than an order of magnitude greater than random guessing on a holdout set, demonstrating that our model can serve as a baseline for comparison against other machine learning models on DEL data. Lastly, we provide a discussion on how we believe this informatics workflow could be applied to benefit researchers in their specific DEL campaigns.

Selinexor-Based Triplet Regimens in Patients With Multiple Myeloma Previously Treated With Anti-CD38 Monoclonal Antibodies.

BACKGROUND: The increasing use of anti-CD38 monoclonal antibodies (αCD38 mAbs) for newly diagnosed or early relapsed multiple myeloma (MM), especially in non-transplant eligible patients, may lead to more patients developing αCD38 mAb-refractory disease earlier in the treatment course with fewer treatment options. PATIENTS AND METHODS: We analyzed the efficacy and safety of selinexor-based triplets (selinexor+dexamethasone [Sd] plus pomalidomide [SPd, n = 23], bortezomib [SVd, n = 16] or carfilzomib (SKd, n = 23]) in a subset of STOMP (NCT02343042) and BOSTON (NCT03110562) study patients treated previously with αCD38 mAbs. RESULTS: Sixty-two patients (median 4 prior therapies, range 1 to 11, 90.3% refractory to αCD38 mAb) were included. Overall response rates (ORR) in the SPd, SVd and SKd cohorts were 52.2%, 56.3%, and 65.2%, respectively. Overall response rate was 47.4% among patients who had MM refractory to the third drug reintroduced in the Sd-based triplet. Median progression-free survival in the SPd, SVd, and SKd cohorts was 8.7, 6.7, and 15.0 months, respectively, and median overall survival was 9.6, 16.9, and 33.0 months, respectively. Median time to discontinuation in the SPd, SVd, and SKd cohorts was 4.4, 5.9, and 10.6 months, respectively. The most common hematological adverse events were thrombocytopenia, anemia, and neutropenia. Nausea, fatigue, and diarrhea were primarily grade 1/2. Adverse events were generally manageable with standard supportive care and dose modifications. CONCLUSION: Selinexor-based regimens may offer effective and well-tolerated therapy to patients with relapsed and/or refractory MM who had disease previously exposed or refractory to αCD38 mAb therapy and could help address the unmet clinical need in these high-risk patients.

The synergy of cybernetical intelligence with medical image analysis for deep medicine: A methodological perspective.

CONCEPTUAL INTRODUCTION: To introduce the concept of cybernetical intelligence, deep learning, development history, international research, algorithms, and the application of these models in smart medical image analysis and deep medicine are reviewed in this paper. This study also defines the terminologies for cybernetical intelligence, deep medicine, and precision medicine. REVIEW OF METHODS: Through literature research and knowledge reorganization, this review explores the fundamental concepts and practical applications of various deep learning techniques and cybernetical intelligence by conducting extensive literature research and reorganizing existing knowledge in medical imaging and deep medicine. The discussion mainly centers on the applications of classical models in this field and addresses the limitations and challenges of these basic models. EVALUATION AND DISCUSSIONS: In this paper, the more comprehensive overview of the classical structural modules in convolutional neural networks is described in detail from the perspective of cybernetical intelligence in deep medicine. The results and data of major research contents of deep learning are consolidated and summarized. CONCLUSION: There are some problems in machine learning internationally, such as insufficient research techniques, unsystematic research methods, incomplete research depth, and incomplete evaluation research. Some suggestions are given in our review to solve the problems existing in the deep learning models. Cybernetical intelligence has proven to be a valuable and promising avenue for advancing various fields, including deep medicine and personalized medicine.

LINC00493-encoded microprotein SMIM26 exerts anti-metastatic activity in renal cell carcinoma.

Human microproteins encoded by long non-coding RNAs (lncRNA) have been increasingly discovered, however, complete functional characterization of these emerging proteins is scattered. Here, we show that LINC00493-encoded SMIM26, an understudied microprotein localized in mitochondria, is tendentiously downregulated in clear cell renal cell carcinoma (ccRCC) and correlated with poor overall survival. LINC00493 is recognized by RNA-binding protein PABPC4 and transferred to ribosomes for translation of a 95-amino-acid protein SMIM26. SMIM26, but not LINC00493, suppresses ccRCC growth and metastatic lung colonization by interacting with acylglycerol kinase (AGK) and glutathione transport regulator SLC25A11 via its N-terminus. This interaction increases the mitochondrial localization of AGK and subsequently inhibits AGK-mediated AKT phosphorylation. Moreover, the formation of the SMIM26-AGK-SCL25A11 complex maintains mitochondrial glutathione import and respiratory efficiency, which is abrogated by AGK overexpression or SLC25A11 knockdown. This study functionally characterizes the LINC00493-encoded microprotein SMIM26 and establishes its anti-metastatic role in ccRCC, and therefore illuminates the importance of hidden proteins in human cancers.

EFHD1, a novel mitochondrial regulator of tumor metastasis in clear cell renal cell carcinoma.

The biological function of many mitochondrial proteins in mechanistic detail has not been well investigated in clear cell renal cell carcinoma (ccRCC). A seven-mitochondrial-gene signature was generated by Lasso regression analysis to improve the prediction of prognosis of patients with ccRCC, using The Cancer Genome Atlas and Clinical Proteomic Tumor Analysis Consortium cohort. Among those seven genes, EFHD1 is less studied and its role in the progression of ccRCC remains unknown. The decreased expression of EFHD1 was validated in clinical samples and was correlated with unfavorable outcome. Overexpression of EFHD1 in ccRCC cells resulted in the reduction of mitochondrial Ca2+ , and the inhibition of cell migration and invasion in vitro and tumor metastasis in vivo. Mechanistically, EFHD1 physically bound to the core mitochondrial calcium transporter (mitochondrial calcium uniporter, MCU) through its N-terminal domain. The interaction between EFHD1 and MCU suppressed the uptake of Ca2+ into mitochondria, and deactivated the Hippo/YAP signaling pathway. Further data revealed that the ectopic expression of EFHD1 upregulated STARD13 to enhance the phosphorylation of YAP protein at Ser-127. The knockdown of STARD13 or the overexpression of MCU partly abrogated the EFHD1-mediated induction of phosphorylation of YAP at Ser-127 and suppression of cell migration. Taken together, the newly identified EFHD1-MCU-STARD13 axis participates in the modulation of the Hippo/YAP pathway and serves as a novel regulator in the progression of ccRCC.

A five-protein prognostic signature with GBP2 functioning in immune cell infiltration of clear cell renal cell carcinoma.

Clear cell renal cell carcinoma (ccRCC) is of poor clinical outcomes, and currently lacks reliable prognostic biomarkers. By analyzing the datasets of the Cancer Genome Atlas (TCGA) and the Clinical Proteomic Tumor Analysis Consortium (CPTAC), we established a five-protein prognostic signature containing GBP2, HLA-DRA, ISG15, ISG20 and ITGAX. Our data indicate that this signature was closely correlated with advanced stage, higher pathological grade, and unfavorable survivals in patients with ccRCC. We further functionally characterized GBP2. Overexpression of GBP2 enhanced the phosphorylation of STAT2 and STAT3 to trigger JAK-STAT signaling and promote cell migration and invasion in ccRCC. Treatment of Ruxolitinib, a specific inhibitor of JAK/STAT, attenuated the GBP2-mediated phenotypes. Patients with high GBP2 expression were accompanied with more infiltration of immune cells positively stained with CD3, CD8, CD68, and immune checkpoint markers PD-1 and CTLA4, which was validated by Opal multiplex immunohistochemistry in ccRCC tissues. More CD8 + T cells and CD68 + macrophages were observed in patients expressing high GBP2. Taken together, a five-protein prognostic signature was constructed in our study. GBP2 has an oncogenic role via modulating JAK-STAT signaling and tumor immune infiltration, and thus may serve as a potential therapeutic target in ccRCC.

Selinexor-based regimens in patients with multiple myeloma after prior anti-B-cell maturation antigen treatment.

There is a lack of consensus on therapy sequencing in previously treated multiple myeloma, particularly after anti-B-cell maturation antigen (BCMA) therapy. Earlier reports on selinexor (X) regimens demonstrated considerable efficacy in early treatment, and after anti-BCMA-targeted chimeric antigen receptor-T cell therapy. Here, we present data from 11 heavily pretreated patients who predominantly received BCMA-antibody-drug conjugate therapy. We observe that X-containing regimens are potent and achieve durable responses with numerically higher overall response and clinical benefit rates, as well as median progression free survival compared to patients' prior anti-BCMA therapies, despite being used later in the treatment course. In an area of evolving unmet need, these data reaffirm the efficacy of X-based regimens following broader anti-BCMA therapy.

Faculty and resident perceptions of surgical resident workload in comparison to objective data.

BACKGROUND: Little is known of the way in which stakeholders in surgical education perceive trainee workload. METHODS: A web-based survey examining the perception of current resident workload (as a percentage of daytime activities) was distributed to the faculty and residents in a Canadian general surgery residency program. The analysis compared the trainee and faculty responses against a 660-hour resident workload observation dataset. RESULTS: A total of 17 residents and 16 faculty completed the survey (74%, 67% participation). The resident estimations of workload were accurate for task categories (r = 0.91) and individual tasks (r = 0.92). The faculty estimations were accurate for task category (r = 0.90) but less so for individual tasks (r = 0.78). The residents perceived that significantly less time was allocated toward educational activities than faculty. Both of the groups underestimated the amount of time spent on indirect patient care (IPC). CONCLUSION: The faculty overestimate educational tasks as a proportion of workload. Both of the groups underestimated IPC tasks. This information can guide resident training program design and be used to bridge gaps between resident and faculty perceptions of resident workload.

Learning Interrupted: Educational Impact of Interruptions on Surgical Residents.

OBJECTIVE: The complex workflow of surgical residents in the workplace-based learning environment makes interruptions an unavoidable aspect of clinical work. Interruptions have been shown to affect cognitive load, surgical performance, and medical error. The purpose of this study was to describe the effects of interruptions on surgical resident education. DESIGN: Junior residents were observed by 2 trained observers using time-motion methodology between September 2018-August 2019. Interruptions were identified and coded retroactively based on predefined criteria. We captured key features of interruptions including frequency, duration, task interrupted, outcome, and learner perceived educational value. SETTING: This study took place at London Health Sciences Centre in London, Ontario, Canada, a tertiary level academic health care center associated with the Schulich School of Medicine & Dentistry at Western University. PARTICIPANTS: Junior residents on a General Surgery service were eligible for participation. Participation was voluntary. 8 residents were observed over 24 clinical periods. RESULTS: A total of 278.2 hours of resident workflow were observed, and 229 interruptions were recorded. Interruptions account for 57.9 minutes/day of a surgical resident (SD = 60.7). Interruptions occur at a frequency of 0.82 interruptions/hour. Disruptive interruptions, that interfere with the continuation or completion of the original task, occur at a frequency of 0.11 interruptions/hour. Disruptive interruptions occurred at a higher frequency of 0.34 interruptions/hour during periods of feedback, coaching and informal teaching. DISCUSSION: We observed that tasks of higher learner perceived educational value are often interrupted by tasks of lower learner perceived education value. Valuable educational experiences such as feedback, coaching and informal teaching are interrupted at a greater rate and experience disruptions at a disproportionate rate. We identified feedback, coaching and informal teaching as an education task vulnerable to disruptive interruptions that would benefit from interventions targeted toward preventing interruption. Suggested interventions include "formalizing" feedback, coaching and informal teaching.

Comparison of RNAscope and immunohistochemistry for evaluation of the UPK2 status in urothelial carcinoma tissues.

BACKGROUND: UPK2 exhibits excellent specificity for urothelial carcinoma (UC). UPK2 evaluation can be useful in making the correct diagnosis of UC. However, UPK2 detection by immunohistochemistry (IHC) has relatively low sensitivity. This paper aimed to compare the diagnostic sensitivity of RNAscope and IHC for evaluation of the UPK2 status in UC. METHODS: Tissue blocks from 127 conventional bladder UCs, 45 variant bladder UCs, 24 upper tract UCs and 23 metastatic UCs were selected for this study. IHC and RNAscope were used to detect the UPK2 status in UCs. Then, comparisons of the two methods were undertaken. RESULTS: There was no significant difference between RNAscope and IHC for the evaluation of the UPK2 positivity rate in UC (68.0% vs. 62.6%, P = 0.141). Correlation analysis revealed a moderate positive correlation for detection of UPK2: RNAscope vs. IHC (P < 0.001, R = 0.441). Our results showed a trend toward a higher positive UPK2 rate detected by RNAscope (53.3%) than by IHC (35.6%) in variant bladder UCs. Disappointingly, the P value did not indicate a significant difference (P = 0.057). CONCLUSIONS: RNAscope for UPK2 appeared to perform similarly to IHC, with a marginally higher positive rate, suggesting it could be used as an alternative or adjunct to UPK2 IHC.

Once weekly selinexor, carfilzomib and dexamethasone in carfilzomib non-refractory multiple myeloma patients.

BACKGROUND: Proteasome inhibitors (PIs), including carfilzomib, potentiate the activity of selinexor, a novel, first-in-class, oral selective inhibitor of nuclear export (SINE) compound, in preclinical models of multiple myeloma (MM). METHODS: The safety, efficacy, maximum-tolerated dose (MTD) and recommended phase 2 dose (RP2D) of selinexor (80 or 100 mg) + carfilzomib (56 or 70 mg/m2) + dexamethasone (40 mg) (XKd) once weekly (QW) was evaluated in patients with relapsed refractory MM (RRMM) not refractory to carfilzomib. RESULTS: Thirty-two patients, median prior therapies 4 (range, 1-8), were enrolled. MM was triple-class refractory in 38% of patients and 53% of patients had high-risk cytogenetics del(17p), t(4;14), t(14;16) and/or gain 1q. Common treatment-related adverse events (all/Grade 3) were thrombocytopenia 72%/47% (G3 and G4), nausea 72%/6%, anaemia 53%/19% and fatigue 53%/9%, all expected and manageable with supportive care and dose modifications. MTD and RP2D were identified as selinexor 80 mg, carfilzomib 56 mg/m2, and dexamethasone 40 mg, all QW. The overall response rate was 78% including 14 (44%) ≥ very good partial responses. Median progression-free survival was 15 months. CONCLUSIONS: Weekly XKd is highly effective and well-tolerated. These data support further investigation of XKd in patients with MM.

A Combined Deep Learning and Anatomical Inch Measurement Approach to Robotic Acupuncture Points Positioning.

Acupuncture therapy is one of the cornerstones in traditional Chinese medicine. It requires rich experiences from Chinese medicine practitioner. However, repeatability among different practitioners are low. Meanwhile, there is a large variety of skin conditions in terms of color, diseases, size, etc. In recent year, deep neural network for acupuncture point detection is proposed. However, it is difficult to localize multiple acupuncture points. In this paper, a high repeatability robot with a new approach of acupuncture points positioning is proposed which can be adaptive to variety skin conditions and achieve multiple acupuncture points' localization.Clinical Relevance- This system can provide identical acupuncture therapy to different patients. Thus, the quality of the therapy can be practitioner independent. Furthermore, the machine operation is simple therefore manual error can be reduced significantly. As the result, the efficiency and accuracy of therapy can be increased.

α-Fetoprotein mRNA in situ hybridisation is a highly specific marker of hepatocellular carcinoma: a multi-centre study.

BACKGROUND: Pathologic diagnosis of hepatocellular carcinoma (HCC) can be challenging in differentiating from benign and non-hepatocytic malignancy lesions. The aim of this study was to investigate the potential utility of α-fetoprotein (AFP) mRNA RNAscope, a sensitive and specific method, in the diagnosis of HCC. METHODS: Three independent retrospective cohorts containing 2216 patients with HCC, benign liver lesions, and non-hepatocytic tumours were examined. AFP was detected using ELISA, IHC (Immunohistochemistry), and RNAscope. Glypican3 (GPC3), hepatocyte paraffin-1 (HepPar-1), and arginase-1 (Arg-1) proteins were detected using IHC. RESULTS: AFP RNAscope improved the HCC detection sensitivity by 24.7-32.7% compared with IHC. In two surgical cohorts, a panel of AFP RNAscope and GPC3 provided the best diagnostic value in differentiating HCC from benign hepatocytic lesions (AUC = 0.905 and 0.811), and a panel including AFP RNAscope, GPC3, HepPar-1, and Arg-1 yielded the best AUC (0.971 and 0.977) when distinguishing HCC from non-hepatocytic malignancies. The results from the liver biopsy cohort were similar, and additional application of AFP RNAscope improved the sensitivity by 18% when distinguishing HCC from benign hepatocytic lesions. CONCLUSIONS: AFP mRNA detected by RNAscope is highly specific for hepatocytic malignancy and may serve as a novel diagnostic biomarker for HCC.

A Zic2/Runx2/NOLC1 signaling axis mediates tumor growth and metastasis in clear cell renal cell carcinoma.

Clear cell renal cell carcinoma (ccRCC) is one of the most common malignancies with rapid growth and high metastasis, but lacks effective therapeutic targets. Here, using public sequencing data analyses, quantitative real-time PCR assay, western blotting, and IHC staining, we characterized that runt-related transcription factor 2 (Runx2) was significantly upregulated in ccRCC tissues than that in normal renal tissues, which was associated with the worse survival of ccRCC patients. Overexpression of Runx2 promoted malignant proliferation and migration of ccRCC cells, and inversely, interfering Runx2 with siRNA attenuates its oncogenic ability. RNA sequencing and functional studies revealed that Runx2 enhanced ccRCC cell growth and metastasis via downregulation of tumor suppressor nucleolar and coiled-body phosphoprotein 1 (NOLC1). Moreover, increased Zic family member 2 (Zic2) was responsible for the upregulation of Runx2 and its oncogenic functions in ccRCC. Kaplan-Meier survival analyses indicated that ccRCC patients with high Zic2/Runx2 and low NOLC1 had the worst outcome. Therefore, our study demonstrates that Zic2/Runx2/NOLC1 signaling axis promotes ccRCC progression, providing a set of potential targets and prognostic indicators for patients with ccRCC.