Tumor Typing Research Articles

Pathological Diagnostic Testing and Biomarkers for Perioperative System Therapy

Optimal personalized treatment planning for resectable lung cancer requires quality-assured, standardized and prompt processing of tissue samples in pathological laboratories, as well as the determination of relevant predictive and prognostic biomarkers. Pathological diagnostic testing includes histological tumor typing, staging and tumor grading, resection status and, if necessary, regression grading after neoadjuvant systemic therapy. Histopathological typing is performed according to the current WHO classification and includes adenocarcinomas, squamous cell carcinomas, other non-small cell lung carcinomas (NSCLCs), carcinoids, small cell and large cell neuroendocrine carcinomas. Standardized tumor grading currently plays an important role in invasive non-mucinous adenocarcinoma in particular and enables prognostic risk assessment. The R classification and regression grading are also prognostically relevant. In the early stages of NSCLC, molecular biomarkers such as EGFR, ALK and PD-L1, are relevant for decisions on individual treatment. Testing is performed on FFPE tissue samples and must be carried out in a quality-assured manner and in accordance with international standards.

Applications of machine learning to MR imaging of pediatric low-grade gliomas.

Machine learning (ML) shows promise for the automation of routine tasks related to the treatment of pediatric low-grade gliomas (pLGG) such as tumor grading, typing, and segmentation. Moreover, it has been shown that ML can identify crucial information from medical images that is otherwise currently unattainable. For example, ML appears to be capable of preoperatively identifying the underlying genetic status of pLGG. In this chapter, we reviewed, to the best of our knowledge, all published works that have used ML techniques for the imaging-based evaluation of pLGGs. Additionally, we aimed to provide some context on what it will take to go from the exploratory studies we reviewed to clinically deployed models. Multiple studies have demonstrated that ML can accurately grade, type, and segment and detect the genetic status of pLGGs. We compared the approaches used between the different studies and observed a high degree of variability throughout the methodologies. Standardization and cooperation between the numerous groups working on these approaches will be key to accelerating the clinical deployment of these models. The studies reviewed in this chapter detail the potential for ML techniques to transform the treatment of pLGG. However, there are still challenges that need to be overcome prior to clinical deployment.

Deciphering the SOX4/MAPK1 regulatory axis: a phosphoproteomic insight into IQGAP1 phosphorylation and pancreatic Cancer progression

ObjectiveThis study aims to elucidate the functional role of IQGAP1 phosphorylation modification mediated by the SOX4/MAPK1 regulatory axis in developing pancreatic cancer through phosphoproteomics analysis.MethodsProteomics and phosphoproteomics data of pancreatic cancer were obtained from the Clinical Proteomic Tumor Analysis Consortium (CPTAC) database. Differential analysis, kinase-substrate enrichment analysis (KSEA), and independent prognosis analysis were performed on these datasets. Subtype analysis of pancreatic cancer patients was conducted based on the expression of prognostic-related proteins, and the prognosis of different subtypes was evaluated through prognosis analysis. Differential analysis of proteins in different subtypes was performed to identify differential proteins in the high-risk subtype. Clinical correlation analysis was conducted based on the expression of prognostic-related proteins, pancreatic cancer typing results, and clinical characteristics in the pancreatic cancer proteomics dataset. Functional pathway enrichment analysis was performed using GSEA/GO/KEGG, and most module proteins correlated with pancreatic cancer were selected using WGCNA analysis. In cell experiments, pancreatic cancer cells were grouped, and the expression levels of SOX4, MAPK1, and the phosphorylation level of IQGAP1 were detected by RT-qPCR and Western blot experiments. The effect of SOX4 on MAPK1 promoter transcriptional activity was assessed using a dual-luciferase assay, and the enrichment of SOX4 on the MAPK1 promoter was examined using a ChIP assay. The proliferation, migration, and invasion functions of grouped pancreatic cancer cells were assessed using CCK-8, colony formation, and Transwell assays. In animal experiments, the impact of SOX4 on tumor growth and metastasis through the regulation of MAPK1-IQGAP1 phosphorylation modification was studied by constructing subcutaneous and orthotopic pancreatic cancer xenograft models, as well as a liver metastasis model in nude mice.ResultsPhosphoproteomics and proteomics data analysis revealed that the kinase MAPK1 may play an important role in pancreatic cancer progression by promoting IQGAP1 phosphorylation modification. Proteomics analysis classified pancreatic cancer patients into two subtypes, C1 and C2, where the high-risk C2 subtype was associated with poor prognosis, malignant tumor typing, and enriched tumor-related pathways. SOX4 may promote the occurrence of the high-risk C2 subtype of pancreatic cancer by regulating MAPK1-IQGAP1 phosphorylation modification. In vitro cell experiments confirmed that SOX4 promoted IQGAP1 phosphorylation modification by activating MAPK1 transcription while silencing SOX4 inhibited the proliferation, migration, and invasion of pancreatic cancer cells by reducing the phosphorylation level of MAPK1-IQGAP1. In vivo, animal experiments further confirmed that silencing SOX4 suppressed the growth and metastasis of pancreatic cancer by reducing the phosphorylation level of MAPK1-IQGAP1.ConclusionThe findings of this study suggest that SOX4 promotes the phosphorylation modification of IQGAP1 by activating MAPK1 transcription, thereby facilitating the growth and metastasis of pancreatic cancer.

Multiple RNA Rapid In Situ Imaging Based on Cas9 Code Key System.

Existing RNA in situ imaging strategies mostly utilize parallel repetitive nucleic acid self-assembly to achieve multiple analysis, with limitations of complicated systems and cumbersome steps. Here, a Cas9 code key system with key probe (KP) encoder and CRISPR/Cas9 signal exporter is developed. This system triggers T-protospacer adjacent motif (T-PAMstructural transitions of multiple KP encoders to form coding products with uniform single-guide RNA (sgRNA) target sequences as tandem nodes. Only single sgRNA/Cas9 complex is required to cleave multiple coding products, enabling efficient "many-to-one" tandem signaling, and non-collateral cleavage activity-dependent automatic signaling output through active introduction of mismatched bases. Compared with conventional parallel multiple signaling analysis model, the proposed system greatly simplifies reaction process and enhances detection efficiency. Further, a rapid multiple RNA in situ imaging system is developed by combining the Cas9 code key system with a T-strand displacement amplification (T-SDA) signal amplifier. The constructed system is applied to tumor cells and clinicopathology slices, generating clear multi-mRNA imaging profiles in less than an hour with just one step. Therefore, this work provides reliable technical support for clinical tumor typing and molecular mechanism investigation.

MHD-Net: Memory-Aware Hetero-Modal Distillation Network for Thymic Epithelial Tumor Typing With Missing Pathology Modality.

Fusing multi-modal radiology and pathology data with complementary information can improve the accuracy of tumor typing. However, collecting pathology data is difficult since it is high-cost and sometimes only obtainable after the surgery, which limits the application of multi-modal methods in diagnosis. To address this problem, we propose comprehensively learning multi-modal radiology-pathology data in training, and only using uni-modal radiology data in testing. Concretely, a Memory-aware Hetero-modal Distillation Network (MHD-Net) is proposed, which can distill well-learned multi-modal knowledge with the assistance of memory from the teacher to the student. In the teacher, to tackle the challenge in hetero-modal feature fusion, we propose a novel spatial-differentiated hetero-modal fusion module (SHFM) that models spatial-specific tumor information correlations across modalities. As only radiology data is accessible to the student, we store pathology features in the proposed contrast-boosted typing memory module (CTMM) that achieves type-wise memory updating and stage-wise contrastive memory boosting to ensure the effectiveness and generalization of memory items. In the student, to improve the cross-modal distillation, we propose a multi-stage memory-aware distillation (MMD) scheme that reads memory-aware pathology features from CTMM to remedy missing modal-specific information. Furthermore, we construct a Radiology-Pathology Thymic Epithelial Tumor (RPTET) dataset containing paired CT and WSI images with annotations. Experiments on the RPTET and CPTAC-LUAD datasets demonstrate that MHD-Net significantly improves tumor typing and outperforms existing multi-modal methods on missing modality situations.

EpiDiP/NanoDiP: a versatile unsupervised machine learning edge computing platform for epigenomic tumour diagnostics

DNA methylation analysis based on supervised machine learning algorithms with static reference data, allowing diagnostic tumour typing with unprecedented precision, has quickly become a new standard of care. Whereas genome-wide diagnostic methylation profiling is mostly performed on microarrays, an increasing number of institutions additionally employ nanopore sequencing as a faster alternative. In addition, methylation-specific parallel sequencing can generate methylation and genomic copy number data. Given these diverse approaches to methylation profiling, to date, there is no single tool that allows (1) classification and interpretation of microarray, nanopore and parallel sequencing data, (2) direct control of nanopore sequencers, and (3) the integration of microarray-based methylation reference data. Furthermore, no software capable of entirely running in routine diagnostic laboratory environments lacking high-performance computing and network infrastructure exists. To overcome these shortcomings, we present EpiDiP/NanoDiP as an open-source DNA methylation and copy number profiling suite, which has been benchmarked against an established supervised machine learning approach using in-house routine diagnostics data obtained between 2019 and 2021. Running locally on portable, cost- and energy-saving system-on-chip as well as gpGPU-augmented edge computing devices, NanoDiP works in offline mode, ensuring data privacy. It does not require the rigid training data annotation of supervised approaches. Furthermore, NanoDiP is the core of our public, free-of-charge EpiDiP web service which enables comparative methylation data analysis against an extensive reference data collection. We envision this versatile platform as a useful resource not only for neuropathologists and surgical pathologists but also for the tumour epigenetics research community. In daily diagnostic routine, analysis of native, unfixed biopsies by NanoDiP delivers molecular tumour classification in an intraoperative time frame.

Integrative HLA typing of tumor and adjacent normal tissue can reveal insights into the tumor immune response

BackgroundThe HLA complex is the most polymorphic region of the human genome, and its improved characterization can help us understand the genetics of human disease as well as the interplay between cancer and the immune system. The main function of HLA genes is to recognize “non-self” antigens and to present them on the cell surface to T cells, which instigate an immune response toward infected or transformed cells. While sequence variation in the antigen-binding groove of HLA may modulate the repertoire of immunogenic antigens presented to T cells, alterations in HLA expression can significantly influence the immune response to pathogens and cancer.MethodsRNA sequencing was used here to accurately genotype the HLA region and quantify and compare the level of allele-specific HLA expression in tumors and patient-matched adjacent normal tissue. The computational approach utilized in the study types classical and non-classical Class I and Class II HLA alleles from RNA-seq while simultaneously quantifying allele-specific or personalized HLA expression. The strategy also uses RNA-seq data to infer immune cell infiltration into tumors and the corresponding immune cell composition of matched normal tissue, to reveal potential insights related to T cell and NK cell interactions with tumor HLA alleles.ResultsThe genotyping method outperforms existing RNA-seq-based HLA typing tools for Class II HLA genotyping. Further, we demonstrate its potential for studying tumor-immune interactions by applying the method to tumor samples from two different subtypes of breast cancer and their matched normal breast tissue controls.ConclusionsThe integrative RNA-seq-based HLA typing approach described in the study, coupled with HLA expression analysis, neoantigen prediction and immune cell infiltration, may help increase our understanding of the interplay between a patient’s tumor and immune system; and provide further insights into the immune mechanisms that determine a positive or negative outcome following treatment with immunotherapy such as checkpoint blockade.

Clinicopathological and Immunohistochemical Analysis of Lung Cancer Biopsies at a Tertiary Care Centre in Central Gujarat, India

Introduction: Lung biopsy is a reliable and highly accurate tool for diagnosing and subtyping lung lesions. Two-thirds of lung cancer patients present with advanced disease, and small biopsies remain the main diagnostic tool. Histopathological and immunohistochemical examination of lung biopsies plays an important role in accurately diagnosing lung malignancies. Aim: The aim of present study was to analyse the clinicopathological and immunohistochemical features of lung cancer. Materials and Methods: This cross-sectional study was conducted in the Department of Pathology, Medical College, Baroda, over a period of two years and three months, from March 2019 to May 2021. A total of 53 cases of malignant lung cancer biopsies were included in this study. Clinical parameters, such as patient age, complaints, smoking history, clinical presentation, radiological findings, histopathology reports, and IHC findings, were retrieved. All cases underwent histopathological examination, and tumour typing was done according to the WHO classification 2021 of lung tumours. Immunohistochemistry (IHC) markers like CK5/6, CK7, TTF1, P63, synaptophysin, chromogranin, and pankeratin were used for further subtyping. Data analysis was done by Microsoft Excel 2019 spreadsheet. Results: A total of 53 cases of malignant lung cancer were included in this study. Of these, 38 (71.69%) cases were male, and 15 (28.30%) cases were female. 43 (81.13%) cases were over 50 years old, with a mean age of 61 years. After IHC, adenocarcinoma was seen in 27 cases (50.9%), followed by Squamous Cell Carcinoma (SCC) in 19 cases (35.8%), five cases of small cell carcinoma, one case of large cell neuroendocrine carcinoma, and one case of Non Small Cell Carcinoma-Not Otherwise Specified (NSCC-NOS). In adenocarcinoma, positivity for CK7, TTF1, and p63 was 100%, 74.07%, and 16.6%, respectively. In squamous cell carcinoma, positivity for CK5/6 and p63 was 100% and 78.94%, respectively. Conclusion: In the present study, the majority of cases were males. Adenocarcinoma was the most common subtype of Non Small Cell Lung Cancer (NSCLC). A panel of IHC markers is helpful in differentiating NSCLC into adenocarcinoma and squamous cell carcinoma on small true-cut lung biopsies. CK5/6 and p63 aid in detecting squamous cell carcinoma, while CK7 and TTF-1 help in detecting adenocarcinoma. The recommended immunohistochemical profile for NSCLC includes TTF1, CK5/6, CK7, and p63.

Non-papillary thyroid carcinoma diagnoses in The Bethesda System for Reporting Thyroid Cytopathology categories V and VI: An institutional experience

BackgroundThe non-papillary thyroid carcinoma (PTC) subgroups of The Bethesda System for Reporting Thyroid Cytopathology (TBSRTC) categories V (Suspicious for malignancy) and VI (Malignant) are rare, and specific tumor typing is difficult. We aimed to analyze histologic outcomes and to investigate the points of caution. MethodsWe reviewed the electronic database and identified 12,215 cases of thyroid fine-needle aspiration cytology between 2013 and 2022. In total, 2783 patients were diagnosed with TBSRTC V or VI. Of these, 51 patients with non-PTC diagnosis were identified. Histological outcomes were analyzed with the cytologic findings. ResultsThe subgroups of non-PTC diagnoses in TBSRTC category V or VI consisted of medullary thyroid carcinoma (MTC) (13/51, 25.5 %), anaplastic thyroid carcinoma (3/51, 5.9 %), lymphoma (2/51, 3.9 %), metastatic tumor (4/51, 7.8 %), and malignant, not otherwise specified (NOS) (29/51, 56.9 %). The concordance rate of the histological outcomes was 30 % (12/40), predominantly comprising MTC cases. The obscuring factors for specific tumor typing in the suspicious for malignancy/malignant NOS cytology diagnosis group was mixed pattern of well differentiated thyroid carcinoma and less differentiated carcinoma cells (9/24, 37.5 %), low cellularity (7/24, 29.2 %) and a history of non-thyroid organ malignancy (6/24, 25 %). The less differentiated carcinoma component in mixed pattern consisted of 2 poorly differentiated thyroid carcinomas, 2 anaplastic thyroid carcinomas, 4 high-grade PTCs and 1 high-grade MTC. ConclusionThe high-grade feature of PTC or MTC cytology is a noteworthy obscuring factor in specific tumor typing of non-PTC cytology diagnosis.

Infection with COVID-19 promotes the progression of pancreatic cancer through the PI3K-AKT signaling pathway

ObjectiveTo investigate the effect of COVID-19 infection on pancreatic cancer.MethodsBased on the mRNA-Seq data of COVID-19 patients and pancreatic cancer (PC) patients in the GEO database, we used a support vector machine (SVM), LASSO-Cox regression analysis and random forest tree (RF) to screen the common signature genes of the two diseases and further investigate their effects and functional characteristics on PC, respectively. The above procedures were performed in R software.ResultsThe proteins COL10A1/FAP/FN1 were found to be common signature genes for COVID-19 and PC, were significantly up-regulated in both diseases and showed good diagnostic efficacy for PC. The risk model based on COL10A1/FAP/FN1 showed good PC risk prediction ability and clinical application potential. Tumor typing based on COL10A1/FAP/FN1 expression levels effectively classified PC into different subtypes and showed significant differences between the two subtypes in terms of survival prognosis, immune levels, immune checkpoint expression levels, mutation status of common tumor mutation sites, and drug sensitivity analysis. While pathway analysis also revealed that FN1 as an extracellular matrix component may be involved in the biological process of PC by regulating the PI3K-AKT signaling axis.ConclusionThe upregulated expression of COL10A1/FAP/FN1, the characteristic genes of COVID-19, are potential diagnostic targets for PC, and the upregulated expression of FN1 may promote the progression of PC by activating the PI3K-AKT signaling pathway. The COL10A1/FAP/FN1-based typing provides a new typing approach for PC, and also provides a good reference and idea for the refinement of PC treatment and subsequent clinical research.

Classification of high-risk depressed mood groups in cancer patients based on Health Ecology Model

BackgroundDepressed mood affects a significant number of patients with cancer, and can impair their quality of life and interfere with successful treatment. Our study aims to create a predictive model for identifying high-risk groups of depressed mood in cancer patients, offering a theoretical support for preventing depressed mood in these individuals. MethodsThe China Health and Retirement Longitudinal Study (CHARLS) provided the data for this research, which used CES-D as a tool to identify individuals with depressed mood. Influencing factors of depressed mood in cancer patients was analyzed using a binary logistic regression model. Using the Harvard Cancer Index, we classified the high-risk patients for depressed mood. ResultsIn present study, 52.96 % of cancer patients met criteria for depressed mood based on the CES-D. Significant correlations were found between depressed mood and factors such as gender, self-rated health, sleep duration, exercise, satisfaction with family, residence, education, life satisfaction, and medical insurance. Utilizing the Harvard Cancer Index, we classified patients into five risk levels for depressed mood, revealing a significant variation in the number of depressive patients across these levels (x2=99.82, P < 0.05). Notably, the incidence of depressed mood increased with the risk level among cancer patients (x2=103.40, P < 0.05). LimitationsLack of data on tumor typing and subgroups makes it unlikely to explore the specifics of depressed mood in patients with various types of cancer. ConclusionThe determinants of depressed mood in cancer patients are multi-dimensional. The Harvard Cancer Index may be helpful in identifying high-risk populations.

The success rate of small renal mass core needle biopsy and its impact on lowering benign resection rate

BackgroundSmall renal mass (SRM) biopsy remains under-utilized due to stigma. Meanwhile, the alarmingly high benign findings in resected kidney masses highlight the need for improved preoperative diagnosis and patient selection.MethodsThe purpose of this study is to review the success rate of SRM biopsy and to evaluate its impact on patient management. A total of 168 percutaneous image-guided core needle biopsies (CNBs) of SRMs were retrieved at a tertiary academic center between 2015 and 2019. Subsequent treatment choices, side effects and outcomes were retrospectively reviewed.ResultsThe diagnostic rate of CNB was 86.9%. Benign neoplasms accounted for a significant portion (14.3%) of SRM. Renal cell carcinomas (RCCs) were the most common diagnoses (69.6%) as expected. In biopsy-resection correlation, the positive predictive value of CNB was 100%. Tumor typing and subtyping by CNB were highly accurate, 100% and 98.3% respectively. Nuclear grading for clear cell RCC was accurate in 83.8% cases. The CNB results had significant impact on treatment. Most patients with RCCs underwent either resection (54.1%) or ablation (33.9%), in contrast to observation in benign neoplasms (90.5%). Most importantly, the benign resection rate (3.2%) in this series was much lower than the national average.ConclusionCNB provided accurate diagnoses for the majority of SRMs and revealed benign diagnoses in a subset of clinically suspicious lesions. Employment of CNB in suspicious SRM may help avoid overtreatment for benign lesions.

Pathologic Processing of Lung Cancer Resection Specimens After Neoadjuvant Therapy

Neoadjuvant treatment of non-small cell lung cancer challenges the traditional processing of pathology specimens. Induction therapy before resection allows evaluation of the efficacy of neoadjuvant agents at the time of surgery. Many clinical trials use pathologic tumor response, measured as major pathologic response (MPR, ≤10% residual viable tumor [RVT]) or complete pathologic response (CPR, 0% RVT) as a surrogate of clinical efficacy. Consequently, accurate pathologic evaluation of RVT is crucial. However, pathologic assessment has not been uniform, which is particularly true for sampling of the primary tumor, which instead of the traditional processing, requires different tissue submission because the focus has shifted from tumor typing alone to RVT scoring. Using a simulation study, we analyzed the accuracy rates of %RVT, MPR, and CPR of 31 pretreated primary lung tumors using traditional grossing compared with the gold standard of submitting the entire residual primary tumor and identified the minimum number of tumor sections to be submitted to ensure the most accurate scoring of %RVT, MPR, and CPR. Accurate %RVT, MPR, and CPR calls were achieved in 52%, 87%, and 81% of cases, respectively, using the traditional grossing method. Accuracy rates of at least 90% for these parameters require either submission of all residual primary tumor or at least 20 tumor sections. Accurate %RVT, MPR, and CPR scores cannot be achieved with traditional tumor grossing. Submission of the entire primary tumor, up to a maximum of 20 sections, is required for the most accurate reads.

Identification and validation of a novel anoikis-related signature for predicting prognosis and immune landscape in ovarian serous cystadenocarcinoma

BackgroundOvarian serous cystadenocarcinoma (OSC) is the most prevalent histological subtype of ovarian cancer (OV) and presents a serious threat to women's health. Anoikis is an essential component of metastasis, and tumor cells can get beyond it to become viable. The impact of anoikis on OSC, however, has only been the topic of a few studies. MethodsThe mRNA sequencing and clinical information of OSC came from The Cancer Genome Atlas Target Genotype-Tissue Expression (TCGA TARGET GTEx) dataset. Anoikis-related genes (ARGs) were collected by Harmonizome and GeneCards websites. Centered on these ARGs, we used unsupervised consensus clustering to explore potential tumor typing and filtered hub ARGs to create a model of predictive signature for OSC patients. Furthermore, we presented clinical specialists with a novel nomogram based on ARGs, revealing the underlying clinical relevance of this signature. Finally, we explored the immune microenvironment among various risk groups. ResultsWe identified 24 ARGs associated with the prognosis of OSC and classified OSC patients into three subtypes, and the subtype with the best prognosis was more enriched in immune-related pathways. Seven ARGs (ARHGEF7, NOTCH4, CASP2, SKP2, PAK4, LCK, CCDC80) were chosen to establish a risk model and a nomogram that can provide practical clinical decision support. Risk scores were found to be an independent and significant prognostic factor in OSC patients. The CIBERSORTx result revealed an inflammatory microenvironment is different for risk groups, and the proportion of immune infiltrates of Macrophages M1 is negatively correlated with risk score (rs = −0.21, P < 0.05). Ultimately, quantitative reverse transcription polymerase chain reaction (RT-PCR) was utilized to validate the expression of the seven pivotal ARGs. ConclusionIn this study, based on seven ARGs, a risk model and nomogram established can be used for risk stratification and prediction of survival outcomes in patients with OSC, providing a reliable reference for individualized therapy of OSC patients.

Supervised topological data analysis for MALDI mass spectrometry imaging applications

BackgroundMatrix-assisted laser desorption/ionization mass spectrometry imaging (MALDI MSI) displays significant potential for applications in cancer research, especially in tumor typing and subtyping. Lung cancer is the primary cause of tumor-related deaths, where the most lethal entities are adenocarcinoma (ADC) and squamous cell carcinoma (SqCC). Distinguishing between these two common subtypes is crucial for therapy decisions and successful patient management.ResultsWe propose a new algebraic topological framework, which obtains intrinsic information from MALDI data and transforms it to reflect topological persistence. Our framework offers two main advantages. Firstly, topological persistence aids in distinguishing the signal from noise. Secondly, it compresses the MALDI data, saving storage space and optimizes computational time for subsequent classification tasks. We present an algorithm that efficiently implements our topological framework, relying on a single tuning parameter. Afterwards, logistic regression and random forest classifiers are employed on the extracted persistence features, thereby accomplishing an automated tumor (sub-)typing process. To demonstrate the competitiveness of our proposed framework, we conduct experiments on a real-world MALDI dataset using cross-validation. Furthermore, we showcase the effectiveness of the single denoising parameter by evaluating its performance on synthetic MALDI images with varying levels of noise.ConclusionOur empirical experiments demonstrate that the proposed algebraic topological framework successfully captures and leverages the intrinsic spectral information from MALDI data, leading to competitive results in classifying lung cancer subtypes. Moreover, the framework’s ability to be fine-tuned for denoising highlights its versatility and potential for enhancing data analysis in MALDI applications.

MulGT: Multi-Task Graph-Transformer with Task-Aware Knowledge Injection and Domain Knowledge-Driven Pooling for Whole Slide Image Analysis

Whole slide image (WSI) has been widely used to assist automated diagnosis under the deep learning fields. However, most previous works only discuss the SINGLE task setting which is not aligned with real clinical setting, where pathologists often conduct multiple diagnosis tasks simultaneously. Also, it is commonly recognized that the multi-task learning paradigm can improve learning efficiency by exploiting commonalities and differences across multiple tasks. To this end, we present a novel multi-task framework (i.e., MulGT) for WSI analysis by the specially designed Graph-Transformer equipped with Task-aware Knowledge Injection and Domain Knowledge-driven Graph Pooling modules. Basically, with the Graph Neural Network and Transformer as the building commons, our framework is able to learn task-agnostic low-level local information as well as task-specific high-level global representation. Considering that different tasks in WSI analysis depend on different features and properties, we also design a novel Task-aware Knowledge Injection module to transfer the task-shared graph embedding into task-specific feature spaces to learn more accurate representation for different tasks. Further, we elaborately design a novel Domain Knowledge-driven Graph Pooling module for each task to improve both the accuracy and robustness of different tasks by leveraging different diagnosis patterns of multiple tasks. We evaluated our method on two public WSI datasets from TCGA projects, i.e., esophageal carcinoma and kidney carcinoma. Experimental results show that our method outperforms single-task counterparts and the state-of-theart methods on both tumor typing and staging tasks.

Nanoprobes based on fluorescent semiconductor nanocrystals and single-domain antibodies for highly sensitive detection of epidermal growth factor receptor in tumor cells

Background. The development of highly oriented conjugates of quantum dots (QDs) and single-domain antibodies (sdAbs) as innovative fluorescence imaging nanoprobes that specifically recognize tumor biomarkers, in particular, epidermal growth factor receptor (EGFR), is a promising approach to improving immunohistochemical tumor typing.Aim. The study was aimed at developing fluorescent nanoprobes based on QDs and sdAbs that specifically recognize EGFR, as well as evaluating their functional characteristics (size and optical properties) and functional activity.Materials and methods. QDs were obtained using high-temperature organometallic synthesis and transferred into the aqueous phase by means of stepwise replacement of ligands on the QD surface. The QDs and sdAbs were conjugated in an oriented manner using a bifunctional cross-linking agent. Detailed characteristics of the resulting conjugates were analyzed by the dynamic light scattering and immunoassay methods. Functional activity was assessed on the model human epidermoid carcinoma cells line A431.Results. The QD–sdAb conjugates have been standardized in terms of control parameters determining their functional activity, in particular, hydrodynamic diameter and efficiency of binding with target tumor cells. They are characterized by high dispersity, homogeneity, and specific functional activity towards their molecular target.Conclusion. The results demonstrate the potential use of the designed QD–sdAb conjugates for EGRF detection in immunohistochemical typing of tumor.

Identification of osteosarcoma m6A-related prognostic biomarkers using artificial intelligence: RBM15

Osteosarcoma has the worst prognosis among malignant bone tumors, and effective biomarkers are lacking. Our study aims to explore m6A-related and immune-related biomarkers. Gene expression profiles of osteosarcoma and healthy controls were downloaded from multiple public databases, and their m6A-based gene expression was utilized for tumor typing using bioinformatics. Subsequently, a prognostic model for osteosarcoma was constructed using the least absolute shrinkage and selection operator and multivariate Cox regression analysis, and its immune cell composition was calculated using the CIBERSORTx algorithm. We also performed drug sensitivity analysis for these two genes. Finally, analysis was validated using immunohistochemistry. We also examined the RBM15 gene by qRT-PCR in an in vitro experiment. We collected routine blood data from 1738 patients diagnosed with osteosarcoma and 24,344 non-osteosarcoma patients and used two independent sample t tests to verify the accuracy of the CIBERSORTx analysis for immune cell differences. The analysis based on m6A gene expression tumor typing was most reliable using the two typing methods. The prognostic model based on the two genes constituting RNA-binding motif protein 15 (RBM15) and YTDC1 had a much lower survival rate for patients in the high-risk group than those in the low-risk group (P < 0.05). CIBERSORTx immune cell component analysis demonstrated that RBM15 showed a negative and positive correlation with T cells gamma delta and activated natural killer cells, respectively. Drug sensitivity analysis showed that these two genes showed varying degrees of correlation with multiple drugs. The results of immunohistochemistry revealed that the expression of these two genes was significantly higher in osteosarcoma than in paraneoplastic tissues. The results of qRT-PCR experiments showed that the expression of RBM15 was significantly higher in both osteosarcomas than in the control cell lines. Absolute lymphocyte value, lymphocyte percentage, hematocrit and erythrocyte count were lower in osteosarcoma than in the control group (P < 0.001). RBM15 and YTHDC1 can serve as potential prognostic biomarkers associated with m6A in osteosarcoma.

The novel subclusters based on cancer-associated fibroblast for pancreatic adenocarcinoma.

Pancreatic adenocarcinoma (PAAD) is a fatal disease characterized by promoting connective tissue proliferation in the stroma. Activated cancer-associated fibroblasts (CAFs) play a key role in fibrogenesis in PAAD. CAF-based tumor typing of PAAD has not been explored. We extracted single-cell sequence transcriptomic data from GSE154778 and CRA001160 datasets from Gene Expression Omnibus or Tumor Immune Single-cell Hub to collect CAFs in PAAD. On the basis of Seurat packages and new algorithms in machine learning, CAF-related subtypes and their top genes for PAAD were analyzed and visualized. We used CellChat package to perform cell-cell communication analysis. In addition, we carried out functional enrichment analysis based on clusterProfiler package. Finally, we explored the prognostic and immunotherapeutic value of these CAF-related subtypes for PAAD. CAFs were divided into five new subclusters (CAF-C0, CAF-C1, CAF-C2, CAF-C3, and CAF-C4) based on their marker genes. The five CAF subclusters exhibited distinct signaling patterns, immune status, metabolism features, and enrichment pathways and validated in the pan-cancer datasets. In addition, we found that both CAF-C2 and CAF-C4 subgroups were negatively correlated with prognosis. With their top genes of each subclusters, the sub-CAF2 had significantly relations to immunotherapy response in the patients with pan-cancer and immunotherapy. We explored the heterogeneity of five subclusters based on CAF in signaling patterns, immune status, metabolism features, enrichment pathways, and prognosis for PAAD.

Immune depletion of the methylated phenotype of colon cancer is closely related to resistance to immune checkpoint inhibitors.

BackgroundMolecular typing based on single omics data has its limitations and requires effective integration of multiple omics data for tumor typing of colorectal cancer (CRC).MethodsTranscriptome expression, DNA methylation, somatic mutation, clinicopathological information, and copy number variation were retrieved from TCGA, UCSC Xena, cBioPortal, FireBrowse, or GEO. After pre-processing and calculating the clustering prediction index (CPI) with gap statistics, integrative clustering analysis was conducted via MOVICS. The tumor microenvironment (TME) was deconvolved using several algorithms such as GSVA, MCPcounter, ESTIMATE, and PCA. The metabolism-relevant pathways were extracted through ssGSEA. Differential analysis was based on limma and enrichment analysis was carried out by Enrichr. DNA methylation and transcriptome expression were integrated via ELMER. Finally, nearest template or hemotherapeutic sensitivity prediction was conducted using NTP or pRRophetic.ResultsThree molecular subtypes (CS1, CS2, and CS3) were recognized by integrating transcriptome, DNA methylation, and driver mutations. CRC patients in CS3 had the most favorable prognosis. A total of 90 differentially mutated genes among the three CSs were obtained, and CS3 displayed the highest tumor mutation burden (TMB), while significant instability across the entire chromosome was observed in the CS2 group. A total of 30 upregulated mRNAs served as classifiers were identified and the similar diversity in clinical outcomes of CS3 was validated in four external datasets. The heterogeneity in the TME and metabolism-related pathways were also observed in the three CSs. Furthermore, we found CS2 tended to loss methylations while CS3 tended to gain methylations. Univariate and multivariate Cox regression revealed that the subtypes were independent prognostic factors. For the drug sensitivity analysis, we found patients in CS2 were more sensitive to ABT.263, NSC.87877, BIRB.0796, and PAC.1. By Integrating with the DNA mutation and RNA expression in CS3, we identified that SOX9, a specific marker of CS3, was higher in the tumor than tumor adjacent by IHC in the in-house cohort and public cohort.ConclusionThe molecular subtypes based on integrated multi-omics uncovered new insights into the prognosis, mechanisms, and clinical therapeutic targets for CRC.