Hasil untuk "Neoplasms. Tumors. Oncology. Including cancer and carcinogens"

S. Woo, J. Hellstein, J. Kalmar

Aleksandra Weronika Nielsen, Hafez Eslami Manoochehri, Hua Zhong et al.

The ability of tumors to evolve and adapt by developing subclones in different genetic and epigenetic states is a major challenge in oncology. Traditional tools like multi-regional sequencing used to study tumor evolution and the resultant intra-tumor heterogeneity (ITH) are often impractical because of their resource-intensiveness and limited scalability. Here, we present MorphoITH, a novel framework that leverages histopathology slides to deconvolve molecular ITH through tissue morphology. MorphoITH integrates a self-supervised deep learning similarity measure to capture phenotypic variation across multiple dimensions (cytology, architecture, and microenvironment) with rigorous methods to eliminate spurious sources of variation. Using a prototype of ITH, clear cell renal cell carcinoma (ccRCC), we show that MorphoITH captures clinically-significant biological features, such as vascular architecture and nuclear grades. Furthermore, we find that MorphoITH recognizes differential biological states corresponding to subclonal changes in key driver genes (BAP1/PBRM1/SETD2). Finally, by applying MorphoITH to a multi-regional sequencing experiment, we postulate evolutionary trajectories that largely recapitulate genetic evolution. In summary, MorphoITH provides a scalable phenotypic lens that bridges the gap between histopathology and genomics, advancing precision oncology.

S. Z. Pirasteh, Ali A. Kiaei, Mahnaz Bush et al.

Background: Gastric neoplasm, primarily adenocarcinoma, is an aggressive cancer with high mortality, often diagnosed late, leading to complications like metastasis. Effective drug combinations are vital to address disease heterogeneity, enhance efficacy, reduce resistance, and improve patient outcomes. Methods: The RAIN method integrated Graph SAGE to propose drug combinations, using a graph model with p-value-weighted edges connecting drugs, genes, and proteins. NLP and systematic literature review (PubMed, Scopus, etc.) validated proposed drugs, followed by network meta-analysis to assess efficacy, implemented in Python. Results: Oxaliplatin, fluorouracil, and trastuzumab were identified as effective, supported by 61 studies. Fluorouracil alone had a p-value of 0.0229, improving to 0.0099 with trastuzumab, and 0.0069 for the triple combination, indicating superior efficacy. Conclusion: The RAIN method, combining AI and network meta-analysis, effectively identifies optimal drug combinations for gastric neoplasm, offering a promising strategy to enhance treatment outcomes and guide health policy.

Pei Liu, Luping Ji, Jiaxiang Gou et al.

Whole-Slide Image (WSI) is an important tool for estimating cancer prognosis. Current studies generally follow a conventional cancer-specific paradigm in which each cancer corresponds to a single model. However, this paradigm naturally struggles to scale to rare tumors and cannot leverage knowledge from other cancers. While multi-task learning frameworks have been explored recently, they often place high demands on computational resources and require extensive training on ultra-large, multi-cancer WSI datasets. To this end, this paper shifts the paradigm to knowledge transfer and presents the first preliminary yet systematic study on cross-cancer prognosis knowledge transfer in WSIs, called CROPKT. It comprises three major parts. (1) We curate a large dataset (UNI2-h-DSS) with 26 cancers and use it to measure the transferability of WSI-based prognostic knowledge across different cancers (including rare tumors). (2) Beyond a simple evaluation merely for benchmarking, we design a range of experiments to gain deeper insights into the underlying mechanism behind transferability. (3) We further show the utility of cross-cancer knowledge transfer, by proposing a routing-based baseline approach (ROUPKT) that could often efficiently utilize the knowledge transferred from off-the-shelf models of other cancers. CROPKT could serve as an inception that lays the foundation for this nascent paradigm, i.e., WSI-based prognosis prediction with cross-cancer knowledge transfer. Our source code is available at https://github.com/liupei101/CROPKT.

Jun Wang, Godwin Imade, Alani S. Akanmu et al.

Abstract Objectives Cervical cancer is one of the most frequently diagnosed cancers and a leading cause of cancer-related deaths in women in low- and middle-income countries (LMICs), accounting for nearly 85% of the global cervical cancer burden. High-risk human papillomavirus (hrHPV) infection is the main cause of cervical cancer. Easy-to-use, rapid, scalable, high-throughput, and cost-effective HPV tests are urgently needed for low-resource settings. Atila Biosystems’ clinically validated ScreenFire HPV Risk Stratification (RS) assay identifies 13 hrHPV in 4 groups based on their oncogenic risk (i.e., HPV16, HPV18/45, HPV31/33/35/52/58, and HPV51/59/39/56/68). While the current standard format is subject to laboratory contamination Atila has developed an innovative, contamination-preventive Zebra BioDome format. Recently we published the analytical performance of ScreenFire RS Zebra BioDome on the BioRad CFX-96 real-time PCR instrument. This current study evaluated its analytical performance on three additional qPCR platforms: Atila Portable iAMP-PS96, Atila Powergene9600 Plus, and Thermo Fisher Quantstudio-7. Methods We tested 173 DNA samples from Nigerian women with cervical cancer. These samples were tested simultaneously using the ScreenFire HPV Zebra BioDome assay (M5FHPV-96) on four different real-time PCR machines (Atila portable iAMP-PS96, Atila Powergene9600 Plus, Thermo Fisher QuantStudio-7, and BioRad CFX-96). We used overall agreement rate and unweighted kappa values to compare different platforms. Results The overall agreement for detection of hrHPV using Atila portable iAMP-PS96 was 96.5% with kappa value 0.95 (95% confidence interval: 0.91–0.99) compared to Thermo Fisher QuantStudio-7, and 97.1% with kappa value 0.96 (95% confidence interval: 0.92–0.99) compared to BioRad CFX-96. For genotype HPV16 and risk stratification (RS) genotype groups (HPV18/45, HPV31/33/35/52/58, and HPV51/59/39/56/68) agreement rates were all > 98.3%. For Atila Powergene9600 Plus the overall agreement was 98.8% with a kappa value of 0.98 (95% confidence interval: 0.96–1.0) compared to Thermo Fisher QuantStudio-7, and 96.5% with a kappa value of 0.96 (95% confidence interval: 0.94–0.99) compared to BioRad CFX-96. The agreements for the HPV16 and RS genotype groups (HPV18/45, HPV31/33/35/52/58, and HPV39/51/56/59/68) were at least 98.3%. Conclusion The novel ScreenFire HPV Zebra BioDome format produced highly concordant hrHPV positivity and RS genotype results on all four qPCR platforms. The data suggests that this innovative technology has the potential to improve HPV testing uptake in low-resource settings without further investment in purchasing new equipment.

Ting-ting Liu, Zhi Zhang, Jing Deng et al.

Abstract The inflammatory microenvironment influences dendritic cell-mediated antigen presentation to regulate asthma Th2 inflammation. The scavenger receptor is expressed on DCs and regulates antigen presentation and T priming. However, whether the transmembrane scavenger receptor (SR-PSOX/CXCL16) regulates the phenotype and antigen presentation function of DCs remains unclear. We found that CXCL16 is mainly expressed on DCs in the lung tissues of asthma patients and asthma mice. CXCL16 knockout led to the suppression of airway inflammation, mucus overproduction, and airway hyperresponsiveness in Aspergillus-induced asthma. In addition, the adoptive transfer of Aspergillus-pulsed DCs shows the CXCL16+ DCs exerted a promoting role in airway inflammation, the CXCL16− DCs inhibit airway inflammation. Additionally, RNA sequencing and flow cytometry data revealed that CXCL16 knockout inhibits airway inflammation by suppressing the antigen processing and presentation function of DCs, which was mediated by MHC II chaperone H2-DM. Furthermore, we found CXCL16 knockout suppressed dendritic cells differentiated forward to cDC2b subtype which is mainly charged with antigen presentation to T cell. In conclusion, we found that CXCL16 downregulated the capacity of DC antigen processing and presentation to suppress airway inflammation by reducing H2-DM expression which mediated DC differentiation. The study suggested that inhibition of CXCL16 can be a potential therapy for asthma.

Fadillah Maani, Anees Ur Rehman Hashmi, Mariam Aljuboory et al.

Automated segmentation proves to be a valuable tool in precisely detecting tumors within medical images. The accurate identification and segmentation of tumor types hold paramount importance in diagnosing, monitoring, and treating highly fatal brain tumors. The BraTS challenge serves as a platform for researchers to tackle this issue by participating in open challenges focused on tumor segmentation. This study outlines our methodology for segmenting tumors in the context of two distinct tasks from the BraTS 2023 challenge: Adult Glioma and Pediatric Tumors. Our approach leverages two encoder-decoder-based CNN models, namely SegResNet and MedNeXt, for segmenting three distinct subregions of tumors. We further introduce a set of robust postprocessing to improve the segmentation, especially for the newly introduced BraTS 2023 metrics. The specifics of our approach and comprehensive performance analyses are expounded upon in this work. Our proposed approach achieves third place in the BraTS 2023 Adult Glioma Segmentation Challenges with an average of 0.8313 and 36.38 Dice and HD95 scores on the test set, respectively.

Rongqi Guo, Yihui Wei, Yating Du et al.

Death receptor-mediated extrinsic apoptosis system had been developed as a promising therapeutic strategy in clinical oncology, such as TRAIL therapy. However, multiple studies have demonstrated that TRAIL resistance is the biggest problem for disappointing clinical trials despite preclinical success. Targeting cellular FLICE inhibitory protein (cFLIP) is one strategy of combinatorial therapies to overcome resistance to DR-mediated apoptosis due to its negative regulator of extrinsic apoptosis. E × 527 (Selisistat) is a specific inhibitor of SIRT1 activity with safe and well tolerance in clinical trials. Here, we show that E × 527 could strengthen significantly activation of rhFasL-mediated apoptotic signaling pathway and increased apoptotic rate of T leukemia cells with high expression of cFLIP. Mechanically, Inhibition of SIRT1 by E × 527 increased polyubiquitination level of cFLIP via increasing acetylation of Ku70, which could promote proteosomal degradation of cFLIP protein. It implied that combinatorial therapies of E × 527 plus TRAIL may have a potential as a novel clinical application for TRAIL-resistant hematologic malignancies.

Feng Wang, Ruifang Pang, Xudong Zhao et al.

Background Early identification and therapy can significantly improve the outcome for gastric cancer. However, there is still no perfect biomarker available for the detection of early gastric cancer. This study aimed to investigate the alterations in the plasma metabolites of early gastric cancer using metabolomics and lipidomics based on high-performance liquid chromatography-mass spectrometry (HPLC-MS), which detected potential biomarkers that could be used for clinical diagnosis. Methods To investigate the changes in metabolomics and lipidomics, a total of 30 plasma samples were collected, consisting of 15 patients with early gastric cancer and 15 healthy controls. Extensive HPLC-MS-based untargeted metabolomic and lipidomic investigations were conducted. Differential metabolites and metabolic pathways were uncovered through the utilization of statistical analysis and bioinformatics analysis. Candidate biomarker screening was performed using support vector machine-based multivariate receiver operating characteristic analysis. Results A disturbance was observed in a combined total of 19 metabolites and 67 lipids of the early gastric cancer patients. The analysis of KEGG pathways showed that the early gastric cancer patients experienced disruptions in the arginine biosynthesis pathway, the pathway for alanine, aspartate, and glutamate metabolism, as well as the pathway for glyoxylate and dicarboxylate metabolism. Plasma metabolomics and lipidomics have identified multiple biomarker panels that can effectively differentiate early gastric cancer patients from healthy controls, exhibiting an area under the curve exceeding 0.9. Conclusion These metabolites and lipids could potentially serve as biomarkers for the screening of early gastric cancer, thereby optimizing the strategy for the detection of early gastric cancer. The disrupted pathways implicated in early gastric cancer provide new clues for additional understanding of gastric cancer's pathogenesis. Nonetheless, large-scale clinical data are required to prove our findings.

N. Ognerubov, M. A. Ognerubova

Background. Solid organ transplantation recipients have a high risk of non-melanoma skin tumors. Patients after heart transplantation are prone to a higher incidence of malignant skin tumors due to intensive immunosuppressive therapy. The most common histological type is squamous cell carcinoma, followed by basal cell carcinoma. These tumors have a more aggressive clinical course, including the frequency of recurrence and metastasis, and a tendency to multifocal lesions. Materials and methods. We present a clinical case of primary multiple squamous cell carcinoma with metastatic lesions of regional lymph nodes in a patient after heart transplantation. Results. A 67-year-old patient underwent an orthotopic heart transplant in September 2018 for ischemic cardiomyopathy. Subsequently, triple immunosuppressive therapy was administered, including tacrolimus combined with mycophenolate mofetil and prednisolone. In May 2022, a solid tumor with ulceration occurred on the skin of the right scapular region. After some time, similar tumors appeared on the skin of the temporal region on the left, the posterior surface of the auricle and the parietal region on the left. The patient later found a solid, painless tumor on the left jaw angle. As a part of the examination in the oncology dispensary, a biopsy of the scapular skin tumor, scrapings from tumors, and aspiration biopsy of the submandibular lymph nodes were performed. Histological and cytological studies of all neoplasms showed squamous cell keratinizing cancer with metastases to the submandibular lymph nodes. Additional examination methods showed no signs of progression. The diagnosis was made: primary multiple synchronous skin cancer: right scapular area, stage III, cT3N0M0; left parietal area, stage II, cT2N0M0; occipital area, stage I, cT1N0M0; left auricle, stage IV, cT1N2M0. Considering the localization of tumors, surgical treatment was performed, including of excision of tumors in the scapular and parietal regions. Radiation therapy was performed on lymph nodes with metastases. After 6 months, a tumor recurrence was detected in the irradiation area. Conclusion. After heart transplantation, squamous cell carcinoma of the skin is common. Usually, it affects the scalp and neck with metastases to the regional lymph nodes and is prone to recurrence. The primary treatment method is surgical and radiation therapy.

Arad Iranmehr, Fateme Jafari, Abolfazl Paeinmahali et al.

Renato Bellotti, Jonas Willmann, Antony J. Lomax et al.

Creating high quality treatment plans is crucial for a successful radiotherapy treatment. However, it demands substantial effort and special training for dosimetrists. Existing automated treatment planning systems typically require either an explicit prioritization of planning objectives, human-assigned objective weights, large amounts of historic plans to train an artificial intelligence or long planning times. Many of the existing auto-planning tools are difficult to extend to new planning goals. A new spot weight optimisation algorithm, called JulianA, was developed. The algorithm minimises a scalar loss function that is built only based on the prescribed dose to the tumour and organs at risk (OARs), but does not rely on historic plans. The objective weights in the loss function have default values that do not need to be changed for the patients in our dataset. The system is a versatile tool for researchers and clinicians without specialised programming skills. Extending it is as easy as adding an additional term to the loss function. JulianA was validated on a dataset of 19 patients with intra- and extracerebral neoplasms within the cranial region that had been treated at our institute. For each patient, a reference plan which was delivered to the cancer patient, was exported from our treatment database. Then JulianA created the auto plan using the same beam arrangement. The reference and auto plans were given to a blinded independent reviewer who assessed the acceptability of each plan, ranked the plans and assigned the human-/machine-made labels. The auto plans were considered acceptable in 16 out of 19 patients and at least as good as the reference plan for 11 patients. Whether a plan was crafted by a dosimetrist or JulianA was only recognised for 9 cases. The median time for the spot weight optimisation is approx. 2 min (range: 0.5 min - 7 min).

Reza Khanmohammadi, Mohammad M. Ghassemi, Kyle Verdecchia et al.

Natural Language Processing (NLP) is a key technique for developing Medical Artificial Intelligence (AI) systems that leverage Electronic Health Record (EHR) data to build diagnostic and prognostic models. NLP enables the conversion of unstructured clinical text into structured data that can be fed into AI algorithms. The emergence of the transformer architecture and large language models (LLMs) has led to remarkable advances in NLP for various healthcare tasks, such as entity recognition, relation extraction, sentence similarity, text summarization, and question answering. In this article, we review the major technical innovations that underpin modern NLP models and present state-of-the-art NLP applications that employ LLMs in radiation oncology research. However, these LLMs are prone to many errors such as hallucinations, biases, and ethical violations, which necessitate rigorous evaluation and validation before clinical deployment. As such, we propose a comprehensive framework for assessing the NLP models based on their purpose and clinical fit, technical performance, bias and trust, legal and ethical implications, and quality assurance, prior to implementation in clinical radiation oncology. Our article aims to provide guidance and insights for researchers and clinicians who are interested in developing and using NLP models in clinical radiation oncology.

Shorya Consul, John Robertson, Haris Vikalo

It is estimated that approximately 15% of cancers worldwide can be linked to viral infections. The viruses that can cause or increase the risk of cancer include human papillomavirus, hepatitis B and C viruses, Epstein-Barr virus, and human immunodeficiency virus, to name a few. The computational analysis of the massive amounts of tumor DNA data, whose collection is enabled by the recent advancements in sequencing technologies, have allowed studies of the potential association between cancers and viral pathogens. However, the high diversity of oncoviral families makes reliable detection of viral DNA difficult and thus, renders such analysis challenging. In this paper, we introduce XVir, a data pipeline that relies on a transformer-based deep learning architecture to reliably identify viral DNA present in human tumors. In particular, XVir is trained on genomic sequencing reads from viral and human genomes and may be used with tumor sequence information to find evidence of viral DNA in human cancers. Results on semi-experimental data demonstrate that XVir is capable of achieving high detection accuracy, generally outperforming state-of-the-art competing methods while being more compact and less computationally demanding.

Zifan Chen, Jiazheng Li, Jie Zhao et al.

**Background:** Accurate 3D CT scan segmentation of gastric tumors is pivotal for diagnosis and treatment. The challenges lie in the irregular shapes, blurred boundaries of tumors, and the inefficiency of existing methods. **Purpose:** We conducted a study to introduce a model, utilizing human-guided knowledge and unique modules, to address the challenges of 3D tumor segmentation. **Methods:** We developed the PropNet framework, propagating radiologists' knowledge from 2D annotations to the entire 3D space. This model consists of a proposing stage for coarse segmentation and a refining stage for improved segmentation, using two-way branches for enhanced performance and an up-down strategy for efficiency. **Results:** With 98 patient scans for training and 30 for validation, our method achieves a significant agreement with manual annotation (Dice of 0.803) and improves efficiency. The performance is comparable in different scenarios and with various radiologists' annotations (Dice between 0.785 and 0.803). Moreover, the model shows improved prognostic prediction performance (C-index of 0.620 vs. 0.576) on an independent validation set of 42 patients with advanced gastric cancer. **Conclusions:** Our model generates accurate tumor segmentation efficiently and stably, improving prognostic performance and reducing high-throughput image reading workload. This model can accelerate the quantitative analysis of gastric tumors and enhance downstream task performance.

Okyaz Eminaga, Mahmoud Abbas, Christian Kunder et al.

Prostate cancer pathology plays a crucial role in clinical management but is time-consuming. Artificial intelligence (AI) shows promise in detecting prostate cancer and grading patterns. We tested an AI-based digital twin of a pathologist, vPatho, on 2,603 histology images of prostate tissue stained with hematoxylin and eosin. We analyzed various factors influencing tumor-grade disagreement between vPatho and six human pathologists. Our results demonstrated that vPatho achieved comparable performance in prostate cancer detection and tumor volume estimation, as reported in the literature. Concordance levels between vPatho and human pathologists were examined. Notably, moderate to substantial agreement was observed in identifying complementary histological features such as ductal, cribriform, nerve, blood vessels, and lymph cell infiltrations. However, concordance in tumor grading showed a decline when applied to prostatectomy specimens (kappa = 0.44) compared to biopsy cores (kappa = 0.70). Adjusting the decision threshold for the secondary Gleason pattern from 5% to 10% improved the concordance level between pathologists and vPatho for tumor grading on prostatectomy specimens (kappa from 0.44 to 0.64). Potential causes of grade discordance included the vertical extent of tumors toward the prostate boundary and the proportions of slides with prostate cancer. Gleason pattern 4 was particularly associated with discordance. Notably, grade discordance with vPatho was not specific to any of the six pathologists involved in routine clinical grading. In conclusion, our study highlights the potential utility of AI in developing a digital twin of a pathologist. This approach can help uncover limitations in AI adoption and the current grading system for prostate cancer pathology.

Kwang-Hyun Uhm, Hyunjun Cho, Zhixin Xu et al.

In 2023, it is estimated that 81,800 kidney cancer cases will be newly diagnosed, and 14,890 people will die from this cancer in the United States. Preoperative dynamic contrast-enhanced abdominal computed tomography (CT) is often used for detecting lesions. However, there exists inter-observer variability due to subtle differences in the imaging features of kidney and kidney tumors. In this paper, we explore various 3D U-Net training configurations and effective post-processing strategies for accurate segmentation of kidneys, cysts, and kidney tumors in CT images. We validated our model on the dataset of the 2023 Kidney and Kidney Tumor Segmentation (KiTS23) challenge. Our method took second place in the final ranking of the KiTS23 challenge on unseen test data with an average Dice score of 0.820 and an average Surface Dice of 0.712.

Fatma Nihan Akkoc Mustafayev, Diane D. Liu, Angelica M. Gutierrez et al.

Objective:Risk-reducing therapy with selective estrogen receptor (ER) modulators and aromatase inhibitors reduce breast cancer risk. However, the effects are limited to ER-positive breast cancer. Therefore, new agents with improved toxicity profiles that reduce the risk in ER-negative breast cancers are urgently needed. The aim of this prospective, short-term, prevention study was to evaluate the effect of dasatinib, an inhibitor of the tyrosine kinase Src, on biomarkers in normal (but increased risk) breast tissue and serum of women at high risk for a second, contralateral primary breast cancer.Materials and Methods:Women with a history of unilateral stage I, II, or III ER-negative breast cancer, having no active disease, and who completed all adjuvant therapies were eligible. Patients underwent baseline fine-needle aspiration (FNA) of the contralateral breast and serum collection for biomarker analysis and were randomized to receive either no treatment (control) or dasatinib at 40 or 80 mg/day for three months. After three months, serum collection and breast FNA were repeated. Planned biomarker analysis consisted of changes in cytology and Ki-67 on breast FNA, and changes in serum levels of insulin-like growth factor 1 (IGF-1), IGF-binding protein 1, and IGF-binding protein 3. The primary objective was to evaluate changes in Ki-67 and secondary objective included changes in cytology in breast tissue and IGF-related serum biomarkers. Toxicity was also evaluated.Results:Twenty-three patients started their assigned treatments. Compliance during the study was high, with 86.9% (20/23) of patients completing their assigned doses. Dasatinib was well tolerated and no drug-related grade 3 and 4 adverse events were observed. Since only one patient met the adequacy criteria for the paired FNA sample, we could not evaluate Ki-67 level or cytological changes. No significant change in serum biomarkers was observed among the three groups.Conclusion:Dasatinib was well tolerated but did not induce any significant changes in serum biomarkers. The study could not fulfill its primary objective due to an inadequate number of paired FNA samples. Further, larger studies are needed to evaluate the effectiveness of Src inhibitors in breast cancer prevention.

Yue Han, Haowei Wang, Hui Chen et al.

Chemotherapy is one of the most commonly treatments of advanced colorectal cancer (CRC). However, the drug resistant following chemotherapeutic treatment is a significant challenge in the clinical management of CRC. Therefore, understanding the resistance mechanisms and developing new strategies for enhancing the sensitivity are urgently needed to improve CRC outcome. Connexins contribute to the formation of gap junctions among neighboring cells and then advance gap junctional intercellular communication (GJIC) for transportation of ions and small molecules. Although the drug resistance resulted from GJIC dysfunctional by aberrant expression of connexins is relatively well understood, the underlying mechanisms of mechanical stiffness mediated by connexin responsible for chemoresistance are largely unknown in CRC. Here, we demonstrated that connexin 43 (CX43) expression was downregulated in CRC and that loss of CX43 expression was positively correlated with metastasis and poor prognosis of CRC patients. The CX43 overexpressing suppressed CRC progression and increased the sensitivity to 5-fluorouracil (5-FU) via enhanced GJIC in vitro and in vivo. Moreover, we also highlight that the downregulation of CX43 in CRC increases the stemness of cells via reducing the cell stiffness, thus promoting the drug resistance. Our results further suggest that both effects, that is changes in the mechanical stiffness of the cell and GJIC mediated by CX43 deregulated, are closely related to drug resistance in CRC, which indicating CX43 as a target against cancer growth and chemoresistance in CRC.

Hai Yang, Yuhang Sheng, Yi Jiang et al.

Motivation: Cancer is heterogeneous, affecting the precise approach to personalized treatment. Accurate subtyping can lead to better survival rates for cancer patients. High-throughput technologies provide multiple omics data for cancer subtyping. However, precise cancer subtyping remains challenging due to the large amount and high dimensionality of omics data. Results: This study proposed Subtype-Former, a deep learning method based on MLP and Transformer Block, to extract the low-dimensional representation of the multi-omics data. K-means and Consensus Clustering are also used to achieve accurate subtyping results. We compared Subtype-Former with the other state-of-the-art subtyping methods across the TCGA 10 cancer types. We found that Subtype-Former can perform better on the benchmark datasets of more than 5000 tumors based on the survival analysis. In addition, Subtype-Former also achieved outstanding results in pan-cancer subtyping, which can help analyze the commonalities and differences across various cancer types at the molecular level. Finally, we applied Subtype-Former to the TCGA 10 types of cancers. We identified 50 essential biomarkers, which can be used to study targeted cancer drugs and promote the development of cancer treatments in the era of precision medicine.