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

Polycarp Nalela, Deepthi Rao, Praveen Rao

Cancer remains a leading global health challenge and a major cause of mortality. This study leverages machine learning (ML) to predict the survivability of cancer patients with metastatic patterns using the comprehensive MSK-MET dataset, which includes genomic and clinical data from 25,775 patients across 27 cancer types. We evaluated five ML models-XGBoost, Naïve Bayes, Decision Tree, Logistic Regression, and Random Fores using hyperparameter tuning and grid search. XGBoost emerged as the best performer with an area under the curve (AUC) of 0.82. To enhance model interpretability, SHapley Additive exPlanations (SHAP) were applied, revealing key predictors such as metastatic site count, tumor mutation burden, fraction of genome altered, and organ-specific metastases. Further survival analysis using Kaplan-Meier curves, Cox Proportional Hazards models, and XGBoost Survival Analysis identified significant predictors of patient outcomes, offering actionable insights for clinicians. These findings could aid in personalized prognosis and treatment planning, ultimately improving patient care.

Si‐Yuan Chen, Peng‐Peng Xu, Ru Feng et al.

Abstract Background Diffuse large B‐cell lymphoma (DLBCL) is the most common subtype of aggressive non‐Hodgkin's lymphoma with distinct clinical and molecular heterogeneity. DLBCL that arises in extranodal organs is particularly linked to poor prognosis. This study aimed to determine the clinical and molecular characteristics of extranodal involvement (ENI) in DLBCL and assess the actual survival status of the patients. Methods In this population‐based cohort study, we investigated the clinical features of 5,023 patients newly diagnosed with DLBCL. Their clinical conditions, eligibility criteria, and sociodemographic details were recorded and analyzed. Gene panel sequencing was performed on 1,050 patients to discern molecular patterns according to ENI. Results The 2‐year overall survival (OS) rate was 76.2% [95% confidence interval (CI), 74.0%‐78.2%], and the 5‐year OS rate was 67.9% (95% CI, 65.2%‐70.4%). The primary treatment was immunochemotherapy with rituximab. Specific lymphoma involvement sites, especially the bones, bone marrow, and central nervous system, were identified as independent adverse prognostic factors. A high prevalence of non‐germinal center B‐cell (non‐GCB) phenotype and myeloid differentiation primary response 88 (MYD88)/CD79B mutations were noted in lymphomas affecting the breasts, skin, uterus, and immune‐privileged sites. Conversely, the thyroid and gastrointestinal tract showed a low occurrence of non‐GCB phenotype. Remarkably, patients with multiple ENIs exhibited a high frequency of MYD88, tet methylcytosine dioxygenase 2 (TET2), CREB binding protein (CREBBP) mutations, increased MYD88L265P and CD79B mutation (MCD)‐like subtypes, and poor prognosis. Genetic subtype‐guided immunochemotherapy showed good efficacy in subgroup analyses after propensity score matching with 5‐year OS and progression‐free survival rates of 85.0% (95% CI, 80.6%‐89.5%) and 72.1% (95% CI, 67.3%‐76.7%). Conclusions In the rituximab era, this large‐scale retrospective analysis from Asia confirmed the poor prognosis of DLBCL with multiple ENIs and underscored the efficacy of genetic subtype‐guided immunochemotherapy in treating extranodal DLBCL.

Hongjie Tang, Hairong Hao, Yue Han

PurposeSepsis is a leading cause of mortality, especially among immunocompromised patients with lung cancer. We aimed to establish machine learning (ML) based model to accurately forecast ICU mortality in patients with sepsis combined lung cancer.MethodsWe incorporated patients with sepsis combined lung cancer from Medical Information Mart for Intensive Care IV (MIMIC IV) database. Univariate and multivariate logistic analysis were employed to select variables. Recursive Feature Elimination (RFE) method based on 6 ML algorithms was used for feature selection. We harnessed 13 ML algorithms to construct prediction model, which were assessed by area under the curve (AUC), accuracy, sensitivity, specificity, precision, cross-entropy and Brier scores. The best ML model was constructed to predict ICU mortality, and the predictive results were interpretated by SHapley Additive exPlanations (SHAP) framework.ResultsA sum of 1096 lung cancer patients combined sepsis from MIMIC IV database and 251 patients from the external validation set were included. We utilized 13 clinical variables to establish prediction model for ICU mortality. CatBoost model was identified as the prime prediction model with the highest AUC in the training (0.931 [0.921, 0.945]), internal validation (0.698 [0.673, 0.724]) and external validation (0.794 [0.725, 0.879]) cohorts. Oxford Acute Severity of Illness Score (OASIS) had the greatest influence on ICU mortality according to SHAP interpretation.ConclusionsOur ML models demonstrate excellent accuracy and reliability, facilitating more rigorous personalized prognostic forecast to lung cancer patients combined sepsis.

Yuejia Zhao, Simeng Zhao, Zhe Fan et al.

Abstract In recent years, the incidence of thyroid cancer has steadily increased. However, the detailed mechanisms of pathogenesis are still unclear. Therefore, a comprehensive understanding of the underlying carcinogenesis mechanisms of thyroid cancer is required. Programmed cell death (PCD) is a cell death process mediated by specific molecular program, regulated by specific genes within the cell. Accumulating evidence suggests that PCD plays an indispensable role in thyroid cancer, maintaining intracellular stability by regulating genes and eliminating damaged or aged cells. In this review, we summarize six identified forms of PCD, analyze biomarkers for different PCD pathways in thyroid cancer, and briefly elucidate the roles of various PCD pathways in targeted therapies for thyroid cancers with a poor prognosis. We also provide an outlook on future treatments for drug-resistant thyroid cancer, poorly differentiated thyroid cancer, and iodine-refractory thyroid cancer, aiming to accurately identify targets and offer effective targeted therapeutic strategies.

Ying Li, Yu-Meng Liu, Yu-Lin Gao et al.

Abstract Background Post-hepatectomy liver failure (PHLF) is a leading cause of perioperative mortality following liver resection. Early detection and prediction of clinically relevant post-hepatectomy liver failure (CR-PHLF) remain critical but challenging. Lactate has shown promise as a biomarker, but its predictive power when combined with other factors remains unclear. Methods This study retrospectively analyzed 915 patients who underwent liver resection at Zhejiang Provincial People’s Hospital. Variables including demographics, liver function markers, intraoperative blood loss, and postoperative lactate levels were assessed. Multivariate logistic regression identified significant predictors for CR-PHLF, and a nomogram was created. The model’s performance was evaluated using ROC curves and decision curve analysis. Results In this study, Multivariate logistic regression was applied to select 6 predictors from the relevant variables, which were gender, ICGR-15, intraoperative blood loss, transfusion, resection extent, and lactate. In the training set, the AUC of the model was 0.781, significantly outperforming traditional models like ALBI and APRI. In the validation set, the model’s AUC was 0.812, indicating robust predictive accuracy. Conclusions The integrated model combining lactate and intraoperative factors provides a more accurate prediction of CR-PHLF risk. It outperforms existing models and has significant potential for improving preoperative risk assessment and intraoperative decision-making.

Şentürk A, Harmantepe AT, Gonullu E et al.

Adem Şentürk,1 Ahmet Tarik Harmantepe,2 Emre Gonullu,2 Alp Omer Canturk,3 Fuldem Mutlu,4 Onur Taydas4 1Sakarya University Training and Research Hospital, Department of Surgical Oncology, Sakarya, Turkey; 2Sakarya University Training and Research Hospital, Department of Gastroenterology Surgery, Sakarya, Turkey; 3Sakarya University Training and Research Hospital, Department of General Surgery, Sakarya, Turkey; 4Sakarya University Training and Research Hospital, Department of Radiology, Sakarya, TurkeyCorrespondence: Adem Şentürk, Sakarya University Training and Research Hospital, Sakarya, Turkey, Tel +905360290724, Email dr.adem.senturk@gmail.comBackground: The prognostic value of negative surgical margins in soft tissue sarcomas in terms of disease course is well known. However, there is a lack of consensus in the literature regarding the impact of preoperative radiological surgical margins on recurrence rates and overall survival The aim of the present study was to determine whether soft tissue density at the margin of abdominal sarcomas using Hounsfield Unit (HU) measurement on CT is associated with recurrence after tumor resection.Material and Methods: Seventeen patients who underwent resectional surgery for abdominal sarcoma between May 2014 and May 2024 were retrospectively analyzed. Patients were compared with their preoperative CT scans for postoperative local recurrence according to soft tissue density at the margins of the sarcomas.Results: Of the 17 patients, nine (52.9%) had recurrence. No significant difference was found for gender in terms of recurrence (p> 0.05). As the median age decreases, recurrence increases significantly. (60 years (23– 70) vs 73 years (44– 79); p= 0.044). Increased preoperative tissue density (width 3 to 5 cm) at sarcoma margin measured by CT was significantly associated with recurrence after tumor resection (with at margin: 3cm; p=0.047, 4cm; p=0.019, 5 cm; p=0.018). The cut-of value of density measured by preoperative CT for soft tissue at sarcoma margin with recurrence was − 98.8 hounsfield Unit (HU), whereas cut-of value of density was − 109.6 hU with a 91.5% sensitivity, 58.9% specificity, 23.2% positive predictive value (PPV), 76.8% negative predictive value (NPV), and 0.83 accuracy, respectively.Conclusion: Study results suggest that the risk of recurrence after tumor resection can be predicted by measuring soft tissue density at the sarcoma margin on preoperative CT scans.There appears to be a linear relationship between increased preoperative soft tissue density at the sarcoma margin and recurrence after tumor resection. This measurement method offers a perspective that reveals a new approach to this subject. Multicenter studies consisted of larger patient populations are needed to reach a definitive conclusion.Keywords: soft tissue sarcoma, recurrence, computerized tomography, Hounsfield Unit

Zi Li, Ying Chen, Zeli Chen et al.

In the radiation therapy of nasopharyngeal carcinoma (NPC), clinicians typically delineate the gross tumor volume (GTV) using non-contrast planning computed tomography to ensure accurate radiation dose delivery. However, the low contrast between tumors and adjacent normal tissues necessitates that radiation oncologists manually delineate the tumors, often relying on diagnostic MRI for guidance. % In this study, we propose a novel approach to directly segment NPC gross tumors on non-contrast planning CT images, circumventing potential registration errors when aligning MRI or MRI-derived tumor masks to planning CT. To address the low contrast issues between tumors and adjacent normal structures in planning CT, we introduce a 3D Semantic Asymmetry Tumor segmentation (SATs) method. Specifically, we posit that a healthy nasopharyngeal region is characteristically bilaterally symmetric, whereas the emergence of nasopharyngeal carcinoma disrupts this symmetry. Then, we propose a Siamese contrastive learning segmentation framework that minimizes the voxel-wise distance between original and flipped areas without tumor and encourages a larger distance between original and flipped areas with tumor. Thus, our approach enhances the sensitivity of features to semantic asymmetries. % Extensive experiments demonstrate that the proposed SATs achieves the leading NPC GTV segmentation performance in both internal and external testing, \emph{e.g.}, with at least 2\% absolute Dice score improvement and 12\% average distance error reduction when compared to other state-of-the-art methods in the external testing.

Opeyemi Sheu Alamu, Bismar Jorge Gutierrez Choque, Syed Wajeeh Abbs Rizvi et al.

Breast cancer remains a significant global health challenge, with prognosis and treatment decisions largely dependent on clinical characteristics. Accurate prediction of patient outcomes is crucial for personalized treatment strategies. This study employs survival analysis techniques, including Cox proportional hazards and parametric survival models, to enhance the prediction of the log odds of survival in breast cancer patients. Clinical variables such as tumor size, hormone receptor status, HER2 status, age, and treatment history were analyzed to assess their impact on survival outcomes. Data from 1557 breast cancer patients were obtained from a publicly available dataset provided by the University College Hospital, Ibadan, Nigeria. This dataset was preprocessed and analyzed using both univariate and multivariate approaches to evaluate survival outcomes. Kaplan-Meier survival curves were generated to visualize survival probabilities, while the Cox proportional hazards model identified key risk factors influencing mortality. The results showed that older age, larger tumor size, and HER2-positive status were significantly associated with an increased risk of mortality. In contrast, estrogen receptor positivity and breast-conserving surgery were linked to better survival outcomes. The findings suggest that integrating these clinical variables into predictive models improvesthe accuracy of survival predictions, helping to identify high-risk patients who may benefit from more aggressive interventions. This study demonstrates the potential of survival analysis in optimizing breast cancer care, particularly in resource-limited settings. Future research should focus on integrating genomic data and real-world clinical outcomes to further refine these models.

Jacqueline Lammert, Nicole Pfarr, Leonid Kuligin et al.

Rare gynecological tumors (RGTs) present major clinical challenges due to their low incidence and heterogeneity. The lack of clear guidelines leads to suboptimal management and poor prognosis. Molecular tumor boards accelerate access to effective therapies by tailoring treatment based on biomarkers, beyond cancer type. Unstructured data that requires manual curation hinders efficient use of biomarker profiling for therapy matching. This study explores the use of large language models (LLMs) to construct digital twins for precision medicine in RGTs. Our proof-of-concept digital twin system integrates clinical and biomarker data from institutional and published cases (n=21) and literature-derived data (n=655 publications with n=404,265 patients) to create tailored treatment plans for metastatic uterine carcinosarcoma, identifying options potentially missed by traditional, single-source analysis. LLM-enabled digital twins efficiently model individual patient trajectories. Shifting to a biology-based rather than organ-based tumor definition enables personalized care that could advance RGT management and thus enhance patient outcomes.

Chi-en Amy Tai, Alexander Wong

Breast cancer was diagnosed for over 7.8 million women between 2015 to 2020. Grading plays a vital role in breast cancer treatment planning. However, the current tumor grading method involves extracting tissue from patients, leading to stress, discomfort, and high medical costs. A recent paper leveraging volumetric deep radiomic features from synthetic correlated diffusion imaging (CDI$^s$) for breast cancer grade prediction showed immense promise for noninvasive methods for grading. Motivated by the impact of CDI$^s$ optimization for prostate cancer delineation, this paper examines using optimized CDI$^s$ to improve breast cancer grade prediction. We fuse the optimized CDI$^s$ signal with diffusion-weighted imaging (DWI) to create a multiparametric MRI for each patient. Using a larger patient cohort and training across all the layers of a pretrained MONAI model, we achieve a leave-one-out cross-validation accuracy of 95.79%, over 8% higher compared to that previously reported.

Zian Kang, Yuqing Jin, Huihui Yu et al.

Abstract Background Novel antibody-drug conjugates (ADCs) drugs present a promising anti-cancer treatment, although survival benefits for HER2-positive advanced breast cancer (BC) remain controversial. The aim of this meta-analysis was to evaluate the comparative effect of ADCs and other anti-HER2 therapy on progression-free survival (PFS) and overall survival (OS) for treatment of HER2-positive locally advanced or metastatic BC. Methods Relevant randomized controlled trials (RCTs) were retrieved from five databases. The risk of bias was assessed with the Cochrane Collaboration’s tool for RCTs by RevMan5.4 software. The hazard ratio (HR) and 95% confidence intervals (CIs) were extracted to evaluate the benefit of ADCs on PFS and OS in HER2-positive advanced BC by meta-analysis. Results Meta-analysis of six RCTs with 3870 patients revealed that ADCs significantly improved PFS (HR: 0.63, 95% CI: 0.49–0.80, P = 0.0002) and OS (HR: 0.79, 95% CI: 0.72–0.86, P < 0.0001) of patients with HER2-positive locally advanced or metastatic BC. Subgroup analysis showed that PFS and OS were obviously prolonged for patients who previously received HER2-targeted therapy. Sensitivity analysis and publication bias suggested that the results were stable and reliable. Conclusion Statistically significant benefits for PFS and OS were observed with ADCs in HER2-positive locally advanced or metastatic BC, especially for those who received prior anti-HER2 treatment.

Min Cen, Xingyu Li, Bangwei Guo et al.

In digital pathology tasks, transformers have achieved state-of-the-art results, surpassing convolutional neural networks (CNNs). However, transformers are usually complex and resource intensive. In this study, we developed a novel and efficient digital pathology classifier called DPSeq, to predict cancer biomarkers through fine-tuning a sequencer architecture integrating horizon and vertical bidirectional long short-term memory (BiLSTM) networks. Using hematoxylin and eosin (H&E)-stained histopathological images of colorectal cancer (CRC) from two international datasets: The Cancer Genome Atlas (TCGA) and Molecular and Cellular Oncology (MCO), the predictive performance of DPSeq was evaluated in series of experiments. DPSeq demonstrated exceptional performance for predicting key biomarkers in CRC (MSI status, Hypermutation, CIMP status, BRAF mutation, TP53 mutation and chromosomal instability [CING]), outperforming most published state-of-the-art classifiers in a within-cohort internal validation and a cross-cohort external validation. Additionally, under the same experimental conditions using the same set of training and testing datasets, DPSeq surpassed 4 CNN (ResNet18, ResNet50, MobileNetV2, and EfficientNet) and 2 transformer (ViT and Swin-T) models, achieving the highest AUROC and AUPRC values in predicting MSI status, BRAF mutation, and CIMP status. Furthermore, DPSeq required less time for both training and prediction due to its simple architecture. Therefore, DPSeq appears to be the preferred choice over transformer and CNN models for predicting cancer biomarkers.

Jiuqi Yan

Artificial intelligence is constantly evolving and can provide effective help in all aspects of people's lives. The experiment is mainly to study the use of artificial intelligence in the field of medicine. The purpose of this experiment was to compare which of MobileNetV1 and MobileNetV2 models was better at detecting histopathological images of the breast downloaded at Kaggle. When the doctor looks at the pathological image, there may be errors that lead to errors in judgment, and the observation speed is slow. Rational use of artificial intelligence can effectively reduce the error of doctor diagnosis in breast cancer judgment and speed up doctor diagnosis. The dataset was downloaded from Kaggle and then normalized. The basic principle of the experiment is to let the neural network model learn the downloaded data set. Then find the pattern and be able to judge on your own whether breast tissue is cancer. In the dataset, benign tumor pictures and malignant tumor pictures have been classified, of which 198738 are benign tumor pictures and 78, 786 are malignant tumor pictures. After calling MobileNetV1 and MobileNetV2, the dataset is trained separately, the training accuracy and validation accuracy rate are obtained, and the image is drawn. It can be observed that MobileNetV1 has better validation accuracy and overfit during MobileNetV2 training. From the experimental results, it can be seen that in the case of processing this dataset, MobileNetV1 is much better than MobileNetV2.

Liumei Shou MD, PhD, Tianyu Shao MD, Jialu Chen MD et al.

Objectives: To compare the effects of the Shouzu Ning Decoction (SND) and Halometasone plus Celecoxib (Hal/Cxb) as therapy in patients with grade 2 hand-foot skin reaction (HFSR). Materials and Methods: Fifty patients with grade 2 HFSR participated in a randomized, single-center, open-label study. Patients were randomly assigned in a 1:1 ratio to receive the SND or Hal/Cxb treatment, twice daily for 4 weeks, followed by 4 weeks of post-treatment follow-up. The primary endpoint was clinical remission of HFSR at the end of the fourth week (W4). The secondary endpoints were recurrence rate, quality of life (QoL), pain intensity, and safety. Results: In this study, 46 patients successfully completed the study, and 4 patients were excluded. There was no statistically significant difference between the 2 groups on demographic and baseline clinical characteristics. In the SND group, 56.52% of patients showed clinical remission at W4, which was significantly superior to that achieved in the Hal/Cxb group (26.09%, P  = .036). In addition, the HF-QoL score was statistically lower in the SND group compared to the Hal/Cxb group at W2 ( P  = .007), W3 ( P  = .005), and W4 ( P  = .005), respectively. In line with this, the inter-group difference in NRS score was statistically significant ( P  = .004). Conclusion: In the present study, SND treatment has been observed to be effective and well tolerated for patients with grade 2 HFSR. Thus, SND treatment could be considered a suitable option for HFSR patients. Trial registration: Chinese Clinical Trial Registry, ChiCTR1900027518. Registered on 17 Nov 2019.

J.I. Martínez Salamanca, G. Maiolino, E. Compérat et al.

Mirza Mumtaz Zahoor, Saddam Hussain Khan

Brain tumor classification is crucial for clinical analysis and an effective treatment plan to cure patients. Deep learning models help radiologists to accurately and efficiently analyze tumors without manual intervention. However, brain tumor analysis is challenging because of its complex structure, texture, size, location, and appearance. Therefore, a novel deep residual and regional-based Res-BRNet Convolutional Neural Network (CNN) is developed for effective brain tumor (Magnetic Resonance Imaging) MRI classification. The developed Res-BRNet employed Regional and boundary-based operations in a systematic order within the modified spatial and residual blocks. Moreover, the spatial block extract homogeneity and boundary-defined features at the abstract level. Furthermore, the residual blocks employed at the target level significantly learn local and global texture variations of different classes of brain tumors. The efficiency of the developed Res-BRNet is evaluated on a standard dataset; collected from Kaggle and Figshare containing various tumor categories, including meningioma, glioma, pituitary, and healthy images. Experiments prove that the developed Res-BRNet outperforms the standard CNN models and attained excellent performances (accuracy: 98.22%, sensitivity: 0.9811, F-score: 0.9841, and precision: 0.9822) on challenging datasets. Additionally, the performance of the proposed Res-BRNet indicates a strong potential for medical image-based disease analyses.

Yuexin Xu, Alicia J. Morales, Andrea M. H. Towlerton et al.

Objective responses of metastatic renal cell carcinoma (RCC) associated with systemic immunotherapies suggest the potential for T-cell-mediated tumor clearance. Recent analyses associate clonally expanded T cells present in the tumor at diagnosis with responses to immune checkpoint inhibitors (ICIs). To identify and further characterize tumor-associated, clonally expanded T cells, we characterized the density, spatial distribution, T-cell receptor (TCR) repertoire, and transcriptome of tumor-infiltrating T cells from 14 renal tumors at the time of resection and compared them with T cells in peripheral blood and normal adjacent kidney. Multiplex immunohistochemistry revealed that T-cell density was higher in clear cell RCC (ccRCC) than in other renal tumor histologies with spatially nonuniform T-cell hotspots and exclusion zones. TCR repertoire analysis also revealed increased clonal expansion in ccRCC tumors compared with non-clear cell histologies or normal tissues. Expanded T-cell clones were most frequently CD8+ with some detectable in peripheral blood or normal kidney and others found exclusively within the tumor. Divergent expression profiles for chemokine receptors and ligands and the Ki67 proliferation marker distinguished tumor-restricted T-cell clones from those also present in blood suggesting a distinct phenotype for subsets of clonally expanded T cells that also differed for upregulated markers of T-cell activation and exhaustion. Thus, our single-cell level stratification of clonally expanded tumor infiltrating T-cell subpopulations provides a framework for further analysis. Future studies will address the spatial orientation of these clonal subsets within tumors and their association with treatment outcomes for ICIs or other therapeutic modalities.

Maria E. Figueroa, Amy E. DeZern

Long JM, Ebrahimzadeh J, Stanich PP et al.

Jessica M Long,1,&ast; Jessica Ebrahimzadeh,1,&ast; Peter P Stanich,2 Bryson W Katona3 1Division of Hematology and Oncology, Penn Medicine, Philadelphia, PA, USA; 2Division of Gastroenterology, Hepatology & Nutrition, The Ohio State University, Wexner Medical Center, Columbus, OH, USA; 3Division of Gastroenterology and Hepatology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA&ast;These authors contributed equally to this workCorrespondence: Bryson W Katona, Division of Gastroenterology and Hepatology, Perelman School of Medicine, University of Pennsylvania, 3400 Civic Center Blvd, 751 South Pavilion, Philadelphia, PA, 19104, USA, Tel +1-215-349-8222, Fax +1-215-349-5915, Email bryson.katona@pennmedicine.upenn.eduAbstract: Gastric cancer is one of the most significant causes of cancer-related morbidity and mortality worldwide. Recognized modifiable risk factors include Helicobacter pylori infection, geographic location, select dietary factors, tobacco use and alcohol consumption. In addition, multiple hereditary cancer predisposition syndromes are associated with significantly elevated gastric cancer risk. Endoscopic surveillance in hereditary gastric cancer predisposition syndromes has the potential to identify gastric cancer at earlier and more treatable stages, as well as to prevent development of gastric cancer through identification of precancerous lesions. However, much uncertainty remains regarding use of endoscopic surveillance in hereditary gastric cancer predisposition syndromes, including whether or not it should be routinely performed, the surveillance interval and age of initiation, cost-effectiveness, and whether surveillance ultimately improves survival from gastric cancer for these high-risk individuals. In this review, we outline the hereditary gastric cancer predisposition syndromes associated with the highest gastric cancer risks. Additionally, we cover current evidence and guidelines addressing hereditary gastric cancer risk and surveillance in these syndromes, along with current challenges and limitations that emphasize a need for continued research in this field.Keywords: hereditary diffuse gastric cancer syndrome, Lynch syndrome, familial adenomatous polyposis, Peutz-Jeghers syndrome, juvenile polyposis syndrome, Li-Fraumeni syndrome

David Oniani, Chen Wang, Yiqing Zhao et al.

Cancer is responsible for millions of deaths worldwide every year. Although significant progress has been achieved in cancer medicine, many issues remain to be addressed for improving cancer therapy. Appropriate cancer patient stratification is the prerequisite for selecting appropriate treatment plan, as cancer patients are of known heterogeneous genetic make-ups and phenotypic differences. In this study, built upon deep phenotypic characterizations extractable from Mayo Clinic electronic health records (EHRs) and genetic test reports for a collection of cancer patients, we developed a system leveraging a joint of phenotypic and genetic features for cancer patient subgrouping. The workflow is roughly divided into three parts: feature preprocessing, cancer patient classification, and cancer patient clustering based. In feature preprocessing step, we performed filtering, retaining the most relevant features. In cancer patient classification, we utilized joint categorical features to build a patient-feature matrix and applied nine different machine learning models, Random Forests (RF), Decision Tree (DT), Support Vector Machine (SVM), Naive Bayes (NB), Logistic Regression (LR), Multilayer Perceptron (MLP), Gradient Boosting (GB), Convolutional Neural Network (CNN), and Feedforward Neural Network (FNN), for classification purposes. Finally, in the cancer patient clustering step, we leveraged joint embeddings features and patient-feature associations to build an undirected feature graph and then trained the cancer feature node embeddings.