检索结果-内蒙古大学图书馆

Suboptimal Engraftment and Toxicity after Post-Transplant Cyclophosphamide

Transplantation and Cellular Therapy 2025年第2期31卷 S141-S142页

作者： Scarpetti, Lauren, MD Kim, Haesook T., PhD Ikeda, Daniel Collier, Kerry Chung, Jooho DeFilipp, Zachariah, MD El-Jawahri, Areej R., MD Frigault, Matthew J., MD McAfee, Steven L., MD Newcomb, Richard Andrew, MD Spitzer, Thomas R, MD Westervelt, Peter O'Donnell, Paul Chen, Yi-Bin, MD Hematopoetic Cell Transplant & Cell Therapy Program Massachusetts General Hospital Boston MA Department of Data Science Dana-Farber Cancer Institute Harvard School of Public Health Boston MA Massachusetts General Hospital Boston MA Hematopoietic Cell Transplant & Cell Therapy Program Massachusetts General Hospital Boston MA Hematopoietic Cell Transplant and Cellular Therapy Program Massachusetts General Hospital Boston MA Medicine Division of Hematology & Oncology Massachusetts General Hospital Boston MA Blood and Marrow Transplant Program Massachusetts General Hospital Boston MA Hematopoietic Cell Transplant and Cell Therapy Program Massachusetts General Hospital Boston MA

BackgroundPost-transplant cyclophosphamide (PTCy) after hematopoietic cell transplantation (HCT) to prevent graft-versus-host disease (GVHD) is increasingly considered standard. We conducted a single-center retrospective cohort study to evaluate suboptimal engraftment (SE) and toxicity after PTCy. MethodsPatients receiving PTCy from Feb 2016 to Apr 2023 were included. Exclusion criteria were history of aplasia after chimeric antigen receptor T-cell therapy, prior HCT complicated by graft failure and 2 nd HCT. The primary objective was to define the incidence of specific early events defined as suboptimal engraftment (SE): primary or secondary graft failure, death before neutrophil (NE) or platelet engraftment and NE 3 21 days. The secondary objective was identification of clinical factors associated with SE including early organ toxicity. Results170 patients met eligibility criteria. HCT was most often for acute myeloid leukemia (AML; 38%) and myelodysplastic syndrome (MDS; 12%). Patients received reduced intensity (60%), non-myeloablative (24%), and myeloablative (17%) conditioning. Grafts were haploidentical (49%), mismatched unrelated (30%), matched unrelated (13%) and matched related (8%). Most grafts were peripheral blood progenitor cells (86%). Median follow-up for survivors was 47 months. 35 patients (21%) experienced an SE event. 5 (3%) patients had primary graft failure, and 4 (2%) had secondary graft failure. 6 (4%) died before NE: 3 from infection, 2 from renal failure, and 1 from cirrhosis. Median time to NE was 16 days (12-29). 19 patients achieved NE ≥ 21 days after HCT (11% of all NE). 4 (2%) died before platelet engraftment - 2 also had delayed NE. Median time to platelet engraftment was 25 days (10-194). Univariable analysis showed age ≥ 70 (OR 2.58, p = 0.044) and diagnosis of MDS (OR 3.03, p = 0.044) were significantly associated with SE. Myeloablative conditioning was protective against SE (OR 0.118, p = 0.039). When analysis was restricted to AM

关键词：

来源：评论

学校读者我要写书评

暂无评论

A randomized phase 2 study of neratinib with or without fulvestrant for patients with HER2-positive, estrogen receptor-positive metastatic breast cancer

引用

Clinical Breast cancer 2025年

作者： Stefania Morganti Xiangying Chu Tarah J. Ballinger Nisha Unni Sarah Sinclair Robert Wesolowski Alyssa M. Pereslete Paulina Lange Nabihah Tayob Nancy U. Lin Jose P. Leone Ian E. Krop Sara M. Tolaney Heather A. Parsons Department of Medical Oncology Dana-Farber Cancer Institute Boston MA Breast Oncology Program Dana-Farber Brigham Cancer Center Boston MA Harvard Medical School Boston MA Broad Institute of MIT and Harvard Cambridge MA Data Science Dana-Farber Cancer Institute Boston MA Indiana University School of Medicine Indianapolis IN University of Texas Southwestern Medical Center Dallas TX Northern Light Cancer Care Brewer ME Department of Medicine The Ohio State University College of Medicine Columbus OH Florida International University Herbert Wertheim College of Medicine Miami FL Boston Medical Center Boston MA Yale Cancer Center New Haven CT

Background Most HER2-positive breast cancers co-express estrogen receptor (ER). Given crosstalk between HER2 and ER signaling pathways, dual blockade may be beneficial. Methods In this randomized, open-label, phase 2 clinical trial, patients with ER-positive (ER≥10%), HER2-positive metastatic breast cancer were randomized (1:1) to neratinib (240mg daily) or the same dose of neratinib with fulvestrant. Any number of prior therapies was allowed; prior trastuzumab, pertuzumab and trastuzumab emtansine were required. The primary endpoint was progression-free survival (PFS). Secondary endpoints included overall survival (OS), overall response rate, and duration of response. Exploratory objectives included the identification of predictive biomarkers via circulating tumor DNA (ctDNA). Results Of 21 patients enrolled, 18 were evaluable for outcomes and safety (neratinib-fulvestrant arm, n=8; neratinib-only arm, n=10). The study was closed before completing enrollment due to slow accrual. Median PFS did not differ between treatment arms (2.79 months with neratinib-fulvestrant versus 5.55 months with neratinib only [HR 0.94; 95% CI 0.24-3.64; P=0.98]). Grade 3 adverse events occurred in 1 (12.5%) patient in the neratinib-fulvestrant arm and 6 (60%) patients in the neratinib-only arm, with diarrhea being the most frequent. Median OS did not differ between the two arms (p=0.91). Clearance of ctDNA was associated with PFS and OS. Conclusions The combination of neratinib and fulvestrant is safe and tolerable. Due to early study closure, this study was underpowered to detect the benefit of adding fulvestrant to neratinib. Chemotherapy-free regimens targeting ER and HER2 warrant further investigation, along with prospective studies investigating ctDNA dynamics may guide treatment switch.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Feasibility of Distress Thermometer Screening for Caregivers of Patients Undergoing Inpatient or Outpatient Hematopoietic Stem Cell Transplantation (HSCT)

引用

Transplantation and Cellular Therapy 2025年第2期31卷 S493-S494页

作者： Gray, Tamryn Fowler, RN, PhD, MPH Heiling, Hillary, PhD Mazzola, Emanuele, PhD Amonoo, Hermioni Lokko, MD, MPP Sullivan, Lauren, MSN, RN, AGCNS-BC, OCN, BMTCN Kelkar, Amar H., MD, MPH, FACP Pirl, William, MD, MPH Hammer, Marilyn J. Tulsky, James A., MD El-Jawahri, Areej, M.D. Cutler, Corey S., MD MPH Partridge, Ann H., MD, MPH School of Nursing UNC Chapel Hill Chapel Hill NC Data Science Dana-Farber Cancer Institute Boston MA Dana-Farber Cancer Institute Boston MA Harvard Medical School Boston MA Commercial AbbVie North Chicago IL Department of Medical Oncology Dana-Farber Cancer Institute Boston MA Psychosocial Oncology and Palliative Care Dana-Farber Cancer Institute Boston MA Supportive Oncology Dana-Farber Cancer Institute Boston MA Medicine Division of Hematology & Oncology Massachusetts General Hospital Boston MA Department of Medical Oncology Dana-Farber Cancer Institute and Harvard Medical School Boston MA Division of Transplantation and Cellular Therapy Department of Medical Oncology Dana-Farber Cancer Institute Boston MA Stem Cell/Bone Marrow Transplantation Program Dana-Farber Cancer Institute Boston MA Division of Hematological Malignancies Department of Medical Oncology Dana Farber Cancer Institute Boston MA

Topic Significance & Study Purpose/Background/RationaleCaregivers of adults undergoing HSCT face considerable emotional challenges, yet few are screened for distress and limited data exists on whether caregiver distress differs between inpatient or outpatient HSCT. The aim of this study is to assess the feasibility of using the Distress Thermometer (DT) and Problem List to screen for distress in caregivers of HSCT recipients at Day -6 before HSCT and Days +1, +8, +15, and +28 post-HSCT. Methods, Intervention, & AnalysisIn this longitudinal study conducted between 2021-2023, caregivers completed the FamilyStrong caregiver distress thermometer adapted from the NCCN DT and Problem List, including 36 items organized within practical, family, spiritual/religious concerns, and physical problems. Caregivers were asked to rate their distress on a 0-10 numeric scale. Feasibility was defined as enrollment of ≥ 70% of eligible caregivers, and completion of ≥60% of study assessments. Findings & InterpretationAmong 86 eligible, 60 HCT caregivers enrolled (70% female, 35.0% aged 60-69, 96.7% White, 68.3% spouse/partnered, and 60% outpatient caregivers). DT completion rates across five time points were 100.0%, 91.7%, 93.3%, 80.0%, and 86.7, respectively. Within the sample, Day +1 post-HCT had the largest proportion of caregivers reporting high distress (72.7% (n=40/55) followed by Day +8 with 67.9% (n=38/56). Among inpatient caregivers, 63.6% (n=14/22) had high distress on Day +1 and 68.4% (n=13/19) reported high distress on Day +15. Among outpatient caregivers, 78.8% (n=26/33) had high distress on Day +1 and 79.4% (n=27/34) reported high distress on Day +8. The most common concerns at baseline were practical issues (66.7% in both inpatient and outpatient caregivers), information and skill needs (54.2% inpatient, 63.9% outpatient), and emotional challenges (54.2% inpatient, 50% outpatient). Discussion & ImplicationsThe adapted FamilyStrong caregiver DT is feasible to use and demon

关键词：

来源：评论

学校读者我要写书评

暂无评论

A T cell resilience model associated with response to immunotherapy in multiple tumor types (May, 10.1038/s41591-022-01799-y, 2022)

引用

NATURE MEDICINE 2022年第10期28卷 2219-2219页

作者： Zhang, Yu Vu, Trang Palmer, Douglas C. Kishton, Rigel J. Gong, Lanqi Huang, Jiao Nguyen, Thanh Chen, Zuojia Smith, Cari Livak, Ferenc Paul, Rohit Day, Chi-Ping Wu, Chuan Merlino, Glenn Aldape, Kenneth Guan, Xin-yuan Jiang, Peng Cancer Data Science Laboratory Center for Cancer Research National Cancer Institute National Institutes of Health Bethesda MD USA Department of Clinical Oncology The University of Hong Kong Hong Kong China Sun Yat-sen University Cancer Center State Key Laboratory of Oncology in South China Collaborative Innovation Center for Cancer Medicine Guangzhou China Surgery Branch National Cancer Institute National Institutes of Health Bethesda MD USA Laboratory of Pathology Center for Cancer Research National Cancer Institute National Institutes of Health Bethesda MD USA Experimental Immunology Branch Center for Cancer Research National Cancer Institute National Institutes of Health Bethesda MD USA Laboratory Animal Science Program Leidos Biomedical Research Inc Frederick MD USA Flow Cytometry Core Center for Cancer Research National Cancer Institute National Institutes of Health Bethesda MD USA Office of the Director Center for Cancer Research National Cancer Institute National Institutes of Health Bethesda MD USA Laboratory of Cancer Biology and Genetics Center for Cancer Research National Cancer Institute National Institutes of Health Bethesda MD USA

Despite breakthroughs in cancer immunotherapy, most tumor-reactive T cells cannot persist in solid tumors due to an immunosuppressive environment. We developed Tres (tumor-resilient T cell, https://***/), a computational model utilizing single-cell transcriptomic data to identify signatures of T cells that are resilient to immunosuppressive signals, such as transforming growth factor-β1, tumor necrosis factor-related apoptosis-inducing ligand and prostaglandin E2. Tres reliably predicts clinical responses to immunotherapy in melanoma, lung cancer, triple-negative breast cancer and B cell malignancies using bulk T cell transcriptomic data from pre-treatment tumors from patients who received immune-checkpoint inhibitors (n = 38), infusion products for chimeric antigen receptor T cell therapies (n = 34) and pre-manufacture samples for chimeric antigen receptor T cell or tumor-infiltrating lymphocyte therapies (n = 84). Further, Tres identified FIBP, whose functions are largely unknown, as the top negative marker of tumor-resilient T cells across many solid tumor types. FIBP knockouts in murine and human donor CD8+ T cells significantly enhanced T cell-mediated cancer killing in in vitro co-cultures. Further, Fibp knockout in murine T cells potentiated the in vivo efficacy of adoptive cell transfer in the B16 tumor model. Fibp knockout T cells exhibit reduced cholesterol metabolism, which inhibits effector T cell function. These results demonstrate the utility of Tres in identifying biomarkers of T cell effectiveness and potential therapeutic targets for immunotherapies in solid tumors.

关键词： cancer immunotherapy Computational biology and bioinformatics Predictive markers T cells

来源：评论

学校读者我要写书评

暂无评论

Core variables for real-world clinicogenomic data collection in precision oncology

ESMO Real World Data and Digital Oncology

引用

ESMO Real World data and Digital Oncology 2025年 7卷 100117-100117页

作者： R. Dienstmann A. Hackshaw J.-Y. Blay C. Le Tourneau OC Precision Medicina Oncoclínicas & Co São Paulo Brazil PhD Program Doctoral School University of Vic—Central University of Catalonia Barcelona Spain Oncology Data Science Vall d’Hebron Institute of Oncology Barcelona Spain Cancer Research UK and UCL Cancer Trials Centre University College London London UK Centre Léon Bérard and Université Claude Bernard Lyon France Institut Curie Department of Drug Development and Innovation (D3i) Paris-Saclay University Paris France

Background Precision oncology evidence gaps may be bridged using real-world clinicogenomic data; however, current limitations compromise real-world data collection and evidence generation. We aimed to define a set of precision oncology core variables that can aid in fit-for-purpose real-world data collection to be used for a range of purposes and by various stakeholders based on the approach adopted when designing the WAYFIND-R registry (NCT04529122). Materials and methods A panel of precision oncology experts created a list of variables based on standard dictionaries and vocabularies, which was aligned with the European Medical Agency’s draft guidelines for registry-based studies. A list of core variables was selected from the initial broad list of variables. Results The initial overall list (∼500 variables) was consolidated to ∼150 variables that covered the entirety of the patient journey in oncology. The panel gave highest priority to patient demographics, socioeconomic information and comorbidities, cancer details, molecular information, particularly predictive biomarkers in routine use and next-generation sequencing technical aspects, systemic cancer therapies and other treatments, outcome assessments and survival outcomes. Conclusions The core oncology variables represent a harmonized list of key clinicogenomic data elements that can be collected during a patient’s journey in oncology and leveraged for their primary intended purpose of molecular tumor board decision making, and might be used for further secondary research purposes, if legal requirements allow. Definition of this list of core variables will help generate evidence from research- or regulatory-oriented precision oncology studies by ensuring synergy, dataset connectivity, and interoperability across different data.

关键词： precision oncology real-world data solid tumors next-generation sequencing molecular tumor boards

来源：评论

学校读者我要写书评

暂无评论

Special Section Guest Editorial: LUNGx Challenge for computerized lung nodule classification: Reflections and lessons learned

引用

Journal of Medical Imaging 2015年第2期2卷

作者： Armato, Samuel G. Hadjiiski, Lubomir Tourassi, Georgia D. Drukker, Karen Giger, Maryellen L. Li, Feng Redmond, George Farahani, Keyvan Kirby, Justin S. Clarke, Laurence P. University of Chicago Department of Radiology MC 2026 5841 S. Maryland Avenue ChicagoIL60637 United States University of Michigan Department of Radiology 1500 E. Medical Center Drive Ann ArborMI48109 United States Biomedical Science and Engineering Center Health Data Sciences Institute Oak Ridge National Laboratory Oak RidgeTN37831 United States National Cancer Institute Division of Cancer Treatment and Diagnosis Cancer Imaging Program 9609 Medical Center Drive BethesdaMD20892 United States Frederick National Laboratory for Cancer Research Leidos Biomedical Research Inc Cancer Imaging Program FrederickMD21702 United States

来源：评论

学校读者我要写书评

暂无评论

Multimodal Co-Attention Transformer for Survival Prediction in Gigapixel Whole Slide Images

Multimodal Co-Attention Transformer for Survival Prediction ...

引用

International Conference on Computer Vision (ICCV)

作者： Richard J. Chen Ming Y. Lu Wei-Hung Weng Tiffany Y. Chen Drew FK. Williamson Trevor Manz Maha Shady Faisal Mahmood Brigham and Women’s Hospital Harvard Medical School Cancer Program Broad Institute of Harvard and MIT Cancer Data Science Program Dana-Farber Cancer Institute Computer Science and Artificial Intelligence Laboratory MIT

ISBN: (纸本)9781665428132

Survival outcome prediction is a challenging weakly-supervised and ordinal regression task in computational pathology that involves modeling complex interactions within the tumor microenvironment in gigapixel whole slide images (WSIs). Despite recent progress in formulating WSIs as bags for multiple instance learning (MIL), representation learning of entire WSIs remains an open and challenging problem, especially in overcoming: 1) the computational complexity of feature aggregation in large bags, and 2) the data heterogeneity gap in incorporating biological priors such as genomic measurements. In this work, we present a Multimodal Co-Attention Transformer (MCAT) framework that learns an interpretable, dense co-attention mapping between WSIs and genomic features formulated in an embedding space. Inspired by approaches in Visual Question Answering (VQA) that can attribute how word embed-dings attend to salient objects in an image when answering a question, MCAT learns how histology patches attend to genes when predicting patient survival. In addition to visualizing multimodal interactions, our co-attention trans-formation also reduces the space complexity of WSI bags, which enables the adaptation of Transformer layers as a general encoder backbone in MIL. We apply our proposed method on five different cancer datasets (4,730 WSIs, 67 million patches). Our experimental results demonstrate that the proposed method consistently achieves superior performance compared to the state-of-the-art methods.

关键词： Representation learning Visualization Histopathology Genomics Predictive models Transformers Biological information theory

来源：评论

学校读者我要写书评

暂无评论

Phase I Study of Ixazomib Added to Chemotherapy in the Treatment of Acute Lymphoblastic Leukemia in Older Adults

引用

BLOOD 2020年 136卷 41-42页

作者： Amrein, Philip C. Ballen, Karen K. Stevenson, Kristen E. Blonquist, Traci M. Brunner, Andrew M. Hobbs, Gabriela S. Hock, Hanno R. McAfee, Steven L. Moran, Jenna A. Bergeron, Meghan Foster, Julia E. Bertoli, Christina McGregor, Kristin Macrae, Molly Burke, Meghan Behnan, Tanya T. Som, Tina T. Ramos, Aura Y. Vartanian, Megan K. Story, Jennifer Lombardi Connolly, Christine Graubert, Timothy A. Neuberg, Donna S. Fathi, Amir T. Leukemia Center/MGH Cancer Center Massachusetts General Hospital Belmont MA Division of Hematology/Oncology University of Virginia Health Center Charlottesville VA Department of Data Science Dana-Farber Cancer Institute Boston MA Data Science Dana-Farber Cancer Institute Boston MA Center for Leukemia Massachusetts General Hospital Boston MA Leukemia Center Massachusetts General Hospital Boston MA Leukemia Center Massachusetts General Hospital / Harvard Medical School Boston MA Massachusetts General Hospital Blood and Marrow Transplant Program Boston MA Leukemia Center Massachusetts General Hospital Center for Leukemia Cambridge MA

Introduction: While progress has been made in the treatment of childhood leukemia, the outlook for patients >60 years of age with acute lymphoblastic leukemia (ALL) is poor with complete remission rates (CR) of approximately 60% and 3-year survivals (OS) of less than 15%. Intensified treatment in a later CALGB trial showed little improvement with a CR=61% and 5-year OS=6% (Stock, cancer 2013). Ixazomib is an oral proteasome inhibitor, which has shown single agent activity and promising combination activity in pediatric ALL patients (Messinger, Blood 2012). We sought to assess the safety and tolerability, as well as early efficacy of adding ixazomib to a current MGH-DFCI/HCC multi-agent regimen for older adults with ALL. Methods: Patients aged 51 to 75 years of age with newly diagnosed B-ALL and T-ALL were screened for eligibility. Patients with mature ALL (including Burkitt's) were excluded. Patients with Philadelphia chromosome positive ALL (BCR-ABL1+) were eligible, and dasatinib was added to the chemotherapy on Day 10 for these patients. The chemotherapy treatment schedule from induction through maintenance is outlined in Table 1. A standard 3 + 3 patient cohort dose escalation design was used to determine the maximum tolerated dose (MTD) of ixazomib during induction for these patients, the primary objective of the trial. After consolidation I, patients in complete remission (CR) with a suitable donor were offered a hematopoietic stem cell transplantation (HSCT) as per institutional guidelines. Those not going to HSCT continued therapy as noted in the table. Results: There were 19 patients with B-ALL enrolled, none with T-ALL. Among these patients, 7 harbored BCR-ABL1 rearrangements. The median age was 65 years, 74% were male, and 90% had a performance status 0 or 1. The MTD was 2.3 mg of ixazomib, as 2 patients at 3.0 mg developed DLT's: a grade 3 peripheral neuropathy and a grade 5 acute kidney injury (Table 2). Grade 3 and 4 toxicities encountered at any time

关键词：

来源：评论

学校读者我要写书评

暂无评论

Choosing between HLA-Mismatched Unrelated and Haploidentical Donors: Donor Age Considerations

引用

Transplantation and Cellular Therapy 2025年第2期31卷 S258-S259页

作者： Mehta, Rohtesh S., MD Wang, Tao, PhD Schmidt, Gabrielle, MPH Gadalla, Shahinaz M., MD, PhD Betts, Brian Christopher, MD Benjamin, Cara L., PhD Arrieta-Bolaños, Esteban, MQC, PhD Bolon, Yung-Tsi Lee, Stephanie J, MD, MPH Department of Stem Cell Transplantation and Cellular Therapy The University of Texas MD Anderson Cancer Center Houston TX CIBMTR® (Center for International Blood and Marrow Transplant Research) Medical College of Wisconsin Milwaukee WI Division of Biostatistics Data Science Institute Medical College of Wisconsin Milwaukee WI Center for International Blood and Marrow Transplant Research (CIBMTR) National Marrow Donor Program/Be The Match Minneapolis MN Clinical Genetics Branch； Division of Cancer Epidemiology and Genetics National Cancer Institute NIH Bethesda MD Roswell Park Comprehensive Cancer Center Buffalo NY Department of Medicine Univeristy of Miami Miller School of Medicine Miami FL Institute for Experimental Cellular Therapy University Hospital Essen Essen Düsseldorf Germany CIBMTR® (Center for International Blood and Marrow Transplant Research) NMDP Minneapolis MN Clinical Research Division Fred Hutchinson Cancer Center Seattle WA

BACKGROUNDHLA-mismatched unrelated donor (mmURD) and haploidentical donor (haplo) are the two most common mismatched donor options for hematopoietic cell transplantation (HCT). Substantial evidence suggests that donor age significantly influences HCT outcomes. However, a direct comparison of these two donor choices while accounting for donor age is lacking. METHODSWe compared the outcomes of mmURD and haplo donor HCT with post-transplantation cyclophosphamide (PTCy)-based graft-versus-host disease (GVHD) prophylaxis using data from the Center for International Blood and Marrow Transplant Research. Donors in each group were categorized as younger (<30 years) vs. older (≥30 years) based on the maximal likelihood ratio test. The groups compared were: ‘younger’ haplo (n=1870) vs. ‘older’ mmURD (n=460), and ‘older’ haplo (n=4186) vs. ‘younger’ mmURD (n=600). All patients received PTCy; in vivo or ex vivo T cell depletion was excluded. Most patients had acute leukemia in all groups (65-73%), intermediate disease risk index (58-66%) and received peripheral blood graft (76-82%) [Baseline characteristics, Fig 1]. The primary outcome was GVHD-free relapse-free survival (GRFS), a composite of death, relapse, grade III-IV acute GVHD (aGVHD), and systemic therapy-requiring chronic GVHD (cGVHD). Secondary endpoints were aGVHD, cGVHD, relapse, non-relapse mortality (NRM), progression-free survival, and overall survival. To adjust for multiple testing, a significance level of 0.01 was used for statistical association of the main variable. RESULTSYounger haplo (ref) vs. older mmURD: The 3-year GRFS was 33% (95% confidence interval 31-36) vs. 24% (20-29), respectively. In multivariate analyses [Fig 2], the older mmURD group had significantly worse GRFS (HR 1.20, p=.003) [Fig 3], higher risk of NRM (HR 1.49, p=.005), and grade III-IV aGVHD (HR 2.88, p=.003). Older haplo (ref) vs. younger mmURD: The 3-year GRFS was 24% (23-26) vs. 29% (25-33), respectively, with HR 0.87, p=.02 in multi

关键词：

来源：评论

学校读者我要写书评

暂无评论

Whole slide images are 2D point clouds: Context-aware survival prediction using patch-based graph convolutional networks

arXiv

引用

arXiv 2021年

作者： Chen, Richard J. Lu, Ming Y. Shaban, Muhammad Chen, Chengkuan Chen, Tiffany Y. Williamson, Drew F.K. Mahmood, Faisal Department of Pathology Brigham and Women's Hospital Department of Biomedical Informatics Harvard Medical School Cancer Data Science Program Dana-Farber Cancer Institute Cancer Program Broad Institute of Harvard MIT

cancer prognostication is a challenging task in computational pathology that requires context-aware representations of histology features to adequately infer patient survival. Despite the advancements made in weakly-supervised deep learning, many approaches are not context-aware and are unable to model important morphological feature interactions between cell identities and tissue types that are prognostic for patient survival. In this work, we present Patch-GCN, a context-aware, spatially-resolved patch-based graph convolutional network that hierarchically aggregates instance-level histology features to model local- and global-level topological structures in the tumor microenvironment. We validate Patch-GCN with 4,370 gigapixel WSIs across five different cancer types from the cancer Genome Atlas (TCGA), and demonstrate that Patch-GCN outperforms all prior weakly-supervised approaches by 3.58-9.46%. Our code and corresponding models are publicly available at https://***/mahmoodlab/Patch-GCN. Copyright © 2021, The Authors. All rights reserved.

关键词： Pathology

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：