scholarly journals NIMG-09. PREDICTING OVERALL SURVIVAL OF GLIOBLASTOMA PATIENTS ON MULTI-INSTITUTIONAL HISTOPATHOLOGY STAINED SLIDES USING DEEP LEARNING AND POPULATION-BASED NORMALIZATION

2020 ◽  
Vol 22 (Supplement_2) ◽  
pp. ii148-ii148
Author(s):  
Jie Hao ◽  
Jose Agraz ◽  
Caleb Grenko ◽  
Ji Won Park ◽  
Angela Viaene ◽  
...  
Keyword(s):  
Overall Survival ◽  
Deep Learning ◽  
Clinical Data ◽  
Clinical Decision Making ◽  
Population Based ◽  
Adult Brain ◽  
The Cancer Genome Atlas ◽  
World Health ◽  
Feature Maps ◽  
Testing Data

Abstract MOTIVATION Glioblastoma is the most common and aggressive adult brain tumor. Clinical histopathologic evaluation is essential for tumor classification, which according to the World Health Organization is associated with prognostic information. Accurate prediction of patient overall survival (OS) from clinical routine baseline histopathology whole slide images (WSI) using advanced computational methods, while considering variations in the staining process, could contribute to clinical decision-making and patient management optimization. METHODS We utilize The Cancer Genome Atlas glioblastoma (TCGA-GBM) collection, comprising multi-institutional hematoxylin and eosin (H&E) stained frozen top-section WSI, genomic, and clinical data from 121 subjects. Data are randomly split into training (80%), validation (10%), and testing (10%) sets, while proportionally keeping the ratio of censored patients. We propose a novel deep learning algorithm to identify survival-discriminative histopathological patterns in a WSI, through feature maps, and quantitatively integrate them with gene expression and clinical data to predict patient OS. The concordance index (C-index) is used to quantify the predictive OS performance. Variations in slide staining are assessed through a novel population-based stain normalization approach, informed of glioblastoma distinct histologic sub-regions and their appearance from 509 H&E stained slides with corresponding anatomical annotations from the Ivy Glioblastoma Atlas Project (IvyGAP). RESULTS C-index was equal to 0.797, 0.713, and 0.703 for the training, validation, and testing data, respectively, prior to stain normalization. Following normalization, staining variations in H&E and ‘E’ gained significant improvements in IvyGAP (pWilcoxon< 0.01) and TCGA-GBM (pWilcoxon< 0.0001) data, respectively. These improvements contributed to further optimizing the C-index to 0.871, 0.777, and 0.780 for the training, validation, and testing data, respectively. CONCLUSIONS Appropriate normalization and integrative deep learning yield accurate OS prediction of glioblastoma patients through H&E slides, generalizable in multi-institutional data, potentially contributing to patient stratification in clinical trials. Our computationally-identified survival-discriminative histopathological patterns can contribute in further understanding glioblastoma.

scholarly journals Multimodal 3D DenseNet for IDH Genotype Prediction in Gliomas

Genes ◽  
2018 ◽  
Vol 9 (8) ◽  
pp. 382 ◽  
Author(s):  
Sen Liang ◽  
Rongguo Zhang ◽  
Dayang Liang ◽  
Tianci Song ◽  
Tao Ai ◽  
...  
Keyword(s):  
Deep Learning ◽  
Clinical Decision Making ◽  
Three Dimensional ◽  
Area Under The Curve ◽  
Image Data ◽  
Clinical Decision ◽  
The Cancer Genome Atlas ◽  
World Health ◽  
Validation Dataset ◽  
Who Grade

Non-invasive prediction of isocitrate dehydrogenase (IDH) genotype plays an important role in tumor glioma diagnosis and prognosis. Recently, research has shown that radiology images can be a potential tool for genotype prediction, and fusion of multi-modality data by deep learning methods can further provide complementary information to enhance prediction accuracy. However, it still does not have an effective deep learning architecture to predict IDH genotype with three-dimensional (3D) multimodal medical images. In this paper, we proposed a novel multimodal 3D DenseNet (M3D-DenseNet) model to predict IDH genotypes with multimodal magnetic resonance imaging (MRI) data. To evaluate its performance, we conducted experiments on the BRATS-2017 and The Cancer Genome Atlas breast invasive carcinoma (TCGA-BRCA) dataset to get image data as input and gene mutation information as the target, respectively. We achieved 84.6% accuracy (area under the curve (AUC) = 85.7%) on the validation dataset. To evaluate its generalizability, we applied transfer learning techniques to predict World Health Organization (WHO) grade status, which also achieved a high accuracy of 91.4% (AUC = 94.8%) on validation dataset. With the properties of automatic feature extraction, and effective and high generalizability, M3D-DenseNet can serve as a useful method for other multimodal radiogenomics problems and has the potential to be applied in clinical decision making.

scholarly journals Prognostic Association between Common Laboratory Tests and Overall Survival in Elderly Men with De Novo Metastatic Castration Sensitive Prostate Cancer: A Population-Based Study in Canada

Cancers ◽  
2021 ◽  
Vol 13 (11) ◽  
pp. 2844
Author(s):  
Christopher J. D. Wallis ◽  
Bobby Shayegan ◽  
Scott C. Morgan ◽  
Robert J. Hamilton ◽  
Ilias Cagiannos ◽  
...  
Keyword(s):  
Prostate Cancer ◽  
Overall Survival ◽  
Clinical Decision Making ◽  
Cox Regression ◽  
De Novo ◽  
Population Based ◽  
Population Based Study ◽  
Cox Regression Analysis ◽  
Lymphocyte Ratio ◽  
Laboratory Markers

De novo cases of metastatic prostate cancer (mCSPC) are associated with poorer prognosis. To assist in clinical decision-making, we aimed to determine the prognostic utility of commonly available laboratory-based markers with overall survival (OS). In a retrospective population-based study, a cohort of 3556 men aged ≥66 years diagnosed with de novo mCSPC between 2014 and 2019 was identified in Ontario (Canada) administrative database. OS was assessed by using the Kaplan–Meier method. Multivariate Cox regression analysis was performed to evaluate the association between laboratory markers and OS adjusting for patient and disease characteristics. Laboratory markers that were assessed include neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), albumin, hemoglobin, serum testosterone and PSA kinetics. Among the 3556 older men with de novo mCSPC, their median age was 77 years (IQR: 71–83). The median survival was 18 months (IQR: 10–31). In multivariate analysis, a statistically significant association with OS was observed with all the markers (NLR, PLR, albumin, hemoglobin, PSA decrease, reaching PSA nadir and a 50% PSA decline), except for testosterone levels. Our findings support the use of markers of systemic inflammation (NLR, PLR and albumin), hemoglobin and PSA metrics as prognostic indicators for OS in de novo mCSPC.

Incidence and Outcome of Chronic Myeloproliferative Disorders: A Population-Based Study from a Cancer Registry in Northern Italy.

Blood ◽  
2007 ◽  
Vol 110 (11) ◽  
pp. 4650-4650
Author(s):  
Alessia Bari ◽  
Raffaella Marcheselli ◽  
Ivan Rashid ◽  
Goretta Bonacorsi ◽  
Orsola Bonanno ◽  
...  
Keyword(s):  
Overall Survival ◽  
Cancer Registry ◽  
Relative Survival ◽  
Epidemiological Data ◽  
Myeloproliferative Disorders ◽  
Population Based ◽  
Age At Diagnosis ◽  
World Health ◽  
Pathology Report ◽  
Chronic Myeloproliferative Disorders

Abstract Background Because in the past Chronic Myeloproliferative Disorders (CMPD) were not considered to be malignant conditions, cancer registries rarely recorded data on these diseases. Thus, information on incidence and outcome of CMPD in the population is limited. The aim of the present study was to better define epidemiological data of CMPD by examining all cases identified by the Modena Cancer Registry (MCR). Materials and methods We considered all cases of CMPD diagnosed in the Province of Modena (population 633.993 at 2001 Census). Cases, except Chronic Myeloid Leukemia, diagnosed from 1997 to 2005, were identified using the MCR database and the archival files of the centralized hemolymphopathological laboratory at Modena Cancer Centre according to ICD-O-3 codes 9950, 9960–64. Death certificate, cytology and histology report, both local and national reports of hospital admission, ICD-9 code reported in medical records were used as sources for identifying new CMPD cases and their outcome. All cases were checked and validated by a hematologist (A.B.) and a pathologist (G.B.) by a review of the original pathology report. Uniform diagnostic criteria were adopted, because the large majority of bone marrow aspirate and biopsy were examined by the same pathologist (G.B.). Clinical and follow-up data were retrieved by active search of discharge letters, review of hospital records and interview of general practitioners. Information on vital status was achieved from official population registries. Age standardized rates (ASR) were calculated according to the World Standard population. The dates of diagnosis and death or the closing date of study (December 2006) were used to estimate survival. Observed survival and relative survival were calculated according to Kaplan-Meier method and the Hakulinen approach, respectively. Results According to the 2001 World Health Organization (WHO) classification, a total of 380 cases of CMPD were identified. There were 155 Essential Thrombocythemia (ET) (41% of all CMPD), 114 Policythemia Vera (PV) (30%), 75 Idiopathic Myelofibrosis (20%), 2 Hypereosinophilic Syndrome/Chronic Eosinophilic Leukaemia (0.5%), 1 Chronic Neutrophilic Leukemia (0.3%) and 31 CMPD not otherwise specified (8%). The ASR of CMPD was 3.2/100,000 varying slightly (from 2.5 to 4.1/100,000) during the study period (p = 0.11); the crude incidence rate was 6.6/100,000. Median age at diagnosis was 69 years. No statistically significant differences were observed between sex regarding incidence and age at diagnosis. Overall relative survival was 97%, 89% and 88% at 1, 3 and 5 years after diagnosis, respectively. Analyzing CMPD, we observed a better survival for ET and PV in comparison with other subtypes (p = 0.01). Conclusions To our knowledge, this study is the first in Italy providing information on the incidence and outcome of CMPD using population-based data. Our results confirm that the risk of developing CMPD increases with age. The incidence of CMPD was substantially stable during the study period. Overall survival patterns reflect the well known chronic course of these diseases. As expected, we observed important differences in overall survival by WHO subtypes. We believe that the availability of precise epidemiological data, in particular those regarding outcome could help clinicians in choosing the most appropriate cost-effective treatments.

scholarly journals Prognostic power of molecular and clinical data across cancer types

2020 ◽  
Author(s):  
Xinsen Xu ◽  
Wei Wang ◽  
Min He ◽  
Linhua Yang ◽  
Wei Chen ◽  
...  
Keyword(s):  
Precision Medicine ◽  
Clinical Data ◽  
Clinical Decision Making ◽  
Human Cancer ◽  
Clinical Decision ◽  
Molecular Data ◽  
The Cancer Genome Atlas ◽  
Low Grade ◽  
Cancer Types ◽  
Prognostic Power

Abstract Background Precision medicine holds promise in prognostication of human cancer. By analyzing the Cancer Genome Atlas (TCGA) data, we evaluated the prognostic power of molecular and clinical data across 33 cancer types.Methods The clinical and molecular data of more than 11,000 patients were obtained from the TCGA database. Top features associated with overall survival were identified. Concordance index of each data type was calculated to investigate the prognostic power. The performance differences among clinical data, molecular data and combination data (integration of molecular data with clinical data) were evaluated.Results The prognostic power of combination data was significantly higher than the molecular data in 108 of 163 comparisons. However, it was only significantly higher than the clinical data in 27 of 163 comparisons. The clinical data seemed to be the most informative prognostic variable in almost half cancer types (14/33). Deeper insights into the low grade glioma models showed that integration of clinical data with molecular data yielded better prognostic modelling than either data used alone. From the pan-cancer level, the combination data was shown to be the most informative prognostic predictor when the sample size was large. In addition, mutation data also showed significant prognostic value.Conclusions Molecular markers complement the traditional diagnostic approaches in the pursuit of precision medicine. The combination of reliable clinical data, multidimensional genomic measurements and mature bioinformatics algorithms may confer more robust prognostic value that will inform clinical decision making.

scholarly journals A New Clinical Nomogram From the TCGA Database to Predict the Prognosis of Hepatocellular Carcinoma

2021 ◽  
Vol 11 ◽  
Author(s):  
Dingde Ye ◽  
Jiamu Qu ◽  
Jian Wang ◽  
Guoqiang Li ◽  
Beicheng Sun ◽  
...  
Keyword(s):  
Risk Factors ◽  
Hepatocellular Carcinoma ◽  
Overall Survival ◽  
Clinical Decision Making ◽  
Clinical Decision ◽  
Assessment Tools ◽  
The Cancer Genome Atlas ◽  
High Recurrence Rate ◽  
Concordance Index ◽  
Discriminative Ability

Background and AimHepatocellular carcinoma is a common malignant tumor of the digestive system with a poor prognosis. The high recurrence rate and metastasis after surgery reduce the survival time of patients. Therefore, assessing the overall survival of patients with hepatocellular carcinoma after hepatectomy is critical to clinicians’ clinical decision-making. Conventional hepatocellular carcinoma assessment systems (such as tumor lymph node metastasis and Barcelona clinical hepatocellular carcinoma) are obviously insufficient in assessing the overall survival rate of patients. This research is devoted to the development of nomogram assessment tools to assess the overall survival probability of patients undergoing liver resection.MethodsWe collected the clinical and pathological information of 438 hepatocellular carcinoma patients undergoing surgery from The Cancer Genome Atlas (TCGA) database, then excluded 87 patients who did not meet inclusion criteria. Univariate and multivariate analyses were performed on patient characteristics and related pathological factors. Finally, we developed a nomogram model to predict patient’s prognosis.ResultsA retrospective analysis of 438 consecutive samples from the TCGA database of patients with hepatocellular carcinoma who underwent potentially curative liver resections. Six risk factors were included in the final model. In the training set, the discriminative ability of the nomogram was very good (concordance index = 0.944), and the external verification method (concordance index = 0.962) was used for verification. At the same time, the internal and external calibration of the model was verified, showing that the model was well calibrated. The calibration between the evaluation of the nomogram and the actual observations was good. According to the patient’s risk factors, we determined the patient’s Kaplan-Meyer survival analysis curve. Finally, the clinical decision curve was used to compare the benefits of two different models in evaluating patients’ clinical outcomes.ConclusionsThe nomogram can be used to evaluate the post-hepatectomy 1-, 3-, and 5-year survival rates of patients with hepatocellular carcinoma. The Kaplan-Meyer curve can intuitively display the survival differences among patients with various risk factors. The clinical decision curve is a good reference guide for clinical application.

Outcome and survival following primary and repeat surgery for World Health Organization Grade III meningiomas

2010 ◽  
Vol 113 (2) ◽  
pp. 202-209 ◽  
Author(s):  
Michael E. Sughrue ◽  
Nader Sanai ◽  
Gopal Shangari ◽  
Andrew T. Parsa ◽  
Mitchel S. Berger ◽  
...  
Keyword(s):  
Overall Survival ◽  
Clinical Data ◽  
World Health ◽  
Gross Total Resection ◽  
Total Resection ◽  
Repeat Surgery ◽  
Who Grade ◽  
Kaplan Meier ◽  
Who Grade Iii ◽  
Grade Iii

Object Despite an increased understanding of the biology of malignant meningioma tumor progression, there is a paucity of published clinical data on factors affecting outcomes following treatment for these lesions. The authors present the largest case series to date dealing with these tumors, providing analysis of 63 patients. Methods The authors identified all patients undergoing resection of WHO Grade III tumors at their institution over a 16-year period. They analyzed clinical data from these patients, and performed Kaplan-Meier and Cox regression analyses to determine the impact of different clinical characteristics and different treatment modalities on survival following initial and repeat surgery for these lesions. Results Sixty-three patients met inclusion criteria and were analyzed further. The median clinical follow-up time was 5 years (range 1–22 years). The 2-, 5-, and 10-year overall survival rates following initial operation were 82, 61, and 40%, respectively. Kaplan-Meier analysis demonstrated a marked survival benefit with repeat operation (53 vs 25 months, p = 0.02). Interestingly, patients treated with near-total resection experienced improved overall survival when compared with patients treated with gross-total resection at initial (p = 0.035) and repeat operations (p = 0.005). Twelve (19%) of 63 patients experienced significant neurological morbidity referable to the resection of their tumors. Conclusions Surgery is an effective treatment for WHO Grade III meningiomas at presentation and recurrence; however, aggressive attempts to achieve gross-total resection can be associated with significant neurological risk.

scholarly journals Clinical characteristics and overall survival prognostic nomogram for invasive cribriform carcinoma of breast: a SEER population-based analysis

BMC Cancer ◽  
2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Jiameng Liu ◽  
Xiaobin Zheng ◽  
Zhonghua Han ◽  
Shunguo Lin ◽  
Hui Han ◽  
...  
Keyword(s):  
Overall Survival ◽  
Clinical Data ◽  
Validation Cohort ◽  
Ductal Carcinoma ◽  
Tumor Location ◽  
Population Based ◽  
Survival Outcomes ◽  
Seer Database ◽  
Survival Status ◽  
The Difference

Abstract Background The prognositc factors in patient with invasive cribriform carcinoma (ICC) of breast is still remain controversal. The study aims to establish a nomogram to predict the survival outcomes in patients with ICC based on the Surveillance, Epidemiology and End Results (SEER) database. Methods We retrieved SEER database for clinical data about patients including ICC and infiltrating ductal carcinoma (IDC) from 2004 to 2015. Kaplan-Meier survival was used to compare the difference survival outcomes between ICC and IDC. ICC patients were randomly allocated to training cohort and validation cohort. A nomogram was built to predict individual patient’s 3-year and 5-year survival status for ICC. The established TMN model and the newly established nomogram was further evaluated by the concordance index (C-index) and the decision curve analysis (DCA). Results Comparing the baseline clinical data between IDC and ICC, a significant of smaller tumor mass, less infiltrated lymph nodes, lower metastases rate, better tumor differentiation degree, higher proportion of estrogen receptor (ER) and progesterone receptor (PR) positive and lower rate of chemotherapy and radiotherapy was found in ICC. Age at diagnosis, marriage status, tumor location, T stage, M stage, ER status, surgery were independent significant prognostic factors for the overall survival (OS). A significantly higher C-index was found in nomogram compared with established TNM model in validation cohort. Conclusions The prognosis of ICC patients is better than that of IDC patients. The nomogram is recommended for future patient with ICC to survival analysis.

scholarly journals PATH-31. ARTIFICIAL INTELLIGENCE IN GLIOMA PATHOLOGY IMAGE ANALYSIS FOR RISK PREDICTION

2020 ◽  
Vol 22 (Supplement_2) ◽  
pp. ii171-ii171
Author(s):  
Pranathi Chunduru ◽  
Joanna Phillips ◽  
Annette Molinaro
Keyword(s):  
Overall Survival ◽  
Deep Learning ◽  
Tumor Tissue ◽  
Cox Regression ◽  
Clinical Care ◽  
Cox Proportional Hazards ◽  
The Cancer Genome Atlas ◽  
Patient Risk ◽  
Cox Regression Analysis ◽  
Whole Slide Images

Abstract Pathological evaluation of tumor tissue images stained with hematoxylin and eosin (H&E) is pivotal in diagnosis and predictive of outcome, yet only a small fraction of the rich phenotypic information on the slide is currently used for clinical care. In this study, we developed a computational approach based on deep learning to predict overall survival within distinct molecular subtypes of glioma patients and to extract prognostic biomarkers from microscopic images of tissue biopsies. Whole-slide images from 766 unique patients [IDH: 336 IDH-wildtype, 364 IDH-mutant, 1p/19q: 142 1p/19q-codeleted, 620 1p/19q-non-codeleted] were obtained from The Cancer Genome Atlas (TCGA). Sub-images that were free of artifacts and that contained viable tumor with descriptive histologic characteristics were extracted, which were used for training and testing the deep neural-network. Our unified survival deep learning framework (SDL) uses a residual CNN network integrated with a traditional survival model to predict patient risk from digitized whole-slide images. We employed statistical sampling techniques and randomized transformation of images to address challenges in learning from histology images. Univariable and multivariable Cox proportional-hazards regression models were used to evaluate the significance of predicted patient risk with and without controlling for known prognostic factors. The integrated SDL framework showed substantial prognostic power achieving a median c-index of 0.79 [95 % CI 0.77 - 0.81]. In multivariable Cox regression analysis, SDL risk was significantly associated with overall survival (hazard ratio of 1.65, 95% CI 1.49-1.83, p < 0.001) after adjusting for age, grade, IDH status, ATRX status, 1p19q codeletion and CDKN2A/2B status. Only IDH-status and age were also significant in the multivariable model. Preliminary findings highlight the emerging role of AI in precision medicine and suggest the utility for computational analysis of tumor tissue images for objective and accurate prediction of outcome for glioma patients and also for risk stratification for targeted clinical therapy.

scholarly journals Identification of Mutator-Derived Alternative Splicing Signatures of Genomic Instability for Improving the Clinical Outcome of Cholangiocarcinoma

2021 ◽  
Vol 11 ◽  
Author(s):  
Zijing Lin ◽  
Jianping Gong ◽  
Guochao Zhong ◽  
Jiejun Hu ◽  
Dong Cai ◽  
...  
Keyword(s):  
Overall Survival ◽  
Alternative Splicing ◽  
Genomic Instability ◽  
Clinical Data ◽  
Risk Score ◽  
Pearson Correlation ◽  
Negative Regulator ◽  
The Cancer Genome Atlas ◽  
Integrated Analysis ◽  
Prognostic Signature

BackgroundCholangiocarcinoma is an aggressive carcinoma with increasing incidence and poor outcomes worldwide. Genomic instability and alternative splicing (AS) events are hallmarks of carcinoma development and progression. The relationship between genomic instability, AS events, and tumor immune microenvironment remain unclear.MethodsThe splicing profiles of patients with cholangiocarcinoma were obtained from The Cancer Genome Atlas (TCGA) spliceSeq database. The transcriptomics, simple nucleotide variation (SNP) and clinical data of patients with cholangiocarcinoma were obtained from TCGA database. Patients were divided into genomic unstable (GU-like) and genomic stable (GS-like) groups according to their somatic mutations. Survival-related differential AS events were identified through integrated analysis of splicing profiling and clinical data. Kyoto Encyclopedia of Genes and Genomes enrichment analysis was used to identify AS events occurring in genes enriched in cancer pathways. Pearson correlation was applied to analyze the splicing factors regulating AS events. CIBERSORT was used identify differentially infiltrating immune cells.ResultsA prognostic signature was constructed with six AS events. Using this signature, the hazard ratio of risk score for overall survival is 2.362. For TCGA patients with cholangiocarcinoma, the area under the receiver operating characteristic curve is 0.981. CDK11A is a negative regulator of survival associated AS events. Additionally, the CD8+ T cell proportion and PD-L1 expression are upregulated in patients with cholangiocarcinoma and high splicing signatures.ConclusionWe provide a prognostic signature for cholangiocarcinoma overall survival. The CDK11A splicing factor and SLC46A1-39899-ES and IARS-86836-ES AS events may be potential targets for cholangiocarcinoma therapy. Patients with high AS risk score may be more sensitive to anti-PD-L1/PD1 immunotherapy.

Sign in / Sign up

Export Citation Format

Share Document