scholarly journals Reimagining T Staging Through Artificial Intelligence and Machine Learning Image Processing Approaches in Digital Pathology

2020 ◽  
pp. 1039-1050
Author(s):  
Kaustav Bera ◽  
Ian Katz ◽  
Anant Madabhushi
Keyword(s):  
Artificial Intelligence ◽  
Domain Knowledge ◽  
Imaging Techniques ◽  
Prognostic Significance ◽  
Digital Pathology ◽  
Staging System ◽  
Tumor Stage ◽  
Tnm Staging ◽  
Molecular Alterations ◽  
Prognostic Prediction

Tumor stage and grade, visually assessed by pathologists from evaluation of pathology images in conjunction with radiographic imaging techniques, have been linked to outcome, progression, and survival for a number of cancers. The gold standard of staging in oncology has been the TNM (tumor-node-metastasis) staging system. Though histopathological grading has shown prognostic significance, it is subjective and limited by interobserver variability even among experienced surgical pathologists. Recently, artificial intelligence (AI) approaches have been applied to pathology images toward diagnostic-, prognostic-, and treatment prediction–related tasks in cancer. AI approaches have the potential to overcome the limitations of conventional TNM staging and tumor grading approaches, providing a direct prognostic prediction of disease outcome independent of tumor stage and grade. Broadly speaking, these AI approaches involve extracting patterns from images that are then compared against previously defined disease signatures. These patterns are typically categorized as either (1) handcrafted, which involve domain-inspired attributes, such as nuclear shape, or (2) deep learning (DL)–based representations, which tend to be more abstract. DL approaches have particularly gained considerable popularity because of the minimal domain knowledge needed for training, mostly only requiring annotated examples corresponding to the categories of interest. In this article, we discuss AI approaches for digital pathology, especially as they relate to disease prognosis, prediction of genomic and molecular alterations in the tumor, and prediction of treatment response in oncology. We also discuss some of the potential challenges with validation, interpretability, and reimbursement that must be addressed before widespread clinical deployment. The article concludes with a brief discussion of potential future opportunities in the field of AI for digital pathology and oncology.

scholarly journals Is Preoperative Fibrinogen Associated with the Survival Prognosis of Gastric Cancer Patients? A Multi-centered, Propensity Score-Matched Retrospective Study

2019 ◽  
Vol 44 (1) ◽  
pp. 213-222
Author(s):  
Lin-Yong Zhao ◽  
Yong-Liang Zhao ◽  
Jun-Jiang Wang ◽  
Qi-Di Zhao ◽  
Wen-Qi Yi ◽  
...  
Keyword(s):  
Gastric Cancer ◽  
Propensity Score ◽  
Cancer Patients ◽  
Prognostic Significance ◽  
Staging System ◽  
Tnm Staging ◽  
Plasma Fibrinogen ◽  
Tnm Staging System ◽  
N Stage ◽  
Gastric Cancer Patients

Abstract Background The prognostic significance of preoperative plasma fibrinogen in patients with operable gastric cancer remains under debate. This study aimed to elucidate the prognostic value of fibrinogen in gastric cancer patients underwent gastrectomy. Methods A total of 4351 patients with gastric cancer collected from three comprehensive medical centers were retrospectively evaluated. Patients were categorized by minimum P value using X-tile, while the baseline confounders for fibrinogen was balanced through propensity score matching (PSM). The relationships between fibrinogen and other clinicopathologic features were evaluated, and nomogram was constructed to assess its prognostic improvement compared with TNM staging system. Results Fibrinogen was significantly correlated with macroscopic type, tumor differentiation, tumor size, and T and N stage. The factors, fibrinogen and T stage as well as N stage, were identified to be independent prognostic factors after PSM. Nomogram based on fibrinogen demonstrated a smaller Akaike information criterion (AIC) and a larger concordance index (C-index) than TNM staging system, illustrating that fibrinogen might be able to improve the prognostic accuracy. Conclusions Preoperative plasma fibrinogen levels in gastric cancer patients were significantly correlated with tumor progression, which could be regarded as a reliable marker for survival prognostic prediction.

scholarly journals Nomogram based on Autophagy Related Genes for Predicting the Survival in Melanoma

2021 ◽  
Author(s):  
Guangtong Deng ◽  
Wenhua Wang ◽  
Yayun Li ◽  
Huiyan Sun ◽  
Furong Zeng
Keyword(s):  
Roc Curve ◽  
Staging System ◽  
Degradation Pathway ◽  
Tnm Staging ◽  
High Risk Group ◽  
Curve Analysis ◽  
The Cancer Genome Atlas ◽  
Kaplan Meier ◽  
Clinical Benefits ◽  
Prognostic Prediction

Abstract Background: Autophagy, a highly conserved lysosomal degradation pathway, isassociated with the prognosis of melanoma. However, prognostic prediction modelsbased on autophagy related genes (ARGs) have never been recognized in melanoma.In the present study, we aimed to establish a novel nomogram to predict the prognosisof melanoma based on ARGs signature and clinical parameters.Methods: Data from The Cancer Genome Atlas (TCGA) and the Genotype-TissueExpression (GTEx) databases were extracted to identify the differentially expressedARGs. Univariate, least absolute shrinkage and selection operator (LASSO) andmultivariate analysis were used to select the prognostic ARGs. ARG signature, ageand stage were then enrolled to establish a nomogram to predict the survivalprobabilities of melanoma. The nomogram was evaluated by concordance index(C-index), receiver operating characteristic (ROC) curve and calibration curve.Decision curve analysis (DCA) was performed to assess the clinical benefits of thenomogram and TNM stage model. The nomogram was validated in GEO cohorts.Results: Five prognostic ARGs were selected to construct ARGs signature model andvalidated in the GEO cohort. Kaplan-Meier survival analysis suggested that patientsin high-risk group had significantly worse overall survival than those in low-riskgroup in TCGA cohort (P = 5.859 × 10-9) and GEO cohort (P = 3.075 × 10-9).We then established and validated a novel promising prognostic nomogram throughcombining ARGs signature and clinical parameters. The C-index of the nomogramwas 0.717 in TCGA training cohort and 0.738 in GEO validation cohort.TCGA/GEO-based ROC curve and decision curve analysis (DCA) demonstrated thatthe nomogram was better than traditional TNM staging system for melanomaprognosis.Conclusion: We firstly developed and validated an ARGs signature based-nomogramfor individualized prognosis prediction in melanoma patients, which could assist withdecision making for clinicians.

scholarly journals Prognostic significance for colorectal carcinoid tumors based on the 8th edition TNM staging system

2020 ◽  
Vol 9 (21) ◽  
pp. 7979-7987
Author(s):  
Piqing Gong ◽  
Chunhua Chen ◽  
Zhan Wang ◽  
Xukun Zhang ◽  
Wenxin Hu ◽  
...  
Keyword(s):  
Prognostic Significance ◽  
Staging System ◽  
Tnm Staging ◽  
Carcinoid Tumors ◽  
Tnm Staging System ◽  
8Th Edition ◽  
Colorectal Carcinoid

scholarly journals Prognostic Significance of the Tumor-Stromal Ratio in Invasive Breast Cancer and a Proposal of a New Ts-TNM Staging System

2020 ◽  
Vol 2020 ◽  
pp. 1-10 ◽  
Author(s):  
Qian Xu ◽  
Jing-Ping Yuan ◽  
Yuan-Yuan Chen ◽  
Hong-Yan Zhang ◽  
Lin-Wei Wang ◽  
...  
Keyword(s):  
Breast Cancer ◽  
Tumor Cells ◽  
Invasive Breast Cancer ◽  
Prognostic Significance ◽  
Staging System ◽  
Tnm Staging ◽  
Lymph Node Status ◽  
Tumor Node Metastasis ◽  
Node Metastasis ◽  
Tnm Staging System

Background. Previous studies have demonstrated that the tumor-stromal ratio (TSR) was an independent prognostic factor in several types of carcinomas. This study aimed at exploring the prognostic significance of the TSR in invasive breast cancer using immunohistochemistry (IHC)-stained tissue microarrays (TMAs) and integrating the TSR into the traditional tumor-node-metastasis (TNM) staging system. Methods. The prepared 7 TMAs containing 240 patients with 480 invasive BC specimens were stained with cytokeratin (CK) by the IHC staining method. The ratio of tumor cells and stromal cells was visually assessed. TSR > 1 and TSR ≤ 1 were categorized as the high TSR (low stroma) and low TSR (high stroma) groups, respectively, and the prognostic value of the TSR at 5-year disease-free survival (5-DFS) was analyzed. A new Ts-TNM (tumor stroma-tumor-node-metastasis) staging system was established and assessed. Results. IHC staining of CK could specifically label tumor cells with clear contrast, making it easy to manually assess TSR. High TSR (low stroma) and low TSR (high stroma) were observed in 52.5% (n = 126) and 47.5 (n = 114) of the cases, according to the division of value 1. A Kaplan–Meier analysis showed that patients in the low TSR group had a worse 5-DFS compared with patients in the high TSR group (P=0.022). Multivariable analysis indicated that the T stage (P=0.014), N status (P<0.001), histological grade (P<0.001), estrogen receptor status (P=0.015), and TSR (P=0.011) were independent prognostic factors of invasive BC patients. The new Ts-TNM staging system combining TSR, tumor staging, lymph node status, and metastasis staging was established. The receiver operating characteristic (ROC) curve analysis demonstrated that the ability of the Ts-TNM staging system to predict recurrence was not lower than that of the TNM staging system. Conclusions. This study confirms that the TSR is a prognostic indicator for invasive breast cancer. The Ts-TNM staging system containing stromal and tumor information may optimize risk stratification for invasive breast cancer.

Immune infiltration and angiogenesis as markers of outcome in the post-nephrectomy setting: Transcriptomic data from patients receiving placebo on a randomized phase III trial (PROTECT).

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. 5082-5082
Author(s):  
A. Ari Hakimi ◽  
Martin H Voss ◽  
Fengshen Kuo ◽  
Andrew W. Silagy ◽  
Mahtab Marker ◽  
...  
Keyword(s):  
Cox Regression ◽  
Expression Profiles ◽  
Univariate Analysis ◽  
Prognostic Significance ◽  
Staging System ◽  
Tumor Stage ◽  
Phase Iii ◽  
Cell Renal Cell Carcinoma ◽  
Metastatic Recurrence ◽  
Definitive Surgery

5082 Background: Defined stromal and immune features of the tumor microenvironment (TME) have proven relevant for outcomes with systemic therapy in advanced clear cell renal cell carcinoma (ccRCC). We hypothesized that these may matter beyond therapeutic applications and could be relevant much earlier in the disease course. We sought to study the TME in high risk ccRCC patients undergoing definitive surgery. Methods: Clinical, pathologic, immunohistochemical, and whole-genome microarray data were acquired on 236 out of 769 patients in the Placebo arm of PROTECT trial (NCT01235962 - pazopanib vs placebo). Transcriptomic scores assessing angiogenesis and myeloid infiltration with individual annotations above/below median were used to categorize patients into four groups (angiogenesis high vs. low; myeloid high vs. low). We tested categorical association with disease free (DFS) and overall survival (OS) using logrank testing and assessed interdependence with relevant clinicopathologic variables, including the UCLA Integrated Staging System (UISS) in a cox regression model. Results: Tumors from236 patients were available for analysis. Overall, 37% developed metastatic recurrence and 81% were alive at last follow up. On univariate analysis increasing tumor stage, higher UISS score, and angiogenesis/myeloid subgroups (high – H and low – L) were associated with worse DFS and OS (all p values <0.05). On multivariate analysis TME subgroups remained significant for worse DFS and OS (Table). Conclusions: Microenvironmental subgroups stratified into angiogenic and myeloid expression profiles carry independent prognostic significance and should be further explored to guide future biomarker-directed adjuvant trials. Clinical trial information: NCT01235962 . [Table: see text]

scholarly journals Integrating Data from Multidisciplinary Management of Malignant Pleural Mesothelioma: a cohort study

2020 ◽  
Author(s):  
David Michael Abbott ◽  
Claudio Valizia ◽  
Chandra Bortolotto ◽  
Stefano Tomaselli ◽  
Laura Saracino ◽  
...  
Keyword(s):  
Malignant Pleural Mesothelioma ◽  
Strong Association ◽  
Prognostic Significance ◽  
Staging System ◽  
Tnm Staging ◽  
Disease Stage ◽  
Pleural Mesothelioma ◽  
Imaging Data ◽  
Number Of Patients ◽  
Treatment Research

Abstract Background. Malignant pleural mesothelioma (MPM) is a rare and aggressive malignancy that most commonly affects the pleural lining of the lungs. MPM has a strong association with asbestos being at least 80% of cases caused by exposure to its biopersistent fibers. Individuals with a chronic exposure to asbestos generally have a 20-40-year latency with no or few symptoms. Such has been the case in the Piedmont and Lombardy regions of Italy where industrial production of materials laden with asbestos, mainly cements, has created a large number of patients. The fibers from this production did not only pose a substantial risk to the workers breathing the fibers, but also to the relatives and the surrounding community. Since 1992, the use of asbestos has been illegal in Italy, but such a long latency means that our center in the university town of Pavia, which is quite close to two former asbestos factories, is currently receiving a large number of patients. Methods. Since 2017 in Pavia, a multidisciplinary team has been collecting data on over 100 patients with MPM, including presentation, pathology type and the diagnostic method used. In addition, imaging data has also been analyzed in 49 patients who underwent CT with contrast prior to treatment. In this study, we bring together the abundant epidemiologic, histologic, radiologic, and other patient data to compare it with similar studies and draw correlations with predictive and prognostic significance. The basic descriptive statistical analysis was conducted through the Excel “Data Analysis” addon package. Results. Overall, the median survival (OS) was of 13.7 months. It was statistically influenced by the detection of pleural effusion at baseline (p<0.0001) and by histology (survival was higher in epidermoid forms (p<0.0001), irrespective of TNM disease stage. Conclusions. Quite unexpectedly, no statistically significant association could be demonstrated between OS and TNM disease stage at diagnosis. This result confirms that the TNM staging system is probably not adequate to manage MPM alone. Multidisciplinary approach to MPM has allowed us to create robust databases that will help guide radiomic projects, treatment research and thus improved patient management

Sign in / Sign up

Export Citation Format

Share Document