Dataset Information

Automated clear cell renal carcinoma grade classification with prognostic significance.

ABSTRACT: We developed an automated 2-tiered Fuhrman's grading system for clear cell renal cell carcinoma (ccRCC). Whole slide images (WSI) and clinical data were retrieved for 395 The Cancer Genome Atlas (TCGA) ccRCC cases. Pathologist 1 reviewed and selected regions of interests (ROIs). Nuclear segmentation was performed. Quantitative morphological, intensity, and texture features (n = 72) were extracted. Features associated with grade were identified by constructing a Lasso model using data from cases with concordant 2-tiered Fuhrman's grades between TCGA and Pathologist 1 (training set n = 235; held-out test set n = 42). Discordant cases (n = 118) were additionally reviewed by Pathologist 2. Cox proportional hazard model evaluated the prognostic efficacy of the predicted grades in an extended test set which was created by combining the test set and discordant cases (n = 160). The Lasso model consisted of 26 features and predicted grade with 84.6% sensitivity and 81.3% specificity in the test set. In the extended test set, predicted grade was significantly associated with overall survival after adjusting for age and gender (Hazard Ratio 2.05; 95% CI 1.21-3.47); manual grades were not prognostic. Future work can adapt our computational system to predict WHO/ISUP grades, and validating this system on other ccRCC cohorts.

SUBMITTER: Tian K

PROVIDER: S-EPMC6776313 | biostudies-literature | 2019

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Automated clear cell renal carcinoma grade classification with prognostic significance.

Tian Katherine K Rubadue Christopher A CA Lin Douglas I DI Veta Mitko M Pyle Michael E ME Irshad Humayun H Heng Yujing J YJ

PloS one 20191003 10

We developed an automated 2-tiered Fuhrman's grading system for clear cell renal cell carcinoma (ccRCC). Whole slide images (WSI) and clinical data were retrieved for 395 The Cancer Genome Atlas (TCGA) ccRCC cases. Pathologist 1 reviewed and selected regions of interests (ROIs). Nuclear segmentation was performed. Quantitative morphological, intensity, and texture features (n = 72) were extracted. Features associated with grade were identified by constructing a Lasso model using data from cases ...[more]

PMID: 31581201

Similar Datasets

Project description:BACKGROUND: The aim of this study was to determine whether Src family kinases (SFK) are expressed in renal cell cancer and to assess their prognostic significance. METHODS: mRNA expression levels were investigated for the 8 SFK members by quantitative real-time PCR in 19 clear cell cancer tissue samples. Immunohistochemical staining was utilised to assess expression of Src kinase, dephosphorylated Src kinase at Y(530) (SrcY(530)), phosphorylated Src at Y(419) (SrcY(419)) and the downstream focal adhesion kinase (FAK) marker at the Y(861) site (FAK Y(861)) in a cohort of 57 clear cell renal cancer specimens. Expression was assessed using the weighted histoscore method. RESULTS: Src, Lyn, Hck, Fgr and Fyn were the most highly expressed in renal cancer. All members were more highly expressed in T2 disease, and furthermore expression levels between T2 and T3 disease showed a significant decrease for Lck, Lyn, Fyn, Blk and Yes (P=0.032). Assessment of membrane, cytoplasm and nuclear expression of Src kinase, SrcY(530) and SrcY(419) were not significantly associated with cancer-specific survival. High expression of cytoplasmic FAK Y(861) was associated with decreased cancer-specific survival (P=0.001). On multivariate analysis, cytoplasmic FAK Y(861) was independently associated with cancer-specific survival (hazard ratio 3.35, 95% CI 1.40-7.98, P=0.006). CONCLUSION: We have reported that all SFK members are expressed in renal cell carcinoma. The SFK members had their highest levels of expression before the disease no longer being organ confined. We hypothesise that these SFK members are upregulated before the cancer spreading out-with the organ and given that Src itself is not associated with cancer-specific survival but the presence of FAK Y(861), a downstream marker for SFK member activity is associated with decreased cancer-specific survival, we hypothesise that another SFK member is associated with decreased cancer-specific survival in renal cell cancer.

Project description:BackgroundRenal cell carcinoma (RCC) is a fatal disease, in which the PI3K/AKT/mTOR signaling pathway serves an important role in the tumorigenesis. Previous studies have reported the prognostic significance of PI3K/AKT/mTOR signaling pathway members in RCC; however, there is insufficient evidence to date to confirm this. Thus, the present study aimed to systematically investigate the prognostic roles of multiple PI3K/AKT/mTOR signaling proteins in clear cell RCC (ccRCC) using online large-scale databases.MethodsThe mRNA expression profiles of PI3K/AKT/mTOR signaling pathway proteins PTEN, PIK3CA, PIK3CB, PIK3CD, PIK3CG, AKT1, AKT2, AKT3 and mTOR were investigated using the Gene Expression Profiling Interactive Analysis (GEPIA) and Oncomine databases, and the protein expression levels of PI3K, AKT and mTOR were detected using western blotting (WB) analysis. In addition, the correlation between mRNA or protein expression levels and the prognostic significance was analyzed using the Kaplan-Meier (K-M) plotter (n = 530), the Human Protein Atlas (HPA; n = 528) and The Cancer Protein Atlas (TCPA; n = 445) databases.ResultsThe GEPIA revealed that the mRNA expression of major PI3K/AKT/mTOR pathway members, including PTEN, PIK3CA, PIK3CB, AKT1, AKT2 and AKT3, were negatively correlated with ccRCC stages (P < 0.05), though most of their mRNA and protein expression levels were notsignificantly different between ccRCC and normal tissues using GEPIA, Oncomine and WB analyses (P < 0.05). Meanwhile, using the K-M plotter and HPA prognostic analysis, it was found that the mRNA expression levels of the majority of the PI3K/AKT/mTOR signaling pathway members, including PTEN, PIK3CA, PIK3CB, PIK3CG, AKT3 and mTOR were positively correlated with overall survival (OS), whereas PIK3CD mRNA expression was negatively correlated with OS (P < 0.05). Furthermore, TCPA prognostic analysis observed that several of the key molecules of the PI3K/AKT/mTOR signaling pathway [PTEN, p-AKT (S473) and p-mTOR (S2448)] were also positively correlated with OS in patients with ccRCC (P < 0.05). In conclusion, the present study suggested that several members of the PI3K/AKT/mTOR signaling pathway, especially PTEN, may be favorable prognostic factors in ccRCC, which indicated that the PI3K/AKT/mTOR signaling pathway may be implicated in ccRCC initiation and progression.

Project description:Background: Calcium signaling is implicated in multiple processes including immune response that important in tumor progression. Kidney renal clear cell carcinoma (KIRC) is the most frequent histological type of renal cell carcinoma with up to a third of cases develop metastases. As a result of a lack of in-depth understanding of the mechanisms underlying KIRC, treatment options have been limited. Here, we aim to comprehensively investigate the landscape of Ca2+ channels, pumps and exchangers in KIRC patients. Methods: The mRNA expression profiles and gene variations of 58 calcium-related genes (CRGs) in KIRC patients and normal control cases were downloaded from TCGA database. CRGs-related risk score was constructed to quantify calcium patterns by using least absolute shrinkage and selection operator (LASSO) regression. The prognostic value, biological functions, immune landscape and therapeutic sensitivities based on CRGs-related risk score were then evaluated using multiple methods. Finally, key gene of CRGs was identified by weighted gene co-expression network analysis (WGCNA). TCGA-CPTAC, GSE53757 datasets, as well as human tissues were used for validation. Results: KIRC patients had significant differences in CRG expression, prognosis, and biological functions between two CRG clusters. CRGs-related risk score was then determined. The prognosis, tumor mutation burden, immune cell infiltration, immune checkpoints, and the response of targeted inhibitors were remarkably different between high and low CRGs-related risk subtypes. CRGs-related high-risk subtype was characterized by immunosuppressive microenvironment with poor prognosis. Meanwhile, several targeted drugs showed distinct sensitivity between CRGs-related risk subtypes. Finally, TRPM3 was identified as a key CRG based on risk score in KIRC patients. TRPM3 mRNA and protein expression were significantly lower in KIRC tumors than in normal controls. Low TRPM3 expression was associated with poor prognosis in KIRC patients. Conclusion: Our study highlighted the promising prognostic value of CRGs in KIRC tumors. The evaluation of CRGs-related risk score will contribute to predicting prognosis and clinical therapy in KIRC patients.

Dataset Information

Automated clear cell renal carcinoma grade classification with prognostic significance.

Publications

Automated clear cell renal carcinoma grade classification with prognostic significance.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets