Artificial intelligence prediction model for overall survival of clear cell renal cell carcinoma based on a 21-gene molecular prognostic score system

Machine Learning-based Framework Develops a Tumor Thrombus Coagulation Signature in Multicenter Cohorts for Renal Cancer

Background: Renal cell carcinoma (RCC) is frequently accompanied by tumor thrombus in the venous system with an extremely dismal prognosis. The current Tumor Node Metastasis (TNM) stage and Mayo clinical classification do not appropriately identify preference-sensitive treatment. Therefore, there is an urgent need to develop a better ideal model for precision medicine. Methods: In this study, we developed a coagulation tumor thrombus signature for RCC with 10 machine-learning algorithms (101 combinations) based on a novel computational framework using multiple independent cohorts. Results: The established tumor thrombus coagulation-related risk stratification (TTCRRS) signature comprises 10 prognostic coagulation-related genes (CRGs). This signature could predict survival outcomes in public and in-house protein cohorts and showed high performance compared to 129 published signatures. Additionally, the TTCRRS signature was significantly related to some immune landscapes, immunotherapy response, and chemotherapy. Furthermore, we also screened out hub genes, transcription factors, and small compounds based on the TTCRRS signature. Meanwhile, CYP51A1 can regulate the proliferation and migration properties of RCC. Conclusions: The TTCRRS signature can complement the traditional anatomic TNM staging system and Mayo clinical stratification and provide clinicians with more therapeutic options.

Development and implementation of a prognostic model for clear cell renal cell carcinoma based on heterogeneous TLR4 expression

Objective

Clear cell renal cell carcinoma (ccRCC) is the most common subtype among renal cell carcinomas and has the worst prognosis, originating from renal tubular epithelial cells. Toll-like receptor 4 (TLR4) plays a crucial role in ccRCC proliferation, infiltration, and metastasis. The aim of this study was to construct a prognostic scoring model for ccRCC based on TLR4 expression heterogeneity and to explore its association with immune infiltration, thereby providing insights for the treatment and prognostic evaluation of ccRCC.

Methods

Using R software, a differential analysis was conducted on normal samples and ccRCC samples, and in conjunction with the KEGG database, a correlation analysis for the clear cell renal cell carcinoma pathway (hsa05211) was carried out. We observed the expression heterogeneity of TLR4 in the TCGA-KIRC cohort and identified its related differential genes (TRGs). Based on the expression levels of TRGs, consensus clustering was employed to identify TLR4-related subtypes, and further clustering heatmaps, principal component, and single-sample gene set enrichment analyses were conducted. Overlapping differential genes (ODEGs) between subtypes were analysed, and combined with survival data, univariate Cox regression, LASSO, and multivariate Cox regression were used to establish a prognostic risk model for ccRCC. This model was subsequently evaluated through ROC analysis, risk factor correlation analysis, independent prognostic factor analysis, and intergroup differential analysis. The ssGSEA model was employed to explore immune heterogeneity in ccRCC, and the performance of the model in predicting patient prognosis was evaluated using box plots and the oncoPredict software package.

Results

In the TCGA-KIRC cohort, TLR4 expression was notably elevated in ccRCC samples compared to normal samples, correlating with improved survival in the high-expression group. The study identified distinct TLR4-related differential genes and categorized ccRCC into three subtypes with varied survival outcomes. A risk prognosis model based on overlapping differential genes was established, showing significant associations with immune cell infiltration and key immune checkpoints (PD-1, PD-L1, CTLA4). Additionally, drug sensitivity differences were observed between risk groups.

Conclusion

In the TCGA-KIRC cohort, the expression of TLR4 in ccRCC samples exhibited significant heterogeneity. Through clustering analysis, we identified that the primary immune cells across subtypes are myeloid-derived suppressor cells, central memory CD4 T cells, and regulatory T cells. Furthermore, we successfully constructed a prognostic risk model for ccRCC composed of 17 genes. This model provides valuable references for the prognosis prediction and treatment of ccRCC patients.

The Present and Future of Artificial Intelligence in Urological Cancer

Artificial intelligence has drawn more and more attention for both research and application in the field of medicine. It has considerable potential for urological cancer detection, therapy, and prognosis prediction due to its ability to choose features in data to complete a particular task autonomously. Although the clinical application of AI is still immature and faces drawbacks such as insufficient data and a lack of prospective clinical trials, AI will play an essential role in individualization and the whole management of cancers as research progresses. In this review, we summarize the applications and studies of AI in major urological cancers, including tumor diagnosis, treatment, and prognosis prediction. Moreover, we discuss the current challenges and future applications of AI.

Prognostic role of the endothelial cell-specific molecule-1 histopathologic expression in renal cell cancer

INTRODUCTION

To measure the level of endothelial cell-specific molecule-1 (ESM-1) expression among the Renal Cell Cancer (RCC) variants using by immunohistochemical method and determine the relationship between ESM-1 expression and RCC prognosis.

MATERIALS AND METHODS

ESM-1 immunoreactivity scores (IR) were measured in appropriate renal tumoral tissue blocks of 153 consecutive RCC patients in this retrospective analysis of prospectively collected data. Mean ESM-1 IR scores were calculated in patients who were pathologically diagnosed with clear cell RCC (ccRCC), papillary RCC (pRCC), and chromophobe RCC (chRCC). Progression-free survival and overall survival were evaluated using the log-rank test according to ESM-1 IR scores. Survival rates were calculated using Kaplan-Meier survival analysis.

RESULTS

In the ccRCC group, the mean ESM-1 IR scores of those with local invasion were significantly higher than those without local invasion (P = 0.014). The mean ESM-1 IR score of patients with metastatic ccRCC was significantly higher than those with non-metastatic ccRCC (P < 0.001). Considering all patients regardless of RCC subtype pathologies, the mean ESM-1 IR score in clinical stage 1 tumor was 3.82 ± 1.98, 4.87 ± 1.74 in clinical stage 2, 5.88 ± 2 in clinical stage 3, and 6.60 ± 2.23 in clinical stage 4. The mean ESM-1 IR score of patients with metastatic ccRCC was significantly higher than those with non-metastatic ccRCC (P < 0.001). The mean follow-up period for all patients in this study was 71 months (range 1-120 months). It has been shown that the higher the ESM-1 IR score, the lower the 10-year overall survival and disease-free survival rates (P = 0.026, P = 0.005).

CONCLUSION

Immunohistochemical expression of ESM-1 may be a promising prognostic biomarker in RCC. Currently, some prognostic scoring systems are available for patients with localized and metastasized RCC. Incorporating ESM-1 expression in RCC into these existing prognostic scoring systems could improve these models and enhance the quality of individual oncologic management in RCC patients.

High SPATA18 Expression and its Diagnostic and Prognostic Value in Clear Cell Renal Cell Carcinoma

BACKGROUND SPATA18 (spermatogenesis-associated 18, also called Mieap) encodes a protein that can induce lysosome-like organelles within mitochondria, which plays an important role in tumor growth. We measured the expression of SPATA18 in ccRCC, and assessed its diagnostic and prognostic clinical value in patients with clear cell renal cell carcinoma (ccRCC). MATERIAL AND METHODS We analyzed SPATA18 expression using data from the TCGA-KIRC cohort, GEO database, and UALCAN database. Immunohistochemistry was carried out to verify the expression in the ccRCC patients. The diagnostic value of SPATA18expression was evaluated by a receiver operating characteristic (ROC) curve. The correlation between clinical characteristics and SPATA18 expression was calculated by chi-square test. The prognostic value of SPATA18 expression was assessed by Kaplan-Meier analysis and Cox analysis. We conducted gene set enrichment analysis (GSEA) using TCGA database. RESULTS SPATA18 gene exhibited a higher expression in ccRCC tissues than in normal tissues. SPATA18 showed a substantial diagnostic value in ccRCC. SPATA18 expression was correlated with histological grade, clinical stage, T classification, and distant metastasis of ccRCC. Furthermore, high SPATA18 expression was associated with favorable overall survival. Multivariate analysis showed that SPATA18 was an independent risk factor for ccRCC. Gene set enrichment analysis (GSEA) showed that B cell receptors, WNT targets, extracellular matrix, oxidative phosphorylation, calcium metabolism, iron uptake and transport, potassium channels, and insulin receptor were differently enriched in the phenotype that was negatively correlated with SPATA18. CONCLUSIONS Our study indicated that high SPATA18 expression in ccRCC was associated with a good prognosis, and it could be a positive prognostic biomarker for ccRCC.

Deep learning-based predictions of clear and eosinophilic phenotypes in clear cell renal cell carcinoma

We have recently shown that histological phenotypes focusing on clear and eosinophilic cytoplasm in clear cell renal cell carcinoma (ccRCC) correlated with prognosis and the response to angiogenesis inhibition and checkpoint blockade. This study aims to objectively show the diagnostic utility of clear or eosinophilic phenotypes of ccRCC by developing an artificial intelligence (AI) model using the TCGA-ccRCC dataset and to demonstrate if the clear or eosinophilic predicted phenotypes correlate with pathological factors and gene signatures associated with angiogenesis and cancer immunity. Before the development of the AI model, histological evaluation using hematoxylin and eosin whole-slide images of the TCGA-ccRCC cohort (n = 435) was performed by a urologic pathologist. The AI model was developed as follows. First, the highest-grade area on each whole slide image was captured for image processing. Second, the selected regions were cropped into tiles. Third, the AI model was trained using transfer learning on a deep convolutional neural network, and clear or eosinophilic predictions were scaled as AI scores. Next, we verified the AI model using a validation cohort (n = 95). Finally, we evaluated the accuracy of the prognostic predictions of the AI model and revealed that the AI model detected clear and eosinophilic phenotypes with high accuracy. The AI model stratified the patients' outcomes, and the predicted eosinophilic phenotypes correlated with adverse clinicopathological characteristics and high immune-related gene signatures. In conclusion, the AI-based histologic subclassification accurately predicted clear or eosinophilic phenotypes of ccRCC, allowing for consistently reproducible stratification for prognostic and therapeutic stratification.