Use of artificial intelligence for the prediction of lymph node metastases in early-stage colorectal cancer: systematic review

BACKGROUND

Risk evaluation of lymph node metastasis for early-stage (T1 and T2) colorectal cancers is critical for determining therapeutic strategies. Traditional methods of lymph node metastasis prediction have limited accuracy. This systematic review aimed to review the potential of artificial intelligence in predicting lymph node metastasis in early-stage colorectal cancers.

METHODS

A comprehensive search was performed of papers that evaluated the potential of artificial intelligence in predicting lymph node metastasis in early-stage colorectal cancers. Studies were appraised using the Joanna Briggs Institute tools. The primary outcome was summarizing artificial intelligence models and their accuracy. Secondary outcomes included influential variables and strategies to address challenges.

RESULTS

Of 3190 screened manuscripts, 11 were included, involving 8648 patients from 1996 to 2023. Due to diverse artificial intelligence models and varied metrics, no data synthesis was performed. Models included random forest algorithms, support vector machine, deep learning, artificial neural network, convolutional neural network and least absolute shrinkage and selection operator regression. Artificial intelligence models' area under the curve values ranged from 0.74 to 0.9993 (slide level) and 0.9476 to 0.9956 (single-node level), outperforming traditional clinical guidelines.

CONCLUSION

Artificial intelligence models show promise in predicting lymph node metastasis in early-stage colorectal cancers, potentially refining clinical decisions and improving outcomes.

PROSPERO REGISTRATION NUMBER

CRD42023409094.

Novel and Annotated Long Noncoding RNAs Associated with Ischemia in the Human Heart

BACKGROUND

Long noncoding RNAs (lncRNAs) have been implicated in the pathogenesis of cardiovascular diseases. We aimed to identify novel lncRNAs associated with the early response to ischemia in the heart.

METHODS AND RESULTS

RNA sequencing data gathered from 81 paired left ventricle samples from patients undergoing cardiopulmonary bypass was collected before and after a period of ischemia. Novel lncRNAs were validated with Oxford Nanopore Technologies long-read sequencing. Gene modules associated with an early ischemic response were identified and the subcellular location of selected lncRNAs was determined with RNAscope. A total of 2446 mRNAs, 270 annotated lncRNAs and one novel lncRNA differed in response to ischemia (adjusted p < 0.001, absolute fold change >1.2). The novel lncRNA belonged to a gene module of highly correlated genes that also included 39 annotated lncRNAs. This module associated with ischemia (Pearson correlation coefficient = -0.69, p = 1 × 10^-23) and activation of cell death pathways (p < 6 × 10^-9). A further nine novel cardiac lncRNAs were identified, of which, one overlapped five cis-eQTL eSNPs for the gene RWD Domain-Containing Sumoylation Enhancer (RWDD3) and was itself correlated with RWDD3 expression (Pearson correlation coefficient -0.2, p = 0.002).

CONCLUSION

We have identified 10 novel lncRNAs, one of which was associated with myocardial ischemia and may have potential as a novel therapeutic target or early marker for myocardial dysfunction.

Assessment of intra-tumoural colorectal cancer prognostic biomarkers using RNA in situ hybridisation

Genome-wide expression studies using microarrays and RNAseq have increased our understanding of colorectal cancer development. Translating potential gene biomarkers from these studies for clinical utility has typically relied on PCR-based technology and immunohistochemistry. Results from these techniques are limited by tumour sample heterogeneity and the lack of correlation between mRNA transcript abundance and corresponding protein levels. The aim of this research was to investigate the clinical utility of the RNA in situ hybridisation technique, RNAscope^®, for measuring intra-tumoural gene expression of potential prognostic markers in a colorectal cancer cohort. Two candidate gene markers (GFI1 and TNFRSF11A) assessed in this study were identified from a previous study led by the The Cancer Genome Atlas (TCGA) Network, and analysis was performed on 112 consecutively collected, archival FFPE colorectal cancer tumour samples. Consistent with the TCGA Network study, we found reduced GFI1 expression was associated with high-grade and left-sided tumours, and reduced TNFRSF11A expression was associated with metastasis and high nodal involvement. RNAscope^® combined with image analysis also enabled quantification of GFI1 and TNFRSF11A mRNA expression levels at the single cell level, allowing cell-type determination. These data showed that reduced mRNA transcript abundance measured in patients with poorer prognosis occurred in carcinoma cells, and not lymphocytes, stromal cells or normal epithelial cells. To our knowledge, this is the first study to assess the intra-tumoural expression patterns of GFI1 and TNFRSF11A and to validate their microarray expression profiles using RNAscope. We also demonstrate the utility of RNAscope^® technology to show that expression differences are derived from carcinoma cells rather than from cells located in the tumour microenvironment.

Abstract 2645: Intratumoral assessment of candidate colorectal cancer prognostic markers using RNA in situ hybridization

Genome-wide expression studies using microarrays and RNAseq have increased our understanding of colorectal cancer development. Translating potential gene biomarkers from these studies for clinical utility has typically relied on PCR-based technology and immunohistochemistry. However, these techniques often suffer from limitations because of tumor sample heterogeneity, or a lack of correlation between protein and corresponding mRNA transcript levels. RNAscope® (Advanced Cell Diagnostics) is an RNA in situ hybridization method that overcomes these limitations by enabling single-molecule detection, while preserving cellular and tissue morphology. The aim of this research was to investigate the clinical utility of RNAscope to measure gene expression of two potential prognostic markers (GFI1 and TNFRSF11A) that were identified from a previously reported study from The Cancer Genome Atlas (TCGA) Network. We analyzed 112 consecutively collected archival formalin-fixed, paraffin-embedded tumors from a collection of well-characterized colorectal cancer cases. Consistent with TCGA Network data, we found significant associations for reduced GFI1 and TNFRSF11A mRNA expression and clinicopathologic features in our cohort. Reduced GFI1 expression measured by RNAscope was associated with high-grade tumor (P = 0.02) and left-sided tumor (P = 0.01). Reduced TNFRSF11A expression was associated with metastasis (P = 0.04) and high nodal involvement (P = 0.02). RNAscope was combined with semiautomated image analysis to provide quantification of mRNA expression at the single-cell level, allowing for cell-type determination of mRNA expression levels within a tumor. These data showed that GFI1 and TNFRSF11A were expressed at a significantly higher level in carcinoma cells compared to noncarcinoma cells (lymphocytes, stromal cells and normal cells). Thus, we were able to demonstrate that reduced expression measured in patients with poorer prognosis was not due to contamination of tumor samples with noncarcinoma cells. Limited information is available regarding TNFRSF11A's involvement in colorectal cancer. Gene knockdown of TNFRSF11A in colorectal cell line models showed changes in cell growth properties. To our knowledge, this is the first study to assess the intercellular expression patterns of GFI1 and TNFRSF11A as candidate prognostic markers in colorectal tumor. Furthermore, we show that intratumoral expression analysis of candidate genes by RNAscope is an informative technology for validating results from genome-wide expression studies.

Citation Format: Arthur Morley-Bunker, John Pearson, Margaret Currie, Helen Morrin, Martin Whitehead, Tim Eglinton, Logan Walker. Intratumoral assessment of candidate colorectal cancer prognostic markers using RNA in situ hybridization [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 2645.

Translating colorectal cancer genetics into clinically useful biomarkers

Colorectal cancer (CRC) is a major health problem worldwide accounting for over a million deaths annually. While many patients with Stage II and III CRC can be cured with combinations of surgery, radiotherapy and chemotherapy, this is morbid costly treatment and a significant proportion will suffer recurrence and eventually die of CRC. Increased understanding of the molecular pathogenesis of CRC has the potential to identify high risk patients and target therapy more appropriately. Despite increased understanding of the molecular events underlying CRC development, established molecular techniques have only produced a limited number of biomarkers suitable for use in routine clinical practice to predict risk, prognosis and response to treatment. Recent rapid technological developments, however, have made genomic sequencing of CRC more economical and efficient, creating potential for the discovery of genetic biomarkers that have greater diagnostic, prognostic and therapeutic capabilities for the management of CRC. This paper reviews the current understanding of the molecular pathogenesis of CRC, and summarizes molecular biomarkers that surgeons will encounter in current clinical use as well as those under development in clinical and preclinical trials. New molecular technologies are reviewed together with their potential impact on the understanding of the molecular pathogenesis of CRC and their potential clinical utility in classification, diagnosis, prognosis and targeting of therapy.