ColoGuidePro: a prognostic 7-gene expression signature for stage III colorectal cancer patients

Transcriptomic subtyping of gastrointestinal malignancies

Gastrointestinal (GI) cancers are highly heterogeneous at multiple levels. Tumor heterogeneity can be captured by molecular profiling, such as genetic, epigenetic, proteomic, and transcriptomic classification. Transcriptomic subtyping has the advantage of combining genetic and epigenetic information, cancer cell-intrinsic properties, and the tumor microenvironment (TME). Unsupervised transcriptomic subtyping systems of different GI malignancies have gained interest because they reveal shared biological features across cancers and bear prognostic and predictive value. Importantly, transcriptomic subtypes accurately reflect complex phenotypic states varying not only per tumor region, but also throughout disease progression, with consequences for clinical management. Here, we discuss methodologies of transcriptomic subtyping, proposed taxonomies for GI malignancies, and the challenges posed to clinical implementation, highlighting opportunities for future transcriptomic profiling efforts to optimize clinical impact.

Discovery and validation of a 10-gene predictive signature for response to adjuvant chemotherapy in stage II and III colon cancer

Identifying patients with stage II and III colon cancer who will benefit from 5-fluorouracil (5-FU)-based adjuvant chemotherapy is crucial for the advancement of personalized cancer therapy. We employ a semi-supervised machine learning approach to analyze a large dataset with 933 stage II and III colon cancer samples. Our analysis leverages gene regulatory networks to discover an 18-gene prognostic signature and to explore a 10-gene signature that potentially predicts chemotherapy benefits. The 10-gene signature demonstrates strong prognostic power and shows promising potential to predict chemotherapy benefits. We establish a robust clinical assay on the NanoString nCounter platform, validated in a retrospective formalin-fixed paraffin-embedded (FFPE) cohort, which represents an important step toward clinical application. Our study lays the groundwork for improving adjuvant chemotherapy and potentially expanding into immunotherapy decision-making in colon cancer. Future prospective studies are needed to validate and establish the clinical utility of the 10-gene signature in clinical settings.

Identification of Copper Homeostasis-Related Gene Signature for Predicting Prognosis in Patients with Epithelial Ovarian Cancer

Objectives

This research aims to establish a copper homeostasis-related gene signature for predicting the prognosis of epithelial ovarian cancer and to investigate its underlying mechanisms.

Methods

We mainly constructed the copper homeostasis-related gene signature by LASSO regression analysis. Then multiple methods were used to evaluate the independent predictive ability of the model and explored the mechanisms.

Results

The 15-copper homeostasis-related gene (15-CHRG) signature was successfully established. Utilizing an optimal cut-off value of 0.35, we divided the training dataset into high-risk and low-risk subgroups. Kaplan-Meier analysis revealed that survival times for the high-risk subgroup were significantly shorter than those in the low-risk group (P < .05). Additionally, the Area Under the Curve (AUC) of the 15-CHRG signature achieved 0.822 at 1 year, 0.762 at 3 years, and 0.696 at 5 years in the training set. COX regression analysis confirmed the 15-CHRG signature as both accurate and independent. Gene set enrichment (GSEA), Kyoto Encyclopedia of Gene and Genome (KEGG) and Gene Ontology (GO) analysis showed that there were significant differences in apoptosis, p53 pathway, protein synthesis, hydrolase and transport-related pathways between high-risk group and low-risk group. In tumor immune cell (TIC) analysis, the increased expression of resting mast cells was positively correlated with the risk score.

Conclusion

Consequently, the 15-CHRG signature shows significant potential as a method for accurately predicting clinical outcomes and treatment responses in patients with epithelial ovarian cancer.

Development of a novel senescence-related gene signature to predict clinical outcomes, immune landscape, and chemotherapeutic sensitivity in oral squamous cell carcinoma

BACKGROUND

Cellular senescence significantly associates with tumor initiation, progression, and therapeutic response across multiple cancers. Here, we sought to develop a novel senescence-related genes (SRGs)-derived signature for oral squamous cell carcinoma (OSCC) prognostication and therapeutic response prediction.

METHODS

OSCC-specific SRG prognostic signature was established with univariate Cox regression, Kaplan-Meier survival, and LASSO-penalized multivariate Cox regression analyses. A SRG nomogram integrating this signature and selected clinicopathological parameters were constructed by multivariate Cox regression. SiRNA-mediated gene knockdown was exploited to validate its function in vitro. The utilities of SRG signature in predicting immune status and chemotherapeutic sensitivities were analyzed.

RESULTS

The prognostic performance of SRG signature/nomogram was satisfactory in multiple independent cohorts. CDK1 knockdown induced senescence phenotype in vitro. Moreover, SRG signature scores negatively correlated with tumor-infiltrating immune cells and associated with multiple chemotherapeutic drug sensitivities.

CONCLUSIONS

Our results established SRG-derived signature/nomogram as powerful predictors for prognosis and chemotherapeutic response for OSCC.

A novel immune-related lncRNA signature predicts the prognosis and immune landscape in ccRCC

BACKGROUND

As one of the most common tumors, the pathogenesis and progression of clear cell renal cell carcinoma (ccRCC) in the immune microenvironment are still unknown.

METHODS

The differentially expressed immune-related lncRNA (DEirlncRNA) was screened through co-expression analysis and the limma package of R, which based on the ccRCC project of the TCGA database. Then, we designed the risk model by irlncRNA pairs. In RCC patients, we have compared the area under the curve, calculated the Akaike Information Criterion (AIC) value of the 5-year receiver operating characteristic curve, determined the cut-off point, and established the optimal model for distinguishing the high-risk group from the low-risk group. We used the model for immune system assessment, immune point detection and drug sensitivity analysis after verifying the feasibility of the above model through clinical features.

RESULTS

In our study, 1541 irlncRNAs were included. 739 irlncRNAs were identified as DEirlncRNAs to construct irlncRNA pairs. Then, 38 candidate DEirlncRNA pairs were included in the best risk assessment model through improved LASSO regression analysis. As a result, we found that in addition to age and gender, T stage, M stage, N stage, grade and clinical stage are significantly related to risk. Moreover, univariate and multivariate Cox regression analysis results reveals that in addition to gender, age, grade, clinical stage and risk score are independent prognostic factors. The results show that patients in the high-risk group are positively correlated with tumor infiltrating immune cells when the above model is applied to the immune system. But they are negatively correlated with endothelial cells, macrophages M2, mast cell activation, and neutrophils. In addition, the risk model was positively correlated with overexpressed genes (CTLA, LAG3 and SETD2, P<0.05). Finally, risk models can also play as an important role in predicting the sensitivity of targeted drugs.

CONCLUSIONS

The new risk model may be a new method to predict the prognosis and immune status of ccRCC.

Potential value of mitochondrial regulatory pathways in the clinical application of clear cell renal cell carcinoma: a machine learning-based study

BACKGROUND

Renal clear cell carcinoma (RCC) is a common cancer in urinary system with increasing incidence. At present, targeted therapy and immunotherapy are the main therapeutic programs in clinical therapy. To develop novel drugs and provide new ideas for clinical therapy, the identification of potential ccRCC subtypes and potential target genes or pathways has become a current research focus.

AIM

The aim of this study was to explore the underlying mechanisms of mitochondrial function in ccRCC. This regulatory pathway is closely related to tumor development and metastasis in ccRCC patients, and their abnormal changes may affect the prognosis of cancer patients. Therefore, we decided to construct a prognostic model of ccRCC patients based on mitochondrial regulatory genes, aiming to provide new methods and ideas for clinical therapy.

RESULT

The 5-year survival prediction model based on iterative LASSO reached 0.746, and the cox model based on coxph reached C-index = 0.77, integrated c/D AUC = 0.61, and integrated brier score = 0.14. The rsf model based on randomForestSRC was built with C-index = 0.82, integrated c/D AUC = 0.69, and integrated brier score = 0.11. The results show that mitochondrial regulatory pathway is a potential target pathway for clinical therapy of ccRCC, which can provide guidelines for clinical targeted therapy, immunotherapy and other first-line therapy.

ReProMSig: an integrative platform for development and application of reproducible multivariable models for cancer prognosis supporting guideline-based transparent reporting

Adequate reporting is essential for evaluating the performance and clinical utility of a prognostic prediction model. Previous studies indicated a prevalence of incomplete or suboptimal reporting in translational and clinical studies involving development of multivariable prediction models for prognosis, which limited the potential applications of these models. While reporting templates introduced by the established guidelines provide an invaluable framework for reporting prognostic studies uniformly, there is a widespread lack of qualified adherence, which may be due to miscellaneous challenges in manual reporting of extensive model details, especially in the era of precision medicine. Here, we present ReProMSig (Reproducible Prognosis Molecular Signature), a web-based integrative platform providing the analysis framework for development, validation and application of a multivariable prediction model for cancer prognosis, using clinicopathological features and/or molecular profiles. ReProMSig platform supports transparent reporting by presenting both methodology details and analysis results in a strictly structured reporting file, following the guideline checklist with minimal manual input needed. The generated reporting file can be published together with a developed prediction model, to allow thorough interrogation and external validation, as well as online application for prospective cases. We demonstrated the utilities of ReProMSig by development of prognostic molecular signatures for stage II and III colorectal cancer respectively, in comparison with a published signature reproduced by ReProMSig. Together, ReProMSig provides an integrated framework for development, evaluation and application of prognostic/predictive biomarkers for cancer in a more transparent and reproducible way, which would be a useful resource for health care professionals and biomedical researchers.

Principles of Molecular Utility for CMS Classification in Colorectal Cancer Management

Colorectal cancer (CRC) is the second cause of cancer-related deaths in both sexes globally and presents different clinical outcomes that are described by a range of genomic and epigenomic alterations. Despite the advancements in CRC screening plans and treatment strategies, the prognosis of CRC is dismal. In the last two decades, molecular biomarkers predictive of prognosis have been identified in CRC, although biomarkers predictive of treatment response are only available for specific biological drugs used in stage IV CRC. Translational clinical trials mainly based on "omic" strategies allowed a better understanding of the biological heterogeneity of CRCs. These studies were able to classify CRCs into subtypes mainly related to prognosis, recurrence risk, and, to some extent, also to treatment response. Accordingly, the comprehensive molecular characterizations of CRCs, including The Cancer Genome Atlas (TCGA) and consensus molecular subtype (CMS) classifications, were presented to improve the comprehension of the genomic and epigenomic landscapes of CRCs for a better patient management. The CMS classification obtained by the CRC subtyping consortium categorizes CRC into four consensus molecular subtypes (CMS1-4) characterized by different prognoses. In this review, we discussed the CMS classification in different settings with a focus on its relationships with precursor lesions, tumor immunophenotype, and gut microbiota, as well as on its role in predicting prognosis and/or response to pharmacological treatments, as a crucial step towards precision medicine.

Development of a novel prognostic signature derived from enhancer RNA-regulated genes in head neck squamous cell carcinoma

BACKGROUND

Enhancer RNAs (eRNAs) are increasingly recognized as prognostic biomarkers-across human cancers. Here, we sought to develop a novel eRNA-regulated genes (ERGs)-derived prognostic signature for head neck squamous cell carcinoma (HNSCC).

METHODS

Candidate ERGs were identified via co-expression between individual survival-related eRNAs and their putative targets by Spearman's correlation analyses. The ERG signature was developed by univariate Cox regression, Kaplan-Meier survival analysis and maximum AUC in 1000 iterations of LASSO-penalized multivariate Cox regression. An ERG nomogram incorporating ERG signature and selected clinicopathological parameters were constructed by multivariate Cox regression. Biological roles of eRNA of interest were further explored in vitro.

RESULTS

The ERG signature successfully stratified patients into subgroups with distinct survival in multiple cohorts. An ERG nomogram was developed with satisfactory performance in prognostication. Inhibition of ENSR00000165816 significantly reduced transcript level of SLC2A9 and impaired cell proliferation and invasion.

CONCLUSION

Our results establish ERG signature and nomogram as powerful prognostic predictors for HNSCC.

MicroRNA-129-3p Inhibits Colorectal Cancer Proliferation

MicroRNA-129-3p plays a pro-cancer role in colorectal cancer by down-regulating BIM. This study intends to assess its role in colorectal cancer cells. A total of 30 colorectal cancers and 10 paracancerous samples were obtained to measure MicroRNA-129-3p expression by PCR. Colorectal cancer cells were transfected with miR-129-3p mimic or inhibitor followed by analysis of cell growth, apoptosis. miR-129-3p expression was significantly lower in colorectal cancer tissues than that in cancer adjacent tissues (P <0.05). miR-129-3p overexpression after mimic transfection significantly inhibited cancer cell viability and promoted apoptosis (P < 0.05). Moreover, it also significantly downregulated E2F5, BIM and FoxO3a in colorectal cancer cells. Furthermore, E2F5 was targeted by miR-129-3p. In conclusion, miR-129-3p inhibits colorectal cancer cell proliferation via targeting E2F5 to downregulate BIM, indicating that it might be a target for treating colorectal cancer.

Systematic evaluation of tumor microenvironment and construction of a machine learning model to predict prognosis and immunotherapy efficacy in triple-negative breast cancer based on data mining and sequencing validation

Background: The role of the tumor microenvironment (TME) in predicting prognosis and therapeutic efficacy has been demonstrated. Nonetheless, no systematic studies have focused on TME patterns or their function in the effectiveness of immunotherapy in triple-negative breast cancer.

Methods: We comprehensively estimated the TME infiltration patterns of 491 TNBC patients from four independent cohorts, and three cohorts that received immunotherapy were used for validation. The TME subtypes were comprehensively evaluated based on immune cell infiltration levels in TNBC, and the TRG score was identified and systematically correlated with representative tumor characteristics. We sequenced 80 TNBC samples as an external validation cohort to make our conclusions more convincing.

Results: Two TME subtypes were identified and were highly correlated with immune cell infiltration levels and immune-related pathways. More representative TME-related gene (TRG) scores calculated by machine learning could reflect the fundamental characteristics of TME subtypes and predict the efficacy of immunotherapy and the prognosis of TNBC patients. A low TRG score, characterized by activation of immunity and ferroptosis, indicated an activated TME phenotype and better prognosis. A low TRG score showed a better response to immunotherapy in TNBC by TIDE (Tumor Immune Dysfunction and Exclusion) analysis and sensitivity to multiple drugs in GDSC (Genomics of Drug Sensitivity in Cancer) analysis and a significant therapeutic advantage in patients in the three immunotherapy cohorts.

Conclusion: TME subtypes played an essential role in assessing the diversity and complexity of the TME in TNBC. The TRG score could be used to evaluate the TME of an individual tumor to enhance our understanding of the TME and guide more effective immunotherapy strategies.

A 9-LncRNA Signature for Predicting Prognosis and Immune Response in Diffuse Large B-Cell Lymphoma

Diffuse large B-cell lymphoma (DLBCL) is a biologically and clinically heterogeneous disease that requires personalized clinical treatment. To assign patients into different risk categories, cytogenetic abnormalities and genetic mutations have been widely applied to the prognostic stratification of DLBCL. Increasing evidence has demonstrated that deregulated epigenetic modifications and long noncoding RNAs (lncRNAs) contribute to the initiation and progression of DLBCL. However, specific lncRNAs that affect epigenetic regulation and their value in predicting prognosis and therapy response remain uncertain. Here, 2,025 epigenetic-related genes were selected, and 9 lncRNAs (PRKCQ-AS1, C22orf34, HCP5, AC007389.3, APTR, SNHG19, ELFN1-AS1, LINC00487, and LINC00877) were tested and validated to establish an lncRNA-regulating epigenetic event signature (ELncSig). ELncSig, which was established based on independent lymphoma datasets, could distinguish different survival outcomes. Functional characterization of ELncSig showed that it could be an indicator of the immune microenvironment and is correlated with distinctive mutational characteristics. Univariate and multivariate analyses showed that ELncSig was independent of traditional prognostic factors. The novel immune-related ELncSig exhibits promising clinical prognostic value for DLBCL.

A novel immune-related microRNA signature for prognosis of thymoma

INTRODUCTION

Immune microenvironment and microRNAs serve as common predictors for diagnosis and prognosis of tumors.

METHODS

Expression of 122 genes and 126 microRNAs in thymoma was obtained from TCGA database. The proportion of tumor-infiltrating cells was calculated, and IMRS was constructed. TREM2hi score was calculated before functional enrichment analysis on gene sets.

RESULTS

IMRS3, TREM2hi score, and CD8+ T lymphocyte abundance were significantly different among WHO classifications. WHO classification, Masaoka staging, and miR-130b-5p, miR-1307-3p, miR-425-5p, CD8, CD68, and CCL18 expression were prognostic factors for relapse-free survival and overall survival. IMRS3 upregulation polarized macrophages into M2, which rejected CD8+ T and other effector lymphocytes to promote thymoma malignant progression.

CONCLUSIONS

BRRS may present a novel immune-related microRNA signature for TET prognosis.

Colorectal cancer concurrent gene signature based on coherent patterns between genomic and transcriptional alterations

Background

The aim of the study was to enhance colorectal cancer prognostication by integrating single nucleotide polymorphism (SNP) and gene expression (GE) microarrays for genomic and transcriptional alteration detection; genes with concurrent gains and losses were used to develop a prognostic signature.

Methods

The discovery dataset comprised 32 Taiwanese colorectal cancer patients, of which 31 were assayed for GE and copy number variations (CNVs) with Illumina Human HT-12 BeadChip v4.0 and Omni 25 BeadChip v1.1. Concurrent gains and losses were declared if coherent manners were observed between GE and SNP arrays. Concurrent genes were also identified in The Cancer Genome Atlas Project (TCGA) as the secondary discovery dataset (n = 345).

Results

The “universal” concurrent genes, which were the combination of z-transformed correlation coefficients, contained 4022 genes. Candidate genes were evaluated within each of the 10 public domain microarray datasets, and 1655 (2000 probe sets) were prognostic in at least one study. Consensus across all datasets was used to build a risk predictive model, while distinct relapse-free/overall survival patterns between defined risk groups were observed among four out of five training datasets. The predictive accuracy of recurrence, metastasis, or death was between 61 and 86% (cross-validation area under the receiver operating characteristic (ROC) curve: 0.548-0.833) from five independent validation studies.

Conclusion

The colorectal cancer concurrent gene signature is prognostic in terms of recurrence, metastasis, or mortality among 1746 patients. Genes with coherent patterns between genomic and transcriptional contexts are more likely to provide prognostication for colorectal cancer.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-022-09627-9.

An m6A/m5C/m1A/m7G-Related Long Non-coding RNA Signature to Predict Prognosis and Immune Features of Glioma

Background: Gliomas are the most common and fatal malignant type of tumor of the central nervous system. RNA post-transcriptional modifications, as a frontier and hotspot in the field of epigenetics, have attracted increased attention in recent years. Among such modifications, methylation is most abundant, and encompasses N6-methyladenosine (m6A), 5-methylcytosine (m5C), N1 methyladenosine (m1A), and 7-methylguanosine (m7G) methylation.

Methods: RNA-sequencing data from healthy tissue and low-grade glioma samples were downloaded from of The Cancer Genome Atlas database along with clinical information and mutation data from glioblastoma tumor samples. Forty-nine m6A/m5C/m1A/m7G-related genes were identified and an m6A/m5C/m1A/m7G-lncRNA signature of co-expressed long non-coding RNAs selected. Least absolute shrinkage and selection operator Cox regression analysis was used to identify 12 m6A/m5C/m1A/m7G-related lncRNAs associated with the prognostic characteristics of glioma and their correlation with immune function and drug sensitivity analyzed. Furthermore, the Chinese Glioma Genome Atlas dataset was used for model validation.

Results: A total of 12 m6A/m5C/m1A/m7G-related genes (AL080276.2, AC092111.1, SOX21-AS1, DNAJC9-AS1, AC025171.1, AL356019.2, AC017104.1, AC099850.3, UNC5B-AS1, AC006064.2, AC010319.4, and AC016822.1) were used to construct a survival and prognosis model, which had good independent prediction ability for patients with glioma. Patients were divided into low and high m6A/m5C/m1A/m7G-LS groups, the latter of which had poor prognosis. In addition, the m6A/m5C/m1A/m7G-LS enabled improved interpretation of the results of enrichment analysis, as well as informing immunotherapy response and drug sensitivity of patients with glioma in different subgroups.

Conclusion: In this study we constructed an m6A/m5C/m1A/m7G-LS and established a nomogram model, which can accurately predict the prognosis of patients with glioma and provides direction toward promising immunotherapy strategies for the future.

Application of the Machine-Learning Model to Improve Prediction of Non-Sentinel Lymph Node Metastasis Status Among Breast Cancer Patients

Background

Performing axillary lymph node dissection (ALND) is the current standard option after a positive sentinel lymph node (SLN). However, whether 1–2 metastatic SLNs require ALND is debatable. The probability of metastasis in non-sentinel lymph nodes (NSLNs) can be calculated using nomograms. In this study, we developed an individualized model using machine-learning (ML) methods to select potential variables, which influence NSLN metastasis.

Materials and Methods

Cohorts of patients with early breast cancer who underwent SLN biopsy and ALND between 2012 and 2021 were created (training cohort, N 157 and validation cohort, N 58) for the development of the nomogram. Three ML methods were trained in the training set to create a strong predictive model. Finally, the multiple iterations of the least absolute shrinkage and selection operator regression method were used to determine the variables associated with NSLN status.

Results

Four independent variables (positive SLN number, absence of lymph node hilum, lymphovascular invasion (LVI), and total number of SLNs harvested) were combined to generate the nomogram. The area under the receiver operating characteristic curve (AUC) value of 0.759 was obtained in the entire set. The AUC values for the training set and the test set were 0.782 and 0.705, respectively. The Hosmer-Lemeshow test of the model fit accuracy was identified with p = 0.759.

Conclusion

This study developed a nomogram that incorporates ultrasound (US)-related variables using the ML method and serves to clinically predict the non-metastatic status of NSLN and help in the selection of the appropriate treatment option.

A new risk model based on a 11-m6A-related lncRNA signature for predicting prognosis and monitoring immunotherapy for gastric cancer

Objective

N⁶-methyladenosine (m⁶A) mRNA modification triggers malignant behaviors of tumor cells and thereby drives malignant progression in gastric cancer (GC). However, data regarding the prognostic values of m⁶A RNA methylation-related long non-coding RNAs (lncRNAs) in GC are very limited in the literature. We aimed to investigate the prognostic potential of m⁶A-related lncRNAs in predicting prognosis and monitoring immunotherapy efficacy in GC patients.

Methods

Transcriptome and clinical data were obtained from GC biopsies from Cancer Genome Atlas (TCGA). M⁶A-related lncRNAs associated with GC were identified by constructing a co-expression network, and the gene pairs differentially expressed in GC were selected using univariate analysis. We constructed a risk model based on prognosis-related lncRNA pairs selected using the LASSO algorithm and quantified the best cutoff by comparing the area under the curve (AUC) for risk stratification. A risk model with the optimal discrimination between high- and low-risk GC patients was established. Its feasibility for overall survival prediction and discrimination of clinicopathological features, tumor-infiltrating immune cells, and biomarkers of immune checkpoint inhibitors between high- and low-risk groups were assessed.

Results

Finally, we identified 11 m⁶A-related lncRNA pairs associated with GC prognosis based on transcriptome analysis of 375 GC specimens and 32 normal tissues. A risk model was constructed with an AUC of 0.8790. We stratified GC patients into high- and low-risk groups at a cutoff of 1.442. As expected, patients in the low-risk group had longer overall survival versus the high-risk group. Infiltration of cancer-associated fibroblasts, endothelial cells, macrophages, particularly M2 macrophages, and monocytes was more severe in high-risk patients than low-risk individuals, who exhibited high CD4⁺ Th1 cell infiltration in GC. Altered expressions of immune-related genes were observed in both groups. PD-1 and LAG3 expressions were found higher in low-risk patients than high-risk patients. Immunotherapy, either single or combined use of PD-1 or CTLA4 inhibitors, had better efficacy in low-risk patients than high-risk patients.

Conclusion

The new risk model based on a 11-m⁶A-related lncRNA signature can serve as an independent predictor for GC prognosis prediction and may aid in the development of personalized immunotherapy strategies for patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-021-09062-2.

Construction of a new immune-related lncRNA model and prediction of treatment and survival prognosis of human colon cancer

Background

An increasing number of studies have shown that immune-related long noncoding RNAs (lncRNAs) do not require a unique expression level. This finding may help predict the survival and drug sensitivity of patients with colon cancer.

Methods

We retrieved original transcriptome and clinical data from The Cancer Genome Atlas (TCGA), sorted the data, differentiated mRNAs and lncRNAs, and then downloaded immune-related genes. Coexpression analysis predicted immune-related lncRNAs (irlncRNAs) and univariate analysis identified differentially expressed irlncRNAs (DEirlncRNAs). We have also amended the lasso pending region. Next, we compared the areas under the curve (AUCs), counted the Akaike information standard (AIC) value of the 3-year receiver operating characteristic (ROC) curve, and determined the cutoff point to establish the best model to differentiate the high or low disease risk group of colon cancer patients.

Results

We reevaluated the patients regarding the survival rate, clinicopathological features, tumor-infiltrating immune cells, immunosuppressive biomarkers, and chemosensitivity. A total of 155 irlncRNA pairs were confirmed, 31 of which were involved in the Cox regression model. After the colon cancer patients were regrouped according to the cutoff point, we could better distinguish the patients based on adverse survival outcomes, invasive clinicopathological features, the specific tumor immune cell infiltration status, high expression of immunosuppressive biomarkers, and low chemosensitivity.

Conclusions

In this study, we established a characteristic model by pairing irlncRNAs to better predict the survival rate, chemotherapy efficacy, and prognostic value of patients with colon cancer.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12957-022-02508-2.

The Expression Pattern of Hypoxia-Related Genes Predicts the Prognosis and Mediates Drug Resistance in Colorectal Cancer

Background: Colorectal cancer (CRC) is one of the leading causes of cancer-related deaths worldwide. However, due to the heterogeneity of CRC, the clinical therapy outcomes differ among patients. There is a need to identify predictive biomarkers to efficiently facilitate CRC treatment and prognosis.

Methods: The expression profiles from Gene Expression Omnibus (GEO) database were used to identify cancer hallmarks associated with CRC outcomes. An accurate gene signature based on the prognosis related cancer hallmarks was further constructed.

Results: Hypoxia was identified to be the primary factor that could influence CRC outcomes. Sixteen hypoxia-related genes were selected to construct a risk gene signature (HGS) associated with individuals’ prognosis, which was validated in three independent cohorts. Further, stromal and immune cells in tumor microenvironment (TME) were found to be associated with hypoxia. Finally, among the 16 hypoxia-related genes, six genes (DCBLD2, PLEC, S100A11, PLAT, PPAP2B and LAMC2) were identified as the most attributable ones to drug resistance.

Conclusion: HGS can accurately predict CRC prognosis. The expression of the drug resistance-related genes is critical in CRC treatment decision-making.

Construction and Comprehensive Prognostic Analysis of a Novel Immune-Related lncRNA Signature and Immune Landscape in Gastric Cancer

Gastric cancer (GC) is a malignant tumor with high mortality and poor prognosis. Immunotherapies, especially immune checkpoint inhibitors (ICI), are widely used in various tumors, but patients with GC do not benefit much from immunotherapies. Therefore, effective predictive biomarkers are urgently needed for GC patients to realize the benefits of immunotherapy. Recent studies have indicated that long noncoding RNAs (lncRNAs) could be used as biomarkers in the immune landscape of multiple tumors. In this study, we constructed a novel immune-related lncRNA (irlncRNA) risk model to predict the survival and immune landscape of GC patients. First, we identified differentially expressed irlncRNAs (DEirlncRNAs) from RNA-Seq data of The Cancer Genome Atlas (TCGA). By using various algorithms, we constructed a risk model with 11 DEirlncRNA pairs. We then tested the accuracy of the risk model, demonstrating that the risk model has good efficiency in predicting the prognosis of GC patients. Inner validation sets were further used to confirm the effectiveness of the risk model. In addition, our risk model has a preferable performance in predicting the immune infiltration status of tumors, immune checkpoint status of the patients, and immunotherapy score. In conclusion, our risk model may provide insights into the prognosis of and immunotherapy strategy for GC.

Construction of a novel immune-related lncRNA signature and its potential to predict the immune status of patients with hepatocellular carcinoma

Background

The accuracy of existing biomarkers for predicting the prognosis of hepatocellular carcinoma (HCC) is not satisfactory. It is necessary to explore biomarkers that can accurately predict the prognosis of HCC.

Methods

In this study, original transcriptome data were downloaded from The Cancer Genome Atlas (TCGA) database. Immune-related long noncoding ribonucleic acids (irlncRNAs) were identified by coexpression analysis, and differentially expressed irlncRNA (DEirlncRNA) pairs were distinguished by univariate analysis. In addition, the least absolute shrinkage and selection operator (LASSO) penalized regression was modified. Next, the cutoff point was determined based on the area under the curve (AUC) and Akaike information criterion (AIC) values of the 5-year receiver operating characteristic (ROC) curve to establish an optimal model for identifying high-risk and low-risk groups of HCC patients. The model was then reassessed in terms of clinicopathological features, survival rate, tumor-infiltrating immune cells, immunosuppressive markers, and chemotherapy efficacy.

Results

A total of 1009 pairs of DEirlncRNAs were recognized in this study, 30 of these pairs were included in the Cox regression model for subsequent analysis. After regrouping according to the cutoff point, we could more effectively identify factors such as aggressive clinicopathological features, poor survival outcomes, specific immune cell infiltration status of tumors, high expression level of immunosuppressive biomarkers, and low sensitivity to chemotherapy drugs in HCC patients.

Conclusions

The nonspecific expression level signature involved with irlncRNAs shows promising clinical value in predicting the prognosis of HCC patients.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12885-021-09059-x.

A risk model based on autophagy-related lncRNAs for predicting prognosis and efficacy of immunotherapy and chemotherapy in gastric cancer patients

Long non-coding RNAs (lncRNAs) are a class of non-protein-coding RNAs essential to the occurrence and development of gastric cancer (GC). We aimed to identify critical lncRNA pairs to construct a prognostic model and assess its performances in prognosis and efficacy prediction in GC patients receiving immunotherapy and chemotherapy. We searched transcriptome and clinical data of GC patients from The Cancer Genome Atlas (TCGA) database. Autophagy-related lncRNAs were identified using co-expression network analysis, and lncRNA pairs with prognostic value were selected using pairwise transcriptome analysis. The gene pairs were subjected to LASSO algorithm for identification of optimal gene pairs for risk model construction. Patients were classified into the low-risk and high-risk groups with the RiskScore as a cutoff. Finally, 9 optimal gene pairs were identified in the LASSO algorithm model for construction of a lncRNA prognostic risk model. For predictive performances, it successfully predicted a shorter survival of high-risk patients than that obtained in low-risk individuals (P < 0.001). It showed moderate AUC (area under the curve) values for 1-, 2-, and 3-year overall survival prediction of 0.713 and could serve as an independent predictor for GC prognosis. Compared to the low-risk group, high-risk patients had higher expressions of marker genes for immune checkpoint inhibitors (ICIs) and showed higher sensitivity to the chemotherapy agents, rapamycin, bexarotene, and bicalutamide. Our findings demonstrate a robust prognostic model based on nine autophagy-related lncRNA pairs for GC. It acts as an independent predictor for survival and efficacy prediction of immunotherapy and chemotherapy in GC patients.

LncRNAs Associated with Chemoradiotherapy Response and Prognosis in Locally Advanced Rectal Cancer

BACKGROUND

There are only limited studies on the long non-coding RNAs (lncRNAs) associated with neoadjuvant chemoradiotherapy (NCRT) response and prognosis of locally advanced rectal cancer (LARC) patients. This study identified lncRNAs associated with NCRT response and prognosis in CRC patients and explored their potential predictive mechanisms.

METHODS

The study subjected the LncRNA expression profiles from our previous gene chip data to LASSO and identified a four-lncRNA signature that predicted NCRT response and prognosis. A Cox regression model was subsequently performed to identify the prognostic risk factors. The function of LINC00909, the lncRNA with the most powerful predictive ability, was finally identified in vivo and in vitro using CRC cell lines.

RESULTS

A comparison of the relative lncRNA expression of NCRT-responsive and non-responsive patients revealed four hub lncRNAs: DBET, LINC00909, FLJ33534, and HSD52 with AUC = 0.68, 0.73, 0.73, and 0.70, respectively (all p < 0.05). COX regression analysis further demonstrated that DBET, LINC00909 and FLJ33534 were associated with the DFS in CRC patients. The expression of the four lncRNAs was also significant in LARC patients who had not undergone NCRT (all p < 0.05). A risk score model was subsequently constructed based on the results of the multivariate COX analysis and used to predict NCRT response and prognosis in the CRC and LARC patients. The expression and prognosis of DBET, LINC00909 and FLJ33534 in the CRC tissues were further validated in the R2 platform and Oncomine database. Notably, overexpression of the LINC00909 increased the cell line resistance to the 5-FU and radiotherapy in vivo and in vitro.

CONCLUSION

DBET, LINC00909, and FLJ33534 are potential novel biomarkers for predicting NCRT response and prognosis in CRC patients. In particular, LINC00909 is an effective oncogene in CRC that could be used as a novel therapeutic target to enhance NCRT response.

Analysis of core genes for colorectal cancer prognosis based on immune and stromal scores

Background

Colorectal cancer (CRC) is one of the most common malignancies.An early diagnosis and an accurate prognosis are major focuses of CRC research. Tumor microenvironment cells and the extent of infiltrating immune and stromal cells contribute significantly to the tumor prognosis.

Methods

Immune and stromal scores were calculated based on the ESTIMATE algorithm using the sample expression profile of the The Cancer Genome Atlas (TCGA) database. GSE102479 was used as the validation database. Differentially expressed genes whose expression was significantly associated with the prognosis of CRC patients were identified based on the immune matrix score. Survival analysis was conducted on the union of the differentially expressed genes. A protein–protein interaction (PPI) network was constructed using the STRING database to identify the closely connected modules. To conduct functional enrichment analysis of the relevant genes, GO and KEGG pathway analyses were performed with Cluster Profiler. Pivot analysis of the ncRNAs and TFs was performed by using the RAID2.0 database and TRRUST v2 database. TF-mRNA regulatory relationships were analyzed in the TRRUST V2 database. Hubgene targeting relationships were screened in the TargetScan, miRTarBase and miRDB databases. The SNV data of the hub genes were analyzed by using the R maftools package. A ROC curve was drawn based on the TCGA database. The proportion of immune cells was estimated using CIBERSORT and the LM22 feature matrix.

Results

The results showed that the matrix score was significantly correlated with colorectal cancer stage T. A total of 789 differentially expressed genes and 121 survival-related prognostic genes were identified. The PPI network showed that 22 core genes were related to the CRC prognosis. Furthermore, four ncRNAs that regulated the core prognosis genes, 11 TFs with regulatory effects on the core prognosis genes, and two drugs, quercetin and pseudoephedrine, that have regulatory effects on colorectal cancer were also identified.

Conclusions

We obtained a list of tumor microenvironment-related genes for CRC patients. These genes could be useful for determining the prognosis of CRC patients. To confirm the function of these genes, additional experiments are necessary.

Unsupervised Learning Framework With Multidimensional Scaling in Predicting Epithelial-Mesenchymal Transitions

Clustering tumor metastasis samples from gene expression data at the whole genome level remains an arduous challenge, in particular, when the number of experimental samples is small and the number of genes is huge. We focus on the prediction of the epithelial-mesenchymal transition (EMT), which is an underlying mechanism of tumor metastasis, here, rather than tumor metastasis itself, to avoid confounding effects of uncertainties derived from various factors. In this paper, we propose a novel model in predicting EMT based on multidimensional scaling (MDS) strategies and integrating entropy and random matrix detection strategies to determine the optimal reduced number of dimension in low dimensional space. We verified our proposed model with the gene expression data for EMT samples of breast cancer and the experimental results demonstrated the superiority over state-of-the-art clustering methods. Furthermore, we developed a novel feature extraction method for selecting the significant genes and predicting the tumor metastasis. The source code is available at "https://github.com/yushanqiu/yushan.qiu-szu.edu.cn".

Immune-Related LncRNAs Affect the Prognosis of Osteosarcoma, Which Are Related to the Tumor Immune Microenvironment

Background: Abnormal expression of lncRNA is closely related to the occurrence and metastasis of osteosarcoma. The tumor immune microenvironment (TIM) is considered to be an important factor affecting the prognosis and treatment of osteosarcoma. This study aims to explore the effect of immune-related lncRNAs (IRLs) on the prognosis of osteosarcoma and its relationship with the TIM.

Methods: Ninety-five osteosarcoma samples from the TARGET database were included. Iterative LASSO regression and multivariate Cox regression analysis were used to screen the IRLs signature with the optimal AUC. The predict function was used to calculate the risk score and divide osteosarcoma into a high-risk group and low-risk group based on the optimal cut-off value of the risk score. The lncRNAs in IRLs signature that affect metastasis were screened for in vitro validation. Single sample gene set enrichment analysis (ssGSEA) and ESTIMATE algorithms were used to evaluate the role of TIM in the influence of IRLs on osteosarcoma prognosis.

Results: Ten IRLs constituted the IRLs signature, with an AUC of 0.96. The recurrence and metastasis rates of osteosarcoma in the high-risk group were higher than those in the low-risk group. In vitro experiments showed that knockdown of lncRNA (AC006033.2) could increase the proliferation, migration, and invasion of osteosarcoma. ssGSEA and ESTIMATE results showed that the immune cell content and immune score in the low-risk group were generally higher than those in the high-risk group. In addition, the expression levels of immune escape-related genes were higher in the high-risk group.

Conclusion: The IRLs signature is a reliable biomarker for the prognosis of osteosarcoma, and they alter the prognosis of osteosarcoma. In addition, IRLs signature and patient prognosis may be related to TIM in osteosarcoma. The higher the content of immune cells in the TIM of osteosarcoma, the lower the risk score of patients and the better the prognosis. The higher the expression of immune escape-related genes, the lower the risk score of patients and the better the prognosis.

Characterization of Immune-Related Long Non-coding RNAs to Construct a Novel Signature and Predict the Prognosis and Immune Landscape of Soft Tissue Sarcoma

Background: Increasing evidence has demonstrated that immune-related long non-coding RNAs (irlncRNAs) are critically involved in tumor initiation and progression and associated with the prognosis of various cancers. However, their role in soft tissue sarcoma (STS) remains significantly uninvestigated.

Materials and Methods: Gene expression profiles were extracted from The Cancer Genome Atlas (TCGA) and Genotype-Tissue Expression (GTEx) for the identification of irlncRNAs. Univariate analysis and modified least absolute shrinkage and selection operator (LASSO) penalized regression analysis were employed to determine differently expressed irlncRNA (DEirlncRNA) pairs of prognostic value, and subsequently, a risk signature based on DEirlncRNA pairs was established. Furthermore, Kaplan–Meier analysis and the area under the receiver operating characteristic curve (AUC) were used to assess survival differences and the predictive accuracy of the risk signature, respectively. Lastly, the correlation of irlncRNAs with immune characteristics and chemosensitivity in patients with STS were further investigated.

Results: A total of 1088 irlncRNAs were identified, and 311 irlncRNAs were distinguished as DEirlncRNAs. A total of 130 DEirlncRNA pairs were further identified as prognostic markers, and 14 pairs were selected for establishing a risk signature. The irlncRNA-based risk signature functioned as an independent prognostic marker for STS. Compared with the patients in the high-risk group, those in the low-risk group exhibited a better prognosis and were more sensitive to several chemotherapeutic agents. In addition, the irlncRNA-based risk signature was significantly associated with immune scores, infiltrating immune cells, and the expression of several immune checkpoints.

Conclusion: In conclusion, our data revealed that the irlncRNA-based risk signature resulted in reliable prognosis, effectively predicted the immune landscape of patients with STS and was significantly correlated with chemosensitivity, thus providing insights into the potential role of irlncRNAs as prognostic biomarkers and novel therapeutic targets for STS.

A Novel CpG Methylation Risk Indicator for Predicting Prognosis in Bladder Cancer

Purpose

Bladder cancer (BLCA) is one of the most common cancers worldwide. In a large proportion of BLCA patients, disease recurs and/or progress after resection, which remains a major clinical issue in BLCA management. Therefore, it is vital to identify prognostic biomarkers for treatment stratification. We investigated the efficiency of CpG methylation for the potential to be a prognostic biomarker for patients with BLCA.

Patients and Methods

Overall, 357 BLCA patients from The Cancer Genome Atlas (TCGA) were randomly separated into the training and internal validation cohorts. Least absolute shrinkage and selector operation (LASSO) and support vector machine-recursive feature elimination (SVM-RFE) were used to select candidate CpGs and build the methylation risk score model, which was validated for its prognostic value in the validation cohort by Kaplan–Meier analysis. Hazard curves were generated to reveal the risk nodes throughout the follow-up. Gene Set Enrichment Analysis (GSEA) was used to reveal the potential biological pathways associated with the methylation model. Quantitative real-time polymerase chain reaction (PCR) and western blotting were performed to verify the expression level of the methylated genes.

Results

After incorporating the CpGs obtained by the two algorithms, CpG methylation of eight genes corresponding to TNFAIP8L3, KRTDAP, APC, ZC3H3, COL9A2, SLCO4A1, POU3F3, and ADARB2 were prominent candidate predictors in establishing a methylation risk score for BLCA (MRSB), which was used to divide the patients into high- and low-risk progression groups (p < 0.001). The effectiveness of the MRSB was validated in the internal cohort (p < 0.001). In the MRSB high-risk group, the hazard curve exhibited an initial wide, high peak within 10 months after treatment, whereas some gentle peaks around 2 years were noted. Furthermore, a nomogram comprising MRSB, age, sex, and tumor clinical stage was developed to predict the individual progression risk, and it performed well. Survival analysis implicated the effectiveness of MRSB, which remains significant in all the subgroup analysis based on the clinical features. A functional analysis of MRSB and the corresponding genes revealed potential pathways affecting tumor progression. Validation of quantitative real-time PCR and western blotting revealed that TNFAIP8L3 was upregulated in the BLCA tissues.

Conclusion

We developed the MRSB, an eight-gene-based methylation signature, which has great potential to be used to predict the post-surgery progression risk of BLCA.

Construction of an immune-related lncRNA signature as a novel prognosis biomarker for LUAD

The tumor immune microenvironment of lung cancer is associated with prognosis and immunotherapy efficacy. Long noncoding RNAs are identified as prognostic biomarkers associated with immune functions. We constructed a signature comprising differentially expressed immune-related lncRNAs to predict the prognosis of patients with lung adenocarcinoma. We established the immune-related lncRNA signature by pairing immune-related lncRNAs regardless of expression level and lung adenocarcinoma patients were divided into high- and low-risk groups. The prognosis of patients in the two groups was significantly different; The immune-related lncRNA signature could serve as an independent lung adenocarcinoma prognostic indicator. The signature correlated negatively with B cell, CD4+ T cell, M2 macrophage, neutrophil, and monocyte immune infiltration. Patients with low risk scores had a higher abundance of immune cells and stromal cells around the tumor. Gene set enrichment analysis showed that samples from low-risk group were more active in the IgA production in intestinal immune network and the T and B cell receptor signaling pathway. High-risk groups had significant involvement of the cell cycle, DNA replication, adherens junction, actin cytoskeleton regulation, pathways in cancer, and TGF-β signaling pathways. High risk scores correlated significantly negatively with high CTLA-4 and HAVCR2 expression and higher median inhibitory concentration of common anti-tumor chemotherapeutics (e.g., cisplatin, paclitaxel, gemcitabine) and targeted therapy (e.g., erlotinib and gefitinib). We identified a reliable immune-related lncRNA lung adenocarcinoma prognosis model, and the immune-related lncRNA signature showed promising clinical prediction value.

Predicting the immune landscape of invasive breast carcinoma based on the novel signature of immune-related lncRNA

Background

The composition of the population of immune‐related long non‐coding ribonucleic acid (irlncRNA) generates a signature, irrespective of expression level, with potential value in predicting the survival status of patients with invasive breast carcinoma.

Methods

The current study uses univariate analysis to identify differentially expressed irlncRNA (DEirlncRNA) pairs from RNA‐Seq data from The Cancer Genome Atlas (TCGA). 36 pairs of DEirlncRNA pairs were identified. Using various algorithms to construct a model, we have compared the area under the curve and calculated the 5‐year curve of Akaike information criterion (AIC) values, which allows determination of the threshold indicating the maximum value for differentiation. Through cut‐off point to establish the optimal model for distinguishing high‐risk or low‐risk groups among breast cancer patients. We assigned individual patients with invasive breast cancer to either high risk or low risk groups depending on the cut‐off point, re‐evaluated the tumor immune cell infiltration, the effectiveness of chemotherapy, immunosuppressive biomarkers, and immunotherapy.

Results

After re‐assessing patients according to the threshold, we demonstrated an effective means of distinguish the severity of the disease, and identified patients with different clinicopathological characteristics, specific tumor immune infiltration states, high sensitivity to chemotherapy,wellpredicted response to immunotherapy and thus a more favorable survival outcome.

Conclusions

The current study presents novel findings regarding the use of irlncRNA without the need to predict precise expression levels in the prognosis of breast cancer patients and to indicate their suitability for anti‐tumor immunotherapy.

Construction of an immune-related lncRNA pairs model to predict prognosis and immune landscape of lung adenocarcinoma patients

The model of immune-related lncRNA pairs (IRLPs) seems to be an available predictor in lung adenocarcinoma (LUAD) patients. The aim of our study was to construct a model with IRLPs to predict the survival status and immune landscape of LUAD patients. Based on TCGA-LUAD dataset, a risk assessment model with IRLPs was established. Then, ROC curves were used to assess the predictive accuracy and effectiveness of our model. Next, we identified the difference of survival, immune cell infiltration, immune checkpoint inhibitor-related (ICI-related) biomarkers, and chemotherapeutics between high-risk group and low-risk group. Finally, A nomogram was built for predicting the survival rates of LUAD patients. 464 LUAD samples were randomly and equally divided into a training set and a test set. Six IRLPs were screened out to construct a risk model. K-M analysis and risk-plot suggested the prognosis of high-risk group was worse than low-risk group (p < 0.001). Multivariate analysis shows risk score was independent risk factor of LUAD (p < 0.001). In addition, the expression of immune cell infiltration, ICI-related biomarkers, chemotherapeutics all demonstrate significant difference in two groups. A nomogram was built that could predict the 1-, 3-, and 5-year survival rates of LUAD patients. Our immune-related lncRNA pairs risk model is expected to be a reliable model for predicting the prognosis and immune landscape of LUAD patients.

A nine-lncRNA signature predicts distant relapse-free survival of HER2-negative breast cancer patients receiving taxane and anthracycline-based neoadjuvant chemotherapy

Multi-gene prognostic signatures of long non-coding RNAs (lncRNAs) provide new insights into mechanisms of HER2-negative breast cancer development and progression, and predict distant relapse-free survival (DRFS) of patients receiving taxane and anthracycline-based neoadjuvant chemotherapy. The aim of this study was to develop such a multi-lncRNAs signature. Optimal multiple candidate signature lncRNAs associated with DRFS were firstly identified by a univariate Cox proportional hazard regression survival analysis and a robust likelihood-based survival analysis of the GEO dataset GSE25055. A nine-lncRNA prognostic risk score model Risk Score = 0.0289 × EXPLOC100507388 - 0.0814 × EXPLINC00094 - 0.2422 × EXPSMG7-AS1 - 0.2433 × EXPPP14571 + 0.4690 × EXPASAP1-IT1 - 0.2483 × EXPLOC103344931 - 0.2464 × EXPFAM182A + 0.3349 × EXPHCG26 - 0.0216 × EXPLINC00963 was built according to the coefficients of multivariate survival analysis of the association between the candidate lncRNAs and survival. EXPlncRNA was the standardized log2-transformed expression level of the gene. According to this model, higher scores predicted lower survival probability. The area under Receiver operating characteristic (ROC) curve (AUC) was 0.777 to 0.823 from 1- to 7- year survival rate. The model and its individual lncRNAs differentiated survival probability between the higher scores (expression) and the lower scores (expression). The nine-lncRNA signature had the robust prognostic power compared with ER, PR, tumor size (T), lymph node invasion (N), TNM stage, pathologic response, chemosensitivity prediction and PAM50 signature. These results were consistent with those based on the GEO dataset GSE25065. The predictive nomograms integrating both the nine-lncRNA signature classifier and clinical-pathological risk factors were robust in predicting 1-, 3- and 5- year survival probabilities. These results supported that the nine-lncRNA signature was a robust and effective model in predicting DRFS of patients with HER2-negative breast cancer following taxane and anthracycline-based neoadjuvant chemotherapy.

3 CpG methylation biomarkers for the diagnosis of polycystic ovary syndrome (PCOS) in blood

BACKGROUND

Polycystic ovary syndrome (PCOS) is a common endocrine disease in women that seriously interferes with the patient's metabolic and reproductive functions. The current diagnostic criteria for PCOS are expert-based and still disputed. Previous studies have identified changes in DNA methylation in peripheral blood of women with PCOS, but their diagnostic potential for PCOS remains to be studied.

OBJECTIVE

The present study aimed to identify potential methylation biomarkers for the diagnosis of PCOS in blood.

METHODS

Methylation profiles of peripheral blood were downloaded from a public database, Gene Expression Omnibus (GEO), of 30 PCOS patients (diagnosed with the revised 2003 Rotterdam consensus criteria), and 30 age-matched healthy women were recruited from the Centre of Reproductive Medicine, Linyi People's Hospital, Shandong, China. Weighted gene co-expression network analysis (WGCNA) was utilized to identify PCOS-related co-methylation CpG sites (co-MPs). Functional enrichment analysis was performed on the localized genes of PCOS-related co-MPs. The least absolute shrinkage and selection operator (LASSO) regression was used to screen CpG methylation signatures for PCOS diagnosis and receiver operating characteristic (ROC) analysis was conducted to evaluate their diagnostic accuracy, respectively. To assess the accuracy of the combination of the investigated indicators, multivariate ROC analysis was performed on the predicted probability values obtained using binary logistic regression on the methylation levels of selected CpGs together.

RESULTS

Seven co-methylation modules were obtained, most relevant to PCOS of which was the turquoise module, containing 194 co-MPs. The genes that these co-MPs located in were mainly associated with the immune-related pathway. According to LASSO regression, three Co-MPs (cg23464743, cg06834912, cg00103771) were identified as potential diagnostic biomarkers of PCOS. ROC analysis showed an AUC (area under the curve) of 0.7556 (sensitivity 60.0%, specificity 83.3%) for cg23464743, 0.7822 (sensitivity 70.0%, specificity 80.0%) for cg06834912, and 0.7611 (sensitivity 63.3%, specificity 83.3%) for cg00103771. The diagnostic accuracy of the combination of these 3 indicators presented to be higher than any single one of them, with an AUC of 0.8378 (sensitivity 73.3%, specificity 93.3%).

CONCLUSION

The combination of 3 CpG methylation signatures in blood was identified with a good diagnostic accuracy for PCOS, which may bring new insight into the development of PCOS diagnostic markers in the future.

Signature based on metabolic-related gene pairs can predict overall survival of osteosarcoma patients

BACKGROUND

Osteosarcoma is a tumour of malignant origin in children and adolescents. Recent progression indicates that it is necessary to develop new therapies to improve the patient's prognosis rather than strengthen anti-tumour chemotherapy. Researchers recently realised that cancer is a kind of disease with a metabolic disorder, and metabolic reprogramming is becoming a new cancer hallmark. Hence, our study's primary purpose is to explore the value of genes related to osteosarcoma metabolism.

METHODS

From public databases, three osteosarcoma datasets with adequate clinical information were obtained. Besides, the IMvigor dataset through the 'IMvigor' package as a supplement was downloaded, the metabolic-related genes were identified, and these genes were used to construct the metabolic-related gene pairs (MRGP). Based on the prognosis-related MRGP, two molecular subtypes were identified. There are significant differences in the metabolic characteristics between the two molecular subtypes. Subsequently, the MRGP signature is constructed using the least absolute shrinkage and selection operator regression method. Finally, use SubMap analysis to evaluate the response of patients in the MRPG signature group to immunotherapy.

RESULTS

The MRGP signature can reliably predict overall survival in patients with osteosarcoma. The MRGP signature is also associated with osteosarcoma patients' metastatic status and can be used for subsequent risk classification of metastatic patients. The immunotherapy is more likely to benefit the patients in the MRGP low-risk group.

CONCLUSION

Metabolic-related gene pairs signature can assess the prognosis of patients with osteosarcoma.

An Immune-Related Long Non-Coding RNA Signature to Predict the Prognosis of Ewing's Sarcoma Based on a Machine Learning Iterative Lasso Regression

The aim of this study was to construct a new immune-associated long non-coding RNA (lncRNA) signature to predict the prognosis of Ewing sarcoma (ES) and explore its molecular mechanisms. We downloaded transcriptome and clinical prognosis data from the Gene Expression Omnibus (GSE17679, which included 88 ES samples and 18 matched normal skeletal muscle samples), and used it as a training set to identify immune-related lncRNAs with different expression levels in ES. Univariable Cox regression was used to screen immune-related lncRNAs related to ES prognosis, and an immune-related lncRNA signature was constructed based on machine learning iterative lasso regression. An external verification set was used to confirm the predictive ability of the signature. Clinical feature subgroup analysis was used to explore whether the signature was an independent prognostic factor. In addition, CIBERSORT was used to explore immune cell infiltration in the high- and low-risk groups, and to analyze the correlations between the lncRNA signature and immune cell levels. Gene set enrichment and variation analyses were used to explore the possible regulatory mechanisms of the immune-related lncRNAs in ES. We also analyzed the expression of 17 common immunotherapy targets in the high- and low-risk groups to identify any that may be regulated by immune-related lncRNAs. We screened 35 immune-related lncRNAs by univariate Cox regression. Based on this, an immune-related 11-lncRNA signature was generated by machine learning iterative lasso regression. Analysis of the external validation set confirmed its high predictive ability. DPP10 antisense RNA 3 was negatively correlated with resting dendritic cell, neutrophil, and γδ T cell infiltration, and long intergenic non-protein coding RNA 1398 was positively correlated with resting dendritic cells and M2 macrophages. These lncRNAs may affect ES prognosis by regulating GSE17721_CTRL_VS_PAM3CSK4_12H_BMDC_UP, GSE2770_IL4_ACT_VS_ACT_CD4_TCELL_48H_UP, GSE29615_CTRL_VS_DAY3_ LAIV_IFLU_VACCINE_PBMC_UP, complement signaling, interleukin 2-signal transducer and activator of transcription 5 signaling, and protein secretion. The immune-related 11-lncRNA signature may also have regulatory effects on the immunotherapy targets CD40 molecule, CD70 molecule, and CD276 molecule. In conclusion, we constructed a new immune-related 11-lncRNA signature that can stratify the prognoses of patients with ES.

Predicting the Risk of Acute Kidney Injury in Patients After Percutaneous Coronary Intervention (PCI) or Cardiopulmonary Bypass (CPB) Surgery: Development and Assessment of a Nomogram Prediction Model

Background

We sought to create a model that incorporated ultrasound examinations to predict the risk of acute kidney injury (AKI) after percutaneous coronary intervention (PCI) or cardiopulmonary bypass (CPB) surgery.

Material/Methods

A total of 292 patients with AKI after PCI or CPB surgery were enrolled for the study. Afterwards, treatment-related information, including data pertaining to ultrasound examination, was collected. A random forest model and multivariate logistic regression analysis were then used to establish a predictive model for the risk of AKI. Finally, the predictive quality and clinical utility of the model were assessed using calibration plots, receiver-operating characteristic curve, C-index, and decision curve analysis.

Results

Predictive factors were screened and the model was established with a C-index of 0.955 in the overall sample set. Additionally, an area under the curve of 0.967 was obtained in the training group. Moreover, decision curve analysis also revealed that the prediction model had good clinical applicability.

Conclusions

The prediction model was efficient in predicting the risk of AKI by incorporating ultrasound examinations and a number of factors. Such included operation methods, age, congestive heart failure, body mass index, heart rate, white blood cell count, platelet count, hemoglobin, uric acid, and peak intensity (kidney cortex as well as kidney medulla).

Tumor Immunometabolism Characterization in Ovarian Cancer With Prognostic and Therapeutic Implications

Metabolic dysregulation in the tumor microenvironment has significant impact on immune infiltration and immune responses. However, interaction between immunity and metabolism in the ovarian microenvironment requires further exploration. We constructed an immunometabolism gene set and ovarian cancer cohort from The Cancer Genome Atlas (TCGA) and classified these into three immunometabolism subtypes. We explored the relationships between immune infiltration and metabolic reprogramming. Additionally, we built risk score and nomogram as prognostic signatures. Three distinctive immunometabolism subtypes were identified with therapeutic and prognostic implications. Subtype 1, the "immune suppressive-glycan metabolism subtype," featured high levels of immunosuppressive cell infiltration and glycan metabolism activation; Subtype 2, the "immune inflamed-amino acid metabolism subtype," showed abundant adaptive immune cell infiltration and amino acid metabolism activation; Subtype 3, the "immune desert-endocrine subtype," was characterized by low immune cell infiltration and upregulation of hormone biosynthesis. Furthermore, we found that epinephrine biosynthesis displayed a significantly negative correlation with MHC molecules, which may result in defective antigen presentation. We proposed immunometabolism subtypes with prognostic implications and provided new perspectives for the ovarian cancer microenvironment.

Prognostic Cancer Gene Expression Signatures: Current Status and Challenges

Current staging systems of cancer are mainly based on the anatomical extent of disease. They need refinement by biological parameters to improve stratification of patients for tumor therapy or surveillance strategies. Thanks to developments in genomic, transcriptomic, and big-data technologies, we are now able to explore molecular characteristics of tumors in detail and determine their clinical relevance. This has led to numerous prognostic and predictive gene expression signatures that have the potential to establish a classification of tumor subgroups by biological determinants. However, only a few gene signatures have reached the stage of clinical implementation so far. In this review article, we summarize the current status, and present and future challenges of prognostic gene signatures in three relevant cancer entities: breast cancer, colorectal cancer, and hepatocellular carcinoma.

Downregulation of PAICS due to loss of chromosome 4q is associated with poor survival in stage III colorectal cancer

Phosphoribosylaminoimidazole carboxylase, phosphoribosylaminoimidazole succinocarboxamide synthetase (PAICS) encodes an enzyme that catalyzes de novo purine biosynthesis. Although PAICS has been implicated as a potential therapeutic target in several cancers, its clinical and prognostic significance in colorectal cancer (CRC) is not fully understood. To elucidate the roles of PAICS in CRC, we investigated PAICS expression in four cohorts consisting of a total of 1659 samples based on quantitative RT-PCR, microarray and RNA-seq analysis. Despite upregulated PAICS levels in tumor compared to those of normal mucosa, we found a decreasing trend of PAICS expression during tumor progression and metastasis. We conducted immunohistochemistry on 252 specimens, showing that PAICS protein was strongly expressed in the majority of CRCs, but not in adjacent mucosa. Notably, 29.0% of tumors lacked PAICS staining, and PAICS-negative expression in tumor had significant prognostic impact on poor cancer-specific survival in stage III CRC. Correspondingly, decreased levels of PAICS transcript were also correlated with poor relapse-free survival particularly in stage III patients, and this finding was robustly confirmed in three microarray datasets of a total of 802 stage II-III patients. Bioinformatics analysis of CRC tissues and cell lines consistently indicated a correlation between decreased PAICS expression and copy number loss of chromosome arm 4q. In conclusion, our results suggest that PAICS expression is downregulated during tumor progression due to genetic deletion of chromosome 4q in microsatellite stable but chromosomally unstable tumors. Furthermore, decreased expression of PAICS transcript or loss of PAICS protein may provide prognostic stratification for postoperative patients with stage III CRC.

The clinical relevance of gene expression based prognostic signatures in colorectal cancer

Colorectal cancer (CRC) is one of the most prevalent cancers, with more than one million new cases every year. In the last few decades, several advancements in therapeutic and preventative levels have reduced the mortality rate, but new biomarkers are required for improved prognosis. The alterations at the genetic and epigenetic level have been recognized as major players in tumorigenesis. The products of gene expression in the form of mRNA, microRNA, and long-noncoding RNA, have started to emerge as important regulatory molecules, playing an important role in cancer. Gene-expression based prognostic risk scores, which quantify and compare their expression, have emerged as promising biomarkers with enormous clinical value. These composite multi-gene models in which more than one gene is used to predict prognosis have been shown to be significantly effective in identifying patients with multiple clinico-pathological risks like overall mortality, response to chemotherapy, risk of metastasis, etc. The advent of microarray and advanced sequencing technologies have led to the generation of large datasets like TCGA (The Cancer Genome Atlas) and GEO (Gene Expression Omnibus), which have fueled the search for new biomarkers. Continuous evaluation of these candidate biomarkers in clinical settings is promising to improve the management of CRC. These composite gene signatures provide potential in identifying high-risk patients, which might help clinicians to better manage these patients and design appropriate personalized therapeutic interventions. In this review, we emphasize on composite prognostic scores from diverse resources with clinical utility in CRC.

PPP2R2B downregulation is associated with immune evasion and predicts poor clinical outcomes in triple-negative breast cancer

Background

Although immune checkpoint blockade has emerged as a novel promising strategy for triple-negative breast cancer (TNBC), many patients fail response or acquire resistance to current agents. Consequently, our focus need to shift toward alternative inhibitory targets, predictor for responsiveness, and immune suppressive mechanisms.

Methods

In this study, we performed systematic bioinformatics analyses to identify PPP2R2B as a robust tumor suppressor in TNBC. Meanwhile, breast cancer progression cell line model was applied in our research. Quantitative real-time PCR assay (Q-PCR) was carried out to assess the role of PPP2R2B in the onset and progression of breast cancer. Furthermore, we validated the effect of PPP2R2B on immune activity via in vitro experiments based on macrophages. To further decipher the roles of PPP2R2B in TNBC, we investigated the transcriptome level, genomic profiles, and its clinical prognostic value.

Results

In TNBC tissues, PPP2R2B expression was significantly downregulated compared to normal breast tissues. Kaplan‐Meier survival analysis revealed that patients with low PPP2R2B expression had shorter survival time than those with high PPP2R2B expression. Q-PCR analysis suggested that PPP2R2B downregulation could play a key role in breast-cancer initiation and progression. Additionally, our findings showed that PPP2R2B was positively related with CD8 T cells, CD4 Th1 helper cells, and M1 macrophages, but negatively related with M2 macrophages. Subsequent results identified that PPP2R2B was strongly related with immune inhibitor genes (GZMA, PRF1, and IFNG), which could improve T lymphocytes antitumor function and restrict immune evasion. Meanwhile, T cell receptor signaling pathway and antigen processing and presentation signaling pathway were significantly suppressed in low PPP2R2B expression group. Afterwards, distinct subgroups based on PPP2R2B expression exhibited several unique features in somatic mutations, copy numbers alterations, extent of copy number burden, and promoter methylation level.

Conclusion

Our results indicated that PPP2R2B could serve as a promising biomarker for TNBC, and help predict immunotherapeutic response and guide personalized strategies in TNBC treatment.

Immune-Related lncRNA to Construct Novel Signature and Predict the Immune Landscape of Human Hepatocellular Carcinoma

The signature composed of immune-related long noncoding ribonucleic acids (irlncRNAs) with no requirement of specific expression level seems to be valuable in predicting the survival of patients with hepatocellular carcinoma (HCC). Here, we retrieved raw transcriptome data from The Cancer Genome Atlas (TCGA), identified irlncRNAs by co-expression analysis, and recognized differently expressed irlncRNA (DEirlncRNA) pairs using univariate analysis. In addition, we modified Lasso penalized regression. Then, we compared the areas under curve, counted the Akaike information criterion (AIC) values of 5-year receiver operating characteristic curve, and identified the cut-off point to set up an optimal model for distinguishing the high- or low-disease-risk groups among patients with HCC. We then reevaluated them from the viewpoints of survival, clinic-pathological characteristics, tumor-infiltrating immune cells, chemotherapeutics efficacy, and immunosuppressed biomarkers. 36 DEirlncRNA pairs were identified, 12 of which were included in a Cox regression model. After regrouping the patients by the cut-off point, we could more effectively differentiate between them based on unfavorable survival outcome, aggressive clinic-pathological characteristics, specific tumor immune infiltration status, low chemotherapeutics sensitivity, and highly expressed immunosuppressed biomarkers. The signature established by paring irlncRNA regardless of expression levels showed a promising clinical prediction value.

An immune-related gene signature for determining Ewing sarcoma prognosis based on machine learning

PURPOSE

Ewing sarcoma (ES) is one of the most common malignant bone tumors in children and adolescents. The immune microenvironment plays an important role in the development of ES. Here, we developed an optimal signature for determining ES patient prognosis based on immune-related genes (IRGs).

METHODS

We analyzed the ES gene expression profile dataset, GSE17679, from the GEO database and extracted differential expressed IRGs (DEIRGs). Then, we conducted functional correlation and protein-protein interaction (PPI) analyses of the DEIRGs and used the machine learning algorithm-iterative Lasso Cox regression analysis to build an optimal DEIRG signature. In addition, we applied ES samples from the ICGC database to test the optimal gene signature. We performed univariate and multivariate Cox regressions on clinicopathological characteristics and optimal gene signature to evaluate whether signature is an important prognostic factor. Finally, we calculated the infiltration of 24 immune cells in ES using the ssGSEA algorithm, and analyzed the correlation between the DEIRGs in the optimal gene signature and immune cells.

RESULTS

A total of 249 DEIRGs were screened and an 11-gene signature with the strongest correlation with patient prognoses was analyzed using a machine learning algorithm. The 11-gene signature also had a high prognostic value in the ES external verification set. Univariate and multivariate Cox regression analyses showed that 11-gene signature is an independent prognostic factor. We found that macrophages and cytotoxic, CD8 T, NK, mast, B, NK CD56bright, TEM, TCM, and Th2 cells were significantly related to patient prognoses; the infiltration of cytotoxic and CD8 T cells in ES was significantly different. By correlating prognostic biomarkers with immune cell infiltration, we found that FABP4 and macrophages, and NDRG1 and Th2 cells had the strongest correlation.

CONCLUSION

Overall, the IRG-related 11-gene signature can be used as a reliable ES prognostic biomarker and can provide guidance for personalized ES therapy.

Worse prognosis in young patients with locally advanced rectal cancer following neoadjuvant chemoradiotherapy: A comparative study

To determine the efficacy of neoadjuvant chemoradiotherapy (NCRT) between young and old patients with locally advanced rectal cancer (LARC) in terms of tumor response and survival outcome.LARC patients undergoing NCRT and radical surgery from 2011 to 2015 were included and divided into: young (aged ≤50 years) and old group (aged >50 years). Multivariate analyses were performed to identify risk factors for local recurrence. Least absolute shrinkage and selection operator analysis was performed to identify risk factors for overall survival. Predicting nomograms and time-indepent receiver operating characteristic curve analysis were performed to compare the models containing with/withour age groups.A total of 572 LARC patients were analyzed. The young group was associated with higher pathological TNM stage, poorly differentiated tumors, and higher rate of positive distal resection margin (P = .010; P = .019; P = .023 respectively). Young patients were associated with poorer 5-year disease-free survival and local recurrence rates (P = .023, P = .003 respectively). Cox regression analysis demonstrated that age ≤50 years (Hazard ratio = 2.994, P = .038) and higher pathological TNM stage (Hazard ratio = 3.261, P = .005) were significantly associated with increased risk for local recurrence. Least absolute shrinkage and selection operator analysis and the time-indepent receiver operating characteristic curve analysis demonstrated that including the age group were superior than that without age group.Young patients were associated with poorer disease free survival (DFS) and a higher risk for local recurrence in LARC following NCRT. The predicting model basing based on the age group had a better predictive ability. More intense adjuvant treatment could be considered to improve DFS and local control for young patients with LARC following NCRT.

Gene dysregulation analysis builds a mechanistic signature for prognosis and therapeutic benefit in colorectal cancer

The implementation of cancer precision medicine requires biomarkers or signatures for predicting prognosis and therapeutic benefits. Most of current efforts in this field are paying much more attention to predictive accuracy than to molecular mechanistic interpretability. Mechanism-driven strategy has recently emerged, aiming to build signatures with both predictive power and explanatory power. Driven by this strategy, we developed a robust gene dysregulation analysis framework with machine learning algorithms, which is capable of exploring gene dysregulations underlying carcinogenesis from high-dimensional data with cooperativity and synergy between regulators and several other transcriptional regulation rules taken into consideration. We then applied the framework to a colorectal cancer (CRC) cohort from The Cancer Genome Atlas. The identified CRC-related dysregulations significantly covered known carcinogenic processes and exhibited good prognostic effect. By choosing dysregulations with greedy strategy, we built a four-dysregulation (4-DysReg) signature, which has the capability of predicting prognosis and adjuvant chemotherapy benefit. 4-DysReg has the potential to explain carcinogenesis in terms of dysfunctional transcriptional regulation. These results demonstrate that our gene dysregulation analysis framework could be used to develop predictive signature with mechanistic interpretability for cancer precision medicine, and furthermore, elucidate the mechanisms of carcinogenesis.

A 16-mRNA signature optimizes recurrence-free survival prediction of Stages II and III gastric cancer

High-throughput messenger RNA (mRNA) analysis has become a powerful tool for exploring tumor recurrence or metastasis mechanisms. Here, we constructed a signature to predict the recurrence risk of Stages II and III gastric cancer (GC) patients. A least absolute shrinkage and selection operator method Cox regression model was utilized to construct the signature. Using this method, a 16-mRNA signature was identified to be associated with the relapse-free survival of Stages II and III GCs in training dataset GSE62254 (n = 194). Then this signature was validated in an independent Gene Expression Omnibus cohort GSE26253 (n = 297) and a dataset of The Cancer Genome Atlas (TCGA; n = 235). This classifier could successfully screen out the high-risk Stages II and III GCs in the training cohort (hazard ratio [HR] = 40.91; 95% confidence interval [CI] = 5.58-299.7; p < .0001). Analysis in two independent validation cohorts yielded consistent results (GSE26253: HR = 1.69, 95% CI = 1.17-2.43,; p = .0045; TCGA: HR = 2.01, 95% CI = 1.13-3.56, p = .0146). Cox regression analyses revealed that the risk score derived from this signature was an independent risk factor in Stages II and III GCs. Besides, a nomogram was constructed to serve clinical practice. Through gene set variation analysis, we found several gene sets associated with chemotherapeutic drug resistance and tumor metastasis significantly enriched in high-risk patients. In summary, this 16-mRNA signature can be used as a powerful tool for prognostic evaluation and help clinicians identify high-risk patients.

Network Properties of Cancer Prognostic Gene Signatures in the Human Protein Interactome

Prognostic gene signatures are critical in cancer prognosis assessments and their pinpoint treatments. However, their network properties remain unclear. Here, we obtained nine prognostic gene sets including 1439 prognostic genes of different cancers from related publications. Four network centralities were used to examine the network properties of prognostic genes (PG) compared with other gene sets based on the Human Protein Reference Database (HPRD) and String networks. We also proposed three novel network measures for further investigating the network properties of prognostic gene sets (PGS) besides clustering coefficient. The results showed that PG did not occupy key positions in the human protein interaction network and were more similar to essential genes rather than cancer genes. However, PGS had significantly smaller intra-set distance (IAD) and inter-set distance (IED) in comparison with random sets (p-value < 0.001). Moreover, we also found that PGS tended to be distributed within network modules rather than between modules (p-value < 0.01), and the functional intersection of the modules enriched with PGS was closely related to cancer development and progression. Our research reveals the common network properties of cancer prognostic gene signatures in the human protein interactome. We argue that these are biologically meaningful and useful for understanding their molecular mechanism.

Prognostic and Predictive Molecular Biomarkers for Colorectal Cancer: Updates and Challenges

Colorectal cancer (CRC) is a leading cause of death among cancer patients. This heterogeneous disease is characterized by alterations in multiple molecular pathways throughout its development. Mutations in RAS, along with the mismatch repair gene deficiency, are currently routinely tested in clinics. Such biomarkers provide information for patient risk stratification and for the choice of the best treatment options. Nevertheless, reliable and powerful prognostic markers that can identify “high-risk” CRC patients, who might benefit from adjuvant chemotherapy, in early stages, are currently missing. To bridge this gap, genomic information has increasingly gained interest as a potential method for determining the risk of recurrence. However, due to several limitations of gene-based signatures, these have not yet been clinically implemented. In this review, we describe the different molecular markers in clinical use for CRC, highlight new markers that might become indispensable over the next years, discuss recently developed gene expression-based tests and highlight the challenges in biomarker research.

Technical differences between sequencing and microarray platforms impact transcriptomic subtyping of colorectal cancer

Gene expression profiling has increasing relevance in the molecular screening of patients with colorectal cancer (CRC). We investigated potential platform-specific effects on transcriptomic subtyping according to established frameworks by comparisons of expression profiles from RNA sequencing and exon-resolution microarrays in 126 primary microsatellite stable CRCs. There was a strong platform correspondence in global gene expression levels, albeit with systematic technical bias likely attributed to few sequencing reads covering short (<2000 nucleotides) and/or lowly expressed genes (<1 FPKM), as well as over-saturation of highly expressed genes on microarrays. Classification concordances according to both the consensus molecular subtypes and CRC intrinsic subtypes (CRIS) were also strong, but with disproportionate subtype distributions between platforms caused by frequent disagreements in adherence to sample classification thresholds. Subtypes defined largely by genes expressed at low levels, including the CRIS-D subtype and the estimated level of tumor-infiltrating cytotoxic lymphocytes, had a weaker correspondence in classification metrics between platforms. In conclusion, even subtle differences between platforms suggest that clinical translation of transcriptomic CRC subtyping frameworks is dependent on assay standardization, and systematic technical biases reinforce the need for careful selection of classifier genes.