survcomp: an R/Bioconductor package for performance assessment and comparison of survival models

Changes in the tumor immune microenvironment during disease progression in clear cell ovarian cancer

OBJECTIVE

The tumor immune microenvironment in ovarian clear cell carcinoma has not been clearly defined. We analyzed the immunological changes from treatment-naive to recurrence to correlate them with clinical outcomes.

METHOD

We compared the changes in immune infiltration of advanced-stage ovarian clear cell carcinoma samples before treatment and at the time of recurrence via immunohistochemistry (Programmed Cell Death-ligand 1 (PD-L1), cluster of differentiation 8 (CD8+), forkhead box P3 (Foxp3+)), tumor-infiltrating lymphocytes (TIL), and next-generation sequencing (54 patients). We analyzed the association between platinum sensitivity status and tumor immune microenvironment.

RESULTS

Immunohistochemistry revealed significantly increased PD-L1 (p=0.048) and CD8+T cells (p=0.022) expression levels after recurrence. No significant differences were observed in TIL density or Foxp3+T cells. There was no significant correlation between TIL, PD-L1, CD8+T cell, and Foxp3+T cell levels in treatment-naive tumors and survival outcomes. The most common genomic alterations were PIK3CA (41.7%) and ARID1A (41.7%) mutations. There were no differences in the immunological changes or survival outcomes according to PIK3CA and ARID1A mutations. Patients with recurrent platinum-sensitive disease showed higher TIL expression levels. There were no significant differences in PD-L1, CD8+T cells, or Foxp3+T cells between platinum-sensitive and platinum-resistant diseases.

CONCLUSION

We characterized the tumor immune microenvironment in patients with advanced-stage ovarian clear cell carcinoma. PD-L1 and CD8+T cell expression significantly increased after recurrence. Whether this could be used to select patients for immunotherapy in the recurrence setting should be investigated.

NMR metabolomics-guided DNA methylation mortality predictors

BACKGROUND

1H-NMR metabolomics and DNA methylation in blood are widely known biomarkers predicting age-related physiological decline and mortality yet exert mutually independent mortality and frailty signals.

METHODS

Leveraging multi-omics data in four Dutch population studies (N = 5238, ∼40% of which male) we investigated whether the mortality signal captured by 1H-NMR metabolomics could guide the construction of DNA methylation-based mortality predictors.

FINDINGS

We trained DNA methylation-based surrogates for 64 metabolomic analytes and found that analytes marking inflammation, fluid balance, or HDL/VLDL metabolism could be accurately reconstructed using DNA-methylation assays. Interestingly, a previously reported multi-analyte score indicating mortality risk (MetaboHealth) could also be accurately reconstructed. Sixteen of our derived surrogates, including the MetaboHealth surrogate, showed significant associations with mortality, independent of relevant covariates.

INTERPRETATION

The addition of our metabolic analyte-derived surrogates to the well-established epigenetic clock GrimAge demonstrates that our surrogates potentially represent valuable mortality signal.

FUNDING

BBMRI-NL, X-omics, VOILA, Medical Delta, NWO, ERC.

A Fast Survival Support Vector Regression Approach to Large Scale Credit Scoring via Safe Screening

Survival models have found wider and wider applications in credit scoring recently due to their ability to estimate the dynamics of risk over time. In this research, we propose a Buckley-James safe sample screening support vector regression (BJS4VR) algorithm to model large-scale survival data by combing the Buckley-James transformation and support vector regression. Different from previous support vector regression survival models, censored samples here are imputed using a censoring unbiased Buckley-James estimator. Safe sample screening is then applied to discard samples that guaranteed to be non-active at the final optimal solution from the original data to improve efficiency. Experimental results on the large-scale real lending club loan data have shown that the proposed BJS4VR model outperforms existing popular survival models such as RSFM, CoxRidge and CoxBoost in terms of both prediction accuracy and time efficiency. Important variables highly correlated with credit risk are also identified with the proposed method.

Multinational evaluation of anthropometric age (AnthropoAge) as a measure of biological age in the USA, England, Mexico, Costa Rica, and China: a population-based longitudinal study

OBJECTIVE

To validate AnthropoAge, a new metric of biological age (BA), for prediction of all-cause mortality and age-related outcomes and characterize population-specific aging patterns using multinational longitudinal cohorts.

METHODS

We analyzed harmonized multinational data from the Gateway to Global Aging, including studies from the US, England, Mexico, Costa Rica, and China. We used body mass index and waist-to-height ratio to estimate AnthropoAge and AnthropoAgeAccel in participants aged 50-90 years old as proxies of BA and age acceleration, respectively. We compared the predictive capacity for all-cause mortality of AnthropoAge and chronological age (CA) using Cox models, described aging trends in all countries and explored the utility of longitudinal assessments of AnthropoAgeAccel to predict new-onset functional decline and age-related diseases using generalized estimating equations (GEE).

FINDINGS

Using data from 55,628 participants, we found AnthropoAge (c-statistic 0.772) outperformed CA (0.76) for prediction of mortality independently of comorbidities, sex, race/ethnicity, education, and lifestyle; this result was replicated in most countries individually except for Mexico. Individuals with accelerated aging had a ~39% higher risk of death, and AnthropoAge also identified trends of faster biological aging per year. In longitudinal analyses, higher AnthropoAgeAccel values were independently predictive of self-reported health deterioration and new-onset deficits in basic/instrumental activities of daily living (ADL/IADL), diabetes, hypertension, cancer, chronic lung disease, myocardial infarction, and stroke.

CONCLUSIONS

AnthropoAge is a robust and reproducible BA metric associated with age-related outcomes. Its implementation could facilitate modeling trends of biological aging acceleration in different populations, although recalibration may enhance its utility in underrepresented populations such as individuals from Latin America.

Plasma pTau181 enhances the prediction of future clinical decline in amyloid-positive mild cognitive impairment

Abstract

Plasma pTau181, a marker of amyloid and tau burden, was evaluated as a prognostic predictor of clinical decline and Alzheimer's disease (AD) progression of amyloid-positive (Aβ+) patients with mild cognitive impairment (MCI). The training cohort for constructing the Bayesian prediction models comprised 135 Aβ+ MCI clinical trial placebo subjects. Performance was evaluated in two validation cohorts. An 18-month ≥1 increase in the Clinical Dementia Rating Sum of Boxes was the clinical decline criterion. Baseline plasma pTau181 concentration matched clinical assessments' prediction performance. Adding pTau181 to clinical assessments significantly improved the prediction of an 18-month clinical decline and the 36-month progression from Aβ+ MCI to AD. The area under the receiver operating characteristic curve for the latter increased from 71.8% to 79%, and the hazard ratio for time-to-progression improved from 2.26 to 3.11 (p < 0.0001). Baseline plasma pTau181 has the potential for identifying Aβ+ MCI subjects with faster clinical decline over time.

Highlights

This study assessed pTau181 as a prognostic predictor of 18-month clinical decline and extended progression to Alzheimer's disease (AD) in amyloid-positive patients with mild cognitive impairment (Aβ+ MCI).The research findings underscore the promise of baseline plasma pTau181 as a screening tool for identifying Aβ+ MCI individuals with accelerated clinical decline within a standard 18-month clinical trial period. The predictive accuracy is notably enhanced when combined with clinical assessments.Similar positive outcomes were noted in forecasting the extended progression of Aβ+ MCI subjects to AD.

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

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

Magnetic resonance imaging-based prognostic model for subsequent distant metastasis in patients with ipsilateral breast tumor recurrence following breast-conserving surgery

Background

Ipsilateral breast tumor recurrence (IBTR) following breast-conserving surgery (BCS) has been considered a risk factor for distant metastasis (DM). Limited data are available regarding the subsequent outcomes after IBTR. Therefore, this study aimed to determine the clinical course after IBTR and develop a magnetic resonance imaging (MRI)-based predictive model for subsequent DM.

Methods

We retrospectively extracted quantitative features from MRI to construct a radiomics cohort, with all eligible patients undergoing preoperative MRI at time of primary tumor and IBTR between 2010 and 2018. Multivariate Cox analysis was performed to identify factors associated with DM. Three models were constructed using different sets of clinicopathological, qualitative, and quantitative MRI features and compared. Additionally, Kaplan-Meier analysis was performed to assess the prognostic value of the optimal model.

Results

Among the 183 patients who experienced IBTR, 47 who underwent MRI for both primary and recurrent tumors were enrolled. Multivariate analysis demonstrated that the independent prognostic factors were human epidermal growth factor receptor 2 (HER2) status [hazard ratio (HR) =5.40] and background parenchymal enhancement (BPE) (HR =7.94) (all P values <0.01). Furthermore, four quantitative MRI features of recurrent tumors were selected through the least absolute shrinkage and selection operator (LASSO) method. The combined model exhibited superior performance [concordance index (C-index) 0.77] compared to the clinicoradiological model (C-index 0.71; P=0.006) and radiomics model (C-index 0.70; and P=0.01). Furthermore, the combined model successfully categorized patients into low- and high-risk subgroups with distinct prognoses (P<0.001).

Conclusions

The clinicopathological and MRI features of IBTR were associated with secondary events following surgery. Additionally, the MRI-based combined model exhibited the highest predictive efficacy. These findings could be helpful in risk stratification and tailoring follow-up strategies in patients with IBTR.

Comprehensive meta-analysis reveals distinct gene expression signatures of MASLD progression

Metabolic dysfunction-associated steatotic liver disease (MASLD) and its progressive form, metabolic dysfunction-associated steatohepatitis (MASH), pose significant risks of severe fibrosis, cirrhosis, and hepatocellular carcinoma. Despite their widespread prevalence, the molecular mechanisms underlying the development and progression of these common chronic hepatic conditions are not fully understood. Here, we conducted the most extensive meta-analysis of hepatic gene expression datasets from liver biopsy samples to date, integrating 10 RNA-sequencing and microarray datasets (1,058 samples). Using a random-effects meta-analysis model, we compared over 12,000 shared genes across datasets. We identified 685 genes differentially expressed in MASLD versus normal liver, 1,870 in MASH versus normal liver, and 3,284 in MASLD versus MASH. Integrating these results with genome-wide association studies and coexpression networks, we identified two functionally relevant, validated coexpression modules mainly driven by SMOC2, ITGBL1, LOXL1, MGP, SOD3, and TAT, HGD, SLC25A15, respectively, the latter not previously associated with MASLD and MASH. Our findings provide a comprehensive and robust analysis of hepatic gene expression alterations associated with MASLD and MASH and identify novel key drivers of MASLD progression.

Integrative bioinformatics approach yields a novel gene expression risk model for prognosis and progression prediction in prostate cancer

Prostate cancer (PCa), a prevalent malignancy among elderly males, exhibits a notable rate of advancement, even when subjected to conventional androgen deprivation therapy or chemotherapy. An effective progression prediction model would prove invaluable in identifying patients with a higher progression risk. Using bioinformatics strategies, we integrated diverse data sets of PCa to construct a novel risk model predicated on gene expression and progression-free survival (PFS). The accuracy of the model was assessed through validation using an independent data set. Eight genes were discerned as independent prognostic factors and included in the prediction model. Patients assigned to the high-risk cohort demonstrated a diminished PFS, and the areas under the curve of our model in the validation set for 1-year, 3-year, and 5-year PFS were 0.9325, 0.9041 and 0.9070, respectively. Additionally, through the application of single-cell RNA sequencing to two castration-related prostate cancer (CRPC) samples and two hormone-related prostate cancer (HSPC) samples, we discovered that luminal cells within CRPC exhibited an elevated risk score. Subsequent molecular biology experiments corroborated our findings, illustrating heightened SYK expression levels within tumour tissues and its contribution to cancer cell migration. We found that the knockdown of SYK could inhibit migration in PCa cells. Our progression-related risk model demonstrated the potential prognostic value of SYK and indicated its potential as a target for future diagnosis and treatment strategies in PCa management.

Prognositic value of anoikis and tumor immune microenvironment-related gene in the treatment of osteosarcoma

OBJECTIVES

Osteosarcoma is a highly aggressive primary malignant bone tumor commonly seen in children and adolescents, with a poor prognosis. Anchorage-dependent cell death (anoikis) has been proven to be indispensable in tumor metastasis, regulating the migration and adhesion of tumor cells at the primary site. However, as a type of programmed cell death, anoikis is rarely studied in osteosarcoma, especially in the tumor immune microenvironment. This study aims to clarify prognostic value of anoikis and tumor immune microenvironment-related gene in the treatment of osteosarcoma.

METHODS

Anoikis-related genes (ANRGs) were obtained from GeneCards. Clinical information and ANRGs expression profiles of osteosarcoma patients were sourced from the therapeutically applicable research to generate effective therapies and Gene Expression Omnibus (GEO) databases. ANRGs highly associated with tumor immune microenvironment were identified by the estimate package and the weighted gene coexpression network analysis (WGCNA) algorithm. Machine learning algorithms were performed to construct long-term survival predictive strategy, each sample was divided into high-risk and low-risk subgroups, which was further verified in the GEO cohort. Finally, based on single-cell RNA-seq from the GEO database, analysis was done on the function of signature genes in the osteosarcoma tumor microenvironment.

RESULTS

A total of 51 hub ANRGs closely associated with the tumor microenvironment were identified, from which 3 genes (MERTK, BNIP3, S100A8) were selected to construct the prognostic model. Significant differences in immune cell activation and immune-related signaling pathways were observed between the high-risk and low-risk groups based on tumor microenvironment analysis (all P<0.05). Additionally, characteristic genes within the osteosarcoma microenvironment were identified in regulation of intercellular crosstalk through the GAS6-MERTK signaling pathway.

CONCLUSIONS

The prognostic model based on ANRGs and tumor microenvironment demonstrate good predictive power and provide more personalized treatment options for patients with osteosarcoma.

Metabolic pathway-based subtypes associate glycan biosynthesis and treatment response in head and neck cancer

Head and Neck Squamous Cell Carcinoma (HNSCC) is a heterogeneous malignancy that remains a significant challenge in clinical management due to frequent treatment failures and pronounced therapy resistance. While metabolic dysregulation appears to be a critical factor in this scenario, comprehensive analyses of the metabolic HNSCC landscape and its impact on clinical outcomes are lacking. This study utilized transcriptomic data from four independent clinical cohorts to investigate metabolic heterogeneity in HNSCC and define metabolic pathway-based subtypes (MPS). In HPV-negative HNSCCs, MPS1 and MPS2 were identified, while MPS3 was enriched in HPV-positive cases. MPS classification was associated with clinical outcome post adjuvant radio(chemo)therapy, with MPS1 consistently exhibiting the highest risk of therapeutic failure. MPS1 was uniquely characterized by upregulation of glycan (particularly chondroitin/dermatan sulfate) metabolism genes. Immunohistochemistry and pilot mass spectrometry imaging analyses confirmed this at metabolite level. The histological context and single-cell RNA sequencing data identified the malignant cells as key contributors. Globally, MPS1 was distinguished by a unique transcriptomic landscape associated with increased disease aggressiveness, featuring motifs related to epithelial-mesenchymal transition, immune signaling, cancer stemness, tumor microenvironment assembly, and oncogenic signaling. This translated into a distinct histological appearance marked by extensive extracellular matrix remodeling, abundant spindle-shaped cancer-associated fibroblasts, and intimately intertwined populations of malignant and stromal cells. Proof-of-concept data from orthotopic xenotransplants replicated the MPS phenotypes on the histological and transcriptome levels. In summary, this study introduces a metabolic pathway-based classification of HNSCC, pinpointing glycan metabolism-enriched MPS1 as the most challenging subgroup that necessitates alternative therapeutic strategies.

System biology approach to identify the novel biomarkers in glioblastoma multiforme tumors by using computational analysis

Introduction: The most common primary brain tumor in adults is glioblastoma multiforme (GBM), accounting for 45.2% of all cases. The characteristics of GBM, a highly aggressive brain tumor, include rapid cell division and a propensity for necrosis. Regretfully, the prognosis is extremely poor, with only 5.5% of patients surviving after diagnosis. Methodology: To eradicate these kinds of complicated diseases, significant focus is placed on developing more effective drugs and pinpointing precise pharmacological targets. Finding appropriate biomarkers for drug discovery entails considering a variety of factors, including illness states, gene expression levels, and interactions between proteins. Using statistical techniques like p-values and false discovery rates, we identified differentially expressed genes (DEGs) as the first step in our research for identifying promising biomarkers in GBM. Of the 132 genes, 13 showed upregulation, and only 29 showed unique downregulation. No statistically significant changes in the expression of the remaining genes were observed. Results: Matrix metallopeptidase 9 (MMP9) had the greatest degree in the hub biomarker gene identification, followed by (periostin (POSTN) at 11 and Hes family BHLH transcription factor 5 (HES5) at 9. The significance of the identification of each hub biomarker gene in the initiation and advancement of glioblastoma multiforme was brought to light by the survival analysis. Many of these genes participate in signaling networks and function in extracellular areas, as demonstrated by the enrichment analysis.We also identified the transcription factors and kinases that control proteins in the proteinprotein interactions (PPIs) of the DEGs. Discussion: We discovered drugs connected to every hub biomarker. It is an appealing therapeutic target for inhibiting MMP9 involved in GBM. Molecular docking investigations indicated that the chosen complexes (carmustine, lomustine, marimastat, and temozolomide) had high binding affinities of -6.3, -7.4, -7.7, and -8.7 kcal/mol, respectively, the mean root-mean-square deviation (RMSD) value for the carmustine complex and marimastat complex was 4.2 Å and 4.9 Å, respectively, and the lomustine and temozolomide complex system showed an average RMSD of 1.2 Å and 1.6 Å, respectively. Additionally, high stability in root-mean-square fluctuation (RMSF) analysis was observed with no structural conformational changes among the atomic molecules. Thus, these in silico investigations develop a new way for experimentalists to target lethal diseases in future.

A multi-parametric prognostic model based on clinicopathologic features: vessels encapsulating tumor clusters and hepatic plates predict overall survival in hepatocellular carcinoma patients

BACKGROUND

Vessels encapsulating tumor clusters (VETC) is a newly described vascular pattern that is distinct from microvascular invasion (MVI) in patients with hepatocellular carcinoma (HCC). Despite its importance, the current pathological diagnosis report does not include information on VETC and hepatic plates (HP). We aimed to evaluate the prognostic value of integrating VETC and HP (VETC-HP model) in the assessment of HCC.

METHODS

A total of 1255 HCC patients who underwent radical surgery were classified into training (879 patients) and validation (376 patients) cohorts. Additionally, 37 patients treated with lenvatinib were studied, included 31 patients in high-risk group and 6 patients in low-risk group. Least absolute shrinkage and selection operator (LASSO) regression analysis was used to establish a prognostic model for the training set. Harrell's concordance index (C-index), time-dependent receiver operating characteristics curve (tdROC), and decision curve analysis were utilized to evaluate our model's performance by comparing it to traditional tumor node metastasis (TNM) staging for individualized prognosis.

RESULTS

A prognostic model, VETC-HP model, based on risk scores for overall survival (OS) was established. The VETC-HP model demonstrated robust performance, with area under the curve (AUC) values of 0.832 and 0.780 for predicting 3- and 5-year OS in the training cohort, and 0.805 and 0.750 in the validation cohort, respectively. The model showed superior prediction accuracy and discrimination power compared to TNM staging, with C-index values of 0.753 and 0.672 for OS and disease-free survival (DFS) in the training cohort, and 0.728 and 0.615 in the validation cohort, respectively, compared to 0.626 and 0.573 for TNM staging in the training cohort, and 0.629 and 0.511 in the validation cohort. Thus, VETC-HP model had higher C-index than TNM stage system(p < 0.01).Furthermore, in the high-risk group, lenvatinib alone appeared to offer less clinical benefit but better disease-free survival time.

CONCLUSIONS

The VETC-HP model enhances DFS and OS prediction in HCC compared to traditional TNM staging systems. This model enables personalized temporal survival estimation, potentially improving clinical decision-making in surveillance management and treatment strategies.

Intertumoral Heterogeneity Based on MRI Radiomic Features Estimates Recurrence in Hepatocellular Carcinoma

BACKGROUND

Hepatocellular carcinoma (HCC) heterogeneity impacts prognosis, and imaging is a potential indicator.

PURPOSE

To characterize HCC image subtypes in MRI and correlate subtypes with recurrence.

STUDY TYPE

Retrospective.

POPULATION

A total of 440 patients (training cohort = 213, internal test cohort = 140, external test cohort = 87) from three centers.

FIELD STRENGTH/SEQUENCE

1.5-T/3.0-T, fast/turbo spin-echo T2-weighted, spin-echo echo-planar diffusion-weighted, contrast-enhanced three-dimensional gradient-recalled-echo T1-weighted with extracellular agents (Gd-DTPA, Gd-DTPA-BMA, and Gd-BOPTA).

ASSESSMENT

Three-dimensional volume-of-interest of HCC was contoured on portal venous phase, then coregistered with precontrast and late arterial phases. Subtypes were identified using non-negative matrix factorization by analyzing radiomics features from volume-of-interests, and correlated with recurrence. Clinical (demographic and laboratory data), pathological, and radiologic features were compared across subtypes. Among clinical, radiologic features and subtypes, variables with variance inflation factor above 10 were excluded. Variables (P < 0.10) in univariate Cox regression were included in stepwise multivariate analysis. Three recurrence estimation models were built: clinical-radiologic model, subtype model, hybrid model integrating clinical-radiologic characteristics, and subtypes.

STATISTICAL TESTS

Mann-Whitney U test, Kruskal-Wallis H test, chi-square test, Fisher's exact test, Kaplan-Meier curves, log-rank test, concordance index (C-index). Significance level: P < 0.05.

RESULTS

Two subtypes were identified across three cohorts (subtype 1:subtype 2 of 86:127, 60:80, and 36:51, respectively). Subtype 1 showed higher microvascular invasion (MVI)-positive rates (53%-57% vs. 26%-31%), and worse recurrence-free survival. Hazard ratio (HR) for the subtype is 6.10 in subtype model. Clinical-radiologic model included alpha-fetoprotein (HR: 3.01), macrovascular invasion (HR: 2.32), nonsmooth tumor margin (HR: 1.81), rim enhancement (HR: 3.13), and intratumoral artery (HR: 2.21). Hybrid model included alpha-fetoprotein (HR: 2.70), nonsmooth tumor margin (HR: 1.51), rim enhancement (HR: 3.25), and subtypes (HR: 5.34). Subtype model was comparable to clinical-radiologic model (C-index: 0.71-0.73 vs. 0.71-0.73), but hybrid model outperformed both (C-index: 0.77-0.79).

CONCLUSION

MRI radiomics-based clustering identified two HCC subtypes with distinct MVI status and recurrence-free survival. Hybrid model showed superior capability to estimate recurrence.

LEVEL OF EVIDENCE

3 TECHNICAL EFFICACY STAGE: 2.

Large-scale environmental signals in seagrass blue carbon stocks are hidden by high variability at local scales

Increasing focus on nature-based climate change mitigation and adaptation strategies has led to the recognition of seagrasses as globally significant organic carbon (Corg) stocks. However, estimates of carbon stocks have been generally confined to a few regions, with few African studies represented in global datasets. In addition, the extent to which biogeographical and environmental variation shape carbon stocks in marine vegetated environments remains uncertain. For South Africa, Zostera capensis is the dominant seagrass species with limited mapping and quantification of its Corg stocks. Here, we measured Z. capensis Corg stocks at six South African estuaries spanning ∼1800 km of the cool-temperate to subtropical marine environmental gradient. Targeting the intertidal zone of the upper and lower estuary reaches, we collected Z. capensis sediments to a depth of 50 cm and measured the Corg, with the median Corg stock estimated at 24.11 Mg C ha-1 (40.4 ± 53.02; mean ± SD). While this is lower than the global average, these data demonstrate that Z. capensis ecosystems are important contributors to blue carbon stocks in the region. Measured Corg stocks showed significant differences between sampling sites for estuaries; however, we did not detect significant differences between estuaries due to high intra-estuarine Corg variability. Examination of biogeographical regions, terrestrial and marine environmental variables as drivers of Corg variability revealed that annual mean sea surface temperature may explain variation in Corg stocks. Furthermore, we found evidence of signals of biogeographical regions and precipitation driving some of the variability in Corg stocks; however, this requires further investigation. Overall, our estimates for Z. capensis add to ongoing national and global efforts to quantify seagrass Corg stocks across environmental and biogeographic gradients to better determine their contributions as nature-based solutions to climate change.

Integrated analysis of purine metabolism assists in predicting prognosis and treatment decisions for patients with lung adenocarcinoma

The incidence of lung cancer, especially lung adenocarcinoma (LUAD), has recently increased. Targeted therapy and immunotherapy combined with conventional treatment have shown surprising benefits in enhancing the LUAD patient's prognosis. For the purpose of guiding treatment planning and the prognosis of LUAD, more research is required. The particular aim of this work was to establish a purine metabolism scoring (PMS) model for the purpose of individually forecasting treatment outcomes and overall survival for patients who have LUAD. Clinical and whole genome data were obtained from the TCGA-LUAD cohort via "UCSC". The 25 driver purine metabolism-related prognostic genes were determined founded on univariate Cox regression. Then PMS was developed through stepwise LASSO Cox regression. Survival analysis indicated that patients who have PMS experienced worse outcomes. We validated the PGM2 effect on lung adenocarcinoma malignancy in in vitro experiments. Univariate as well as multivariate Cox regression suggested that PMS was an independent prognostic indicator for LUAD patients, which was confirmed in subgroup analysis. Functional assay demonstrated that immune response as well as cytotoxicity pathways have a connection with lower PMS, and patients who have low PMS possess an active immune microenvironment. Moreover, the LUAD patients who have low PMS showed greater sensitivity to immunotherapy, targeted therapy, as well as chemotherapy. Knockdown of PGM2 was discovered to decrease the proliferation, invasion, as well as migration of lung adenocarcinoma cells in an in vitro assay. Pertaining to this particular research, we created a PMS model and conducted a thorough analysis of purine metabolism in LUAD in order to determine prognosis and offer recommendations for treatment. This finding offered a fresh concept for the clinical management of LUAD and novel therapy protocols.

Development and validation of nomograms to evaluate the survival outcome of HCC patients undergoing selective postoperative adjuvant TACE

PURPOSE

The objective of this study was to develop and validate a novel prognostic nomogram to evaluate the survival benefit of hepatocellular carcinoma (HCC) patients receiving postoperative adjuvant transarterial chemoembolization (PA-TACE).

MATERIALS AND METHODS

Clinical data of HCC patients who underwent hepatectomy at four medical centers were retrospectively analyzed, including those who received PA-TACE and those who did not. These two categories of patients were randomly allocated to the development and validation cohorts in a 7:3 ratio.

RESULTS

A total of 1505 HCC patients who underwent hepatectomy were included in this study, comprising 723 patients who did not receive PA-TACE and 782 patients who received PA-TACE. Among them, patients who received PA-TACE experienced more adverse events, although these events were mild and manageable (Grade 1-2, all p < 0.05). Nomograms were constructed and validated for patients with and without PA-TACE using independent predictors that influenced disease-free survival (DFS) and overall survival (OS). These two nomograms had C-indices greater than 0.800 in the development cohort and exhibited good calibration and discrimination ability compared to six conventional HCC staging systems. Patients in the intermediate-to-high-risk group in the nomogram who received PA-TACE had higher DFS and OS (all p < 0.05). In addition, tumor recurrence was significantly controlled in the intermediate-to-high-risk group of patients who received PA-TACE, while there was no significant difference in the low-risk group of patients who received PA-TACE.

CONCLUSION

The nomograms were developed and validated based on large-scale clinical data and can serve as online decision-making tools to predict survival benefits from PA-TACE in different subgroups of patients.

Development and prognostic validation of a three-level NHG-like deep learning-based model for histological grading of breast cancer

BACKGROUND

Histological grade is a well-known prognostic factor that is routinely assessed in breast tumours. However, manual assessment of Nottingham Histological Grade (NHG) has high inter-assessor and inter-laboratory variability, causing uncertainty in grade assignments. To address this challenge, we developed and validated a three-level NHG-like deep learning-based histological grade model (predGrade). The primary performance evaluation focuses on prognostic performance.

METHODS

This observational study is based on two patient cohorts (SöS-BC-4, N = 2421 (training and internal test); SCAN-B-Lund, N = 1262 (test)) that include routine histological whole-slide images (WSIs) together with patient outcomes. A deep convolutional neural network (CNN) model with an attention mechanism was optimised for the classification of the three-level histological grading (NHG) from haematoxylin and eosin-stained WSIs. The prognostic performance was evaluated by time-to-event analysis of recurrence-free survival and compared to clinical NHG grade assignments in the internal test set as well as in the fully independent external test cohort.

RESULTS

We observed effect sizes (hazard ratio) for grade 3 versus 1, for the conventional NHG method (HR = 2.60 (1.18-5.70 95%CI, p-value = 0.017)) and the deep learning model (HR = 2.27, 95%CI 1.07-4.82, p-value = 0.033) on the internal test set after adjusting for established clinicopathological risk factors. In the external test set, the unadjusted HR for clinical NHG 2 versus 1 was estimated to be 2.59 (p-value = 0.004) and clinical NHG 3 versus 1 was estimated to be 3.58 (p-value < 0.001). For predGrade, the unadjusted HR for predGrade 2 versus 1 HR = 2.52 (p-value = 0.030), and 4.07 (p-value = 0.001) for preGrade 3 versus 1 was observed in the independent external test set. In multivariable analysis, HR estimates for neither clinical NHG nor predGrade were found to be significant (p-value > 0.05). We tested for differences in HR estimates between NHG and predGrade in the independent test set and found no significant difference between the two classification models (p-value > 0.05), confirming similar prognostic performance between conventional NHG and predGrade.

CONCLUSION

Routine histopathology assessment of NHG has a high degree of inter-assessor variability, motivating the development of model-based decision support to improve reproducibility in histological grading. We found that the proposed model (predGrade) provides a similar prognostic performance as clinical NHG. The results indicate that deep CNN-based models can be applied for breast cancer histological grading.

Using multi-scale genomics to associate poorly annotated genes with rare diseases

BACKGROUND

Next-generation sequencing (NGS) has significantly transformed the landscape of identifying disease-causing genes associated with genetic disorders. However, a substantial portion of sequenced patients remains undiagnosed. This may be attributed not only to the challenges posed by harder-to-detect variants, such as non-coding and structural variations but also to the existence of variants in genes not previously associated with the patient's clinical phenotype. This study introduces EvORanker, an algorithm that integrates unbiased data from 1,028 eukaryotic genomes to link mutated genes to clinical phenotypes.

METHODS

EvORanker utilizes clinical data, multi-scale phylogenetic profiling, and other omics data to prioritize disease-associated genes. It was evaluated on solved exomes and simulated genomes, compared with existing methods, and applied to 6260 knockout genes with mouse phenotypes lacking human associations. Additionally, EvORanker was made accessible as a user-friendly web tool.

RESULTS

In the analyzed exomic cohort, EvORanker accurately identified the "true" disease gene as the top candidate in 69% of cases and within the top 5 candidates in 95% of cases, consistent with results from the simulated dataset. Notably, EvORanker outperformed existing methods, particularly for poorly annotated genes. In the case of the 6260 knockout genes with mouse phenotypes, EvORanker linked 41% of these genes to observed human disease phenotypes. Furthermore, in two unsolved cases, EvORanker successfully identified DLGAP2 and LPCAT3 as disease candidates for previously uncharacterized genetic syndromes.

CONCLUSIONS

We highlight clade-based phylogenetic profiling as a powerful systematic approach for prioritizing potential disease genes. Our study showcases the efficacy of EvORanker in associating poorly annotated genes to disease phenotypes observed in patients. The EvORanker server is freely available at https://ccanavati.shinyapps.io/EvORanker/ .

Exploring the hub genes and potential drugs involved in Fanconi anemia using microarray datasets and bioinformatics analysis

Fanconi anemia (FA) is a genetic disorder that occurs when certain genes responsible for repairing DNA replication and promoting homologous recombination fail to function properly. This leads to severe clinical symptoms and a wide range of cancer-related characteristics. Recent treatment approaches for FA involve hematopoietic stem cell transplantation (HSCT), which helps restore the population of stem cells. A survival study using p-values indicated that specific hub genes play a significant role in diagnosing and predicting the disease. To find potential medications that interact with the identified hub genes, researchers inferred drugs. Among hub genes, TP53 was found to be particularly promising through computational analysis. Further investigation focused on two drugs, Topiramate and Tocofersolan predicted based on drug bank database analysis. Molecular docking strategies were employed to assess the best binding pose of these drugs with TP53. Topiramate showed a binding affinity of -6.5 kcal/mol, while Tocofersolan showed -8.5 kcal/mol against the active residues within the binding pocket. Molecular dynamics (MD) simulations were conducted to observe the stability of each drug's interaction with the TP53 protein over time. Both drugs exhibited stable confirmation with only slight changes in the loop region of the TP53 protein during the simulation intervals. Results also shows that there was a high fluctuation observed during apo-sate simulation time intervals as compared to complex system. Hence, it is suggested that the exploration of structure-based drug design holds promising results to specific target. This could potentially lead to a breakthrough in future experimental approaches for FA treatment.Communicated by Ramaswamy H. Sarma.

Human and mouse neutrophils share core transcriptional programs in both homeostatic and inflamed contexts

Neutrophils are frequently studied in mouse models, but the extent to which findings translate to humans remains poorly defined. In an integrative analysis of 11 mouse and 13 human datasets, we find a strong correlation of neutrophil gene expression across species. In inflammation, neutrophils display substantial transcriptional diversity but share a core inflammation program. This program includes genes encoding IL-1 family members, CD14, IL-4R, CD69, and PD-L1. Chromatin accessibility of core inflammation genes increases in blood compared to bone marrow and further in tissue. Transcription factor enrichment analysis implicates members of the NF-κB family and AP-1 complex as important drivers, and HoxB8 neutrophils with JunB knockout show a reduced expression of core inflammation genes in resting and activated cells. In independent single-cell validation data, neutrophil activation by type I or type II interferon, G-CSF, and E. coli leads to upregulation in core inflammation genes. In COVID-19 patients, higher expression of core inflammation genes in neutrophils is associated with more severe disease. In vitro treatment with GM-CSF, LPS, and type II interferon induces surface protein upregulation of core inflammation members. Together, we demonstrate transcriptional conservation in neutrophils in homeostasis and identify a core inflammation program shared across heterogeneous inflammatory conditions.

Electrocardiography score based on the Minnesota code classification system predicts cardiovascular mortality in an asymptomatic low-risk population

BACKGROUND

The use of a single abnormal finding on electrocardiography (ECG) is not recommended for stratifying the risk of cardiovascular (CV) events in low-risk general populations because of its low discriminative power. However, the value of a scoring system containing multiple abnormal ECG findings for predicting CV death has not been sufficiently evaluated.

METHODS

In a prospective community-based cohort study, 8417 participants without atherosclerotic CV diseases (ASCVDs) and any related symptoms were followed for 18 years. The standard 12-lead ECGs were recorded at baseline and the ECG findings were categorized using the Minnesota code classification. CV deaths were defined as death from myocardial infarction (MI), chronic ischemic heart disease, heart failure, fatal arrhythmia, cerebrovascular event, pulmonary thromboembolism, peripheral vascular disease and sudden cardiac arrest and identified using the Korean National Statistical Office (KOSTAT) database.

RESULTS

In a multivariate Cox proportional hazard (CPH) model, major and minor ST-T wave abnormalities, atrial fibrillation (AF), Q waves in the anterior leads, the lack of Q waves in the posterior leads, high amplitudes of the left and right precordial leads, left axis deviation and sinus tachycardia were associated with higher risks of CV deaths. The ECG score consisted of these findings showed modest predictive values represented by C-statistics that ranged from 0.632 to 760 during the follow-up and performed better in the early follow-up period. The ECG score independently predicted CV death after adjustment for relevant covariates in a multivariate model, and improved the predictive performance of the 10-year ASCVD risk estimator and a model of conventional risk factors including age, diabetes and current smoking. The combined ECG score (Harrell's C-index: 0.852, 95% confidence interval [CI], 0.828-0.876) composed of the ECG score and the conventional risk factors outperformed the 10-year ASCVD risk estimator (Harrell's C-index: 0.806; 95% CI, 0.780-0.833) and the model of the conventional risk factors (Harrell's C-index: 0.841, 95% CI, 0.817-0.865) and exhibited an excellent goodness of fit between the predicted and observed probabilities of CV death.

CONCLUSIONS

The ECG score could be useful to predict CV death independently and may add value to the conventional CV risk estimators regarding the risk stratification of CV death in asymptomatic low-risk general populations.

Machine Learning Models for Predicting Stroke Mortality in Malaysia: An Application and Comparative Analysis

Background Stroke is a significant public health concern characterized by increasing mortality and morbidity. Accurate long-term outcome prediction for acute stroke patients, particularly stroke mortality, is vital for clinical decision-making and prognostic management. This study aimed to develop and compare various prognostic models for stroke mortality prediction. Methods In a retrospective cohort study from January 2016 to December 2021, we collected data from patients diagnosed with acute stroke from five selected hospitals. Data contained variables on demographics, comorbidities, and interventions retrieved from medical records. The cohort comprised 950 patients with 20 features. Outcomes (censored vs. death) were determined by linking data with the Malaysian National Mortality Registry. We employed three common survival modeling approaches, the Cox proportional hazard regression (Cox), support vector machine (SVM), and random survival forest (RSF), while enhancing the Cox model with Elastic Net (Cox-EN) for feature selection. Models were compared using the concordance index (C-index), time-dependent area under the curve (AUC), and discrimination index (D-index), with calibration assessed by the Brier score. Results The support vector machine (SVM) model excelled among the four, with three-month, one-year, and three-year time-dependent AUC values of 0.842, 0.846, and 0.791; a D-index of 5.31 (95% CI: 3.86, 7.30); and a C-index of 0.803 (95% CI: 0.758, 0.847). All models exhibited robust calibration, with three-month, one-year, and three-year Brier scores ranging from 0.103 to 0.220, all below 0.25. Conclusion The support vector machine (SVM) model demonstrated superior discriminative performance, suggesting its efficacy in developing prognostic models for stroke mortality. This study enhances stroke mortality prediction and supports clinical decision-making, emphasizing the utility of the support vector machine method.

Interpretable Machine Learning for Choosing Radiation Dose-volume Constraints on Cardio-pulmonary Substructures Associated with Overall Survival in NRG Oncology RTOG 0617

PURPOSE

Our objective was to use interpretable machine learning for choosing dose-volume constraints on cardiopulmonary substructures (CPSs) associated with overall survival (OS) in radiation therapy for locally advanced non-small cell lung cancer.

METHODS AND MATERIALS

A total of 428 patients with non-small cell lung cancer were randomly divided into training/validation/test subsets (n = 230/149/49) in Radiation Therapy Oncology Group 0617. Manual or automated contouring was performed to segment CPSs, including heart, atria, ventricles, aorta, left/right ventricle/atrium (LV+RV+LA+RA), inferior/superior vena cava, pulmonary artery, and pericardium. Peri (pericardium-heart), rest (heart-[LV+RV+LA+RA]), clinical target volume (CTV), and lungs-CTV contours were also obtained. Dose-volume histogram features were extracted, including minimum/mean dose to the hottest x% volume (Dx%[Gy]/MOHx%[Gy]), minimum/mean/maximum dose, percent volume receiving at least xGy (VxGy[%]), and overlapping volume of each CPS with planning target volume (PTV_Voverlap[%]). Clinical parameters were collected from the National Clinical Trials Network/Community oncology research program data archive. Feature selection was performed using a series of multiblock sparse partial least squares regression, stability selection supervised principal component analysis, and Boruta. Explainable boosting machine (EBM) was trained using a conditional survival distribution-based approach for imputing censored data, treating survival analysis as a regression problem. Harrell's C-index was used to evaluate OS discrimination performance of EBM, Cox proportional hazards (CPH), random survival forest, extreme gradient boosting survival embeddings, and CPH deep neural network (DeepSurv) models in the test set. Dose-volume constraints were selected using the binary change point detection algorithm in Shapley additive explanations-based partial dependence functions.

RESULTS

Selected features included LA_V60Gy(%), pericardium_D30%(Gy), lungs-CTV_PTV_Voverlap(%), RA_V55Gy(%), and received_cons_chemo. All models ranked LA_V60Gy(%) as the most important feature. EBM achieved the best performance for predicting OS, followed by extreme gradient boosting survival embeddings, random survival forest, DeepSurv, and CPH (C-index = 0.653, 0.646, 0.642, 0.638, and 0.632). EBM global explanations suggested that LA_V60Gy(%) < 25.6, lungs-CTV_PTV_Voverlap(%) < 1.1, pericardium_D30%(Gy) < 18.9, RA_V55Gy(%) < 19.5, and received_cons_chemo = 'Yes' for improved OS.

CONCLUSIONS

EBM can be used to discriminate OS while also guiding dose-volume constraint selection for optimal management of cardiac toxicity in lung cancer radiation therapy.

Hybridizing machine learning in survival analysis of cardiac PET/CT imaging

BACKGROUND

Machine Learning (ML) allows integration of the numerous variables delivered by cardiac PET/CT, while traditional survival analysis can provide explainable prognostic estimates from a restricted number of input variables. We implemented a hybrid ML-and-survival analysis of multimodal PET/CT data to identify patients who developed myocardial infarction (MI) or death in long-term follow up.

METHODS

Data from 739 intermediate risk patients who underwent coronary CT and selectively stress 15O-water-PET perfusion were analyzed for the occurrence of MI and all-cause mortality. Images were evaluated segmentally for atherosclerosis and absolute myocardial perfusion through 75 variables that were integrated through ML into an ML-CCTA and an ML-PET score. These scores were then modeled along with clinical variables through Cox regression. This hybridized model was compared against an expert interpretation-based and a calcium score-based model.

RESULTS

Compared with expert- and calcium score-based models, the hybridized ML-survival model showed the highest performance (CI .81 vs .71 and .64). The strongest predictor for outcomes was the ML-CCTA score.

CONCLUSION

Prognostic modeling of PET/CT data for the long-term occurrence of adverse events may be improved through ML imaging score integration and subsequent traditional survival analysis with clinical variables. This hybridization of methods offers an alternative to traditional survival modeling of conventional expert image scoring and interpretation.

A prognostic model for ovarian neoplasms established by an integrated analysis of 1580 transcriptomic profiles

Even after debulking surgery combined with chemotherapy or new adjuvant chemotherapy paired with internal surgery, the average year of disease free survival in advanced ovarian cancer was approximately 1.7 years1. The development of a molecular predictor of early recurrence would allow for the identification of ovarian cancer (OC) patients with high risk of relapse. The Ovarian Cancer Disease Free Survival Predictor (ODFSP), a predictive model constructed from a special set of 1580 OC tumors in which gene expression was assessed using both microarray and sequencing platforms, was created by our team. To construct gene expression barcodes that were resistant to biases caused by disparate profiling platforms and batch effects, we employed a meta-analysis methodology that was based on the binary gene pair technique. We demonstrate that ODFSP is a reliable single-sample predictor of early recurrence (1 year or less) using the largest pool of OC transcriptome data sets available to date. The ODFSP model showed significantly high prognostic value for binary recurrence prediction unaffected by clinicopathologic factors, with a meta-estimate of the area under the receiver operating curve of 0.64 (P  =  4.6E-05) and a D-index (robust hazard ratio) of 1.67 (P  =  9.2E-06), respectively. GO analysis of ODFSP's 2040 gene pairs (collapsed to 886 distinct genes) revealed the involvement in small molecular catabolic process, sulfur compound metabolic process, organic acid catabolic process, sulfur compound biosynthetic process, glycosaminoglycan metabolic process and aminometabolic process. Kyoto encyclopedia of genes and genomes pathway analysis of ODFSP's signature genes identified prominent pathways that included cAMP signaling pathway and FoxO signaling pathway. By identifying individuals who might benefit from a more aggressive treatment plan or enrolment in a clinical trial but who will not benefit from standard surgery or chemotherapy, ODFSP could help with treatment decisions.

A novel tumor microenvironment-related gene signature with immune features for prognosis of lung squamous cell carcinoma

PURPOSE

Lung squamous cell carcinoma (LUSC) is an aggressive subset of non-small-cell lung cancer (NSCLC). The tumor microenvironment (TME) plays an important role in the development of LUSC. We aim to identify potential therapeutic targets and a TME-related prognostic signature and for LUSC.

METHODS

TME-related genes were obtained from TCGA-LUSC dataset. LUSC samples were clustered by the non-negative matrix clustering algorithm (NMF). The prognostic signature was constructed through univariate Cox regression, multivariate Cox regression, and the least absolute shrinkage and selection operator (LASSO) analyses. Gene set enrichment analysis (GSEA) was carried out to explore the enrichment pathways.

RESULTS

This study constructed a prognostic signature which contained 12 genes: HHIPL2, PLK4, SLC6A4, LSM1, TSLP, P4HA1, AMH, CLDN5, NRTN, CDH2, PTGIS, and STX1A. Patients were classified into high-risk and low-risk groups according to the median risk score of this signature. Compared with low-risk group patients, patients in high-risk group patients had poorer overall survival, which demonstrated this signature was an independent prognostic factor. Besides, correlation analysis and GSEA results revealed that genes of this signature were correlated with immune cells and drug response.

CONCLUSION

Our novel signature based on 12 TME-related genes might be applied as an independent prognostic indicator. Importantly, the signature could be a promising biomarker and accurately predict the prognosis of LUSC patients.

A deep learning approach based on multi-omics data integration to construct a risk stratification prediction model for skin cutaneous melanoma

PURPOSE

Skin cutaneous melanoma (SKCM) is a highly aggressive melanocytic carcinoma whose high heterogeneity and complex etiology make its prognosis difficult to predict. This study aimed to construct a risk subtype typing model for SKCM.

METHODS

The study proposes a deep learning framework combining early fusion feature autoencoder (AE) and late fusion feature AE for risk subtype prediction of SKCM. The deep learning framework integrates mRNA, miRNA, and DNA methylation data of SKCM patients from The Cancer Genome Atlas (TCGA), and clusters the screened multi-omics features associated with survival prognosis to identify risk subtypes. Differential expression analysis and functional enrichment analysis were performed between risk subtypes, while SVM classifiers were constructed between differentially expressed genes (DEGs) obtained by Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression screening and risk subtype labels inferred from multi-omics data, and the predictive robustness of risk subtypes inferred from the risk subtype classification prediction model was validated using two independent datasets.

RESULTS

The deep learning framework that combined early fusion feature AE with late fusion feature AE distinguished the two best risk subtypes compared to the multi-omics integration approach with single strategy AE or PCA. A promising C-index (C-index = 0.748) and a significant difference in survival (log-rank P value = 4.61 × 10-9) were found between the identified risk subtypes. The DEGs with the top significance values together with differentially expressed miRNAs provided the biological interpretation of risk subtypes on SKCM. Finally, the framework was applied to predict risk subtypes in two independent test datasets of SKCM patients, all of which showed good predictive power (C-index > 0.680) and significant survival differences (log-rank P value < 0.01).

CONCLUSION

The SKCM risk subtypes identified by integrating multi-omics data based on deep learning can not only improve the understanding of the molecular mechanisms of SKCM, but also provide clinicians with assistance in treatment decisions.

Multitask Learning with Convolutional Neural Networks and Vision Transformers Can Improve Outcome Prediction for Head and Neck Cancer Patients

Neural-network-based outcome predictions may enable further treatment personalization of patients with head and neck cancer. The development of neural networks can prove challenging when a limited number of cases is available. Therefore, we investigated whether multitask learning strategies, implemented through the simultaneous optimization of two distinct outcome objectives (multi-outcome) and combined with a tumor segmentation task, can lead to improved performance of convolutional neural networks (CNNs) and vision transformers (ViTs). Model training was conducted on two distinct multicenter datasets for the endpoints loco-regional control (LRC) and progression-free survival (PFS), respectively. The first dataset consisted of pre-treatment computed tomography (CT) imaging for 290 patients and the second dataset contained combined positron emission tomography (PET)/CT data of 224 patients. Discriminative performance was assessed by the concordance index (C-index). Risk stratification was evaluated using log-rank tests. Across both datasets, CNN and ViT model ensembles achieved similar results. Multitask approaches showed favorable performance in most investigations. Multi-outcome CNN models trained with segmentation loss were identified as the optimal strategy across cohorts. On the PET/CT dataset, an ensemble of multi-outcome CNNs trained with segmentation loss achieved the best discrimination (C-index: 0.29, 95% confidence interval (CI): 0.22-0.36) and successfully stratified patients into groups with low and high risk of disease progression (p=0.003). On the CT dataset, ensembles of multi-outcome CNNs and of single-outcome ViTs trained with segmentation loss performed best (C-index: 0.26 and 0.26, CI: 0.18-0.34 and 0.18-0.35, respectively), both with significant risk stratification for LRC in independent validation (p=0.002 and p=0.011). Further validation of the developed multitask-learning models is planned based on a prospective validation study, which has recently completed recruitment.

The predictive significance of a 5-m6A RNA methylation regulator signature in colorectal cancer

Colorectal cancer attacks the colon or rectum, with increasing morbidity and mortality globally. The RNA modification 6-methyladenine (m6A) is related to RNA modifications, playing a critical role in colorectal cancer. We aimed to identify prognostic signatures for colorectal cancer using risk prediction algorithms, and to validate these signatures using independent datasets and clinical samples. In this study, 175 cases in GSE17536 were assigned into two clusters using consistent clustering and PCA analysis. A multivariate Cox risk regression model revealed that among 21 m6A RNA methylation regulators, RBM15B, FTO, IGF2BP2, ZCCHC4, and KIAA1429 were remarkably associated with colorectal cancer patients' overall survival (OS); however, Kaplan-Meier (KM) survival assessment showed no significant association between these five regulators and colorectal cancer patients' prognosis. A 5-m6A RNA methylation regulator signature was established using LASSO algorithm. Risk scores of cases in GSE17536, GSE17537 and GSE75500 were calculated, and lower risk scores were associated with better DSS/OS. receiver operating characteristic (ROC) curve and the nomogram revealed the satisfactory predictive efficiency of the risk score model. The risk score could distinguish cases in Cluster1 and Cluster2 and normal and tumor tissues based on GSE37182. The prognostic variables for colorectal cancer patients were assessed using both univariate and multivariate Cox's proportional hazard regression models, which revealed that the stage and risk score were significant risk factors. In this study, a comprehensive set of integrative bioinformatics analyses was conducted to investigate the prognostic and diagnostic potential of a panel of 5 m6A RNA methylated regulators in colorectal cancer patients. The conducted studies included the use of several statistical methods, such as the LASSO regression model, KM survival evaluation, ROC curve, and univariate and multivariate Cox's proportional hazard regression analyses. The findings from these analyses collectively established the prognostic marker, highlighting its significance in predicting patient outcomes and diagnosing colorectal cancer.

Identification of lncRNAs associated with T cells as potential biomarkers and therapeutic targets in lung adenocarcinoma

Lung adenocarcinoma (LUAD) is the most common and deadliest subtype of lung cancer. To select more targeted and effective treatments for individuals, further advances in classifying LUAD are urgently needed. The number, type, and function of T cells in the tumor microenvironment (TME) determine the progression and treatment response of LUAD. Long noncoding RNAs (lncRNAs), may regulate T cell differentiation, development, and activation. Thus, our aim was to identify T cell-related lncRNAs (T cell-Lncs) in LUAD and to investigate whether T cell-Lncs could serve as potential stratifiers and therapeutic targets. Seven T cell-Lncs were identified to further establish the T cell-related lncRNA risk score (TRS) in LUAD. Low TRS individuals were characterized by robust immune status, fewer genomic alterations, and remarkably longer survival than high TRS individuals. The excellent accuracy of TRS in predicting overall survival (OS) was validated in the TCGA-LUAD training cohort and the GEO-LUAD validation cohort. Our data demonstrated the favorable predictive power of the TRS-based nomogram, which had important clinical significance in estimating the survival probability for individuals. In addition, individuals with low TRS could respond better to chemotherapy and immunotherapy than those with high TRS. LINC00525 was identified as a valuable study target, and the ability of LUAD to proliferate or invade was significantly attenuated by downregulation of LINC00525. In conclusion, the TRS established by T cell-Lncs could unambiguously classify LUAD patients, predict their prognosis and guide their management. Moreover, our identified T cell-Lncs could provide potential therapeutic targets for LUAD.

Comparison of deep learning-based recurrence-free survival with random survival forest and Cox proportional hazard models in Stage-I NSCLC patients

BACKGROUND

The curative treatment for Stage I non-small cell lung cancer (NSCLC) is surgical resection. Even for Stage I patients, the probability of recurrence after curative treatment is around 20%.

METHODS

In this retrospective study, we included 268 operated Stage I NSCLC patients between January 2008 and June 2018 to analyze the prognostic factors (pathological stage, histological type, number of sampled mediastinal lymph node stations, type of resection, SUVmax of the lesion) that may affect relapse with three different methods, Cox proportional hazard (CoxPH), random survival forest (RSF), DeepSurv, and to compare the performance of these methods with Harrell's C-index. The dataset was randomly split into two sets, training and test sets.

RESULTS

In the training set, DeepSurv showed the best performance among the three models, the C-index of the training set was 0.832, followed by RSF (0.675) and CoxPH (0.672). In the test set, RSF showed the best performance among the three models, followed by DeepSurv with 0.677 and CoxPH methods with 0.625.

CONCLUSION

In conclusion, machine-learning techniques can be useful in predicting recurrence for lung cancer and guide clinicians both in choosing the adjuvant treatment options and best follow-up programs.

Algorithmically Reconstructed Molecular Pathways as the New Generation of Prognostic Molecular Biomarkers in Human Solid Cancers

Individual gene expression and molecular pathway activation profiles were shown to be effective biomarkers in many cancers. Here, we used the human interactome model to algorithmically build 7470 molecular pathways centered around individual gene products. We assessed their associations with tumor type and survival in comparison with the previous generation of molecular pathway biomarkers (3022 "classical" pathways) and with the RNA transcripts or proteomic profiles of individual genes, for 8141 and 1117 samples, respectively. For all analytes in RNA and proteomic data, respectively, we found a total of 7441 and 7343 potential biomarker associations for gene-centric pathways, 3020 and 2950 for classical pathways, and 24,349 and 6742 for individual genes. Overall, the percentage of RNA biomarkers was statistically significantly higher for both types of pathways than for individual genes (p < 0.05). In turn, both types of pathways showed comparable performance. The percentage of cancer-type-specific biomarkers was comparable between proteomic and transcriptomic levels, but the proportion of survival biomarkers was dramatically lower for proteomic data. Thus, we conclude that pathway activation level is the advanced type of biomarker for RNA and proteomic data, and momentary algorithmic computer building of pathways is a new credible alternative to time-consuming hypothesis-driven manual pathway curation and reconstruction.

The prognostic value of 11th Japanese classification and 8th AJCC staging systems in Chinese patients with esophageal squamous cell carcinoma

BACKGROUND

Two staging systems, the 8th staging system by the American Joint Committee on Cancer (AJCC) and the 11th Japanese classification by Japan Esophageal Society (JES), are currently applied in the clinic for predicting the prognosis of patients with esophageal squamous cell carcinoma (ESCC). The differences between the two staging systems have been widely researched. However, little studies focus on the differences in specific staging between the two systems. Therefore, we aimed to compare the performance of different staging in predicting overall survival (OS) of Chinese patients with ESCC.

METHODS

This retrospective study included 268 patients who underwent radical esophagectomy and mediastinal lymph node dissection for ESCC between January 2008 and December 2013. Patients were staged by the 8th AJCC and 11th JES staging systems. OS was estimated using the Kaplan-Meier method and compared between N stages and between stage groupings using the log-rank test. Cox proportional hazards regression analysis was performed to identify factors independently related to outcome. Further, we compared the concordance indexes (C-indexes) of the two staging systems.

RESULTS

The mean age was 61.25 ± 7.056 years, median follow-up was 44.82 months, and 5-year OS rate was 47%. The OS was well predicted by the 8th AJCC N staging (P < 0.001) and the 11th JES N staging (P < 0.001), with a c-index of 0.638 (95% CI: 0.592-0.683) for AJCC N staging and 0.627 (95% CI: 0.583-0.670) for JES N staging (P = 0.13). In addition, the OS was also well predicted by stage groupings of the 8th AJCC (P < 0.001) and the 11th JES systems (P < 0.001), with a c-index of 0.658 (95% CI: 0.616-0.699) for 8th AJCC stage grouping and 0.629 (95% CI: 0.589-0.668) for the11th JES stage grouping (P = 0.211).

CONCLUSIONS

The prognostic effect of 11th JES staging system is comparable with that of AJCC 8th staging system for patients with ESCC. Therefore, both systems are applicable to clinical practice.

Novel prognostic gene signature for pancreatic ductal adenocarcinoma based on hypoxia

BACKGROUND

Currently, there is lack of marker to accurately assess the prognosis of patients diagnosed with pancreatic ductal adenocarcinoma (PDAC). This study aims to establish a hypoxia-related risk scoring model that can effectively predict the prognosis and chemotherapy outcomes of PDAC patients.

METHODS

Using unsupervised consensus clustering algorithms, we comprehensively analyzed The Cancer Genome Atlas (TCGA) data to identify two distinct hypoxia clusters and used the weighted gene co-expression network analysis (WGCNA) to examine gene sets significantly associated with these hypoxia clusters. Then univariate Cox regression, the least absolute shrinkage and selection operator (LASSO) Cox regression and multivariate Cox regression were used to construct a signature and its efficacy was evaluated using the International Cancer Genome Consortium (ICGC) PDAC cohort. Further, the correlation between the risk scores obtained from the signature and carious clinical, pathological, immunophenotype, and immunoinfiltration factors as well as the differences in immunotherapy potential and response to common chemotherapy drugs between high-risk and low-risk groups were evaluated.

RESULTS

From a total of 8 significantly related modules and 4423 genes, 5 hypoxia-related signature genes were identified to construct a risk model. Further analysis revealed that the overall survival rate (OS) of patients in the low-risk group was significantly higher than the high-risk group. Univariate and multivariate Cox regression analysis showed that the risk scoring signature was an independent factor for prognosis prediction. Analysis of immunocyte infiltration and immunophenotype showed that the immune score and the anticancer immune response in the high-risk were significantly lower than that in the low-risk group.

CONCLUSION

The constructed hypoxia-associated prognostic signature demonstrated could be used as a potential risk classifier for PDAC.

Identification and verification of novel immune-related ferroptosis signature with excellent prognostic predictive and clinical guidance value in hepatocellular carcinoma

Background: Immunity and ferroptosis often play a synergistic role in the progression and treatment of hepatocellular carcinoma (HCC). However, few studies have focused on identifying immune-related ferroptosis gene biomarkers. Methods: We performed weighted gene co-expression network analysis (WGCNA) and random forest to identify prognostic differentially expressed immune-related genes (PR-DE-IRGs) highly related to HCC and characteristic prognostic differentially expressed ferroptosis-related genes (PR-DE-FRGs) respectively to run co-expression analysis for prognostic differentially expressed immune-related ferroptosis characteristic genes (PR-DE-IRFeCGs). Lasso regression finally identified 3 PR-DE-IRFeCGs for us to construct a prognostic predictive model. Differential expression and prognostic analysis based on shared data from multiple sources and experimental means were performed to further verify the 3 modeled genes' biological value in HCC. We ran various performance testing methods to test the model's performance and compare it with other similar signatures. Finally, we integrated composite factors to construct a comprehensive quantitative nomogram for accurate prognostic prediction and evaluated its performance. Results: 17 PR-DE-IRFeCGs were identified based on co-expression analysis between the screened 17 PR-DE-FRGs and 34 PR-DE-IRGs. Multi-source sequencing data, QRT-PCR, immunohistochemical staining and testing methods fully confirmed the upregulation and significant prognostic influence of the three PR-DE-IRFeCGs in HCC. The model performed well in the performance tests of multiple methods based on the 5 cohorts. Furthermore, our model outperformed other related models in various performance tests. The immunotherapy and chemotherapy guiding value of our signature and the comprehensive nomogram's excellent performance have also stood the test. Conclusion: We identified a novel PR-DE-IRFeCGs signature with excellent prognostic prediction and clinical guidance value in HCC.

Immunometabolic coevolution defines unique microenvironmental niches in ccRCC

Tumor cell phenotypes and anti-tumor immune responses are shaped by local metabolite availability, but intratumoral metabolite heterogeneity (IMH) and its phenotypic consequences remain poorly understood. To study IMH, we profiled tumor/normal regions from clear cell renal cell carcinoma (ccRCC) patients. A common pattern of IMH transcended all patients, characterized by correlated fluctuations in the abundance of metabolites and processes associated with ferroptosis. Analysis of intratumoral metabolite-RNA covariation revealed that the immune composition of the microenvironment, especially the abundance of myeloid cells, drove intratumoral metabolite variation. Motivated by the strength of RNA-metabolite covariation and the clinical significance of RNA biomarkers in ccRCC, we inferred metabolomic profiles from the RNA sequencing data of ccRCC patients enrolled in 7 clinical trials, and we ultimately identifyied metabolite biomarkers associated with response to anti-angiogenic agents. Local metabolic phenotypes, therefore, emerge in tandem with the immune microenvironment, influence ongoing tumor evolution, and are associated with therapeutic sensitivity.

Integrating machine learning algorithms to systematically assess reactive oxygen species levels to aid prognosis and novel treatments for triple -negative breast cancer patients

Introduction

Breast cancer has become one of the top health concerns for women, and triple-negative breast cancer (TNBC) leads to treatment resistance and poor prognosis due to its high degree of heterogeneity and malignancy. Reactive oxygen species (ROS) have been found to play a dual role in tumors, and modulating ROS levels may provide new insights into prognosis and tumor treatment.

Methods

This study attempted to establish a robust and valid ROS signature (ROSig) to aid in assessing ROS levels. The driver ROS prognostic indicators were searched based on univariate Cox regression. A well-established pipeline integrating 9 machine learning algorithms was used to generate the ROSig. Subsequently, the heterogeneity of different ROSig levels was resolved in terms of cellular communication crosstalk, biological pathways, immune microenvironment, genomic variation, and response to chemotherapy and immunotherapy. In addition, the effect of the core ROS regulator HSF1 on TNBC cell proliferation was detected by cell counting kit-8 and transwell assays.

Results

A total of 24 prognostic ROS indicators were detected. A combination of the Coxboost+ Survival Support Vector Machine (survival-SVM) algorithm was chosen to generate ROSig. ROSig proved to be the superior risk predictor for TNBC. Cellular assays show that knockdown of HSF1 can reduce the proliferation and invasion of TNBC cells. The individual risk stratification based on ROSig showed good predictive accuracy. High ROSig was identified to be associated with higher cell replication activity, stronger tumor heterogeneity, and an immunosuppressive microenvironment. In contrast, low ROSig indicated a more abundant cellular matrix and more active immune signaling. Low ROSig has a higher tumor mutation load and copy number load. Finally, we found that low ROSig patients were more sensitive to doxorubicin and immunotherapy.

Conclusion

In this study, we developed a robust and effective ROSig model that can be used as a reliable indicator for prognosis and treatment decisions in TNBC patients. This ROSig also allows a simple assessment of TNBC heterogeneity in terms of biological function, immune microenvironment, and genomic variation.

Pacpaint: a histology-based deep learning model uncovers the extensive intratumor molecular heterogeneity of pancreatic adenocarcinoma

Two tumor (Classical/Basal) and stroma (Inactive/active) subtypes of Pancreatic adenocarcinoma (PDAC) with prognostic and theragnostic implications have been described. These molecular subtypes were defined by RNAseq, a costly technique sensitive to sample quality and cellularity, not used in routine practice. To allow rapid PDAC molecular subtyping and study PDAC heterogeneity, we develop PACpAInt, a multi-step deep learning model. PACpAInt is trained on a multicentric cohort (n = 202) and validated on 4 independent cohorts including biopsies (surgical cohorts n = 148; 97; 126 / biopsy cohort n = 25), all with transcriptomic data (n = 598) to predict tumor tissue, tumor cells from stroma, and their transcriptomic molecular subtypes, either at the whole slide or tile level (112 µm squares). PACpAInt correctly predicts tumor subtypes at the whole slide level on surgical and biopsies specimens and independently predicts survival. PACpAInt highlights the presence of a minor aggressive Basal contingent that negatively impacts survival in 39% of RNA-defined classical cases. Tile-level analysis ( > 6 millions) redefines PDAC microheterogeneity showing codependencies in the distribution of tumor and stroma subtypes, and demonstrates that, in addition to the Classical and Basal tumors, there are Hybrid tumors that combine the latter subtypes, and Intermediate tumors that may represent a transition state during PDAC evolution.

A longitudinal circulating tumor DNA-based model associated with survival in metastatic non-small-cell lung cancer

One of the great challenges in therapeutic oncology is determining who might achieve survival benefits from a particular therapy. Studies on longitudinal circulating tumor DNA (ctDNA) dynamics for the prediction of survival have generally been small or nonrandomized. We assessed ctDNA across 5 time points in 466 non-small-cell lung cancer (NSCLC) patients from the randomized phase 3 IMpower150 study comparing chemotherapy-immune checkpoint inhibitor (chemo-ICI) combinations and used machine learning to jointly model multiple ctDNA metrics to predict overall survival (OS). ctDNA assessments through cycle 3 day 1 of treatment enabled risk stratification of patients with stable disease (hazard ratio (HR) = 3.2 (2.0-5.3), P < 0.001; median 7.1 versus 22.3 months for high- versus low-intermediate risk) and with partial response (HR = 3.3 (1.7-6.4), P < 0.001; median 8.8 versus 28.6 months). The model also identified high-risk patients in an external validation cohort from the randomized phase 3 OAK study of ICI versus chemo in NSCLC (OS HR = 3.73 (1.83-7.60), P = 0.00012). Simulations of clinical trial scenarios employing our ctDNA model suggested that early ctDNA testing outperforms early radiographic imaging for predicting trial outcomes. Overall, measuring ctDNA dynamics during treatment can improve patient risk stratification and may allow early differentiation between competing therapies during clinical trials.

Multimodal deep learning to predict prognosis in adult and pediatric brain tumors

BACKGROUND

The introduction of deep learning in both imaging and genomics has significantly advanced the analysis of biomedical data. For complex diseases such as cancer, different data modalities may reveal different disease characteristics, and the integration of imaging with genomic data has the potential to unravel additional information than when using these data sources in isolation. Here, we propose a DL framework that combines these two modalities with the aim to predict brain tumor prognosis.

METHODS

Using two separate glioma cohorts of 783 adults and 305 pediatric patients we developed a DL framework that can fuse histopathology images with gene expression profiles. Three strategies for data fusion were implemented and compared: early, late, and joint fusion. Additional validation of the adult glioma models was done on an independent cohort of 97 adult patients.

RESULTS

Here we show that the developed multimodal data models achieve better prediction results compared to the single data models, but also lead to the identification of more relevant biological pathways. When testing our adult models on a third brain tumor dataset, we show our multimodal framework is able to generalize and performs better on new data from different cohorts. Leveraging the concept of transfer learning, we demonstrate how our pediatric multimodal models can be used to predict prognosis for two more rare (less available samples) pediatric brain tumors.

CONCLUSIONS

Our study illustrates that a multimodal data fusion approach can be successfully implemented and customized to model clinical outcome of adult and pediatric brain tumors.

Golgi-apparatus genes related signature for predicting the progression-free interval of patients with papillary thyroid carcinoma

BACKGROUND

We aimed to build a novel model with golgi apparatus related genes (GaGs) signature and relevant clinical parameters for predicting progression-free interval (PFI) after surgery for papillary thyroid carcinoma (PTC).

METHODS

We performed a bioinformatic analysis of integrated PTC datasets with the GaGs to identify differentially expressed GaGs (DE-GaGs). Then we generated PFI-related DE-GaGs and established a novel GaGs based signature. After that, we validated the signature on multiple external datasets and PTC cell lines. Further, we conducted uni- and multivariate analyses to identify independent prognostic characters. Finally, we established a signature and clinical parameters-based nomogram for predicting the PFI of PTC.

RESULTS

We identified 260 DE-GaGs related to PFI in PTC. The functional enrichment analysis showed that the DE-MTGs were associated with an essential oncogenic glycoprotein biosynthetic process. Consequently, we established and optimized a novel 11 gene signature that could distinguish patients with poorer prognoses and predicted PFI accurately. The novel signature had a C-index of 0.78, and the relevant nomogram had a C-index of 0.79. Also, it was closely related to the pivotal clinical characters of and anaplastic potential in datasets and PTC cell lines. And the signature was confirmed a significant independent prognostic factor in PTC. Finally, we built a nomogram by including the signature and relevant clinical factors. Validation analysis showed that the nomogram's efficacy was satisfying in predicting PTC's PFI.

CONCLUSION

The GaGs signature and nomogram were closely associated with PTC prognosis and may help clinicians improve the individualized prediction of PFI, especially for high-risk patients after surgery.

WGCNA combined with machine learning algorithms for analyzing key genes and immune cell infiltration in heart failure due to ischemic cardiomyopathy

Background

Ischemic cardiomyopathy (ICM) induced heart failure (HF) is one of the most common causes of death worldwide. This study aimed to find candidate genes for ICM-HF and to identify relevant biomarkers by machine learning (ML).

Methods

The expression data of ICM-HF and normal samples were downloaded from Gene Expression Omnibus (GEO) database. Differentially expressed genes (DEGs) between ICM-HF and normal group were identified. Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment and gene ontology (GO) annotation analysis, protein–protein interaction (PPI) network, gene pathway enrichment analysis (GSEA), and single-sample gene set enrichment analysis (ssGSEA) were performed. Weighted gene co-expression network analysis (WGCNA) was applied to screen for disease-associated modules, and relevant genes were derived using four ML algorithms. The diagnostic values of candidate genes were assessed using receiver operating characteristic (ROC) curves. The immune cell infiltration analysis was performed between the ICM-HF and normal group. Validation was performed using another gene set.

Results

A total of 313 DEGs were identified between ICM-HF and normal group of GSE57345, which were mainly enriched in biological processes and pathways related to cell cycle regulation, lipid metabolism pathways, immune response pathways, and intrinsic organelle damage regulation. GSEA results showed positive correlations with pathways such as cholesterol metabolism in the ICM-HF group compared to normal group and lipid metabolism in adipocytes. GSEA results also showed a positive correlation with pathways such as cholesterol metabolism and a negative correlation with pathways such as lipolytic presentation in adipocytes compared to normal group. Combining multiple ML and cytohubba algorithms yielded 11 relevant genes. After validation using the GSE42955 validation sets, the 7 genes obtained by the machine learning algorithm were well verified. The immune cell infiltration analysis showed significant differences in mast cells, plasma cells, naive B cells, and NK cells.

Conclusion

Combined analysis using WGCNA and ML identified coiled-coil-helix-coiled-coil-helix domain containing 4 (CHCHD4), transmembrane protein 53 (TMEM53), acid phosphatase 3 (ACPP), aminoadipate-semialdehyde dehydrogenase (AASDH), purinergic receptor P2Y1 (P2RY1), caspase 3 (CASP3) and aquaporin 7 (AQP7) as potential biomarkers of ICM-HF. ICM-HF may be closely related to pathways such as mitochondrial damage and disorders of lipid metabolism, while the infiltration of multiple immune cells was identified to play a critical role in the progression of the disease.

Identification of fatty acid metabolism-based molecular subtypes and prognostic signature to predict immune landscape and guide clinical drug treatment in renal clear cell carcinoma

Three subtypes of samples were generated based on genes involved in fatty acid metabolism in The Cancer Genome Atlas (TCGA)-RCC patients using a non-negative matrix factorization (NMF) algorithm. 32 co-expressed modules were identified using WCGNA. We constructed a four-gene signature in our training set using least absolute shrinkage selection operator regression analysis and verified it in our testing and overall sets. A relevant study analysis in clinical trials was conducted, which showed the model had good stability and potential application value for predicting outcomes. We analyzed the immune microenvironment using MCPcounter, CIBERSORT, quanTIseq, TIMER and ESTIMATE algorithms, and the result indicated risk was positively related to T cells, B-lineage, and fibroblasts and negatively correlated with monocytic lineage, myeloid dendritic cells, neutrophils, and endothelial cells, and CPT1B was positively related to T cells, CD8 + T cells, Cytotoxic lymphocytes and NK cells, and negatively correlated with myeloid dendritic cells, fibroblasts, endothelial cells. Tumor mutation burden was positively related to risk score and the expression of CPT1B using the R packages corrplot, circlize. Through the R package pRRophetic, drug sensitivity tests showed that the low-risk score group would benefit more from sunitinib and less from pazopanib, sorafenib, temsirolimus, gemcitabine and doxorubicin than the high-risk score group. We performed the relevant basic assay validation for CPT1B, and the proliferation ability of RCC cells was inhibited after the knockdown of protein expression of CPT1B. In conclusion, we established a four-gene model that can predict outcomes of RCC with potential applications in diagnosis and treatment.

Reduced methylation correlates with diabetic nephropathy risk in type 1 diabetes

Diabetic nephropathy (DN) is a polygenic disorder with few risk variants showing robust replication in large-scale genome-wide association studies. To understand the role of DNA methylation, it is important to have the prevailing genomic view to distinguish key sequence elements that influence gene expression. This is particularly challenging for DN because genome-wide methylation patterns are poorly defined. While methylation is known to alter gene expression, the importance of this causal relationship is obscured by array-based technologies since coverage outside promoter regions is low. To overcome these challenges, we performed methylation sequencing using leukocytes derived from participants of the Finnish Diabetic Nephropathy (FinnDiane) type 1 diabetes (T1D) study (n = 39) that was subsequently replicated in a larger validation cohort (n = 296). Gene body-related regions made up more than 60% of the methylation differences and emphasized the importance of methylation sequencing. We observed differentially methylated genes associated with DN in 3 independent T1D registries originating from Denmark (n = 445), Hong Kong (n = 107), and Thailand (n = 130). Reduced DNA methylation at CTCF and Pol2B sites was tightly connected with DN pathways that include insulin signaling, lipid metabolism, and fibrosis. To define the pathophysiological significance of these population findings, methylation indices were assessed in human renal cells such as podocytes and proximal convoluted tubule cells. The expression of core genes was associated with reduced methylation, elevated CTCF and Pol2B binding, and the activation of insulin-signaling phosphoproteins in hyperglycemic cells. These experimental observations also closely parallel methylation-mediated regulation in human macrophages and vascular endothelial cells.

Development of preoperative prognostic models including radiological features for survival of singular nodular HCC patients

BACKGROUND

Early singular nodular hepatocellular carcinoma (HCC) is an ideal surgical indication in clinical practice. However, almost half of the patients have tumor recurrence, and there is no reliable prognostic prediction tool. Besides, it is unclear whether preoperative neoadjuvant therapy is necessary for patients with early singular nodular HCC and which patient needs it. It is critical to identify the patients with high risk of recurrence and to treat these patients preoperatively with neoadjuvant therapy and thus, to improve the outcomes of these patients. The present study aimed to develop two prognostic models to preoperatively predict the recurrence-free survival (RFS) and overall survival (OS) in patients with singular nodular HCC by integrating the clinical data and radiological features.

METHODS

We retrospective recruited 211 patients with singular nodular HCC from December 2009 to January 2019 at Eastern Hepatobiliary Surgery Hospital (EHBH). They all met the surgical indications and underwent radical resection. We randomly divided the patients into the training cohort (n =132) and the validation cohort (n = 79). We established and validated multivariate Cox proportional hazard models by the preoperative clinicopathologic factors and radiological features for association with RFS and OS. By analyzing the receiver operating characteristic (ROC) curve, the discrimination accuracy of the models was compared with that of the traditional predictive models.

RESULTS

Our RFS model was based on HBV-DNA score, cirrhosis, tumor diameter and tumor capsule in imaging. RFS nomogram had fine calibration and discrimination capabilities, with a C-index of 0.74 (95% CI: 0.68-0.80). The OS nomogram, based on cirrhosis, tumor diameter and tumor capsule in imaging, had fine calibration and discrimination capabilities, with a C-index of 0.81 (95% CI: 0.74-0.87). The area under the receiver operating characteristic curve (AUC) of our model was larger than that of traditional liver cancer staging system, Korea model and Nomograms in Hepatectomy Patients with Hepatitis B Virus-Related Hepatocellular Carcinoma, indicating better discrimination capability. According to the models, we fitted the linear prediction equations. These results were validated in the validation cohort.

CONCLUSIONS

Compared with previous radiography model, the new-developed predictive model was concise and applicable to predict the postoperative survival of patients with singular nodular HCC. Our models may preoperatively identify patients with high risk of recurrence. These patients may benefit from neoadjuvant therapy which may improve the patients' outcomes.

Identification and validation of a muscle failure index to predict prognosis and immunotherapy in lung adenocarcinoma through integrated analysis of bulk and single-cell RNA sequencing data

Background

It was previously reported that the production of exerkines is positively associated with the beneficial effects of exercise in lung adenocarcinoma (LUAD) patients. This study proposes a novel scoring system based on muscle failure-related genes, to assist in clinical decision making.

Methods

A comprehensive analysis of bulk and single cell RNA sequencing (scRNA-seq) of early, advanced and brain metastatic LUAD tissues and normal lung tissues was performed to identify muscle failure-related genes in LUAD and to determine the distribution of muscle failure-related genes in different cell populations. A novel scoring system, named MFI (Muscle failure index), was developed and validated. The differences in biological functions, immune infiltration, genomic alterations, and clinical significance of different subtypes were also investigated.

Results

First, we conducted single cell analysis on the dataset GSE131907 and identified eight cell subpopulations. We found that four muscle failure-related genes (BDNF, FNDC5, IL15, MSTN) were significantly increased in tumor cells. In addition, IL15 was widely distributed in the immune cell population. And we have validated it in our own clinical cohort. Then we created the MFI model based on 10 muscle failure-related genes using the LASSO algorithm, and MFI remained an independent prognostic factor of OS in both the training and validation cohorts. Moreover, we generated MFI in the single-cell dataset, in which cells with high MFI received and sent more signals compared to those with low MFI. Biological function analysis of both subtypes revealed stronger anti-tumor immune activity in the low MFI group, while tumor cells with high MFI had stronger metabolic and proliferative activity. Finally, we systematically assessed the immune cell activity and immunotherapy responses in LUAD patients, finding that the low MFI group was more sensitive to immunotherapy.

Conclusion

Overall, our study can improve the understanding of the role of muscle failure-related genes in tumorigenesis and we constructed a reliable MFI model for predicting prognosis and guiding future clinical decision making.

An immune, stroma, and epithelial-mesenchymal transition-related signature for predicting recurrence and chemotherapy benefit in stage II-III colorectal cancer

BACKGROUND

Debates exist on the treatment decision of the stage II/III colorectal cancer (CRC) due to the insufficiency of the current TNM stage-based risk stratification system. Epithelial-mesenchymal transition (EMT) and tumor microenvironment (TME) have both been linked to CRC progression in recent studies. We propose to improve the prognosis prediction of CRC by integrating TME and EMT.

METHODS

In total, 2382 CRC patients from seven datasets and one in-house cohort were collected, and 1640 stage II/III CRC patients with complete survival information and gene expression profiles were retained and divided into a training cohort and three independent validation cohorts. Integrated analysis of 398 immune, stroma, and epithelial-mesenchymal transition (ISE)-related genes identified an ISE signature independently associated with the recurrence of CRC. The underlying biological mechanism of the ISE signature and its influence on adjuvant chemotherapy was further explored.

RESULTS

We constructed a 26-gene signature which was significantly associated with poor outcome in Training cohort (p < 0.001, HR [95%CI] = 4.42 [3.25-6.01]) and three independent validation cohorts (Validation cohort-1: p < 0.01, HR [95%CI] = 1.70 [1.15-2.51]; Validation cohort-2: p < 0.001, HR [95% CI] = 2.30 [1.67-3.16]; Validation cohort-3: p < 0.01, HR [95% CI] = 2.42 [1.25-4.70]). After adjusting for known clinicopathological factors, multivariate cox analysis confirmed the ISE signature's independent prognostic value. Subgroup analysis found that stage III patients with low ISE score might benefit from adjuvant chemotherapy (p < 0.001, HR [95%CI] = 0.15 [0.04-0.55]). Hypergeometric test and enrichment analysis revealed that low-risk group was enriched in thr immune pathway while high-risk group was associated with the EMT pathway and CMS4 subtype.

CONCLUSION

We proposed an ISE signature for robustly predicting the recurrence of stage II/III CRC and help treatment decision by identifying patients who will not benefit from current standard treatment.

Systematic integration of machine learning algorithms to develop immune escape-related signatures to improve clinical outcomes in lung adenocarcinoma patients

Background

Immune escape has recently emerged as one of the barriers to the efficacy of immunotherapy in lung adenocarcinoma (LUAD). However, the clinical significance and function of immune escape markers in LUAD have largely not been clarified.

Methods

In this study, we constructed a stable and accurate immune escape score (IERS) by systematically integrating 10 machine learning algorithms. We further investigated the clinical significance, functional status, TME interactions, and genomic alterations of different IERS subtypes to explore potential mechanisms. In addition, we validated the most important variable in the model through cellular experiments.

Results

The IERS is an independent risk factor for overall survival, superior to traditional clinical variables and published molecular signatures. IERS-based risk stratification can be well applied to LUAD patients. In addition, high IERS is associated with stronger tumor proliferation and immunosuppression. Low IERS exhibited abundant lymphocyte infiltration and active immune activity. Finally, high IERS is more sensitive to first-line chemotherapy for LUAD, while low IERS is more sensitive to immunotherapy.

Conclusion

In conclusion, IERS may serve as a promising clinical tool to improve risk stratification and clinical management of individual LUAD patients and may enhance the understanding of immune escape.

Evaluation and comparison of different breast cancer prognosis scores based on gene expression data

BACKGROUND

Breast cancer is one of the three most common cancers worldwide and is the most common malignancy in women. Treatment approaches for breast cancer are diverse and varied. Clinicians must balance risks and benefits when deciding treatments, and models have been developed to support this decision-making. Genomic risk scores (GRSs) may offer greater clinical value than standard clinicopathological models, but there is limited evidence as to whether these models perform better than the current clinical standard of care.

METHODS

PREDICT and GRSs were adapted using data from the original papers. Univariable Cox proportional hazards models were produced with breast cancer-specific survival (BCSS) as the outcome. Independent predictors of BCSS were used to build multivariable models with PREDICT. Signatures which provided independent prognostic information in multivariable models were incorporated into the PREDICT algorithm and assessed for calibration, discrimination and reclassification.

RESULTS

EndoPredict, MammaPrint and Prosigna demonstrated prognostic power independent of PREDICT in multivariable models for ER-positive patients; no score predicted BCSS in ER-negative patients. Incorporating these models into PREDICT had only a modest impact upon calibration (with absolute improvements of 0.2-0.8%), discrimination (with no statistically significant c-index improvements) and reclassification (with 4-10% of patients being reclassified).

CONCLUSION

Addition of GRSs to PREDICT had limited impact on model fit or treatment received. This analysis does not support widespread adoption of current GRSs based on our implementations of commercial products.