Independent External Cohort Research Articles

A Novel Signature Integrated of Immunoglobulin, Glycosylation and Anti-Viral Genes to Predict Prognosis for Breast Cancer.

Background: Aberrant glycosylation is significantly related to the occurrence, progression, metastasis, and drug resistance of tumors. It is essential to identify glycosylation and related genes with prognostic value for breast cancer. Objective: We aimed to construct and validate a prognostic model based on glycosylation and related genes, and further investigate its prognosis values in validation set and external independent cohorts. Materials and Methods: The transcriptome and clinical data of breast cancer patients were downloaded from The Cancer Genome Atlas (TCGA, n = 1072), Molecular Taxonomy of Breast Cancer International Consortium (METABRIC, n = 1451), and GSE2741 (n = 120). Glycosylation-related genes were downloaded from the Genecards website. Differentially expressed glycosylation-related geneswere identified by comparing the tumor tissues with the adjacent tissues. The TCGA data were randomly divided into training set and validation set in a 1:1 ratio for further analysis. The glycosylation risk-scoring prognosis model was constructed by univariate and multivariate Cox regression analysis, followed by confirmation in TCGA validation, METABRIC, and GEO datasets. Gene set enrichment analysis (GSEA) and Gene ontology analysis for identifying the affected pathways in the high- and low-risk groups were performed. Results: We attained 1072 breast cancer samples from the TCGA database and 786 glycosylation genes from the Genecards website. A signature contains immunoglobulin, glycosylation and anti-viral related genes was constructed to separate BRCA patients into two risk groups. Low-risk patients had better overall survival than high-risk patients (p < 0.001). A nomogram was constructed with risk scores and clinical characteristics. The area under time-dependent ROC curve reached 0.764 at 1 year, 0.744 at 3 years, and 0.765 at 5 years in the training set. Subgroup analysis showed differences in OS between the high- and low-risk patients in different subgroups. Moreover, the risk score was confirmed as an independent prognostic indicator of BRCA patients and was potentially correlated with immunotherapy response and drug sensitivity. Conclusion: We identified a novel signature integrated of immunoglobulin (IGHA2), glycosylation-related (SLC35A2) and anti-viral gene (BST2) that was an independent prognostic indicator for BRCA patients. The risk-scoring model could be used for predicting prognosis and immunotherapy in BRCA, thus providing a powerful instrument for combating BRCA.

Identification of clinical and molecular features of recurrent serous borderline ovarian tumour

SummaryBackgroundSerous borderline ovarian tumour (SBOT) is the most common type of BOT. Fertility sparing surgery (FSS) is an option for patients with SBOT, though it may increase the risk of recurrence. The clinical and molecular features of its recurrence are important and need to be investigated in detail.MethodsAn internal cohort of 319 patients with SBOT was collected from Aug 1, 2009 to July 31, 2019 from the Obstetrics and Gynecology Hospital of Fudan University in China. An external cohort of 100 patients with SBOT was collected from Aug 1, 2009 to Nov 30, 2019 from the Shandong Provincial Hospital in China. The risk factors for the recurrence were identified by multivariate cox analysis. Several computational methods were tested to establish a prediction tool for recurrence. Whole genome sequencing, RNA-seq, metabolomics and lipidomics were used to understand the molecular characteristics of the recurrence of SBOT.FindingsFive factors were significantly correlated with SBOT recurrence in a Han population: micropapillary pattern, advanced stage, FSS, microinvasion, and lymph node invasion. A random forest-based online recurrence prediction tool was established and validated using an internal cohort and an independent external cohort for patients with SBOT. The multi-omics analysis on the original SBOT samples revealed that recurrence is related to metabolic regulation of immunological suppression.InterpretationOur study identified several important clinical and molecular features of recurrent SBOT. The prediction tool we established could help physicians to estimate the prognosis of patients with SBOT. These findings will contribute to the development of personalised and targeted therapies to improve prognosis.FundingJL was funded by MOST 2020YFA0803600, 2018YFA0801300, NSFC 32071138, and SKLGE-2118 to Jin Li; JY was funded by the Initial Project for Young and Middle-aged Medical Talents of Wuhan City, Hubei Province ([2014] 41); HH was funded by MOST 2019YFA0801900 and 2020YF1402600 to He Huang; JS was funded by NSFC 22,104,080; CG was funded by Natural Science Foundation of Shanghai 20ZR1408800 and NSFC82171633; BL was funded by Natural Science Foundation of Shanghai 19ZR1406800.

Identification of non-small-cell lung cancer subtypes by unsupervised clustering of CT image features with distinct prognoses and gene pathway activities

ObjectiveIntra-tumour heterogeneity exists in non-small-cell lung cancer (NSCLC). Medical imaging can provide a comprehensive and detailed description of the tumour. Moreover, quantitative features can reflect the potentially valuable information of tumours that is neglected by visual diagnosis. To this end, we identified imaging subtypes of NSCLC by computed tomography (CT) images. MethodsWe analysed CT image data from a single-centre discovery cohort (n = 239), an internal validation cohort (n = 80) and an independent multi-centre external validation cohort (n = 63). Quantitative image features were extracted to identify intra-tumour heterogeneity. An unsupervised consensus clustering algorithm was used to identify and validate robust imaging subtypes. Then, we assessed the clinical relevance and prognostic value of these subtypes. Furthermore, differential gene expression profiles with copy number variation (CNV) data were used to identify the underlying biological patterns associated with subtypes and to map molecular activation patterns in dysregulated pathways. ResultsThree distinct imaging subtypes, spherical tumour clusters, irregular large tumour clusters and hollow tumour clusters, were identified and validated in the discovery and validation cohorts. Imaging subtypes showed significant survival differences in all cohorts. Moreover, each imaging subtype had a specific activated pattern in dysregulated pathways that can be targeted. ConclusionThe imaging subtypes provide complimentary prognostic value to establish non-invasive histopathological subtypes with molecular signatures and may help stratify NSCLC patients for targeted diagnosis and treatment. SignificanceThis study identified three imaging NSCLC subtypes that implied different prognosis and dysregulated pathways. These results may contribute to personalized diagnosis and treatment for NSCLC patients.

The Expression Pattern of Ferroptosis-Related Genes in Colon Adenocarcinoma: Highly Correlated to Tumor Microenvironment Characteristics.

In the latest literatures, ferroptosis caused by T cells in cancerous cells provided new insights of improving curative effect of the PD-1/PD-L1 antibody. The microenvironment on which tumor cells develop and survive was also emphasized as its crucial role in tumor occurrence, development, metastasis and immune escape. Thus, the interaction of ferroptosis related genes and tumor microenvironment (TME) was urgently be detected in a comprehensive perspective. We comprehensively evaluated the transcriptional feature of ferroptosis related genes in colon adenocarcinoma (COAD), and systematically associated these ferroptosis subtypes with DNA damage repair (DDR) and TME characteristics. We found two unique patterns of ferroptosis characterized by distinct biological pathways activation. We also demonstrated that FRG score constructed based on ferroptosis subtypes has a significant correlation with prognosis of colon cancer and could act as an independent prognostic biomarker for predicting patients’ survival. The higher immune infiltrating level, immune functional pathways activation was observed in the high FRG score group. Furthermore, these results were verified by an independent external GEO cohort. This work revealed ferroptosis was highly associated with TME complexity and diversity. A novel ferroptosis subtypes related gene scoring system can be used for prognostic prediction in COAD. Targeting ferroptosis may be a therapeutic alternative for COAD.

Deep Learning-Based Classification of Hepatocellular Nodular Lesions on Whole-Slide Histopathologic Images

Hepatocellular nodular lesions (HNLs) constitute a heterogeneous group of disorders. Differential diagnosis among these lesions, especially high-grade dysplastic nodules (HGDNs) and well-differentiated hepatocellular carcinoma (WD-HCC), can be challenging, letalone biopsy specimens. We aimed to develop a deep learning system to solve these puzzles, improving the histopathologic diagnosis of HNLs (WD-HCC, HGDN, low-grade DN, focal nodular hyperplasia, hepatocellular adenoma), and background tissues (nodular cirrhosis, normal liver tissue). The samples consisting of surgical and biopsy specimens were collected from 6 hospitals. Each specimen was reviewed by 2 to 3 subspecialists. Four deep neural networks (ResNet50, InceptionV3, Xception, and the Ensemble) were used. Their performances were evaluated by confusion matrix, receiver operating characteristic curve, classification map, and heat map. The predictive efficiency of the optimal model was further verified by comparing with that of 9 pathologists. We obtained 213,280 patches from 1115 whole-slide images of 738 patients. An optimal model was finally chosen based on F1 score and area under the curve value, named hepatocellular-nodular artificial intelligence model (HnAIM), with the overall 7-category area under the curve of 0.935 in the independent external validation cohort. For biopsy specimens, the agreement rate with subspecialists' majority opinion was higher for HnAIM than 9 pathologists on both patch level and whole-slide images level. We first developed a deep learning diagnostic model for HNLs, which performed well and contributed to enhancing the diagnosis rate of early HCC and risk stratification of patients with HNLs. Furthermore, HnAIM had significant advantages in patch-level recognition, with important diagnostic implications for fragmentary or scarce biopsy specimens.

Neutrophil-to-lymphocyte ratio is a novel predictor of venous thrombosis in polycythemia vera

We investigated the neutrophil-to-lymphocyte ratio (NLR) as a predictor of thrombosis in polycythemia vera (PV). After a median follow-up of 2.51 years, of 1508 PV patients enrolled in the ECLAP study, 82 and 84 developed arterial and venous thrombosis, respectively. Absolute counts of total leukocytes, neutrophils, lymphocytes, platelets, and the NLR were tested by generalized additive models (GAM) to evaluate their trend in continuous scale of thrombotic risk. Only for venous thrombosis, we showed that baseline absolute neutrophil and lymphocyte counts were on average respectively higher (median: 6.8 × 109/L, p = 0.002) and lower (median: 1.4 × 109/L, p = 0.001), leading to increased NLR values (median: 5.1, p = 0.002). In multivariate analysis, the risk of venous thrombosis was independently associated with previous venous events (HR = 5.48, p ≤ 0.001) and NLR values ≥5 (HR = 2.13, p = 0.001). Moreover, the relative risk in both low- and high-standard risk groups was almost doubled in the presence of NLR ≥ 5. These findings were validated in two Italian independent external cohorts (Florence, n = 282 and Rome, n = 175) of contemporary PV patients. Our data support recent experimental work that venous thrombosis is controlled by innate immune cells and highlight that NLR is an inexpensive and easily accessible prognostic biomarker of venous thrombosis.

Integration of Single-Cell RNA Sequencing and Bulk RNA Sequencing Data to Establish and Validate a Prognostic Model for Patients With Lung Adenocarcinoma.

Background: Lung adenocarcinoma (LUAD) remains a lethal disease worldwide, with numerous studies exploring its potential prognostic markers using traditional RNA sequencing (RNA-seq) data. However, it cannot detect the exact cellular and molecular changes in tumor cells. This study aimed to construct a prognostic model for LUAD using single-cell RNA-seq (scRNA-seq) and traditional RNA-seq data. Methods: Bulk RNA-seq data were downloaded from The Cancer Genome Atlas (TCGA) database. LUAD scRNA-seq data were acquired from Gene Expression Omnibus (GEO) database. The uniform manifold approximation and projection (UMAP) was used for dimensionality reduction and cluster identification. Weighted Gene Correlation Network Analysis (WGCNA) was utilized to identify key modules and differentially expressed genes (DEGs). The non-negative Matrix Factorization (NMF) algorithm was used to identify different subtypes based on DEGs. The Cox regression analysis was used to develop the prognostic model. The characteristics of mutation landscape, immune status, and immune checkpoint inhibitors (ICIs) related genes between different risk groups were also investigated. Results: scRNA-seq data of four samples were integrated to identify 13 clusters and 9cell types. After applying differential analysis, NK cells, bladder epithelial cells, and bronchial epithelial cells were identified as significant cell types. Overall, 329 DEGs were selected for prognostic model construction through differential analysis and WGCNA. Besides, NMF identified two clusters based on DEGs in the TCGA cohort, with distinct prognosis and immune characteristics being observed. We developed a prognostic model based on the expression levels of six DEGs. A higher risk score was significantly correlated with poor survival outcomes but was associated with a more frequent TP53 mutation rate, higher tumor mutation burden (TMB), and up-regulation of PD-L1. Two independent external validation cohorts were also adopted to verify our results, with consistent results observed in them. Conclusion: This study constructed and validated a prognostic model for LUAD by integrating 10× scRNA-seq and bulk RNA-seq data. Besides, we observed two distinct subtypes in this population, with different prognosis and immune characteristics.

Prediction of biochemical nonresolution in patients with chronic drug-induced liver injury: A large multicenter study.

To clarify high-risk factors and develop a nomogram model to predict biochemical resolution or biochemical nonresolution (BNR) in patients with chronic DILI. Retrospectively, 3655 of 5326 patients with chronic DILI were enrolled from nine participating hospitals, of whom 2866 underwent liver biopsy. All of these patients were followed up for over 1 year and their clinical characteristics were retrieved from electronic medical records. The endpoint was BNR, defined as alanine aminotransferase or aspartate aminotransferase >1.5× upper limit of normal or alkaline phosphatase >1.1× ULN, at 12 months from chronic DILI diagnosis. The noninvasive high-risk factors for BNR identified by multivariable logistic regression were used to establish a nomogram, which was validated in an independent external cohort. Finally, 19.3% (707 of 3655) patients presented with BNR. Histologically, with the increase in liver inflammation grades and fibrosis stages, the proportion of BNR significantly increased. The risk of BNR was increased by 21.3-fold in patients with significant inflammation compared to none or mild inflammation (p<0.001). Biochemically, aspartate aminotransferase and total bilirubin, platelets, prothrombin time, sex, and age were associated with BNR and incorporated to construct a nomogram model (BNR-6) with a concordance index of 0.824 (95% CI, 0.798-0.849), which was highly consistent with liver histology. These results were successfully validated both in the internal cohort and external cohort. Significant liver inflammation is a robust predictor associated with biochemical nonresolution. The established BNR-6 model provides an easy-to-use approach to assess the outcome of chronic DILI.

Development and Validation of a Three-Gene-Based Prognostic Model for Predicting the Overall Survival of Head and Neck Squamous Cell Carcinoma Through Bioinformatics Analysis.

Head and neck squamous cell carcinoma (HNSCC) is one of the most common cancer worldwide and seriously threats public health safety. Despite the improvement of diagnostic and treatment methods, the overall survival for advanced patients has not improved yet. This study aimed to sort out prognosis-related molecular biomarkers for HNSCC and establish a prognostic model to stratify the risk hazards and predicate the prognosis for these patients, providing a theoretical basis for the formulation of individual treatment plans. We firstly identified differentially expressed genes (DEGs) between HNSCC tissues and normal tissues via joint analysis based on GEO databases. Then a total of 11 hub genes were selected for single-gene prognostic analysis to identify the prognostic genes. Later, the clinical information and transcription information of HNSCC were downloaded from the TCGA database. With the application of least absolute shrinkage and selection operator (LASSO) algorithm analyses for the prognostic genes on the TCGA cohort, a prognostic model consisting of three genes (COL4A1, PLAU and ITGA5) was successfully established and the survival analyses showed that the prognostic model possessed a robust performance in the overall survival prediction. Afterward, the univariate and multivariate regression analysis indicated that the prognostic model could be an independent prognostic factor. Finally, the predicative efficiency of this model was well confirmed in an independent external HNSCC cohort.

Predictive model to identify multiple failure to biological therapy in patients with rheumatoid arthritis.

Background:Despite advances in the treatment of rheumatoid arthritis (RA) and the wide range of therapies available, there is a percentage of patients whose treatment presents a challenge for clinicians due to lack of response to multiple biologic and target-specific disease-modifying antirheumatic drugs (b/tsDMARDs).Objective:To develop and validate an algorithm to predict multiple failure to biological therapy in patients with RA.Design:Observational retrospective study involving subjects from a cohort of patients with RA receiving b/tsDMARDs.Methods:Based on the number of prior failures to b/tsDMARDs, patients were classified as either multi-refractory (MR) or non-refractory (NR). Patient characteristics were considered in the statistical analysis to design the predictive model, selecting those variables with a predictive capability. A decision algorithm known as ‘classification and regression tree’ (CART) was developed to create a prediction model of multi-drug resistance. Performance of the prediction algorithm was evaluated in an external independent cohort using area under the curve (AUC).Results:A total of 136 patients were included: 51 MR and 85 NR. The CART model was able to predict multiple failures to b/tsDMARDs using disease activity score-28 (DAS-28) values at 6 months after the start time of the initial b/tsDMARD, as well as DAS-28 improvement in the first 6 months and baseline DAS-28. The CART model showed a capability to correctly classify 94.1% NR and 87.5% MR patients with a sensitivity = 0.88, a specificity = 0.94, and an AUC = 0.89 (95% CI: 0.74–1.00). In the external validation cohort, 35 MR and 47 NR patients were included. The AUC value for the CART model in this cohort was 0.82 (95% CI: 0.73–0.9).Conclusion:Our model correctly classified NR and MR patients based on simple measurements available in routine clinical practice, which provides the possibility to characterize and individualize patient treatments during early stages.

Development and Validation of a Nomogram Incorporating Colloid Osmotic Pressure for Predicting Mortality in Critically Ill Neurological Patients.

Backgrounds: The plasma colloid osmotic pressure (COP) values for predicting mortality are not well-estimated. A user-friendly nomogram could predict mortality by incorporating clinical factors and scoring systems to facilitate physicians modify decision-making when caring for patients with serious neurological conditions.Methods: Patients were prospectively recruited from March 2017 to September 2018 from a tertiary hospital to establish the development cohort for the internal test of the nomogram, while patients recruited from October 2018 to June 2019 from another tertiary hospital prospectively constituted the validation cohort for the external validation of the nomogram. A multivariate logistic regression analysis was performed in the development cohort using a backward stepwise method to determine the best-fit model for the nomogram. The nomogram was subsequently validated in an independent external validation cohort for discrimination and calibration. A decision-curve analysis was also performed to evaluate the net benefit of the insertion decision using the nomogram.Results: A total of 280 patients were enrolled in the development cohort, of whom 42 (15.0%) died, whereas 237 patients were enrolled in the validation cohort, of which 43 (18.1%) died. COP, neurological pathogenesis and Acute Physiology and Chronic Health Evaluation II (APACHE II) score were predictors in the prediction nomogram. The derived cohort demonstrated good discriminative ability, and the area under the receiver operating characteristic curve (AUC) was 0.895 [95% confidence interval (CI), 0.840–0.951], showing good correction ability. The application of this nomogram to the validation cohort also provided good discrimination, with an AUC of 0.934 (95% CI, 0.892–0.976) and good calibration. The decision-curve analysis of this nomogram showed a better net benefit.Conclusions : A prediction nomogram incorporating COP, neurological pathogenesis and APACHE II score could be convenient in predicting mortality for critically ill neurological patients.

Utilizing phenotypic characteristics of metastatic brain tumors to improve the probability of detecting circulating tumor DNA from cerebrospinal fluid in non-small-cell lung cancer patients: development and validation of a prediction model in a prospective cohort study

BackgroundCirculating tumor DNA (ctDNA) in cerebrospinal fluid (CSF) has become a promising surrogate for genomic profiling of central nervous system tumors. However, suboptimal ctDNA detection rates from CSF limit its clinical utility. Thus precise screening of suitable patients is needed to maximize the clinical benefit.Patients and methodsBetween February 2017 and December 2020, 66 newly diagnosed non-small-cell lung cancer (NSCLC) patients with brain parenchymal metastases were prospectively enrolled as a training cohort and 30 additional patients were enrolled as an external validation cohort. CSF samples and matched primary tumor tissues were collected before treatment and subjected to next-generation sequencing (NGS). The imageological characteristics of patients’ brain tumors were evaluated by radiologists using enhanced magnetic resonance imaging images. The clinical and imageological characteristics were evaluated by complete subsets regression, Akaike information criteria, and Bayesian information criteria methods to establish the prediction model. A nomogram was then built for CSF ctDNA detection prediction.ResultsThe somatic mutation detection rate of genes covered by our targeted NGS panel was significantly lower in CSF ctDNA (59.09%) than tumor tissue (91.84%). The Tsize (diameter of the largest intracranial lesion) and LVDmin (minimum lesion–ventricle distance for all intracranial lesions) were significantly associated with positive CSF ctDNA detection, and thus, were selected to establish the prediction model, which achieved an area under the ROC curve (AUC) of 0.819 and an accuracy of 0.800. The model’s predictive ability was further validated in the independent external cohort (AUC of 0.772, accuracy of 0.767) and by internal cross-validation. The CSF ctDNA detection rate was significantly improved from 58.18% (32/55) to 81.81% (27/33) in patients after model selection (P = 0.022).ConclusionsThis study developed a regression model to predict the probability of detecting CSF ctDNA using the phenotypic characteristics of metastatic brain lesions in NSCLC patients, thus, maximizing the benefits of CSF liquid biopsies.

A Novel Prognostic and Predictive Signature for Lung Adenocarcinoma Derived from Combined Hypoxia and Infiltrating Immune Cell-Related Genes in TCGA Patients.

BackgroundThe hypoxia and immune status of the lung adenocarcinoma (LUAD) microenvironment appear to have combined impacts on prognosis. Therefore, deriving a prognostic signature by integrating hypoxia- and immune infiltrating cell-related genes (H&IICRGs) may add value over prognostic indices derived from genes driving either process alone.MethodsDifferentially expressed H&IICRGs (DE-H&IICRGs) were identified in The Cancer Genome Atlas transcriptomic data using limma, CIBERSORT, weighted gene co-expression network analysis, and intersection analysis. A stepwise Cox regression model was constructed to identify prognostic genes and to produce a gene signature based on DE-H&IICRGs. The potential biological functions associated with the gene signature were explored using functional enrichment analysis. The prognostic signature was externally validated in a separate cohort from the Gene Expression Omnibus database.ResultsFive prognostic genes associated with overall survival in LUAD were used in the DE-H&IICRG-based prognostic signature. Patients in the high-risk group had an inferior prognosis, which was validated in an independent external cohort, and had lower expression of most immune checkpoint genes. In multivariate analysis, only risk score and T stage were independent prognostic factors. Differentially expressed genes (DEGs) associated with the risk score were enriched for pathways related to cell cycle, hypoxia regulation, and immune response. TIDE analyses showed that low-risk LUAD patients might also respond better to immunotherapy.ConclusionThis study establishes and validates a prognostic profile for LUAD patients that combines hypoxia and immune infiltrating cell-related genes. This signature may have clinical application both for prognostication and guiding individualized immunotherapy.

CSF protein panels reflecting multiple pathophysiological mechanisms for early and specific diagnosis of Alzheimer’s disease

AbstractBackgroundThe development of disease‐modifying therapies against Alzheimer’s disease (AD) requires biomarker panels that reflect the diverse pathological pathways specifically involved in AD. Here we aimed to identify and validate panels of cerebrospinal fluid (CSF) proteins covering different molecular pathways for early and specific diagnosis of AD.MethodWe measured 665 proteins in 797 CSF samples from patients with mild cognitive impairment with abnormal amyloid (MCI(Aβ+): n=50), AD‐dementia (n=230), non‐AD dementias (n=322; 123 DLB and 199 FTD) and cognitively‐unimpaired controls (n=195; classical AD CSF biomarkers negative) using proximity ligation‐based multiplex immunoassays. Nested and penalized linear modeling were used to identify protein differences (q&lt;0,05) and translatable classification signatures.ResultWe detected highly dysregulated CSF proteins in MCI(Aβ+) or AD compared to controls (112 and 288 proteins respectively, lowest q:1‐15 and 1‐23), as well as between AD and non‐AD dementias (469 proteins; lowest q:1‐29). We confirmed previous findings (e.g., DDAH1, ENO2, PARK7), but we also identified novel proteins especially associated to the prodromal and/or dementia AD stages (e.g., ABL1, SCD4, ENTPD5). Proteins dysregulated in MCI(Aβ+) were primarily related to oxidative stress and energy metabolism, while those specifically dysregulated in later stages of AD dementia were related to cell remodeling, vascular function and immune system. Using penalised generalised linear modeling we identified the minimal number of markers with maximal power to discriminate clinical groups: for MCI(Aβ+) vs. controls: 10‐CSF proteins, AUC:0.99 (95%CI:0.97‐1); for AD vs. controls: 8‐CSF proteins, AUC:0.95 (95%CI:0.92‐0.99) and for AD vs. non‐AD dementias: 9‐CSF proteins, AUC:0.87 (95%CI:0.81‐0.93). The CSF panel discriminating AD from cognitively unimpaired controls was validated in an independent external cohort (n=62, AUC:0.94). Only 3 proteins overlapped across panels, suggesting that the panels with the strongest discriminative power are specific for disease stage and clinical group.ConclusionThis unprecedented large and protein‐rich CSF study indicates that CSF proteome follows a highly dynamic trajectory with distinct CSF profiles associated to different biological processes along the AD continuum. We unveil optimal CSF biomarker panels reflecting the specific multifactorial nature of AD, which can be now translated into customised assays for widespread validation and potential use in heterogeneous diagnostic settings or clinical trials.

Association of Tumor-Associated Collagen Signature With Prognosis and Adjuvant Chemotherapy Benefits in Patients With Gastric Cancer

The current TNM staging system provides limited information for prognosis prediction and adjuvant chemotherapy benefits for patients with gastric cancer (GC). To develop a tumor-associated collagen signature of GC (TACSGC) in the tumor microenvironment to predict prognosis and adjuvant chemotherapy benefits in patients with GC. This retrospective cohort study included a training cohort of 294 consecutive patients treated between January 1, 2012, and December 31, 2013, from Nanfang Hospital, Southern Medical University, People's Republic of China, and a validation cohort of 225 consecutive patients treated between October 1, 2010, and December 31, 2012, from Fujian Provincial Cancer Hospital, Fujian Medical University, People's Republic of China. In total, 146 collagen features in the tumor microenvironment were extracted with multiphoton imaging. A TACSGC was then constructed using the least absolute shrinkage and selection operator Cox proportional hazards regression model in the training cohort. Data analysis was conducted from October 1, 2020, to April 30, 2021. The association of TACSGC with disease-free survival (DFS) and overall survival (OS) was assessed. An independent external cohort was included to validate the results. Interactions between TACSGC and adjuvant chemotherapy were calculated. This study included 519 patients (median age, 57 years [IQR, 49-65 years]; 360 [69.4%] male). A 9 feature-based TACSGC was built. A higher TACSGC level was significantly associated with worse DFS and OS in both the training (DFS: hazard ratio [HR], 3.57 [95% CI, 2.45-5.20]; OS: HR, 3.54 [95% CI, 2.41-5.20]) and validation (DFS: HR, 3.10 [95% CI, 2.26-4.27]; OS: HR, 3.24 [95% CI, 2.33-4.50]) cohorts (continuous variable, P < .001 for all comparisons). Multivariable analyses found that carbohydrate antigen 19-9, depth of invasion, lymph node metastasis, distant metastasis, and TACSGC were independent prognostic predictors of GC, and 2 integrated nomograms that included the 5 predictors were established for predicting DFS and OS. Compared with clinicopathological models that included only the 4 clinicopathological predictors, the integrated nomograms yielded an improved discrimination for prognosis prediction in a C index comparison (training cohort: DFS, 0.80 [95% CI, 0.73-0.88] vs 0.78 [95% CI, 0.71-0.85], P = .03; OS, 0.81 [95% CI, 0.75-0.88] vs 0.80 [95% CI, 0.73-0.86], P = .03; validation cohort: DFS, 0.78 [95% CI, 0.70-0.87] vs 0.76 [95% CI, 0.67-0.84], P = .006; OS, 0.78 [95% CI, 0.69-0.86] vs 0.75 [95% CI, 0.67-0.84], P = .002). Patients with stage II and III GC and low TACSGC levels rather than high TACSGC levels had a favorable response to adjuvant chemotherapy (DFS: HR, 0.65 [95% CI, 0.43-0.96]; P = .03; OS: HR, 0.55 [95% CI, 0.36-0.82]; P = .004; dichotomized variable, P < .001 for interaction for both comparisons). The findings suggest that TACSGC provides additional prognostic information for patients with GC and may distinguish patients with stage II and III disease who are more likely to derive benefits from adjuvant chemotherapy.

Deep learning predicts epidermal growth factor receptor mutation subtypes in lung adenocarcinoma.

This study aimed to explore the predictive ability of deep learning (DL) for the common epidermal growth factor receptor (EGFR) mutation subtypes in patients with lung adenocarcinoma. A total of 665 patients with lung adenocarcinoma (528/137) were recruited from two different institutions. In the training set, an 18-layer convolutional neural network (CNN) and fivefold cross-validation strategy were used to establish a CNN model. Subsequently, an independent external validation cohort from the other institution was used to evaluate the predictive efficacy of the CNN model. Grad-weighted class activation mapping (Grad-CAM) technology was used for the visual interpretation of the CNN model. In addition, this study also compared the prediction abilities of the radiomics and CNN models. Receiver operating characteristic (ROC) curves, accuracy and precision values, and recall and F1-score were used to evaluate the effectiveness of the CNN model and compare its performance with that of the radiomics model. In the validation set, the micro- and macroaverage values of the area under the ROC curve of the CNN model to identify the three EGFR subtypes were 0.78 and 0.79, respectively. All evaluation indicators of the CNN model were better than those of the radiomics model. Our study confirmed the potential of DL for predicting the EGFR mutation status in lung adenocarcinoma. The imaging phenotypes of the three mutation subtypes were found to be different, which can provide a basis for choosing more accurate and personalized treatment in patients with lung adenocarcinoma.

Development and External Validation of a Prognostic Nomogram for Angioimmunoblastic T-Cell Lymphoma

Introduction: Heterogeneity exists in prognosis of Angioimmunoblastic T-cell lymphoma (AITL) patients. Thus, a personalized prognostic model is crucial for survival prediction for each AITL patient. Nomogram is a powerful mathematical tool to predict survival. In this study, we aimed to develop a prognostic nomogram for AITL based on data from a large clinical database and validate it in an independent external cohort. In addition, we compared the accuracy of the nomogram with previous prognostic models used in AITL including International Prognostic Index (IPI) and Prognostic Index of T-cell lymphoma (PIT) model.Methods: Totally, 1071 patients were retrieved from the Surveillance, Epidemiology, and End Results (SEER) database as the training cohort; 156 patients diagnosed from 2009-2021 in Sun Yat-Sen University Cancer Center or The First Affiliated Hospital of Guangzhou Medical University were recruited as the external validation cohort. 105 patients with IPI information in the training cohort were used to compare the nomogram and IPI. 156 patients in the external cohort were used to compare the nomogram and IPI or PIT. The Prognostic risk factors in the nomogram were identified by cox proportional hazards model. Concordance index (C-index) and calibration curves were used for internal and external validation. C-index and decision curve analysis (DCA) curves were used to compare the models.Results: Age, sex, systematic symptoms, Ann Arbor stage and chemotherapy were risk factors finally included to develop the nomogram. C-indexes of the nomogram were 0.676 and 0.652 in the training cohort and the validation cohort. Favorable agreement of nomogram-predicted and actual probability of overall survival (OS) was detected by calibration curves in both training and validation cohorts. In the cohort of 105 patients, C-indexes of the nomogram and IPI were 0.696 vs 0.616 (P<0.05); in the cohort of 156 patients, C-indexes were 0.652 vs 0.597 (P=0.08) of the nomogram and IPI while 0.652 vs 0.616 (P=0.176) of the nomogram and PIT. Decision curve analysis (DCA) showed superiority of the nomogram as compared with the IPI or the PIT model in both 105 and 156 patients' cohorts. A web calculator was published for convenient clinical application. Based on the prognostic scores calculated by the nomogram, three cut points were identified by X-tile program to establish a classification system that could significantly distinguish patients in four risk groups.Conclusion: We establish a nomogram for survival prediction of AITL, which may assist in treatment strategy making and survival consultation. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Nutrition-Incorporated Pink-E Predicts Survival of Extranodal NK/T-Cell Lymphoma

▪Introduction: Patients with aggressive lymphomas are at high risk of losing body resources, resulting in malnutrition, immunodeficiency, impaired life quality and inferior outcomes. Nutritional status is closely associated with the development and treatment arrangements of malignancies, but often neglected in the prognostic assessment. Thus, this study aimed to explore the prognostic significance of nutrition-related parameters in extranodal NK/T-cell lymphoma (ENKTL), an aggressive lymphoma with dismal outcomes.Methods: We retrospectively analyzed patients diagnosed with ENKTL between 2011 and 2020 from Shandong Provincial Hospital and Affiliated Hospital of Qingdao University. The observation endpoints were overall survival (OS) and progression-free survival (PFS). Univariate (UVA) and multivariate (MVA) Cox regression analyses were conducted to examine the significance of variables at diagnosis on OS and PFS. A reformative nutrition assessment-incorporated score system was constructed based on the MVA results. Its performance was tested in the training and validation cohort from multiple aspects, including calibration, discrimination and clinical utility.Results: A total of 160 patients with a median age of 50 (39, 58.75) years and male predominance (67.5%) were included. The prognostic significance of two emerging nutritional indexes, prognostic nutritional index (PNI) and controlling nutritional status (CONUT) score, was assessed. UVA analysis showed PNI was a meaningful marker to OS (HR 2.275, P=0.014) but not to PFS (HR 1.653, P=0.089) while CONUT score was a significant predictor not only to OS (HR 29.385, P＜0.001) but to PFS (HR 12.516, P＜0.001). MVA analysis further verified that CONUT score could independently predict OS (HR 10.247, P=0.001) and PFS (HR 5.587, P=0.001) in addition to prognostic index of natural killer lymphoma plus EBV (PINK-E), another independent marker (PINK-E=2, HR 3.842, P=0.034, PINK-E=3/4/5, HR 9.185, P=0.006 for OS; PINK-E=2, HR 2.308, P=0.06, PINK-E=3/4/5, HR 4.535, P=0.004 for PFS) (Figure 1).Based on this finding, a modified prognosis scoring tool, CONUT-PINK-E, was developed from the training cohort (TC) and tested in the independent external validation cohort (VC). The novel model contains six factors, age≥60 years old, Ann Arbor III/IV stage, distant lymph node involvement, non-nasal type, detectable EBV-DNA in blood and CONUT score≥5 (moderate and severe malnutrition). CONUT-PINK-E differentiated patients into low-, intermediate- and high-risk grades according to the imparities of survival probability revealed by Kaplan-Meier curves. Significant survival differences existed among three risk grades that the median OS was 74, 36, 5 months and median PFS was 31, 14, 4 months, respectively, for the low-, intermediate-, high-risk group (P＜0.001) in TC and the corresponding data were 72, 26, 7 months and 35, 14, 4 months in VC (P＜0.001, Figure 2).Subsequently, the superiority of CONUT-PINK-E was examined by comparison with international prognostic index (IPI), Korean Prognostic Index (KPI), Prognostic index of natural killer lymphoma (PINK) and PINK-E. The 1- to 5-year time-dependent areas under receiver operating characteristic curve (AUCs) hinted that CONUT-PINK-E (TC: 0.832-0.961, VC: 0.773-0.937) possessed higher discrimination than IPI (TC: 0.710-0.785, P<0.001; VC: 0.646-0.840, P<0.001), KPI (TC: 0.659-0.826, P<0.001; VC: 0.623-0.691, P<0.001), PINK (TC: 0.802-0.865, P<0.001; VC: 0.663-0.842, P<0.001) and PINK-E(TC: 0.815-0.902, P<0.001; VC: 0.741-0.894, P<0.001) (Figure 3A, 3B). Decision curve analysis plainly showed CONUT-PINK-E brought higher clinical net benefits than IPI, KPI, PINK and PINK-E in forecasting OS and PFS of ENKTL (Figure 3C, 3D). In addition, net reclassification index (NRI) and integrated discrimination improvement (IDI) conformably demonstrated that CONUT-PINK-E achieved significant improvements than the prior stratification tools.Conclusion: This is the first time that CONUT score was recognized capable of acting as an independent prognostic marker in newly-diagnosed ENKTL patients. More importantly, the first risk score covering nutritional status assessment for ENKTL, CONUT-PINK-E, was proposed and presented striking performance. The newly-built model could deliver vital reference to the risk stratification and comprehensive managements of ENKTL patients. [Display omitted] DisclosuresNo relevant conflicts of interest to declare.

Neutrophil-to-Lymphocyte Ratio (NLR) Is a Risk Factor for Venous Thrombosis in Polycythemia Vera

▪Background. The tendency towards thrombosis in myeloproliferative neoplasms (MPNs) is linked to the JAK2-mutant clone which leads to hypercellularity and functional interplay between abnormal erythrocytes, platelets, leukocytes and dysfunctional endothelium. The resulting cell activation that also involves stromal cells in their microenvironment, has been shown associated with chronic, systemic, subclinical pro-inflammatory state, and has been implicated in the pathogenesis of thrombosis in MPN. Neutrophil to lymphocyte ratio (NLR) is a novel inflammatory marker found to be associated with the severity and prognosis of cardiovascular diseases but there is little evidence for its prognostic significance in MPN. We investigated whether NLR could predict the onset of arterial and venous thrombosis in polycythemia vera (PV).Methods. Subjects of this study were 1508 patients included in the ECLAP trail with NLR evaluation available. In addition to standard statistical methods, we used proportional hazards additive models (GAM) as they can provide an excellent fit in the presence of nonlinear relationships, for the prediction of total, arterial and venous thrombosis as smooth functions with cubic splines of different blood parameters.Results. After a median follow-up of 2.76 years, 169 thrombotic events (10.3%) were objectively diagnosed and analyzed: 87 arterial (MI, Stroke, TIA, PAT) and 88 venous thrombosis (DVT±PE, superficial thrombophlebitis). •In univariate analysis, arterial thrombosis was associated with age (p=0.003) and previous thrombosis, especially if arterial (p<0.001), and with the presence of at least one cardiovascular risk factor (p=0.009). No association between baseline platelet and leukocyte counts and NLR with arterial events was found. Conversely, in venous thrombosis, beside age (p=0.039) and history of venous thrombosis (p<0.001), additional risk factors were low hemoglobin (p=0.039) and increasing values of NLR (p=0.002) resulting from a simultaneous increase of neutrophils (p=0.002) and to a decrease of absolute number lymphocytes (p=0.002).•To predict outcomes, we tested total leukocytes, neutrophils, lymphocytes, platelets and NLR by GAM to evaluate their trend in continuous scale of thrombotic risk. A significant association between the risk of venous thrombosis and the progressive lower counts of lymphocytes (p=0.002) emerged, leading to increase the NLR values (p=0.005). However, given the greater precision of the NLR estimates and the markedly wide confidence interval of lymphocyte counts estimates, NLR was used in multivariate model. Conversely, neither absolute values of neutrophils and lymphocytes nor NLR were found associated with arterial events.•In multivariate model, adjusted for age, gender and treatments at baseline, the risk of venous thrombosis was independently associated with previous venous events (HR=5.48, p ≤0.001) and NLR values ≥5 -the best cut-off resulted applying the Liu's method (HR=2.13, p=0.001). NLR effect for total venous thrombosis was not modified by excluding superficial venous thrombosis (HR=2.90, p=0.001) in a sensitivity analysis. Older age (i.e., ≥65 years, HR=1.88, p=0.005) and previous arterial thrombosis (HR=1.92, p=0.003) retained the prognostic statistical significance for arterial thrombosis.•Our learning cohort findings, indicating a correlation between NLR values and venous thrombosis, have been validated in two Italian independent external cohorts (Florence, n=282 and Rome, n=173) of contemporary PV patients (Figure 1).Conclusions. The NLR is an inexpensive and easily accessible test compared to other inflammatory markers. We found that NLR-alone is an independent predictor of venous thrombosis and could be included in a new scoring system. In addition to guiding the stratification of thrombotic risk, these data confirm that inflammation is a relevant target of antithrombotic therapy in PV. [Display omitted] DisclosuresVannucchi: Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees; BMS: Honoraria, Membership on an entity's Board of Directors or advisory committees; Incyte: Honoraria, Membership on an entity's Board of Directors or advisory committees; AbbVie: Membership on an entity's Board of Directors or advisory committees. Barbui: AOP Orphan: Membership on an entity's Board of Directors or advisory committees, Research Funding; Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding.

Transcription Factor Signatures May Predict the Prognosis and Status of the Immune Microenvironment of Primary Lower-Grade Gliomas.

AimTranscription factor (TF) in glioma, including proliferation, invasion/migration, and tumor microenvironment, has been receiving increasing attention. However, there are still no systematical analyses based on global TF. Herein, using global TF target gene sets, we comprehensively investigated their relationship with prognosis and potential biological effect in lower-grade glioma (LGG). We aimed to develop a less-biased prognostic model and provide new insight for personalized management of this disease.MethodsTF target gene sets were collected from MSigDB and GRID database followed by ssGSEA calculating normalized enrichment score. Comprehensive survival analysis was combined with Kaplan–Meier and Cox algorithms. Consensus cluster and lasso regression were performed to develop prognostic signatures with validation of ROC and independent external cohort. Approaches of xCell/CIBERSORT/TIMER were involved in analyzing the immune microenvironment. We also correlated identified prognostic signatures with tumor mutational burden (TMB) and m6A genes.ResultsFourteen TFs were significantly screened based on survival. Patients were classified into 2 prognosis-related clusters based on 14-TFs features. The function of differentially expressed TF target genes between Cluster1/2 was enriched mostly on glioma invasion/migration. The prognostic model was trained by 6 out of 14-TFs followed by generating risk-score as an independent prognostic indicator. We found differences between the high/low-risk group in TMB and the immune microenvironment, where the high-risk group represented “hot-tumor”. Besides, 6-TFs were correlated with m6A regulation genes.ConclusionOur findings suggested that the 6-TFs model could be used to predict prognosis and predict the status of the immune microenvironment in LGG.