Ferroptosis-related gene signature as a prognostic marker for lower-grade gliomas

Ferroptosis is a newly discovered form of programmed cell death, which has unique biological effects on metabolism and redox biology. In this study, the prognostic value of ferroptosis-related genes was investigated in lower-grade gliomas (LGG). We downloaded the ferroptosis-related genes from the FerrDb dataset. Univariate Cox and LASSO regression analyses were applied to identify genes correlated with overall survival (OS). Subsequently, 12 ferroptosis-related genes were screened to establish the prognostic signature using stepwise multivariate Cox regression. According to the median value of risk scores, patients were divided into low- and high-risk subgroups. The Kaplan-Meier curves showed the high-risk group had a lower OS. The predictive power of the risk model was validated using the CGGA. Functional analysis revealed that the terms associated with plasma membrane receptor complex, immune response and glutamate metabolic process were primarily related to the risk model. Moreover, we established a nomogram that had a strong forecasting ability for the 1-, 3- and 5-year OS. In addition, we compared the risk scores between different clinical features. We also detected infiltration of macrophages and monocytes in different subgroups. Overall, our study identified the prognostic signature of 12 ferroptosis-related genes, which has the potential to predict the prognosis of LGG.     

A risk signature-based on metastasis-associated genes to predict survival of patients with osteosarcoma

Osteosarcoma (OS) is the most common primary solid malignant bone tumor, and its metastasis is a prominent cause of high mortality in patients. In this study, a prognosis risk signature was constructed based on metastasis-associated genes. Four microarrays datasets with clinical information were downloaded from Gene Expression Omnibus, and 256 metastasis-associated genes were identified by limma package. Further, a protein-protein interaction network was constructed, and survival analysis was performed using data from the Therapeutically Applicable Research to Generate Effective Treatments data matrix, identifying 19 genes correlated with prognosis. Six genes were selected by the least absolute shrinkage and selection operator regression for multivariate cox analysis. Finally, a three-gene (MYC, CPE, and LY86) risk signature was constructed, and datasets GSE21257 and GSE16091 were used to validate the prediction efficiency of the signature. The survival times of low- and high-risk groups were significantly different in the training set and validation set. Additionally, gene set enrichment analysis revealed that the genes in the signature may affect the cell cycle, gap junctions, and interleukin-6 production. Therefore, the three-gene survival risk signature could potentially predict the prognosis of patients with OS. Further, proteins encoded by CPE and LY86 may provide novel insights into the prediction of OS prognosis and therapeutic targets.     

Prognostic Value of E-cadherin-, CD44-, and MSH2-associated Nomograms in Patients With Stage II and III Colorectal Cancer

BACKGROUND: To evaluate the prognostic value of E-cadherin, CD44, and MSH2 expression for colorectal cancer (CRC) and construct nomograms that can predict prognosis. METHODS: We retrospectively analyzed the expression of E-cadherin, CD44, and MSH2 in 223 paraffin-embedded stage II and III CRC specimens using immunohistochemistry in the training cohort. Their prognostic values were assessed using Kaplan–Meier curves and univariate and multivariate COX regression models. Moreover, a number of risk factors were used to form nomograms to evaluate survival, and Harrell's concordance index (C-index) was used to evaluate the predictive accuracy. Further validation of the nomograms was performed in an independent cohort of 115 cases. RESULTS: Low E-cadherin expression and low CD44 expression were significantly associated with diminished overall survival (OS) and disease-free survival (DFS) in stage II and III CRC patients and patients with negative MSH2 expression had better clinical outcomes. Moreover, the multivariate COX analysis identified E-cadherin, CD44 and MSH2 expression as independent prognostic factors for DFS and OS. Using these three markers and three clinicopathological risk variables, two nomograms were constructed and externally validated for predicting OS and DFS (C-index: training cohort, 0.779 (95% CI 0.722–0.835) and 0.771 (0.720–0.822), respectively; validation cohort, 0.773 (0.709–0.837) and 0.670 (0.594–0.747), respectively). CONCLUSION: The expression levels of E-cadherin, CD44 and MSH2 were independent prognostic factors for stage II and III CRC patients. By incorporating clinicopathological features and these biomarkers, we have established two nomograms that could be used to make individualized predictions for OS and DFS.     

内质网应激相关基因能预测骨肉瘤预后并与肿瘤免疫微环境相关

以内质网应激相关基因构建骨肉瘤患者的风险模型,探索其与肿瘤免疫微环境的关系。采用生物信息学分析法,训练集的转录组数据及临床数据下载于UCSC Xena数据库,验证集的相应数据下载于GEO数据库(GSE21257,GSE39058)。采用单因素COX回归分析、LASSO回归分析及多因素COX回归分析提取风险特征基因构建风险模型,使用决策曲线分析、受试者工作特征曲线分析验证模型的准确性,随后构建列线图进一步预测骨肉瘤患者预后;根据风险评分将患者分为高、低风险组,使用Kaplan-Meier生存曲线评估高、低风险组间的生存差异,对差异表达基因(Differentially expressed genes, DEGs)进行GO/KEGG联合富集分析、基因集富集分析(Gene set enrichment analysis, GSEA)及基因集变异分析(Gene set variation analysis, GSVA);采用ESTIMATE算法、微环境种群计数器(Microenvironment cell population counter, MCP counter)方法、单样本基因集富集分析(Single sample gene set enrichment analysis, ssGSEA)进行免疫分析;最终在验证集中验证上述结果。6个风险特征基因中VEGFA、PTGIS及SERPINH1与骨肉瘤患者的不良预后相关,而TMED10、MAPK10及TOR1B与与骨肉瘤患者的良好预后相关,高、低风险组患者之间具有显著生存差异;GO/KEGG联合富集分析、GSVA、GSEA结果表明DEGs与免疫状态相关;免疫分析显示高风险组具有更低的免疫评分及免疫景观;列线图进一步准确地预测了骨肉瘤患者的预后。内质网应激相关基因构建的风险模型能准确预测骨肉瘤患者预后,并与肿瘤免疫微环境相关。     

Integrating Genetic,Neuropsychological and Neuroimaging Data to Model Early-Onset Obsessive Compulsive Disorder Severity

We propose an integrative approach that combines structural magnetic resonance imaging data (MRI), diffusion tensor imaging data (DTI), neuropsychological data, and genetic data to predict early-onset obsessive compulsive disorder (OCD) severity. From a cohort of 87 patients, 56 with complete information were used in the present analysis. First, we performed a multivariate genetic association analysis of OCD severity with 266 genetic polymorphisms. This association analysis was used to select and prioritize the SNPs that would be included in the model. Second, we split the sample into a training set (N = 38) and a validation set (N = 18). Third, entropy-based measures of information gain were used for feature selection with the training subset. Fourth, the selected features were fed into two supervised methods of class prediction based on machine learning, using the leave-one-out procedure with the training set. Finally, the resulting model was validated with the validation set. Nine variables were used for the creation of the OCD severity predictor, including six genetic polymorphisms and three variables from the neuropsychological data. The developed model classified child and adolescent patients with OCD by disease severity with an accuracy of 0.90 in the testing set and 0.70 in the validation sample. Above its clinical applicability, the combination of particular neuropsychological, neuroimaging, and genetic characteristics could enhance our understanding of the neurobiological basis of the disorder.     

Development and Validation of a Nomogram for Predicting Overall Survival in Pancreatic NeuroendocrineTumors

BACKGROUND: The objective of current study was to develop and validate a nomogram to predict overall survival in pancreatic neuroendocrine tumors (PNETs). METHODS: The Surveillance, Epidemiology, and End Results (SEER) database was queried for patients with PNETs between 2004 and 2015. Patients were randomly separated into the training set and the validation set. Cox regression model was used in training set to obtain independent prognostic factors to develop a nomogram for predicting overall survival (OS). The discrimination and calibration plots were used to evaluate the predictive accuracy of the nomogram. RESULTS: A total of 3142 patients with PNETs were collected from the SEER database. Sex, age, marital status, primary site, TNM stage, tumor grade, and therapy were associated with OS in the multivariate models. A nomogram was constructed based on these variables. The nomogram for predicting OS displayed better discrimination power than the Tumor-Node-Metastasis (TNM) stage systems 7th edition in the training set and validation set. The calibration curve indicated that the nomogram was able to accurately predict 3- and 5-year OS. CONCLUSIONS: The nomogram which could predict 3- and 5-year OS were established in this study. Our nomogram showed a good performance, suggesting that it could be served as an effective tool for prognostic evaluation of patients with PNETs.     

Predicting outcomes of prostate cancer immunotherapy by personalized mathematical models

Background

Therapeutic vaccination against disseminated prostate cancer (PCa) is partially effective in some PCa patients. We hypothesized that the efficacy of treatment will be enhanced by individualized vaccination regimens tailored by simple mathematical models.

Methodology/Principal Findings

We developed a general mathematical model encompassing the basic interactions of a vaccine, immune system and PCa cells, and validated it by the results of a clinical trial testing an allogeneic PCa whole-cell vaccine. For model validation in the absence of any other pertinent marker, we used the clinically measured changes in prostate-specific antigen (PSA) levels as a correlate of tumor burden. Up to 26 PSA levels measured per patient were divided into each patient''s training set and his validation set. The training set, used for model personalization, contained the patient''s initial sequence of PSA levels; the validation set contained his subsequent PSA data points. Personalized models were simulated to predict changes in tumor burden and PSA levels and predictions were compared to the validation set. The model accurately predicted PSA levels over the entire measured period in 12 of the 15 vaccination-responsive patients (the coefficient of determination between the predicted and observed PSA values was R ² = 0.972). The model could not account for the inconsistent changes in PSA levels in 3 of the 15 responsive patients at the end of treatment. Each validated personalized model was simulated under many hypothetical immunotherapy protocols to suggest alternative vaccination regimens. Personalized regimens predicted to enhance the effects of therapy differed among the patients.

Conclusions/Significance

Using a few initial measurements, we constructed robust patient-specific models of PCa immunotherapy, which were retrospectively validated by clinical trial results. Our results emphasize the potential value and feasibility of individualized model-suggested immunotherapy protocols.     

A prognostic 11 long noncoding RNA expression signature for breast invasive carcinoma

Breast cancer, the most common cancer in women worldwide, is associated with high mortality. The long non-coding RNAs (lncRNAs) with a little capacity of coding proteins is playing an increasingly important role in the cancer paradigm. Accumulating evidences demonstrate that lncRNAs have crucial connections with breast cancer prognosis while the studies of lncRNAs in breast cancer are still in its primary stage. In this study, we collected 1052 clinical patient samples, a comparatively large sample size, including 13 159 lncRNA expression profiles of breast invasive carcinoma (BRCA) from The Cancer Genome Atlas database to identify prognosis-related lncRNAs. We randomly separated all of these clinical patient samples into training and testing sets. In the training set, we performed univariable Cox regression analysis for primary screening and played the model for Robust likelihood-based survival for 1000 times. Then 11 lncRNAs with a frequency more than 600 were selected for prediction of the prognosis of BRCA. Using the analysis of multivariate Cox regression, we established a signature risk-score formula for 11 lncRNA to identify the relationship between lncRNA signatures and overall survival. The 11 lncRNA signature was validated both in the testing and the complete set and could effectively classify the high-/low-risk group with different OS. We also verified our results in different stages. Moreover, we analyzed the connection between the 11 lncRNAs and the genes of ESR1, PGR, and Her2, of which protein products (ESR, PGR, and HER2) were used to classify the breast cancer subtypes widely. The results indicated correlations between 11 lncRNAs and the gene of PGR and ESR1. Thus, a prognostic model for 11 lncRNA expression was developed to classify the BRAC clinical patient samples, providing new avenues in understanding the potential therapeutic methods of breast cancer.     

An angiogenesis‐related long noncoding RNA signature correlates with prognosis in patients with hepatocellular carcinoma

Hepatocellular carcinoma (HCC) is one of the most prevalent and lethal cancers worldwide. Neovascularization is closely related to the malignancy of tumors. We constructed a signature of angiogenesis-related long noncoding RNA (lncRNA) to predict the prognosis of patients with HCC. The lncRNA expression matrix of 424 HCC patients was downloaded from The Cancer Genome Atlas (TCGA). First, gene set enrichment analysis (GSEA) was used to distinguish the differentially expressed genes of the angiogenesis genes in liver cancer and adjacent tissues. Next, a signature of angiogenesis-related lncRNAs was constructed using univariate and multivariate analyses, and receiver operating characteristic (ROC) curves were used to assess the accuracy. The signature and relevant clinical information were used to construct the nomogram. A 5-lncRNA signature was highly correlated with overall survival (OS) in HCC patients and performed well in evaluations using the C-index, areas under the curve, and calibration curves. In summary, the 5-lncRNA model can serve as an accurate signature to predict the prognosis of patients with liver cancer, but its mechanism of action must be further elucidated by experiments.     

Identification and Validation of a Prognostic Model Based on Three MVI-Related Genes in Hepatocellular Carcinoma

MVI has significant clinical value for treatment selection and prognosis evaluation in hepatocellular carcinoma (HCC). We aimed to construct a model based on MVI-Related Genes (MVIRGs) for risk assessment and prognosis prediction in patients with HCC. This study utilized various statistical analysis methods for prognostic model construction and validation in the Cancer Genome Atlas (TCGA) and International Cancer Genome Consortium (ICGC) cohorts, respectively. In addition, immunohistochemistry and qRT-PCR were used to analyze and identify the value of the model in our cohort. After the analyses, 153 differentially expressed MVIRGs were identified, and three key genes were selected to construct a prognostic model. The high-risk group showed significantly lower overall survival (OS), and this trend was observed in all subgroups: different age groups, genders, stages, and grades. Risk score was a risk factor independent of age, gender, stage, and grade. Moreover, the ICGC cohort validated the prognostic value of the model corresponding to the TCGA. In our cohort, qRT-PCR and immunohistochemistry showed that all three genes had higher expression levels in HCC samples than in normal controls. High expression levels of genes and high-risk scores showed significantly lower recurrence-free survival (RFS) and OS, especially in MVI-positive HCC samples. Therefore, the prognostic model constructed by three MVIRGs can reliably predict the RFS and OS of patients with HCC and is valuable for guiding clinical treatment selection and prognostic assessment of HCC.     

Development and validation of two nomograms for predicting overall survival and cancer-specific survival in gastric cancer patients with liver metastases: A retrospective cohort study from SEER database

BackgroundGastric cancer is heterogeneous and aggressive, especially with liver metastasis. This study aims to develop two nomograms to predict the overall survival (OS) and cancer-specific survival (CSS) of gastric cancer with liver metastasis (GCLM) patients.MethodsFrom January 2000 to December 2018, a total of 1936 GCLM patients were selected from the Surveillance, Epidemiology, and End Results Program (SEER) database. They were further divided into a training cohort and a validation cohort, with the OS and CSS serving as the study's endpoints. The correlation analyses were used to determine the relationship between the variables. The univariate and multivariate Cox analyses were used to confirm the independent prognostic factors. To discriminate and calibrate the nomogram, calibration curves and the area under the time-dependent receiver operating characteristic curve (time-dependent AUC) were used. DCA curves were used to examine the accuracy and clinical benefits. The clinical utility of the nomogram and the AJCC Stage System was compared using net reclassification improvement (NRI) and integrated differentiation improvement (IDI) (IDI). Finally, the nomogram and the AJCC Stage System risk stratifications were compared.ResultsThere was no collinearity among the variables that were screened. The results of multivariate Cox regression analysis showed that six variables (bone metastasis, lung metastasis, surgery, chemotherapy, grade, age) and five variables (lung metastasis, surgery, chemotherapy, grade, N stage) were identified to establish the nomogram for OS and CSS, respectively. The calibration curves, time-dependent AUC curves, and DCA revealed that both nomograms had pleasant predictive power. Furthermore, NRI and IDI confirmed that the nomogram outperformed the AJCC Stage System.ConclusionBoth nomograms had satisfactory accuracy and were validated to assist clinicians in evaluating the prognosis of GCLM patients.     

Prediction of butterfly diversity hotspots in Belgium: a comparison of statistically focused and land use-focused models

Aim We evaluate differences between and the applicability of three linear predictive models to determine butterfly hotspots in Belgium for nature conservation purposes. Location The study is carried out in Belgium for records located to Universal Transverse Mercator (UTM) grid cells of 5 × 5 km. Methods We first determine the relationship between factors correlated to butterfly diversity by means of modified t‐tests and principal components analysis; subsequently, we predict hotspots using linear models based on land use, climate and topographical variables of well‐surveyed UTM grid cells (n = 197). The well‐surveyed squares are divided into a training set and an evaluation set to test the model predictions. We apply three different models: (1) a ‘statistically focused’ model where variables are entered in descending order of statistical significance, (2) a ‘land use‐focused’ model where land use variables known to be related to butterfly diversity are forced into the model and (3) a ‘hybrid’ model where the variables of the ‘land use‐focused model’ are entered first and subsequently complemented by the remaining variables entered in descending order of statistical significance. Results A principal components analyses reveals that climate, and to a large extent, land use are locked into topography, and that topography and climate are the variables most strongly correlated with butterfly diversity in Belgium. In the statistically focused model, biogeographical region alone explains 65% of the variability; other variables entering the statistically focused model are the area of coniferous and deciduous woodland, elevation and the number of frost days; the statistically focused model explains 77% of the variability in the training set and 66% in the evaluation set. In the land use‐focused model, biogeographical region, deciduous and mixed woodland, natural grassland, heathland and bog, woodland edge, urban and agricultural area and biotope diversity are forced into the model; the land use‐focused model explains 68% of the variability in the training set and 57% in the evaluation set. In the hybrid model, all variables from the land use‐focused model are entered first and the covariates elevation, number of frost days and natural grassland area are added on statistical grounds; the hybrid model explains 78% of the variability in the training set and 67% in the evaluation set. Applying the different models to determine butterfly diversity hotspots resulted in the delimitation of spatially different areas. Main conclusions The best predictions of butterfly diversity in Belgium are obtained by the hybrid model in which land use variables relevant to butterfly richness are entered first after which climatic and topographic variables were added on strictly statistical grounds. The land use‐focused model does not predict butterfly diversity in a satisfactory manner. When using predictive models to determine butterfly diversity, conservation biologists need to be aware of the consequences of applying such models. Although, in conservation biology, land use‐focused models are preferable to statistically focused models, one should always check whether the applied model makes sense on the ground. Predictive models can target mapping efforts towards potentially species‐rich sites and permits the incorporation of un‐surveyed sites into nature conservancy policies. Species richness distribution maps produced by predictive modelling should therefore be used as pro‐active conservation tools.     

A Novel Model to Combine Clinical and Pathway-Based Transcriptomic Information for the Prognosis Prediction of Breast Cancer

Breast cancer is the most common malignancy in women worldwide. With the increasing awareness of heterogeneity in breast cancers, better prediction of breast cancer prognosis is much needed for more personalized treatment and disease management. Towards this goal, we have developed a novel computational model for breast cancer prognosis by combining the Pathway Deregulation Score (PDS) based pathifier algorithm, Cox regression and L1-LASSO penalization method. We trained the model on a set of 236 patients with gene expression data and clinical information, and validated the performance on three diversified testing data sets of 606 patients. To evaluate the performance of the model, we conducted survival analysis of the dichotomized groups, and compared the areas under the curve based on the binary classification. The resulting prognosis genomic model is composed of fifteen pathways (e.g. P53 pathway) that had previously reported cancer relevance, and it successfully differentiated relapse in the training set (log rank p-value = 6.25e-12) and three testing data sets (log rank p-value<0.0005). Moreover, the pathway-based genomic models consistently performed better than gene-based models on all four data sets. We also find strong evidence that combining genomic information with clinical information improved the p-values of prognosis prediction by at least three orders of magnitude in comparison to using either genomic or clinical information alone. In summary, we propose a novel prognosis model that harnesses the pathway-based dysregulation as well as valuable clinical information. The selected pathways in our prognosis model are promising targets for therapeutic intervention.     

Comparative study of radiomics,tumor morphology,and clinicopathological factors in predicting overall survival of patients with rectal cancer before surgery

We compared the ability of a radiomics model, morphological imaging model, and clinicopathological risk model to predict 3-year overall survival (OS) in 206 patients with rectal cancer who underwent radical surgery and had magnetic resonance imaging, clinicopathological, and OS data available. The patients were randomized to a training cohort (n = 146) and a verification cohort (n = 60). Radiomics features were extracted from preoperative T2-weighted images, and a radiomics score model was constructed. Factors that were significant in the Cox multivariate analysis were used to construct the final morphological tumor model and clinicopathological model. A comprehensive model in the form of a line chart was established by combining the three models. Ten radiomics features significantly related to OS were selected to construct the radiomics feature model and calculate the radiomics score. In the morphological model, mesorectal extension depth and distance between the lower tumor margin and the anal margin were significant prognostic factors. N stage was the only significant clinicopathological factor. The comprehensive model combined with the above factors had the best prediction performance for OS. The C-index had a predictive performance of 0.872 (95% confidence interval [CI]: 0.832–0.912) in the training cohort and 0.944 (95% CI: 0.890–0.990) in the verification cohort, which was better than for any single model. The comprehensive model was divided into high-risk and low-risk groups. Kaplan-Meier curve analysis showed that all factors were significantly correlated with poor OS in the high-risk group. A comprehensive nomogram based on multi-model radiomics features can predict 3-year OS after rectal cancer surgery.     

丝氨酸蛋白酶抑制剂(SERPIN)家族相关基因在结肠腺癌中预后模型的建立及应用

应用生物信息学方法,构建结肠腺癌(COAD)丝氨酸蛋白酶抑制剂(SERPIN)家族相关基因预后模型。从TCGA数据库和GEO数据库下载结肠腺癌(COAD)转录组和临床数据,根据数据中SERPINs家族基因的表达量对COAD患者进行一致性聚类分析;将数据随机均分为训练集(Train)组和验证集(Test)组,基于两个亚型的差异基因,利用Train组进行COX回归和Lasso回归构建预后模型,根据模型风险评分中位值将样本分为高、低风险两组,绘制高低风险组患者生存曲线;通过ROC曲线评价模型预测能力;利用Test组数据验证模型;构建列线图,评估患者生存率模型预测值与实际值的一致性;并利用利用ESTIMATE算法和CIBERSORT算法评估风险评分和肿瘤微环境(TME)以及免疫浸润的相关性。通过34个SERPIN基因确定了两个亚型,基于2个亚型筛选出了436个预后相关分型差异基因,通过Lasso回归确定出了11个预后相关基因参与风险模型的构建,根据模型评分区分的高低风险组具有明显的生存差异,列线图可以准确预测1、3和5年生存率。肿瘤微环境分析和免疫浸润分析显示高风险评分组患者免疫活性差。SERPIN家族相关基因构建的风险评分模型能够预测COAD的预后,有利于进一步指导临床对COAD的诊治,提高患者生存率。     

Gene-expression profiles predict survival of patients with lung adenocarcinoma

Histopathology is insufficient to predict disease progression and clinical outcome in lung adenocarcinoma. Here we show that gene-expression profiles based on microarray analysis can be used to predict patient survival in early-stage lung adenocarcinomas. Genes most related to survival were identified with univariate Cox analysis. Using either two equivalent but independent training and testing sets, or 'leave-one-out' cross-validation analysis with all tumors, a risk index based on the top 50 genes identified low-risk and high-risk stage I lung adenocarcinomas, which differed significantly with respect to survival. This risk index was then validated using an independent sample of lung adenocarcinomas that predicted high- and low-risk groups. This index included genes not previously associated with survival. The identification of a set of genes that predict survival in early-stage lung adenocarcinoma allows delineation of a high-risk group that may benefit from adjuvant therapy.     

Integrated analysis of multiple microarray datasets identifies a reproducible survival predictor in ovarian cancer

Background

Public data integration may help overcome challenges in clinical implementation of microarray profiles. We integrated several ovarian cancer datasets to identify a reproducible predictor of survival.

Methodology/Principal Findings

Four microarray datasets from different institutions comprising 265 advanced stage tumors were uniformly reprocessed into a single training dataset, also adjusting for inter-laboratory variation (“batch-effect”). Supervised principal component survival analysis was employed to identify prognostic models. Models were independently validated in a 61-patient cohort using a custom array genechip and a publicly available 229-array dataset. Molecular correspondence of high- and low-risk outcome groups between training and validation datasets was demonstrated using Subclass Mapping. Previously established molecular phenotypes in the 2^nd validation set were correlated with high and low-risk outcome groups. Functional representational and pathway analysis was used to explore gene networks associated with high and low risk phenotypes. A 19-gene model showed optimal performance in the training set (median OS 31 and 78 months, p<0.01), 1^st validation set (median OS 32 months versus not-yet-reached, p = 0.026) and 2^nd validation set (median OS 43 versus 61 months, p = 0.013) maintaining independent prognostic power in multivariate analysis. There was strong molecular correspondence of the respective high- and low-risk tumors between training and 1^st validation set. Low and high-risk tumors were enriched for favorable and unfavorable molecular subtypes and pathways, previously defined in the public 2^nd validation set.

Conclusions/Significance

Integration of previously generated cancer microarray datasets may lead to robust and widely applicable survival predictors. These predictors are not simply a compilation of prognostic genes but appear to track true molecular phenotypes of good- and poor-outcome.