Predicting survival from colorectal cancer histology slides using deep learning: A retrospective multicenter study

PLoS Medicine, Journal Year: 2019, Volume and Issue: 16(1), P. e1002730 - e1002730

Published: Jan. 24, 2019

Background For virtually every patient with colorectal cancer (CRC), hematoxylin–eosin (HE)–stained tissue slides are available. These images contain quantitative information, which is not routinely used to objectively extract prognostic biomarkers. In the present study, we investigated whether deep convolutional neural networks (CNNs) can prognosticators directly from these widely available images. Methods and findings We hand-delineated single-tissue regions in 86 CRC slides, yielding more than 100,000 HE image patches, train a CNN by transfer learning, reaching nine-class accuracy of >94% an independent data set 7,180 25 patients. With this tool, performed automated decomposition representative multitissue 862 500 stage I–IV patients The Cancer Genome Atlas (TCGA) cohort, large international multicenter collection tissue. Based on output neuron activations CNN, calculated "deep stroma score," was factor for overall survival (OS) multivariable Cox proportional hazard model (hazard ratio [HR] 95% confidence interval [CI]: 1.99 [1.27–3.12], p = 0.0028), while same manual quantification stromal areas gene expression signature cancer-associated fibroblasts (CAFs) were only specific tumor stages. validated cohort 409 "Darmkrebs: Chancen der Verhütung durch Screening" (DACHS) study who recruited between 2003 2007 multiple institutions Germany. Again, score OS (HR 1.63 [1.14–2.33], 0.008), CRC-specific 2.29 [1.5–3.48], 0.0004), relapse-free (RFS; HR 1.92 [1.34–2.76], 0.0004). A prospective validation required before biomarker be implemented clinical workflows. Conclusions our retrospective show that assess human microenvironment predict prognosis histopathological

Language: Английский

---

Metabolic phenotyping in clinical and surgical environments

Nature, Journal Year: 2012, Volume and Issue: 491(7424), P. 384 - 392

Published: Nov. 13, 2012

Language: Английский

---

Current practice of liquid chromatography–mass spectrometry in metabolomics and metabonomics

Journal of Pharmaceutical and Biomedical Analysis, Journal Year: 2013, Volume and Issue: 87, P. 12 - 25

Published: July 17, 2013

Language: Английский

---

Targeted Metabolomics for Biomarker Discovery

Angewandte Chemie International Edition, Journal Year: 2010, Volume and Issue: 49(32), P. 5426 - 5445

Published: July 13, 2010

Metabolomics is a truly interdisciplinary field of science, which combines analytical chemistry, platform technology, mass spectrometry, and NMR spectroscopy with sophisticated data analysis. Applied to biomarker discovery, it includes aspects pathobiochemistry, systems biology/medicine, molecular diagnostics requires bioinformatics multivariate statistics. While successfully established in the screening inborn errors neonates, metabolomics now widely used characterization diagnostic research an ever increasing number diseases. In this Review we highlight important technical prerequisites as well recent developments analysis special emphasis on their utility identification qualification, targeted by employing high-throughput spectrometry.

Language: Английский

---

Ultrahigh Performance Liquid Chromatography−Tandem Mass Spectrometry Method for Fast and Robust Quantification of Anionic and Aromatic Metabolites

Analytical Chemistry, Journal Year: 2010, Volume and Issue: 82(11), P. 4403 - 4412

Published: April 30, 2010

Quantification of metabolites is pivotal relevance in biology, where it complements more established omics techniques such as transcriptomics and proteomics. Here, we present a 25 min ion-pairing ultrahigh performance liquid chromatography−tandem mass spectrometry method that was developed for comprehensive coverage central metabolism (glycolysis, pentose phosphate pathway, tricarboxylic acid cycle) closely related biosynthetic reactions. We demonstrate quantification 138 compounds, including carboxylic acids, amino sugar phosphates, nucleotides, functionalized aromatics. Biologically relevant isomers phosphates are individually quantified by combining chromatographic separation fragmentation. The obtained sensitivity robustness enabled the detection than half all tested compounds each eight diverse biological samples 0.5−50 mg dry cell weight. recommend this routine yet primary wide variety matrices.

Language: Английский

---

High‐Throughput Metabolomics by 1D NMR

Angewandte Chemie International Edition, Journal Year: 2018, Volume and Issue: 58(4), P. 968 - 994

Published: July 12, 2018

Metabolomics deals with the whole ensemble of metabolites (the metabolome). As one -omic sciences, it relates to biology, physiology, pathology and medicine; but are chemical entities, small organic molecules or inorganic ions. Therefore, their proper identification quantitation in complex biological matrices requires a solid ground. With respect for example, DNA, much more prone oxidation enzymatic degradation: we can reconstruct large parts mammoth's genome from specimen, unable do same its metabolome, which was probably largely degraded few hours after animal's death. Thus, need standard operating procedures, good skills sample preparation storage subsequent analysis, accurate analytical broad knowledge chemometrics advanced statistical tools, at least two metabolomic techniques, MS NMR. All these traditionally cultivated by chemists. Here focus on metabolomics standpoint restrict ourselves From point view, NMR has pros cons does provide peculiar holistic perspective that may speak future adoption as population-wide health screening technique.

Language: Английский

---

Integration of Metabolomics and Transcriptomics Revealed a Fatty Acid Network Exerting Growth Inhibitory Effects in Human Pancreatic Cancer

Clinical Cancer Research, Journal Year: 2013, Volume and Issue: 19(18), P. 4983 - 4993

Published: Aug. 6, 2013

Abstract Purpose: To identify metabolic pathways that are perturbed in pancreatic ductal adenocarcinoma (PDAC), we investigated gene-metabolite networks with integration of metabolomics and transcriptomics. Experimental Design: We conducted global metabolite profiling analysis on two independent cohorts resected PDAC cases to critical metabolites alteration may contribute the progression cancer. then searched for gene surrogates were significantly correlated key metabolites, by integrating expression profiles. Results: Fifty-five consistently altered tumors as compared adjacent nontumor tissues a test cohort (N = 33) an validation 31). Weighted network revealed unique set free fatty acids (FFA) highly coregulated decreased PDAC. Pathway 157 differentially expressed lipid metabolism network, including lipolytic enzymes PNLIP, CLPS, PNLIPRP1, PNLIPRP2. Gene expressions these lipases tissues, leading reduced FFAs. More importantly, lower PNLIP was associated poorer survival cohorts. further showed saturated FFAs, palmitate stearate, induced TRAIL expression, triggered apoptosis, inhibited proliferation cancer cells. Conclusions: Our results suggest impairment pathway involving lipases, play important role development provide potential targets therapeutic intervention. Clin Cancer Res; 19(18); 4983–93. ©2013 AACR.

Language: Английский

---

A Review of Applications of Metabolomics in Cancer

Metabolites, Journal Year: 2013, Volume and Issue: 3(3), P. 552 - 574

Published: July 5, 2013

Cancer is a devastating disease that alters the metabolism of cell and surrounding milieu. Metabolomics growing powerful technology capable detecting hundreds to thousands metabolites in tissues biofluids. The recent advances metabolomics technologies have enabled deeper investigation into cancer better understanding how cells use glycolysis, known as "Warburg effect," advantageously produce amino acids, nucleotides lipids necessary for tumor proliferation vascularization. Currently, research being used discover diagnostic biomarkers clinic, understand its complex heterogeneous nature, pathways involved could be new targets monitor metabolic during therapeutic intervention. These approaches may also provide clues personalized treatments by providing useful information clinician about patient's response medical interventions.

Language: Английский

---

Metabolomics Analysis for Biomarker Discovery: Advances and Challenges

Current Medicinal Chemistry, Journal Year: 2013, Volume and Issue: 20(2), P. 257 - 271

Published: Jan. 1, 2013

Over the last decades there has been a change in biomedical research with search for single genes, transcripts, proteins, or metabolites being substituted by coverage of entire genome, transcriptome, proteome, and metabolome "omics" approaches. The emergence metabolomics, defined as comprehensive analysis all system, is still recent compared to other fields, but its particular features improvement both analytical techniques pattern recognition methods contributed greatly increasingly use. feasibility metabolomics biomarker discovery supported assumption that are important players biological systems diseases cause disruption biochemical pathways, which not new concepts. In fact, meaning parallel assessment multiple metabolites, shown have benefits various clinical areas. Compared classical diagnostic approaches conventional biomarkers, offers potential advantages sensitivity specificity. Despite potential, retains several intrinsic limitations great impact on widespread implementation - these experimental measurements. This review will provide an insight characteristics, strengths, limitations, advances always keeping mind application clinical/ health areas tool.

Language: Английский

---

Effects of pre-analytical processes on blood samples used in metabolomics studies

Analytical and Bioanalytical Chemistry, Journal Year: 2015, Volume and Issue: 407(17), P. 4879 - 4892

Published: March 3, 2015

Every day, analytical and bio-analytical chemists make sustained efforts to improve the sensitivity, specificity, robustness, reproducibility of their methods. Especially in targeted non-targeted profiling approaches, including metabolomics analysis, these objectives are not easy achieve; however, robust reproducible measurements low coefficients variation (CV) crucial for successful approaches. Nevertheless, all from analysts vain if sample quality is poor, i.e. preanalytical errors made by partner during collection. Preanalytical risks more common than expected, even when standard operating procedures (SOP) used. This risk particularly high clinical studies, poor may heavily bias CV final results, leading disappointing outcomes study consequently, although unjustified, critical questions about performance approach who provided samples. review focuses on phase liquid chromatography-mass spectrometry-driven analysis body fluids. Several important factors that seriously affect profile investigated metabolome fluids, before collection, blood drawing, subsequent handling whole (transportation), processing plasma serum, inadequate conditions storage, will be discussed. In addition, a detailed description latent effects stability suggestion practical procedure circumvent given.

Language: Английский

---