Feeds:
Posts
Comments

Archive for the ‘Meta-analysis of transcriptome data’ Category


Deep Learning extracts Histopathological Patterns and accurately discriminates 28 Cancer and 14 Normal Tissue Types: Pan-cancer Computational Histopathology Analysis

Reporter: Aviva Lev-Ari, PhD, RN

Pan-cancer computational histopathology reveals mutations, tumor composition and prognosis

Yu Fu1, Alexander W Jung1, Ramon Viñas Torne1, Santiago Gonzalez1,2, Harald Vöhringer1, Mercedes Jimenez-Linan3, Luiza Moore3,4, and Moritz Gerstung#1,5 # to whom correspondence should be addressed 1) European Molecular Biology Laboratory, European Bioinformatics Institute (EMBL-EBI), Hinxton, UK. 2) Current affiliation: Institute for Research in Biomedicine (IRB Barcelona), Parc Científic de Barcelona, Barcelona, Spain. 3) Department of Pathology, Addenbrooke’s Hospital, Cambridge, UK. 4) Wellcome Sanger Institute, Hinxton, UK 5) European Molecular Biology Laboratory, Genome Biology Unit, Heidelberg, Germany.

Correspondence:

Dr Moritz Gerstung European Molecular Biology Laboratory European Bioinformatics Institute (EMBL-EBI) Hinxton, CB10 1SA UK. Tel: +44 (0) 1223 494636 E-mail: moritz.gerstung@ebi.ac.uk

Abstract

Pan-cancer computational histopathology reveals mutations, tumor composition and prognosis

Here we use deep transfer learning to quantify histopathological patterns across 17,396 H&E stained histopathology image slides from 28 cancer types and correlate these with underlying genomic and transcriptomic data. Pan-cancer computational histopathology (PC-CHiP) classifies the tissue origin across organ sites and provides highly accurate, spatially resolved tumor and normal distinction within a given slide. The learned computational histopathological features correlate with a large range of recurrent genetic aberrations, including whole genome duplications (WGDs), arm-level copy number gains and losses, focal amplifications and deletions as well as driver gene mutations within a range of cancer types. WGDs can be predicted in 25/27 cancer types (mean AUC=0.79) including those that were not part of model training. Similarly, we observe associations with 25% of mRNA transcript levels, which enables to learn and localise histopathological patterns of molecularly defined cell types on each slide. Lastly, we find that computational histopathology provides prognostic information augmenting histopathological subtyping and grading in the majority of cancers assessed, which pinpoints prognostically relevant areas such as necrosis or infiltrating lymphocytes on each tumour section. Taken together, these findings highlight the large potential of PC-CHiP to discover new molecular and prognostic associations, which can augment diagnostic workflows and lay out a rationale for integrating molecular and histopathological data.

SOURCE

https://www.biorxiv.org/content/10.1101/813543v1

Key points

● Pan-cancer computational histopathology analysis with deep learning extracts histopathological patterns and accurately discriminates 28 cancer and 14 normal tissue types

● Computational histopathology predicts whole genome duplications, focal amplifications and deletions, as well as driver gene mutations

● Wide-spread correlations with gene expression indicative of immune infiltration and proliferation

● Prognostic information augments conventional grading and histopathology subtyping in the majority of cancers

 

Discussion

Here we presented PC-CHiP, a pan-cancer transfer learning approach to extract computational histopathological features across 42 cancer and normal tissue types and their genomic, molecular and prognostic associations. Histopathological features, originally derived to classify different tissues, contained rich histologic and morphological signals predictive of a range of genomic and transcriptomic changes as well as survival. This shows that computer vision not only has the capacity to highly accurately reproduce predefined tissue labels, but also that this quantifies diverse histological patterns, which are predictive of a broad range of genomic and molecular traits, which were not part of the original training task. As the predictions are exclusively based on standard H&E-stained tissue sections, our analysis highlights the high potential of computational histopathology to digitally augment existing histopathological workflows. The strongest genomic associations were found for whole genome duplications, which can in part be explained by nuclear enlargement and increased nuclear intensities, but seemingly also stems from tumour grade and other histomorphological patterns contained in the high-dimensional computational histopathological features. Further, we observed associations with a range of chromosomal gains and losses, focal deletions and amplifications as well as driver gene mutations across a number of cancer types. These data demonstrate that genomic alterations change the morphology of cancer cells, as in the case of WGD, but possibly also that certain aberrations preferentially occur in distinct cell types, reflected by the tumor histology. Whatever is the cause or consequence in this equation, these associations lay out a route towards genomically defined histopathology subtypes, which will enhance and refine conventional assessment. Further, a broad range of transcriptomic correlations was observed reflecting both immune cell infiltration and cell proliferation that leads to higher tumor densities. These examples illustrated the remarkable property that machine learning does not only establish novel molecular associations from pre-computed histopathological feature sets but also allows the localisation of these traits within a larger image. While this exemplifies the power of a large scale data analysis to detect and localise recurrent patterns, it is probably not superior to spatially annotated training data. Yet such data can, by definition, only be generated for associations which are known beforehand. This appears straightforward, albeit laborious, for existing histopathology classifications, but more challenging for molecular readouts. Yet novel spatial transcriptomic44,45 and sequencing technologies46 bring within reach spatially matched molecular and histopathological data, which would serve as a gold standard in combining imaging and molecular patterns. Across cancer types, computational histopathological features showed a good level of prognostic relevance, substantially improving prognostic accuracy over conventional grading and histopathological subtyping in the majority of cancers. It is this very remarkable that such predictive It is made available under a CC-BY-NC 4.0 International license. (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. bioRxiv preprint first posted online Oct. 25, 2019; doi: http://dx.doi.org/10.1101/813543. The copyright holder for this preprint signals can be learned in a fully automated fashion. Still, at least at the current resolution, the improvement over a full molecular and clinical workup was relatively small. This might be a consequence of the far-ranging relations between histopathology and molecular phenotypes described here, implying that histopathology is a reflection of the underlying molecular alterations rather than an independent trait. Yet it probably also highlights the challenges of unambiguously quantifying histopathological signals in – and combining signals from – individual areas, which requires very large training datasets for each tumour entity. From a methodological point of view, the prediction of molecular traits can clearly be improved. In this analysis, we adopted – for the reason of simplicity and to avoid overfitting – a transfer learning approach in which an existing deep convolutional neural network, developed for classification of everyday objects, was fine tuned to predict cancer and normal tissue types. The implicit imaging feature representation was then used to predict molecular traits and outcomes. Instead of employing this two-step procedure, which risks missing patterns irrelevant for the initial classification task, one might directly employ either training on the molecular trait of interest, or ideally multi-objective learning. Further improvement may also be related to the choice of the CNN architecture. Everyday images have no defined scale due to a variable z-dimension; therefore, the algorithms need to be able to detect the same object at different sizes. This clearly is not the case for histopathology slides, in which one pixel corresponds to a defined physical size at a given magnification. Therefore, possibly less complex CNN architectures may be sufficient for quantitative histopathology analyses, and also show better generalisation. Here, in our proof-of-concept analysis, we observed a considerable dependence of the feature representation on known and possibly unknown properties of our training data, including the image compression algorithm and its parameters. Some of these issues could be overcome by amending and retraining the network to isolate the effect of confounding factors and additional data augmentation. Still, given the flexibility of deep learning algorithms and the associated risk of overfitting, one should generally be cautious about the generalisation properties and critically assess whether a new image is appropriately represented. Looking forward, our analyses revealed the enormous potential of using computer vision alongside molecular profiling. While the eye of a trained human may still constitute the gold standard for recognising clinically relevant histopathological patterns, computers have the capacity to augment this process by sifting through millions of images to retrieve similar patterns and establish associations with known and novel traits. As our analysis showed this helps to detect histopathology patterns associated with a range of genomic alterations, transcriptional signatures and prognosis – and highlight areas indicative of these traits on each given slide. It is therefore not too difficult to foresee how this may be utilised in a computationally augmented histopathology workflow enabling more precise and faster diagnosis and prognosis. Further, the ability to quantify a rich set of histopathology patterns lays out a path to define integrated histopathology and molecular cancer subtypes, as recently demonstrated for colorectal cancers47 .

Lastly, our analyses provide It is made available under a CC-BY-NC 4.0 International license. (which was not peer-reviewed) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity.

bioRxiv preprint first posted online Oct. 25, 2019; doi: http://dx.doi.org/10.1101/813543.

The copyright holder for this preprint proof-of-concept for these principles and we expect them to be greatly refined in the future based on larger training corpora and further algorithmic refinements.

SOURCE

https://www.biorxiv.org/content/biorxiv/early/2019/10/25/813543.full.pdf

 

Other related articles published in this Open Access Online Scientific Journal include the following: 

 

CancerBase.org – The Global HUB for Diagnoses, Genomes, Pathology Images: A Real-time Diagnosis and Therapy Mapping Service for Cancer Patients – Anonymized Medical Records accessible to anyone on Earth

Reporter: Aviva Lev-Ari, PhD, RN

https://pharmaceuticalintelligence.com/2016/07/28/cancerbase-org-the-global-hub-for-diagnoses-genomes-pathology-images-a-real-time-diagnosis-and-therapy-mapping-service-for-cancer-patients-anonymized-medical-records-accessible-to/

 

631 articles had in their Title the keyword “Pathology”

https://pharmaceuticalintelligence.com/?s=Pathology

 

Read Full Post »


Single-cell RNA-seq helps in finding intra-tumoral heterogeneity in pancreatic cancer

Reporter and Curator: Dr. Sudipta Saha, Ph.D.

 

Pancreatic cancer is a significant cause of cancer mortality; therefore, the development of early diagnostic strategies and effective treatment is essential. Improvements in imaging technology, as well as use of biomarkers are changing the way that pancreas cancer is diagnosed and staged. Although progress in treatment for pancreas cancer has been incremental, development of combination therapies involving both chemotherapeutic and biologic agents is ongoing.

 

Cancer is an evolutionary disease, containing the hallmarks of an asexually reproducing unicellular organism subject to evolutionary paradigms. Pancreatic ductal adenocarcinoma (PDAC) is a particularly robust example of this phenomenon. Genomic features indicate that pancreatic cancer cells are selected for fitness advantages when encountering the geographic and resource-depleted constraints of the microenvironment. Phenotypic adaptations to these pressures help disseminated cells to survive in secondary sites, a major clinical problem for patients with this disease.

 

The immune system varies in cell types, states, and locations. The complex networks, interactions, and responses of immune cells produce diverse cellular ecosystems composed of multiple cell types, accompanied by genetic diversity in antigen receptors. Within this ecosystem, innate and adaptive immune cells maintain and protect tissue function, integrity, and homeostasis upon changes in functional demands and diverse insults. Characterizing this inherent complexity requires studies at single-cell resolution. Recent advances such as massively parallel single-cell RNA sequencing and sophisticated computational methods are catalyzing a revolution in our understanding of immunology.

 

PDAC is the most common type of pancreatic cancer featured with high intra-tumoral heterogeneity and poor prognosis. In the present study to comprehensively delineate the PDAC intra-tumoral heterogeneity and the underlying mechanism for PDAC progression, single-cell RNA-seq (scRNA-seq) was employed to acquire the transcriptomic atlas of 57,530 individual pancreatic cells from primary PDAC tumors and control pancreases. The diverse malignant and stromal cell types, including two ductal subtypes with abnormal and malignant gene expression profiles respectively, were identified in PDAC.

 

The researchers found that the heterogenous malignant subtype was composed of several subpopulations with differential proliferative and migratory potentials. Cell trajectory analysis revealed that components of multiple tumor-related pathways and transcription factors (TFs) were differentially expressed along PDAC progression. Furthermore, it was found a subset of ductal cells with unique proliferative features were associated with an inactivation state in tumor-infiltrating T cells, providing novel markers for the prediction of antitumor immune response. Together, the findings provided a valuable resource for deciphering the intra-tumoral heterogeneity in PDAC and uncover a connection between tumor intrinsic transcriptional state and T cell activation, suggesting potential biomarkers for anticancer treatment such as targeted therapy and immunotherapy.

 

References:

 

https://www.ncbi.nlm.nih.gov/pubmed/31273297

 

https://www.ncbi.nlm.nih.gov/pubmed/21491194

 

https://www.ncbi.nlm.nih.gov/pubmed/27444064

 

https://www.ncbi.nlm.nih.gov/pubmed/28983043

 

https://www.ncbi.nlm.nih.gov/pubmed/24976721

 

https://www.ncbi.nlm.nih.gov/pubmed/27693023

 

Read Full Post »


5:00 – 5:45 PM Early Diagnosis Through Predictive Biomarkers, NonInvasive Testing

Reporter: Stephen J. Williams, Ph.D.

 

Diagnosing cancer early is often the difference between survival and death. Hear from experts regarding the new and emerging technologies that form the next generation of cancer diagnostics.

Moderator: Heather Rose, Director of Licensing, Thomas Jefferson University
Speakers:
Bonnie Anderson, Chairman and CEO, Veracyte @BonnieAndDx
Kevin Hrusovsky, Founder and Chairman, Powering Precision Health @KevinHrusovsky

Bonnie Anderson and Veracyte produces genomic tests for thyroid and other cancer diagnosis.  Kevin Hrusovksy and Precision Health uses peer reviewed evidence based medicine to affect precision medicine decision.

Bonnie: aim to get a truth of diagnosis.  Getting tumor tissue is paramount as well as properly preserved tissue.  They use deep RNA sequencing  and machine learning  in their clinically approved tests.

Kevin: Serial biospace entrepreneur.  Two diseases, cancer and neurologic, have been diseases which have been hardest to get reproducible and validated biomarkers of early disease.  He concentrates on protein biomarkers.

Heather:  FDA has recently approved drugs for early disease intervention.  However the use of biomarkers can go beyond patient stratification in clinical trials.

Kevin: 15 approved drugs for MS but the markers are scans looking for brain atrophy which is too late of an endpoint.  So we need biomarkers of early disease progression.  We can use those early biomarkers of disease progression so pharma can target those early biomarkers and or use those early biomarkers of disease progression  for endpoint

Bonnie: exciting time in the early diagnostics field. She prefers transcriptomics to DNA based methods such as WES or WGS (whole exome or whole genome sequencing).  It was critical to show data on the cost savings imparted by their transcriptomic based thryoid cancer diagnostic test for payers to consider this test eligible for reimbursement.

Kevin: There has been 20 million  CAT scans for  cancer but it is estimated 90% of these scans led to misdiagnosis. Biomarker  development  has revolutionized diagnostics in this disease area.  They have developed a breakthrough panel of ten protein biomarkers in serum which he estimates may replace 5 million mammograms.

All panelists agreed on the importance of regulatory compliance and the focus of new research should be on early detection.  In addition they believe that Dr. Gotlieb’s appointment to the FDA is a positive for the biomarker development field, as Dr. Gotlieb understands the potential and importance of early detection and prevention of disease.  Kevin also felt Dr. Gotlieb understands the importance of incorporating biomarkers as endpoints in clinical trials.  Over 750 phase 1,2, and 3 clinical trials use biomarker endpoints but the pharma companies still need to prove the biomarkers clinical relevance to the FDA.They also agreed it would be helpful to involve advocacy groups in putting more pressure on the healthcare providers and policy makers on this importance of diagnostics as a preventative measure.

In addition, the discovery and use of biomarkers as disease endpoints has led to a resurgence of Alzheimer’s disease drug development by companies which have previously given up on these type of neurodegenerative diseases.

Kevin feels proteomics offers great advantages over DNA-based diagnostics, especially in cancer such as ovarian cancer, where a high degree of specificity for a diagnostic test is required to ascertain if a woman should undergo prophylactic oophorectomy.  He suggests that a new blood-based protein biomarker panel is being developed for early detection of some forms of ovarian cancer.

Please follow on Twitter using the following #hash tags and @pharma_BI

#MCConverge

#cancertreatment

#healthIT

#innovation

#precisionmedicine

#healthcaremodels

#personalizedmedicine

#healthcaredata

And at the following handles:

@pharma_BI

@medcitynews

 

Please see related articles on Live Coverage of Previous Meetings on this Open Access Journal

LIVE – Real Time – 16th Annual Cancer Research Symposium, Koch Institute, Friday, June 16, 9AM – 5PM, Kresge Auditorium, MIT

Real Time Coverage and eProceedings of Presentations on 11/16 – 11/17, 2016, The 12th Annual Personalized Medicine Conference, HARVARD MEDICAL SCHOOL, Joseph B. Martin Conference Center, 77 Avenue Louis Pasteur, Boston

Tweets Impression Analytics, Re-Tweets, Tweets and Likes by @AVIVA1950 and @pharma_BI for 2018 BioIT, Boston, 5/15 – 5/17, 2018

BIO 2018! June 4-7, 2018 at Boston Convention & Exhibition Center

https://pharmaceuticalintelligence.com/press-coverage/

Read Full Post »


SNP-based Study on high BMI exposure confirms CVD and DM Risks – no associations with Stroke

Reporter: Aviva Lev-Ari, PhD, RN

Genes Affirm: High BMI Carries Weighty Heart, Diabetes Risk – Mendelian randomization study adds to ‘burgeoning evidence’

by Crystal Phend, Senior Associate Editor, MedPage Today, July 05, 2017

 

The “genetically instrumented” measure of high BMI exposure — calculated based on 93 single-nucleotide polymorphisms associated with BMI in prior genome-wide association studies — was associated with the following risks (odds ratios given per standard deviation higher BMI):

  • Hypertension (OR 1.64, 95% CI 1.48-1.83)
  • Coronary heart disease (CHD; OR 1.35, 95% CI 1.09-1.69)
  • Type 2 diabetes (OR 2.53, 95% CI 2.04-3.13)
  • Systolic blood pressure (β 1.65 mm Hg, 95% CI 0.78-2.52 mm Hg)
  • Diastolic blood pressure (β 1.37 mm Hg, 95% CI 0.88-1.85 mm Hg)

However, there were no associations with stroke, Donald Lyall, PhD, of the University of Glasgow, and colleagues reported online in JAMA Cardiology.

The associations independent of age, sex, Townsend deprivation scores, alcohol intake, and smoking history were found in baseline data from 119,859 participants in the population-based U.K. Biobank who had complete medical, sociodemographic, and genetic data.

“The main advantage of an MR approach is that certain types of study bias can be minimized,” the team noted. “Because DNA is stable and randomly inherited, which helps to mitigate errors from reverse causality and confounding, genetic variation can be used as a proxy for lifetime BMI to overcome limitations such as reverse causality and confounding, a process that hampers observational analyses of obesity and its consequences.”

 

Other related articles published in this Open Access Online Scientific Journal include the following:

9 results for Kindle Store : “Aviva Lev-Ari”

Sort by 
Relevance
Featured
Price: Low to High
Price: High to Low
Avg. Customer Review
Publication Date
  • Product Details

    Etiologies of Cardiovascular Diseases: Epigenetics, Genetics and Genomics

    Nov 28, 2015 | Kindle eBook

    by Justin D. Pearlman MD ME PhD MA FACC and Stephen J. Williams PhD
    Subscribers read for free.
    Auto-delivered wirelessly
    Sold by: Amazon Digital Services LLC
  • Product Details

    Perspectives on Nitric Oxide in Disease Mechanisms (Biomed e-Books Book 1)

    Jun 20, 2013 | Kindle eBook

    by Margaret Baker PhD and Tilda Barliya PhD
    Subscribers read for free.
    Auto-delivered wirelessly
    Sold by: Amazon Digital Services LLC
  • Product Details

    Cancer Therapies: Metabolic, Genomics, Interventional, Immunotherapy and Nanotechnology in Therapy Delivery (Series C Book 2)

    May 13, 2017 | Kindle eBook

    by Larry H. Bernstein and Demet Sag
    Subscribers read for free.
    Auto-delivered wirelessly
    Sold by: Amazon Digital Services LLC
  • Product Details

    Metabolic Genomics & Pharmaceutics (BioMedicine – Metabolomics, Immunology, Infectious Diseases Book 1)

    Jul 21, 2015 | Kindle eBook

    by Larry H. Bernstein MD FCAP and Prabodah Kandala PhD
    Subscribers read for free.
    Auto-delivered wirelessly
    Sold by: Amazon Digital Services LLC
  • Product Details

    Milestones in Physiology: Discoveries in Medicine, Genomics and Therapeutics (Series E: Patient-Centered Medicine Book 3)

    Dec 26, 2015 | Kindle eBook

    by Larry H. Bernstein MD FACP and Aviva Lev-Ari PhD RN
    Subscribers read for free.
    Auto-delivered wirelessly
    Sold by: Amazon Digital Services LLC
  • Product Details

    Genomics Orientations for Personalized Medicine (Frontiers in Genomics Research Book 1)

    Nov 22, 2015 | Kindle eBook

    by Sudipta Saha PhD and Ritu Saxena PhD
    Subscribers read for free.
    Auto-delivered wirelessly
    Sold by: Amazon Digital Services LLC
  • Product Details

    Cancer Biology and Genomics for Disease Diagnosis (Series C: e-Books on Cancer & Oncology Book 1)

    Aug 10, 2015 | Kindle eBook

    by Larry H Bernstein MD FCAP and Prabodh Kumar Kandala PhD
    Subscribers read for free.
    Auto-delivered wirelessly
    Sold by: Amazon Digital Services LLC
  • Product Details

    Regenerative and Translational Medicine: The Therapeutic Promise for Cardiovascular Diseases

    Dec 26, 2015 | Kindle eBook

    by Justin D. Pearlman MD ME PhD MA FACC and Stephen J. Williams
    Subscribers read for free.
    Auto-delivered wirelessly
    Sold by: Amazon Digital Services LLC
  • Product Details

    Cardiovascular Original Research: Cases in Methodology Design for Content Co-Curation: The Art of Scientific & Medical Curation

    Nov 29, 2015 | Kindle eBook

    by Larry H. Bernstein MD FCAP and Aviva Lev-Ari PhD RN
    Subscribers read for free.
    Auto-delivered wirelessly

 

Read Full Post »


Genomic Diagnostics: Three Techniques to Perform Single Cell Gene Expression and Genome Sequencing Single Molecule DNA Sequencing

Curator: Aviva Lev-Ari, PhD, RN

 

This article presents Three Techniques to Perform Single Cell Gene Expression and Genome Sequencing Single molecule DNA sequencing

Read Full Post »


The BioPharma Industry’s Unrealized Wealth of Data, by Ben Szekely, Vice President, Cambridge Semantics

Reporter: Aviva Lev-Ari, PhD, RN

 

 

The BioPharma Industry’s Unrealized Wealth of Data

by Ben Szekely, Vice President of Solutions and Pre-sales, Cambridge Semantics

 

Solving the great medical challenges of our time reside within patient data. Clinical trial data, real-world evidence, patient feedback, genetic data, wearables data and adverse event reports contain signals to target medicines at the right patient populations, improve overall safety, and uncover the next blockbuster therapy for unmet medical needs.

However, data sources are large, diverse, multi-structured, messy and highly regulated presenting numerous challenges. As result, extracting value from data are slow to come and require manual work or long-poll dependencies on IT and Data Science teams.

Fortunately, there are new ways being adopted to take better advantage of the ever-growing volumes of patient data.  Called ‘Smart’ Patient Data Lakes (SPDL), these tools create an Enterprise Knowledge Graph built upon foundational and open Semantic Web technology standards, providing rich descriptions of data and flexibility end-to-end.  With the SPDL, biopharma researchers can:

  • Quickly on-board new data without requiring up-front modeling or mapping, ingesting data from any source versus months or weeks of preparation
  • Dynamically map and prepare data at analytics time
  • Horizontally scale in cloud or on-prem infrastructure to 100’s of nodes – allowing billions of facts to be analyzed, queried and explored in real-time   

The world’s BioPharma and research institutions are sitting on a wealth of highly differentiating and life-saving data and should begin to realize its value via Smart Patient Data Lakes (SPDL).

 

 

CONTACT: Nadia Haidar

Global Results Communications ∙ 949-278-7328 ∙ nhaidar@globalresultspr.com

 

Read Full Post »


A novel 5-gene pancreatic adenocarcinoma classifier: Meta-analysis of transcriptome data – Clinical Genomics Research @BIDMC

Curator: Tilda Barliya, PhD

Analysis of  Bhasin et al paper and Literature search

Table 1: 5-genes classifiers as biomarkers for PDAC:

Gene symbol Gene name Subcellular localization
ECT2 Epithelial cell transforming sequence 2 oncogene Nucleus, cytoplasm
AHNAK2 AHNAKE nucleoprotein 2 Plasma membrane, cytoplasm
POSTN Periostin, osteoblast specific factor Extracellular space
TMPRSS4 Transmembrane protease, serine 4 Plasma membrane

 

SERPINB5 Serpin peptidase inhibitor, clade B (ovalbumin) member 5 Extracellular space


Introduction
:

  • Bhasin et al, conducted a beautiful study using a powerful meta-analysis from different sources to identify the “important/classifier” genes associated with Pancreatic Cancer (PDAC).
  • The authors identified 5 genes that were considered as good classifiers (table 1).
  • It is important to note that the meta-analysis was performed on tissue and microdissection samples.
  • In their summary, the authors aim to validate these genes in blood/urine samples.
  • While these genes might be over expressed in tissue samples it may not be true to their existence in blood and careful examination and validation is required.
  • Liquid biopsies are emerging as the go-to use tools for disease detection, mostly aimed for early diagnosis.
  • Liquid biopsies are non-invasive biopsies of blood, urine (potentially saliva) and their “exotic” components, i.e miRNA, exosomes etc.
  • Since Liquid biopsies are non-invasive, they are painless and patients are more complied.
  • It is important to note that there is a gap between the expression of a gene or a protein in tissue section and their expression in the blood and may not necessarily correlate.
  • It will be very interesting to follow their research and future outcomes.

Additional References:

  • TMPRSS4: an emerging potential therapeutic target in cancer.

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4453593/

  • The tumour trail left in blood

http://www.nature.com/nature/journal/v532/n7598/full/532269a.html

Aashir Awan, PhD, wrote on 12/28/2016

I was wondering if these same 5 genes were upregulated in the pancreatic ductal adenocarcinoma cell lines that are available out there.  In doing cell biology work, there is always a dilemma whether cancer cell lines correctly re-capitulate in vivo tumors or not.  Personally, I prefer primary cell lines to do analysis but this finding can be used to test primary vs cell line.  In addition, I’ve attached the gene network for Ect2.  If you look carefully, the two big proteins that jump out are RACGAP1 and KIF23.  I think in designing therapies, combinatorial targets can yield the best outcomes.  Drugs directed towards two or more targets would seem ideal in my opinion.

ect2

Gene Network for Ect2

Original Research
Oncotarget. 2016 Apr 26;7(17):23263-81. doi: 10.18632/oncotarget.8139.

Meta-analysis of transcriptome data identifies a novel 5-gene pancreatic adenocarcinoma classifier.

Abstract

PURPOSE:

Pancreatic ductal adenocarcinoma (PDAC) is largely incurable due to late diagnosis. Superior early detection biomarkers are critical to improving PDAC survival and risk stratification.

EXPERIMENTAL DESIGN:

Optimized meta-analysis of PDAC transcriptome datasets identified and validated key PDAC biomarkers. PDAC-specific expression of a 5-gene biomarker panel was measured by qRT-PCR in microdissected patient-derived FFPE tissues. Cell-based assays assessed impact of two of these biomarkers, TMPRSS4 and ECT2, on PDAC cells.

RESULTS:

A 5-gene PDAC classifier (TMPRSS4, AHNAK2, POSTN, ECT2, SERPINB5) achieved on average 95% sensitivity and 89% specificity in discriminating PDAC from non-tumor samples in four training sets and similar performance (sensitivity = 94%, specificity = 89.6%) in five independent validation datasets. This classifier accurately discriminated PDAC from chronic pancreatitis (AUC = 0.83), other cancers (AUC = 0.89), and non-tumor from PDAC precursors (AUC = 0.92) in three independent datasets. Importantly, the classifier distinguished PanIN from healthy pancreas in the PDX1-Cre;LSL-KrasG12D PDAC mouse model. Discriminatory expression of the PDAC classifier genes was confirmed in microdissected FFPE samples of PDAC and matched surrounding non-tumor pancreas or pancreatitis. Notably, knock-down of TMPRSS4 and ECT2 reduced PDAC soft agar growth and cell viability and TMPRSS4 knockdown also blocked PDAC migration and invasion.

CONCLUSIONS:

This study identified and validated a highly accurate 5-gene PDAC classifier for discriminating PDAC and early precursor lesions from non-malignant tissue that may facilitate early diagnosis and risk stratification upon validation in prospective clinical trials. Cell-based experiments of two overexpressed proteins encoded by the panel, TMPRSS4 and ECT2, suggest a causal link to PDAC development and progression, confirming them as potential therapeutic targets.

KEYWORDS:

bioinformatics; biomarkers; meta-analysis; pancreatic cancer; transcriptome

PMID:
26993610
PMCID:
PMC5029625
DOI:
10.18632/oncotarget.8139

SOURCE

Oncotarget, Vol. 7, No. 17 – Referred as PDF, above

 

BIDMC Researchers Discover Early Indicators of Pancreatic Cancer

LibermannBhasin_PancreasCancerStudy

Markers may help doctors detect pancreatic cancer before it becomes deadly

In photo: First author Manoj Bhasin, PhD, and co-senior author Towia Libermann, PhD, Co-Director and Director of BIDMC’s Genomics, Proteomics, Bioinformatics and Systems Biology Center.

SOURCE

http://www.bidmc.org/News/PRLandingPage/2016/March/Libermann-Pancreatic-Cancer-Research-2016.aspx

BOSTON – Pancreatic cancer, the fourth leading cause of cancer death in the United States, is often diagnosed at a late stage, when curative treatment is no longer possible. A team led by investigators at Beth Israel Deaconess Medical Center (BIDMC) has now identified and validated an accurate 5-gene classifier for discriminating early pancreatic cancer from non-malignant tissue. Described online in the journal Oncotarget, the finding is a promising advance in the fight against this typically fatal disease.

“Pancreatic cancer is a devastating disease with a death rate close to the incidence rate,” said co-senior author Towia Libermann, PhD, Director of the Genomics, Proteomics, Bioinformatics and Systems Biology Center at BIDMC and Associate Professor of Medicine at Harvard Medical School (HMS). “Because more than 90 percent of pancreatic cancer cases are diagnosed at the metastatic stage, when there are only limited therapeutic options, earlier diagnosis is anticipated to have a major impact on extending life expectancy for patients. There has been a lack of reliable markers, early indicators and risk factors associated with pancreatic cancer, but this new way of differentiating between healthy and malignant tissue offers hope for earlier diagnosis and treatment.”

The investigators used a number of publicly available gene expression datasets for pancreatic cancer and developed a strategy to reanalyze these datasets together, applying rigorous statistical criteria to compare different datasets from different laboratories and different platforms with each other. The team then selected a subset of data for developing a panel for differentiating between pancreatic cancer and healthy pancreas tissue and thereafter applied this “Pancreatic Cancer Predictor” to the remaining datasets for independent validation to confirm the accuracy of the markers.

After demonstrating and independently validating that a 5-gene pancreatic cancer predictor discriminated between cancerous and healthy tissue, the researchers applied the predictor to datasets that also included benign lesions of the pancreas, including pancreatitis and early stage cancer. The predictor accurately differentiated pancreatic cancer, benign pancreatic lesions, early stage pancreatic cancer and healthy tissue. The predictor achieved on average 95 percent sensitivity and 89 percent specificity in discriminating pancreatic cancer from non-tumor samples in four training sets and similar performance (94 percent sensitivity, 90 percent specificity) in five independent validation datasets.

“Using innovative data normalization and gene selection approaches, we combined the statistical power of multiple genomic studies and masked their variability and batch effects to identify robust early diagnostic biomarkers of pancreatic cancer,” said first author Manoj Bhasin, PhD, Co-Director of BIDMC’s Genomics, Proteomics, Bioinformatics and Systems Biology Center and Assistant Professor of Medicine at HMS.

“The identification and initial validation of a highly accurate 5-gene pancreatic cancer biomarker panel that can discriminate late and early stages of pancreatic cancer from normal pancreas and benign pancreatic lesions could facilitate early diagnosis of pancreatic cancer,” said co-senior author Roya Khosravi-Far, PhD, Associate Professor of Pathology at BIDMC. “Our findings may open a window of opportunity for earlier diagnosis and, consequently, earlier intervention and more effective treatment of this deadly cancer, leading to higher survival rates.”

The first diagnostic application of the panel may be for analyses of fine needle biopsies routinely used for diagnosing pancreatic cancer and for determining the malignant potential of mostly benign pancreatic cysts that can sometimes be precursors of pancreatic cancer. In addition to providing a new tool for diagnoses, the research may also lead to new insights into how pancreatic cancer arises.

“Because these five genes are ‘turned on’ so early in the development of pancreatic cancer, they may play roles as drivers of this disease and may be exciting targets for therapies,” said Libermann. Most of the five genes—named TMPRSS4, AHNAK2, POSTN, ECT2 and SERPINB5—have been linked to migration, invasion, adhesion, and metastasis of pancreatic or other cancers.

The scientists next plan to evaluate the precise roles of the five genes and to validate the accuracy of their diagnostic assay in a prospective clinical study. “Moving forward, we will explore the potential to convert this tissue-based diagnostic into a noninvasive blood or urine test,” Libermann said.

“To further enhance the diagnostic power of this biomarker, we plan to expand it by including non-coding RNAs, proteins, metabolites and mutations associated with pancreatic cancer. This will result in development of the first of its kind biomarker that gauges pancreatic cancer alterations from multiple genomic angles for making highly accurate diagnoses,” added Bhasin. Such an inexpensive and simple test could help transform the landscape of pancreatic cancer and help prevent many of the estimated 330,000 deaths that it causes worldwide each year.

Study coauthors include BIDMC investigators Kenneth Ndebele, Octavian Bucur, Eric Yee, Jessica Plati, Andrea Bullock, Xuesong Gu, Eduardo Castan, Peng Zhang, Robert Najarian, Maria Muraru and Rebecca Miksad, and the University of Nebraska-Lincoln’s Hasan H. Otu. The work was supported by the National Institutes of Health, National Cancer Institute and Ben and Rose Cole Charitable Pria Foundation.

SOURCE

http://www.bidmc.org/News/PRLandingPage/2016/March/Libermann-Pancreatic-Cancer-Research-2016.aspx

Read Full Post »