The University of Queensland Diamantina Institute Turning scientific - - PowerPoint PPT Presentation

Dr Michelle Hill

The University of Queensland Diamantina Institute

“Turning scientific discoveries into better treatments...”

The University of Queensland Diamantina Institute (UQDI) was established on 1st January 2007 as the sixth research institute of The University of Queensland. The location of UQDI on the Princess Alexandra Hospital campus is key to the institute’s mission, “to translate scientific discoveries into better treatments”. UQDI is now part of Translational Research Institute (TRI), a state-

• f-the-art facility housing four

research institutes to promote better collaborative innovation. TRI houses 650+ researchers and is the first of it’s kind in Australia, allowing biopharmaceuticals and treatments to be discovered, produced, clinically tested and manufactured in the one location.

Research at UQDI

Immune-related diseases

• Rheumatoid arthritis
• Type 1 diabetes
• Infection & immunity

Cancer

• Skin cancers
• Head & Neck cancer
• Cancer vaccine &

immunotherapy

Genomics & Proteomics technology

• Susceptibility genes, disease etiology
• Biomarkers

Omics in translational biomarker research

Biomarker

• Measurable attribute that can be used to

indicate or predict physiological status

– Blood pressure – Imaging – Metabolite/chemicals (e.g. blood glucose) – Genomic (mutation or expression level) – Protein (e.g. PSA)

Simon 2011 EMBO Mol Med 3, 429

Genomic Proteomic Metabolomic Imaging Genomic Pharmacogenomic Proteomic Metabolomic Imaging

Prognostic Theragnostic Predictive

Annual PubMed records for ‘diagnostic biomarkers’ (Moschos 2012 Bioanalysis)

Approved In vitro diagnostics (IVD)

In vitro diagnostic (IVD)

“reagents, instruments, and systems intended for use in the diagnosis of disease or other conditions, including a determination of the state of health, in order to cure, mitigate, treat or prevent disease...”

(FDA guideline, Title 21 of Code of Federal Regulations)

Classification based on:

• Intended use

– what the test measures (biomarkers)

• Indications for use

– why a patient would be tested

Approval requires:

• Preclinical evaluation

– demonstrates accurate and reproducible measurements

• Clinical performance

– shows that the device provides the expected results in a defined patient population for intended use

Mansfield et al. 2005 J Mol Diag

Bridges over the valley of death: From biomarker to IVD

• Clearly defined clinical intended use
• Sufficient preliminary evidence from multiple cohorts
• Select/develop suitable clinical assays
• Design appropriate clinical trial for regulatory approval

Vidal et al. 2012 Clin Proteomics

Multidisciplinary & multi-centre: Clinical – sample collection with controlled standard procedures Technology – establish standard measurement conditions Informatics & Statistics – consistent rigorous analysis Team decision making – when to drop biomarkers

National Cancer Institute Early Detection Research Network (EDRN)

2001-2003

biomarker development paradigm, standardize collection and banking of non-invasive biosamples with comprehensive clinical data

2006-2008

Prospective Specimen Collection Retrospective Blinded Evaluation (PRoBE) design for phase 2 and 3 biomarker validation trials

1998-2000

inception & inauguration

2003-2005

establish partnerships, collaborative projects, bioinformatics tools

2008- present

delivery of clinically useful biomarkers: 300+ passed phase 2 (300+ did not) 1450+ publications 28+ patents, 14+ licences

Adapted from Srivastava 2013 Clin Chem

EDRN five phase biomarker development paradigm

Phase 1

• Preclinical discovery: Distinction between normal and cancer

Phase 2

• Preclinical verification: Reproducibility of markers
• Development of suitable clinical assay: Portability of assay format

Phase 3

• Preclinical validation: Evaluation of sensitivity & specificity for clinical

indication Phase 4

• Clinical evaluation: Estimation of false positive and false negative rates

Phase 5

• Disease Control: Evaluation of overall benefits & risks of the test

# # Number of targets Number of samples

2001-2003

biomarker development paradigm, standardize collection and banking of non-invasive biosamples with comprehensive clinical data

2006-2008

Prospective Specimen Collection Retrospective Blinded Evaluation (PRoBE) design for phase 2 and 3 biomarker validation trials

1998-2000

inception & inauguration

2003-2005

establish partnerships, collaborative projects, bioinformatics tools

2008- present

delivery of clinically useful biomarkers: 300+ passed phase 2 (300+ did not) 1450+ publications 28+ patents, 14+ licences

Adapted from Srivastava 2013 Clin Chem

National Cancer Institute Early Detection Research Network (EDRN)

Omics discovery $$$$ Clinical assay $

EDRN five phase biomarker development paradigm

Phase 1

• Preclinical discovery: Distinction between normal and cancer

Phase 2

• Preclinical verification: Reproducibility of markers
• Development of suitable clinical assay: Portability of assay format

Phase 3

• Preclinical validation: Evaluation of sensitivity & specificity for clinical

indication Phase 4

• Clinical evaluation: Estimation of false positive and false negative

rates Phase 5

• Disease Control: Evaluation of overall benefits & risks of the test

# # Number of targets Number of samples

Moschos 2012 Bioanalysis 4, 2499

Less than ¼ of new molecular in vitro diagnostics approved by US FDA since 1995 use nucleic acid biomarkers

Which Omics for biomarker discovery?

Moschos 2012 Bioanalysis 4, 2499

Assay capable of predicting drug dose, efficacy or safety risk of a drug

Companion diagnostic

Manjili et al. 2012 Future Onc 8, 703

In vitro diagnostic multivariate index assay (IVDMIA)

An IVD that measures 2 or more independent variables in parallel, and a scoring algorithm

Moschos 2012 Bioanalysis 4, 2499

Why proteins?

• Simple, specific and sensitive assay with antibodies
• Proteome but not genome rapidly modulated by

disease/treatment : disease detection vs disease risk

• Can be actively released or shed from cells
• Body fluid detection desirable over tissues because:
• less invasive, allows repeated sampling
• reduced sampling error (tumour heterogeneity)
• ability to detect microenvironmental changes (increase

specificity or sensitivity)

Less than ¼ of new molecular in vitro diagnostics approved by US FDA since 1995 use nucleic acid biomarkers

Anderson & Anderson 2002 MCP

Comparative proteomics in blood is challenging!

Strategies for body fluid protein biomarker discovery

• Depletion of abundant proteins
• Target sub-proteome

– Glycoproteins – Exosome/microvesicles (small circulating vesicles, also contain DNA and RNA)

Glycosylation changes of circulating proteins as biomarker

• Glycosylation changes implicated in cancer pathogenesis
• Change in glycosylation machinery in the cancer cell
• Neo-expression in stromal cells, which has a different profile of

glycosyltransferases

• Lectins as affinity reagent which binds to specific glycan structures: readily

adaptable for clinical assay

• Glycosylation changes more specific than changes in protein, e.g. AFP-L3 test for

fucosylated form of a-fetoprotein

Lectin Abbrev. Ligand moiety Related cancers Aleuria aurantia lectin AAL Core fucosylation Liver, lung, breast, colon, pancreatic, esophageal etc. Helix pomatia agglutinin HPA GalNAc Breast cancer Elderberry lectin SNA α2-6-linked sialic acid Pancreatic cancer

Lectin-magnetic bead array-coupled mass spectrometry (LeMBA-MS) for glyco-biomarker discovery

GlycoSelect database biomarker selection pipeline for LeMBA-MS

Data entry/storage Lectin-protein pairs Analysis

• 1. Patient selection
• 2. Normalize to internal standard
• 3. Sample outlier detection
• 4. Identify on/off changes using group

difference tool

• 5. Ranking of quantitative changes

using sPLS-DA

(sparse Partial Least Squares regression- Discriminant Analysis, Le Cao et al. 2011 BMC Bioinformatics) David Chen, Kim-Anh Le Cao

Phase 1 discovery for oesophageal adenocarcinoma (EAC)

Goal: Obtain a list of differentially glycosylated serum proteins in oesophageal adenocarcinoma (EAC) by comparing with matched samples from the pre-cancer condition Barrett’s oesophagus (BE) and controls. Multidisciplinary team PhD student ( Alok Shah) Epidemiologist & biological samples (David Whiteman, Australian Cancer Study, PROBE-NET) Oncology surgeon (Andrew Barbour) Informatics (David Chen) Biostatistics (Kim-Anh Le Cao) Nanotechnology for diagnostic device (Matt Trau) Discovery cohort

Phase 1 discovery LeMBA-MS/MS outcomes

Phase 1

• Preclinical discovery: Distinction between normal and cancer

Phase 2

• Preclinical verification: Reproducibility of markers
• Development of suitable clinical assay: Portability of assay format

Phase 3

• Preclinical validation: Evaluation of sensitivity & specificity for clinical

indication Phase 4

• Clinical evaluation: Estimation of false positive and false negative

rates Phase 5

• Disease Control: Evaluation of overall benefits & risks of the test

# # Number of targets Number of samples

Summary

• Clinical utility of biomarkers when translated

into IVD.

• Discovery and development requires multi-

disciplinary, multi-institutional team.

• Omics technologies with rigorous statistical

assessment essential in biomarker discovery.