lgc science & technology conceptual challenges of rms in … · 2016-06-03 · proteomic,...
TRANSCRIPT
LGC SCIENCE & TECHNOLOGY
Conceptual challenges of RMs in support of
multiparametric biomarker measurements: A changing healthcare paradigm?
S.Pang, J.Huggett, W.Burkitt, C.Pritchard, G.O’Connor, C.Foy
& HELEN PARKES, LGC, UK
“Future of Reference Materials Science & Innovation”,
IRMM, 24 November 2010
21st Century Healthcare
• Increase in ageing population developing complex diseases such as cancer, Alzheimer’s and heart disease - burden on healthcare systems
• These complex, multifactor disorders are challenging diagnosis, treatment and monitoring
• Advances in ‘omics’ technologies are leading to identification of many novel putative disease related biomarkers /biomarker panels
• Promises significant advances in healthcare management for the future
• However, these emerging clinical diagnostics and therapeutics have a number of related measurement challenges
New Measurement Challenges:
• Increasingly, panels of biomarkers or multiparametric “fingerprints” derived from multiple data types are needed to give accurate diagnosis of complex disorders.
• Novel measurement issues to be addressed including:-
– Multiple analytical platforms - issues of comparability
– Large dynamic range of putative biomarkers in blood
– Detection of overall shifts and trends in the datasets (cf accurate quantification of individual analytes)
– Combination of genomic, transcriptomic, epigenomic, proteomic, metabolomic measurements may be required
Setting new demands for measurement traceability and design of appropriate reference standards
Normal Disease
Development of RNA (gene expression) and
protein biomarker profiling standards - A
quality metric “tool kit” for biomarker discovery and validation
Biomarker Issues
• Rate of approval of validated biomarkers is low as companies often fail to develop robust validation strategies– Assessing measurement
performance characteristics is a critical step
• Broad dynamic range of proteins in plasma – No single platform enables
detection of all proteins within plasma
• Issues span multiple technology platforms
• Overlap of results not guaranteed between platforms– Difficulties in validating
findings
Biomarker Technical Validation
(FDA)
To meet technical requirements the analytical process must be shown to be “fit for purpose”:-
• Performance evaluation of reproducibility, accuracy, sensitivity etc.
• Robustness across multiple labs/platforms/technologies
This requires:-
• Standards and control materials
• Process definition
– Analytical guidelines to use multivariate tools appropriately
• Quality metrics
– To allow objective assessment of data
Candidate protein biomarker
panel - reference material
production
• Preparation of a mixed panel of proteins with characterisation data
– 6 non-human proteins sourced from commercial suppliers (mouse CCL6, mouse lungkine, chicken caronte, mouse soggy, firefly luciferase and chicken egg lysozyme)
– Protein abundance: 50 – 10e7 pg/mL.
– Gravimetrically prepared a 10x stock of protein panel in buffer.
– Homogeneity, short and long term stability studies were performed.
Panel evaluation
• Two immunoassay platforms (Luminex and MSD) and a mass spectrometry
platform (QSTAR) were used for the study.
• Demonstrate applicability of panel for evaluating measurement
performance:
- Reproducibility, sensitivity, accuracy, dynamic range, signal to noise,
matrix effects.
- Platform performance indicator (by spiking protein panel into normal and
ovarian cancer diseased individual donor plasma samples).
Performance of the QC markers
- ovarian cancer biomarkers
Caronte
1
10
100
1000
0 1 2 3 4 5 6 7 8 9 10 11 12
Plasma donor
Mean s
ignal outp
ut (A
U)
Expt 1
Expt 2
Expt 3
Lysozyme
1
10
100
1000
0 1 2 3 4 5 6 7 8 9 10 11 12
Plasma donor
Mean
sig
nal
ou
tpu
t (A
U)
Expt 1
Expt 2
Expt 3
Lungkine
1
10
100
1000
0 1 2 3 4 5 6 7 8 9 10 11 12
Plasma donor
Mean
sig
nal
ou
tpu
t (A
U)
Expt 1
Expt 2
Expt 3
Soggy
1
10
100
1000
10000
100000
0 1 2 3 4 5 6 7 8 9 10 11 12
Plasma donor
Mean
sig
nal
ou
tpu
t (A
U)
Expt 1
Expt 2
Expt 3
CCL6
1
10
100
1000
0 1 2 3 4 5 6 7 8 9 10 11 12
Plasma donor
Mean
sig
nal
ou
tpu
t (A
U)
Expt 1
Expt 2
Expt 3
Luciferase
1
10
100
1000
10000
100000
0 1 2 3 4 5 6 7 8 9 10 11 12
Plasma donor
Expt 1
Expt 2
Expt 3
Donors 1-6: normal plasma;
Donors 7-12: ovarian cancer diseased plasma
• 3 component proteins of the QC material produced the same signal output when spiked in distinct individual donor plasma samples
- these may serve as generic indicators of the platform function
• Others all exhibit similar trends in sensitivity to matrix effects among the 12 donor samples
- these may act to indicate that sample matrix may interfere with the assay for diagnostic biomarker
•Evaluated with IL8 biomarker
Genomic Approach -• Collaboration with NIST using ERCC (External RNA Controls Consortium)
materials– set of 96 RNA control transcripts (non human/artificial) developed to assess
technical performance of gene expression assays
• Calibrator samples for gene expression multi-analyte measurements covering a range of concentrations
– Pool of RNA transcripts (8-20) of variable known abundance
– Multiple pools constructed with relative amounts of transcripts varied
• Used to monitor multi-parameter measurements – evaluate dynamic range, comparability and technical precision of assays
– Microarrays, qRT-PCR, digital PCR, Next Gen Sequencing….
• Transcripts “certified” by a range of approaches:-– International qRT-PCR inter-lab studies (CCQM - P103)
– Digital RT-PCR
• Transcripts spiked-in to a background of total RNA
International comparisons - (CCQM P103)
‘biomarker’
K-61
RNA instead of
DNA
single RNA
transcript,
matched calibrant
single RNA
transcript with
matched
calibrant in
matrix
multiple RNA
transcripts,
matched
calibrant
multiple “biomarker”
panel
RNA spike in a
complex
background of
human RNA
multiple RNAs
in a panel
RNA spikes in a
complex
background of
human RNA
- Multiplex measurement
C
CQ
M
B
AW
G
Development of traceable reference
methods for structural measurements of
clinical protein biomarkers
Biomolecular structure / function:
• WHO report highlighted the increased recognition of structure function links for biologicals
• A protein’s structure, folding state and interactions with other proteins/ligands define its [immuno] activity
• Relationship between the measured “signal” and the “clinical” value, depends on many influences and effects. Variability in protein structure has a significant influence on signal measured.
Proteins - 100-1000 times larger than small molecules
Heterogeneous - single AB may have 35 glycovariants
Complex biomolecular traceability:
• Key issue for biomolecular traceability, is not simply SI traceable measurement of “amount of substance”, as achieved by IDMS for small molecules - challenge is to progress beyond state of the art and develop SI traceable reference measurements and RMs linked to biomolecule “function/activity”
• Requires advances in measurement science, providing a route to traceable complex protein measurements that will impact significantly on healthcare measurement comparability
Complete digestion to constituent peptides
Add isotopically labelled peptides to protein
*
*
*
*
Complete hydrolysis to
constituent amino acids
*
*
*
*
Add isotopically labelled amino acids to peptide
WP3
WP4
Labelled Peptides (Secondary Standard)
Quantitative separation of target
protein from clinical
matrix
WP5Target Protein
H/D exchange and
IMS studies to
characterise
structure
WP6
Immunoassay
StudiesWP7
Co
mp
ari
so
n
IDMS Quantitation of
Amino Acids
Select clinically
relevant model proteins
WP2
Patient Diagnosis & Care
Labelled Amino Acids (Primary Standard)
IDMS Quantitation of
Amino Acids
Labelled Peptides (Secondary Standard)
Pure Protein Standard
Matrix Protein Standard
IDMS
Quantitation
CLINBIOTRACE - Traceability of complex biomolecules and biomarkers in diagnostics – effecting measurement comparability in clinical medicine
hGH
CRP
“Model” clinical biomarkers:
• Human Growth Hormone (hGH / rhGH / somatropin)
– Growth defect marker/ biopharmaceutical
– multimer formation, isoforms
– plasma-based heterogeneity
• C-Reactive Protein (CRP)
– Inflammmation Cardiac marker
– oliogomer formation, additives
Aim: develop methods for traceable quantification of the proteins tertiary structure and relate to protein activity(eg immunoassay)
Structural measurement
• A number of physical and immuno-affinity based separation methods are being investigated:
• HD exchange MS (HDX) enables interrogation of protein structure under physiological conditions.
– By monitoring the exchange rates of different peptides an assessment of solvent proximity and protein folding structure can be determined
– Can be performed in the presence of specified ligands enabling the determination of binding constants and the degree of activity.
• Ion mobility separates ions on the basis of collisional cross section. This enables proteins of the same mass but different folding states to be separated.
• In combination with high resolution chromatography, capable of separating protein isoforms, we aim to traceably quantity the active form
HDX: recombinant human growth hormone (somatropin)
• Issues with protein solubility
• Measured in water – pH of starting solutions unknown
0%
20%
40%
60%
80%
100%
10 100 1000 10000 100000
Time /s
%D
eu
teri
um
In
co
rpo
ratio
n
EP
WHO
time
Differences
0%
20%
40%
60%
80%
100%
10 100 1000 10000 100000
Time /s%
De
ute
riu
m I
nc
orp
ora
tio
n
EP
WHO
Work undertaken by Will Burkitt & Caroline Pritchard,LGC
WHO
EP
drift time
%%
drift time (bins)
m/z
m/z
EP
WHO
IMS differences observed between WHO/EP standards, analysis on SYNAPT TWIMS
Somatotropin/hGH - structure Ion mobility mass spectrometry (IMS)
Work undertaken by Caroline Pritchard,LGC
Clinbiotrace - progress• hGH
– hGH quantitation in serum
– Isoform study undertaken
– Correlating SI traceable LCMS quantified hGH, & HDX structural analysis with routine IVD methods via a commutability study
• CRP
– Structure defining immunoassay result - monomer /pentamer differences
– LC/MS approaches to quantify oligomeric states
– [IMS aggregation measurement)
– Link to commutability study results
Real metrics on how structure can influence in vitrodiagnostic method - inform commutable CRM development & MU determinations
Summary
• We have developed a number of approaches for development of protein and nucleic acid biomarker reference materials -predominantly “calibrants”, in support of biomarker discovery and valid measurement
• Issues of rapidly changing technologies - RMs should be platform independent (microarray experience…)
• Key issue - how to establish traceability? - particularly for multiparametric measurements?
• Clinbiotrace project investigations of MS (IDMS, LCMS, HDEX, IMS) approaches to protein [structure] quantification have started toaddress traceability for complex protein biomarkers and link to immunoassay performance - but long way to go!
• Resource intensive and slow process - collaboration important
Acknowledgements
• Colleagues at LGC
– Will Burkitt
– Carole Foy
– Jim Huggett
– Gavin O’Connor
– Susan Pang
– Caroline Pritchard
– Milena Quaglia
• EMRP collaborators
– Heinz Schimmel, Ingrid Zegers IRMM
– Andre Henrion PTB
Where we are heading…..
• Development of reference methods/materials for trace detection and absolute nucleic acid quantification (digital PCR) - defined copy number
• Development of reference methods for multi-parametric nucleic acid measurements
• Development of reference standards for miRNA analysis
• Developments of reference methods and materials for DNA methylation measurements
• Standardisation framework for NGS (sample and library preparation)
• Validation of alternative DNA amplification approaches for POC diagnostics (microbiology/ infectious disease focus)
• Provision of traceable standards for the standardisation of multi-parametric platforms….