+ Development of a serum protein assay for organ confined - - PowerPoint PPT Presentation

▶
development of a serum protein assay for organ confined
SMART_READER_LITE
LIVE PREVIEW

+ Development of a serum protein assay for organ confined - - PowerPoint PPT Presentation

Dec 07, 2023 221 likes •823 views

+ Development of a serum protein assay for organ confined prostate cancer 15th th June 2014 Steve Pennington UCD Conway Institute, UCD, Dublin + Protein Biomarker Discovery and Development Confirmation Validation/ Approval &

slide-1
SLIDE 1

+

Steve Pennington UCD Conway Institute, UCD, Dublin

Development of a serum protein assay for

  • rgan confined

prostate cancer

15thth June 2014

slide-2
SLIDE 2

+

Discovery

Sample accrual

Protein Discovery

Protein Identification and Characterisation

Other analytes

(anything measurable)

Confirmation Assay development

Antibody based

Western blotting ELISA

Mass Spectrometry based

Multiple Reaction Monitoring (MRM)

Multi-analyte assays

‘Robust‘ high- throughput assays

Additional clinical samples Large Multicentre Cohorts Large Scale Clinical Trials

Validation/ Qualification Approval & Adoption

Regulatory Authorities Clinician Adoption

Impact measurement

Clinical assays

Protein Biomarker Discovery and Development

Sample Numbers

slide-3
SLIDE 3

+

Discovery

Sample accrual

Protein Discovery

Protein Identification and Characterisation

Other analytes

(anything measurable)

Confirmation Assay development

Antibody based

Western blotting ELISA

Mass Spectrometry based

Multiple Reaction Monitoring (MRM)

Multi-analyte assays

‘Robust‘ high- throughput assays

Additional clinical samples Large Multicentre Cohorts Large Scale Clinical Trials

Validation/ Qualification Approval & Adoption

Regulatory Authorities Clinician Adoption

Impact measurement

Clinical assays

Protein Biomarker Discovery and Development

DISCOVERY VALIDATION CANDIDATES QUALIFICATION PANEL

Statistical Methods

slide-4
SLIDE 4

+Biomarker Futility

Specimens Fragmented approach

slide-5
SLIDE 5

+

2006

Clinical Utility

slide-6
SLIDE 6

+

How will we discover them? How will we measure them? How will we validate them? Will the protein biomarkers we discover be useful? How will we proceed to them gaining utility? How will we discover them? How will we measure them? How will we validate them?

2014

Clinical Utility: 8 years on

slide-7
SLIDE 7

+From Biomarkers to Diagnostics

Tests must have analytical validity, clinical value and financial value.

slide-8
SLIDE 8

+From Biomarkers to Diagnostics

Tests must have analytical validity as well as clinical and financial value.

Biomarkers should be fit for purpose and their purpose known

  • 1. Reform regulatory review
  • 2. Increase re-imbursement of tumour tests with

clinical utility

  • 3. Increase investment in research (cf. therapeutics)
  • 4. Increase rigour for assessment - publication
  • 5. Adhere to high-level evidence based

recommendations for use

slide-9
SLIDE 9

+

Can we identify and develop protein biomarkers of clinical value in prostate cancer ?

Tests to guide treatment decisions

slide-10
SLIDE 10

+Imagine this scene …..

slide-11
SLIDE 11

+Imagine the screen

http://www.thewell.ie/executive_ medicals_men.asp

Blood – FBC, Hb & Fe,

cholesterol, glucose, liver & kidney function

Urine Heart Hearing Vision

slide-12
SLIDE 12

+Imagine the screen

http://www.thewell.ie/executive_ medicals_men.asp

slide-13
SLIDE 13

+Imagine the screen

http://www.thewell.ie/executive_ medicals_men.asp

slide-14
SLIDE 14

+All clear doc?

✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓

PSA

✓ ✓ ✓ ✓ ✓ ✓ ✓

slide-15
SLIDE 15

+All clear doc?

✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓

PSA 14.2ng/ml

slide-16
SLIDE 16

+DRE

slide-17
SLIDE 17

+DRE

slide-18
SLIDE 18

+TRUS Biopsy

slide-19
SLIDE 19

+Gleason Scoring of Biopsy 3 4 5

slide-20
SLIDE 20

+So, the result…..

Gleason 3 + 4 “What now?”

DRE – abnormal PSA 14.2ng/ml

slide-21
SLIDE 21

+Decisions….

 The patient's treatment decision is a momentous one.  He must gather all the reliable information he can so he can

participate in the diagnostic process, then ultimately select the therapy most reasonable under the circumstances.

 As the patient confronts his condition - and he must do so - he

should take into account his personal goals regarding the available therapies and their peculiar morbidities.

 In his decision process he may get differing medical opinions

http://www.pccnc.org/patient_resources/ understanding_diagnosis/

Prostate Cancer Coalition

slide-22
SLIDE 22

+NCI Statistics

In Ireland: about 3000 men are diagnosed with prostate cancer very year (UK >25,000)

In the UK: one man dies of prostate cancer every hour

http://seer.cancer.gov/statfacts/html/prost.html

slide-23
SLIDE 23

+Personalised - Population

All 7’s aren’t equal

3+4 ≠ 4+3

slide-24
SLIDE 24

+NCI Statistics

In Ireland: about 3000 men are diagnosed with prostate cancer very year (UK >25,000)

In the UK: one man dies of prostate cancer every hour

http://seer.cancer.gov/statfacts/html/prost.html

Over-diagnosis and over-treatment is a major problem Most men die with rather than of prostate cancer But, there is currently no effective treatment for metastatic prostate cancer

slide-25
SLIDE 25

+Decisions, Decisions, Decisions Radical Prostatectomy (RP) Radiation (with hormones) No treatment (Active Surveillance)

slide-26
SLIDE 26

+

Normal BPH Confined Prostate cancer Non Confined Prostate cancer

PSA DRE Biopsy Surgery Active Surveillance Radiation

Diagnostic Test Diagnosis Treatment

Diagnosis and Treatment

26

RP no RP

slide-27
SLIDE 27

+

Can we identify and develop protein biomarkers of clinical value in prostate cancer ?

To guide treatment decisions

Accessible, Repeatable, Reliable

slide-28
SLIDE 28

PCa Multidisciplinary Teams

National Prostate Cancer Research Group

Prostate Cancer Research Consortium

UCD Conway Teams

slide-29
SLIDE 29

+Define the Clinical Question First

RP No RP

slide-30
SLIDE 30

+

PCRC Serum Sample Bioresource Biomarker Candidate list Biomarker discovery

2D-DIGE Label-free LC-MS/MS

Biomarker Panel Development

slide-31
SLIDE 31

+

 50 age matched serum samples from PCRC

 14 BPH, 36 PCa patients (Organ Confined and Non Organ Confined)

14 BPH and 36PCa patients

Discovery: 2D-DIGE

slide-32
SLIDE 32

+2D-DIGE candidates

slide-33
SLIDE 33

+

Affinity Depletion using MARS 14 column

Create reference pool sample from each pool depleted sample

GS5 (n = 10) GS7 OC (n = 10) GS7 NOC (n = 10) Depleted serum samples Protein concentration normalization

Protein assay and 1D gel In-solution digestion

Label-free LC-MS/MS

  • n Q-TOF

Peptide/protei n expression profile

Progenesis, database search and result filtering

Serum samples In-house MS/MS spectral library

TPP and Skyline

Public MS/MS spectral library

Trypsin digestion

Trans- Proteomic Pipeline

Mars 14 column

Discovery: Label free LC-MS/MS

slide-34
SLIDE 34

+Label free LC-MS/MS data

  • >90,000 features
  • Ion counting for quantification
  • Alignment using Progenesis
  • Mascot search for protein id
  • Mascot Score > 34 (FDR = 3.08%)
  • Remove non-unique mapping peptides
  • MS/MS library construction
  • Trans-Proteomic Pipeline (TPP)
  • Peptide to protein roll up
  • Analysis of differential protein expression
  • 59 Proteins differentially expressed (p-value<0.05)

Principle Component Analysis

1.E-08 1.E-07 1.E-06 1.E-05 1.E-04 1.E-03 1.E-02 1.E-01 1.E+00 0.1 1 10 ANOVA p-value Fold change ratio - drug treated/vehicle control 2 0.5

Protein Expression Changes Feature Alignment Feature Selection

(a) (b) (c)

Feature Quantification

slide-35
SLIDE 35

+

PCRC Serum Sample Bioresource Biomarker Candidate list Biomarker discovery

2D-DIGE Label-free LC-MS/MS Literatur e review

64 Candidate Proteins

PCRC OC Biomarker Candidates

slide-36
SLIDE 36

+

PCRC Serum Sample Bioresource Biomarker Candidate list Biomarker discovery

2D-DIGE Label-free LC-MS/MS Literatur e review

64 Candidate Proteins

MRM

Biomarker Validation

PCRC OC Biomarker Candidates

slide-37
SLIDE 37

+MRM

 Targeted approach for measuring multiple

proteins simultaneously

 Features:

 Dynamic range of >4 orders of magnitude  Up to 50 proteins per assay (more)  Can be quantitative: moles of protein of interest/g of

protein sample

 Very robust: CV’s of less than 10%  NOT as sensitive as ELISA in most cases

 Identify and measure peptide which is unique to

the protein of interest and measure it (mass/charge ratio) and fragments of it generated in the MS

slide-38
SLIDE 38

+Multiplexed quantification

16 Cytochrome P450’s

slide-39
SLIDE 39

+Another protein panel assembly

slide-40
SLIDE 40

+

Proteins - 57 Peptides - 174 Transitions - 1681

  • 8-10 transitions per peptide
  • 1-5 peptides per protein

15 injections of pooled sample (~ 13 hours instrument time) Survey run – determine detectability of peptides

Proteins - 52 Peptides - 119 Transitions - 609

  • 5 transitions per peptide
  • 1-5 peptides per protein

Collision energy optimisation 16 injections of pooled sample (~14 hours of instrument time)

Proteins - 48 Peptides - 109 Transitions - 545

  • 5 transitions per peptide
  • 1-5 peptides per protein

Final SRM method Technical variance measurement 10 injections pooled sample (~17 hours instrument time) Mean CV = 5.7 % Measurement in 30 individual samples (~51 hours instrument time)

  • drug treated or vehicle control)

Refined method Initial SRM method

MRM development pipeline

slide-41
SLIDE 41

+

0.1 1 10 100 1000 FC - veh/ veh D4 FC - veh/ veh D15 FC - low/ veh D4 FC - low/ veh D15 FC - high/ veh D4 FC - high/ veh D15

Fold change ratio Treatment group

Housekeeping

0.1 1 10 100 1000 FC - veh/ veh D4 FC - veh/ veh D15 FC - low/ veh D4 FC - low/ veh D15 FC - high/ veh D4 FC - high/ veh D15

Fold change ratio Treatment group

Proteomics (label-free LC-MS)

0.1 1 10 100 1000 FC - veh/ veh D4 FC - veh/ veh D15 FC - low/ veh D4 FC - low/ veh D15 FC - high/ veh D4 FC - high/ veh D15

Fold change ratio Treatment group

Transcriptomics (Affy array)

0.1 1 10 100 1000 FC - veh/ veh D4 FC - veh/ veh D15 FC - low/ veh D4 FC - low/ veh D15 FC - high/ veh D4 FC - high/ veh D15

Fold change ratio Treatment group

Literature

MRM measurement:48 proteins

slide-42
SLIDE 42

+

Workflow Map

31 Candidates MRM Transitions

PCa OC Candidate Biomarkers

slide-43
SLIDE 43

+Candidate Biomarker MRMs

slide-44
SLIDE 44

+

OC (GS6 and 7) and NOC (GS7) OC(GS7) and NOC (GS7)

PLS-DA with 200 times bootstrapping

AUC=0.82 AUC=0.78

Prediction of Organ Confinement (initial data)

slide-45
SLIDE 45

+

Biomarker Candidate list Biomarker discovery

2D-DIGE Label-free LC-MS/MS Literatur e review

64 Candidate Proteins

MRM

Biomarker Validation

Use global data to assemble panel

136 Candidate Proteins Movember GAP

1st Generation 2nd Generation

slide-46
SLIDE 46

+

Biomarker Prioritization Biomarker assembly

Biomarker measurement (now)

Agilent 6490 Triple Quad with UPLC: Agilent Partner Lab X Candidate Proteins

MRM

Biomarker Validation

Samples

Assembly of Reference Pool (method development and QC) Test (150) Samples: False Indolent; True Indolent

slide-47
SLIDE 47

+Conclusion?

✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓ ✓

PSA 14.2ng/ml Blood Test for Organ Confinement Clinical assay

Best Decision for Individual Patient

slide-48
SLIDE 48

+

 ‘End user’ driven question/ clinical need  Design of discovery experiment(s) to match clinical

question

 Well planned validation strategy ….. sample numbers

and type

 Incorporation of appropriate statistical methods

 For selection of candidates from discovery  For selection of signatures from candidate panels

 Then, science ends … product development begins

Clinical Utility: What will it take?

slide-49
SLIDE 49

PCa Multidisciplinary Teams

National Prostate Cancer Research Group

Prostate Cancer Research Consortium

Movember GAP

UCD Conway Teams

slide-50
SLIDE 50

Acknowledgements

Prostate Cancer Research Consortium

Teams: Nurses, clinicians, pathologists, training clinician scientists, non-clinical scientists, research assistants

The PATIENTS Cathy Rooney

Giuliano Elia Kieran Wynne Christine Miller

Ben Collins Yue Fan Brian Morrissey Rosanna Inzitari Lisa Staunton Claire Tonry Belinda Long Andrew Parnell

Opeyemi Ademowo, Jian Chen, Trevor Clancy, Moyez Dharsee, Ken Evans, Lorelei Mucci, Kristen Tasken, Bill Watson, Brian Flately

Movember Serum GAP Team

slide-51
SLIDE 51
slide-52
SLIDE 52

+

“The philosophies of one age have become the absurdities of the next…..”

William Osler

slide-53
SLIDE 53

+MRM for Lung Cancer

  • Used a systems biology strategy to identify 371 protein

candidates

  • Developed a multiple reaction monitoring (MRM) assay for each.
  • MRM assays applied in a three-site discovery study (n = 143)
  • Used plasma samples from patients with benign and stage IA lung

cancer

  • Produced a 13-protein classifier.
  • Classifier validated on an independent set of plasma samples (n =

104) exhibiting a negative predictive value (NPV) of over 90%.

slide-54
SLIDE 54

MRM run order: Randomised

B1 B2 QC B3 21 B3 22 B4 23 B5 24 B6 25 B6 26 B7 27 B8 28 B9 29 B10 30 B11 31 B12 32 B13 QC B14 33 B15 34 B16 35 B17

36 B18 55 B19 56 B20 57 B21 58 B22 62 B23 63 B24 64 B25 65 B28 QC B29 B30 QC Crude Serum Sample Blank

slide-55
SLIDE 55

+Current Biomarker Pipeline

Discovery Assay Development Validation Approval /Adoption

Programme

Prostate Cancer

63/64 47/102

Psoriatic Arthritis Pre-Clinical Tox (Liver)

48/48

Cardiovascular

24/24

Breast Cancer = Intellectual Property Filings

Numbers: MRMs developed/Candidates

14 P450’s

Cytochrome P450s MIAMI

500 patient samples

slide-56
SLIDE 56

+

Anderson and Anderson, Mol. Cell. Proteomics 2002 1: 845

Abundant protein removal

slide-57
SLIDE 57

+Serum Proteins: Dynamic Range

slide-58
SLIDE 58

+Serum Proteins: Dynamic Range

MRM Luminex (120)

slide-59
SLIDE 59

+Serum Proteins: Dynamic Range

MRM Luminex (120)