[PPT] - Princess Marga ret Hospital and Ontario Can Ontario Can ncer PowerPoint Presentation

SLIDE 1

Princess Marga Ontario Can Ontario Can 50th Annive 50th Annive 16 – 18 Oc

“The Unexpected Consequence PMH/OCI – Psychoso

Peter

ret Hospital and ncer Institute ncer Institute ersary Event ersary Event ctober 2008

es of Being a Research Fellow at

cial and Proteomics”

Selby

SLIDE 2

Q ti Question

Does measuring “patien formally, lead to improv nt-centred” variables ved outcomes?

SLIDE 3

Patient-centred, self-r li i l clinical

A personal perspective A personal perspective

there is a NEED for PCSR
it can increase DISCOVER
it is now LOGISTICALLY p

th d t manage the data

its use MAY IMPROVE IMP

settings settings

we are NOT YET VERY GO

reported information in ti practice

information in practice RY of the relevant issues possible to collect and PORTANT OUTCOMES in some OOD at collecting or using the data

SLIDE 4

Patient-centred, self-r li i l clinical

Challenges

clear conceptual frameworks
what to collect and when
how to train clinicians and alt
how to adapt healthcare syst

p y

how to change the culture in
how to ENHANCE (not replac
how to ENHANCE (not replac

patients, carers, healthcare p

reported information in ti practice

and choice of topics ter clinical behaviour ems healthcare ce) the interactions between ce) the interactions between professionals and healthcare systems

SLIDE 5

Patient-centred, self-r clinical clinical

DISCOVERY DISCOVERY

Fatigue Pain N & V Dysnoea Sleep Appetite Constipation Constipation Diarrhoea Finance

reported information in practice practice

114 Patients reporting

n QLQ C30

moderate problems % in notes 34 29 24 58 35 14 18 56 27 7 18 39 13 8 13 8 4 25 14 7

JCO 2002

SLIDE 6

LOGISTICS

Touch-screen data collection

C S ( C O

Comparison TS vs paper (J Clin Onco
Patient compliance with regular QOL

(J Clin Oncol, 2003)

100 50 60 70 80 90 100 ent complete (%)

1264 198 339 110 48 242

10 20 30 40 1st visit 2nd visit 3rd visit 4th visit 5 QL assessme

98 147 114 8

1st visit 2nd visit 3rd visit 4th visit 5 visit

)

l, 1999)

L collection

17 tients

120 110

5th visit

83 Number of pat

100 90 80 70 60 50 40 30

5th visit

40 20

20
40
60
100

30 20 10

SLIDE 7

Randomised Tria

Patients starting chem Patients starting chem

I t ti 50% Att ti t Intervention 50% Attention-cont

EORTC QLQ-C30 EORTC QLQ-C3 HADS on Touchscreen HADS on TS F db k N f db k Feedback No feedback

Process outcomes: tape-recording of con Patient outcomes

FACT G (QOL Questionnaire) FACT-G (QOL Questionnaire) Continuity & Co-ordination of Care Satisfaction Satisfaction

al – Study Design

mo-/biological treatment mo /biological treatment

randomized t l 25% C t l 25% trol 25% Control 25%

30 No QL in clinic nsultations – content analysis @ baseline t 3 i t ti post 3 interventions @ 4 months @ 6 months

SLIDE 8

Results – Patient well Proportion of patients

NNT

Proportion of patients

r deterioration in FAC

NNT

Interv vs Attn-contr +Contr p=0.007,

Improvement No

14% 7%

80% 100%

46% 61%

60% 80%

40% 32%

20% 40%

32%

0%

Intervention Attention-c

being

s with improvement

T = 4 2

s with improvement CT-G scores

T = 4.2

Interv + Attn-contr vs Control p=0.003

change

Deterioration

31% 45% 45% 24% control Control

SLIDE 9

Further analysis of commun

Symptoms

Providing QOL data lead to

g more consistent discussion of

Insomnia (p=0.003)

Dyspnoea (p=0 03)

Dyspnoea (p=0.03)
Symptoms more often raised by

doctor

Discussion of common

symptoms depended mainly on whether the problem was raised at baseline

Pain, fatigue, nausea,

appetite appetite

ication and decision making

Functioning

Providing QOL data lead to
Providing QOL data lead to

more consistent discussion of

Physical function (p=0.006)
Emotional function (p=0.03)
Not initiated by doctor

y

No effect on social function

SLIDE 10

Health-Related Quality o Patient-Physician y

Detmar, Aaro

Randomised Trial: 214 patien Use of QLQ-C30 led to:

more HRQL issues disc
more health problems id
support from patients a
NO CHANGE IN QL (o

NO CHANGE IN QL (o

improvements in EMOT

and ROLE FUNCTION

f Life Assessments and Communication

nson et al, 2002

nts / 10 physicians cussed dentified nd staff

n SF 36)
n SF 36)

TIONAL FUNCTION (.04) (.05) ( )

SLIDE 11

Health-Related Qua and Satisfaction S

140 100 120 140 60 80 100

F L IC

20 40 60 1 2 3 Mont

Control-FLIC Structured Intervention-FLIC Assessment Control-Satisfaction

ality of Life (HRQL) Scores Over Time

100 70 80 90

n

40 50 60 70

S a tis fa c tio

20 30 40 4 5 6 th

Assessessment Control - FLIC Control-Satisfaction Structured Interview-Satisfaction

SLIDE 12

The Social Difficulties Inventory (SDI)

SLIDE 13

A meaningful SD g

Individual items

Frequency of so

700 400 500 600

f people

100 200 300 number o n d e p e n d e n c e m e s t i c c h

r

e s P e r s

n

a l c a r e

f

d e p e n d e n t s

r

d e p e n d e n t s e l f a r e b e n e f i t s F i n a n c e s n c i a l s e r v i c e s W

r

k n i n g t h e f u t u r e i c a t i

n

g c l

s

e n i c a t i n g

t

h e r s S e x u a l m a t t e

h

a v e I n d D

m

e P e C a r e

f

S u p p

r

t f

r

W e l f F i n a n c P l a n n i C

m

m u n i c C

m

m u n i c S e x P l a n s t

h

SDI items

DI scoring system g y

cial difficulties

very much difficulty quite a bit of difficulty a little difficulty no difficulty t e r s v e a f a m i l y B

d

y i m a g e I s

l

a t i

n

G e t t i n g a r

u

n d W h e r e y

u

l i v e R e c r e a t i

n

H

l

i d a y s O t h e r B G e W h

SLIDE 14

A meaningful SDI scorin

Independence Domestic chores Personal care Care of dependents Support for those close to you Welfare benefits Finances Financial services Work Planning the future Communication with close Comm nication ith others Communication with others Sexual matters Plans to have a family Body image Isolation Isolation Getting around Where you live Recreation Holidays Holidays Other

ng system-Rasch analysis

Measure of social distress (SD)

16 items Unidimensional 72% of the variance Differences in scores are equally spaced Interval scale Interval scale Age, gender, stage & site of disease, deprivation

SLIDE 15

Results: deriv

Social Distress (SD) ( )

(16 Rasch items)

T 10% f h SD

Top 10% of researcher SD

≥ 14 (gold standard for SD)

Using this gold standard

the best ‘cut-point’ was the best cut-point was patient SD ≥ 10

ving a cut point g p

Area under the curve = .850

1.0

ROC Curve

0.6 0.8

vity

0.4

Sensitiv

0.0 0.2 0.4 0.6 0.8 1.0

1 S ifi it

0.0 0.2

sensitivity = 800

1 - Specificity sensitivity = .800

specificity = .755
positive predictive value =.29

SLIDE 16

Summ

We have the technology Routine QL assessment does im There is potential for making que i di id l ti t individual patients Future studies Future studies

Making assessments more individu
The roles undertaken by different
The roles undertaken by different

members of the multidisciplinary te

How people make decisions

p p

Staff training & patient information

needs

Widening access—web based

systems

mary

mprove QL in some studies estions asked more specific to

ual am

SLIDE 17

Galina Velikova Penny Wright

SLIDE 18

Question Question Can proteomic approach Can proteomic approach and evaluation improve patients and their outco patients and their outco hes to biomarker discovery hes to biomarker discovery

ur management of
mes?
mes?

SLIDE 19

Leed

Renal Cancer S r C

( h

Sample banking & clinical data collection

h

Other diseases and clinical trials

(area-specific clinical issues)

ds Proteomic Activities

Specific focussed projects e.g. VHL,

Underlying pathogenesis

projects e.g. VHL, response to TKIs…

(“hypothesis driven” )

pathogenesis

Clinical proteomic profiling

(“hypothesis driven and h h i i ”)

Marker & Target Dis

hypothesis generating”)

Discovery

Underlying pathogenesis p g

SLIDE 20

Why Proteomics? Why Proteomics?

Functional entities in biological sys Post translational modifications A i bi l i l fl id Access in biological fluids

Why not Proteomics alone?

Absences of amplification steps (s Absences of amplification steps (s Highly multi-professional teams ne stems

?

sensitivity) sensitivity) eeded and expensive equipment

SLIDE 21

Clinical Proteomics

– essentially profiling e.g comparison differences due to disease, drug trea – key in marker discovery but also pr y y p

“Cell-Mapping” Proteomics pp g

– addressing the question of protein e.g. isolating/characterising protein c g g g p Much of proteomics research is very p y 2D-PAGE ICAT MALDI MS SELDI MS iTRA LC/LC-MS/MS Nove Protein/antibody arrays TECHNOLOGY n of samples to find tment etc rovides biological information g function from interactions complexes p technology-dependent: gy p T AQ el prefractionation

SLIDE 22

Novel Biomark

Protein extract e.g. conditioned 3 pI 10 media

kers: 2D PAGE

mW 3 pI 10

SLIDE 23

Protein Identification Protein Identification

Excise protein – digest – MALDI-TOF pe

r MS/MS sequencing – database query
r MS/MS sequencing

database query

T23 T31 T27 T21 T11 T15

– Mass Spectrometry – Mass Spectrometry

ptide mass fingerprint (PMF) – protein ID. protein ID.

7 T10 T37 T36 T32

SLIDE 24

Surface Enhanced Lase Su ace a ced ase (SELDI™) Mas

ProteinChips™ 8/16 spot ar

surfaces – selective protein

Analysed by time-of-flight m
Optimal <20 kDa

er Desorption/Ionisation e eso pt o / o sat o ss Spectrometry

rrays - different chromatological binding mass spectrometry

SLIDE 25

SELDI ProteinChip™ Arra p

By Surface ‘Chemistries’ By Surface ‘Chemistries’

Hydrophobic Sites Hydrophobic Sites Ioni Ioni

By Surface ‘Biologies’ By Surface ‘Biologies’

Antibody Chips Antibody Chips DNA Chips DNA Chips Enzym Enzym

ays y

SO SO NR NR

3

ic Sites ic Sites IMAC Sites IMAC Sites

SO SO 4 NR NR

Mixed Sites Mixed Sites

Cu Cu2+

2+

Cu Cu3+

3+

scFv Chips scFv Chips

” ”

me Chips me Chips Receptor Chips Receptor Chips

SLIDE 26

Peak Identification

Serum-CM10

PBS-II 4000 QSTAR PCI

3,000 – 5,000 m/z

Profiling may be useful but ultimately identification of di i i k i k discriminant peaks is key

QSTAR-PCI

3156.61 Fragment of PK-120 (inter-alpha trypsin inhibitor H4)

Ideal for examining the “ tid ” “i t t ” “peptidome” or “interactome”. MW<6kDa for sequence. MW<40kDa for MS profile

SLIDE 27

Multidisciplinary

Biostatistics f Bioinformatics M Translation Clinical Scientific s Clinical Integration Biochemists Data Manager Sample technicians

SLIDE 28

Leed

Renal Cancer S r C

( h

Sample banking & clinical data collection

h

Other diseases and clinical trials

(area-specific clinical issues)

ds Proteomic Activities

Specific focussed projects e.g. VHL,

Underlying pathogenesis

projects e.g. VHL, response to TKIs…

(“hypothesis driven” )

pathogenesis

Clinical proteomic profiling

(“hypothesis driven and h h i i ”)

Marker & Target Dis

hypothesis generating”)

Discovery

Underlying pathogenesis p g

SLIDE 29

Von Hippel Linda

Antibody validation of QPRT

u (VHL)

VHL

+VHL Quinolinate phosphoribosyl transferase

SLIDE 30

L-Tryptophan

N H

COOH NH2

L-Tryptophan

N H

COOH NH2

Q

TDO IDO F lk i TDO IDO F lk i

Lo

Formylkynurenine Formylkynurenine

Lo

tum

Q

Quinolinic acid QPRT

COOH COOH

Quinolinic acid QPRT

COOH COOH

Q

As

QPRT QPRT

As T

NAD NAD

Ta

PRT IN Renal Cancer

ss of QPRT is common in
ss of QPRT is common in

mours Quinolinic acid accumulates Q stimulates N-methyl-D- spartate receptors spartate receptors t f th argets for cancer therapy

SLIDE 31

Leed

Renal Cancer S r C

( h

Sample banking & clinical data collection

h

Other diseases and clinical trials

(area-specific clinical issues)

ds Proteomic Activities

Specific focussed projects e.g. VHL,

Underlying pathogenesis

projects e.g. VHL, response to TKIs…

(“hypothesis driven” )

pathogenesis

Clinical proteomic profiling

(“hypothesis driven and h h i i ”)

Marker & Target Dis

hypothesis generating”)

Discovery

Underlying pathogenesis p g

SLIDE 32

2500 5000 7500 10000

RCC Urine – diagnosis SELDI. WCX2 chip

Profiles, peak identification, bio

normal normal2 25 50 75 4772.0+H 5092.6+H 6144.4+H 6544.3+H 8510.2+H 25 50 75 4771.3+H 5091.2+H 6143.1+H 6540.9+H 8506.6+H

N N

rcc 20 40 60 2797.2+H 3383.0+H 4151.8+H 4770.4+H 6143.6+H 6541.2+H 8506.6+H 50 75 3357 2+H 4147.5+H 4766.7+H 6139.9+H 6535.1+H 8502.8+H

RCC RCC

2500 5000 7500 10000 rcc(2) 25 50 2796.1+H 3357.2+H

RCC Serum–prognosis SELDI.CM10 chip

2 50 0 5 00 0 75 0 0 1 0 00 0 1 25 00 3799 urea

25 50 75 100 75 100

p g p

3799 pH7 1/10 3867 urea

25 50 75 25 50 75 100

2 50 0 5 00 0 75 0 0 1 0 00 0 1 25 00 3867 pH7 1/10

25 50 75 100

S iti it S ifi it

Initial Results (neural network)

informatics and statistics

Sensitivity Specificity Training RCC (48) 100% 100% vs Controls (38) Initial Blind RCC (12) 83.3% 85.0% ( ) vs Controls/benign (20)

Several discriminatory peaks

SLIDE 33

“Fraction 8” “Fraction 53”

Identification of Identification of Prognostic Peaks

“Fraction 62”

SLIDE 34

Arc Arc But chival FFPE tissue a major resource chival FFPE tissue – a major resource t – limited suitability

Still problems to overcome Excellent correlation in results

dependence on specific protein
optimisation of normalisation

SLIDE 35

NIHR Applied pp

Biomarkers have major poten the NHS, particularly in co , p y and/or “stratified” medicine a may supplement

r

replac imaging tests for: imaging tests for:

accurate and early diag
measurement of the ac
indication of prognosis

selection and predictio

selection and predictio
monitoring for treatmen

disease progression p g

d Programme g

ntial benefits for patients and

ntributing to "personalised”

g p and improved safety. They ce invasive procedures

r

gnosis ctivity and extent of disease s

n of optimal treatments
n of optimal treatments

nt response/toxicity or

SLIDE 36

NIHR Applied

Potential to improve patient care and h li d b th th li ki realised because the pathway linking research is still quite poorly defined.

Methodological work to define current be
Clinical biochemistry to rapidly identify

y p y y characteristics-a biomarker biobank for re

A RCT of 3 biomarkers for liver fibrosis a

0 8 0.9 1 0.3 0.4 0.5 0.6 0.7 0.8 nsitivity (true positives) 0.1 0.2 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1 - Specificity (false positives) Sen

d Programme

ealth service provision is not yet being bi k h t h lth i biomarker research to health services st practice and explore innovations y protein biomarkers with good clinical y p g nal and liver disease. nd cirrhosis-the ELF test.

Compound serum ELISAs HA, TIMP-1, PIIINP

SLIDE 37

ELF – liver relate

Survival Odds

1 0

n=500 Odds

1.0 0.8 0.6 0.4 0.2

4.00 2.00 0.00

Time years

0.0

ed events at 7 yrs

1 Hazard Ratio

1 7

1 .0 ELF score 4.16—8.33 2.0 ELF score 8.34-10.425 3.0 ELF score 10.426-12.51 4.0 ELF score 12.52-16.67

25 >100

8.00 6.00

Parkes et al Submitted

SLIDE 38

SLIDE 39

The Clinical Proteomics Exp

Pre-fractionation and protein profiling

4000 6000 8000 10000

Bioinform & Biostati

Clinical questions l & samples Clinical and NHS Benefits Trials and HSR

Multiplex assa development & cli

periment

HYPOTHESIS, TECHNOLOGY, and CLINICAL DATA

RCC i l

Association of “peaks” with clinical “outcome”

0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.2 0.4 0.6 0.8 1.0

RCC - survival

Peak 1 Peak 2 Peak 3 Peak 4

atics istics

1000 2000 1000 2000 0 1000 2000 0 1000 2000

Protein purification I iti l

Identification of “peaks”

Initial validation

Identification of peaks by mass spectrometry

ay/drug inical trials

SLIDE 40

Cancer Research UK Ce

Vassilis Aggelis gg Andrew Bernard Ann Bogue Glenn Bonney Janet Brown David Cairns Fiona Collinson Rachel Craven Rosie Ferguson g Geoff Hall Pat Harnden Sharon Jackson Satinder Jagdev g Alan Liu David Perkins Jianhe Peng Rumana Rafiq

Roz Ba

q Liz Sheldon Sheryl Sim Annie Stanley Douglas Thompson g p Nav Vasudev Alison Young Medical Oncology and Urology Consultants

entre

NIHR Programme g

Professor Doug Altman Professor Roz Banks Dr Ian Barnes Professor Jon Deeks Dr Walter Gregory Professor Jenny Hewison Professor Philip Johnson p Professor Chris McCabe Professor William Rosenberg Professor Peter Selby Dr Catharine Sturgeon

anks

g Dr Doug Thompson