Predict ive biomarker f or molecular t arget drugs- prot eomic and - - PowerPoint PPT Presentation

Dept Genome Biology

Predict ive biomarker f or molecular t arget drugs- prot eomic and glycobiological approach

Depar t ment of Genome Biology Kinki University School of Medicine

Kazut o Nishio, MD PhD

ETH Zurich – J ST Workshop on Medical Research

15t h – 16th of Sept ember 2008, ETH Zurich Session 3

Dept Genome Biology

Biomarkers Def init ion

Biomarkers Definitions Working Group (2001) Clin. Pharmacol. Ther. 69, 89–95

Biomarkers are

‘ a characteristic that is objectively measured and evaluated as an indicator of normal biologic

processes, pathogenic processes,

• r

pharmacologic responses to a therapeutic intervention’.

The need f or bet t er predict ive markers The need f or bet t er predict ive markers

Slide: Paul Warning, Genent ech (modif ied)

The average response rate to dr ug t reat ment is not accept able.

Dept Genome Biology

分子標的薬 Target : Molecules Molecular t arget t herapy

Dept Genome Biology

Oncology New Drug Approvals ( FDA ) 2001-2006 Chabner AACR 2006

• Molecular Targeted Drug
• Non- Targeted Drug

( 27 % Patient Selection )

33%

67 % ( 2015 ＞75% )

Dept Genome Biology

D r u g s T a r g e t A p p l i c a t i

• n

* *

F D A J a p a n R i t u x a n C D 2 B C L L 1 9 9 7 2 1 H e r c e p t i nH e r 2 * b r e a s t c a 1 9 9 8 2 1 G l e e v e c B c r

• A

b l / K i t * C M L , G I S T 2 1 2 1 I r e s s a E G F R * N S C L C 2 3 2 2 V e l c a d e P r

• t

e a s

• m

e M M 2 3 年 2 7 2 7 A v a s t i n V E G F C R C 2 4 2 8 E r b i t u x E G F R * C R C 2 4 a p p l i e d T a r c e v a E G F R E G F R * N S C L C , p a n c c a 2 5 2 7 N e x a v a r M u l t i

• k

i n a s e s *R C C 2 5 2 7 S u t e n t M u l t i

• k

i n a s e s *G I S T , R C C 2 6 a p p l i e d S p r y c e l B c r

• A

b l / S r c

* C

M L

* *

, P h + A L L 2 6 a p p l i e d

*

k i n a s e i n h i b i t

• r

k i n a s e i n h i b i t

• r

*

*

G l e e v e c

• r

e s i s t a n c e

A p p r

• v

e d M

• l

T a r g e t D r u g s （ s m a l l m

• l

e c u l e s a n d a n t i b

• d

i e s ）

Compounds Target Tumors Diagnosis Herceptin

Humanized ant i HER2 Ab Overexpression of HER2 I H (Hercep t est ) FI SH

Rituxisan

Chimeric ant i CD30 m- Ab CD20 (+) B- cell non Hodking lymphoma I H FCM

Gleevec

bcr- abl / c- kit - TKI

• 1. CML
• 2. GI ST wit h KI T

（ C D117） +g Chromosomal t est Gene analysis I H

I rinotecan

Topo I inhibit or NSCLC or ovarian ca.

• e. g.

I nvader assay (UGT1A1gene polymorphism)

Biomarkers f or patient selection

New, more inf ormat ive t rial designs

Approach: Pair diagnostic with therapeutic

• I dent if y r esponder s and non-r esponder s
• Pr event t oxicit y
• Monit or r esponse

Answer ser ies of quest ions, e.g.,

• Which dose is cor r ect f or which sub-populat ion?
• Which sub-populat ion should be t r eat ed?

Dept Genome Biology

Biomarker study f or target based drugs

Operative Biopsy Cytology PBMC (Blood) Serum (Blood) Plasma (Blood) Pleural ef f usion

• Feasibility
• Power f or prediction
• Sensitivity
• Accuracy : f alse positive e. g.

We f ocused on

Search f or biomarkers in surrogat e t issue (circulat ing samples) Search f or biomarkers in surrogat e t issue (circulat ing samples)

• 1. EGFR somat ic mut at ion

in circulat ing t umor cells in lung cancer

• 2. Gene expression prof ile in PBMC
• 3. Serum prot eomics
• 4. Mult iplex ELI SA using bio beads
• 5. Glycoprof iling

Dept Genome Biology

EGFR tyrosine kinase inhibitors (EGFR EGFR tyrosine kinase inhibitors (EGFR-

• TKI s)

TKI s)

Kinase Kinase

lig a nd lig a nd

Kinase Kinase

kina se kina se kina se kina se

R R R R P P

Cell cycle Cell cycle progression progression Prolif eration Prolif eration Decreased Decreased apoptosis apoptosis Angiogenesis Angiogenesis Metast asis Metast asis EGFR expression in NSCLC

Squamous cell ca. Squamous cell ca. Adenocarcinoma Adenocarcinoma CM CM Small molecule tyrosine kinase inhibitors Gef itinib (I ressa) Erlotinib (Tarceva)

EGFR

EGFR mutation

EXON 18 EXON 24 EXON 23 EXON 22 EXON 19 EXON 21 EXON 20

G719X 4% Del746-750 2235 20% Del746-750 2236 11% Ot her Exon 19 del 13%. 2237_2251del15; 2237_2254del18; 2237_2255> T (complex); 2236_2250del15; 2238_2255del18; 2238_2248> GC (complex); 2238_2252> GCA (compl ex); 2239_2247del9; 2239_2253del15; 2239_2256del18; 2239_2248TTAAGAGAAG> C (complex) 2239_2258> CA (complex); 2240_2251del12; 2240_2257del18; 2240_2254del15; 2239_2251> C (complex); 2235_2252>AAT (complex) G768I 1% T790M 1% Exon 20 I ns 5% L858R Mutati ons 39% L861Q Mutati ons 1%

Dept Genome Biology

Sensitivity of transf ected cells with deletional mutation to EGFR- TKI

ZD1839 (µM) Gr owth r ati o 100 10 1 .1 .01 0.2 0.4 0.6 0.8 1.0 1.2 AG1478 (µM) Gr owth r ati o 100 10 1 .1 .01 0.2 0.4 0.6 0.8 1.0 1.2

gef itinib AG1478

M W D M W D M W D 0.01 0.1

M:293_pcDNA3. 1/ Zeo W:293_pcDNA3. 1/ EGFR wt D:293_pcDNA3. 1/ EGFR15d

ZD1839 uM f or 3h Det ect ion of pEGFR (1068)

Ar ao et al. Cancer Res ‘05

Paez et al., Science 2004 Lynch et al.,NEJM 2004 Pao et al.,PNAS 2004 Huang et al., CCR 2004 Tokumo et al., CCR 2005 Mitsudomi et al., JCO 2005 Han et al., JCO 2005 Kim et al., CCR 2005 Cotes-Funes et al., Ann Oncol 2005 Cappuzzo et al, JNCI 2005 Chou et al., CCR 2005 Taron et al., CCR 2005 Takano et al., JCO 2005

I mpact of EGFR mutation on the response to EGFR- TKI (N=616)

Tumor response (%)

Wild-type

10 20 30 40 50 60 70 80 90 100

77% 12%

EGFR Mutation

EGFR Wild type

Dept Genome Biology

Analysis of circulat ing DNA

Dept Genome Biology

200 400 600 800 20 40 60 80 100 Mut at ion + Mut at ion - P* = 0. 078 Percent age Overall Survival (Days) 200 400 600 800 20 40 60 80 100 Mut at ion + Mut at ion - P* = 0. 005 Percent age Progression Free Survival (Days)

+ + + + + + + + + +

PFS OS

Detection of EGFR mutation by Scorpion- ARMS in serum samples

Search f or biomarkers in surrogat e t issue (circulat ing samples) Search f or biomarkers in surrogat e t issue (circulat ing samples)

• 1. EGFR somat ic mut at ion

in circulat ing t umor cells in lung cancer

• 2. Gene expression prof ile in PBMC
• 3. Serum prot eomics
• 4. Mult iplex ELI SA using bio beads
• 5. Glycoprof iling

prot eomics using LC-MS/ MS prot eomics using LC-MS/ MS

RT: 0.00 - 90.02 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 Time (min ) 50 100 50 100 50 100 50 100 Relative Abundance 50 100 50 100 68.33 1113.38 67.18 1113.23 39.38 738.23 69.03 1113.46 57.35 836.12 18.21 842.56 49.44 762.70 19.53 842.54 41.95 971.30 34.24 877.76 13.53 523.64 77.72 1326.17 25.63 453.44 4.56 515.41 81.78 845.85 85.53 738.15 68.54 1113.34 39.50 738.29 64.35 937.47 19.35 842.59 57.33 836.14 69.27 1113.58 49.51 762.67 34.54 877.72 17.79 715.12 42.02 971.27 22.08 743.49 31.41 745.14 77.65 1326.13 81.70 531.33 4.89 515.53 12.19 561.16 85.49 743.35 39.76 738.26 19.39 842.53 34.37 877.72 69.21 1113.36 40.47 742.94 65.24 937.40 17.58 715.16 57.56 836.19 49.67 837.37 21.74 743.50 31.17 745.18 69.66 1113.52 44.48 746.73 78.06 971.23 12.22 502.63 8.29 512.57 81.68 531.16 85.45 899.19 39.54 738.28 69.21 1113.37 39.72 738.33 18.96 842.58 57.48 836.09 49.60 762.68 64.81 937.42 34.15 877.73 17.20 715.14 42.03 971.25 21.58 743.53 69.98 1113.52 26.42 690.69 78.03 971.37 4.74 515.49 81.66 531.46 10.85 561.11 84.91 743.44 68.78 1113.34 39.63 738.24 64.81 937.44 57.43 836.14 17.81 715.14 49.53 762.71 19.88 842.52 34.17 877.74 56.42 836.13 71.78 1064.80 47.80 762.42 31.01 745.15 77.88 971.17 13.97 523.58 4.77 515.44 81.57 845.88 86.16 836.29 39.44 738.27 68.55 1113.34 40.07 742.93 34.02 877.70 49.26 837.38 17.25 715.12 56.93 836.25 64.50 937.36 69.06 1113.43 19.74 842.51 30.67 745.16 44.05 746.67 59.91 1022.21 77.67 971.24 13.73 523.55 4.45 515.43 79.77 1163.82 86.21 743.62

NL: 4.63E7 Base Peak F: MS Control_01 NL: 3.87E7 Base Peak F: MS control_02 NL: 1.95E7 Base Peak F: MS iressa_01 NL: 2.67E7 Base Peak F: MS iressa_02 NL: 2.46E7 Base Peak F: MS tarceva_01 NL: 1.88E7 Base Peak F: MS tarceva_02

Base- peak Chromatograms of Chemical Pulldown Samples

Control_01 Control_02 Iressa_01 Tarceva_02 Iressa_02 Tarceva_01

I dentif ication of t arget proteins bind to EGFR- TKI

gerfitib (Iressa) Control erlotinib (Tarceva)

1st line gef itinib monotherapy f or advanced NSCLC LC- MS/ MS analysis of serum samples

Proteomic approach to detect biomarkers to predict gef itinib- response

• 0.7 24

1 1 2 7 2 2 2

PR vs. SD (Pre)

PR SD

PCA

OS ~ T (818. 1 / 38. 1) pre

T T R 7 6 5 1 2 1 6 2

R2=0. 73

A 1 B G _ 7 6 8 2 3 1 5 8

OS ~ G (1076. 5 / 74. 8) pre

■Mult icohort Cross-I nst it ut ional St udy ■St udy Design To dist inguish prognost ic and predict ive biomarkers

Study design

• 1. Biomarkers not available
• 2. Biomarkers available
• 3. Enrichment

Biomarkers (+) のみが有効である こ と が示さ れている 場合 New Dr ugs P lacebo R New Dr ugs P lacebo R P atients B (＋) B (－) New Dr ugs P lacebo R P atients B (＋) B (－) P atients B (＋) (－) R B (＋) (－) New Dr ugs P lacebo

Pneumonitis induced by gef itinib

I LD: interstitial lung disease

Pretreatment Af ter 35 Days

Dept Genome Biology 10 20 30 40 50 60 70 80 90 100 10 20 30 40 50 60 70 80 90 100 Relative Abundance(%)

47.28 807.50 27.13 570.10 59.22 694.43 20.14 529.75 32.54 838.72 79.68 990.24 68.20 757.10 66.98 724.96 42.21 808.45 13.14 608.70 16.33 530.09 56.86 853.55 79.14 894.87 70.28 757.42 62.75 984.53 11.35 580.05 5.75 588.57

Cont rol- Pre Case- Pre

32.18 838.65 26.70 596.88 36.08 612.85 47.01 807.62 28.05 859.51 25.43 491.18 18.07 620.45 38.03 725.30 65.00 822.02 14.66 462.68 44.09 639.49 59.07 694.49 24.66 732.13 71.11 1292.89 79.48 937.11 51.69 834.07 9.89 472.00 5.98 638.12

Relative Abundance(%)

Comparison of chromat ogr am

• f t ypical I LD / non-I LD baseline samples

non- I LD I LD

Dept Genome Biology

A Comprehensive I on Signal Map f r om each dat a point The Map consist s of quant it at ive signals and physicochemical pr oper t ies (like m/ z) der ived f r om t he LC

• MS

measur ement

Cases Controls

Target MS/ MS

t o ident if y pr ot eins f r om t he signals

Mining signif icant signals by compar ing t he

t wo t emplat es Signif icant LC- MS signals An ‘aver aged’ map as a t emplat e of

posit ive gr oup

An ‘aver aged’ map as a t emplat e of

negat ive gr oup List of t he pr ot eins of which amount s ar e signif icantly dif f er ent bet ween t he two groups

How to identif y group- specif ic peptide signals i- OPAL approach

Dept Genome Biology

Applicat ion t o t he gef it inib I LD CCS Prof ile chart of candidat e marker signals

Align LC

• MS signals of all samples

LC-MS measur ement of case / cont r ol samples Apply st at i st ical t est t o each aligned signal point P ickup signal point s of which int ensit y is signif icant ly dif f er ent bet ween cases & cont r ol s

Dept Genome Biology

L t o cont r ol

5 10 15 20 25 30 5 10 15 20 25 30

L t o case

baseline post baseline post Case-P r e(n=11) Case-P

• st (n=11)

Cont r ol-P r e（ n=12) Cont r ol-P

• st (n=13)

I LD specif ic area non- I LD specif ic area

Distribution of 47 patients prof iles using tentative signal set

(Case-P r e 11, Case-P

• st 11, Cont r ol-P

r e 12, Cont r ol-P

• st 13sample)

Search f or biomarkers in surrogat e t issue (circulat ing samples) Search f or biomarkers in surrogat e t issue (circulat ing samples)

• 1. EGFR somat ic mut at ion

in circulat ing t umor cells in lung cancer

• 2. Gene expression prof ile in PBMC
• 3. Serum prot eomics
• 4. Mult iplex ELI SA using bio beads
• 5. Glycoprof iling

Dept Genome Biology

Multiplexing with Colored Bead Sets Bio-PlexTM ELI SA Assay Syst em

Dept Genome Biology

The levels of cyt okines in plasma f rom NSCLC pat ient s received gef it inib det ect ed by BioPlex (Luminex) The levels of cyt okines in plasma f rom NSCLC pat ient s received gef it inib det ect ed by BioPlex (Luminex)

Dept Genome Biology

The cytokine levels at pre- treatment in NSCLC patients

I L-1b I L-2 I L-4 I L-5 I L-6 I L-7 I L-8 I L-10 I L-12 I L-13 I L-17 G-CSF GM-CSF TNF-a I FN-γ MCP

• 1 MI P
• 1b

The minimum det ect able dose f or each cyt okine was 0.01mg/ ml.

Dept Genome Biology

MI P- 1b levels between patients with and without gef itinib- induced dermatitis

50 100 150 200 250 300

With dermatitis Without dermatitis

P = 0. 042 Plasma MI P- 1β (pg/ mL) (n=8) (n=16)

◆MI P-1b is a candidate f or predictive marker to predict skin toxicity.

Kimura et al. Lung Ca, 2005

Dept Genome Biology

Bio- PlexTM ELI SA Assay System

f or angiogenesis inhibitors

１ ． Angio-1 panel （ angiopoietin-2, follistatin, G-CSF, HGF, IL-8, leptin, PDGF- BB, PECAM-1, VEGF） ２ ． Angio-2 panel （ IL-6R、 MMP-9、 TIMP-1、 TIMP-2、 Endostatin、 P-selectin、 ICAM-1、 VCAM-1、 Tie-2、 PAI-1、 MIF、 uPAR）

Search f or biomarkers in surrogat e t issue (circulat ing samples) Search f or biomarkers in surrogat e t issue (circulat ing samples)

• 1. EGFR somat ic mut at ion

in circulat ing t umor cells in lung cancer

• 2. Gene expression prof ile in PBMC
• 3. Serum prot eomics
• 4. Mult iplex ELI SA using bio beads
• 5. Glycoprof iling

Dept Genome Biology

MS HPLC labeling Column chromatography Samples including released glycans.

(Ex. Glycoproteins Serum, plasma Cells, Tissue treated by PNGase)

several days The bottleneck of current glycomics research is time- consuming and non- precise purif ication using column chromatography.

Current situation of glycomics

Dept Genome Biology

Concept of BlotGlyco

MS HPLC

BlotGlyco beads

Samples I ncluding released glycans

(Ex. Glycoproteins Serum, plasma Cells, Tissue treated by PNGase)

Speedy, one- pot solid phase process to obtain perf ectly purif ied and labeled glycan. Precise purif ication and various labeling, 5~7 hours in total

Dept Genome Biology

H z H z H z Hz Hz Hz

Peptide Lipid Glycan Protein DNA

O HO HO HO OH OH OH HO HO HO OH O OH HO HO HO OH N N H

BlotGlyco

Applying BlotGlyco into crude mixture *Hz: hydrazide group (-NHNH2) Labeled glycan

Hz Hz Hz Hz H z Hz

Purif ication Aldehyde of reducing end of glycan is bound to hydrazide of BlotGlyco beads. (covalent bonding) Various labeling f or MS, HPLC

Reaction mechanism of BlotGlyco

Glycan capt ured by BlotGlyo beads endure harsh wash. Every impurities even peptides can be completely removed.

Dept Genome Biology

. E + 1 . E + 6 2 . E + 6 3 . E + 6 4 . E + 6 . 2 . 4 . 6 . 8 . 1 . 1 2 C

• n

c . [ m M ] p e a k a r e a . E + 4 . E + 8 8 . E + 8 1 . 2 E + 9 1 . 6 E + 9 2 . E + 9 1 2 3 4 5 6 C

• n

c . [ m M ] p e a k a r e a

Cor r elation between concentr ation of glycan and peek ar ea

Low concent r at ion ar ea: under 10 mM

～5 mM

* Detectable f rom 0. 1 mM solution (2 pmol Maltoheptaose) * Standard curve shows linearity in the range of 0. 1 mM to 5 mM. *Malt ohept aose solut ion was used.

Quantitative reliability of BlotGlyco

Dept Genome Biology 1250 1500 1750 2000 2250 2500 2750 3000 3250 3500 3750 m/z

Human serum (5 μL)

49 kinds of N-glycans were det ected f rom 5μL human serum.

Mannose Galactose N-Acetylglucosamine N-Acetylgalactosamine Neuramic acid (sialic acid) Fucose

N- glycan prof ile obtained f rom 5μL human serum

Minor components detected:

Estimated structures from mass data (not y et conf irmed).

Dept Genome Biology

Pre CNB (HER2 check)

t rast uzumab

I C CNB:core needle biopsy

Serum samples of breast cancer patients received with t rast uzumab monotherapy Trast uzumab (Herceptin): anti-HER2 Ab I dentif ication of predictive biomarkers f or response to trastuzumab using glycobiological analysis

Dept Genome Biology

PD nonPD

Represent at ive dat a of plasma n-glycan prof ile measured using MALDI -TOF-MS

Dept Genome Biology

Plasma N- glycan and clinical outcome

(Lef t) Expr ession of plasma N-glycan and clinical r esponse. The expr ession of plasma 2534 m/ z N-glycan was signif icant ly lower in patients with pr ogr essive disease (P D). (Right) Kaplan-Meier cur ve of high (detectable) or low (not detectable) plasma N-glycan gr oups f or pr ogr ession-f r ee sur vival (P FS) af t er tr ast uzumab tr eat ment .

p<0. 05

nonPD

PD

N- glycan (% of cont rol)

10 20 30

p<0.05

Time (days) Progression ｆ ree Survival

. 2 . 4 . 6 . 8 1 100 200 300 400 500 600

High 2534m/ z Low 2534m/ z

(PFS)

Dept Genome Biology