DNA MICROARRAY“Gene Expression Profiling:

from gene function to personalized medicine”

Hans Bluyssen

17-03-2018

Genome & Genes

Gene

RNA

Protein

transcription

translation

3’ 5’

+1

promoter

start site

coding sequence

mRNA

RNA polymerase

generation of mRNA from genomic DNA

RNA Transcription

The Transcriptome

Transcriptional Program & Cellular Fate

Breast Cancer cells Red Blood cells Sperm cells

Astrocytes Smooth Muscle cells Neuronal cells

Parallel monitoring of relative levels of thousands of mRNA

species at one time point or condition:

expression profiling

DNA Microarray

Transcriptomics

• Put a large number (~30K) of cDNA sequences or synthetic DNA oligomers onto a glass slide (or other substrate) in known locations on a grid.

• Label an RNA sample and hybridize

• Measure amounts of RNA bound to each square in the grid

• Make comparisons

– Cancerous vs. normal tissue

– Treated vs. untreated

– Time course

• Many applications in both basic and clinical research

DNA Microarrays: Basics

DNA microarrays are ordered assemblies of

DNA sequences immobilized on a solid support

(such as chemically modified glass).

What is a DNA microarray?

The DNA sequences (e.g. PCR products or oligos)

correspond to the transcribed regions of genes.

5’ start

genomic DNA

exon 1 exon 2 exon 3

…ATTTCAGGCGCATGCTCGG… gene X

gene Y

gene Z

What is a DNA microarray?

Each gene represented by single long oligo (60 - 70-mer)

Oligo’s spotted by robot

Long oligo arrays

Each gene represented by 20 different short (25-mer)

oligonucleotides and 20 mismatch controls

Oligonucleotides synthesized on chip by photolithographic

masking

Affymetrix chips

probe pair Mismatch probe cells

(12-20/gene)

Affymetrix chips

3-micron silica beads

that self assemble in

microwells on fiber

optic bundles or planar

silica slides.

When randomly

assembled the beads

have a uniform spacing

of ~5.7 microns.

ILLUMINA

… and BEADS!

Optical fiber with wells…

ILLUMINA

N=20

ILLUMINA

ILLUMINA

Oligo’s synthesized on chip by ink-jet printing

Agilent arrays

Each gene represented by single long oligo

(60 - 70-mer)

Agilent arrays

1.

Sample A cDNA A Cy3

Sample B cDNA B Cy5one microarray

2.

Measurements are relative i.e. a change is

measured, not an absolute amount

Dual color

Microarray hybridizations

Scan 1 Scan 2

Image 1 Image 2

Combined Image

(ratio cy5:cy3)

Statistical analysis

X Y Z

Hybridize

Wash

cy5 cy3

equal expression

higher expression in Cy3

higher expression in Cy5

Overview of data capture

– two different mRNA populations, labeled with different fluors

– excited by a laser

– each fluor excites at a different wavelength, which is captured using a photo detector attached to a filter tuned to the particular fluor

Microarray: Workflow

Controlcells

Experimentalcells

Cy3

Targets

Cy3

ILLUMINA

Microarray: Workflow

Microarray vs NG-Sequencing

1. Find the genes that change expression between experimental and control samples

2. Classify samples based on a gene expression profile

3. Find patterns: Groups of biologically related genes that change expression together across samples/treatments

4. Correlate expression profile to disease state, diagnosis/prognosis or treatment

Goals of a Microarray Experiment

• Type I: (n = 2)

– How is this gene expressed in target 1 as compared to target 2?

– Which genes show up/down regulation between the two targets?

• Type II: (n > 2)

– How does the expression of gene A vary over time, tissues, or treatments?

– Do any of the expression profiles exhibit similar patterns of expression?

Microarray Experiment Design

Aging & Increased Atherosclerosis

Endothelial Cells in Culture

Young Old

Aim: Identify known and novel genes involved in aging ECs

Gene expression?

Growing Cells

Grow HUVECs in culture

Cells were obtained from umbellical vein.Cultered for different passages (1-8).Two independent exp. (11-06-02, 13-02-03)

Isolate total RNA

Microarray analysis (fluor flip)

Run 1: #1 vs #8 (11-06-02)Run 2: #1 vs #8, #1 vs #4 (13-02-03)

Wavelength (nm)

Sample 260 280 ratio amount of RNA (ng)

1 mm 0.047 0.025 1.9 206.8

0.5 mm 0.025 0.014 1.8 11028S

18S

RNA yield & quality

19200, 70-mer oligos

-16752 genes of H. sapiens,

-2448 control spots (external controls for ratios

and normalization,

background controls).

Contains:

Human 19K Oligo Microarray (Holstege et al., Genomics lab, UMCU, 2004)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20

1 EC1 empty 336 337 338 339 340 341 342 343 EC8 344 345 346 347 348 349 empty empty EC1

2 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335

3 299 300 EC2 301 302 303 304 305 306 307 EC16 308 309 310 311 312 313 EC5 314 315

4 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298

5 262 263 264 265 EC12 266 267 268 269 270 EC10 271 272 273 274 EC14 275 276 277 278

6 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261

7 225 226 227 228 229 230 EC4 231 232 233 EC9 234 235 EC7 236 237 238 239 240 241

8 207 208 209 210 211 212 213 empty 214 215 216 217 buffer 218 219 220 221 222 223 224

9 189 190 191 192 193 194 195 196 empty 197 198 EC17 199 200 201 202 203 204 205 206

10 EC11 176 EC3 177 EC15 178 EC6 179 EC13 empty EC18 180 181 182 183 184 185 186 187 188

11 163 164 165 166 167 168 169 170 171 EC18 empty EC13 172 EC6 173 EC15 174 EC3 175 EC11

12 145 146 147 148 149 150 151 152 EC17 153 154 empty 155 156 157 158 159 160 161 162

13 127 128 129 130 131 132 133 buffer 134 135 136 137 empty 138 139 140 141 142 143 144

14 110 111 112 113 114 115 EC7 116 117 EC9 118 119 120 EC4 121 122 123 124 125 126

15 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109

16 73 74 75 76 EC14 77 78 79 80 EC10 81 82 83 84 85 EC12 86 87 88 89

17 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72

18 36 37 EC5 38 39 40 41 42 43 EC16 44 45 46 47 48 49 50 EC2 51 52

19 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35

20 EC1 empty 1 2 3 4 5 6 7 EC8 8 9 10 11 12 13 14 15 empty EC1

1 2 3 4

9

10

11

12

5

6

7

8

1

2

3

4

Barcode

45 46 47 48

41 42 43 44

37 38 39 40

33 34 35 36

29 30 31 32

25 26 27 28

21 22 23 24

17 18 19 20

13 14 15 16

9 10 11 12

5 6 7 8

1 2 3 4

Cy3 Scan

Overlay Cy3 and Cy5 scans

equal expression

higher expression in Cy3

higher expression in Cy5

• Quantitate each spot

• Export a table of fluorescent intensities

for each gene in the array

• Subtract background

• Normalize

• Analyze data further

Data Acquisition

#1 #8

cDNA cDNA

Label mix

IMAGENE

Chip

Imaging analysis

Gridding

Data analysis: Image analysis

Measured differences may be due to variation in

processing, different labels used, different

channels measured etc.

Sample A cDNA A Cy3

Sample B cDNA B Cy5

Requires normalisation between channels

one microarray

Dual color experiments

• Can control for many of the experimental sources of variability

• Bring each image to the same average brightness

• Can use simple math or fancy -

– divide by the mean (whole chip or by sectors)

– House keeping genes (assumed constants?)

– External controls (constants)

– LOESS (locally weighted regression)

– Dye-swap (fluor-flip)

Normalization

#1 vs #8 UPREGULATED

24-01-03: FALSE 4779 (5005); upregulated 223

11-06-02: FALSE 1222; upregulated 373

Mean > 0.7: 125

#1 vs #8 DOWNREGULATED

24-01-03: downregulated 202

11-06-02: downregulated 351

Mean < -0.7: 118

Regulated Genes

Gene_Symbol/name GB_accession/nameDescription Fflip #1 vs #8 (240103)Fflip #1 vs #8 (110602)>0.5 Mean

KIAA177 AB029000 KIAA1077 protein 2,513731797 2,162643424 0 2,338188

MGP X53331 Matrix Gla protein 1,924562345 1,06359492 0 1,494079

CLU M25915 Clusterin (complement lysis inhibitor, SP-40,40, sulfated glycoprotein 2, testosterone-repressed pro1,120027922 1,422007634 0 1,271018

MLLT4 AL161973 Myeloid/lymphoid or mixed-lineage leukemia (trithorax (Drosophila) homolog); translocated to, 40,699761417 1,788515835 0 1,244139

FABP4 J02874 Fatty acid binding protein 4, adipocyte 1,033984581 1,224226734 0 1,129106

LMO2 X61118 LIM domain only 2 (rhombotin-like 1) 1,608864882 0,632131603 0 1,120498

CD34 S53911 CD34 antigen 0,625034943 1,491856907 0 1,058446

PLA2G6 AF064594 Phospholipase A2, group VI (cytosolic, calcium-independent)1,120023278 0,923718755 0 1,021871

FLJ22344 NM_024717Hypothetical protein FLJ22344 0,676278219 1,284356776 0 0,980317

IFITM1 J04164 Interferon induced transmembrane protein 1 (9-27)1,321185077 0,612176398 0 0,966681

FADS2 AF126799 Fatty acid desaturase 2 0,703455048 1,19919164 0 0,951323

GNG11 U31384 Guanine nucleotide binding protein 11 0,912465229 0,980640855 0 0,946553

ALDH1A1 M31994 Aldehyde dehydrogenase 1 family, member A11,18344417 0,657748227 0 0,920596

FLJ1111 AK001972 Hypothetical protein FLJ11110 0,995447178 0,799882598 0 0,897665

KIAA692 AB014592 KIAA0692 protein 0,78229028 0,99391436 0 0,888102

COL6A1 X15880 Collagen, type VI, alpha 1 1,159896438 0,599391575 0 0,879644

VWF X04385 Von Willebrand factor 0,938009512 0,777968076 0 0,857989

TROAP U04810 Trophinin associated protein (tastin) 0,737655495 0,939420669 0 0,838538

DKFZP434C211 AL110244 DKFZP434C211 protein 0,593289953 1,058910073 0 0,8261

LIPA X76488 Lipase A, lysosomal acid, cholesterol esterase (Wolman disease)1,07536012 0,566963312 0 0,821162

ARMET M83751 Arginine-rich, mutated in early stage tumors0,717870343 0,875637643 0 0,796754

GCL M81637 Grancalcin 0,811293852 0,777934191 0 0,794614

VCAM1 M73255 Vascular cell adhesion molecule 1 0,654017353 0,926043108 0 0,79003

IRF6 AL022398 Interferon regulatory factor 6 0,969052784 0,604807534 0 0,78693

HEPH AB014598 Hephaestin 0,978667308 0,594992788 0 0,78683

AL022163 Human DNA sequence from clone 551E13 on chromosome Xp11.2-11.3 Contains farnesyl pyrophosphate synth0,707438024 0,841306538 0 0,774372

TAS2R1 AF227129 Taste receptor, type 2, member 1 0,6107639 0,919104935 0 0,764934

PRCP L13977 Prolylcarboxypeptidase (angiotensinase C)0,567628921 0,946099969 0 0,756864

DLK1 U15979 Delta-like homolog (Drosophila) 0,976160319 0,529537279 0 0,752849

DACH AJ005670 Dachshund (Drosophila) homolog 0,831559463 0,667854409 0 0,749707

HMGCR NM_0008593-hydroxy-3-methylglutaryl-Coenzyme A reductase0,900287358 0,565012227 0 0,73265

AL137512 Homo sapiens mRNA; cDNA DKFZp564E0178 (from clone DKFZp564E0178); partial cds0,713181012 0,679221354 0 0,696201

AL137578 Homo sapiens mRNA; cDNA DKFZp564N2464 (from clone DKFZp564N2464)0,643705912 0,722626994 0 0,683166

DHCR7 AF034544 7-dehydrocholesterol reductase 0,77803299 0,552501793 0 0,665267

OR1G1 U53583 Olfactory receptor, family 1, subfamily G, member 10,794092501 0,51583863 0 0,654966

HSPB1 Z23090 Heat shock 27kD protein 1 0,68780744 0,616494742 0 0,652151

HHEX L16499 Hematopoietically expressed homeobox 0,735995153 0,567180638 0 0,651588

STAT1 M97935 Signal transducer and activator of transcription 1, 91kD0,765382029 0,529963687 0 0,647673

Log2 ratio’s

Up-regulated #1 vs #8, >0.7

Gene_Symbol/name GB_accession/nameDescription Fflip #1 vs #8 (240103)Fflip #1 vs #8 (110602)<-0.5 Mean

TGFBI M77349 Transforming growth factor, beta-induced, 68kD-2.33833722 -1.27362528 2 -1.8059812

SGK AJ000512 Serum/glucocorticoid regulated kinase -0.82421584 -2.58943878 2 -1.7068273

PLAU X02419 Plasminogen activator, urokinase -1.6055826 -1.11066064 2 -1.3581216

KEL M64934 Kell blood group -1.31760599 -1.34347572 2 -1.3305409

MMP1 M13509 Matrix metalloproteinase 1 (interstitial collagenase)-0.95061809 -1.58462396 2 -1.267621

AF098641 Homo sapiens CD44 isoform RC (CD44) mRNA, complete cds-1.26297713 -1.20426938 2 -1.2336233

AL136803 Homo sapiens mRNA; cDNA DKFZp434M1820 (from clone DKFZp434M1820); complete cds-0.50233598 -1.92845647 2 -1.2153962

CDKN1A U03106 Cyclin-dependent kinase inhibitor 1A (p21, Cip1)-1.25234383 -0.99838852 2 -1.1253662

XM_007600Homo sapiens alanyl (membrane) aminopeptidase (aminopeptidase N, aminopeptidase M, microsomal aminop-1.30118944 -0.91301997 2 -1.1071047

SPUVE AF015287 Protease, serine, 23 -1.19481818 -0.83773478 2 -1.0162765

AB033089 Homo sapiens mRNA for KIAA1263 protein, partial cds-1.3513174 -0.66492176 2 -1.0081196

PTPRO NM_002848Protein tyrosine phosphatase, receptor type, O-1.12796975 -0.76306015 2 -0.9455149

CGA S70585 Glycoprotein hormones, alpha polypeptide-0.76538247 -1.11468868 2 -0.9400356

CDKN1A Z85996 Cyclin-dependent kinase inhibitor 1A (p21, Cip1)-0.96468421 -0.88752639 2 -0.9261053

ARHB NM_004040Ras homolog gene family, member B -1.24914781 -0.56511585 2 -0.9071318

CAPN2 M23254 Calpain 2, (m/II) large subunit -0.67063344 -1.12861276 2 -0.8996231

SPOCK AF231124 Sparc/osteonectin, cwcv and kazal-like domains proteoglycan (testican)-0.76625413 -1.03297294 2 -0.8996135

ITGB5 X53002 Integrin, beta 5 -0.85883624 -0.93320134 2 -0.8960188

TEM8 AK025429 Tumor endothelial marker 8 -1.1064934 -0.56344376 2 -0.8349686

TADG12 NM_022364Serine protease TADG12 -1.05323657 -0.58052308 2 -0.8168798

FBN2 U03272 Fibrillin 2 (congenital contractural arachnodactyly)-0.99638042 -0.61295108 2 -0.8046657

SMURF2 NM_022739E3 ubiquitin ligase SMURF2 -0.77041438 -0.8217453 2 -0.7960798

AC004528 Homo sapiens chromosome 19, cosmid R32184-0.93405079 -0.64952944 2 -0.7917901

GRB14 L76687 Growth factor receptor-bound protein 14-0.96972574 -0.59117799 2 -0.7804519

FACL1 L09229 Fatty-acid-Coenzyme A ligase, long-chain 1-0.97376797 -0.55574601 2 -0.764757

HMOX1 Z82244 Heme oxygenase (decycling) 1 -0.96184824 -0.5646887 2 -0.7632685

KIAA1151 AB032977 KIAA1151 protein -0.56521377 -0.95738307 2 -0.7612984

CACNA1I AF142567 Calcium channel, voltage-dependent, alpha 1I subunit-0.55503678 -0.93969802 2 -0.7473674

TFEC D43945 Transcription factor EC -0.74754013 -0.74422819 2 -0.7458842

KIAA193 D83777 KIAA0193 gene product -0.91757685 -0.56213729 2 -0.7398571

EPHX1 NM_000120Epoxide hydrolase 1, microsomal (xenobiotic)-0.87217861 -0.60636676 2 -0.7392727

PRO2714 NM_018534Hypothetical protein PRO2714 -0.89962074 -0.51623757 2 -0.7079292

ZNF28 AC003973 Zinc finger protein 208 -0.67517041 -0.7261952 2 -0.7006828

RAC2 M64595 Ras-related C3 botulinum toxin substrate 2 (rho family, small GTP binding protein Rac2)-0.68866119 -0.71043787 2 -0.6995495

AJ276395 Homo sapiens mRNA for MSF-FN70 (FN gene)-0.54169472 -0.84441272 2 -0.6930537

AL034369 Human DNA sequence from clone 149D17 on chromosome Xq22.2-23. Contains part of a PLRP2 (PNLIPRP2, Pa-0.60013725 -0.77680433 2 -0.6884708

TRIP4 L40371 Thyroid hormone receptor interactor 4 -0.76642149 -0.60545613 2 -0.6859388

HCLS1 X16663 Hematopoietic cell-specific Lyn substrate 1-0.50070018 -0.8492648 2 -0.6749825

TIMP2 NM_003255Tissue inhibitor of metalloproteinase 2 -0.695665 -0.64558501 2 -0.670625

DKFZp586H623 AL096739 Hypothetical protein DKFZp586H0623 -0.57019613 -0.77100085 2 -0.6705985

KIAA273 D87463 KIAA0273 gene product -0.78756329 -0.53861163 2 -0.6630875

PTGS1 M59979 Prostaglandin-endoperoxide synthase 1 (prostaglandin G/H synthase and cyclooxygenase)-0.68414877 -0.63149194 2 -0.6578204

NFIX L31881 Nuclear factor I/X (CCAAT-binding transcription factor)-0.77217229 -0.54139827 2 -0.6567853

AL163202 Homo sapiens chromosome 21 segment HS21C002-0.74568656 -0.56590513 2 -0.6557958

Down-regulated #1 vs #8, <-0.7

Log2 ratio’s

• Biologists got together a few years ago and developed a sensible system called Genome Ontology (GO) Database

• 3 hierarchical sets of terminology

– Biological Process

– Cellular Component (location within cell)

– Molecular Function

• about 1000 categories of functions

Couple Gene List to Biological function:

Gene Ontology

ECMuPA, tPA Matrix Gla proteinFBN2 ADAM3AMMP1 ADAM12TIMP2 TFPI2COL6A2 COL3A1OPTC COL6A1KERA COL8A1

Cell AdhesionCD44 VWFCDH2 CD34TGFBI VCAM1ITGB5 PCDH1Neuropilin2 GJA4Claudin4 CTNNB1

TROAP

ApoptosisBCL2L1 SSTR3SGK CRDF14PAWR CASP1BOK TNFF8

CLUTNFRSF1B

REDOXHMOX1 NDUFA6TXNRD1 SEPP1

GSTT1CSR1

Cell growthP21 NDRG2CCDN1 ASCL2CCDN3 BMP2CSH2 LMO2IFRD2 IGFBP2QSCN6 IGF2NPPB

Transcr factorsId2b LMO2TTF1 HHEXTRIP4 MEOX2ZNF79 GFI1ZNF43 STAT1HES6 LBX1NFIX HNF3GTFEC IRF6EPAS1

InflammationTSA192 IFITM1FCGR3B NK4FCGR2B SCYA19TNFAIP6 CXCR4TNFSF4 CCR1TLR4 SCYA27

“senescence biological profile in vitro”

Overlap with cardiovascular disease in vivo

Genotype vs Phenotype

Senescent vascular cells in

human atheroma

Senescence-associated b-gal activity

Senescence in vitro Atherosclerosis in vivo

• Type I: (n = 2)

– How is this gene expressed in target 1 as compared to target 2?

– Which genes show up/down regulation between the two targets?

• Type II: (n > 2)

– How does the expression of gene A vary over time, tissues, or treatments?

– Do any of the expression profiles exhibit similar patterns of expression?

Microarray Experiment Design

Iyer et al. (1999) Science, 283: 83

The transcriptional program in the

response of human fibroblasts to serum

Identify genes with similar expression

Grouping unknown genes with known genes

may provide insight into function of unknown

genes

Cluster genes by similar changes - only really

meaningful across multiple treatments or time

points

Cluster samples by similar gene expression

profiles

>8 >6 >4 >2 >8>6>4>20 1 3 6 12 24 72hours:

genes

microarrays

Down Up

0 vs 1

0 vs 3

0 vs 6

etc..

Basics of Data Filtering and Visualization

0 1 3 6 12 24 72hours:

4,124 genes

spot-filtered

clustered

0 1 3 6 12 24 72hours:

464 genes

spot-filtered

ratio-filtered

clustered

0 1 3 6 12 24 72hours:

4,416 genes

Grouping genes: clustering

Microarray

(Clustering)

8600 cDNA clones

• Coordinated gene expression

• Differential gene expression

Iyer et al. (1999) Science, 283: 83

Biological information

Serum treatment ‘in vitro’

Wound healing ‘in vivo’

Expression

signatures

Couple expression to GO

GeneSpring

Visualisation

Expression Landscape of cell-cycle

regulated genes in yeast

GenMapp: Biological Pathways

Microarray and cancer

•Identification of prognostic biomarkers specific

to onset and progression

•Disease classification

•Development of drug resistance

•Risk of relapse assessment

•Metastasis

•Response to treatment

•Survival

Ross DT, et. al., Nature Genetics, (24): 2000, 227-235.

Aim:

to explore the variation in gene expression

of ~ 8,000 genes among 60 human cancer

cell lines (spanning 9 distinct tissues)

Variation in Gene Expression Patterns

in Human Cancer Cell Lines

1,161 genes: 60 cell lines

•Relationship between expression profile and tissue of origin•Recognize previously incorrect classified outliers•Recognize relationships to tumors in vivo

Hierarchical Clustering of Gene Expression Patterns

Groups Cell Lines According to Tissue of Origin

Alizadeh AA, et. al., Nature, (403): 2000,503-511.

Aim:

to determine whether gene expression profiling

could subdivide DLBCL – a clinically heterogeneous

diagnostic category – into molecularly distinct diseases

with more homogeneous clinical behaviors

Only 40% of patients respond well to therapy

“Lymphochip”: -17.856 cDNA clones

-lymphoid cell origin

-cancer + immunology

Distinct Types of Diffuse Large B-cell

Lymphoma (DLBCL)

45 DLBCL biopsies

Different B-cell differentiation stage

Set of ~3000 genes

Clustering Identifies 2 Major

Subgroups of DLBCL

a. Kaplan-Meier plot of overall survival of DLBCL patients grouped on the basis of gene expression profiling.

b. Kaplan-Meier plot of overall survival of DLBCL patients grouped according to the International Prognostic Index.

c. Kaplan-Meier plot of overall survival of low clinical risk DLBCL patients grouped on the basis of gene expression profiles.

DLBCL Subgroups Define

Prognostic Categories

Aim:

To classify breast carcinoma’s based on

expression profiling and to correlate these

to clinical outcome

2001, PNAS

Differential expressed genes: 476

• tumor properties

• patient outcome 85 biopsy samples

Clustering Identifies novel and existing

Subgroups of Breast cancer

Sorlie et al. PNAS 2003

ER++, PR++, G1,2 HER2 ISH pos “triple neg,” CK5/6+

overall survival: relapse-free survival:

Molecular classes are

predictive of outcome

Oncologists would like to use arrays to predict

whether or not a cancer is going to spread in the

body, how likely it will respond to a certain type of

treatment, and how long the patient will probably

survive.

It would be useful if the gene expression signatures

could distinguish between subtypes of tumours that

standard methods, such as histological pathology

from a biopsy, fail to discriminate, and that require

different treatments.

BioArray News (2, no. 35, 2002)

Arrays Hold Promise for Cancer Diagnostics

Gene expression profiling predicts

clinical outcome of breast cancer

Van ‘t Veer, et. al., Nature, (415): 2002,530-536.

Aim:

to determine whether gene expression profiling could

predict disease outcome and provide a strategy to select

patients who would benefit from adjuvant therapy

(metastasis)

Breast Cancer – Survival Pre-menopausal

patients, lymph node negative

~30% die <10 year

~70% survive >10 year

traditional diagnostics

Everyone receives

chemotherapy...!time (years)

su

rviv

al

time (years)

su

rviv

al

Current adjuvant treatment selection criteria:

• NIH (US) consensus criteria: > 95%• St Gallen (EU) consensus criteria: > 80%

receive adjuvant chemo- and hormonal therapy

As only 30% of these patients develop distantmetastases, some 50-65% of patients areover-treated with adjuvant (chemo)therapy

Breast Cancer – Survival Pre-menopausal

patients, lymph node negative

Identification of gene expression

changes in breast cancer

• analyse 98 breast tumors • 34 metastases-

positive <5 year– bad prognosis

• 44 metastases-negative >5 year– good prognosis

• 18 BRCA1 +• 2 BRCA2 +

‘spora

dic

’

Van’t veer, et. al.

98 breast tumors

analysed

34 ‘bad’ vs.

44 ‘good’

18 BRCA1 +

2 BRCA2 +

microarray with

24.000 genes

5.000 genes showed

expressional changes

in tumors

Different classes of

breast tumors...!

70-gene prognosis classifier for predicting risk

of distant metastasis within 5 yearsVan’t veer, et. al.

Supervisedclustering

Poor prognosis

Good prognosis

Microarray classification vs.

NIH classification

Classification of 158

breast cancer

tumors

Less unnecessary

chemo-therapy

Identification of

genes playing a role

in breast cancerClassical

NIH classification

59%

74%

5 % low risk

95 % high risk

96%

50%

Classification based

on microarray

39 % low risk

61 % high risk

Microarray to be used as

routine clinical screen

The Netherlands Cancer Institute in Amsterdam is the first institution in the world

to use microarray techniques for the routine prognostic screening of cancer

patients. Aiming for a June 2003 start date, the center will use a panoply of 70

genes to assess the tumor profile of breast cancer patients and to determine

which women will receive adjuvant treatment after surgery.

by C. M. Schubert

Nature Medicine

9, 9, 2003.

“Though each tumor is molecularly unique,

there exist common transcriptional cassettes

that underlie biological and clinical properties

of tumors that may be of diagnostic,

prognostic and therapeutic significance”.

Also true for other complex diseases

Expression profiling &

clinical application