#3946 Stratification of Metastatic Colorectal Cancer Patients Using DNA and RNA SequencingFang Yin Lo1, Sharon Austin1, Kellie Howard1, Mollie McWhorter1, Heather Collins1, Amanda Leonti1, Lindsey Maassel1, Christopher Subia1, Tuuli Saloranta1, Nicole Christopherson1, Kathryn Shiji1, Shradha Patil1, Saman Tahir1, Sally Dow1, Evan Anderson1, Jon Oblad1, Kerry Deutsch1, Timothy Yeatman2, Steven Anderson3 and Anup Madan1

1Covance, Seattle, WA; 2Gibbs Cancer Center, Spartanburg, SC; 3Covance, Durham, NC

IntroductionColorectal cancer (CRC) is the third most common type of cancer in the United States. Although chemotherapy, radiation and targeted therapies can improve survival rates, recent studies have shown the potential benefit of immunotherapies to improve outcomes for patients with advanced CRC. Targeted therapies that use monoclonal antibodies (mAbs) to EGFR have been shown to benefit some CRC patients.1 Until recently, KRAS has been the only predictive biomarker for anti-EGFR therapy for metastatic CRC. However, 40% to 60% of patients with wild-type KRAS do not respond to anti-EGFR therapy. Therefore, to accurately predict patients’ responses to treatments and improve clinical outcomes, additional prediction and treatment methods are imperative. One of the many efforts to improve prediction for CRC patients’ responses to the anti-EGFR therapy is the development of gene expression based RAS signature scores for identification of RAS activated tumors independent of mutations in the KRAS gene.2,3 In addition to passive immunotherapy using mAb, there have been major advances in targeted active immunotherapy in other tumors, including checkpoint inhibitors and cancer peptide vaccines.4,5 In melanoma, there have been preliminary clinical findings indicating that combined targeted therapies and simultaneous active immunotherapies such as blockade of multiple immune checkpoints could promote therapeutic synergy and improve clinical outcomes for patients. In addition, chromosomal rearrangements have the potential to alter gene function in many different ways. Recently there have been major advances in detecting these chromosomal rearrangements. Fusion genes such as BCR-ABL and EML4-ALK have become targets for therapy in cancer. There is considerable effort being placed on combinatorial ways of tumor stratification to improve responses for these cancers. Similarly, since no single treatment can apply to all CRC patients, we aim to stratify patients using a combination of the following methods:

1. RAS signature score based on the expression profile of 18 genes. This RAS signature score enables measurements of mitogen-activated protein/extracellular signal–regulated kinase (MEK) pathway functional output independent of tumor genotype.

2. Expression profile of immune checkpoint inhibitor target genes, such as PD1 and PD-L1.

3. DNA mutational profiles of genes such as KRAS, APC, BRAF and NRAS. Further, we explored potential gene fusion events in colorectal cancer tumor samples and discovered potential association between RAS gene signature score and the level of chromosomal rearrangements.

Methods and Results55 FFPE samples were selected from a cohort of 468 samples with matching FF samples. These 55 samples have about 1:1:1 ratio of high, medium and low RAS scores. Here we showed our capability to obtain RAS signature scores with concordant results using different platforms including whole transcriptome RNA-seq, Affymetrix® microarray (Affymetrix Inc.), targeted RNA-seq and Nanostring® (Nanostring Technologies, Inc.). We discovered that samples that have RAS activating mutations such as KRAS and BRAF have significantly higher RAS scores (p<0.001). On the contrary, expression of PD-L1 was significantly lower in tumor samples harboring mutations of genes such as MET, PTEN, NRAS, FBXW7 and GNAS. Kruskal-Wallis test showed that the expression of PD-L1 was significantly lower in samples with higher RAS signature scores (p<0.05). Furthermore, using the RNA-sequencing data, we were able to detect gene fusion events in these tumor samples. After filtering out low confidence results, a total of 730 gene fusion events were detected among the 55 tumor samples. While most of the gene fusion events were only detected once within the sample cohost, some were detected in multiple samples. For example, the fusion between KANSL1 and ARL17A was detected in 18 of the 55 samples. This is a relatively new discovery that had just started being mentioned in other cancer research institute reports.6 Other fusions that appeared multiple times include SAMD5 and SASH1. Interestingly, we discovered that significantly fewer fusion events were detected in samples with lower RAS signature scores than samples with higher RAS scores (p < 10-5). 1045 genes are involved in these fusion events. GO enrichment analysis shows that many of the cell cycle and phosphorylation associated pathways are significantly over-represented within these 1045 genes. Further analysis is being carried out for the implication of association between gene fusion events and RAS gene signature scores. Our analysis will lead to the development a combinatorial method for stratifying metastatic CRC patients.

CRC sample cohort

Ras signature scores calculated using FF samples

Samples with high Ras signature scores

Samples with medium Ras signature scores

Samples with lowRas signature scores

Select:- samples that have both FF and FFPE- about equal number from each Ras score group

55 samples for the pilot study

Figure 1. Colorectal cancer samples cohort selection strategy. The cohort was selected by filtering out colorectal cancer samples available as formalin-fixed, paraffin-embedded (FFPE) and flash frozen (FF). Samples were then filtered for known RAS score obtained from Affymetrix® array. Known RAS scores are divided into 3 groups evenly: low (<33% percentile), medium (33%-66% percentile), high (>66% percentile).

References1. Gallagher DJ, Kemeny N. Metastatic colorectal cancer: from improved survival to potential cure. Oncology. 2010;78:237–248.

2. EGFR gene copy number as a prognostic marker in colorectal cancer patients treated with cetuximab or panitumumab: a systematic review and meta-analysis

3. KRAS mutation status is predictive of response to cetuximab therapy in colorectal cancer. Lièvre A, Bachet JB, Le Corre D, Boige V, Landi B, Emile JF, Côté JF, Tomasic G, Penna C, Ducreux M, Rougier P, Penault-Llorca F, Laurent-Puig P. Cancer Res. 2006 Apr 15; 66(8):3992-5.

4. Loboda A et al. A gene expression signature of RAS pathway dependence predicts response to PI3K and RAS pathway inhibitors and expands the population of RAS pathway activated tumors.

5. BMC Medical Genomics 2010, 3:26Dry JR et al. Transcriptional Pathway Signatures Predict MEK Addiction and Response to Selumetinib (AZD6244). Cancer Res. 2010 Mar 15;70(6):2264-73.

6. http://newswise.com/articles/rutgers-cancer-researchers-examine-gene-fusion-and-treatment-implications-for-breast-cancer

7. Genome Medicine 20157:43.

Presented at AACR 2016Covance is the drug development business of Laboratory Corporation of America Holdings (LabCorp). Content of this material was developed by scientists who at the time were affiliated with LabCorp Clinical Trials or Tandem Labs, now part of Covance.

55 FFPE colorectal cancer samples

DNA Analysis RNA Analysis

Exome Analysis

Targeted Mutational Analysis

TruSight mutational panel

Whole Transcriptome Analysis

RNA-seq

Affymetrix® Microarray

Nanostring®

Targeted RNA-seq

Targeted Transcriptome Analysis

Figure 2. Multi-platform comparison. Samples derived from the same 55 FFPE blocks were assayed across multiple platforms. The method design to combine RNA analysis (gene expression signature scores) with DNA analysis (i.e. mutation status) allows for comparison of RAS signature scores and overall gene expression from different platforms.

1FF and 5 FFPE Datasets

Quality Control and Data Normalization

Affy,FF

Affy,FFPE

RNASeq-Acc,FFPE

RNASeq-rRNAdep, FFPE

TargetedRNASeq,FFPE

Nanostring,FFPED

ata

Sou

rce

Gen

e E

xpr S

igna

ture

Sco

reD

NA

and

mut

atio

nal

info

rmat

ion

QC

Affy,FF

Affy,FFPE

RNASeq-Acc,FFPE

RNASeq-rRNAdep, FFPE

TargetedRNASeq,FFPE

Nanostring,FFPE

Ras Signature Score Calculation

Correlation Analysis

Gene Mutation, e.g. KRAS/BRAF/NRAS Mutation Burden Gene Fusion Events

Regression Analysis

Data Integration for Tumor Stratification

Figure 3. Flowchart for the analysis. 55 samples were gone through 5 different platforms for gene expression measurements – Affymetrix®, whole transcriptome RNA-Seq by two different library preparation methods, targeted RNA-Seq, and Nanostring®. Data went through quality control and normalization. For RAS score calculation, 18 genes were used based on previous study.5

targ

eted

RN

Ase

q

Affy

met

rix®

RN

AS

eq.R

NA

Acc

ess

RN

AS

eq.r

RN

Ade

plet

ion

Nan

ostr

ing®

targeted RNAseq

Affymetrix®

RNASeq.RNAAccess

RNASeq.rRNAdepletion

Nanostring®

0.5 0.6 0.7 0.8 0.9 1

Pearson correlation coefficients

RAS

scor

es fr

om Aff

ymet

rix®

RAS scores from RNASeq-RNAAccess

BA

R=0.88

0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

0 0.05 0.1 0.15 0.2

Figure 4. Correlation of RAS signature scores between different platforms. (A) RAS scores calculated from multiple platforms are significantly correlated. All pairwise comparison has rho>=0.5 and p < 0.05. (B) Correlation of RAS scores calculated from RNA-seq and Affymetrix®. Scores between different platforms are highly correlated.

Figure 5. RAS signature scores versus mutation call. (A) Samples with KRAS mutation have significantly higher RAS signature scores compared with samples with wild type KRAS. (B) Combine information from gene expression, RAS signature score and mutation status.

�

�

� �

�

�

�

�

�

�

��

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

��

�

�

�

�

�

�

�

�

��

�

�

�

�

p(aov)= 3.04e�030.04

0.08

0.12

0.16

KR

AS

_mut

KR

AS

_wt

mutation call

RA

Ssc

ore

s

KRAS mutation status vs. RAS scores

DS−

5629

3D

S−52

681

DS−

6029

6D

S−51

803

DS−

5311

4D

S−60

252

DS−

5412

9D

S−52

838

DS−

5198

2D

S−51

941

DS−

5450

3D

S−40

199

DS−

5436

3D

S−54

564

DS−

3363

5D

S−48

764

DS−

5279

0D

S−51

997

DS−

5345

3D

S−49

639

DS−

7029

4D

S−54

041

DS−

4889

3D

S−53

191

DS−

5221

0D

S−47

369

DS−

5478

3D

S−49

315

DS−

4805

5D

S−58

258

DS−

5215

1D

S−48

607

DS−

5834

1D

S−54

589

DS−

6798

0D

S−56

858

DS−

4487

8D

S−50

687

DS−

5696

3D

S−54

051

DS−

5197

7D

S−51

652

DS−

4885

7D

S−56

376

DS−

5285

3D

S−51

043

DS−

5232

0D

S−56

326

DS−

5321

1D

S−49

796

DS−

5092

5D

S−51

797

DS−

4982

6D

S−60

353

DS−

6795

5

LGALS3FXYD5ETV4LZTS1KANK1ETV5SPRY2ELF1DUSP4SLCO4A1ZFP106TRIB2S100A6MAP2K3SERPINB1PROS1PHLDA1DUSP6

−6−4

−20

24

6

log2

inte

nsity

AstraZenica RAS signature gene expression from Affymetrix® microarray (log2)

Mean expression of all 18 genes

Low High

RAS signature score

BRAFAPCKRAS

TP53PIK3CA

NRAS

mutation status

WTMutant

A B

Figure 6. The number of mutation versus KRAS status. KRAS mutant samples have significantly higher number of non synonymous mutations than KRAS wild type samples.

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

��

�

�

�

�

�

�

�

�

�

�

�

�

�

�

��

�

�

�

�

�

�

�

�

��

�

�

p(aov)= 2.5e�02500

1000

1500

KR

AS

_mut

KR

AS

_wt

mutation call

mu

tNu

m

KRAS mutation status vs. mutation number

Figure 7. RAS signature scores and the number of gene fusion events. (A) Distribution of gene fusion events of all samples. Only high and medium confidence gene fusion events based on results from JAFFA7 were considered. (B) Samples with higher number of gene fusion events have significantly higher RAS signature scores (p<0.001).

0

2

4

6

0 10 20 30 40Number of gene fusion events

cou

nt

Distribution of number of gene fusion events

Median = 17

�

�

��

�

�

�

�

�

�

� �

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

��

�

�

�

�

�

�

�

�

��

�

�

�

�

p(aov)= 5.79e�050.04

0.08

0.12

0.16

high

_fus

ion_

num

low

_fus

ion_

num

RA

S s

core

s

RAS scores vs. Number of fusion events

A B

Figure 8. Number of gene fusion events and RAS signature scores. Samples with lower RAS scores have significantly fewer gene fusion events detected than samples with higher RAS scores.

Figure 9. Scatter plot of enriched GO cluster representatives. Multidimensional scaling is applied to the list of significantly enriched GO terms in fusion genes found in the CRC samples.6

�

�

�

��

�

�

��

��

��

�

�

�

�

�

�

�

�

��

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

�

��

�

�

�

�

�

�

�

p(aov)= 9.97e�06

0

10

20

30

40

low

_ras

scor

e

mid

_ras

scor

e

hi_r

assc

ore

Nu

mb

er o

f fu

sio

n e

ven

ts

Number of fusion events vs. RAS scores

Table 2. Fusion Genes That Are Kinases

Associated Gene Name DescriptionABR active BCR-related AKT3 v-akt murine thymoma viral oncogene homolog 3 BAZ1A bromodomain adjacent to zinc finger domain 1A BAZ1B bromodomain adjacent to zinc finger domain 1B BLK BLK proto-oncogene, Src family tyrosine kinase BLVRA biliverdin reductase A BMPR2 bone morphogenetic protein receptor type II CASK calcium/calmodulin-dependent serine protein kinase (MAGUK family) CDK6 cyclin-dependent kinase 6 CDK9 cyclin-dependent kinase 9 CLK3 CDC like kinase 3 DCLK2 doublecortin like kinase 2 FGFR1 fibroblast growth factor receptor 1 FGFR4 fibroblast growth factor receptor 4 INSR insulin receptor JAK2 Janus kinase 2 LATS1 large tumor suppressor kinase 1 LRRK1 leucine-rich repeat kinase 1 LRRK2 leucine-rich repeat kinase 2 LYN LYN proto-oncogene, Src family tyrosine kinase MAP3K7 mitogen-activated protein kinase kinase kinase 7 MAPKAPK2 mitogen-activated protein kinase-activated protein kinase 2 MAPKAPK5 mitogen-activated protein kinase-activated protein kinase 5 MARK3 MAP/microtubule affinity-regulating kinase 3 NEK9 NIMA-related kinase 9 PAK1 p21 protein (Cdc42/Rac)-activated kinase 1 PAN3 PAN3 poly(A) specific ribonuclease subunit PKN2 protein kinase N2 PLK2 polo-like kinase 2 RIOK2 RIO kinase 2 SCYL2 SCY1-like, kinase-like 2 SRPK2 SRSF protein kinase 2 TNIK TRAF2 and NCK interacting kinase

Fusion Frequency

KANSL1:ARL17A 18

SAMD5:SASH1 16

LMO7:EXT2 9

DPP4:FAP 7

C10orf68:CCDC7 6

NOXA1:SLCO4A1 5

NR3C2:NR3C1 5

OSBPL2:OSBPL1A 5

PRPF19:ABR 5

USP7:SPARC 5

PHF20L1:KIAA0753 4

RP11-123O10.4:GRIP1 4

WFDC10B:FA2H 4

AKR1C1:AKR1E2 3

BIN2:MAN2A1 3

Fusion Frequency

CCDC125:MAGED2 3

CEACAM5:CEACAM7 3

CMTM8:CMTM7 3

DLG5:DLG3 3

LPHN2:LPHN3 3

PDE4DIP:RP11-353N4.1 3

RNF123:SERINC4 3

RP11-141M1.3:STARD13 3

RP11-680G10.1:GSE1 3

SAMD5:RP11-307P5.1 3

SF3B2:PHF17 3

SRPK2:KMT2E 3

TFG:GPR128 3

VTI1A:RP11-57H14.3 3

Table 1. Frequency of Gene Fusion Events*

* Gene fusion events that occurred more than 3 times in the sample cohort.