PowerPoint Presentation - Asia Pacific Bioinformatics Network's 10th

Document Sample
PowerPoint Presentation - Asia Pacific Bioinformatics Network's 10th Powered By Docstoc
					Integative Genomic Approaches to Personalized
Cancer Therapy


Patrick Tan, MD PhD




International Conference on Bioinformatics
Singapore, Sept 09 2009
       Genomic Oncology in Singapore : Translating
               Information into Knowledge




                      Disease Genes
Clinical Biomarkers                      Cancer Pathways
Basic Science to Translation

   1) Metastasis Genes
                              - Network Structures

   2) Cancer Classification
                              - Pathway Signatures

   3) Lung Cancer Outcome
                              - Integrative Genomics
              Biological Networks – Robust Yet Fragile




                                                    Hub Gene
  Edge Gene
   Tolerant                                     Ultrasensitive

Wide Variation                                      Low Variation




Can we infer ‘hub-like’ genes in cancer?
                                           Yu Kun
 Identifying Precisely Controlled Genes in Cancer




 Lung       Thyroid   Liver   Esophagus    Breast


                        270 Tumors



Large                                Restricted
Variation                            Variation
Restricted Variation Only in Cancers



                                       Cancer

                                       Non-malignant




48 Precisely Controlled Genes in Cancers
  The PGC is Precisely Controlled in Many Solid Tumors


Tumor                        Significance
Gastric, NPC (99)

Breast (286)

Lung (118)

Ovarian (146)

Breast (189)

Glioma (77)

Colon (100)

                    0 1 2 3 4 5 6 7 8 9 10 11 12 13
  The PGC is NOT Precisely Controlled in Normal Tissues


Normal                     Significance
Novartis (158)

Ge et al (36)




                 0 1 2 3 4 5 6 7 8 9 10 11 12 13
PGC Genes are Enriched in the Integrin Signaling Pathway




                               Growth Factor Regulation
                               RAS/MAPK Signaling
                               PI3K Signaling
                               JNK/SAPK Signaling
                               Cytoskeletal Interactions
                               Cell Motility
 Implications of Precise PGC Regulation


Dedicated Cellular Mechanisms to Ensure Accurate
Expression


A Functional Requirement for Tight PGC Control in
Tumors?



     Are Tumors Ultrasensitive to PGC Activity?
PGC Expression in Breast Cancer Cell Lines
                                  P=0.01
30 Breast Cancer
    Cell Lines
                         Non-invasive      Invasive




                   PGC
PGC Expression in Experimental Metastasis




      HCT116       Splenic      Liver
    Tumor Cells   Injection   Metastases




                                    Adapted from Clark et al (2000)
Reduced PGC Expression Correlates with Metastatic Potential




                                                    P=0.022
siRNA Knockdown of PGC Genes Enhances Metastasis




   p53CSV siRNA                          qRT-PCR
PGC Expression in Primary Tumors
Reduced PGC Expression Predicts Clinical Prognosis


                                   Elevated PGC
                                   Decreased PGC
Are Low-Variance Genes True Hubs?
       (Lessons from Yeast)
                       mRNA variance overlaid on a
                       protein-protein network

                       Black nodes = missing data.

                       A: proteasome regulatory lid
                       B: mediator complex
                       C: SAGA complex
                       D: SWR1 complex




                                Goel and Wilkins,
                                unpublished.



                       Slide Courtesy of Marc Wilkins
Take Home Messages


- A General Strategy for Identifying Tightly Regulated
Genes

- A Precisely Regulated Expression Cassette in Cancer

- Fine-scale alterations potently modulate tumor behaviour
and clinical outcome

-Not discernible by conventional microarray analysis
methods




                                   Yu et al (2008) PLOS Genetics
Basic Science to Translation

   1) Metastasis Genes
                              - Network Structures

   2) Cancer Classification
                              - Pathway Sigantures

   3) Lung Cancer Outcome
                              - Integrative Genomics
    High Prevalence of Gastric Cancer in Asia



Global Cancer Mortality

Lung (1.3 million deaths/year)
Stomach (1 million deaths/year)
Liver (662,000 deaths/year)
Colon (655,000 deaths/year)
Breast (502,000 deaths/year)

               - WHO, 2005




                                   From The Scientist, Sep 22, 2003
Tumor Heterogeneity Impacts Response

    CML                 Gastric Cancer
“One Disease”          “Many Diseases”




  Imatinib                  5-FU




100% Response            20% Response
         Pre-Selecting Patients for Optimal Therapy
                          Gastric Cancer




Subtype A   Subtype B   Subtype C   Subtype D   Subtype E   Subtype F



  Rx 1        Rx 2        Rx 3         Rx 4       Rx 5       Rx 6
Expression Signatures as Cancer Phenotypes




 Tumor Type A             Tumor Type B
  (“State A”)              (“State B”)




                           Genes


                A     B
Expression Signatures
Capture Heterogeneity




                        Tay et al., Cancer Research (2003)
 Using Pathway Signatures to Guide Targeted Therapies


Experimental System          Pathway A
                                                                     1.5



                                                                     1



                                                                     0.5



                                                                     0



                                                                     -0.5



                                                                     -1
                                                                            Chia Huey Ooi
                                                                     -1.5



                              -1.5   -1   -0.5   0   0.5   1   1.5




            Tumor Profiles




Pathway A
Mapping Pathway Signatures to Tumor Profiles

 Pathway A                                      Pathway B                                      Pathway C                                      Pathway D                                      Pathway E
                                         1.5                                            1.5                                            1.5                                            1.5                                            1.5



                                         1                                              1                                              1                                              1                                              1



                                         0.5                                            0.5                                            0.5                                            0.5                                            0.5



                                         0                                              0                                              0                                              0                                              0



                                         -0.5                                           -0.5                                           -0.5                                           -0.5                                           -0.5



                                         -1                                             -1                                             -1                                             -1                                             -1



                                         -1.5                                           -1.5                                           -1.5                                           -1.5                                           -1.5



  -1.5   -1   -0.5   0   0.5   1   1.5           -1.5   -1   -0.5   0   0.5   1   1.5           -1.5   -1   -0.5   0   0.5   1   1.5           -1.5   -1   -0.5   0   0.5   1   1.5           -1.5   -1   -0.5   0   0.5   1   1.5




                         Tumor Profiles




  A

  B

  C

  D
Predominant Oncogenic Pathways in Gastric Cancer

                    200 primary gastric tumors

                                                                                            P21
                                                                                            E2F (a)
                                         Proliferation                                         1
                                                                                            E2F (b)
                                          /stem cell                                        Stem cell (a)
                                                                                               0.8
                                          pathways                                          Stem cell (b)




                                                                       Oncogenic Pathways
                                          activated                                         Myc (a)
                                                                                               0.6
                                                                                            Stem cell (c)
                                                                                            Myc (b)
                                                                                               0.4
                                                                                            NF-kB (a)
               b-catenin                                                                    Wnt
                pathway                                                                        0.2
                                                                                            NF-kB (b)
               activation                                                                   p53 (a)
                                                                                            HDAC
                                                                                               0

                                                                                            b-catenin
  p53 pathway                                                                               Src-0.2
   activation                                                                               Ras
                                                                                            BRCA1
                                                                                               -0.4
                                                                                            HDAC
                                                                                            p53 (b)
                                                                                               -0.6
                                                                                            BRCA1
                                                                                              -0.8

                                  Activation score                                            -1
 -1   -0.8   -0.6   -0.4   -0.2   0      0.2     0.4   0.6   0.8   1
                Validating Oncogenic Pathway Predictions
Pathways   Proliferation   Wnt         NFKB




                       GC cell lines
High Proliferation Scores are Associated with Rapid Growth
        Proliferative capacity vs. combined E2F+Myc+Stemcell activation score for
                                      22 GC cell lines
              4

                                     3.5
                        Proliferative capacity
        Proliferative capacity




                                                 3


                                     2.5

                                                 2

                                     1.5
                                                                                            R = 0.5051
                                                                                            p = 0.0165
                                                 1                                                           48h
                                                     0.2        0.4          0.6          0.8            1   Linear (48h)
                                                           Summarized activation activation the
                                                           E2F+Myc+Stemcell combined score of score
                                                              proliferation/stem cell cluster
High Wnt Scores are Associated with Wnt Activity




      b-catenin pathway activation
                                     0.6




          In-silico prediction of
                                     0.4

                                     0.2

                                       0




                                            AGS




                                                                                              SNU16
                                                     YCC3




                                                                                SNU1

                                                                                       SNU5
                                                                      NCI-N87
                                                            KatoIII
                                     -0.2

                                     -0.4
         TCF7L2 activity




                                                  Relative constitutive TCF7L2 activity
            (folds)




         TCF7L2:
Oncogenic Pathways in Gastric Cancer are Functionally
Significant
                                                      5
                                                                           p=4.549106
                                                      4




                                            Proliferative
                                              capacity
              Cell Lines
                                                      3                                                          MKN1

                                                      2                                                          MKN1


                                                      1
   Pathways




                                                      0
                                     NFKB                                 p65
                                                             Control T72/T0 shRNA
                                                             shRNA
                                                                                      Annexin +ve cells


                                                                          60


                                      Wnt                                 50   Cell Death




                                                            % apoptotic
                                                                          40
                                                                               Assay




                                                               cells
                                                                          30

                                                                          20
                        Neg siRNA                                         10

                   b-catenin siRNA                                         0

                    b-catenin (WB)                                             Neg siRNA
                                                                                 Neg siRNA        b-catenin siRNA
                                                                                                     B-Catenin siRNA

                          GC cell
                        Actin (WB)lines
      Pathway Interactions Influence Survival
  Single                               Pathway
 Pathways                            Combinations

   NFKB
                           NFKB +
                           Prolif.


Proliferation

                           Wnt +
                           Prolif.
     Wnt
 Clinical Validation of Pathway Combinations

                Singapore (200)    Australia (90)


Proliferation
and NKFB




Proliferation
and Wnt
Oncogenic Pathways in Gastric Cancer May Guide Therapy

                    200 primary gastric tumors                                                              Potential
                                                                                                            Therapies
                                                                                            P21
                                                                                            E2F (a)
                                                                                               1            HLM006474
                                                                                            E2F (b)
                                                                                            Stem cell (a)
                                                                                               0.8
                                                                                            Stem cell (b)




                                                                       Oncogenic Pathways
                                                                                            Myc (a)
                                                                                               0.6
                                                                                                            CX-3543
                                                                                            Stem cell (c)
                                                                                            Myc (b)
                                                                                               0.4
                                                                                            NF-kB (a)       RTA-402
                                                                                            Wnt
                                                                                               0.2
                                                                                            NF-kB (b)
                                                                                            p53 (a)
                                                                                            HDAC
                                                                                               0            PXD-101
                                                                                            b-catenin
                                                                                            Src-0.2         KX2-391
                                                                                            Ras             Salirasib
                                                                                            BRCA1
                                                                                               -0.4
                                                                                            HDAC
                                                                                            p53 (b)
                                                                                               -0.6         pifithrin-a
                                                                                            BRCA1
                                                                                              -0.8

                                  Activation score                                            -1
 -1   -0.8   -0.6   -0.4   -0.2   0      0.2     0.4   0.6   0.8   1
Take Home Messages

 •   A framework for mapping defined pathway
     signatures into complex tumor profiles

 •   Signatures are transportable (in vitro to in vivo)

 •   Gastric cancers can be subdivided by pathway
     activity into biologically and clinically relevant
     subgroups

 •   “High-throughput pathway profiling” highlights the
     role of oncogenic pathway combinations in clinical
     behavior
                                    Ooi et al (2009) PLOS Genetics
Basic Science to Translation

   1) Metastasis Genes
                              - Network Structures

   2) Cancer Classification
                              - Pathway Biology

   3) Lung Cancer Outcome
                              - Integrative Genomics
Genomic Classification of Early Stage
            Lung Cancer
               Philippe and Sophine Broet
               INSERM U472, Faculté de Médecine
               Paris-Sud



           Lance Miller
           Wake Forest University, USA




                       Broet et al., (2009) Cancer Research
Adjuvant Chemotherapy in Early-Stage NSCLC

                           Observation
                           (Watch and Wait)
                Surgery
                          40-50% 5-yr Survival

                            Chemotherapy?
  Stage I, II
Study Questions
Clinical questions
Can we use genomics to discriminate between low
risk (pseudo-stage I) & high risk (pseudo-stage
II) groups?

Previous studies on NSCLC prognosis have been
transcriptome centered, not incorporating
genomic alterations
    An Integrated Genomic Strategy to
  Identify “Poor Prognosis” NSCLC Cases
                           Array-CGH



                       Recurrent Amplifications
                            And Deletions




                      Gene Expression Profiling
  Stage IB
  NSLCLCs                            Highly Regulated
(Training Set)                       Genes
Recurrent Genomic Alterations in NSCLC
   1q31      5p13       8q24   11q13




             CyclinD1            WWOX
Genomic Regions Associated with Outcome




Survival associations – “Survival CNAs”
 Gene Expression Associated with Survival-CNAs


  Gene                  Copy Number Driven Expression
Expression       203342_at 205564_at 201699_at        202988_at
                    204322_at        201698_at    203301_at
                       2113458_at              203343_at
                                          201408_at



 Survival CNAs
Predicting Prognosis in Stage IB NSCLC

                      Integrated Signature
                      103 genes (Chr. 7, 16, 20, 22)

                   Good Prognosis

                     P=0.002

                   Poor Prognosis


                               Training Cohort
Validation of the Integrated Signature

   Michigan Series: 73 Stage I A&B NSCLCs


                    Good Prognosis
                               P=0.025
                    Poor Prognosis
Another Validation of the Integrated Signature

       Duke Series: 31 Stage I A&B NSCLCs


              Good
              Prognosis

                P=0.003

                  Poor          Candidates for
                  Prognosis     Chemotherapy?
Implications for Chemotherapy Selection


                                Stage II
                                NSCLC




Poor Prognosis
Stage IB


                    Poor Prognosis Ib Patients
                    Are Comparable to Stage II
                    Patients
            A Genomic Approach
Stage Ib    to Guide Chemotherapy
NSCLC
            in Early-Stage NSCLC


Surgery
                 Good
               Prognosis
                                Observation
            (“Stage Ia-like”)
Genomic
Predictor
                  Poor          Adjuvant
               Prognosis        Chemotherapy
            (“Stage II-like”)
Acknowledgements
  Kun Yu                 Philippe Broet (Paris)
  Kumaresan Ganesan      Sophine Broet (Paris)
                         Lance Miller (GIS)
  Ooi Chia Huey          Elaine Lim (NUH)
  Tatiana Ivanova
                         Wei Chia Lin (GIS)
  Shenli Zhang           Hooi Shing Chuan (NUS)
  Wu Yonghui
                         Alex Boussioutas (Peter Mac, AU)
  Lai Ling Cheng         David Bowtell (Peter Mac, AU)
  Veena Gopalakrishnan   Sun Yong Rha (S. Korea)
  Jun Hao Koo            Heike Grabsch (Leeds)
  Julian Lee
  Ming Hui Lee
  Iain Tan               Support :
  Angie Tan              French-Singapore MERLION program
  Jiong Tao              Singapore Cancer Syndicate
  Jeanie Wu              Biomedical Research Council
  Yansong Zhu            National Medical Research Council

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:14
posted:7/7/2012
language:English
pages:49