Docstoc

Cancer Pharmacogenomics

Document Sample
Cancer Pharmacogenomics Powered By Docstoc
					Cancer Pharmacogenomics

   Caroline F. Thorn, Ph.D.
   Thorn@helix.stanford.edu



      Stanford – South Africa   Biomedical Informatics Program
                       Overview

• Host genome and cancer genome
• Etiology of cancer/genetics of cancer
  predisposition
• Whole genome approaches
  – Patient samples / ex vivo
  – Cell lines / in vitro



              Stanford – South Africa   Biomedical Informatics Program
Host Genome and Cancer Genome

 • What is the cancer genome?
 • How is cancer cell different from host
   cell?
   – genetic change(s) in cancer cell may alter its
     response to drug




              Stanford – South Africa   Biomedical Informatics Program
Host Genome and Cancer Genome

 What makes a cancer cell different?
 • Loss of cell cycle control so cancer cell is not
   contact inhibited and can replicate
 • Evasion of normal methods of cell death, loss
   of tumour suppressors, decreased apoptosis
 • Amplification of detoxifying pathways,
   increased activity of drug transporters
 • Increase in growth factor receptors
   – E.g. Her-2/neu


               Stanford – South Africa   Biomedical Informatics Program
 Host Genome and Cancer Genome

• So what if we target drugs to those
  differences ?




            Stanford – South Africa   Biomedical Informatics Program
    Her/neu and trastuzumab

• Her-2 oncogene, over expressed in ~30%
  breast cancers
• Associated with aggressive tumours and poor
  prognosis
• Is expressed on cancer cell surface
• Drug is antibody to target it - trastuzumab
  aka Herceptin,
• Limits toxicity as targets only the cancer cells
= successful example

              Stanford – South Africa   Biomedical Informatics Program
           Gefitinib (Iressa)

• Inhibits the growth factor receptor EGFR
• Promising early trials - but only improved
  survival in subset of patients, 10-15%
  – Female,
  – Japanese
  – Non-smokers

  Example that shows we have to remember
   about the host genome as well

              Stanford – South Africa   Biomedical Informatics Program
          Gefitinib (Iressa)

• Iressa works better in those with variant
  EGFR that makes it more susceptible to
  the drug




            Stanford – South Africa   Biomedical Informatics Program
         Etiology of cancer

• 2-hit hypothesis, Knudsen : mutation
  (may be germ line) in one tumour
  suppressor gene or oncogene can be
  tolerated, but second somatic change
  then tips balance from controlled growth
  to uncontrolled growth
• More likely to be several hits


            Stanford – South Africa   Biomedical Informatics Program
           Cancer predisposition
• Even though all cancer has genetic changes and there
  are several well known examples of inherited genetic
  risk factors e.g. BRCA1 and breast cancer, only 5-10%
  of cancer is inherited
• Most cancers come from random mutations over
  lifetime
   – mistake when cells are going through cell division
   – response to injuries from environmental agents such as
     radiation or chemicals.
   – viruses



                   Stanford – South Africa   Biomedical Informatics Program
                Other omics

• Worth mentioning efforts to examine
  gene-environment interactions of
  importance in cancer research
  – Nutrigenomics, interaction between diet and
    genes
  – Toxigenomics, interactions between toxins
    (e.g. Carcinogens in environment) and genes



             Stanford – South Africa   Biomedical Informatics Program
         Whole genome approaches
                                                  • Can’t test cancer drugs
                                                    on healthy volunteers in
                                                    same way as other drug
                                                    classes so need to
                                                    exploit other methods
                                                    for PGx discovery
                                                        – Expression profiling in
                                                          patient samples
                                                        – In vitro drug response in
                                                          cell lines
From McLeod review, JCO 2005
                                                        – mouse models (will not
                                                          cover in this course)
                        Stanford – South Africa   Biomedical Informatics Program
        Expression profiling

• Using microarray or RT-PCR in high-
  throughput system to find a set of genes
  that respond to drug (up or down
  regulated) or are found in certain disease
  type and give characteristic “signature”




            Stanford – South Africa   Biomedical Informatics Program
Expression profiling on patient samples




           Stanford – South Africa   Biomedical Informatics Program
 Evans/Relling group, NEJM, 2004

Background
• Recap- leukemia therapy effective for
  ~80% patients
• but reason why the 20% fail is unclear
• Leukemia usually treated with cocktail of
  drugs in successive phases



             Stanford – South Africa   Biomedical Informatics Program
 Evans/Relling group, NEJM, 2004

• Tested leukemia cells with 4
  antineoplastic drugs with different
  mechanisms of action
• Measured sensitivity/resistance to
  – asparaginase (breaks down amino acid)
  – daunorubicin (inhibits DNA and RNA
    synthesis)
  – prednisolone (steroid/anti-inflammatory),
  – vincristine (cell cycle arrest),

             Stanford – South Africa   Biomedical Informatics Program
Evans/Relling group, NEJM, 2004




        Stanford – South Africa   Biomedical Informatics Program
 Evans/Relling group, NEJM, 2004

• Run microarrays of gene expression
• Identify differenially expressed genes
  – Supervised clustering using Wilcoxon rank-sum test
• Build model on 2/3 and test on 1/3 with 1000
  iterations
• Validate in second group of patients




               Stanford – South Africa   Biomedical Informatics Program
Evans/Relling group, NEJM, 2004




 Identified 40 genes that discriminate between
 vincristine sensitive and resistant cells



             Stanford – South Africa   Biomedical Informatics Program
  Evans/Relling group, NEJM, 2004

• Of the 124 differentially expressed genes
  identified only 3 had previously been
  linked with resistance to doxorubicin or
  vincristine




             Stanford – South Africa   Biomedical Informatics Program
Evans/Relling group, NEJM, 2004




        Stanford – South Africa   Biomedical Informatics Program
 Whole genome approaches

• Expression profiling in patient samples
• In vitro drug response in cell lines




          Stanford – South Africa   Biomedical Informatics Program
Watters et al, PNAS 2004




     Stanford – South Africa   Biomedical Informatics Program
  In vitro drug response in cell lines
• Example phenotype:
  drug cytotoxicity
• Easy to measure
- Cells counted, plated at 1 x
  104 /well
- Cells incubated with
  increasing concentrations of
  drug
- Alamar blue vital dye
  indicator added
- Viability relative to
  untreated control calculated
  by spectrophotometry


                     Stanford – South Africa   Biomedical Informatics Program
        Which cells to use?

• Repository of publicly available cell lines
  and DNA samples from families from
  defined populations
• Major benefit of using these samples -
  they have been extensively genotyped
• CEPH data has pedigree (family
  relationship) data


             Stanford – South Africa   Biomedical Informatics Program
Centre d’ Etude du Polymorphisme Human




          Stanford – South Africa   Biomedical Informatics Program
Are the CEPH cell lines a good model?

• Why not just use tumour cell lines?
  – Not enough from the same tissue of origin, don’t
    have genetics data
• But do these cells behave the same way as
  tumour cells?
  – Ran tests to show CEPH cells had the same IC50 as
    some publicly available tumour cell lines
  – Both CEPH cells and publicly available tumour cell
    lines responded to toxicity by same pathway of
    apoptosis


               Stanford – South Africa   Biomedical Informatics Program
    Watters et al, PNAS 2004

• Used Random
  Coefficient Regression
  model to assign
  coefficient for slope of
  each individual's cells
  response
• Used SEGPATH to
  assess heritability of                       Response curve for one individual
  response and linkage


              Stanford – South Africa   Biomedical Informatics Program
Linkage on chromosome 9




     Stanford – South Africa   Biomedical Informatics Program
             New candidate genes
RAD23B        NM_002874        Hs.159087   RAD23 homolog B (S. cerevisiae)                                                       105472558-105473090
FCMD          NM_006731        Hs.55777    Fukuyama type congenital muscular dystrophy (fukutin)                                 103782702-103783131
CDW92         AW165999         Hs.414728   CDw92 antigen                                                                         103525239-103527612
ABCA1         NM_005502        Hs.147259   ATP-binding cassette, sub-family A (ABC1), member 1                                   102923739-102924215
SMC2L1        AU154486         Hs.119023   SMC2 structural maintenance of chromosomes 2-like 1 (yeast)                           102282761-102283241
RNF20         AK022532         Hs.168095   ring finger protein 20                                                                99704963-99705433
ZNF189        NM_003452        Hs.50123    zinc finger protein 189                                                               99552272-99552713
MRPL50        BG028213         Hs.288224   mitochondrial ribosomal protein L50                                                   99532436-99532607
INVS          AF039217         Hs.150744   inversin                                                                              98439227-98443060
TXNDC4        BC005374         Hs.154023   thioredoxin domain containing 4 (endoplasmic reticulum)                               98124676-98149788
SEC61B        NM_006808        Hs.191887   Sec61 beta subunit                                                                    97370032-97372556
ALG2          BE967331         Hs.40919    asparagine-linked glycosylation 2 homolog (yeast, alpha-1,3-mannosyltransferase)         97359553-97359942
TGFBR1        AA604375         Hs.28005                                                                                             97295384-97295719
                                           transforming growth factor, beta receptor I (activin A receptor type II-like kinase, 53kDa)
NANS          NM_018946        Hs.274424   N-acetylneuraminic acid synthase (sialic acid synthase)                                  96222949-96225118
ANP32B        NM_006401        Hs.459987   acidic (leucine-rich) nuclear phosphoprotein 32 family, member B                         96157636-96158013
XPA           NM_000380        Hs.288867   xeroderma pigmentosum, complementation group A                                        95817102-95817607
NCBP1         BC001450         Hs.439203   nuclear cap binding protein subunit 1, 80kDa                                          95808875-95813344
TMOD1         NM_003275        Hs.374849   tropomodulin 1                                                                        95742775-95743279
PCTAIRE2BP    AW129593         Hs.416543   tudor repeat associator with PCTAIRE 2                                                95629328-95638103
CDC14B        AI921238         Hs.22116    CDC14 cell division cycle 14 homolog B (S. cerevisiae)                                94656602-94657833
HABP4         AF241831         Hs.301839   hyaluronan binding protein 4                                                          94630336-94632501
SLC35D2       AJ005866         Hs.386278   solute carrier family 35, member D2                                                   94462851-94463372
FANCC         NM_000136        Hs.253236   Fanconi anemia, complementation group C                                               93202851-93203264
ZNF169        BC019228         Hs.387623   zinc finger protein 169                                                               92382458-92396707
PHF2          AB014562         Hs.93868    PHD finger protein 2                                                                  91782703-91783231
C9orf10       AF214738         Hs.446534   chromosome 9 open reading frame 10                                                    91634518-91635068
BICD2         AI934125         Hs.436939   coiled-coil protein BICD2                                                             90815152-90815659
IARS          NM_013417        Hs.172801   isoleucine-tRNA synthetase                                                            90314089-90326264
SPTLC1        BC007085         Hs.90458    serine palmitoyltransferase, long chain base subunit 1                                90183077-90183244
NFIL3         NM_005384        Hs.79334    nuclear factor, interleukin 3 regulated                                               89512814-89513311
AUH           NM_001698        Hs.81886    AU RNA binding protein/enoyl-Coenzyme A hydratase                                     89317590-89318101
SYK           BF593625         Hs.192182   spleen tyrosine kinase                                                                89001692-89002111
SBP2          BC001189         Hs.59804    SECIS binding protein 2                                                               87430383-87439113
CKS2          NM_001827        Hs.83758    CDC28 protein kinase regulatory subunit 2                                             87415644-87421033
HFSE-1        AF072164         Hs.137570   HFSE-1 protein                                                                        87258704-87259146
SPIN          AL136719         Hs.439052   spindlin                                                                              86550195-86550700
DAPK1         NM_004938        Hs.244318   death-associated protein kinase 1                                                     85780116-85780511




                          Stanford – South Africa                Biomedical Informatics Program
Public phenotype data on CEPH cell lines




                                                                      n
                                                               s oo
                                                    m ing
                                               Co


          Stanford – South Africa   Biomedical Informatics Program
                     Summary

• Complex diseases and their treatments
  require complex studies with variety of
  approaches
• As yet (October 2005) no one has
  published clear pharmacogenetic
  relationship of SNP to phenotype derived
  from whole genome approach



            Stanford – South Africa   Biomedical Informatics Program
               acknowledgments

• Howard McLeod, Pharm.D.
 Functional Polymorphism Analysis in Drug Pathways (CREATE)
 Washington University




                   Stanford – South Africa   Biomedical Informatics Program
                              References
•   Lynch TJ, Bell DW, Sordella R, Gurubhagavatula S, Okimoto RA, Brannigan BW,
    Harris PL, Haserlat SM, Supko JG, Haluska FG, Louis DN, Christiani DC, Settleman
    J, Haber DA. Activating mutations in the epidermal growth factor receptor
    underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med.
    2004 May 20;350(21):2129-39. Epub 2004 Apr 29. PMID: 15118073
•   Walgren RA, Meucci MA, McLeod HL.Pharmacogenomic Discovery Approaches: Will
    the Real Genes Please Stand Up? J Clin Oncol. 2005 Sep 6; PMID: 16145062
•   Cheok MH, Yang W, Pui CH, Downing JR, Cheng C, Naeve CW, Relling MV, Evans
    WE. Treatment-specific changes in gene expression discriminate in vivo drug
    response in human leukemia cells. Nat Genet. 2003 May;34(1):85-90. Erratum in:
    Nat Genet. 2003 Jun;34(2):231. PMID: 12704389
•   Watters JW, Kraja A, Meucci MA, Province MA, McLeod HL. Genome-wide
    discovery of loci influencing chemotherapy cytotoxicity. Proc Natl Acad Sci U S A.
    2004 Aug 10;101(32):11809-14. Epub 2004 Jul 28. PMID: 15282376




                         Stanford – South Africa   Biomedical Informatics Program

				
DOCUMENT INFO