Illumina, Inc. · NASDAQ:ILMN
Illumina — Technical Presentation (2014)
60 slides from Illumina's technical presentation, 2014
Slide 1: Kyoto University
Bt fb AK KYOTO UNIVERSITY
Slide 2: MDS and sAML
sAML MDS • • • • •
Slide 3: MDS and AML
MDS AML
Slide 4: Early mutation studies in myelodysplastic syndrome
A point mutation at codon 13 of the N-ras oncogene in myelodysplastic syndrome Hisamaru Hirai*, Yukio Kebayashi*, Hiroyuki Mano*, Koichi Hagiwara*, Yoshiro Maru*, Mitsuhiro Ominet, Hideaki Mizoguchit, Junji Nishida* & Fumimaro Takaku* * The Third Department of Internal Medicine, Faculty of Medicine,
Slide 5: CBL, TET2 and EZH2 mutations
CBL TET2 Delhommeau F, NEJM, 2009 EZH2 Sanada M, et al., Nature, 2009 Nikoloski G, et al., Nat Genet, 2010
Slide 6: Landscape of gene mutations in MDS 2010
Landscape of gene mutations in MDS 2010 MDS/MPN PIGA MDS EZH2/EED/SUZ12 c‐CBL AA/PNH ASXL1 PTPN11 DNMT3A NF1 RUNX1 TET2 ETV6 N/KRAS IDH1/2 TP53 cMPL FLT3 +8, ‐7/7q‐, ‐5/5q‐, NPM1 20q‐, 3q26, etc JAK2 RUNX1/RUNX1T1 CEBPα MPN CBFb/MYH11 c‐Kit PML/RARa AML MLL fusions Etc…
Slide 7: Genomic profile of cases with myelodysplasia
Genomic Profile of 222 cases with myelodysplasia TET2 RUNX1 RAS ASXL1 TP53 IDH DNMT3A EZH2 c-CBL p 1 cMPL, NRAS 1q+ q p 2 q p 3 q 4 p TET2 p q 5 5q- q Chromosomes 6 q p EZH2 7q- q p q p q p q p q p q p 7 8 +8 9 JAK2 10 11 WT1 cCBL 12 gain 13 FLT3 loss q 14 ?? CN-LOH q 15 qp q q p q p q p q p q p q 1
Slide 8: Moore's Law
Moore's Law National Human ll). Genome Research Institute Ql bi genome.gov/sequencingcosts ie} S (=) [=) o. o. fo} So (s} [=) oo. oO i=} oO is} = om A a $1,000,000
Slide 9: Sequencing platforms GAIIx and HiSeq 2000
2 ss 38 S-e-8e— H g $9896 8 53883 " go ee e f t f=) : Go a O = i e ° e e ° 5 Hen Gel @oe| @e| ®e| @e ° a gl a a i Y x= GAIIx
Slide 10: Types of genetic alterations
Point mutation Indel Homozygous Hemizygous Gain Translocation Exogenous deletion deletion pathogen
Slide 11: Slide 11
Slide 12: Whole exome capture and sequencing
Whole exome capture and sequencing biotinylated ‘bait’ cRNA gDNA Bind target sequences Avidin-coated magnet beads targets Recovery Non-targets Sequencing
Slide 13: Number of somatic mutations / sample
Number of somatic mutations / sample 20 ■Missense ■Nonsense ■Nonflameshift ■Frameshift ■Splice site 18 9.2 / sample 16 14 12 10 8 6 4 2 0 RA RARS RCMD del(5q) RAEB-1 RAEB-2 AML/MRC CMML1 CMML2 RCMD-RS MDS-U
Slide 14: Spliceosome gene mutations
Gene Gene 1 TET2 13 14 STAG2 3 Spliceosome 15 IDH1/2 =32/56 (57.0%) 2 SRSF2 9 2 3 U2AF1 7 16 PHF6 2 4 ZRSR2 7 17 BCOR 2 18 SETBP1 SRSF2 5 SF3B1 9 2 6 EZH2 5 19 HTR1A 2 7 ASXL1 4 20 LUC7L2 2 None U2AF1 8 NRAS 4 21 PCDHAC1 2 9 KRAS 4 22 TCF4 2 ZRSR2 10 CBL 4 23 CACNA1E 2 PRPF40B 11 DNMT3A 3 24 TTN 2 S
Slide 15: Splicing factor mutations by disease subtype
MDS without RS RARS / RCMD-RS CMML SF3B1 mutations (N = 155) (N = 73) (N = 88) SRSF2 (5.5%) ZRSR2 (1.4%) U2AF35 RARS (64~83%) U2AF35 (11.6%) . No Mut (15.1%) (8.0%) RCMD‐RS (57~76%) SRSF2 Dup (2.7%) (11.6%) SFSF2 (28.4%) RARS‐T (68~73%) ZRSR2 (7.7%) No Mut (45.5%) No Mut (56.1%) SF3B1 (6.5%) SF3B1 (
Slide 16: Frequency of mutations
Frequency of mutations 0% 10% 20% 30% 40% 50% 60% 70% SPLICING ASXL1 MDS JAK2 BCOR CBL EZH2 TET2 RUNX1 COHESIN PHF6 AML/MDS TP53 N/KRAS DNMT3A IDH1/2 PTPN11 KIT CEBPA WT1 MDS Gene mutations in MDS vs AML FLT3 AML NPM1 AML (TCGA) t(15;17) t(8;21) inv(16)
Slide 17: Splicing circle
“Splicing circle” (Wahl MC, et al. Cell 136:701, 2009)
Slide 18: Pathway mutations of splicing machinery
Pathway mutations of splicing machinery (N = 7 ) (N = 37 ) (N = 56 ) (N = 7 ) U2 complex U1 complex U2AF35 SRSF2 SF3A1 5’ A SF3B1 YYY AG ESE Exon 3’ Pre-mRNA U2AF65 ZRSR2 (N = 79 ) (N = 5 ) ESE: Exonic splicing enhancer SF1 (N = 23 ) (N = 5 ) RARS/RCMD-RS MDS without RS CMML AML/MDS de novo AML MPD
Slide 19: Reduced chimerism after transplants with U2AF35 mutants
Reduced chimerism after transplants with U2AF35mut Bone marrow CD34(−) KSL cells P < 0.001 30 P < 0.001 * P < 0.001 25 % G FP chim erism Retroviral tranduction of U2AF35 20 FACS sorting of * GFP(+) cells 15 10 Competitive trans- * not significant plantation to irr mice 5 Negative mock WT S34F Q157P
Slide 20: Landscape of mutations in exome sequencing of MDS cases
Landscape of mutations in exome seq in 56 MDS cases 30 20 10 0 SRSF2 1 1 1 1 1 1 1 1 1 U2AF1 1 1 1 1 1 1 1 ZRSR2 1 1 1 1 1 1 1 Splicing factors (~60%) SF3B1 1 1 1 1 1 1 1 1 1 1 SF3A1 1 PRPF40B 1 TET2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 IDH1/2 1 1 ASXL1 1 1 1 1 1 1 1 1 1 1 Epigenetic EZH2 1 1 1 1 1 1
Slide 21: SETBP1
SETBP1 500 1000 1500 1596 aa DSGIGTDNNSTSD SETBP1* AT hook I871T SET SKI homologous region Repeat domain SET binding domain PP2A D868 Mutation G870S N 170kD protein associated with SET protein, which are thought to inhibit PP2A tumor suppressor Translocated/overexpressed in leukemia Major down
Slide 22: SETBP1 mutations are associated with poor survival
SETBP1 mutations are associated with poor survival Whole cohort Whole cohort <60y.o. P<0.001 P<0.001 HR 2.27 HR 5.02 Overall Survival Overall Survival 95%CI 1.56‐3.21 Probability of Probability of 95%CI 2.36‐9.65 WT WT MT MT Months Months (Makishima, et al. Nature genet, 2013)
Slide 23: Leukemogenic potential of SETBP1 mutants
Leukemogenic potential of SETBP1 mutants ** 300 * Cell number (X105) Replating / immortalization assay (normal progenitors) 200 pMYs‐wild type Setbp1 pMYs‐Setbp1 (I871T) 100 Transduction pMYs‐Setbp1 (D868N) SCF pMYs 0 C57BL/6 TPO WT I871T D868N BM IGF‐2 SCF Infection FGF‐1 IL‐3 ** 6 days 4 days 2 da
Slide 24: Origin of gain-of-function
Origin of gain‐of‐function (Piazza R et al., Nat Genet, 2013)
Slide 25: SETBP1 escapes from proteasome degradation
SETBP1 escapes from proteosome degradation β‐TrCP1 SETBP1 E3 Ub Mock WT D868N I871T SETBP1‐HA 250kD β‐TrCP1 E3 150kD * mut Anti actin Proteasome inhibitor − + − + − + − + (Makishima, et al. Nature genet, 2013)
Slide 26: Kon A., et al., Nature Genetics, 2013
(Kon A., et al., Nature gent, 2013)
Slide 27: Cohesin mutations in myeloid neoplasms
Cohesin mutations in myeloid neoplasms Cohesion of sister chromatids Long‐range regulation of gene expression Post‐replicative DNA repair Responsible gene for congenital disorders (Cornelia de Lange syndrome and Roberts syndrome) (A Kon, et al. Nature genet, 2013)
Slide 28: Karyotypes of cohesin mutated cases
Karyotypes of cohesin mutated cases a # of numerical abnormalities Mutated or Noramal Karyotype 0 1 2 3 deleted (N = 58) cohesin status P=0.122 wild type (N = 279) 0% 20% 40% 60% 80% 100% b CMS Kasumi-1 MOLM-13 (A Kon, et al. Nature genet, 2013)
Slide 29: Cohesin for long-range regulation of gene expression
Cohesin for long‐range regulation of gene expression Phillips‐Cremins et al, Cell, 2013
Slide 30: BCOR/BCORL1 mutations in myeloid malignancies
BCOR/BCORL1 mutations in myeloid malignancies • Responsible for Oculofaciocardiodental (OFCD) syndrome • Involved in polycomb regulation • Mutated in 19/208 myeloid malignancies (9.1%) • 18/19(95%) were mutated in advanced disease types (RAEB,CMML,AML/MRC) • 12/19(63%) were inactivating mutations •
Slide 31: Spectrum of somatic mutation in MDS in 2014
Spectrum of somatic mutation in MDS in 2014 MDS Splicing factors AA/PNH RPS14 (SF3B1, SRSF2, U2AF1, ZRSR2, etc) MDS/MPN ETV6 EZH2/EED/SUZ12 SETBP1 RUNX1 ASXL1 c‐CBL DNMT3A PTPN11 PHF6 NF1 BCOR TET2 CUX1 RIT1 Cohesin IDH1/2 N/KRAS c‐Kit TP53 +8, ‐7/7q‐, ‐5/5q‐, cMPL FLT3 20q‐, 3q26, NPM1 LNK etc.. MP
Slide 32: Characteristics of patients with MDS
Characteristics of 944 patients with MDS Training Validation Parameter Total Cohort Cohort Cohort Patient Numbers 944 730 214 72.8 73.0 72.6 Median age (years) (23.3 – 90.8) (24.3–90.8) (23.3–90.4) Patients with Follow-up data 786 611 175 Median overall survival (months) 54.5 54.2 61.6 Median follow
Slide 33: Targeted deep sequencing and somatic mutation calling
Summary of targeted deep sequencing and somatic mutations calling 104 known or putative mutational genes captured using SureSelect HiSeq 2000 (100bp, paired-end)@mean depth of 1000x Align to the human genome reference (hg19) using BWA Discard low-quality reads and low-quality bases • Reads with 5 or
Slide 34: Copy number abnormalities in example cases
A 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 1819 20 2122 X -4q (TET2) B -7q (EZH2) C -5q (IRF1) -7 (LUC7L2, EZH2) D -5q (IRF1) -7q(LUC7L2, EZH2) -12p (ETV6) E +8q -3q (STAG1) -5q (IRF1) -12p (ETV6) -17 (TP53)
Slide 35: Distribution of allele frequencies
(SNVs) and short indels were identified. SRRM1 SETD2 KIT KDR FBXW7 CEBPA ASH1L TP53BP1 SMC1A Non‐missense PRPF8 NPM1 Missense NCOR1 LUC7L2 GNAS Distribution of allele frequencies FANCL BCORL1 PTPRD ARID1B RAD21 PTPN11 CTCF KDM6A ATRX SMC3 LAMB4 Genes NCOR2 KRAS A total of 2,764 single nucleotide var
Slide 36: Mutated genes in 864/944 cases
TET2 SF3B1 ASXL1 SRSF2 DNMT3A RUNX1 U2AF1 ZRSR2 STAG2 TP53 EZH2 CBL JAK2 BCOR IDH2 NRAS MPL NF1 ATM IDH1 KRAS PHF6 BRCC3 ETV6 LAMB4 NCOR2 SMC3 RAD21 CTCF PTPN11 FLT3 GNAS 864/944 cases (91.5%) FBXW7 SMC1A CEBPA NPM1 DCLRE1C KIT FANCL GATA2 GPRC5A U2AF2 IRF1 LUC7L2 GNB1 SF1 PIGA
Slide 37: Concordance between two large studies
Concordance between two large studies Top 12 Genes Top12 Genes in in our results Papaemmanuil et al TET2 SF3B1 SF3B1 TET2 ASXL1 SRSF2 SRSF2 ASXL1 DNMT3A DNMT3A RUNX1 RUNX1 U2AF1 U2AF1 ZRSR2 TP53 STAG2 EZH2 TP53 IDH2 EZH2 STAG2 CBL ZRSR2 N = 944 N = 738 104 genes tested 111 genes tested Torsten Hafer
Slide 38: Median of driver gene mutations per sample
Median of 3 driver gene mutations per semple 250 Observed 200 Expected Number of cases 150 100 50 0 0 1 2 3 4 5 6 7 8 9 10 11 12 Number of mutations
Slide 39: Number of gene mutations correlates with blast counts
Number of gene mutations correlate blast counts MDS subtypes Mean number of mutations 4 3 2 1 0 100% Number of mutations 80% More than 3 60% Frequency 3 40% 2 1 20% 0 0% Low risk High risk
Slide 40: Landscape of gene mutations in MDS (N=944)
Landscape of gene mutations in MDS (N=944)
Slide 41: Correlation among different mutations
Correlation among different mutations 2 3A 1 2 T 3B1 SF2 AF1 SR2 C7L PF8 T2 H1 H2 MT XL1 H2 V6 F1 NX M1 F6 OR 53 CC3TAGPL K2 P R AS AS L 1 MB4 SF SR U2 ZR LU PR TE ID ID DN AS EZ ET IR RU NP PH BC TP BR S M JA PT KR NR CB NF LA SF3B1 SRSF2 U2AF1 ZRSR2 LUC7L2 PRPF8 TET2 IDH1 q < 0.0001 IDH2 q < 0.001
Slide 42: Concordance of coexisting mutations between two studies
Concordance of coexisting mutations between two studies Haferlach et al, Leukemia 2013 Papaemmanuil et al, Blood 2013 (N= 944) (N= 738) TET2 ZRSR2 (q<0.0001), SRSF2 , CBL (q<0.001) SRSF2, ZRSR2 (q<0.01) SF3B1 DNMT3A, JAK2 (q<0.01) DNMT3A (q<0.01) STAG2, ASXL1, RUNX1, IDH2 (q<0.0001) STAG2, IDH2, CUX
Slide 43: Clonal evolution and relapse mechanisms
Mut C Mut A Mut A Mut B Mut C Mut D Mut A Mut B Mut A Mut A Mut D Mut B Mut E Mut E Mechanism of cancer development Relapse Resistance to anti‐tumor therapy
Slide 44: Copy number-adjusted variant allele frequencies
0.6 0.5 variant allele frequencies copy number-adjusted 0.4 0.3 0.2 0.1 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 Cases
Slide 45: Intratumor heterogeneity
Number of 1 2 >=3 Subclones 0.5 0.4 Valiant allele frequencies 0.3 0.2 0.1 0 Frequency Intratumor heterogeneity was evident in 456 cases. (48.3%)
Slide 46: Variations among variant allele frequencies
Variations among variant allele frequencies
Slide 47: Significant difference in allele frequency between common mutations
Significant difference in allele freq between common mutations A B C 0.6 0.6 0.6 U2AF1-ASXL1 TET2-ASXL1 SF3B1-JAK2 0.5 0.5 0.5 0.4 0.4 0.4 q = 6.2 x 10-5 q = 4.2 x 10-7 q = 8.8 x 10-5 ASXL1 ASXL1 JAK2 0.3 0.3 0.3 0.2 0.2 0.2 0.1 0.1 0.1 0 0 0 0 0.1 0.2 0.3 0.4 0.5 0.6 0 0.1 0.2 0.3 0.4 0.5 0.6 0 0.1
Slide 48: Gene pair allele frequency comparisons
Gene 2 SF3B1 SRSF2 U2AF1 ZRSR2 TET2 DNMT3A EZH2 ASXL1 RUNX1 ETV6 NPM1 BRCC3 STAG2 MPL JAK2 KRAS NRAS NF1 CBL eee -@-<e a) @ r o 2) 6 8 @ @ e@ e @ @ C5) ® e@ @e ®@ Sve MAF Mv He Ss LES EME TE SE & Gene © q<0.001 ® q<0.01 ® = q<0.1 Allele frequency Gene 1/Gene 2 -_ 6.0 1.0 Pathways Gl Splicing Ml DNA
Slide 49: Pathway pair comparisons
DNA methylation RAS pathway Transcription modification DNA repair Receptors/ Chromatin Cohesin Kinases Splicing Other Splicing DNA methylation Chromatin modification Transcription Pathway 1/Pathway 2 DNA repair q<0.001 Cohesin q<0.01 Receptors/ Kinases q<0.1 1.0 RAS pathway Other (Torsten, et al., u
Slide 50: Mutation rate per haploid genome per division
〜0.2‐4.5 per haploid genome/devision (Gundry and Vijg, 2012; Lynch, 2010).
Slide 51: Validated exome SNVs per HSPC clone
Validated exome SNVs _ _ Oo on ! J per HSPC clone ° of Clone: 123 123 123 123 123 123 123 Cord1Cord2 20s 30s 40s 50s “(70s (Welch, J et al. Cell, 2012)
Slide 52: Origin of mutations and MDS clones
Origin of mutations and MDS clones HSC EZH2/ASXL1 SETBP1 Splicing factors RUNX11, RAS pathway Progenitor cells TET2/IDH1/2 ETV6, etc.. JAK2/cMPL etc (M. Walter, et al., NEJM, 2012)
Slide 53: Loss of Y in older individuals: clonal hematopoiesis
‐Y is also observed in older individuals : clonal hematopoiesis
Slide 54: Hazard ratios of mutated/deleted genes in univariate analysis
Hazard ratios of mutated/deleted genes in univariate analysis Favorable Worse PTPN11 NPM1 Significant TP53 PRPF8 Not-Significant EZH2 LUC7L2 NRAS KRAS FLT3 RUNX1 NF1 LAMB4 ETV6 GATA2 ASXL1 SMC1A STAG2 CEBPA 25 out of 48 genes tested GPRC5A IRF1 significantly affected overall survival (P < 0.05) CBL
Slide 55: Clonal vs subclonal mutations and overall survival
Subclonal mutation Clonal mutation Not mutated SRSF2 RUNX1 ASXL1 Overall Survival P= 0.786 P= 0.351 P= 0.076 Clonal vs Subclonal Clonal vs Subclonal Clonal vs Subclonal (m) (m) (m) EZH2 TP53 CBL Overall Survival P= 0.224 P= 0.459 P= 0.403 Clonal vs Subclonal Clonal vs Subclonal Clonal vs Subclonal (
Slide 56: Hazard ratios in multivariate analysis
Hazard ratios in multivariate analysis (A) clinical and genetic parameters (Model-1) (B) Only genetic parameters (Model-2) HR HR Gender (Male) 1.3 ASXL1 1.7 Age 1.1 Hb (8-<10) 2.4 CBL 1.3 Clinical parameters Hb (<8) 4.2 1.4 ETV6 1.4 Plt (50-<100) Plt (<50) 2.2 1.3 EZH2 1.3 Blast score (>2-<5) 1.4 KR
Slide 57: Development of a novel prognostic risk classification
Development of a novel prognostic risk classification (Training cohort, N=611) (Validation cohort , N=175) Clinical and genetic parameters (Model-1) The number of very high risk cases 20 Cases Only genetic parameters 10 (Model-2) 0 IPSS-R Model-1 Model-2 Conventional model (IPSS-R)
Slide 58: Model-1 expected to outperform IPSS-R
Model-1 was expected to outperform the IPSS-R 1 Training Cohort (N=611) 2 Validation Cohort (N=175) 1.0 1.0 0.8 0.8 True positive rate True positive rate 0.6 0.6 0.4 3 years 0.4 3 years Model-1 0.821 Model-1 0.778 0.2 IPSS-R 0.742 0.2 IPSS-R 0.675 IPSS 0.688 IPSS 0.661 0 0 0 0.2 0.4 0.6 0.8 1.0 0 0.
Slide 59: Summary
Summary 1. Massively parallel sequencing technologies enabled identification of a full spectrum of gene mutations in MDS. 2. Major mutational targets in MDS included splicing factor genes (SF3B1, SRSF2, U2AF1 and ZRSR2), epigenetic regulators, and other novel gene targets, such as BCOR, SETBP1, PPRF
Slide 60: Acknowledgement
Acknowledgement Cleveland Clinic University of Tokyo Uniformed Services Taussig Cancer Institute Cancer Genomics Project University of the Health Hideki Makishima Kenichi Yoshida Sciences Kwok Peng Ng Yasunobu Nagata Yang Du Mikael A. Seaker Ayana Kon Nhu Nguyen Jaroslow P. Maciejewski Masashi Sanad