IDH1 R132H/ATRX Immunohistochemical validation CIQC/DSM 2016 12 June 2016 0835-0905 Stephen Yip, M.D., Ph.D., FRCPC University of British Columbia
Disclosure Statement I have nothing to disclose I will not discuss off label use and/or investigational use in my presentation.
Objectives To appreciate the algorithm for molecular classification of low- grade glioma Understand the central role of identification of IDH mutation and ATRX loss in LGG classification To describe ciqc effort of ensuring consistent implementation and interpretation of biomarker testing in neuro-oncology
2007
IDH1/2 testing Clinical behaviour Better prognosis? Identification, validation of novel marker IHC, SNaPshot, MLPA testing Sequencing, HRM, IHC Histology #1 Molecular genetics Cytogenetics IDH1 R132H mutation in 2 GBM
Parsons DW, Jones S, Zhang X, et al: An Integrated Genomic Analysis of Human Glioblastoma Multiforme. Science, 2008 Yan H, Parsons DW, Jin G, et al: IDH1 and IDH2 mutations in gliomas. N Engl J Med 360:765-73, 2009
IDH1 R132H mutation in malignant glioma Argument for more unbiased genomic scanning of cancers Yan H, Parsons DW, Jin G, et al: IDH1 and IDH2 mutations in gliomas. N Engl J Med 360:765-73, 2009
loosterhof NK, Bralten LB, Dubbink HJ, et al: Isocitrate dehydrogenase-1 mutations: a fundamentally new understanding of diffuse glioma? Lancet Oncol 12:83-91, 2010
IDH1/2 mutations in gliomas IHC test available for the most common mutation R132H Yan H, Parsons DW, Jin G, et al: IDH1 and IDH2 mutations in gliomas. N Engl J Med 360:765-73, 2009
IDH1 R132H Tumour exome IDH1 R132H ACG->ATG Matched normal H09 Mutation- specific antibody Genotyping assay Sequencing- based assay
Anaplastic astrocytoma (WHO III) IDH1 R132H POSITIVE
Allelic discrimination assays for IDH1/2 hotspot mutations p.r132c Intrahepatic cholangioca, AML, melanoma p.r132s Jessica Que & Julie Ho - VGH
Cancer Genome Atlas Research, et al. (2015). Comprehensive, Integrative Genomic Analysis of Diffuse Lower-Grade Gliomas. N Engl J Med, 372(26), 2481-2498. doi: 10.1056/NEJMoa1402121
Molecular segregation of low grade gliomas 1p19q ROH, IDH wildtype Behaves like GBM 1p19q ROH aka astrocytomas 1p19q LOH aka ODG IDH1 R132H IDH1 R132H IDH1 R132H EGFR ATRX ATRX ATRX
OLIGODENDROGLIOMA H&E ATRX 1p19q Co-deleted IDH1 R132H 1p19q FISH
1p19q ATRX_20X retained ATRX_40X ATRX loss in tumour cells with preservation of expression in entrapped neurons Matija Snuderl NYU Craig Horbinski U Kentucky
2016 BRAF V600E
TMA VGH low grade and high grade gliomas as well as non-neoplastic lesions IDH1 R132H and ATRX IHC status identified IDH1/2 hotspot sequencing performed 1p19q FISH performed on LGG and GBM samples
IDH1 R132H ciqc challenge Lab ID Total n % Scorable Pairwise complete observations Run 1 Concordance with reference (%) Sensitivity Specificity PPV (positive predictive value) NPV (negative predictive value) Cohen's kappa 101 22 100 22 19/22 (86%) 0.73 1 1 0.79 0.73 102 22 90.91 20 20/20 (100%) 1 1 1 1 1 103 22 100 22 22/22 (100%) 1 1 1 1 1 107 22 100 22 22/22 (100%) 1 1 1 1 1 110 22 95.45 21 14/21 (67%) 0.45 0.9 0.83 0.6 0.35 111 22 100 22 19/22 (86%) 0.73 1 1 0.79 0.73 112 22 100 22 21/22 (95%) 1 0.91 0.92 1 0.91 114 22 100 22 22/22 (100%) 1 1 1 1 1 123 22 100 22 22/22 (100%) 1 1 1 1 1 125 22 95.45 21 21/21 (100%) 1 1 1 1 1 126 22 100 22 22/22 (100%) 1 1 1 1 1 149 22 100 22 22/22 (100%) 1 1 1 1 1 162 22 100 22 22/22 (100%) 1 1 1 1 1 175 22 95.45 21 21/21 (100%) 1 1 1 1 1 191 22 100 22 22/22 (100%) 1 1 1 1 1 202 22 100 22 19/22 (86%) 1 0.73 0.79 1 0.73 Sanger sequencing verified
Lab ID Clone Dilution Supplier Ag Retrieval Ab Incubation Time Detection Enhancement Chromogen 101 H09 1:100 Dianova CC1 32 min Ventana Optiview Copper DAB 102 H09 1/250 Dianova DAKO 3in1 High ph 30" DAKO FLEX+ Copper Sulphate DAKO DAB+ 103 IDH R132H 1/50 DIANOVA CC1 64, 32 ULTRA VIEW DAB COPPER DAB 107 H09 1:20 Dianova Ventana CC1 32min 32min Ventana Optiview DAB none DAB 110 H09 1:50 Dianova High ph (ph 9) 30 min Dako Flex Mouse linker DAB CC1-36 111 H09 1/50 Histobiotec minutes 32 minutes Ultraview ---- Dab 112 H09 1:100 Dianova CC1 30 min 40 min iview copper DAB 114 H09 1/200 Dianova CC1 32min 16min Optiview-Ventana Copper DAB 123 H09 1:50 Cedarlane CC1 standard 60 min UltraView none DAB Bond Polymer Refine Detection Bond DAB enhancer 125 111219/18 1/2000 dianova ER1-30 (bond) 15 min DAB 1:500 = 126 H09 0.4ug/ml OPTISTAIN ph9.0 Tris-EDTA 30 minutes Quanto polymer None Dako DAB+ DIA HO9 149 M 1:25 Dianova PT Link, high ph 30 min Envision Flex Yes DAB Ventana CC1 48 Ventana 162 H09 1:80 Dianova min. 32 min, 36 C Ventana OptiView - OptiView DAB 175 H09 1:200 Dianova CC1 32 min Optiview None DAB 191 H09 1/10 Dianova CC1 32' ultraview none DAB 202 H09 1/50 RUN1 HISTOBIOT ECH HIGH PH 16 MIN LEICA REFINE DETECTION KIT NONE DAB
IDH1 R132H ciqc challenge (runs 35 & 44) Lab ID Run 1 Run 2 Concordance with sequencing Cohen's kappa Lab ID Concordance with sequencing Cohen's kappa 101 86% 0.73 101 100% 1 102 100% 1 102 100% 1 103 100% 1 103 100% 1 107 100% 1 107 100% 1 110 67% 0.35 110 100% 1 111 86% 0.73 111 87% 0.74 112 95% 0.91 112 100% 1 114 100% 1 114 100% 1 123 100% 1 123 100% 1 125 100% 1 125 100% 1 126 100% 1 126 100% 1 149 100% 1 149 96% 0.93 162 100% 1 162 96% 0.92 175 100% 1 175 100% 1 191 100% 1 191 96% 0.92 202 86% 0.73 202 100% 1 - - - 215 100% 1 - - - 217 89% 0.78
IDH1 R132H ciqc challenge (runs 35 & 44) Lab ID Run 1 Run 2 Concordance with sequencing Cohen's kappa Lab ID Concordance with sequencing Cohen's kappa 101 86% 0.73 101 100% 1 102 100% 1 102 100% 1 103 100% 1 103 100% 1 107 100% 1 107 100% 1 110 67% 0.35 110 100% 1 111 86% 0.73 111 87% 0.74 112 95% 0.91 112 100% 1 114 100% 1 114 100% 1 123 100% 1 123 100% 1 125 100% 1 125 100% 1 126 100% 1 126 100% 1 149 100% 1 149 96% 0.93 162 100% 1 162 96% 0.92 175 100% 1 175 100% 1 191 100% 1 191 96% 0.92 202 86% 0.73 202 100% 1 - - - 215 100% 1 - - - 217 89% 0.78
IDH1 R132H ciqc challenge (runs 35 & 44) Lab ID Run 1 Run 2 Concordance with sequencing Cohen's kappa Lab ID Concordance with sequencing Cohen's kappa 101 86% 0.73 101 100% 1 102 100% 1 102 100% 1 103 100% 1 103 100% 1 107 100% 1 107 100% 1 110 67% 0.35 110 100% 1 111 86% 0.73 111 87% 0.74 112 95% 0.91 112 100% 1 - In Run 1, Lab 101 possessed weak staining that led to false negative results. They reduced their dilution from 1:100 to 1:50 after ciqc feedback, and achieved 100% concordance with sequencing for Run 2. 114 100% 1 114 100% 1 123 100% 1 123 100% 1 125 100% 1 125 100% 1 126 100% 1 126 100% 1 149 100% 1 149 96% 0.93 162 100% 1 162 96% 0.92 175 100% 1 175 100% 1 191 100% 1 191 96% 0.92 202 86% 0.73 202 100% 1 - - - 215 100% 1 - - - 217 89% 0.78 - In Run 1, Lab 110 had significant background staining and decreased overall sensitivity that led to many discordant results. After complete protocol re-optimization based on another ciqc participant s successful staining protocol, Lab 110 achieved 100% concordance with sequencing for Run 2. - In Run 1, Lab 202 had significant background staining that led to false positive results. They increased their dilution from 1:50 to 1:100 and reduced antibody incubation time after ciqc feedback, and achieved 100% concordance with sequencing for Run 2.
Allelic discrimination assays for IDH1/2 hotspot mutations p.r132c Intrahepatic cholangioca, AML, melanoma p.r132s Jessica Que & Julie Ho - VGH
IDH1/2 Testing Indication for IDH1 R132H IHC testing ANY glioma Differentiate between reactive gliosis, non- infiltrative vs diffuse glioma i.e. confirm diagnosis IDH1 R132H IHC robust reactivity in cytoplasm Indication for IDH1/2 genotyping If Immunonegative Any WHO II/III diffuse glioma Not GBM >55 year old IDH1/2 Sanger sequencing, genotyping, NGS
ATRX
CORE14- Anaplastic ODG (WHO III), IDH1 R132H, 1p19q co-deleted 102 110 113 123 125 149 175 215 All centres agreed on POSITIVE reactivity 217 112
CORE13- Oligoastrocytoma (WHO II), IDH1 R132H 102 110 113 123 125 149 175 215 All centres agreed on NEGATIVE reactivity 217 112
Practice pattern of 1p19q FISH ordering - VGH
Molecular prognostic marker Genotype - IDH mutated ATRX intact IDH mutated ATRX loss IDH mutated ATRX loss IDH wildtype ATRX intact Molecular Dx- Anaplastic ODG Anaplastic Astrocytoma Anaplastic Astrocytoma GBM- like
Conclusion New molecular insights into LGG have altered the practice of clinical neuropathology Molecular segregation based on 1p19q/IDH/ATRX status is superior than histology Multi- institution QA/QC initiatives such as ciqc will ensure consistent and accurate implementation and interpretation of above assays
Acknowledgements VGH Neuropathology Ian MacKenzie, John Maguire, Wayne Moore Max Signaevski ciqc Jennifer Won, John Garratt, Blake Gilks Staff of VGH Immunohistochemistry core Participants of IDH and ATRX challenges
Six participating laboratories 100% concordance of IDH1 R132H IHC calls 2/6 labs had initial discordance in HRM/sequencing calls IHC vs Sequencing/Genotyping Sequencing backup 44/50 cases show concordance between IHC and Sanger sequencing (DNA from frozen tumour) All 6 cases were immunopositive for IDH1 R132H
Correlation of IDH1 R132H/ATRX IHC with 1p19q status FINAL DIAGNOSIS IDH1 R132H STATUS WHO Grading (Present/Not present) ATRX IHC status (Present/Not Present) 1P/1Q 19Q/19P ODG II P N/A 0.71 0.65 ODG II N N/A 0.67 0.7 ODG II P N 0.72 0.64 ODG II P N 0.61 0.63 ODG II P pending 0.67 0.68 ODG II P N/A 0.6 0.54 ANAPLASTIC ODG III P N/A 0.57 0.6 ODG II P P 0.6 0.59 Anaplastic oligoastrocytoma III N P 0.76 0.71 Oligoastrocytoma II P P 0.73 0.67 ODG II P P 0.66 0.61 ODG II P N 0.6 0.58 Anaplastic ODG III P P 0.57 0.6 Anaplastic ODG III P P 0.55 0.52 ANAPLASTIC ODG III P P 0.66 0.55 ODG II P N 0.72 0.68 ODG II P P 0.62 0.59 ODG II N P 0.74 0.73 ODG II P N/A 0.55 0.57 ODG II P P 0.71 0.65 ODG II P N/A 0.62 0.59 ODG II P P 0.69 0.64 ODG II P P 0.79 0.8