1. PRACTICE EVOLUTION: Decentralized Computer-Assisted IHC Image Analysis Liron Pantanowitz, MD, FCAP Director of Pathology Informatics Richard C. Friedberg, MD PhD, FCAP Chairman, Department of Pathology Baystate Health, Springfield, MA Tufts University School of Medicine

2. Why Are We Doing This? Practice BackgroundPractice Background Today’s EnvironmentToday’s Environment Increased technological innovation Increased technological innovation Increased biological information Increased biological information Increased clinical demand Increased clinical demand Convergence of two independent long term trendsConvergence of two independent long term trends

3. Key Trend #1 in the Practice of Anatomic Pathology Evolution along Clinical Pathology linesEvolution along Clinical Pathology lines Greater concern with analytical precision, reproducibility, accuracy, specificity, reliability Greater concern with analytical precision, reproducibility, accuracy, specificity, reliability Qualitative becoming quantitative Qualitative becoming quantitative “Stains” becoming “assays” “Stains” becoming “assays” Results directly tied to treatment, not just prognosis Results directly tied to treatment, not just prognosis Diminishing “guild” mentality with anointed experts Diminishing “guild” mentality with anointed experts ExamplesExamples IHC & ELISA IHC & ELISA Her2/neu & Herceptin Her2/neu & Herceptin

4. Key Trend #2 in the Practice of Anatomic Pathology Evolution along Radiology/Imaging linesEvolution along Radiology/Imaging lines Analog images establish the field Analog images establish the field Market & technology forces start trend to digital imaging Market & technology forces start trend to digital imaging Initially, scanning of analog imagesInitially, scanning of analog images Later, digitally acquired imagesLater, digitally acquired images Digitalization of images allows new applications Digitalization of images allows new applications Significant workload & throughput implications Significant workload & throughput implications ExamplesExamples PACS PACS Convergence imaging Convergence imaging Windowing Windowing Dynamic images Dynamic images Telediagnostics Telediagnostics

5. Expectations EventuallyEventually Every “image-based” pathologist will use computer-assisted analytic tools to assay specimens Every “image-based” pathologist will use computer-assisted analytic tools to assay specimens Intelligently designed PACS will revolutionize pathology workflow Intelligently designed PACS will revolutionize pathology workflow Increased reliance upon pathology Increased reliance upon pathology

6. Breast Cancer & Immunohistochemistry (IHC) Determining breast tumor markers (ER, PR & HER-2/neu) for prognostic & predictive purposes by IHC &/or FISH is the standard of practice. Determining breast tumor markers (ER, PR & HER-2/neu) for prognostic & predictive purposes by IHC &/or FISH is the standard of practice. IHC score/quantification by manual microscopy is currently accepted as the traditional gold standard. IHC score/quantification by manual microscopy is currently accepted as the traditional gold standard. Surgical Pathology workflow involves: Surgical Pathology workflow involves: Pre-analytic preparation (e.g. tissue fixation & processing)Pre-analytic preparation (e.g. tissue fixation & processing) Analysis (i.e. staining of controls & patient slides)Analysis (i.e. staining of controls & patient slides) Post-analytical component (e.g. quantification & reporting)Post-analytical component (e.g. quantification & reporting) Discrepancies between HER2 IHC & FISH mainly reflect errors in manual interpretation & not reagent limitations (Bloom & Harrington. AJCP 2004; 121:620-30). Discrepancies between HER2 IHC & FISH mainly reflect errors in manual interpretation & not reagent limitations (Bloom & Harrington. AJCP 2004; 121:620-30). Inter- & intra-observer differences in scoring occur: Inter- & intra-observer differences in scoring occur: Most notably with borderline & weakly stained casesMost notably with borderline & weakly stained cases Related to fatigue & subjectivity of human observersRelated to fatigue & subjectivity of human observers

7. Accuracy is Required Accuracy = the amount by which a measured value adheres to a standard. Accuracy = the amount by which a measured value adheres to a standard. The need for precise ER, PR & HER2/neu status in breast cancer is required to ensure appropriate therapeutic intervention. The need for precise ER, PR & HER2/neu status in breast cancer is required to ensure appropriate therapeutic intervention. Lay press have communicated concerns over inaccuracies in breast biomarker testing. Lay press have communicated concerns over inaccuracies in breast biomarker testing. Threat of having to refer such testing to reference laboratories. Threat of having to refer such testing to reference laboratories. Is computer assisted image analysis (CAIA) a better (i.e. more accurate & reproducible) method for scoring IHC? Is computer assisted image analysis (CAIA) a better (i.e. more accurate & reproducible) method for scoring IHC?

9. Guidelines ASCO/CAP Guideline Recommendations for HER2/neu testing in breast cancer (Wolff et al. Arch Pathol Lab Med 2007; 131:18) ASCO/CAP Guideline Recommendations for HER2/neu testing in breast cancer (Wolff et al. Arch Pathol Lab Med 2007; 131:18) Image analysis can be an effective tool for achieving consistent interpretationImage analysis can be an effective tool for achieving consistent interpretation A pathologist must confirm the image analysis resultA pathologist must confirm the image analysis result Image analysis equipment (including optical microscopes) must be calibrated, subjected to regular maintenance & internal QC evaluationImage analysis equipment (including optical microscopes) must be calibrated, subjected to regular maintenance & internal QC evaluation Image analysis procedures must be validatedImage analysis procedures must be validated Canadian National Consensus Meeting on HER2/neu testing in breast cancer (Hanna et al. Current Oncology 2007; 14:149-53) Canadian National Consensus Meeting on HER2/neu testing in breast cancer (Hanna et al. Current Oncology 2007; 14:149-53) Use of image analysis systems can be useful to enhance reproducibility of scoringUse of image analysis systems can be useful to enhance reproducibility of scoring Pathologists must supervise all image analysesPathologists must supervise all image analyses FDA clearance for CAIA in vitro diagnostic use of HER-2/neu, ER, and PR IHC has been obtained by several companies FDA clearance for CAIA in vitro diagnostic use of HER-2/neu, ER, and PR IHC has been obtained by several companies

10. CAIA vs. Manual Score Remmele & Schicketanz. Pathol Res Pract 1993; 189:862-6 “Subjective grading of slides is a simple, rapid and useful method for the determination of tissue receptor content and must not be replaced by expensive and time-consuming computer-assisted image analysis in daily practice.” “Subjective grading of slides is a simple, rapid and useful method for the determination of tissue receptor content and must not be replaced by expensive and time-consuming computer-assisted image analysis in daily practice.”

11. Data on CAIA & IHC Early studies showed CAIA was no better than visual analysis Early studies showed CAIA was no better than visual analysis (Schultz et al. Anal Quant Cytol Histol 1992; 14:35-40 ) Few studies have shown that manual & CAIA are comparable Few studies have shown that manual & CAIA are comparable (Diaz et al. Ann Diagn Pathol 2004; 8:23-7) Most studies found CAIA to be superior to manual methods Most studies found CAIA to be superior to manual methods (Taylor & Levenson. Histopathology 2006; 49::411-24; McClelland et al. Cancer Res 1990; 50:3545-50; Kohlberger et al. Anticancer Res 1999; 19:2189-93; Wang et al. Am J Clin Pathol 2001; 116:495-503; Turner et al. USCAP 2008 abstract 1694). Provides effective qualitative & quantitative evaluationProvides effective qualitative & quantitative evaluation More consistent than manual & digital microscopyMore consistent than manual & digital microscopy More precise (scan per scan) than pathologistsMore precise (scan per scan) than pathologists One study showed agreement between different CAIA systems: Chroma Vision ACIS & Applied Imaging Ariol SL-50 One study showed agreement between different CAIA systems: Chroma Vision ACIS & Applied Imaging Ariol SL-50 (Gokhale et al. Appl Immunohistochem Mol Morphol 2007; 15:451-5)

12. Published Considerations Expense of CAIA may be hard to justify where volumes are low Expense of CAIA may be hard to justify where volumes are low Image analysis frequently requires interactive input by the pathologist Image analysis frequently requires interactive input by the pathologist Increased time requirements Increased time requirements Systems may be discrepant when tumor cells have low levels of staining Systems may be discrepant when tumor cells have low levels of staining Interfering non-specific staining within selected areas Interfering non-specific staining within selected areas Images must be free from artifacts Images must be free from artifacts Small amounts of stained tissue can erroneously generate lower scores Small amounts of stained tissue can erroneously generate lower scores

13. CAIA Systems ImageJ (NIH developed freeware)ImageJ (NIH developed freeware) Adobe Photoshop softwareAdobe Photoshop software (Lehr et al J Histochem Cytochem 1997; 45:1559-65) Automated Cellular Imaging System (Chroma Vision)Automated Cellular Imaging System (Chroma Vision) Pathiam (BioImagene)Pathiam (BioImagene) Applied Imaging Ariol (Gentix Systems)Applied Imaging Ariol (Gentix Systems) Spectrum (Aperio)Spectrum (Aperio)

14. Image Analysis & Algorithms Object-Oriented Image Analysis (morphology- based) Object-Oriented Image Analysis (morphology- based) Involves color normalization, background extraction, segmentation, classification & feature selection Involves color normalization, background extraction, segmentation, classification & feature selection Separation of tissue elements (e.g. tumor epithelium) from background (e.g. stroma) permits selection of areas of interest & filtering out of unwanted areas Separation of tissue elements (e.g. tumor epithelium) from background (e.g. stroma) permits selection of areas of interest & filtering out of unwanted areas Region of Interest (ROI) is subject to further image analysis (computation of diagnostic score) Region of Interest (ROI) is subject to further image analysis (computation of diagnostic score) Quantification of results Quantification of results

15. Digital Algorithm Courtesy of BioImagene

18. Validation & Implementation at Baystate Health Distant medical centers Distant medical centers Significant breast IHC caseload Significant breast IHC caseload Need to mimic daily practice Need to mimic daily practice avoid central (single user) image analysisavoid central (single user) image analysis Bandwidth limitations Bandwidth limitations Whole slide imager availability Whole slide imager availability Professional reluctance to read digital images Professional reluctance to read digital images

19. Key Components Multimedia PC upgrade Multimedia PC upgrade Spot Diagnostic digital cameras for each workstation Spot Diagnostic digital cameras for each workstation Pathiam (BioImagene) web-based application Pathiam (BioImagene) web-based application Server (Oracle database + application + image file storage) Server (Oracle database + application + image file storage) Training & Validation Training & Validation

20. WORKFLOW CONTROL IHC PATIENT IHC FOV ANALYSIS REPORT GENERATION

21. NEED FOR STANDARDIZATION

22. Calibrated Workstations

23. FOV IHC Analysis FFPE breast cases routinely stained for ER, PR & HER2-neu FFPE breast cases routinely stained for ER, PR & HER2-neu Standardized camera acquisition settings (calibration) Standardized camera acquisition settings (calibration) Pathologists (n=3) acquired 3-5 FOVs (each at 20x Mag.) Pathologists (n=3) acquired 3-5 FOVs (each at 20x Mag.) Uniform jpg image file formats used (4 Mb) Uniform jpg image file formats used (4 Mb) Post-processing image manipulation was avoided Post-processing image manipulation was avoided Control parameter set defined/IHC run (default/modified) Control parameter set defined/IHC run (default/modified) ER/PR nuclear staining analyzed using the Allred scoring system (i.e. proportion + intensity score = TS) ER/PR nuclear staining analyzed using the Allred scoring system (i.e. proportion + intensity score = TS) HER-2/neu membranous staining evaluated per ASCO/CAP 2007 recommendations (0, 1+, 2+, 3+) HER-2/neu membranous staining evaluated per ASCO/CAP 2007 recommendations (0, 1+, 2+, 3+) Manual vs. CAIA comparison tracked (IHC score, time & problems) Manual vs. CAIA comparison tracked (IHC score, time & problems) FISH for HER2/neu obtained on several cases FISH for HER2/neu obtained on several cases

24. ER/PR Correlation (N=29) Bio- marker ConcordantCases Discordant Cases ER+160 ER - 42* PR+140 PR - 43* * 3 cases

25. HER-2/Neu Results (N=28) Score0/1+2+3+ 0/1+161* 2+31** 3+4 CAIA Manual Scoring FISH RESULTS: * Negative (Ratio 1.04) ** Abnormal (Ratio 6.5)

26. HER-2/Neu FISH Correlation Manual Score CAIA Score FISH Result 00 Negative (1.06) 00 Negative (0.93) 11 Negative (1.04) 10 Negative (1.00) 10 Negative (1.07) 10 Negative (1.66) 21 Negative (1.04) 32 Abnormal (6.5)

27. Challenging Cases Infiltrating Lobular Carcinoma Cytoplasmic Staining Infiltrating Lobular Carcinoma Cytoplasmic Staining

28. Lessons Learned Decentralized CAIA for IHC designed to mimic daily surgical pathology workflow in practice is feasible Decentralized CAIA for IHC designed to mimic daily surgical pathology workflow in practice is feasible Image acquisition requires standardization Image acquisition requires standardization Tissue heterogeneity may impact FOV selection (whether biological or due to IHC variation) Tissue heterogeneity may impact FOV selection (whether biological or due to IHC variation) Pathologists must supervise CAIA systems Pathologists must supervise CAIA systems

29. Future Prospects Adopt virtual workflow-centric systems feasible for routine practice (that may potentially show better results) Adopt virtual workflow-centric systems feasible for routine practice (that may potentially show better results) E.g. Whole slide imaging (WSI) to eliminate the need to standardize different systemsE.g. Whole slide imaging (WSI) to eliminate the need to standardize different systems Automatic ROI selection & image analysis Automatic ROI selection & image analysis Shortened analysis time Shortened analysis time AP-LIS & CAIA system integration AP-LIS & CAIA system integration To improve workflowTo improve workflow Permit disparate systems to access the same digital images & case dataPermit disparate systems to access the same digital images & case data Learning algorithms Learning algorithms Systems that improve with experience following pathologist feedbackSystems that improve with experience following pathologist feedback Clinical outcome studies are needed Clinical outcome studies are needed In one study, CAIA for ER IHC yielded results that did not differ from human scoring against patient outcome gold standards (Turbin et al. Breast Cancer Res Treat 2007)In one study, CAIA for ER IHC yielded results that did not differ from human scoring against patient outcome gold standards (Turbin et al. Breast Cancer Res Treat 2007)

30. Acknowledgements Christopher N. Otis, MD Giovana M. Crisi, MD Andrew Ellithorpe, MHS Peter Marquis, BA BioImagene

32. TRANSFORMING PATHOLOGY: Emerging technology driving practice innovation