On May 4 and 5, 2002, the College of American Pathologists

Transcription

1 College of American Pathologists Conference Conference Summary, Strategic Science Symposium Her-2/neu Testing of Breast Cancer Patients in Clinical Practice Richard J. Zarbo, MD, DMD; M. Elizabeth H. Hammond, MD Context. Practicing pathologists often encounter controversial clinical issues and nonstandardized laboratory approaches to the evolving science of predictive/prognostic tumor marker assays. This dilemma becomes especially acute when the assay is the sole determinant for selection of a specific therapy. Objectives. To summarize the areas of practical agreement and identify opportunities for improvement in Her- 2/neu testing of breast cancer. Design. The College of American Pathologists created a new comprehensive education model, called Strategic Science, with expert speakers integrating new and evolving basic, clinical, and scientific issues of Her-2/neu testing with aspects of laboratory management. Setting. Symposium held May 4 and 5, 2002, in Rosemont, Ill. Participants. Ten speakers and more than 100 attendees. Results. Components addressed were new technology assessment, practice guidelines, quality assurance, regulatory compliance, risk and liability, billing and coding, cost analysis, consultation, information management, and results reporting. Conclusions. This Strategic Science symposium derived areas of practical agreement, defined the current state-ofthe-art, and identified areas for improvement in Her-2/neu testing. (Arch Pathol Lab Med. 2003;127: ) On May 4 and 5, 2002, the College of American Pathologists (CAP) held the first Strategic Science symposium on Her-2/neu testing of breast cancer patients in clinical practice. Strategic Science is a new multimodal continuing education series integrating science with numerous aspects of laboratory management. This new education model, in which expert speakers address new or evolving scientific issues from both the clinicians and pathologists perspectives, is offered in 4 continuing medical education credit-generating formats. These include an online preconference, a 1½-day live symposium, an enduring online postconference, and peer-reviewed publication. Strategic Science is designed for the practicing pathologist and comprehensively addresses pertinent aspects of new technology assessment, practice guidelines, quality assurance, regulatory compliance, risk and liability, billing and coding, cost analysis, consultation, utilization and information management, and results reporting. Although not strictly designed as a consensus conference, this Strategic Science symposium derived areas of practical agreement, defined the current state-of-the-art, and pointed out opportunities for improvement in Her-2/ neu testing. The status of Her-2/neu testing in breast cancer was addressed in the June 1999 CAP Consensus Conference XXXV, entitled Solid Tumor Prognostic Factors Accepted for publication November 15, From the Department of Pathology, Henry Ford Hospital and Medical Group, Detroit, Mich (Dr Zarbo); and the Department of Pathology, Urban Central Region Hospitals, Intermountain Health Care, Salt Lake City, Utah (Dr Hammond). Reprints: Richard J. Zarbo MD, DMD, Henry Ford Hospital, Department of Pathology, 2799 W Grand Blvd, Detroit, MI ( rzarbo1@hfhs.org). Which, How and So What. At that time, Her-2/neu was classified as a promising (category II) test in need of continued validation as a prognostic/predictive marker. 1 Since that time, Her-2/neu testing has assumed a major predictive role in qualifying patients for trastuzumab (Herceptin) therapy. Although commonly used to type newly diagnosed breast cancers, currently the justification for performing this assay is limited to patients with advanced or metastatic breast cancers who might be candidates for trastuzumab therapy or who are enrolled in clinical trials testing the potential efficacy of adjuvant trastuzumab. We believe that implementation of the recommendations from this CAP Strategic Science symposium will secure the position of Her-2/neu testing in breast cancer as an accepted clinical test (category I test), defined as one well supported by the literature and generally used in patient management. What follows is our summary of the key points presented at the symposium. CLINICAL VALUE OF HER-2/neu AS A PROGNOSTIC AND PREDICTIVE FACTOR IN BREAST CANCER It is readily accepted that as a pure prognostic factor, Her-2/neu protein overexpression or gene amplification confers a worse prognosis in both disease-free status and overall survival for both node-negative and node-positive disease. As a factor predictive of therapeutic response, patients whose tumors have an overexpressed/amplified state of Her-2/neu obtain more benefit from trastuzumab, equal or more benefit from treatment with anthracyclines, equivocal benefit from taxanes, and less benefit from nonanthracycline chemotherapy and hormonal therapy. 2 Hammond and Taube 3 recently provided an in-depth re- Arch Pathol Lab Med Vol 127, May 2003 Conference Summary Zarbo & Hammond 549

2 view discussing the process of tumor marker validation and clinical utility assessment. DIRECTING DRUG THERAPY: STANDARDIZATION OF STAND-ALONE LABORATORY TESTS In comparison to the adjunctive use of diagnostic immunohistochemistry (IHC) to confirm a histologic diagnosis, the unique stand-alone nature of Her-2/neu IHC and fluorescence in situ hybridization (FISH) testing to select patients for therapy with trastuzumab (Herceptin) was repeatedly emphasized throughout the conference. The Her-2/neu test(s) stands alone to select patients for trastuzumab therapy, once the sample being tested is histologically confirmed to contain the invasive tumor. This fact subjects these tests to more rigorous quantitation and regulation, as well as to more rigorous training and competency assessment of technologists and pathologists to ensure this consistency. Both IHC and FISH Her-2/neu testing for directing drug therapy choices require rigorous adherence to protocol, which includes standardized protocols, test validation, use of appropriate controls, documentation, and competency assessment. 4,5 Many preanalytic factors strongly influence both assays, mandating that standardized protocols should be adopted that address the preanalytic factors that can alter results. 1. Use of fixatives other than 10% neutral buffered formalin will alter results. Both the Food and Drug Administration (FDA) approved IHC and FISH assays for Her- 2/neu are approved for use with formalin-fixed tissues only Length of formalin fixation and delay in fixation may also impact results; optimal formalin fixation time is 6 to 12 hours. Specimens should be fixed promptly and should not be allowed to sit for prolonged periods of time at room temperature Controls (cell lines) fixed exactly as the test sample with varying surface protein expression must be used to calibrate the assay with each episode of testing Optimal block selection should include selecting well-fixed areas of tumor and benign breast tissue without artifacts and avoidance of decalcified material. 5. Although a false-negative IHC test result would be rare with epitope retrieval, if antigen lability in the block is suspected, then a negative IHC test result could be verified as negative by evaluating ubiquitous antigen preservation using immunostains for such housekeeping antigens as vimentin, cytokeratin, estrogen, or progesterone receptor. 6. The higher interobserver variation attributed to IHC interpretation may be reduced by better training, use of improved interpretation guidelines (as refined by Ken Bloom, MD; Figure), 11 or quantitation by image analysis. 12 Typically, the highest concordance of IHC with Her-2/neu FISH assays will be seen with scores of 0 and Verification procedures should be used to ensure that invasive tumor is being assessed in both IHC and FISH assays, and that artifacts are not being misread as positivity. 13 A scientific analysis of FISH for Her-2/neu gene amplification in the selection of patients for Herceptin therapy leads to additional key quality control considerations. Fluorescence in situ hybridization appears to be less affected by preanalytic factors, because DNA appears to be more robust than protein in fixed specimens. Therefore, The basic Food and Drug Administration approved scoring criteria for the Dako HercepTest are recommended. 11 Illustrated in this algorithm are additional light microscopic features of chicken-wire or complete membrane staining and uniform brown staining that indicate a positive result of a 2 or 3 score. 11 FISH appears to be an optimal method, not only for archived blocks, but also for touch imprints and fine-needle aspirations if fixed in formalin. However, enzymatic digestion failures related to tissue handling and fixation conditions may account for a FISH assay failure rate up to 10%. 14 Excessive enzymatic digestion may result in no signal, whereas too little enzyme digestion may result in tissue autofluorescence, precluding accurate signal evaluation. Although FISH interpretation with actual signal counts may be less subjective than IHC scoring, subjectivity arises from signal heterogeneity, problematic microscopic identification of invasive versus in situ components with UV light, and difficulty with interpretation of borderline amplified cases. For these reasons, it is recommended that equivocal or uninterpretable FISH cases be repeated or reevaluated by IHC. CONSENSUS TESTING ALGORITHM FOR HER-2/neu TESTING Based on evaluation of the current state-of-the-art and the fact that IHC and FISH are confirmatory of each other, we have derived the following consensus on preferred test methods from the symposium. To select the optimum testing procedure in an individual laboratory, performing laboratories should confirm their concordance rates of FISH/ IHC for IHC scores of 3 and 0, or consider doing both tests in all breast cancer cases. If the concordance rates are greater than 90% for 3 /FISH amplified and for 0/FISH nonamplified, the laboratory can select IHC as a screening tool. In such a setting, all 1 and 2 positive cases should be confirmed by FISH. Laboratories with a known concordance rate for 1 positive cases with FISH of 95% may not feel that it is necessary to do this confirmation, but should document their concordance rates in a report comment, as is suggested in Table 1. Chromogenic in situ hybridization (CISH) or gold-enhanced autometallographic ISH (GOLDFISH) are promising technologies for Her-2/neu detection that are currently in development, but they are still in need of standardized protocols and validation BILLING AND CODING FOR HER-2/neu TESTING There is no controversy regarding billing and coding for both tests. An IHC test is coded as CPT (im- 550 Arch Pathol Lab Med Vol 127, May 2003 Conference Summary Zarbo & Hammond

3 Table 1. Her-2/neu Protein Expression by Immunohistochemistry: Report Template Elements* Site Left breast Right breast Specimen type Incisional biopsy Excisional biopsy or resection Mastectomy Histologic type Infiltrating ductal carcinoma Infiltrating lobular carcinoma Other (specify) Do not report pattern in DCIS or LCIS Her-2/neu expression detected by immunohistochemistry Positive (3 ) Equivocal (2 ), FISH will be done and a separate report will be issued with final designation of Her-2/neu status Negative (1 ), FISH will be done and a separate report will be issued with final designation of Her-2/neu status or this laboratory has validated that 95% of these cases are negative by FISH Negative (0) Block examined Surgical pathology No. Block identification No. or letter Test performed HercepTest according to protocol (Dako Corporation, Carpinteria, Calif) Pathway according to protocol (Ventana Medical Systems Inc, Tucson, Ariz) Other antibody: specify clone, manufacturer, epitope retrieval conditions, detection system (Envision, ABC, LSAB, etc) Platform used Dako autostainer Ventana autostainer Manual staining Results Positive (3 ), percent of invasive tumor cells with complete membrane pattern Equivocal (2 ), percent of invasive tumor cells with weak intensity, complete membrane pattern Negative (1 ), percent of invasive tumor cells with weak incomplete membrane or cytoplasmic staining pattern Negative (0), no staining of invasive tumor cells Not evaluable, no staining of invasive tumor; intrinsic staining is negative Do not report staining pattern of DCIS Intrinsic staining pattern Appropriate (normal ducts show weak staining with Her-2/neu, or vimentin stain is positive) Negative (no normal ductal staining with Her-2/neu, or vimentin stain is negative) test will be repeated Not evaluable (no normal ducts in sample; vimentin stain not done); results to be interpreted with caution Control (cell line) staining pattern Appropriate positive and negative control staining is seen Positive control shows excessive staining; assay will be repeated Positive control shows weak staining; assay will be repeated Negative control shows staining; assay will be repeated Interpretation method Manual evaluation of invasive tumor cell staining Image analysis evaluation of invasive tumor cell staining by (name system and reference range for positivity if different than above, eg, ACIS system, range for positive [3 ] is 3.0 units) Table 1. Continued Laboratory quality assurance Laboratory is accredited by CAP/JCAHO/other Her-2/neu by (Herceptest, CB11) is an FDA-approved test to determine Her-2/neu protein overexpression in breast cancer patients. Laboratory personnel have undergone appropriate training Her-2/neu staining of 3 has been shown in this laboratory to have a concordance with FISH amplification of % (eg, 95%). Her-2/neu staining of 0, 1 has concordance rate with FISH amplification of (eg, 1% 5%). Therefore, only 2 cases are confirmed by FISH except in cases in which the tumor is poorly differentiated and estrogenreceptor negative, or specifically requested to be confirmed by the clinician. * DCIS indicates ductal carcinoma in situ; LCIS, lobular carcinoma in situ; FISH, fluorescence in situ hybridization; ACIS, automated cellular imaging system; CAP, College of American Pathologists; JCAHO, Joint Commission on Accreditation of Healthcare Organizations; and FDA, Food and Drug Administration. If concordance rates are less than those stated above, lab should consider running all tests by FISH or sending out confirmatory testing until the concordance rates are known. munocytochemistry, each antibody for professional and technical components). Fluorescence in situ hybridization is coded and billed using molecular cytogenetic codes (tissue in situ hybridization, interpretation, and report) for the professional component, and (DNA probe, each) and (interphase in situ hybridization, analyze cells) for the technical component. Pathologists should be aware that actual reimbursement may vary by payer and by national region. REPORTING OF HER-2/neu TESTING It is recommended that a standardized report format and defined terminology be used for reporting Her-2/ neu test results by all pathologists within a group. Tables 1 and 2 are suggested template elements for Her-2/neu IHC and FISH test reporting. In addition to the basic information related to the patient and morphologic verification of the tumor tested, the pathologist should include sample identification (case, block, slide), fixative type and duration of fixation, method used (test, clone/ probe, vendor), specification of positive and negative controls used, scoring system used, assay result, and reference ranges. The pathologist should also specify if secondary (reflex) testing will be done, and how it will be reported. 18 PATHOLOGIST LIABILITY IN HER-2/neu TESTING The nature of this type of stand-alone testing, on which therapeutic decisions are based, raises a number of liability issues for the pathologist. These concerns arise principally from the use of alternative, home-brew assays. The first is the potential malpractice exposure for use of non FDA-approved tests should the patient experience injury from the therapy. Depending on jurisdiction, use of home-brew Her-2/neu tests may possibly require that the pathologist advise the patient and obtain written informed consent in advance. It would behoove the pathologist to record the scientific rationale for selection of the home-brew test and to document the test validation and quality control. A home-brew test selection decision based solely on cost savings will not likely withstand a legal challenge. It would also be pru- Arch Pathol Lab Med Vol 127, May 2003 Conference Summary Zarbo & Hammond 551

4 Table 2. Her-2/neu Gene Amplification Status by Fluorescence In Situ Hybridization (FISH): Report Template* Site Left breast Right breast Specimen type Incisional biopsy Excisional biopsy or resection Mastectomy Histologic type Infiltrating ductal carcinoma Infiltrating lobular carcinoma Other (specify) Do not report pattern in DCIS or LCIS Her-2 gene amplification status by FISH Amplified (positive) Ratio of Her-2/neu gene signals/chromosome 17, 2.2 Average Her-2/neu gene signals/cell, 4.0 (Ventana) Borderline amplified (borderline) Ratio of Her-2/neu gene signals/chromosome 17, Average Her-2/neu gene signals/cell, (Ventana) Unamplified (negative) Ratio of Her-2/neu gene signals/chromosome 17, 1.8 Average Her-2/neu gene signals/cell, 3.5 (Ventana) Assay cannot be interpreted due to technical problems (see comment section) Aneusomy [increase chromosome 17 number identified (average 2.25 chromosome 17/cell)] Aneusomy absent ( 2 chromosome 17/cell) Aneusomy not evaluated by chromosome 17 probe (Ventana) Block examined Surgical pathology No. Block identification No. or letter Test performed Her-2/neu by Pathvision Her-2/neu by Inform (Ventana) Results Ratio of Her-2/neu gene copies/chromosome 17 Average No. of gene copies per cell (Ventana) Test cannot be interpreted due to poor resolution of FISH signal (test will be repeated or another block of tumor will be requested, or Her-2/neu by immunohistochemistry will be done) No. of invasive tumor regions counted: (1 3) No. of invasive tumor cells counted: (20 60) Tumor histologic evaluation Hematoxylin-eosin section of tissue used for FISH contains invasive tumor in the region counted Hematoxylin-eosin section of tissue used for FISH does not contain invasive tumor; another block of tumor will be requested or assay cannot be interpreted Controls FISH controls were reviewed and showed appropriate levels of Her-2/neu gene amplification Test controls showed equivocal results and test will be repeated dent for pathologists using assays of their own design to check their malpractice insurance coverage for use of a non FDA-approved methodology when an FDA-approved test is available. 19 Table 2. Continued Laboratory quality assurance Laboratory is accredited by CAP/JCAHO/other Her-2/neu by (Pathvision, Inform) is an FDA-approved test to determine Her-2/neu gene amplification status in breast cancer patients Laboratory is manufacturer approved to conduct this test (date). Assay validation was performed using (30 50) cases with independent review by a reference Laboratory and (100%) concordance of results * DCIS indicates ductal carcinoma in situ; LCIS, lobular carcinoma in situ; FISH, fluorescence in situ hybridization; CAP, College of American Pathologists; JCAHO, Joint Commission on Accreditation of Healthcare Organizations; and FDA, Food and Drug Administration. PIVOTAL ROLE FOR PATHOLOGISTS IN HER-2/neu TESTING STRATEGY The role of the pathologist in accurate and appropriate Her-2/neu testing is pivotal. In addition to test selection and quality assurance, the pathologist fulfills a major role in educating the medical staff about testing methods, results, and potential problems. In consultation with the clinical oncologist, the pathologist s role should include determination of which patients would be tested and what reflex testing will be done. In current medical practice, it is appropriate to consider testing every newly diagnosed patient with breast cancer, not just those with metastatic disease. Based on knowledge of optimal testing, the pathologist should determine which pathologic material should be tested and set up appropriate test order and reflex mechanisms to obtain clinically valid test results in each case. The decision to arrange for testing from a reference laboratory or to perform testing in-house should be made based on technical, legal, and reimbursement considerations. In-house testing will require the pathologist to develop the Her-2/neu test algorithm and reporting strategy acceptable to clinical oncologists, to develop or find the test expertise, to validate the method(s), and to determine the FISH/IHC concordance rate to confirm the test algorithm. Should testing be sent out, the pathologist is responsible for identifying a vendor partner, reviewing their protocols and quality assurance process, and developing the acceptable algorithm and reporting strategy. Regardless of which approach is used, it is extremely important for the pathologist to ensure that these specimens are optimally processed. The first step is to control specimen handling and fixation to standardize how quickly specimens are fixed, in what, and for how long. This initial preanalytic factor is so important in both IHC and FISH that fixation parameters should be recorded on each case. Another key role of the pathologist in Her-2/neu testing is the ongoing monitoring of laboratory performance. This practice would include ensuring that 10% formalin fixation is used for breast specimens and that the algorithm and protocol are being followed. Ongoing tracking of Her- 2/neu test quality control parameters should include the percentage of positive cases obtained and whether the percentage varies by different pathologist readers. It is also the responsibility of the pathologist to see that the laboratory is enrolled in an ongoing proficiency testing survey for both IHC and FISH, and that technical staff and pathologist competence is monitored. Based on clinical 552 Arch Pathol Lab Med Vol 127, May 2003 Conference Summary Zarbo & Hammond

5 consultation and laboratory performance monitoring, the Her-2/neu testing strategies should be modified or remediated as necessary. SUMMARY Pathologists should view the current Her-2/neu testing challenge as an emerging opportunity to play a larger and more pivotal role in clinical medicine by providing the laboratory determination responsible for the selection of patients for future targeted therapies. This larger role in pathology practice requires the ongoing education of colleagues regarding the significance of new testing as it becomes available and leadership in algorithm development to provide consistent and accurate testing. This rigorous approach to Her-2/neu testing should become the standard by which we view our role, as future laboratory assays that determine therapeutic decisions become available. We are indebted to the scientific contributions of the program faculty: Noel Weidner, MD; Daniel Hayes, MD; Robert Mass, MD; Jon Askaa, DVM, PhD; Kenneth Bloom, MD; Steven Gutman, MD; Jack Bierig; Raymond Tubbs, DO; and Jeffery Ross, MD. The program and conclusions were also enriched by the participants contributions to panel discussions. We are grateful to the staff of the College of American Pathologists for their professional support of this conference. References 1. Fitzgibbons PL, Page DL, Weaver D, et al. Prognostic factors in breast cancer: College of American Pathologists consensus statement Arch Pathol Lab Med. 2000;124: Hayes DF, Thor AD. c-erb-b-2 in breast cancer: development of a clinically useful marker. Semin Oncol. 2002;29: Hammond MEH, Taube SE. Issues and barriers to development of clinically useful tumor markers: a development pathway proposal. Semin Oncol. 2002;29: Taylor C. The total test approach to standardization of immunohistochemistry. Arch Pathol Lab Med. 2000;124: O Leary TF. Standardization in immunohistochemistry. Appl Immunohistochem Molec Morphol. 2001;9: Path Vysion Her-2 DNA Probe Kit [package insert]. Downers Grove, Ill: Vysis, Inc; Dako HercepTest anti-her2 IHC system for immunoenzymatic staining [package insert]. Carpinteria, Calif: Dako Corp; Pathway Her2 [package insert]. Tucson, Ariz: Ventana Medical Systems Inc; Penault-Llorca F, Adelaide J, Houvenaeghel G, et al. Optimization of immunohistochemical detection of ERBB2 in human breast cancer: impact of fixation. J Pathol. 1994;173: Rhodes A, Jasani B, Couturier J, et al. A formalin fixed, paraffin processed cell line standard for quality control of immunohistochemical assay of Her2/neu expression in breast cancer. Am J Clin Pathol. 2000;117: Dako HercepTest. A Manual for Interpretation. Carpinteria, Calif: Dako Corp; Wang S, Saboorian MH, Frenkel EP, et al. Assessment of Her-2/neu status in breast cancer: automated cellular imaging system (ACIS)-assisted quantitation of immunohistochemical assay achieves high accuracy in comparison with fluorescence in situ hybridization assay as the standard. Am J Clin Pathol. 2001;116: Bloom K. Her2 immunohistochemistry and FISH in clinical practice. Presented at: Strategic Science symposium; May 4 5, 2002; Rosemont, Ill. 14. Roche PC, Suman VJ, Jenkins RB, et al. Concordance between local and central laboratory Her 2 testing in the breast intergroup trial N9831. J Natl Cancer Inst. 2002;94: Tubbs R, Pettay J, Skacel M, et al. Gold-facilitated in situ hybridization: a bright-field autometallographic alternative to fluorescence in situ hybridization for detection of Her-2/neu gene amplification. Am J Pathol. 2002;160: Zhao J, Wu R, Au A, Marquez A, Yu Y, Shi Z. Determination of HER2 gene amplification by chromogenic in situ hybridization (CISH) in archival breast carcinoma. Mod Pathol. 2002;15: Tubbs RR, Pettay J, Roche P, et al. Concomitant oncoprotein detection with fluorescence in situ hybridization (CODFISH): a fluorescence-based assay enabling simultaneous visualization of gene amplification and encoded protein expression. J Mol Diagn. 2000;2: Wang S, Hossein SM, Frenkel EP, et al. Aneusomy 17 in breast cancer: its role in HER-2/neu protein expression and implication for clinical assessment of HER-2/neu status. Mod Pathol. 2002;15: Bierig J. Liability and payment issues in the selection of pathology assays. Arch Pathol Lab Med. 2002;126: Arch Pathol Lab Med Vol 127, May 2003 Conference Summary Zarbo & Hammond 553