Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Immunohistochemical Validation of Rare Tissues and Antigens With Low Frequency of Occurrence: Recommendations From The Anatomic Pathology Patient Interest Association (APPIA)

Immunohistochemical Validation of Rare Tissues and Antigens With Low Frequency of Occurrence:... REVIEW ARTICLE Immunohistochemical Validation of Rare Tissues and Antigens With Low Frequency of Occurrence: Recommendations From The Anatomic Pathology Patient Interest Association (APPIA) Robert L. Lott, BS, HTL(ASCP),* Peter V. Riccelli, PhD,* Elizabeth A. Sheppard, MBA,* Keith A. Wharton Jr, MD,† Eric E. Walk, MD,* George Kennedy, MBA,‡ and Bryce Portier, MD, PhD§ utility in clinical immunohistochemistry laboratories. Participation in Abstract: Laboratories worldwide find it challenging to identify External Quality Assessment program(s) may provide insufficient enough tissues and cases for verification and validation studies of low- numbers or the ability to calculate concordance rates. However, incidence, rare antigens. These antigens have a low frequency of oc- supplementation with in-house tissues can allow a laboratory to reach currence in the population, or have little or no expression in normal the optimal number of cases needed for verification and/or validation tissues. Validation studies are essential to assure testing stand- schemes. An alternative approach is conducting a thorough literature ardization before introducing a new instrument, product, or test into search and correlating staining patterns of the new test to the expected the clinical laboratory. The College of American Pathologists has results. These solutions may be used uniquely or together to assure published comprehensive guidelines for the verification and validation consistent standardized testing. of new immunohistochemical tests introduced into the laboratory Key Words: low incidence, verification, validation, im- menu. Within the guidelines, varied numbers of cases are required for munohistochemistry (IHC), in vitro diagnostic test (IVD), lab- nonpredictive versus predictive markers. However, regarding low-in- oratory developed test (LDT), external quality assurance (EQA) cidence antigens, the laboratory medical director determines the ex- tent of validation required. Recommended practical solutions (Appl Immunohistochem Mol Morphol 2021;29:327–334) available to clinical laboratories for low-incidence validation include developing internal resources using the laboratory information system with retrospective and prospective search(s) of archival material and he Anatomic Pathology Patient Interest Association (AP- purchase of tissue microarray blocks, slides, or cell lines from external TPIA) is a not-for-profit membership organization dedi- resources. Utilization of homemademultitissueblockshas proved to cated to the issues that affect laboratory practices, quality and, be extremely valuable in biomarker research and demonstrated great ultimately, the safety of the patient. In keeping with the mis- sion of the organization, to ensure accuracy and reduce var- iation in test results, proper validation studies are needed at the Received for publication July 23, 2019; accepted October 23, 2019. onset of clinical test implementation. However, laboratories From the *Roche Tissue Diagnostics, Tucson, AZ; †Leica Biosystems, Richmond, IL; ‡Sakura Finetek USA, Inc., Torrance; and §Agilent worldwide face challenges in gathering enough tissues and Technologies, Santa Clara, CA. cases for antigens with rare occurrence in the population or R.L.L.: royalty payments, College of American Pathology Press; from patients with a specific rare disease to perform required Employee, Roche Tissue Diagnostics, Tucson, AZ; and Member, verification and validation studies for new laboratory products CAP Histotechnology Committee. P.V.R. and E.A.S.: are employees at Roche Tissue Diagnostics, Tucson, AZ. K.A.W.: Employee, and immunohistochemical (IHC) tests. Procurement of bio- Leica Biosystems/Danaher, Inc.; Stockholder, Danaher, Inc.; and logical material for these biomarkers is technology-agnostic Stockholder, Novartis AG. E.E.W.: Employee and shareholder, and not unique to IHC: the same challenges exist for molec- F. Hoffmann-La Roche AG. G.K.: Employee, Sakura Finetek ular testing. Finding and procuring biospecimens from rare USA Inc., Torrance, CA. B.P.: Employee and shareholder, Agilent cancersisalsodifficult. When combining a low-frequency Technologies, Santa Clara, CA; Member, CAP Molecular Oncology Committee. occurrence biomarker in a rare cancer, acquiring the required Reprints: Robert L. Lott, BS, HTL(ASCP), Roche Tissue Diagnostics, number of positive samples to properly perform verification 1910 E Innovation Park Drive, Tucson, AZ 85755 (e-mail: robert. and validation seems daunting. For reference, the College of lott@roche.com). American Pathologists (CAP) currently recommends for each Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. This is an open access article distributed under the terms of the predictive test that a laboratory acquire 20 positive and 20 Creative Commons Attribution-Non Commercial-No Derivatives negative cases for verification and validation. License 4.0 (CCBY-NC-ND), where it is permissible to download and In a recent CAP survey publication that assessed share the work provided it is properly cited. The work cannot be changes in IHC analytic validation practices following the changed in any way or used commercially without permission from the journal. publication of an evidence-based laboratory practice guideline, Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 www.appliedimmunohist.com | 327 Lott et al Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 the 2 biggest challenges in implementing the guideline was MSH6. Trying to find enough positive and negative cases difficulty in finding validation cases for rare antigens, and the for 1 or all 4 biomarkers to validate for MMR proteins by other was time/staff resource limitations. IHC can be difficult, as only 1 in 9 (11%) colorectal car- For clarification, verification is defined as the process cinomas contain the deleted MLH1 gene and only 1 in 50 by which a laboratory determines that an IVD assay per- (2%) contain the deleted MSH2 gene. Thus, a laboratory forms according to the specifications set forth by the man- would need to test potentially ≥ 180 unselected cancers to ufacturer. Validation is defined as the process by which a identify 20 MLH1 gene deleted cases, and perhaps 1000 laboratory confirms a laboratory developed test (LDT) or cancers to find 20 MSH2 gene deleted cases. modified Food and Drug Administration (FDA)-cleared/ The V600E activating mutation in BRAF and its de- approved test performs as intended or claimed. tection by IHC may or may not be rare depending on the assay application. It has been detected in ~7% of all solid tumors, including 45% of papillary thyroid carcinomas, 40% BACKGROUND to 60% melanomas, 35% of serous ovarian carcinomas, but Historically, IHC laboratories have experienced diffi- only 5% to 15% of colorectal adenocarcinomas, and only 1% culty in finding enough cases for verification and/or vali- to 3% of lung and other cancers. dation of new tests that use antibodies against specific Several IHC markers are useful for the identification of markers with a low frequency of occurrence. For example, spindle cell tumors and soft tissue tumors. For example, ALK+ positive lymphoma, a type of anaplastic large cell TLE-1 is a sensitive and specific marker for synovial sarco- lymphoma (ALCL), is considered “rare” because it com- ma, yet the incidence is estimated to be only 900 new cases a prises only 2% to 3% of all lymphomas in adults and 10% to year in the United States. The antibody INI-1 is excellent 30% of lymphomas in children. Only about half (50% to for the identification of malignant rhabdoid tumors, and 60%) of those show positivity for the ALK fusion protein Wilms tumors, typically found in young children. However, by IHC (Fig. 1). Finding a single positive case can be the incidence is ~0.19 per million for renal tumors, 0.89 per challenging. In addition, choice of primary antibodies and million for atypical teratoid/rhabdoid tumor (AT/RT), and protocol parameters for the validation of ALK+ lymphomas 0.32 per million for tumors of other sites. Only 650 cases of mayvarysignificantly than those of ALK+ lung carcinomas. Wilms tumor is diagnosed in the United States each year. Lynch syndrome-associated cancers are charac- Soft tissue sarcomas (STS) comprise about 7% of all terized by Mismatch Repair (MMR) deficiency caused by malignancies in children and adolescents under the age of somatic inactivation of the sole remaining wild-type 20 years and rhabdomyosarcoma (RMS) accounts for MMR gene allele. The central step in diagnosing Lynch about 40% of pediatric soft tissue sarcomas. Antibodies syndrome is the analysis of patient tumor tissue, usually used for identification of RMS, such as Myo-D1 and colon or endometrial adenocarcinoma, for the presence of myogenin, are very difficult to validate because the in- MMR deficiency by IHC for MLH1, PMS2, MSH1, and cidence of RMS is only 4.5 cases/million children and adolescents per year. Each year, ~4000 to 5000 adults in the United States are diagnosed with a gastrointestinal stromal tumor (GIST). CD117 (c-kit) and DOG-1 performed by IHC have proven to be reliable and sensitive diagnostic mark- 9,10 ers for detection of GIST. However, GIST accounts for 9,10 <1% of all gastrointestinal tumors. Overall, about 85% of GISTs are reported to have a protein activating mu- tation in KIT or PDGFRA. Studies of PD-L1 expression in lung cancer report this biomarker to be positive in 13% to 70% of cases, partly dependent on how expression is tested for and defined. However, a much smaller number of “positive” cases will meet certain percent-positive cutoffs (1%, 5%, 25%, 50% etc) (Table 1). Patients with non–small cell lung cancer and with anaplastic lymphoma kinase (ALK) gene rearrangements constitute only 4% to 5% of all non–small cell lung cancer 13,14 patients. To introduce and properly validate new tests in the clinical laboratory, the College of American Pathologists (CAP) has published comprehensive guidelines for the verification and validation of IHC tests introduced into the laboratory menu. Note that CAP recommends differ- ent numbers of cases for nonpredictive versus predictive FIGURE 1. ALK protein (+) lymphoma (ALCL) IHC. ALCL indicates anaplastic large cell lymphoma; IHC, immunohistochemical. markers; laboratory verification studies require twice as 328 www.appliedimmunohist.com Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 Low-incidence IHC Validation TABLE 1. PD-L1 Prevalence by Immunohistochemical No. PD-L1 Positive Immune Cell % of IC Cases No. PD-L1 Positive Tumor Cell % of TC Cases Indications n (+) (> 5%) Cases (> 5%) (+) (> 5%) Cases (> 5%) Non–small cell lung 184 48 26 44 24 Cancer Renal cell cancer 88 22 25 9 10 Melanoma 58 21 36 5 3 Head/Neck SCCa 101 28 28 19 19 Gastric cancer 141 25 18 7 5 Colorectal cancer 77 27 35 1 1 Pancreatic cancer 83 10 12 3 4 IC indicates immune cell; SCCa, squamous cell carcinoma; TC, tumor cell. many positive and negative cases for FDA-approved is appropriate for the test’s intended use. Its requirements of predictive assays. If a laboratory chooses to develop their 20 positive and 20 negative cases are descriptive of a strictly own predictive marker assay, such as CD117 or PD-L1, “analytic” validation process. It provides no specifications using a LDT, or chooses to alter an FDA-approved assay, for tissues and tumors or even a range of epitope expression. even more rigorous verification analyses are required. However, it may not be entirely fit for purpose for validation There are specific differences in verification requirements of a predictive biomarker where a more “clinical” type of between a laboratory developing LDTs and using an validation process, designed with tissue specimens that test in vitro diagnostic (IVD) tests from industry, especially the thresholds or cutoffs (eg, <1%, <10%, etc) used in clinical with predictive marker assays. Each laboratory must de- trials or in a comparator validated assay, is required. cide the best option, considering parameters like costs (eg, So rather than just positive and negative, this type of of personnel, material), development time, and skills of clinical validation tests the diagnostic sensitivity and personnel. In some laboratories, neither of these require- specificity of the assay. For comparison, at least 50 pos- ments are practical or feasible, so laboratories have im- itive and 50 negative samples are recommended by the plemented alternative strategies in an attempt to be CLSI EP12-A2 User Protocol for Evaluation of Qual- compliant with these guidelines. The CAP guidelines are itative Test Performance. briefly summarized in Table 2. Recent publications by CAP sponsored surveys for Estrogen receptor, progesterone receptor, and HER-2/ practices in IHC testing laboratories reveal that most labo- neu testing for breast cancer have unique, specific validation ratories are not using the recommended sample sizes for 16,17 guidelines. Other global quality organizations have validation as described in Table 2. In 2 studies for initial adopted similar validation guidelines, especially with regard validation, 75.4% (538 of 714) of laboratories adopted the 18–21 to overall concordance. The CAP guideline recom- 20-case minimum for nonpredictive markers but less than mendations, as described, help ensure that the IHC test ac- half (45.9% of the 579 responding laboratories) adopted the curately measures the analyte of interest and that validation 40-case minimum for predictive markers as outlined in the TABLE 2. Guidelines for Introducing a New In Vitro Diagnostic (IVD) Antibody into the Laboratory Step #1 Optimization Use tissues indicated for the specific intended use or clinical application. For IVD antibodies, the vendor’s package insert protocol should be used as a starting point. For other tests, optimization steps may include the testing of antigen retrieval, 1 degree Ab. titration, detection system, chromogen, amplification, counterstaining, etc Step #2 Verification Use the optimized protocol from above: For nonpredictive assays, laboratories should run a minimum of 10 positive and 10 negative cases For FDA-approved predictive assays, laboratories should run a minimum of 20 positive and 20 negative/low-expressor cases Labs choosing to alter FDA-approved kits or develop their own predictive marker assay(s) (LDTs), should run a minimum of 40 positive and 40 low-expressor/negative cases (0, 1+). This includes LDTs for ER, PR, and HER2 Laboratories must verify new IHC tests before placing them into clinical service Step #3 Validation After verification testing is complete, those slides are compared/correlated (validated) in at least one of the following ways: Compare the new test results with a prior validated assay/test using the same tissue set Compare the new test results with validated results from another laboratory using the same tissue set Compare the new test with an alternate validated non-IHC test (ISH, etc) Compare the new test with the morphology and expected results (ie, from a thorough literature search) Testing is compared with graded results from formal Proficiency Testing challenges For validation, every assay must achieve 90% overall concordance between the new test and the comparator test All tissues/cases in this process should be fixed and processed the same as cases tested clinically. Laboratories may use whole sections, tissue microarrays, multitissue blocks as appropriate. Tissues containing low-incidence antigens may be difficult to find. The characteristics of the tissues or cases used for validation should be similar to those seen in the laboratory’s patient population. Tissues should include relevant normal tissues, if available, and neoplasias which span a range of expression from negative to low to high. FDA indicates Food and Drug Administration; IHC, immunohistochemical; LDT, laboratory developed test. Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. www.appliedimmunohist.com 329 Lott et al Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 2014 evidence-based laboratory practice guideline. As may suffice to cover a group of similar intended uses, as reported by 714 laboratories, the median number of cases long as the overall process of preparation and inter- required for nonpredictive assays was 20, and the median pretation is the same. number of cases required for predictive assays was only 25, Interestingly, recent CAP proficiency testing survey well below the guideline recommendation of 40. results on validation and implementation of Next Gen- Recent review of the FDA IHC reagent classifications, eration Sequencing (NGS) in clinical oncology applica- IHC validation protocols, and quality assurance (QA) proce- tions showed that about half of the 58 laboratories that dures and documentation needed in conjunction with IHC test reported on their variant confirmation procedures are not guidelines for Mohs Labs, IHC reagents are subject to several performing variant confirmation (n = 28; 48%). For those histologic and clinical controls to maximize the accuracy of laboratories that are performing variant confirmation, IHC reagents, with it being at the discretion of laboratory they are employing various orthogonal technologies in- medical director to establish these quality control procedures cluding Sanger sequencing (n = 26), targeted mutation before offering these tests for patient diagnostics. testing (n = 11), pyrosequencing (n = 8), or other tech- In a survey of over 700 laboratories, to establish baseline nologies (n = 5). parameters for IHC validation procedures and practice and to assess feasibility of implementation, Hardy and colleagues, SOLUTIONS noted that for non-FDA approved, nonpredictive IHC tests, There are many IHC tests available for rare, low- 75% of laboratories used 21 or fewer total cases for validation. incidence antigens and tissues. The increasingly demand- Less than 50% of laboratories used ≥25 cases to validate non– ing requirements for verifying or validating new antibody FDA-approved predictive markers (LDTs). Of those, only 47% tests, while necessary, have made it difficult for labo- used cases with any weak to moderate expression levels. When ratories wanting to perform in-house testing to identify the comparing IHC HER2/neu assays with an IHC test performed required number of cases and needed amount of tissue in another laboratory, only 56% of laboratories used a especially those tissues expressing rare antigens. CAP recommended minimum of 25 cases. Seventy-five percent of recognizes this fact and has addressed it in its current respondents had validated the most recently introduced new Anatomic Pathology accreditation checklist antibody, with a median of only 15 cases. Major challenges to meet the proper validation requirements included tissue ANP.22750—Antibody Validation availability (especially for cytological materials), availability of appropriate tissue controls which were highly dependent upon The extent of this validation is at the discretion of the particular antibody used, and time and personnel resources the medical director. If a laboratory does not use 10 to find and qualify appropriate cases. Most respondents felt a positive cases and 10 negative cases for their nonpredictive minimum of 25 cases was appropriate for validation of a pre- marker validation, then the rationale for this needs to be dictive marker and 13 cases for nonpredictive markers as pre- documented in the validation summary. Just because a dictive markers are much less likely to have morphologic clues laboratory uses less than the suggested number of cases for to refute an errant stain a validation does not necessarily mean that they are out of Considering another diagnostic procedure, Whole compliance. Slide Imaging (WSI), CAP has established an evidence- based guideline setting the minimum number of cases to …the laboratory director determines that fewer validation 60 after finding that when an average of 20 cases (range, cases are sufficient for a specific marker (e.g. a rare antigen 10 to 46 cases) was used to compare WSI and glass slide, or tissue), the rationale for that decision needs to be there was significantly lower accuracy (77%) and con- recorded. cordance (75%), but when an average of 60 cases (range, 52 to 90 cases) was used, the studies showed improved CAP does require that ER, PR and HER2 assays accuracy (90%) and far better concordance (95%). Thus, employed as predictive markers on breast carcinoma have a CAP recommended that validation studies for WSI in- minimum of 20 positive cases and 20 negative cases. All clude at least 60 routine cases per application, [eg, hem- other predictive markers are to bevalidated,but theextentof atoxylin eosin (H&E)-stained sections of fixed tissue, that validation is, again, at the discretion of the medical frozen sections, cytology, hematology] that reflects the director and will vary with the antibody, especially those spectrum and complexity of specimen types (biopsies and against low-incidence antigens and rare, hard to find tissue resections) and diagnoses likely to be encountered during samples. If IHC is regularly performed on specimens that are routine practice. If the laboratory intends to use its vali- not fixed or processed in the same manner as the tissues used dated WSI system for another supplemental application for validation (eg, alcohol fixed cell blocks, cytological (eg, to evaluate special stains, IHC or fluorescence stains smears, formalin post fixed tissue, or decalcified tissue), the and H&E-stained sections), then another 20 cases of the laboratory should test a sufficient number of such tissues to “additional” application should be validated. In addition, ensure that assays consistently achieve expected results using the study showed that the specimen preparation type the appropriate protocol for the intended use. Again, the (H&E, Papanicolaou, etc) was a more important perfor- laboratory director is responsible for determining the num- mance variable than the source of the tissue or the specific ber of positive and negative cases and the number of pre- analyte being assessed. Thus, a single validation study dictive and nonpredictive markers to test. 330 www.appliedimmunohist.com Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 Low-incidence IHC Validation There are several solutions available to laboratories sections should be used if TMAs/ MTBs are not appro- wanting to perform verification and/or validation studies priate for the targeted antigen or if the laboratory medical in-house for low-frequency antigens, including: director cannot confirm that the fixation and processing of Using specimen blocks identified via retrospective and TMAs/MTBs is similar to clinical specimens. MTBs prospective LIS search/survey from “in-house” cases. should be employed for external positive and negative Purchase of tissue blocks, multitissue blocks (MTBs), or tissue controls. Use of multitissue external positive con- precut slides from qualified sources, or creating trols by IHC technologists provides consistent positive homemade MTBs. feedback because it simplifies workflow and minimizes Participation in External Quality Assessment (EQA) potential error. TMA external positive controls have also proficiency programs. demonstrated superior performance in staining con- Correlating the new test’s results with expected results sistency and reproducibility. based on literature search, demonstrating same tumor Several qualified vendors offer both TMA slides morphology and antigen distribution. and blocks for purchase in a variety of formats including formalin-fixed, paraffin-embedded tissues (Table 3). These multicore arrays, some of which contain 300+ cores, Retrospective Search include normal, malignant, or metastatic tissues and are A common, almost routine, activity in most aca- suitable for use in validation studies. Multicore TMAs demic medical center, reference, community hospital, and may contain protein, RNA, or DNA and provide an ideal private laboratories is performing a keyword search(s) to high throughput method for rapid analysis of IHC identify in-house, archived cases using the laboratory in- markers or in situ hybridization. Many companies can formation system (LIS). Most systems are able to print or provide custom tissue array constructions for use in save list(s) of cases that can be used to identify and/or immunohistochemistry and in situ hybridization. In screen cases and blocks with low-incidence antigens. addition, vendors can often provide relevant preanalytic As an example, a keyword search for “Anaplastic variables such as fixation times for each tissue. Cost of large cell lymphoma (ALCL),” an aggressive type of non- these products varies considerably and depends upon Hodgkin T-cell lymphoma, may reveal only a few cases in factors such as sample type, availability, size of sample set, the block archive because of its rarity. However, uti- and antigen presentation. lization of in-house, archived tissue blocks, even if rare, Several vendors are able to generate engineered cell best fits the CAP checklist instruction that “laboratories lines with low/negative, intermediate, and high levels of should use validation tissues that have been processed protein using proprietary gene-editing platforms (Fig. 2). using the same fixative and processing methods as cases The raw material is then formalin-fixed, processed in a that will be tested clinically.” traditional manner using alcohol and xylene, and paraffin embedded. These cell buttons, which mimic cytology cell blocks, can be used as reference standards that serve as Purchase surrogates for real patient samples within an IHC testing In a recent review of diagnostic IHC standardization protocol. They can also be used as a renewable and procedures by Lin and Chen, tissue microarray (TMA) consistent point of reference, essentially run controls, blocks containing a grid-like array of various tumors and/ when optimizing and monitoring performance of an IHC or normal tissues have proved to be extremely valuable in assay. biomarker research and have demonstrated great utility in clinical IHC laboratories. At least 4 different prototypes of Multitissue Blocks TMA blocks can be used: Homemade MTBs represent an alternative to whole (1) TMA block containing a broad spectrum of tumors sections or TMAs. MTBs containing just 8 to 12 different and/or normal tissues from various organs, useful for cases of specific tissue-types can be used to supplement screening a new biomarker. (2) TMA block containing 50 to 100 tumors with a specific diagnosis, such as lung adenocarcinoma, TABLE 3. Vendors That Provide TMA Blocks and/or Slides useful for antibody validation, revalidation, estimating Cooperative Human Tissue http://chtn.sites.virginia.edu/tissue- marker prevalence in a specific indication, and in Network microarrays determining the diagnostic sensitivity and specificity of Folio Biosciences http://www.foliobio.com/products/ a newly discovered antibody. tissue-microarrays (3) TMA block containing 5 to 10 cases of a specific type Histocyte Laboratories http://www.histocyte.com Horizon Discovery http://www.horizondiscovery.com of tumor, useful for antibody testing and optimization. Invitrogen http://www.invitrogen.com (4) TMA block containing 5 to 10 cases of selected, mixed Origene http://www.origene.com/tissue/tissue_ tumors and/or normal tissues from various organs, microarrays.aspx useful as external positive and negative control tissues Pantomics http://www.pantomics.com StatLab Medical Products http://www.statlab.com for each antibody. US Biolab http://www.usbiolab.com/custom-tissue- array Laboratories may use whole sections, TMAs, and/or US Biomax http://www.biomax.us/tissue-arrays MTBs in their validation sets as appropriate. Whole Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. www.appliedimmunohist.com 331 Lott et al Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 FIGURE 2. IHC staining of engineered cell lines that express PD-L1 at varying levels, from high (left) to low (right). IHC indicates immunohistochemical. existing in-house tissues for optimization and verification Participation in such programs provides appropriate work, comparison and parallel testing of new antibody support for initial and continuous revalidation of these lots, reverification of assay modifications, and positive and testsaswell. negative tissue control material. The Canadian Immunohistochemistry Quality The concept involves some up-front work on the Control (cIQc) EQA scheme regularly conducts EQA ac- part of laboratory personnel to identify tissues and as- tivities for rare antigen, hard-to-find tissue specimens such semble the MTBs. The blocks consist mainly of normal as (ie, ALK, IDH1, BRAF V600E, PD-L1, MMR) using tissue types easily found in typical pathology labo- medium-to-large multiple-core tissue micro-array slides ratories. The best MTBs are those put together in the (TMAs) (Fig. 4). user laboratory using tissues fixed and processed in that Large TMAs containing multiple cores of positive laboratory. They contain both positive and negative in- and negative tissues processed by standard methods allow ternal controls for a wide range of IHC tests. A single a laboratory to supplement in-house tissue numbers to MTB paraffinblock canbeusedfor verification studies reach the number of cases needed for verification and/or in over 100 routine IHC tests and are useful in supple- validation. They also help improve the precision and ac- menting rare antigen and hard to find tissue searches. As curacy of testing. We presume that any global EQA pro- seen in Figure 3, for example, this simple block consists gram that provides such rare antigen samples would be of: acceptable for test verification and validation. (1) 2 to 3 full cross sections of appendix from different cases The CAP MMR survey for DNA MMR proteins (2) 2 to 3 small-medium sized pieces of liver resection, provides fewer cores per slide, but 2 unique TMA slides including normal liver from different cases rather than just 1 (Fig. 5). TMA slides used for EQA (3) 2 to 3 pieces of tonsil resection, to include squamous assessment are often available for purchase separately epithelium and lymphatic nodules from different cases from the scheme as an aid to laboratories. (4) 2 to 3 sections of normal pancreas containing islets, Participation in EQA programs that do not pro- ducts, and acinar cells from different cases. vide sufficient numbers of tissue samples may be in- formative and useful, but do not provide the opportunity The MTB contains 8 to 12 different patient tissues to calculate concordance rates with adequate power and can be used for a wide variety of verification and unless supplemented with in-house samples, and there- validation activities. Similarly, if sections of normal skin, fore alone do not provide sufficient numbers of cases for melanoma, striated muscle, and brain (cerebellum) are verification or validation. Participation in EQA how- combined in a MTB another 25 to 30 IHC markers can ever, informs the laboratory of the sensitivity and spec- easily be used to verify and validate. ificity of their protocol and their platform choices and serves as an indicator of the need for overall opti- External Quality Assessment mization of the IHC assay. Incidentally, it is essential Internally, if hospitals and reference laboratories that laboratories performing Class II and/or Class III find it difficult to acquire sufficient numbers of tissue IHC testing document participation in EQA programs in samples for initial validation, some EQA programs may the laboratory Quality Management Plan (QMP), and if provide test samples that quantitatively and qual- suboptimal results are achieved, corrective actions are itatively support IHC test verification and validation. taken and documented. 332 www.appliedimmunohist.com Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 Low-incidence IHC Validation FIGURE 3. Slide made from a multitissue block consisting mainly of normal tissue types. 2 to 3 full cross sections of appendix from different cases. 2 to 3 small-medium sized pieces of liver resection, including normal liver from different cases. 2 to 3 pieces of tonsil resection, to include squamous epithelium and lymphatic nodules from different cases. 2 to 3 sections of normal pancreas containing islets, ducts, and acinar FIGURE 4. BRAF V600E TMA slide from cIQc. TMA indicates cells from different cases. MTB indicates multitissue block. tissue microarray. Use of Expected Staining Results laboratories for 725 nonpredictive markers (61%) and 101 Verification activities do not always have to involve predictive markers (46.5%). Comparing the new test stained slides being compared with previously stained results with the same tissue set stained in another labo- slides or slides stained in another laboratory, or results ratory was a distant second used by only 123 (17%) of being judged comparable to an alternative method (ie, laboratories in the survey. FISH) to validate or revalidate a new test, especially if staining is for a low-incidence antigen. An alternative DISCUSSION approach is a thorough literature search, guided by the pathologist(s), to reveal “both” the expected morphology Many laboratories have problems finding sufficient of the tumor(s) under study and the correlation of staining tissues and cases to adequately validate antibodies against patterns to the “expected results” with the new test. rare antigens with a low frequency of occurrence and little Results from a survey of 1085 clinical laboratories, or no expression in normal tissues. There are several comparing and correlating the staining results of the new practical solutions available to clinical laboratories. These test with expected results, regardless of specimen size, was include developing internal resources with retrospective the “most common” validation method used among and prospective search(s) of archival material, purchase of Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. www.appliedimmunohist.com 333 Lott et al Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 11. Kerr KM, Tsao MS, Nicholson AG, et al. Programmed death-ligand 1 immunohistochemistry in lung cancer: in what state is this art? J Thorac Oncol. 2015;10:985–989. 12. Herbst RS, Soria JC, Kowanetz M, et al. Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature. 2014;515:563–567. 13. Chia PL,MitchellP,DobrovicA,etal. Prevalence and natural history of ALK positive non-small-cell lung cancer and the clinical impact of targeted therapy with ALK inhibitors. Clin Epidemiol. 2014;6:423–432. 14. Cheung CC, Garratt J, Won J, et al. Developing ALK immunohis- tochemistry and in situ hybridization proficiency testing for non- small cell lung cancer in Canada: Canadian immunohistochemistry quality control challenges and successes. Appl Immunohistochem Mol Morphol. 2015;23:677–681. 15. Fitzgibbons PL, Bradley LA, Fatheree LA, et al. Principles of analytic validation of immunohistochemical assays: guideline from the College of American Pathologists Pathology and Laboratory Quality Center. Arch Pathol Lab Med. 2014;138:1432–1443. 16. Hammond ME, Hayes DF, Dowsett M, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommen- dations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer. Arch Pathol Lab Med. 2010;134:907–922. 17. Wolff AC, Hammond ME, Schwarz JN, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol. 2007;25:118–145. 18. Hanna W, O’Malley FP, Barnes P, et al. Updated recommendations from the Canadian National Consensus Meeting on HER2/neu testing in breast cancer. Curr Oncol. 2007;14:149–153. 19. Torlakovic EE, Riddell R, Banerjee D, et al. Canadian Association of Pathologists-Association canadienne des pathologistes National Standards Committee/Immunohistochemistry: best practice recom- mendations for standardization of immunohistochemistry tests. Am J FIGURE 5. MMR TMA slides from CAP. CAP indicates College Clin Pathol. 2010;133:354–365. of American Pathologists; MMR, Mismatch Repair; TMA, tissue 20. Ellis IO, Bartlett J, Dowsett M, et al. Best Practice No 176: updated microarray. recommendations for HER2 testing in the UK. J Clin Pathol. 2004;57:233–237. 21. Bilous M, Dowsett M, Hannah W, et al. Current perspectives on TMA blocks and slides from outside resources, and par- HER2 testing: a review of National Testing Guidelines. Mod Pathol. ticipation in EQA schemes, and comparison with pub- 2003;16:173–182. lished literature. These solutions may be used uniquely 22. Torlakovic E. How to validate predictive immunohistochemistry and in combination to achieve verification and validation testing in pathology? Arch Pathol Lab Med. 2019;143:907. 23. Garrett PE, Lasky FD, Meier KL, et al. Clinical and Laboratory requirements for new and ongoing IHC tests. Standards Institute User Protocol for Evaluation of Qualitative Test Performance: Approved Guideline, 2nd ed. Wayne, PA: Clinical and REFERENCES Laboratory Standards Institute; 2008. 1. Fitzgibbons PL, Goldsmith JD, Souers RJ, et al. Analytic validation 24. Stuart LN, Volmar KE, Nowak JA, et al. Analytic validation of of immunohistochemical assays: a comparison of laboratory immunohistochemistry assays: New Benchmark Data from a survey practices before and after introduction of an evidence-based guide- of 1085 laboratories. Arch Pathol Lab Med. 2017;141:1255–1261. line. Arch Pathol Lab Med. 2017;141:1247–1254. 25. Kinas CG, Carroll BT. General guidelines for quality assurance of 2. College of American Pathologists Checklist, Anatomic Pathology, v. immunohistochemistry in a Mohs lab. Dermatol Surg. 2017;43:507–511. 07-29-2013, Definition of Terms. 26. Hardy LB, Fitzgibbons PL, Goldsmith JD, et al. Immunohistochemistry 3. Vose JM. Getting the Facts: Anaplastic Large Cell Lymphoma. New validation procedures and practices: a College of American Pathologists York, NY: Lymphoma Research Foundation; 2009:1–2. survey of 727 laboratories. Arch Pathol Lab Med. 2013;137:19–25. 4. Jasperson KW, Tuohy TM, Neklason DW, et al. Hereditary and 27. Pantanowitz L, Sinard JH, Hendricks WH, et al. Validating whole familial colon cancer. Gastroenterology. 2010;138:2044–2058. slide imaging for diagnostic purposes in pathology: guideline from 5. Pakneshan S, Salajegheh A, Smith RA, et al. Clinicopathological relevance the College of American Pathologists Pathology and Laboratory of BRAF mutations in human cancer. Pathology. 2013;45:346–356. Quality Center. Arch Pathol Lab Med. 2013;137:1710–1722. 6. Cure Sarcoma. Synovial sarcoma. 2019. Available at: www. 28. Nagarajan R, Bartley AN, Bridge JA, et al. A window into clinical curesarcoma.org/patient-resources/sarcoma-subtypes/synovial-sarcoma/. next-generation sequencing-based oncology testing practices. Arch Accessed February 22, 2017. Pathol Lab Med. 2017;141:1679–1685; 2017. 7. Annual incidence of rhabdoid tumors among children <15, 2000-2008. 29. College of American Pathologists, 2017 Anatomic Pathology Surveillance, Epidemiology, and End Results (SEER) Program; Bethesda, Checklist, ANP.22750 – Antibody Validation. MD: 2011. Volume SEER 17. SEER*Stat version 7.0.6. 30. Lin F, Chen Z. Standardization of diagnostic immunohistochemis- 8. Gurney JG, Young JL, Roffers SD, et al SEER Pediatric try: literature review and geisinger experience. Arch Pathol Lab Med. Monograph. National Cancer Institute; 2005. Soft Tissue Sarcomas. 2014;138:1564–1577. 9. Rubin BP, Heinrich MC, Corless CL. Gastrointestinal stromal 31. Torlakovic E, Nielsen S, Vyberg M, et al. Getting controls under tumour. Lancet. 2007;369:1731–1741. control: the time is now for immunohistochemistry. J Clin Pathol. 10. Steigen SE, Eide TJ. Gastrointestinal stromal tumors (GISTs): 2015;68:879–882. a review. APMIS. 2009;117:73–86. 334 www.appliedimmunohist.com Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Immunohistochemistry Wolters Kluwer Health

Immunohistochemical Validation of Rare Tissues and Antigens With Low Frequency of Occurrence: Recommendations From The Anatomic Pathology Patient Interest Association (APPIA)

Download PDF
8 pages

Loading next page...
 
/lp/wolters-kluwer-health/immunohistochemical-validation-of-rare-tissues-and-antigens-with-low-7YEc5ieBnx

References (65)

Publisher
Wolters Kluwer Health
Copyright
Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc.
ISSN
1062-3345
eISSN
1533-4058
DOI
10.1097/pai.0000000000000821
Publisher site
See Article on Publisher Site

Abstract

REVIEW ARTICLE Immunohistochemical Validation of Rare Tissues and Antigens With Low Frequency of Occurrence: Recommendations From The Anatomic Pathology Patient Interest Association (APPIA) Robert L. Lott, BS, HTL(ASCP),* Peter V. Riccelli, PhD,* Elizabeth A. Sheppard, MBA,* Keith A. Wharton Jr, MD,† Eric E. Walk, MD,* George Kennedy, MBA,‡ and Bryce Portier, MD, PhD§ utility in clinical immunohistochemistry laboratories. Participation in Abstract: Laboratories worldwide find it challenging to identify External Quality Assessment program(s) may provide insufficient enough tissues and cases for verification and validation studies of low- numbers or the ability to calculate concordance rates. However, incidence, rare antigens. These antigens have a low frequency of oc- supplementation with in-house tissues can allow a laboratory to reach currence in the population, or have little or no expression in normal the optimal number of cases needed for verification and/or validation tissues. Validation studies are essential to assure testing stand- schemes. An alternative approach is conducting a thorough literature ardization before introducing a new instrument, product, or test into search and correlating staining patterns of the new test to the expected the clinical laboratory. The College of American Pathologists has results. These solutions may be used uniquely or together to assure published comprehensive guidelines for the verification and validation consistent standardized testing. of new immunohistochemical tests introduced into the laboratory Key Words: low incidence, verification, validation, im- menu. Within the guidelines, varied numbers of cases are required for munohistochemistry (IHC), in vitro diagnostic test (IVD), lab- nonpredictive versus predictive markers. However, regarding low-in- oratory developed test (LDT), external quality assurance (EQA) cidence antigens, the laboratory medical director determines the ex- tent of validation required. Recommended practical solutions (Appl Immunohistochem Mol Morphol 2021;29:327–334) available to clinical laboratories for low-incidence validation include developing internal resources using the laboratory information system with retrospective and prospective search(s) of archival material and he Anatomic Pathology Patient Interest Association (AP- purchase of tissue microarray blocks, slides, or cell lines from external TPIA) is a not-for-profit membership organization dedi- resources. Utilization of homemademultitissueblockshas proved to cated to the issues that affect laboratory practices, quality and, be extremely valuable in biomarker research and demonstrated great ultimately, the safety of the patient. In keeping with the mis- sion of the organization, to ensure accuracy and reduce var- iation in test results, proper validation studies are needed at the Received for publication July 23, 2019; accepted October 23, 2019. onset of clinical test implementation. However, laboratories From the *Roche Tissue Diagnostics, Tucson, AZ; †Leica Biosystems, Richmond, IL; ‡Sakura Finetek USA, Inc., Torrance; and §Agilent worldwide face challenges in gathering enough tissues and Technologies, Santa Clara, CA. cases for antigens with rare occurrence in the population or R.L.L.: royalty payments, College of American Pathology Press; from patients with a specific rare disease to perform required Employee, Roche Tissue Diagnostics, Tucson, AZ; and Member, verification and validation studies for new laboratory products CAP Histotechnology Committee. P.V.R. and E.A.S.: are employees at Roche Tissue Diagnostics, Tucson, AZ. K.A.W.: Employee, and immunohistochemical (IHC) tests. Procurement of bio- Leica Biosystems/Danaher, Inc.; Stockholder, Danaher, Inc.; and logical material for these biomarkers is technology-agnostic Stockholder, Novartis AG. E.E.W.: Employee and shareholder, and not unique to IHC: the same challenges exist for molec- F. Hoffmann-La Roche AG. G.K.: Employee, Sakura Finetek ular testing. Finding and procuring biospecimens from rare USA Inc., Torrance, CA. B.P.: Employee and shareholder, Agilent cancersisalsodifficult. When combining a low-frequency Technologies, Santa Clara, CA; Member, CAP Molecular Oncology Committee. occurrence biomarker in a rare cancer, acquiring the required Reprints: Robert L. Lott, BS, HTL(ASCP), Roche Tissue Diagnostics, number of positive samples to properly perform verification 1910 E Innovation Park Drive, Tucson, AZ 85755 (e-mail: robert. and validation seems daunting. For reference, the College of lott@roche.com). American Pathologists (CAP) currently recommends for each Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. This is an open access article distributed under the terms of the predictive test that a laboratory acquire 20 positive and 20 Creative Commons Attribution-Non Commercial-No Derivatives negative cases for verification and validation. License 4.0 (CCBY-NC-ND), where it is permissible to download and In a recent CAP survey publication that assessed share the work provided it is properly cited. The work cannot be changes in IHC analytic validation practices following the changed in any way or used commercially without permission from the journal. publication of an evidence-based laboratory practice guideline, Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 www.appliedimmunohist.com | 327 Lott et al Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 the 2 biggest challenges in implementing the guideline was MSH6. Trying to find enough positive and negative cases difficulty in finding validation cases for rare antigens, and the for 1 or all 4 biomarkers to validate for MMR proteins by other was time/staff resource limitations. IHC can be difficult, as only 1 in 9 (11%) colorectal car- For clarification, verification is defined as the process cinomas contain the deleted MLH1 gene and only 1 in 50 by which a laboratory determines that an IVD assay per- (2%) contain the deleted MSH2 gene. Thus, a laboratory forms according to the specifications set forth by the man- would need to test potentially ≥ 180 unselected cancers to ufacturer. Validation is defined as the process by which a identify 20 MLH1 gene deleted cases, and perhaps 1000 laboratory confirms a laboratory developed test (LDT) or cancers to find 20 MSH2 gene deleted cases. modified Food and Drug Administration (FDA)-cleared/ The V600E activating mutation in BRAF and its de- approved test performs as intended or claimed. tection by IHC may or may not be rare depending on the assay application. It has been detected in ~7% of all solid tumors, including 45% of papillary thyroid carcinomas, 40% BACKGROUND to 60% melanomas, 35% of serous ovarian carcinomas, but Historically, IHC laboratories have experienced diffi- only 5% to 15% of colorectal adenocarcinomas, and only 1% culty in finding enough cases for verification and/or vali- to 3% of lung and other cancers. dation of new tests that use antibodies against specific Several IHC markers are useful for the identification of markers with a low frequency of occurrence. For example, spindle cell tumors and soft tissue tumors. For example, ALK+ positive lymphoma, a type of anaplastic large cell TLE-1 is a sensitive and specific marker for synovial sarco- lymphoma (ALCL), is considered “rare” because it com- ma, yet the incidence is estimated to be only 900 new cases a prises only 2% to 3% of all lymphomas in adults and 10% to year in the United States. The antibody INI-1 is excellent 30% of lymphomas in children. Only about half (50% to for the identification of malignant rhabdoid tumors, and 60%) of those show positivity for the ALK fusion protein Wilms tumors, typically found in young children. However, by IHC (Fig. 1). Finding a single positive case can be the incidence is ~0.19 per million for renal tumors, 0.89 per challenging. In addition, choice of primary antibodies and million for atypical teratoid/rhabdoid tumor (AT/RT), and protocol parameters for the validation of ALK+ lymphomas 0.32 per million for tumors of other sites. Only 650 cases of mayvarysignificantly than those of ALK+ lung carcinomas. Wilms tumor is diagnosed in the United States each year. Lynch syndrome-associated cancers are charac- Soft tissue sarcomas (STS) comprise about 7% of all terized by Mismatch Repair (MMR) deficiency caused by malignancies in children and adolescents under the age of somatic inactivation of the sole remaining wild-type 20 years and rhabdomyosarcoma (RMS) accounts for MMR gene allele. The central step in diagnosing Lynch about 40% of pediatric soft tissue sarcomas. Antibodies syndrome is the analysis of patient tumor tissue, usually used for identification of RMS, such as Myo-D1 and colon or endometrial adenocarcinoma, for the presence of myogenin, are very difficult to validate because the in- MMR deficiency by IHC for MLH1, PMS2, MSH1, and cidence of RMS is only 4.5 cases/million children and adolescents per year. Each year, ~4000 to 5000 adults in the United States are diagnosed with a gastrointestinal stromal tumor (GIST). CD117 (c-kit) and DOG-1 performed by IHC have proven to be reliable and sensitive diagnostic mark- 9,10 ers for detection of GIST. However, GIST accounts for 9,10 <1% of all gastrointestinal tumors. Overall, about 85% of GISTs are reported to have a protein activating mu- tation in KIT or PDGFRA. Studies of PD-L1 expression in lung cancer report this biomarker to be positive in 13% to 70% of cases, partly dependent on how expression is tested for and defined. However, a much smaller number of “positive” cases will meet certain percent-positive cutoffs (1%, 5%, 25%, 50% etc) (Table 1). Patients with non–small cell lung cancer and with anaplastic lymphoma kinase (ALK) gene rearrangements constitute only 4% to 5% of all non–small cell lung cancer 13,14 patients. To introduce and properly validate new tests in the clinical laboratory, the College of American Pathologists (CAP) has published comprehensive guidelines for the verification and validation of IHC tests introduced into the laboratory menu. Note that CAP recommends differ- ent numbers of cases for nonpredictive versus predictive FIGURE 1. ALK protein (+) lymphoma (ALCL) IHC. ALCL indicates anaplastic large cell lymphoma; IHC, immunohistochemical. markers; laboratory verification studies require twice as 328 www.appliedimmunohist.com Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 Low-incidence IHC Validation TABLE 1. PD-L1 Prevalence by Immunohistochemical No. PD-L1 Positive Immune Cell % of IC Cases No. PD-L1 Positive Tumor Cell % of TC Cases Indications n (+) (> 5%) Cases (> 5%) (+) (> 5%) Cases (> 5%) Non–small cell lung 184 48 26 44 24 Cancer Renal cell cancer 88 22 25 9 10 Melanoma 58 21 36 5 3 Head/Neck SCCa 101 28 28 19 19 Gastric cancer 141 25 18 7 5 Colorectal cancer 77 27 35 1 1 Pancreatic cancer 83 10 12 3 4 IC indicates immune cell; SCCa, squamous cell carcinoma; TC, tumor cell. many positive and negative cases for FDA-approved is appropriate for the test’s intended use. Its requirements of predictive assays. If a laboratory chooses to develop their 20 positive and 20 negative cases are descriptive of a strictly own predictive marker assay, such as CD117 or PD-L1, “analytic” validation process. It provides no specifications using a LDT, or chooses to alter an FDA-approved assay, for tissues and tumors or even a range of epitope expression. even more rigorous verification analyses are required. However, it may not be entirely fit for purpose for validation There are specific differences in verification requirements of a predictive biomarker where a more “clinical” type of between a laboratory developing LDTs and using an validation process, designed with tissue specimens that test in vitro diagnostic (IVD) tests from industry, especially the thresholds or cutoffs (eg, <1%, <10%, etc) used in clinical with predictive marker assays. Each laboratory must de- trials or in a comparator validated assay, is required. cide the best option, considering parameters like costs (eg, So rather than just positive and negative, this type of of personnel, material), development time, and skills of clinical validation tests the diagnostic sensitivity and personnel. In some laboratories, neither of these require- specificity of the assay. For comparison, at least 50 pos- ments are practical or feasible, so laboratories have im- itive and 50 negative samples are recommended by the plemented alternative strategies in an attempt to be CLSI EP12-A2 User Protocol for Evaluation of Qual- compliant with these guidelines. The CAP guidelines are itative Test Performance. briefly summarized in Table 2. Recent publications by CAP sponsored surveys for Estrogen receptor, progesterone receptor, and HER-2/ practices in IHC testing laboratories reveal that most labo- neu testing for breast cancer have unique, specific validation ratories are not using the recommended sample sizes for 16,17 guidelines. Other global quality organizations have validation as described in Table 2. In 2 studies for initial adopted similar validation guidelines, especially with regard validation, 75.4% (538 of 714) of laboratories adopted the 18–21 to overall concordance. The CAP guideline recom- 20-case minimum for nonpredictive markers but less than mendations, as described, help ensure that the IHC test ac- half (45.9% of the 579 responding laboratories) adopted the curately measures the analyte of interest and that validation 40-case minimum for predictive markers as outlined in the TABLE 2. Guidelines for Introducing a New In Vitro Diagnostic (IVD) Antibody into the Laboratory Step #1 Optimization Use tissues indicated for the specific intended use or clinical application. For IVD antibodies, the vendor’s package insert protocol should be used as a starting point. For other tests, optimization steps may include the testing of antigen retrieval, 1 degree Ab. titration, detection system, chromogen, amplification, counterstaining, etc Step #2 Verification Use the optimized protocol from above: For nonpredictive assays, laboratories should run a minimum of 10 positive and 10 negative cases For FDA-approved predictive assays, laboratories should run a minimum of 20 positive and 20 negative/low-expressor cases Labs choosing to alter FDA-approved kits or develop their own predictive marker assay(s) (LDTs), should run a minimum of 40 positive and 40 low-expressor/negative cases (0, 1+). This includes LDTs for ER, PR, and HER2 Laboratories must verify new IHC tests before placing them into clinical service Step #3 Validation After verification testing is complete, those slides are compared/correlated (validated) in at least one of the following ways: Compare the new test results with a prior validated assay/test using the same tissue set Compare the new test results with validated results from another laboratory using the same tissue set Compare the new test with an alternate validated non-IHC test (ISH, etc) Compare the new test with the morphology and expected results (ie, from a thorough literature search) Testing is compared with graded results from formal Proficiency Testing challenges For validation, every assay must achieve 90% overall concordance between the new test and the comparator test All tissues/cases in this process should be fixed and processed the same as cases tested clinically. Laboratories may use whole sections, tissue microarrays, multitissue blocks as appropriate. Tissues containing low-incidence antigens may be difficult to find. The characteristics of the tissues or cases used for validation should be similar to those seen in the laboratory’s patient population. Tissues should include relevant normal tissues, if available, and neoplasias which span a range of expression from negative to low to high. FDA indicates Food and Drug Administration; IHC, immunohistochemical; LDT, laboratory developed test. Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. www.appliedimmunohist.com 329 Lott et al Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 2014 evidence-based laboratory practice guideline. As may suffice to cover a group of similar intended uses, as reported by 714 laboratories, the median number of cases long as the overall process of preparation and inter- required for nonpredictive assays was 20, and the median pretation is the same. number of cases required for predictive assays was only 25, Interestingly, recent CAP proficiency testing survey well below the guideline recommendation of 40. results on validation and implementation of Next Gen- Recent review of the FDA IHC reagent classifications, eration Sequencing (NGS) in clinical oncology applica- IHC validation protocols, and quality assurance (QA) proce- tions showed that about half of the 58 laboratories that dures and documentation needed in conjunction with IHC test reported on their variant confirmation procedures are not guidelines for Mohs Labs, IHC reagents are subject to several performing variant confirmation (n = 28; 48%). For those histologic and clinical controls to maximize the accuracy of laboratories that are performing variant confirmation, IHC reagents, with it being at the discretion of laboratory they are employing various orthogonal technologies in- medical director to establish these quality control procedures cluding Sanger sequencing (n = 26), targeted mutation before offering these tests for patient diagnostics. testing (n = 11), pyrosequencing (n = 8), or other tech- In a survey of over 700 laboratories, to establish baseline nologies (n = 5). parameters for IHC validation procedures and practice and to assess feasibility of implementation, Hardy and colleagues, SOLUTIONS noted that for non-FDA approved, nonpredictive IHC tests, There are many IHC tests available for rare, low- 75% of laboratories used 21 or fewer total cases for validation. incidence antigens and tissues. The increasingly demand- Less than 50% of laboratories used ≥25 cases to validate non– ing requirements for verifying or validating new antibody FDA-approved predictive markers (LDTs). Of those, only 47% tests, while necessary, have made it difficult for labo- used cases with any weak to moderate expression levels. When ratories wanting to perform in-house testing to identify the comparing IHC HER2/neu assays with an IHC test performed required number of cases and needed amount of tissue in another laboratory, only 56% of laboratories used a especially those tissues expressing rare antigens. CAP recommended minimum of 25 cases. Seventy-five percent of recognizes this fact and has addressed it in its current respondents had validated the most recently introduced new Anatomic Pathology accreditation checklist antibody, with a median of only 15 cases. Major challenges to meet the proper validation requirements included tissue ANP.22750—Antibody Validation availability (especially for cytological materials), availability of appropriate tissue controls which were highly dependent upon The extent of this validation is at the discretion of the particular antibody used, and time and personnel resources the medical director. If a laboratory does not use 10 to find and qualify appropriate cases. Most respondents felt a positive cases and 10 negative cases for their nonpredictive minimum of 25 cases was appropriate for validation of a pre- marker validation, then the rationale for this needs to be dictive marker and 13 cases for nonpredictive markers as pre- documented in the validation summary. Just because a dictive markers are much less likely to have morphologic clues laboratory uses less than the suggested number of cases for to refute an errant stain a validation does not necessarily mean that they are out of Considering another diagnostic procedure, Whole compliance. Slide Imaging (WSI), CAP has established an evidence- based guideline setting the minimum number of cases to …the laboratory director determines that fewer validation 60 after finding that when an average of 20 cases (range, cases are sufficient for a specific marker (e.g. a rare antigen 10 to 46 cases) was used to compare WSI and glass slide, or tissue), the rationale for that decision needs to be there was significantly lower accuracy (77%) and con- recorded. cordance (75%), but when an average of 60 cases (range, 52 to 90 cases) was used, the studies showed improved CAP does require that ER, PR and HER2 assays accuracy (90%) and far better concordance (95%). Thus, employed as predictive markers on breast carcinoma have a CAP recommended that validation studies for WSI in- minimum of 20 positive cases and 20 negative cases. All clude at least 60 routine cases per application, [eg, hem- other predictive markers are to bevalidated,but theextentof atoxylin eosin (H&E)-stained sections of fixed tissue, that validation is, again, at the discretion of the medical frozen sections, cytology, hematology] that reflects the director and will vary with the antibody, especially those spectrum and complexity of specimen types (biopsies and against low-incidence antigens and rare, hard to find tissue resections) and diagnoses likely to be encountered during samples. If IHC is regularly performed on specimens that are routine practice. If the laboratory intends to use its vali- not fixed or processed in the same manner as the tissues used dated WSI system for another supplemental application for validation (eg, alcohol fixed cell blocks, cytological (eg, to evaluate special stains, IHC or fluorescence stains smears, formalin post fixed tissue, or decalcified tissue), the and H&E-stained sections), then another 20 cases of the laboratory should test a sufficient number of such tissues to “additional” application should be validated. In addition, ensure that assays consistently achieve expected results using the study showed that the specimen preparation type the appropriate protocol for the intended use. Again, the (H&E, Papanicolaou, etc) was a more important perfor- laboratory director is responsible for determining the num- mance variable than the source of the tissue or the specific ber of positive and negative cases and the number of pre- analyte being assessed. Thus, a single validation study dictive and nonpredictive markers to test. 330 www.appliedimmunohist.com Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 Low-incidence IHC Validation There are several solutions available to laboratories sections should be used if TMAs/ MTBs are not appro- wanting to perform verification and/or validation studies priate for the targeted antigen or if the laboratory medical in-house for low-frequency antigens, including: director cannot confirm that the fixation and processing of Using specimen blocks identified via retrospective and TMAs/MTBs is similar to clinical specimens. MTBs prospective LIS search/survey from “in-house” cases. should be employed for external positive and negative Purchase of tissue blocks, multitissue blocks (MTBs), or tissue controls. Use of multitissue external positive con- precut slides from qualified sources, or creating trols by IHC technologists provides consistent positive homemade MTBs. feedback because it simplifies workflow and minimizes Participation in External Quality Assessment (EQA) potential error. TMA external positive controls have also proficiency programs. demonstrated superior performance in staining con- Correlating the new test’s results with expected results sistency and reproducibility. based on literature search, demonstrating same tumor Several qualified vendors offer both TMA slides morphology and antigen distribution. and blocks for purchase in a variety of formats including formalin-fixed, paraffin-embedded tissues (Table 3). These multicore arrays, some of which contain 300+ cores, Retrospective Search include normal, malignant, or metastatic tissues and are A common, almost routine, activity in most aca- suitable for use in validation studies. Multicore TMAs demic medical center, reference, community hospital, and may contain protein, RNA, or DNA and provide an ideal private laboratories is performing a keyword search(s) to high throughput method for rapid analysis of IHC identify in-house, archived cases using the laboratory in- markers or in situ hybridization. Many companies can formation system (LIS). Most systems are able to print or provide custom tissue array constructions for use in save list(s) of cases that can be used to identify and/or immunohistochemistry and in situ hybridization. In screen cases and blocks with low-incidence antigens. addition, vendors can often provide relevant preanalytic As an example, a keyword search for “Anaplastic variables such as fixation times for each tissue. Cost of large cell lymphoma (ALCL),” an aggressive type of non- these products varies considerably and depends upon Hodgkin T-cell lymphoma, may reveal only a few cases in factors such as sample type, availability, size of sample set, the block archive because of its rarity. However, uti- and antigen presentation. lization of in-house, archived tissue blocks, even if rare, Several vendors are able to generate engineered cell best fits the CAP checklist instruction that “laboratories lines with low/negative, intermediate, and high levels of should use validation tissues that have been processed protein using proprietary gene-editing platforms (Fig. 2). using the same fixative and processing methods as cases The raw material is then formalin-fixed, processed in a that will be tested clinically.” traditional manner using alcohol and xylene, and paraffin embedded. These cell buttons, which mimic cytology cell blocks, can be used as reference standards that serve as Purchase surrogates for real patient samples within an IHC testing In a recent review of diagnostic IHC standardization protocol. They can also be used as a renewable and procedures by Lin and Chen, tissue microarray (TMA) consistent point of reference, essentially run controls, blocks containing a grid-like array of various tumors and/ when optimizing and monitoring performance of an IHC or normal tissues have proved to be extremely valuable in assay. biomarker research and have demonstrated great utility in clinical IHC laboratories. At least 4 different prototypes of Multitissue Blocks TMA blocks can be used: Homemade MTBs represent an alternative to whole (1) TMA block containing a broad spectrum of tumors sections or TMAs. MTBs containing just 8 to 12 different and/or normal tissues from various organs, useful for cases of specific tissue-types can be used to supplement screening a new biomarker. (2) TMA block containing 50 to 100 tumors with a specific diagnosis, such as lung adenocarcinoma, TABLE 3. Vendors That Provide TMA Blocks and/or Slides useful for antibody validation, revalidation, estimating Cooperative Human Tissue http://chtn.sites.virginia.edu/tissue- marker prevalence in a specific indication, and in Network microarrays determining the diagnostic sensitivity and specificity of Folio Biosciences http://www.foliobio.com/products/ a newly discovered antibody. tissue-microarrays (3) TMA block containing 5 to 10 cases of a specific type Histocyte Laboratories http://www.histocyte.com Horizon Discovery http://www.horizondiscovery.com of tumor, useful for antibody testing and optimization. Invitrogen http://www.invitrogen.com (4) TMA block containing 5 to 10 cases of selected, mixed Origene http://www.origene.com/tissue/tissue_ tumors and/or normal tissues from various organs, microarrays.aspx useful as external positive and negative control tissues Pantomics http://www.pantomics.com StatLab Medical Products http://www.statlab.com for each antibody. US Biolab http://www.usbiolab.com/custom-tissue- array Laboratories may use whole sections, TMAs, and/or US Biomax http://www.biomax.us/tissue-arrays MTBs in their validation sets as appropriate. Whole Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. www.appliedimmunohist.com 331 Lott et al Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 FIGURE 2. IHC staining of engineered cell lines that express PD-L1 at varying levels, from high (left) to low (right). IHC indicates immunohistochemical. existing in-house tissues for optimization and verification Participation in such programs provides appropriate work, comparison and parallel testing of new antibody support for initial and continuous revalidation of these lots, reverification of assay modifications, and positive and testsaswell. negative tissue control material. The Canadian Immunohistochemistry Quality The concept involves some up-front work on the Control (cIQc) EQA scheme regularly conducts EQA ac- part of laboratory personnel to identify tissues and as- tivities for rare antigen, hard-to-find tissue specimens such semble the MTBs. The blocks consist mainly of normal as (ie, ALK, IDH1, BRAF V600E, PD-L1, MMR) using tissue types easily found in typical pathology labo- medium-to-large multiple-core tissue micro-array slides ratories. The best MTBs are those put together in the (TMAs) (Fig. 4). user laboratory using tissues fixed and processed in that Large TMAs containing multiple cores of positive laboratory. They contain both positive and negative in- and negative tissues processed by standard methods allow ternal controls for a wide range of IHC tests. A single a laboratory to supplement in-house tissue numbers to MTB paraffinblock canbeusedfor verification studies reach the number of cases needed for verification and/or in over 100 routine IHC tests and are useful in supple- validation. They also help improve the precision and ac- menting rare antigen and hard to find tissue searches. As curacy of testing. We presume that any global EQA pro- seen in Figure 3, for example, this simple block consists gram that provides such rare antigen samples would be of: acceptable for test verification and validation. (1) 2 to 3 full cross sections of appendix from different cases The CAP MMR survey for DNA MMR proteins (2) 2 to 3 small-medium sized pieces of liver resection, provides fewer cores per slide, but 2 unique TMA slides including normal liver from different cases rather than just 1 (Fig. 5). TMA slides used for EQA (3) 2 to 3 pieces of tonsil resection, to include squamous assessment are often available for purchase separately epithelium and lymphatic nodules from different cases from the scheme as an aid to laboratories. (4) 2 to 3 sections of normal pancreas containing islets, Participation in EQA programs that do not pro- ducts, and acinar cells from different cases. vide sufficient numbers of tissue samples may be in- formative and useful, but do not provide the opportunity The MTB contains 8 to 12 different patient tissues to calculate concordance rates with adequate power and can be used for a wide variety of verification and unless supplemented with in-house samples, and there- validation activities. Similarly, if sections of normal skin, fore alone do not provide sufficient numbers of cases for melanoma, striated muscle, and brain (cerebellum) are verification or validation. Participation in EQA how- combined in a MTB another 25 to 30 IHC markers can ever, informs the laboratory of the sensitivity and spec- easily be used to verify and validate. ificity of their protocol and their platform choices and serves as an indicator of the need for overall opti- External Quality Assessment mization of the IHC assay. Incidentally, it is essential Internally, if hospitals and reference laboratories that laboratories performing Class II and/or Class III find it difficult to acquire sufficient numbers of tissue IHC testing document participation in EQA programs in samples for initial validation, some EQA programs may the laboratory Quality Management Plan (QMP), and if provide test samples that quantitatively and qual- suboptimal results are achieved, corrective actions are itatively support IHC test verification and validation. taken and documented. 332 www.appliedimmunohist.com Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 Low-incidence IHC Validation FIGURE 3. Slide made from a multitissue block consisting mainly of normal tissue types. 2 to 3 full cross sections of appendix from different cases. 2 to 3 small-medium sized pieces of liver resection, including normal liver from different cases. 2 to 3 pieces of tonsil resection, to include squamous epithelium and lymphatic nodules from different cases. 2 to 3 sections of normal pancreas containing islets, ducts, and acinar FIGURE 4. BRAF V600E TMA slide from cIQc. TMA indicates cells from different cases. MTB indicates multitissue block. tissue microarray. Use of Expected Staining Results laboratories for 725 nonpredictive markers (61%) and 101 Verification activities do not always have to involve predictive markers (46.5%). Comparing the new test stained slides being compared with previously stained results with the same tissue set stained in another labo- slides or slides stained in another laboratory, or results ratory was a distant second used by only 123 (17%) of being judged comparable to an alternative method (ie, laboratories in the survey. FISH) to validate or revalidate a new test, especially if staining is for a low-incidence antigen. An alternative DISCUSSION approach is a thorough literature search, guided by the pathologist(s), to reveal “both” the expected morphology Many laboratories have problems finding sufficient of the tumor(s) under study and the correlation of staining tissues and cases to adequately validate antibodies against patterns to the “expected results” with the new test. rare antigens with a low frequency of occurrence and little Results from a survey of 1085 clinical laboratories, or no expression in normal tissues. There are several comparing and correlating the staining results of the new practical solutions available to clinical laboratories. These test with expected results, regardless of specimen size, was include developing internal resources with retrospective the “most common” validation method used among and prospective search(s) of archival material, purchase of Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc. www.appliedimmunohist.com 333 Lott et al Appl Immunohistochem Mol Morphol Volume 29, Number 5, May/June 2021 11. Kerr KM, Tsao MS, Nicholson AG, et al. Programmed death-ligand 1 immunohistochemistry in lung cancer: in what state is this art? J Thorac Oncol. 2015;10:985–989. 12. Herbst RS, Soria JC, Kowanetz M, et al. Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature. 2014;515:563–567. 13. Chia PL,MitchellP,DobrovicA,etal. Prevalence and natural history of ALK positive non-small-cell lung cancer and the clinical impact of targeted therapy with ALK inhibitors. Clin Epidemiol. 2014;6:423–432. 14. Cheung CC, Garratt J, Won J, et al. Developing ALK immunohis- tochemistry and in situ hybridization proficiency testing for non- small cell lung cancer in Canada: Canadian immunohistochemistry quality control challenges and successes. Appl Immunohistochem Mol Morphol. 2015;23:677–681. 15. Fitzgibbons PL, Bradley LA, Fatheree LA, et al. Principles of analytic validation of immunohistochemical assays: guideline from the College of American Pathologists Pathology and Laboratory Quality Center. Arch Pathol Lab Med. 2014;138:1432–1443. 16. Hammond ME, Hayes DF, Dowsett M, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommen- dations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer. Arch Pathol Lab Med. 2010;134:907–922. 17. Wolff AC, Hammond ME, Schwarz JN, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. J Clin Oncol. 2007;25:118–145. 18. Hanna W, O’Malley FP, Barnes P, et al. Updated recommendations from the Canadian National Consensus Meeting on HER2/neu testing in breast cancer. Curr Oncol. 2007;14:149–153. 19. Torlakovic EE, Riddell R, Banerjee D, et al. Canadian Association of Pathologists-Association canadienne des pathologistes National Standards Committee/Immunohistochemistry: best practice recom- mendations for standardization of immunohistochemistry tests. Am J FIGURE 5. MMR TMA slides from CAP. CAP indicates College Clin Pathol. 2010;133:354–365. of American Pathologists; MMR, Mismatch Repair; TMA, tissue 20. Ellis IO, Bartlett J, Dowsett M, et al. Best Practice No 176: updated microarray. recommendations for HER2 testing in the UK. J Clin Pathol. 2004;57:233–237. 21. Bilous M, Dowsett M, Hannah W, et al. Current perspectives on TMA blocks and slides from outside resources, and par- HER2 testing: a review of National Testing Guidelines. Mod Pathol. ticipation in EQA schemes, and comparison with pub- 2003;16:173–182. lished literature. These solutions may be used uniquely 22. Torlakovic E. How to validate predictive immunohistochemistry and in combination to achieve verification and validation testing in pathology? Arch Pathol Lab Med. 2019;143:907. 23. Garrett PE, Lasky FD, Meier KL, et al. Clinical and Laboratory requirements for new and ongoing IHC tests. Standards Institute User Protocol for Evaluation of Qualitative Test Performance: Approved Guideline, 2nd ed. Wayne, PA: Clinical and REFERENCES Laboratory Standards Institute; 2008. 1. Fitzgibbons PL, Goldsmith JD, Souers RJ, et al. Analytic validation 24. Stuart LN, Volmar KE, Nowak JA, et al. Analytic validation of of immunohistochemical assays: a comparison of laboratory immunohistochemistry assays: New Benchmark Data from a survey practices before and after introduction of an evidence-based guide- of 1085 laboratories. Arch Pathol Lab Med. 2017;141:1255–1261. line. Arch Pathol Lab Med. 2017;141:1247–1254. 25. Kinas CG, Carroll BT. General guidelines for quality assurance of 2. College of American Pathologists Checklist, Anatomic Pathology, v. immunohistochemistry in a Mohs lab. Dermatol Surg. 2017;43:507–511. 07-29-2013, Definition of Terms. 26. Hardy LB, Fitzgibbons PL, Goldsmith JD, et al. Immunohistochemistry 3. Vose JM. Getting the Facts: Anaplastic Large Cell Lymphoma. New validation procedures and practices: a College of American Pathologists York, NY: Lymphoma Research Foundation; 2009:1–2. survey of 727 laboratories. Arch Pathol Lab Med. 2013;137:19–25. 4. Jasperson KW, Tuohy TM, Neklason DW, et al. Hereditary and 27. Pantanowitz L, Sinard JH, Hendricks WH, et al. Validating whole familial colon cancer. Gastroenterology. 2010;138:2044–2058. slide imaging for diagnostic purposes in pathology: guideline from 5. Pakneshan S, Salajegheh A, Smith RA, et al. Clinicopathological relevance the College of American Pathologists Pathology and Laboratory of BRAF mutations in human cancer. Pathology. 2013;45:346–356. Quality Center. Arch Pathol Lab Med. 2013;137:1710–1722. 6. Cure Sarcoma. Synovial sarcoma. 2019. Available at: www. 28. Nagarajan R, Bartley AN, Bridge JA, et al. A window into clinical curesarcoma.org/patient-resources/sarcoma-subtypes/synovial-sarcoma/. next-generation sequencing-based oncology testing practices. Arch Accessed February 22, 2017. Pathol Lab Med. 2017;141:1679–1685; 2017. 7. Annual incidence of rhabdoid tumors among children <15, 2000-2008. 29. College of American Pathologists, 2017 Anatomic Pathology Surveillance, Epidemiology, and End Results (SEER) Program; Bethesda, Checklist, ANP.22750 – Antibody Validation. MD: 2011. Volume SEER 17. SEER*Stat version 7.0.6. 30. Lin F, Chen Z. Standardization of diagnostic immunohistochemis- 8. Gurney JG, Young JL, Roffers SD, et al SEER Pediatric try: literature review and geisinger experience. Arch Pathol Lab Med. Monograph. National Cancer Institute; 2005. Soft Tissue Sarcomas. 2014;138:1564–1577. 9. Rubin BP, Heinrich MC, Corless CL. Gastrointestinal stromal 31. Torlakovic E, Nielsen S, Vyberg M, et al. Getting controls under tumour. Lancet. 2007;369:1731–1741. control: the time is now for immunohistochemistry. J Clin Pathol. 10. Steigen SE, Eide TJ. Gastrointestinal stromal tumors (GISTs): 2015;68:879–882. a review. APMIS. 2009;117:73–86. 334 www.appliedimmunohist.com Copyright © 2020 The Author(s). Published by Wolters Kluwer Health, Inc.

Journal

Applied Immunohistochemistry – Wolters Kluwer Health

Published: May 21, 2021

There are no references for this article.