Quantitative comparison of immunohistochemical HER2-low detection in an interlaboratory study.

AIMS: Recently, human epidermal growth factor 2 (HER2)-low (i.e. HER2 score 1+ or 2+ without amplification) breast cancer patients became eligible for trastuzumab-deruxtecan treatment. To improve assay standardisation and detection of HER2-low in a quantitative manner, we conducted an external quality assessment-like study in the Netherlands. Dynamic range cell lines and immunohistochemistry (IHC) calibrators were used to quantify HER2 expression and to assess interlaboratory variability. METHODS AND RESULTS: Three blank slides with a dynamic range cell line and an IHC calibrator were stained with routine HER2 assays by 35 laboratories. Four different antibody clones were used: 19 (54.3%) 4B5, six (17.1%) A0485, five (14.3%) DG44 (HercepTest) and five (14.3%) SP3. Laboratories used two different detection kits for 4B5 assays: 14 (73.7%) ultraView and five (26.3%) OptiView. Variability of HER2 expression in cell lines, measured with artificial intelligence software, was median (min-max) = negative core 0.5% (0.0-57.0), 1+ core 4.3% (1.6-71.3), 2+ core 42.8% (30.4-92.6) and 3+ core 96.2% (91.8-98.8). The calibrators DG44 and 4B5 OptiView had the highest analytical sensitivity, closely followed by 4B5 ultraView. SP3 was the least sensitive. Calibrators of A0485 assays were not analysable due to background staining. CONCLUSIONS: As assays were validated for detecting HER2-amplified tumours, not all assays and antibodies proved suitable for HER2-low detection. Some tests showed distinct expression in the negative cell line. Dynamic range cell line controls and quantitative analysis using calibrators demonstrated more interlaboratory variability than commonly appreciated. Revalidation of HER2 tests by laboratories is needed to ensure clinical applicable HER2-low assays.

Quantitative comparison of PD-L1 IHC assays against NIST standard reference material 1934.

Companion diagnostic immunohistochemistry (IHC) tests are developed and performed without incorporating the tools and principles of laboratory metrology. Basic analytic assay parameters such as lower limit of detection (LOD) and dynamic range are unknown to both assay developers and end users. We solved this problem by developing completely new tools for IHC-calibrators with units of measure traceable to National Institute of Standards & Technology (NIST) Standard Reference Material (SRM) 1934. In this study, we demonstrate the clinical impact and opportunity for incorporating these changes into PD-L1 testing. Forty-one laboratories in North America and Europe were surveyed with newly-developed PD-L1 calibrators. The survey sampled a broad representation of commercial and laboratory-developed tests (LDTs). Using the PD-L1 calibrators, we quantified analytic test parameters that were previously only inferred indirectly after large clinical studies. The data show that the four FDA-cleared PD-L1 assays represent three different levels of analytic sensitivity. The new analytic sensitivity data explain why some patients' tissue samples were positive by one assay and negative by another. The outcome depends on the assay's lower LOD. Also, why previous attempts to harmonize certain PD-L1 assays were unsuccessful; the assays' dynamic ranges were too disparate and did not overlap. PD-L1 assay calibration also clarifies the exact performance characteristics of LDTs relative to FDA-cleared commercial assays. Some LDTs' analytic response curves are indistinguishable from their predicate FDA-cleared assay. IHC assay calibration represents an important transition for companion diagnostic testing. The new tools will improve patient treatment stratification, test harmonization, and foster accuracy as tests transition from clinical trials to broad clinical use.

Development and Validation of Measurement Traceability for In Situ Immunoassays.

BACKGROUND: Immunoassays for protein analytes measured in situ support a $2 billion laboratory testing industry that suffers from significant interlaboratory disparities, affecting patient treatment. The root cause is that immunohistochemical testing lacks the generally accepted tools for analytic standardization, including reference standards and traceable units of measure. Until now, the creation of these tools has represented an insoluble technical hurdle. METHODS: We address the need with a new concept in metrology-that is, linked traceability. Rather than calculating analyte concentration directly, which has proven too variable, we calculate concentration by measuring an attached fluorescein, traceable to NIST Standard Reference Material 1934, a fluorescein standard. RESULTS: For validation, newly developed estrogen receptor (ER) calibrators were deployed in tandem with an array of 80 breast cancer tissue sections in a national external quality assessment program. Laboratory performance was assessed using both the ER standards and the tissue array. Similar to previous studies, the tissue array revealed substantial discrepancies in ER test results among the participating laboratories. The new ER calibrators revealed a broad range of analytic sensitivity, with the lower limits of detection ranging from 7310 to 74 790 molecules of ER. The data demonstrate, for the first time, that the variable test results correlate with analytic sensitivity, which can now be measured quantitatively. CONCLUSIONS: The reference standard enables precise interlaboratory alignment of immunohistochemistry test sensitivity for measuring cellular proteins in situ. The introduction of a reference standard and traceable units of measure for protein expression marks an important milestone.

Validation of Linear Range HER2/Estrogen Receptor/Progesterone Receptor IHControls for Daily Quality Assurance.

OBJECTIVES: To evaluate a new US Food and Drug Administration (FDA)-cleared immunohistochemistry (IHC) control (IHControls [Boston Cell Standards]) comprising peptide epitopes for HER2, estrogen receptor (ER), and progesterone receptor (PR) attached to cell-sized microspheres and to compare its performance against conventional tissue controls. METHODS: IHControls and tissue/cell line controls for HER2, ER, and PR were compared side by side daily at 5 clinical IHC laboratories for 1 to 2 months. Separately, the sensitivity of the 2 types of controls was evaluated in simulated IHC assay failure experiments by diluting the primary antibody. Additional evaluations included lot-to-lot manufacturing reproducibility of 3 independent lots and specificity against 26 antigenically irrelevant IHC stains. RESULTS: Side-by-side testing revealed a 99.6% concordance between IHControls and tissue controls across 5 IHC laboratories and 766 individual evaluations. Three discordant quality control events were the result of operator error. Simulated assay failure data showed that both IHControls and tissue controls are similarly capable of detecting IHC staining errors. Manufacturing reproducibility of IHControls showed less than 10% variability (coefficient of variation). No cross-reactions were detected from 26 antigenically irrelevant IHC stains. CONCLUSIONS: IHControls, the first FDA-cleared IHC controls, can sensitively and accurately detect IHC assay problems, similar to tissue controls.

Digital Image Analysis and Quantitative Bead Standards in Root Cause Analysis of Immunohistochemical Staining Variability: A Real-world Example.

Assessment of automated immunohistochemical staining platform performance is largely limited to the visual evaluation of individual slides by trained personnel. Quantitative assessment of stain intensity is not typically performed. Here we describe our experience with 2 quantitative strategies that were instrumental in root cause investigations performed to identify the sources of suboptimal staining quality (decreased stain intensity and increased variability). In addition, these tools were utilized as adjuncts in validation of a new immunohistochemical staining instrument. The novel methods utilized in the investigation include quantitative assessment of whole slide images (WSI) and commercially available quantitative calibrators. Over the course of ~13 months, these methods helped to identify and verify correction of 2 sources of suboptimal staining. One root cause of suboptimal staining was insufficient/variable power delivery from our building's electrical circuit. This led us to use uninterruptible power managers for all automated immunostainer instruments, which restored expected stain intensity and consistency. Later, we encountered one instrument that, despite passing all vendor quality control checks and not showing error alerts was suspected of yielding suboptimal stain quality. WSI analysis and quantitative calibrators provided a clear evidence that proved critical in confirming the pathologists' visual impressions. This led to the replacement of the instrument, which was then validated using a combination of standard validation metrics supplemented by WSI analysis and quantitative calibrators. These root cause analyses document 2 variables that are critical in producing optimal immunohistochemical stain results and also provide real-world examples of how the application of quantitative tools to measure automated immunohistochemical stain output can provide a greater objectivity when assessing immunohistochemical stain quality.

Bioinformatic requirements for protein database searching using predicted epitopes from disease-associated antibodies.

We describe a new approach to identify proteins involved in disease pathogenesis. The technology, Epitope-Mediated Antigen Prediction (E-MAP), leverages the specificity of patients' immune responses to disease-relevant targets and requires no prior knowledge about the protein. E-MAP links pathologic antibodies of unknown specificity, isolated from patient sera, to their cognate antigens in the protein database. The E-MAP process first involves reconstruction of a predicted epitope using a peptide combinatorial library. We then search the protein database for closely matching amino acid sequences. Previously published attempts to identify unknown antibody targets in this manner have largely been unsuccessful for two reasons: 1) short predicted epitopes yield too many irrelevant matches from a database search and 2) the epitopes may not accurately represent the native antigen with sufficient fidelity. Using an in silico model, we demonstrate the critical threshold requirements for epitope length and epitope fidelity. We find that epitopes generally need to have at least seven amino acids, with an overall accuracy of >70% to the native protein, in order to correctly identify the protein in a nonredundant protein database search. We then confirmed these findings experimentally, using the predicted epitopes for four monoclonal antibodies. Since many predicted epitopes often fail to achieve the seven amino acid threshold, we demonstrate the efficacy of paired epitope searches. This is the first systematic analysis of the computational framework to make this approach viable, coupled with experimental validation.

Selecting an Optimal Positive IHC Control for Verifying Antigen Retrieval.

Positive immunohistochemistry (IHC) controls are intended to detect problems in both immunostaining and heat-induced epitope retrieval (HIER). However, it is not known what features in a control are important for verifying HIER. Contrary to expectation, the fact that a tissue is formalin-fixed does not necessarily render it suitable in verifying proper HIER. Some tissue controls, for some immunostains, strongly stain even without HIER. Consequently, the control may verify the immunostain but provide little or no information regarding the HIER step. To sort this out, we used formalin-fixed peptide epitopes, a model that provides for precise definition of analyte concentration, epitope composition, and degree of fixation. Our data demonstrate that formalin fixation generates a variable level of protein epitope masking, depending on the epitope recognized by the primary antibody. Some epitopes are highly masked while others hardly at all. Furthermore, the ability of amino acids in the epitope to react with formaldehyde can, at least in part, account for this variability. Most important, we demonstrate the importance of selecting a positive control with a low or intermediate analyte concentration (relative to the immunostain's analytic sensitivity). High analyte concentrations can be insensitive in verifying the HIER step.

Quantitative Assessment of Immunohistochemistry Laboratory Performance by Measuring Analytic Response Curves and Limits of Detection.

CONTEXT: - Numerous studies highlight interlaboratory performance variability in diagnostic immunohistochemistry (IHC) testing. Despite substantial improvements over the years, the inability to quantitatively and objectively assess immunostain sensitivity complicates interlaboratory standardization. OBJECTIVE: - To quantitatively and objectively assess the sensitivity of the immunohistochemical stains for human epidermal growth factor receptor type 2 (HER2), estrogen receptor (ER), and progesterone receptor (PR) across IHC laboratories in a proficiency testing format. We measure sensitivity with parameters that are new to the field of diagnostic IHC: analytic response curves and limits of detection. DESIGN: - Thirty-nine diagnostic IHC laboratories stained a set of 3 slides, one each for HER2, ER, and PR. Each slide incorporated a positive tissue section and IHControls at 5 different concentrations. The IHControls comprise cell-sized clear microbeads coated with defined concentrations of analyte (HER2, ER, and/or PR). The laboratories identified the limits of detection and then mailed the slides for quantitative assessment. RESULTS: - Each commercial immunostain demonstrated a characteristic analytic response curve, reflecting strong reproducibility among IHC laboratories using the same automation and reagents prepared per current Good Manufacturing Practices. However, when comparing different commercial vendors (using different reagents), the data reveal up to 100-fold differences in analytic sensitivity. For proficiency testing purposes, quantitative assessment using analytic response curves was superior to subjective interpretation of limits of detection. CONCLUSIONS: - Assessment of IHC laboratory performance by quantitative measurement of analytic response curves is a powerful, objective tool for identifying outlier IHC laboratories. It uniquely evaluates immunostain performance across a range of defined analyte concentrations.

Analytic Response Curves of Clinical Breast Cancer IHC Tests.

An important limitation in the field of immunohistochemistry (IHC) is the inability to correlate stain intensity with specific analyte concentrations. Clinical immunohistochemical tests are not described in terms of analytic response curves, namely, the analyte concentrations in a tissue sample at which an immunohistochemical stain (1) is first visible, (2) increases in proportion to the analyte concentration, and (3) ultimately approaches a maximum color intensity. Using a new immunostaining tool ( IHControls), we measured the analytic response curves of the major clinical immunohistochemical tests for human epidermal growth factor receptor type II (HER-2), estrogen receptor (ER), and progesterone receptor (PR). The IHControls comprise the analytes HER-2, ER, and PR at approximately log concentration intervals across the range of biological expression, from 100 to 1,000,000 molecules per test microbead. We stained IHControls of various concentrations using instruments, reagents, and protocols from three major IHC vendors. Stain intensity at each analyte concentration was measured, thereby generating an analytic response curve. We learned that for HER-2 and PR, there is significant variability in test results between clinical kits for samples with analyte concentrations of approximately 104 molecules/microbead. We propose that the characterization of immunostains is an important step toward standardization.

The Importance of Epitope Density in Selecting a Sensitive Positive IHC Control.

Clinical Immunohistochemistry (IHC) laboratories face unique challenges in performing accurate and reproducible immunostains. Among these challenges is the use of homemade controls derived from pathological discard samples. Such positive controls have an unknown number of analyte molecules per cell (epitope density). It is unclear how the lack of defined analyte concentrations affects performance of the control. To address this question, we prepared positive IHC controls ( IHControls) for human epidermal growth factor receptor type II (HER-2), estrogen receptor (ER), or progesterone receptor (PR) with well-defined, homogeneous, and reproducible analyte concentrations. Using the IHControls, we examined the effect of analyte concentration on IHC control sensitivity. IHControls and conventional tissue controls were evaluated in a series of simulated primary antibody reagent degradation experiments. The data demonstrate that the ability of a positive IHC control to reveal reagent degradation depends on (1) the analyte concentration in the control and (2) where that concentration falls on the immunostain's analytic response curve. The most sensitive positive IHC controls have analyte concentrations within or close to the immunostain's concentration-dependent response range. Strongly staining positive controls having analyte concentrations on the analytic response curve plateau are less sensitive. These findings emphasize the importance of selecting positive IHC controls that are of intermediate (rather than strong) stain intensity.

A high throughput combinatorial library technique for identifying formalin-sensitive epitopes.

We present a technique for identifying the amino acids responsible for a loss of immunoreactivity in response to treating an antigen with a chemical modifier. This is of particular interest for the chemical formaldehyde, the cross-linking agent in formalin. Formalin is a commonly used fixative to preserve the cellular architecture of cells and tissues and to prevent degradation from proteases and nucleases. Formalin is also routinely used in the preparation of vaccines, to inactivate both toxins and microbes. Formalin fixation attenuates infectivity and pathogenicity by cross-linking while often preserving antigenicity. However, some epitopes are irreversibly modified by formalin while others are not. An understanding of how formalin affects epitope immunoreactivity may be useful in vaccine development or in the development of diagnostic antibody reagents for formalin-fixed tissues. In this report, we describe a method for systematically identifying formalin-sensitive and formalin-insensitive epitopes in a high throughput fashion, for any particular antibody. The data from this effort underscore the importance of certain amino acids, notably lysine, in affecting antibody immunoreactivity after formalin fixation. The method can be generally applicable in exploring the sensitivity of protein epitopes to an agent or condition of interest.

A molecular model of antigen retrieval using a peptide array.

Even though antigen retrieval is highly denaturing, it paradoxically restores immunoreactivity after formalin fixation. It is unclear how this happens. We address this question using a peptide array to model formalin fixation and antigen retrieval. The peptides are linear stretches based on the native protein sequence, containing antibody epitopes of HER-2, estrogen receptor, progesterone receptor, and Ki-67. Of the 7 peptides, 6 retain their immunoreactivity after formalin fixation. However, if formalin fixation is performed in the presence of an irrelevant protein, immunoreactivity is abrogated, regardless of the peptides' amino acid composition. Fixation of an external protein around the antibody epitope prevents antibody binding. Antigen retrieval restores immunoreactivity. These findings demonstrate that native protein conformation is not relevant during antigen retrieval. Moreover, the loss and recovery of immunoreactivity associated with fixation and antigen retrieval, respectively, can be accounted for completely with a model of steric interference by adjacent proteins.

Antibodies immunoreactive with formalin-fixed tissue antigens recognize linear protein epitopes.

It is not clearly understood why some monoclonal antibodies bind to their antigens informalin-fixed, paraffin-embedded tissue sections but others do not. To address this question, we analyzed the protein epitopes of 9 monoclonal antibodies that are immunoreactive after formalin fixation and antigen retrieval. We identified the antibody contact sites by using phage display and synthesized corresponding peptides derived from the GenBank database sequence that contain the predicted antibody binding sites. Our data indicate that all 9 antibodies bind to linear epitopes, ie, composed of contiguous amino acids. In addition, the amino acids proline, tyrosine, glutamine, and leucine are highly represented in these antibody contact sites. The epitopes tend to be mildly to moderately hydrophilic. These findings are the first detailed studies of antibody epitopes associated with antigen retrieval and suggest that antibodies must recognize linear sequences to bind after formalin fixation and antigen retrieval.

Levey-Jennings Analysis Uncovers Unsuspected Causes of Immunohistochemistry Stain Variability.

Almost all clinical laboratory tests use objective, quantitative measures of quality control (QC), incorporating Levey-Jennings analysis and Westgard rules. Clinical immunohistochemistry (IHC) testing, in contrast, relies on subjective, qualitative QC review. The consequences of using Levey-Jennings analysis for QC assessment in clinical IHC testing are not known. To investigate this question, we conducted a 1- to 2-month pilot test wherein the QC for either human epidermal growth factor receptor 2 (HER-2) or progesterone receptor (PR) in 3 clinical IHC laboratories was quantified and analyzed with Levey-Jennings graphs. Moreover, conventional tissue controls were supplemented with a new QC comprised of HER-2 or PR peptide antigens coupled onto 8 µm glass beads. At institution 1, this more stringent analysis identified a decrease in the HER-2 tissue control that had escaped notice by subjective evaluation. The decrement was due to heterogeneity in the tissue control itself. At institution 2, we identified a 1-day sudden drop in the PR tissue control, also undetected by subjective evaluation, due to counterstain variability. At institution 3, a QC shift was identified, but only with 1 of 2 controls mounted on each slide. The QC shift was due to use of the instrument's selective reagent drop zones dispense feature. None of these events affected patient diagnoses. These case examples illustrate that subjective QC evaluation of tissue controls can detect gross assay failure but not subtle changes. The fact that QC issues arose from each site, and in only a pilot study, suggests that immunohistochemical stain variability may be an underappreciated problem.

Standardizing Immunohistochemistry: A New Reference Control for Detecting Staining Problems.

A new standardized immunohistochemistry (IHC) control for breast cancer testing comprises formalin-fixed human epidermal growth factor receptor 2, estrogen receptor, or progesterone receptor peptide antigens covalently attached to 8-µm glass beads. The antigen-coated beads are suspended in a liquid matrix that hardens upon pipetting onto a glass microscope slide. The antigen-coated beads remain in place through deparaffinization, antigen retrieval, and immunostaining. The intensity of the beads' stain provides feedback regarding the efficacy of both antigen retrieval and immunostaining. As a first report, we tested the sensitivity and specificity of the new IHC controls ("IHControls"). To evaluate sensitivity, various staining problems were simulated. IHControls detected primary and secondary reagent degradation similarly to tissue controls. This first group of IHControls behaved similarly to tissue controls expressing high concentrations of the antigen. The IHControls were also able to detect aberrations in antigen retrieval, as simulated by sub-optimal times or temperatures. Specificity testing revealed that each antigen-coated bead was specific for its cognate IHC test antibody. The data support the conclusion that, like tissue controls, IHControls are capable of verifying the analytic components of an immunohistochemical stain. Unlike tissue controls, IHControls are prepared in large bulk lots, fostering day-to-day reproducibility that can be standardized across laboratories.

A novel quality control slide for quantitative immunohistochemistry testing.

We introduce a novel quality control technology that may improve intra- and interlaboratory immunohistochemistry (IHC) standardization. The technology involves the creation of standardized antibody targets that are attached to the same slides as the patient sample. After IHC staining, the targets turn the same color as the stained cells or tissue elements. Unlike current clinical practice, our proposed targets are neither cells nor tissue sections. To create reproducible standards that are available in unlimited supply, we use short constrained peptides as antibody targets. These peptides are attached directly to the glass slide. We show that these peptides simulate the portion of the native antigen to which the antibody binds. They are useful in detecting subtle changes in IHC staining efficacy. Moreover, the peptides do not degrade after deparaffinization or antigen retrieval treatments. This technology may be valuable in creating nationally standardized controls to quantify IHC analytical variability.

A molecular mechanism of formalin fixation and antigen retrieval.

Despite the popularity of antigen-retrieval techniques, the precise molecular mechanism underlying the process remains enigmatic. We examined the molecular features underlying the loss of immunoreactivity following formalin fixation, with subsequent recovery by antigen retrieval. To do this, we first created a molecular model using short peptides that mimic the antibody-binding site of common clinical protein targets. The advantage of this model is that we know the amino acid sequence in and around the antibody-binding site. We observed that some, not all, of the peptides exhibited the formalin-fixation and antigen-retrieval phenomenon. Other peptides did not lose their ability to be recognized by antibody, even after prolonged incubation in formalin. A third, intermediate group exhibited the formalin-fixation and antigen-retrieval phenomenon only if another irrelevant protein was mixed with the peptide before fixation. Amino acid sequence analysis indicates that fixation and antigen retrieval are associated with a tyrosine in or near the antibody-binding site and with an arginine elsewhere, implicating the Mannich reaction as important in fixation and antigen retrieval.

Generation of anti-colorectal cancer fab phage display libraries with a high percentage of diverse antigen-reactive clones.

A combinatorial Fab phage display library was generated from the antibody variable region genes of each of 2 BALB/c mice immunized with the human colorectal cancer cell lines SW480, SW948, and SW837. These libraries were shown to be diverse by nucleotide sequencing and diagnostic restriction enzyme digestion (fingerprinting) of individual members. The two libraries were combined and selected for binding to a suspension of formaldehyde-fixed human colorectal cancer cells in two successive rounds of selection and phage amplification by infection of bacteria. Analysis of the selected libraries as well as individual library clones by ELISA, showed binding to the cancer cell lines in both formaldehyde-fixed and native forms. Fifty five percent and 94% of library clones were positive for colorectal cancer cell binding after the first and second rounds of selection, respectively. Fingerprinting of individual clones showed the first round selected library to be very diverse and the second round selected library to be of more limited diversity. After absorption with normal human cell types, these anti-cancer selected libraries could be used to develop therapeutic and/or diagnostic agents.