The insights and perspectives offered by Magnani and Taylor1 in their editorial in this issue are salient and timely. Immunohistochemistry (IHC) assays are “stains” whose performance just happens to depend on the prevalence and integrity of antigens in the sample, the concentration and affinity/avidity of the reagent antibody, incubation and washing parameters, and other factors shared with clinical immunoassays. With the proliferation of IHC companion diagnostics, assumptions about the intrinsic performance capabilities of IHC are made, and many are apt to forget the “chemistry” in immunohistochemistry.
Also see p. 1229.
Magnani and Taylor1 capably summarized the similarities and differences between IHC and immunoassays and the ways each are regulated differently. They emphasize the vast preanalytic differences between a blood sample (immunoassay) and paraffin tissue section (IHC) and the challenges these pose for precision and accuracy with IHC. Monitoring and maintaining an awareness of fixation parameters from sample to sample, important as that is, does not necessarily lead to consistency in assay results and will always be a challenge for IHC.
The reduced regulatory stringency for IHC assays, in the United States at least, can be traced to policies of various government agencies. Clinical Laboratory Improvement Amendments (CLIA) defines IHC assays as a stain rather than a test. The US Food and Drug Administration considers IHC assays in different class designations (I, II, and III) according to their potential for clinical impact and availability of corroborating support for the result. Although the College of American Pathologists (CAP) and its members can advocate for reconsideration of IHC assays under CLIA, that decision ultimately rests with these agencies and the lawmakers who grant their authority.
Another relevant difference between IHC and immunoassays has to do with signal detection. The fact that IHC is viewed by human eyes attached to human brains, as opposed to a photometer, is a significant reason for the difference in how each is regulated. One of those systems is more easily calibrated than the other. This fact underlies one of the reasons the CLIA definition of IHC as a stain has persisted. Many consider the interpretation of IHC assays to fall under the purview of “the practice of medicine”; the application of medical judgment by a trained pathologist in an individualized approach to a patient sample. As such, the argument goes, these should be exempt from regulation.
All of that said, as pathologists directing and performing IHC in our laboratories, we can and we should do better. Although it is debatable whether the technology and performance capabilities of IHC can reliably distinguish quantitative cut points, such as 1% or 10% thresholds (for example), we can work to be as precise and accurate as the chemistry allows. We can certainly draw an analogy to clinical immunoassay concepts like analytic range, calibrated controls, and specified quality control ranges with regular assay monitoring and incorporate these into the IHC lab.
The CAP IHC Resource Committee is working to advance these concepts. In recent years, the CAP surveys for mismatch repair (MMR) and Ki-67 have adopted new formats using cultured cell line microarrays to circumvent issues with tissue heterogeneity. Other approaches to calibrated IHC analytes, such as synthetic beads, cultured pseudotissues, and xenografts, are also being explored for future surveys and hold promise for developing the kind of calibrated controls spoken of by Magnani and Taylor.1
Where it makes the most sense, we are starting to decouple assessments of the staining/technical aspects of IHC from the interpretive/professional aspects in our surveys. In 2024, a new set of “interpretation only” surveys for estrogen receptor/progesterone receptor (ER/PR) and human epidermal growth factor receptor 2 (HER2) will become available. These will include cases in the “HER2 low” and “ER low positive” range that we have not been able to query before. Efforts to develop surveys focused on the slides produced by participant laboratories in a standardized way are also underway. Preparing slides with homogeneous analyte material at scale and evaluating each lab’s stained slides using digital and artificial intelligence–based methods will be key components of this.
The CAP IHC surveys have played an important role in the quality assurance of IHC assays for decades. To date, it has been challenging to determine whether a lab failure was due to technical assay problems or erroneous interpretation by a pathologist. It is hoped that this decoupling of the assay components will take this to the next level. Both sources of error need to be addressed (and distinguished) if we want to move IHC closer to the standards of clinical immunoassays. The chemistry in immunohistochemistry matters—both the chemistry under the coverslip and the chemistry within the synapses of a pathologist’s mind at the other end of the microscope.
Many of the ideas expressed here were influenced by current and former members of the CAP IHC Committee and Council on Scientific Affairs, particularly Andrew Bellizzi, MD; David Rimm, MD, PhD; Richard Cartun, PhD; Meghan Troxell, MD, PhD; and Bradley Karon, MD, PhD.
The author has no relevant financial interest in the products or companies described in this article.