Since its introduction as a routine diagnostic procedure over 25 yrs ago, immunohistochemistry (IHC) has revolutionized the field of surgical pathology. This powerful technique allows greater precision in the characterization and diagnosis of solid tumors, hematolymphoid neoplasms, and infections than ever before. An increasing number of antibodies directed against normal and abnormal cellular proteins as well as infectious agents is available to the surgical pathologist to diagnose and subclassify disease entities. These markers can be used in a variety of diagnostic and research settings.

We live in a time when a number of diseases can be characterized by a single genetic alteration that is easily assayed in the modern molecular pathology laboratory. Unfortunately, a laboratory with this level of sophistication is not yet readily accessible to the majority of practicing pathologists. Likewise, there are few pathologists that are trained in the performance and evaluation of molecular studies. For surgical pathologists, it is IHC, a test that focuses on recognition of protein products expressed by different cell populations in conjunction with a morphologic examination, that is used as a means to circumvent the need for direct evaluation of nucleic acid alterations. With this method, the products of genes are assayed in tissue sections, often allowing one to characterize a cell population as benign or neoplastic, determine cell lineage, and, in some cases, even determine the nature of the molecular genetic alteration leading to the process.

As with so many areas of medicine, when something seems too good to be true, it probably is. Immunohistochemistry is not a panacea that can provide a concise answer in every questionable case. In fact, if its use is not guided by a working knowledge of morphology, immunohistochemical methods, and their limitations, IHC can be as much a hindrance as a help. In this chapter, we will discuss the history and methodology of IHC, provide a brief introduction to the evaluation of tumors with immunohistochemical studies, and review some specific examples of how it correlates with molecular genetics.


Immunohistochemistry has been around in its rudimentary forms since the 1940s, when Coons experimented with immunofluorescence to detect antigens in frozen tissue sections (1). Unfortunately, immunofluorescence usually requires fresh frozen tissue and lacks the sensitivity required for routine surgical pathology applications. The use of immunohisto-chemical methods on routinely processed surgical tissue sections was first reported by Taylor and colleagues in the mid-1970s (2). The introduction of IHC into daily practice has taken some time, but most anatomic pathologists today would be loath to practice without access to at least some immunohistochemical markers.

A large step forward in the development of IHC techniques occurred in 1975 with the development of hybridoma cell lines capable of producing large quantities of pure monoclonal antibodies (3). Also in 1975, the use of enzyme digestion to help unmask antigens altered by formalin fixation was described by Huang et al. (4). These humble beginnings led to an explosion of literature and a rapidly growing list of IHC "stains" that could delineate cell lineages, specific products of cells, and, in some cases, the difference between benign and malignant cells. With experience and constant investigation, antibodies with greater specificity have been developed and newer techniques such as heat-induced epitope retrieval (HIER) have improved the sensitivity (5). Other advances that improve laboratory logistics such as automated staining platforms, commercial antibody "kits," and polymer detection techniques have improved standardization, decreased turnaround time, and reduced nonspecific staining. A dizzying array of options is currently available for routine diagnostic use in many pathology laboratories.


A complete discussion of antibody production and IHC methodology is beyond the scope of this chapter. For further information, the reader is referred to recent textbooks of immunology and IHC (6,7).

From: Molecular Diagnostics: For the Clinical Laboratorian, Second Edition Edited by: W. B. Coleman and G. J. Tsongalis © Humana Press Inc., Totowa, NJ

3.1. ANTIBODIES Antibodies are proteins produced by B-lymphocytes in response to an antigen, which is a part of a protein, carbohydrate moiety, or some other biochemical construct that can be recognized and bound by an antibody. An immunogen, by contrast, is something that, when encountered by the immune system, stimulates the production of an antibody. Antibodies are immunoglobulin molecules that comprise two identical heavy and two identical light chains, each of which includes a constant region and a variable region. The amino acid sequence of the variable region confers steric and electrochemical properties that help determine how tightly (avidity) and how specifically (affinity) an antibody will interact with an antigen.

Antibodies for immunohistochemical studies are produced by animal hosts in response to an antigen stimulus. Depending on the method of production, antibodies can be monoclonal (produced in a hybridoma cell line in an animal host) or poly-clonal (produced by many different cells in a live animal). As the names imply, the monoclonal antibody is derived from a single clone and produces a single antibody that recognizes one epitope of an antigen, whereas polyclonal antibodies are made by many cells that may recognize different epitopes on the antigen used to stimulate production. Each form has inherent advantages and disadvantages. The monoclonal antibody, usually produced in a mouse host, has the advantages of homogeneity, reproducibility, and lack of admixed nonspecific antibodies. Polyclonal antibodies will have lot-to-lot variability and usually higher background staining because they are not made by a single clone and they recognize multiple epitopes on a given antigen. The latter property can be advantageous if the antigen of interest is susceptible to degradation or alteration during fixation and also because the conditions under which the stains are used are less stringent. Although monoclonal antibodies are frequently preferred because of to specificity and reproducibility, several caveats must be considered. For instance, some antibodies are raised against fresh cells that have antigens that degenerate or are masked as a result of to suboptimal processing conditions or of fixation.

In IHC, antibodies serve two purposes. The first function is as a probe to recognize and bind a specific epitope such as the intermediate filament, cytokeratin, or the calcium-binding protein, S-100, and so forth that is mediated through the variable region. This function is characteristic of the "primary" antibody. Antibodies also serve as antigens wherein the constant region of the primary antibody is recognized by a "secondary"antibody conjugated to a fluorescent dye, or to biotin, which, in turn, binds an avidin-enzyme complex, or directly to an enzyme that can produce a localized visible result when reacted with a chro-mogenic substrate. Allowing visualization of the anatomic and histologic distribution of the antigen of interest is one of the main advantages of this technique.

3.2. IMMUNOHISTOCHEMICAL TECHNIQUES It is not only the specificity of the primary antibody that determines the usefulness of an antibody. Other factors related to the tissue and its processing are equally important. These include the fixative, duration of fixation, tissue processing, the method of antigen retrieval, and the detection system, to name a few (8). The last critical ingredient is a working knowledge of morphology and expected reactions so that results can be interpreted as normal, abnormal, or indeterminate.

Attempts to find alternative fixatives to formalin for preservation of tissue morphology and immunoreactivity have been largely unsuccessful. Formalin, which is inexpensive, is easy to use, and provides good and reproducible morphology, will likely remain the standard fixative used in surgical pathology for the foreseeable future. The fact that current morphologic criteria for diagnoses are based on the appearance of tissues fixed in formalin, embedded in paraffin, and stained with hematoxylin and eosin makes the possibility of developing and using other fixatives even less likely. At the molecular level, formalin fixation results in the formation of crosslinks between proteins or between proteins and nucleic acids, involving hydroxymethylene bridges (9). As a result of the alteration of the three-dimensional structure of the proteins, some anti-genic sites are masked and become inaccessible to antibodies used in IHC. Initial efforts to find a better fixative than formalin were redirected after the discovery that at least some of the changes induced by formalin were reversible with the use of antigen retrieval techniques (10). It is interesting to note that these same techniques might be able to improve yield in nucleic acid-based tests as well (11).

3.2.1. Antigen Retrieval Antigen retrieval (AR) can be a simple and effective method that allows antigens masked during formalin fixation to be recognized by antibodies, but extensive testing is necessary to optimize the retrieval for different types of antibody. The first method for unmasking antigens in formalin-fixed tissue involved the use of proteolytic enzymes (12). A number of enzymes have been used, including trypsin, pronase, pepsin, and ficin. However, the number of variables that have to be considered for enzymatic epitope retrieval make this technique less desirable. Different antibodies require different enzyme techniques, and inappropriate use of proteases could result in false-negative reactions and tissue destruction. The choice of enzyme also depends on the type of fixative used. Furthermore, the length of enzymatic digestion should be proportional to the extent of exposure to fixative, which is difficult to assess and control. Although HIER has taken a leading role in antigen retrieval, enzyme retrieval techniques are still preferred for detection of some antigens (13). The requirements for pretreatment must be established by the institution preparing the immunohistochemical stains, as conditions are highly variable among different laboratories.

The antigen retrieval technique introduced by Shi et al. in 1991 (5) has revolutionized IHC and greatly expanded the number of antigens that can be detected in formalin-fixed, paraffin-embedded tissue (10,14,15). In HIER, the unmasking of antigens is brought about by placing tissue in different types of buffer at temperatures between 80°C and 125°C. Studies have shown that the product of temperature and time, and the pH of the buffer used (rather then its chemical composition) are the major factors that influence antigen retrieval (8). Different heating methods (microwave, pressure cooker, water bath, steamer, or autoclave) can be adjusted to yield similar intensities of staining. Water baths that use a lower temperature (90-95°C) require a longer time but show better preservation of morphology, which is critical to interpretation of results.

The most widely used retrieval buffers are citrate-based at pH = 6.0. The use of high-pH buffers in conjunction with high temperature for low-expression tissue antigens or for overfixed tissue enhances the sensitivity of detection, at the risk of increased tissue loss (tissue tends to detach from the slide with this method) and increased background staining.

The introduction of HIER has increased the sensitivity of staining for a wide range of antibodies and has made possible the retrieval of some antigens that are otherwise negative even with other unmasking pretreatments (Ki-67 [clone MIB-1], bcl-2, estrogen receptor [ER], progesterone receptor [PR], p53, and some CD markers) (16). Increasing the sensitivity generally comes at a cost of decreasing specificity. The delicate balance that must be maintained requires a test battery approach to establish the optimal protocol for pH, time, and temperature that gives reliable, specific immunoreactivity (8).

As always, standardization is a critical issue, especially for quantitative or semiquantitative immunohistochemical studies (Ki-67, ER, PR, p53). One area that has been particularly difficult to standardize is fixation. Which fixative is used and for how long has been virtually impossible to control across institutions. HIER has been used to reverse some of the effects of prolonged fixation. Another reagent that is of utmost importance is the antibody itself. Which epitope or part of an epitope does it recognize? Is it monoclonal or polyclonal? A number of antibodies that claim to recognize the same marker (for instance, CD3) are available with widely variable efficacy. Still another example is the use of antigen retrieval. As noted earlier, variations in time of retrieval, temperature, and pH of the buffer play a major role in the reproducibility and utility of these studies. In some instances, the use of AR requires re-evaluation of the clinical interpretation of IHC (e.g., wild-type p53 might be detected after using AR) (17). The more cynical in the pathology community will often quip that any antigen they want to see as positive can be made so with the right preparation.

3.2.2. Detection Methods A wide variety of systems has been developed for the visualization of antigen-antibody reactions in tissue sections, and more are currently being investigated. Initially, direct staining methods using only a labeled primary antibody offered too little sensitivity. Secondary labeled antibodies directed against the primary antibody were introduced to increase the amount of signal generated by allowing more than one labeled antibody to bind for a given amount of antigen. Using an additional labeled antibody to react with the secondary antibody allowed further amplification.

The two most widely used antigen detection methods are (1) unlabeled, which include enzyme-antienzyme methods and (2) labeled, including the avidin-biotin complex (ABC), which eliminated the need to conjugate all primary antibodies (6). Because of their low sensitivity and complex methodology, enzyme-antienzyme methods such as peroxidase-antiperoxi-dase (PAP) and alkaline phosphatase-antialkaline phosphatase (APAAP) techniques have largely been abandoned. The ABC and labeled streptavidin-biotin (LSAB) techniques have been used extensively. Newer detection systems are aimed at increasing sensitivity while limiting nonspecific background staining. Other practical goals are simplicity, reproducibility, and speed, all with less expense.

The ABC and LSAB techniques have had a significant role in the advancement of IHC and are still in use today (18). ABC is based on the high-affinity binding between biotin and avidin. In this method, the secondary antibodies are conjugated with biotin and function as a link between the primary antibody, which is bound to the tissue, and the avidin-biotin complex. Streptavidin can be used instead of avidin with the advantage that there is less nonspecific tissue binding. A drawback of the ABC and LSAB methods is the fact that high levels of endogenous biotin in tissues, especially liver, kidney, thyroid, and brain, will produce background staining. Furthermore, HIER techniques appear to increase the reactivity of endogenous biotin to levels that can interfere with interpretation.

Newer, polymer-based amplifications are replacing the old three-step detection systems (19-22). Depending on the method, either the primary or the secondary antibody is attached to an inert polymer (e.g., dextran) decorated with additional molecules of antibody and an enzyme such as horseradish per-oxidase. Because the polymer can be bound to many antibody and enzyme molecules, the signal of an antigen with low-level expression can be greatly amplified. DakoCytomation (Carpenteria, CA) has pioneered this method by introducing the Enhanced Polymer One-Step Staining (EPOS) system. This resembles the early direct methods in that it requires only the primary antibody, with the difference of markedly increased signal in a single step. The methodology is easy, fast [might even be used for frozen section diagnosis (23)], and sensitive, but is limited by the number of labeled primary antibodies available. To address this problem, an indirect method that uses a standardized secondary antibody attached to dextran was developed. This EnVision system, from the same company, has been gaining popularity for its speed because of shorter incubation times, reduced complexity with only two steps, high sensitivity, and its ability to be used with a wide variety of primary antibodies (20,21). Additional benefits include a potential reduction in primary antibody costs, because higher dilutions can be used while maintaining sensitivity, and in background staining, because endogenous biotin is avoided. The latter also allows for work with stronger HIER without concern for activation of endogenous biotin. A second generation of polymer-based detection reagents, PowerVision (22), has been developed by ImmunoVision Technologies (Daly, CA). It uses small, multifunctional polymeric linkers that have better tissue penetration than dextran and, therefore, a higher sensitivity in detection.

Finally, signal amplification techniques such as tyramide amplification (24,25) can greatly enhance the sensitivity of classical methods such as ABC or LSAB. This technique is based on the generation of highly reactive biotinyl-tyramide intermediates in the vicinity of the peroxidase enzyme. Although this method can greatly increase the sensitivity of detection, nonspecific staining is also increased. Introduction of these methods requires careful evaluation of the interpretation criteria and of their clinical significance. For now, they remain primarily research tools.

3.2.3. Automation The fact that there are so many variables in IHC and that controlling these variables is essential to standardizing results has led to the development of automated staining devices (26). These devices reduce day-to-day staining variability and can potentially improve intralaboratory and interlaboratory reproducibility of results. There are many variations on the theme of automation. Most are based either on the use of capillary action or pipets to deliver reagents. They are devised as either closed systems that work only with proprietary reagent kits or more flexible open systems that can use antibodies and other reagents from different manufacturers. In general, the requirements for an automated staining device include analytical flexibility, "walk-away" operation, a user-friendly interface, and safety (27). Taking automation to the next level has led to the production of automated instruments that will deparaffinize, perform antigen retrieval, and perform IHC staining. Automation can lead to more consistent quality by improving standardization, especially with regard to certain quantitative IHC stains that are potential prognostic indicators and therapeutic targets.

My First Baby

My First Baby

Are You Prepared For Your First Baby? Endlessly Searching For Advice and Tips On What To Expect? Then You've Landed At The Right Place With All The Answers! Are you expecting? Is the time getting closer to giving birth to your first baby? So many mothers to be are completely unprepared for motherhood and the arrival of a little one, but stress not, we have all the answers you need!

Get My Free Ebook

Post a comment