Immunohistochemical analysis is a laboratory method that uses antibodies to detect specific proteins in tissue samples, making invisible molecular targets visible under a microscope. Pathologists rely on it every day to diagnose cancers, identify infections, and determine which treatments a patient’s tumor is likely to respond to. The technique works by pairing an antibody that recognizes a target protein with a color-producing system, so that when the antibody locks onto its target, the result is a visible stain in the exact location where that protein sits within the tissue. It has become one of the most widely used tools in diagnostic medicine, yet the process behind each stained slide involves a surprisingly finicky chain of chemistry, biology, and interpretation.
How the Technique Works in Plain Terms
Think of it as a molecular search party. A pathologist wants to know whether a particular protein is present in a patient’s tissue and, if so, where exactly it is. To find it, the lab applies an antibody designed to recognize that protein. Antibodies are Y-shaped molecules that bind to a specific molecular shape the way a key fits a lock. Once the antibody finds and latches onto its target, a detection system translates that binding event into a color the human eye can see under a microscope. The color appears only where the target protein exists, producing a map of protein expression across the tissue.
The concept dates back to 1941, when Albert Coons and colleagues first attached a fluorescent tag to antibodies and used them to locate bacterial antigens in infected tissue.1PubMed. Immunohistochemistry in Historical Perspective: Knowing the Past to Understand the Present That original approach used fluorescence, which required a specialized microscope and faded quickly. Over the following decades, researchers developed enzyme-based labels that produce a permanent colored deposit visible under an ordinary light microscope, making the technique practical for routine diagnostic work.
Preparing the Tissue
Before any antibodies touch the tissue, the sample has to be preserved and sliced thin enough for light to pass through. Most diagnostic labs fix tissue in formalin, a formaldehyde solution that cross-links proteins and prevents decay. The fixed tissue is then embedded in paraffin wax and cut into sections just a few micrometers thick, mounted on glass slides. These formalin-fixed, paraffin-embedded (FFPE) samples are the standard currency of pathology labs worldwide because they can be stored at room temperature for years and still yield useful results.
Formalin preservation comes with a catch: the chemical cross-links it creates can mask the very proteins the antibodies need to find. The target protein is still there, but the cross-links physically block the antibody from reaching it. That is why almost every antibody used on FFPE tissue requires a step called antigen retrieval before staining begins.2PubMed Central. Molecular mechanisms of antigen retrieval: antigen retrieval reverses steric interference caused by formalin-induced cross-links Antigen retrieval typically involves heating the tissue slide in a buffer solution. The heat breaks apart some of the formalin-induced cross-links and unfolds proteins, exposing the sites where antibodies can bind.3PubMed Central. Antigen retrieval causes protein unfolding: evidence for a linear epitope model of recovered immunoreactivity
Storage conditions matter more than most people realize. Research on long-stored tissue sections has shown that some proteins lose their detectability over time. Proteins located in the cell nucleus or on the cell membrane tend to degrade faster than those in the cytoplasm, and cold storage of the cut slides slows that degradation compared to storage at room temperature.4PubMed. Factors affecting immunoreactivity in long-term storage of formalin-fixed paraffin-embedded tissue sections Cytoplasmic structural proteins like keratins, by contrast, remain reliably detectable even after years on the shelf. This means a pathologist pulling an archival slide from a decade ago to re-test for a nuclear marker like estrogen receptor may get a falsely weak or negative result, not because the protein was never there, but because the target has deteriorated.
Detection Systems and Color
The antibody that recognizes the target protein is called the primary antibody. On its own, it is invisible. To generate a visible signal, labs apply a detection system, which usually involves a secondary antibody directed against the primary, linked to an enzyme. When a chemical substrate is added, the enzyme converts it into a colored deposit right at the spot where the primary antibody is bound.
The most common enzyme used is horseradish peroxidase, and its most popular substrate is a chemical called DAB, which produces a brown deposit. Other substrates produce red, blue, or other colors. Among the various options, DAB-based protocols with certain enhancers have proven to be among the most sensitive for peroxidase-driven reactions, while for the alternative enzyme alkaline phosphatase, a blue-purple substrate called BCIP-TNBT tends to perform best.5PubMed. Sensitivity of various visualization methods for peroxidase and alkaline phosphatase activity in immunoenzyme histochemistry The brown-stained tissue is then counterstained with hematoxylin, which colors the cell nuclei blue. This two-tone result lets the pathologist see exactly which cells contain the target protein and where in those cells it sits.
Earlier generations of detection systems relied on avidin-biotin complexes to amplify the signal. These worked well but had a practical annoyance: biotin is a naturally occurring molecule in many human tissues, so endogenous biotin could produce false background staining, especially after aggressive antigen retrieval.6PubMed. Comparison of two polymer-based immunohistochemical detection systems: ENVISION+ and ImmPRESS Newer polymer-based detection systems sidestep this problem entirely. They use synthetic polymer backbones loaded with many enzyme molecules, delivering stronger signal without relying on biotin and producing cleaner results than the older avidin-biotin methods.7PubMed Central. Detection Systems in Immunohistochemistry – Section: Polymer-based immunohistochemistry
What It Is Used to Diagnose
The single largest use of immunohistochemistry is in cancer diagnosis. When a pathologist looks at a tumor under the microscope, the cells sometimes look so abnormal that it is hard to tell where the cancer originally came from. This matters because treatment depends heavily on the tissue of origin. Systematic panels of antibodies can classify most of these otherwise unrecognizable tumors, helping pinpoint whether a cancer started in the lung, breast, colon, kidney, or elsewhere.8PubMed Central. Immunohistochemistry for Diagnosis of Metastatic Carcinomas of Unknown Primary Site A pathologist might apply an antibody for a protein that is typically expressed only in thyroid cells, for example. If the tumor lights up, it likely originated in the thyroid even if it was biopsied from a lymph node in the neck.
Beyond cancer classification, immunohistochemistry is also used to detect pathogens directly in tissue. A review documented detection of 59 different infectious agents across 22 different tissues and organs using this approach. It is especially valuable for organisms that are difficult or impossible to grow in culture, including intracellular bacteria. One creative variation, called auto-immunohistochemistry, uses the patient’s own serum as the primary antibody on surgically removed infected tissue, which has helped identify organisms responsible for culture-negative heart valve infections.9PubMed Central. Immunohistochemical diagnosis of human infectious diseases: a review
Guiding Cancer Treatment Decisions
Immunohistochemistry has moved well beyond diagnosis into the territory of predicting which therapies will work. Certain antibodies stain for proteins that directly determine a patient’s treatment options. Two of the best-known examples are HER2 and PD-L1 in breast cancer.
HER2 is a growth-promoting receptor found on the surface of some breast cancer cells. When the immunohistochemistry score is high, the patient is eligible for targeted drugs that block HER2. The scoring system runs from 0 to 3+, and only tumors scoring 3+ are considered clearly positive. The stakes of getting this score right are high: it determines access to drugs that can meaningfully extend survival.
PD-L1 is a protein that helps tumors evade the immune system. In triple-negative breast cancer, a specific immunohistochemistry test measures how much PD-L1 the immune cells surrounding the tumor express. If those immune cells cover at least one percent of the tumor area, the patient qualifies for a combination of immunotherapy and chemotherapy.10PubMed Central. Understanding PD-L1 Testing in Breast Cancer: A Practical Approach One persistent frustration with PD-L1 testing is that different antibody clones and scoring systems can give different answers on the same tissue. In lung cancer, for instance, PD-L1 and ALK are established immunohistochemistry-based predictive tests, but PD-L1 interpretation remains specific to the assay, the testing platform, and the cutoff used.11Biomolecules and Biomedicine. Predictive immunohistochemistry in non-small cell lung cancer – Biomarkers, companion diagnostics, and molecular testing: A narrative review This means a patient tested with one commercial antibody kit could receive a different result than if tested with another, which is an uncomfortable reality for a test that gates access to treatment.
Seeing More Than One Protein at a Time
Traditional immunohistochemistry stains for one protein per slide. That works fine for many diagnostic questions, but modern cancer research often needs to understand how multiple cell types interact within the same neighborhood of a tumor. If you want to know whether immune T cells are sitting right next to tumor cells and whether those T cells are active or exhausted, you need to see several markers simultaneously on the same tissue section.
Multiplex immunohistochemistry and immunofluorescence techniques have been developed to do exactly this. They allow detection of multiple markers on a single tissue section, enabling researchers to study cell composition, cell function, and cell-to-cell interactions all at once.12PubMed Central. Overview of multiplex immunohistochemistry/immunofluorescence techniques in the era of cancer immunotherapy One platform demonstrated simultaneous detection of six protein markers plus a nuclear stain, with automated imaging and computer-driven quantification of hundreds of thousands of individual cells on a single slide.13Scientific Reports. Systems pathology by multiplexed immunohistochemistry and whole-slide digital image analysis These techniques have become particularly important for understanding why some patients respond to immunotherapy and others do not, because the spatial relationships between immune cells and tumor cells seem to matter as much as the raw numbers.
Computers Reading the Slides
Pathologists have traditionally scored immunohistochemistry results by eye, estimating the percentage of cells that stain positive and how intensely they stain. This visual scoring is subjective. Two experienced pathologists looking at the same slide can disagree, and even the same pathologist may score a slide differently on two different days. Digital image analysis aims to reduce that variability.
For HER2 scoring in breast cancer, automated digital analysis has shown near-perfect agreement with visual evaluation, and in some cases the computer was actually more accurate at detecting borderline cases that visual scoring would have missed.14PubMed Central. Membrane connectivity estimated by digital image analysis of HER2 immunohistochemistry is concordant with visual scoring and fluorescence in situ hybridization results More recently, deep learning algorithms have been developed that can take an immunohistochemistry image and output a standardized score in seconds, with precision and consistency comparable to experienced pathologists but at a fraction of the cost and time.15Modern Pathology. Deep Learning–Based H-Score Quantification of Immunohistochemistry-Stained Images The algorithm categorizes the staining intensity of each pixel and calculates the final score based on the distribution. For busy pathology departments processing thousands of slides a week, this kind of automation is becoming less of a novelty and more of a necessity.
Where Things Go Wrong
Immunohistochemistry looks deceptively simple in concept, but in practice there are many places the process can break down. Quality control studies have evaluated staining quality by scoring slides for intensity, uniformity, specificity, background, and counterstaining, categorizing results as optimal, borderline, or unacceptable.16Journal of Clinical Pathology. Audit and internal quality control in immunohistochemistry Suboptimal results can stem from many sources: tissue that sat too long before fixation, antigen retrieval that was too aggressive or too mild, expired reagents, or poorly calibrated automated staining machines.
One area where conventional wisdom turns out to be wrong is blocking. Most immunohistochemistry protocols include a step where the tissue is incubated with a protein solution before the primary antibody is applied, supposedly to prevent antibodies from sticking nonspecifically to the tissue. A study that systematically tested this found, somewhat surprisingly, that routinely fixed cell and tissue samples showed no nonspecific antibody binding regardless of whether a blocking step was performed.17Scientific Reports. Non-specific binding of antibodies in immunohistochemistry: fallacies and facts The standard fixation process appears to destroy the receptors that would otherwise grab antibodies indiscriminately. This suggests that the blocking step so many labs treat as sacred may be unnecessary in routine work, though old habits die hard in clinical protocols.
The Mouse-on-Mouse Problem
A quirky but common technical headache arises when researchers need to stain mouse tissue using mouse-derived antibodies. Most monoclonal antibodies used in research are made in mice, and that is fine when the tissue being stained comes from a human. But when you apply a mouse antibody to mouse tissue, the secondary detection antibody cannot tell the difference between the antibody you applied on purpose and the immunoglobulins that were naturally present in the mouse’s own tissue. The result is a mess of false background staining.18PubMed Central. A flexible mouse-on-mouse immunohistochemical staining technique adaptable to biotin-free reagents, immunofluorescence, and multiple antibody staining
Specialized protocols have been developed to work around this. One approach involves pre-complexing the primary mouse antibody with a fragment of an anti-mouse antibody in a tube before applying it to the tissue, then mopping up any uncomplexed secondary antibody with mouse serum. Another uses fluorescent-dye-conjugated secondary antibodies that reduce the background noise, offering a less expensive alternative to commercial kits based on antibody fragments.19PubMed. Optimization of Mouse-on-Mouse Immunohistochemistry by Utilizing Fluorescent-dye Conjugated Secondary Anti-Mouse Antibody The issue matters mainly in preclinical research, where mouse models of disease are ubiquitous and reliable staining of mouse tissue is essential for interpreting experimental results.
Where the Technology Is Heading
One of the more ambitious frontiers involves combining immunohistochemistry with mass spectrometry, a technique that identifies molecules by their mass. A workflow called MALDI-IHC uses targeted immunohistochemistry imaging to guide an untargeted mass-spectrometry-based analysis of proteins across a tissue section, revealing extensive molecular diversity within a single tumor.20PubMed Central. MALDI-IHC-Guided In-Depth Spatial Proteomics: Targeted and Untargeted MSI Combined A 2025 preprint extended this idea to over 100 protein markers simultaneously, demonstrating that the spatial patterns captured by MALDI-IHC correspond to the tissue architecture seen in traditional pathology images.21bioRxiv. Spatial proteomics with 100+ markers with Highly Multiplexed MALDI-IHC Mapping over a hundred proteins across a tissue section goes far beyond what conventional immunohistochemistry can achieve, and it begins to approach a comprehensive molecular portrait of a tumor’s ecosystem rather than a test for one or two targets at a time.
These spatial proteomics approaches are still largely research tools rather than clinical diagnostics, but they illustrate the direction the field is moving: from asking “is protein X present?” to asking “what is the full molecular landscape of this tissue, and how does it vary from one microscopic neighborhood to the next?” If the technology matures and becomes affordable enough for clinical use, it could reshape how pathologists characterize tumors and how oncologists choose treatments, moving from a handful of yes-or-no biomarker tests toward a high-resolution molecular map of each patient’s disease.