What Is HEp-2 and Why Is It Used for ANA Testing?

HEp-2 is a line of human epithelial cells grown on glass slides and used as the standard substrate for antinuclear antibody (ANA) testing, the primary screening test for autoimmune diseases like lupus. When a patient’s blood is placed on these cells and viewed under a fluorescence microscope, any autoantibodies present will bind to structures inside the cells and glow, creating distinctive patterns that help clinicians figure out which autoimmune condition might be at work. The story of how this particular cell line became the global workhorse for autoimmune diagnostics involves a case of mistaken identity, a contamination scandal, and decades of international effort to make the test reliable.

A Cell Line With an Unexpected Identity

HEp-2 stands for “Human Epithelial type 2.” The cell line was first described in 1954, supposedly derived from a laryngeal (throat) cancer. For over a decade, researchers used it believing they were working with laryngeal cancer cells. Then, in 1966, it was revealed that the line had been overtaken by HeLa cells, the famously aggressive cervical cancer cells originally taken from Henrietta Lacks in the 1950s.¹ In other words, what labs call HEp-2 is actually a derivative of HeLa, not a throat cancer line at all. This cross-contamination was a widespread problem in early cell culture work, and HEp-2 is one of the most prominent examples.

The name stuck anyway. By the time the contamination was confirmed, HEp-2 cells had already become entrenched in immunology research and diagnostic testing. For the purposes of ANA testing, the cells’ actual tissue of origin does not matter much. What matters is that they grow quickly, divide reliably, and display a rich variety of cellular structures that autoantibodies can target. The HeLa heritage actually contributes to some of these useful properties, including vigorous growth and large, well-defined nuclei.

Why HEp-2 Replaced Older Methods

Before HEp-2 became the standard, labs performed ANA testing on thin slices of animal tissue, typically mouse liver or kidney. These tissue sections worked, but they had real drawbacks. The cells were small and tightly packed, making it hard to see fine details under the microscope. Different tissue preparations from different animals introduced variability between labs and even between batches at the same lab. Two technologists reading the same slide might disagree on the pattern they were seeing.

HEp-2 cells solved several of these problems at once. Because they are a cultured cell line grown in a monolayer, each slide looks essentially the same. The cells and their nuclei are much larger than those in tissue sections, making patterns far easier to read.² By the early 2000s, most labs worldwide had made the switch. The uniformity of a cell line grown under controlled conditions also meant that results from a hospital in Tokyo could, at least in theory, be compared with results from a clinic in São Paulo.

There is another advantage that gets less attention outside the lab. Because HEp-2 cells are constantly dividing, a single slide captures cells at every stage of the cell cycle, including cells caught in the act of dividing (mitosis). This is useful because some autoantibodies target structures that only appear during mitosis, and those antibodies would be missed entirely on resting tissue slices.

How the Test Actually Works

The basic process is called indirect immunofluorescence, or IIF. A technologist dilutes a patient’s blood serum and applies it to a glass slide coated with a monolayer of fixed HEp-2 cells. If the serum contains autoantibodies, they bind to their target molecules inside the cells. After washing away unbound material, a second antibody tagged with a fluorescent dye is added. This secondary antibody sticks to any human antibodies already bound to the cells. When the slide is placed under a fluorescence microscope, the bound autoantibodies light up, and the pattern of fluorescence tells the lab which cellular structures are being targeted.

The test also measures titer, which is the highest dilution of serum at which fluorescence is still visible. A titer of 1:80 means the serum was diluted 80-fold and still glowed. Higher titers generally suggest a stronger autoimmune response, though the clinical significance depends heavily on context. One study of healthy individuals found that about 13% tested positive for ANA, underscoring that a positive result alone does not mean disease.³ That same study defined a titer of 1:160 as the threshold for an abnormal result, which is the cutoff many labs use to separate background noise from potentially meaningful findings.

Reading the Patterns

The real diagnostic power of the HEp-2 test lies not just in whether it is positive, but in the fluorescence pattern. Different autoantibodies target different structures, and each produces a distinctive glow. A homogeneous pattern, where the entire nucleus lights up evenly, is often associated with antibodies against DNA or histone proteins and is commonly seen in lupus. A speckled pattern, with dots scattered across the nucleus, suggests antibodies against extractable nuclear antigens and can point to conditions like mixed connective tissue disease or Sjögren’s syndrome. A centromere pattern, with discrete dots corresponding to the centromere regions of chromosomes, is strongly linked to a form of scleroderma called limited cutaneous systemic sclerosis.

HEp-2 cells display so many different antigens that multiple autoantibodies in a single patient’s serum can produce overlapping patterns on the same slide. This makes interpretation genuinely difficult. A large study examining complex patterns noted that correct interpretation requires significant training and experience, because serum samples reacting against more than one antigen are common in daily lab work.⁴ A tired technologist at the end of a long shift may read a combined homogeneous-and-speckled pattern differently than a fresh pair of eyes would.

The test is not limited to nuclear patterns, either. Because HEp-2 cells display their entire cytoplasm and mitotic apparatus, the assay detects autoantibodies directed at structures outside the nucleus. This has led some experts to argue that the old name “antinuclear antibody test” is a misnomer. The term “anti-cell antibody test” has been proposed as a more accurate replacement, reflecting the fact that the HEp-2 method detects autoantibodies against the cytoplasm and mitotic structures too.⁵

Where Labs Struggle With Accuracy

A survey of laboratory performance showed a telling gap in how well technologists identify patterns depending on what part of the cell is involved. Accuracy for recognizing nuclear patterns was excellent, at about 99%. But for cytoplasmic patterns, accuracy dropped to about 78%, and mitotic patterns fell in between at roughly 93%.⁶ The reason is straightforward: nuclear patterns are the bread and butter of ANA testing, and technologists see them constantly. Cytoplasmic patterns are rarer, less standardized, and often subtler. When labs misread these patterns, patients can end up with delayed diagnoses or unnecessary follow-up testing.

One specific pattern that creates confusion is the dense fine speckled (DFS) pattern. It looks alarming under the microscope because it brightly stains both the nucleus and the area around chromosomes during cell division. But a study examining this pattern found that among samples with a standard DFS pattern at low titer, only one patient actually had an autoimmune disease.⁷ The DFS pattern is now recognized as one of the most common “false alarm” results, frequently appearing in healthy individuals. The antibody responsible, anti-DFS70, is actually considered a marker of health rather than disease in most contexts.

Technical Factors That Change Results

Something as seemingly mundane as how the cells are preserved on the slide can meaningfully affect what the test picks up. HEp-2 cells must be fixed (chemically preserved) before serum is applied, and the choice of fixative matters. Research comparing different fixation methods found that aldehyde fixatives preserved cell structures best, while acetone fixatives caused notable changes in cell and nuclear shape.⁸

This matters most for detecting certain antibodies. A study comparing acetone fixation to an alcohol-acetone mixture found that pure acetone produced 97.5% sensitivity for detecting anti-SSA/Ro antibodies, while the alcohol-acetone mix dropped to only 81.3%.⁹ Anti-SSA/Ro antibodies are clinically important because they are associated with Sjögren’s syndrome and neonatal lupus, and missing them can delay treatment. The practical takeaway is that two labs using different commercial HEp-2 slide preparations may not produce identical results for the same patient sample, which is a known source of diagnostic confusion.

The ICAP Standardization Effort

Recognizing that pattern interpretation varied too much between labs and between countries, the International Consensus on ANA Patterns (ICAP) initiative was established in 2014 during a workshop in São Paulo, Brazil.¹⁰ ICAP created a standardized classification system with numbered codes for each fluorescence pattern. A homogeneous nuclear pattern, for instance, became AC-1. A centromere pattern became AC-3. Cytoplasmic and mitotic patterns each received their own codes.

The system now includes over two dozen classified patterns, organized by whether they are competent (strongly associated with autoimmune disease and clinically useful to report) or expert-level (real but rare, requiring specialized interpretation). ICAP also provides an online resource where technologists can compare their slides against reference images, which has helped reduce disagreement between labs.

Engineered Variants of HEp-2

Standard HEp-2 cells express many antigens, but not all of them in high enough quantities for reliable detection. One notable gap is the Ro60 antigen, a target of anti-SSA/Ro antibodies. To address this, a modified version called HEp-2000 was created by genetically engineering HEp-2 cells to overexpress the Ro60 protein. A study of anti-SSA/Ro-positive patients found that HEp-2000 detected the antibody with about 77% sensitivity and correlated well with traditional precipitation methods.¹¹

HEp-2000 slides are used as a supplemental tool rather than a replacement for standard HEp-2. In some labs, they serve as a bridge between the initial IIF screening and confirmatory testing with specific antibody assays.¹² The concept of engineering the substrate to improve detection for specific targets is an active area of development, though standard HEp-2 remains the foundation.

HEp-2 IIF Versus Newer Automated Methods

The fluorescence microscope approach, for all its diagnostic richness, is labor-intensive and subjective. This has driven development of alternative platforms, including ELISA (a plate-based method) and multiplex bead assays that can test for many specific antibodies simultaneously in an automated fashion. The question of whether these newer methods can replace HEp-2 IIF has been debated for years, and the evidence suggests each approach has different strengths.

In lupus patients specifically, IIF on HEp-2 consistently shows higher sensitivity than ELISA or multiplex methods. One head-to-head comparison found that IIF had a diagnostic sensitivity of 0.97 for lupus, versus 0.80 for ELISA and 0.83 for a multiplex platform. The false-negative rate was only 3% for IIF, compared to 20% for ELISA.¹³ A separate study found similar sensitivity between IIF and ELISA for lupus detection but noted that IIF had lower specificity, meaning it was more likely to flag healthy people as positive.¹⁴

In pediatric patients with juvenile idiopathic arthritis, the discrepancy was even more striking. IIF identified nearly twice as many positive patients as a multiplex assay: 98 IIF positives versus 48 multiplex positives.¹⁵ This happens because IIF detects antibodies against the full range of cellular antigens, while ELISA and multiplex panels only test for the specific antigens included in their design. An antibody targeting an unusual or less-characterized antigen will show up on an HEp-2 slide but be invisible to a panel that does not include that antigen.

The trade-off is that IIF’s broad sensitivity comes at the cost of specificity and labor. ELISA and multiplex methods are faster, fully automated, and less dependent on individual technologist skill. Many labs now use a tiered approach: HEp-2 IIF for initial screening, followed by specific antibody assays to identify exactly which autoantibodies are present.

Cross-Species Reactivity

An underappreciated feature of HEp-2 cells is that they work across species boundaries. Research using mouse and rat models of autoimmune disease demonstrated that human HEp-2 cells effectively detected autoantibodies in mouse and rat serum, with no difference in diagnostic performance compared to species-matched cell substrates like mouse neuroblastoma cells or Chinese hamster lung cells.¹⁶ This cross-reactivity exists because the nuclear proteins targeted by autoantibodies are highly conserved across mammals. It also makes HEp-2 IIF a valuable tool in pharmaceutical development, where companies need to screen drug candidates for the potential to trigger autoimmune side effects in animal models.

Which Diseases Depend on This Test

A positive HEp-2 IIF result is formally built into the classification criteria for several autoimmune conditions. It is part of the classification system for systemic lupus erythematosus and juvenile idiopathic arthritis, and it also features in the diagnostic criteria for autoimmune hepatitis and primary biliary cholangitis.¹⁷ In other words, for lupus and these other conditions, you cannot fully meet the diagnostic or classification criteria without an ANA result, and the recommended method for obtaining that result is HEp-2 IIF.

This does not mean the test is useful for every patient with vague symptoms. In children especially, the ANA test is over-ordered. A widely cited pediatric review recommended against ordering the ANA test in most children, advising that if it is ordered and comes back positive at a low titer (below 1:640), the result can be safely ignored in an otherwise healthy child without features of systemic illness.¹⁸ The problem is that a positive result causes anxiety in families and can lead to unnecessary specialist referrals, imaging, and follow-up testing, all triggered by a test that probably should not have been ordered in the first place.

Artificial Intelligence and the Future of Pattern Reading

The subjectivity of human pattern reading has made HEp-2 IIF a natural target for automation. Computer-aided systems that photograph slides and use algorithms to classify patterns have been in development for years, and recent deep-learning approaches have pushed accuracy to impressive levels. One supervised-learning framework designed to recognize all 14 ICAP-recommended pattern classes achieved about 92% accuracy on validation data and 87% on an independent test set, which surpassed the performance of human examiners evaluated on the same images.¹⁹

These systems are not yet replacing human readers in most clinical labs, but they are beginning to serve as a second opinion, flagging discrepancies between the algorithm’s read and the technologist’s call. The hope is that automation will reduce the variability between labs and make the test more consistent, especially for the cytoplasmic and mitotic patterns where human accuracy lags behind. The AI approach also opens the door to standardized digital image archives, so that a slide read in one country could be electronically reviewed by a specialist anywhere in the world.

Cost and Accessibility Considerations

IIF on HEp-2 requires a fluorescence microscope, trained personnel, and commercial slide kits, all of which add up. ELISA-based ANA testing is cheaper per sample and can be run on standard automated chemistry analyzers that most labs already own. A comparison of the two approaches found that ELISA had lower sensitivity than IIF (roughly 59% versus 80% in one study, and about 51% versus 78% when measured against a confirmatory antibody profile), while specificity was comparable or slightly better for ELISA.²⁰

In resource-limited settings, this creates a genuine dilemma. ELISA misses more true positives, but IIF costs more and requires skilled personnel who may not be available in every facility. Some smaller labs have adopted ELISA as their primary screen and only send ambiguous or high-clinical-suspicion samples out to reference labs for HEp-2 IIF. This is a pragmatic compromise, but it means that in parts of the world where autoimmune disease is common but laboratory infrastructure is thin, patients with real disease may receive false-negative screening results simply because the more sensitive test was not available locally.