Rethinking HLA-B27 Testing: What HLA-B27 Can—and Cannot—Tell Us

HLA-B27: a cornerstone in spondyloarthritis

Nearly 50 years after its discovery as a major genetic risk factor for spondyloarthritis (SpA),¹ the precise contribution of HLA-B27 to disease development and progression remains only partially understood. Proposed mechanisms include aberrant antigen presentation, misfolding-induced endoplasmic reticulum stress and modulation of innate immune responses.² Despite these uncertainties, HLA-B27 is one of the only diagnostic biomarkers routinely used in clinical practice and is part of the most recent classification criteria for both axial SpA (axSpA) and peripheral SpA (pSpA).^3,4 HLA-B27 is also incorporated into several referral strategies aimed at assisting nonrheumatologists in identifying individuals who may benefit from rheumatology evaluation for suspected SpA.⁵ HLA-B27 also holds prognostic and theragnostic value, given its association with greater structural damage, higher inflammatory burden, and better response to tumor necrosis factor and interleukin 17 inhibitors.⁶

Yet, like any biomarker, it must be interpreted with a clear understanding of its limitations. Its sensitivity is incomplete (approximately 70-90% of patients with axSpA are HLA-B27 positive⁷), and its specificity is limited by a relatively high frequency in the general population (8-10% among individuals of European ancestry).⁸ This frequency varies considerably across ethnic groups,⁹ introducing potential biases in interpretation based on ancestry.¹⁰ Most importantly, the diagnostic value of HLA-B27 depends heavily on pretest probability, which itself is strongly influenced by the highly polymorphic clinical presentation of SpA. The test is most informative in patients with typical axSpA features, such as inflammatory back pain in young adults, but its diagnostic value is more limited in cases with peripheral arthritis or when SpA is associated with psoriasis or inflammatory bowel disease, where the association with HLA-B27 is weaker. To reach a clinically meaningful posttest probability (eg, > 90%), the physician’s prior suspicion should already be above 50%.¹¹ As shown by Rudwaleit et al, in patients with chronic back pain, at least 2-3 SpA features are typically needed to reach a 90% diagnostic probability.¹² In this context, HLA-B27 was one of the features with the highest positive likelihood ratio (LR+ ~9). However, a more recent metaanalysis reported lower LR+ values for HLA-B27 of around 3.¹³ It is also worth noting that HLA-B27–positive SpA cases may represent a more phenotypically homogeneous subgroup, with a more consistent pattern of axial involvement and inflammatory burden. Taken together, these data reinforce that HLA-B27, although valuable, should never be interpreted in isolation.

Lessons learned from HLA-B27 referral patterns

In this issue of The Journal of Rheumatology, McDermott and colleagues report the indications for HLA-B27 testing and subsequent rheumatology referral patterns through a retrospective analysis of 1960 HLA-B27 tests over 1 year in a North American hospital.¹⁴ They found that HLA-B27 testing was more frequently prescribed by nonrheumatologists, often for a broad and poorly defined range of indications, including unexpected ones such as temporomandibular joint disease. Back pain, the hallmark symptom of axSpA, ranked only third among testing indications; many tests were instead ordered for peripheral arthritis or temporomandibular joint symptoms, which carry lower predictive value. Additionally, HLA-B27 was frequently cotested with antinuclear antibody, anticyclic citrullinated peptide, and rheumatoid factor, suggesting its use as part of a general rheumatologic screen rather than a targeted diagnostic tool. Notably, in most clinical contexts where testing was performed, the HLA-B27 positivity rate fell within the confidence interval of its background prevalence in the general population (5-10%), underscoring the likely inappropriate use of the test without sufficient clinical suspicion to provide meaningful diagnostic value.

Beyond issues of test indication, the findings of McDermott et al¹⁴ reveal a concerning disconnect between test results and subsequent clinical decision making. Among patients who tested positive for HLA-B27, only 53% were referred to rheumatology when the test had been ordered by nonrheumatologists,¹⁴ suggesting that even when results point to a potential diagnosis of SpA, appropriate follow-up is frequently lacking. The problem was particularly evident in ophthalmology, a specialty where the link between HLA-B27 and acute anterior uveitis (AAU) is long-established and well documented.¹⁵ Despite this, 60% of HLA-B27–positive patients in ophthalmologic settings were not referred for rheumatologic evaluation. Just as concerning is the inverse scenario: patients who tested negative for HLA-B27 but presented with clinical features suggestive of SpA were sometimes not pursued further, potentially missing diagnoses that do not depend on HLA-B27 positivity. These patterns highlight the dual risk of misapplication: a false sense of reassurance from negative results, and failure to act upon positive ones. Both outcomes delay diagnosis, hinder timely treatment, and can ultimately compromise patient outcomes.

HLA-B27 testing in clinical practice: when, how, and what next?

When to test. As already mentioned, the clinical utility of HLA-B27 testing is highly dependent on pretest probability. As summarized in the Table, testing is most informative in the presence of features suggestive of SpA, such as inflammatory back pain in young adults, AAU, or characteristic patterns of arthritis and enthesitis. Conversely, indiscriminate use of HLA-B27 testing, such as in mechanical back pain, isolated joint complaints without SpA features, or as part of a nontargeted rheumatologic screen, offers little diagnostic value and may contribute to misinterpretation or unnecessary referrals.

A common question from patients with HLA-B27–positive SpA is whether their relatives should be screened. Routine HLA-B27 testing in asymptomatic family members is not recommended, as most carriers never develop SpA—even among first-degree relatives.¹⁶ Without symptoms, a positive result does not change management and may lead to unnecessary anxiety or investigations. Instead, patients should be encouraged to inform relatives, particularly children and adolescents, about key warning signs such as inflammatory back pain, joint swelling, or AAU. If such symptoms occur, clinical evaluation and targeted HLA-B27 testing may then be appropriate.

How to test. HLA-B27 testing is commonly performed using flow cytometry or molecular techniques such as PCR-based genotyping. Flow cytometry detects surface expression of the HLA-B27 antigen using monoclonal antibodies and is widely available, but it carries a risk of false-positive or false-negative results due to cross-reactivity with other HLA-B alleles or low antigen expression. Molecular typing offers greater specificity and sensitivity by directly identifying the HLA-B27 gene.¹⁷ In cases where flow cytometry results are discordant with the clinical presentation, confirmatory testing with a molecular method may be warranted.¹⁸ Additionally, before ordering an HLA-B27 test, clinicians should verify whether the test has already been performed—either in prior specialist evaluations or documented in medical records—to avoid redundant testing and help reduce unnecessary healthcare costs.

What next? As highlighted in the study by McDermott et al,¹⁴ HLA-B27 test results do not always lead to appropriate follow-up. A positive result should not automatically prompt rheumatology referral, but it should at least lead clinicians to consider the possibility of SpA, particularly in contexts such as ophthalmology, where this association is well known. Conversely, a negative HLA-B27 result does not rule out SpA and should not preclude referral when musculoskeletal symptoms are present. Rheumatology evaluation remains valuable to assess for HLA-B27–negative SpA or to consider alternative musculoskeletal diagnoses.

How to improve HLA-B27 use in clinical practice

Enhancing the clinical utility of HLA-B27 testing requires a more judicious, context-sensitive approach to both its use and interpretation. A first step is targeted education for nonrheumatologists, particularly general practitioners, ophthalmologists, dermatologists, and gastroenterologists, all of whom frequently encounter early features of SpA. Training should clarify when testing is appropriate, how to interpret results based on clinical context, and when to refer to rheumatology. Collaborative initiatives between rheumatology societies and other professional bodies can help standardize messaging and promote shared referral tools.

Beyond education, institutional and laboratory systems have a role to play. Embedding decision support tools into electronic ordering systems such as symptom checklists, alerts for prior test results, or prompts based on pretest probability, can reduce inappropriate and redundant testing. Audit and feedback mechanisms can further reinforce best practices, monitor guideline adherence, and identify areas for improvement. Together, these measures can help improve the use of HLA-B27 testing.