• systemic lupus erythematosus
• treatment
• disease activity

We very much appreciate the comments of Willems et al 1 on our article on the variability of antinuclear antibody (ANA) testing for patients with established systemic lupus erythematosus (SLE).2 Others have provided their perspective on test variability.3–6 In their letter, Willems et al discuss the importance of antibody titre especially in the context of patient classification. In the new classification criteria under development, a positive ANA at a titre of 1:80 or higher is required.7 In view of our findings on assay variability, we have also commented on the use of serology as the initial element in classification.8 Without specification of the kit used and its performance characteristics, uncertainty about this key element of classification can occur.

The study presented by Willems et al provides valuable data on ANA detection by two assay formats—an immunofluorescence assay or IFA and a solid phase assay—for the broad range of patients with autoantibody-associated systemic rheumatic diseases (AASRD). Of note, among their patients with established SLE, 10% showed negative results, consistent with our findings2; interestingly, six out of eight of the IFA-negative patients showed antibody positivity with the solid phase assay. Together, these studies highlight the complexity of serological testing and emphasise that, whatever the purpose of such determinations—disease classification or clinical trial eligibility—it is important to take into account the performance characteristics of any particular assay, including the titres obtained.

Serological testing is a key element in the diagnosis and management of patients with AASRD because of the utility of the ANA as a biomarker especially when coupled with assays for antibodies to specific antigens. Unfortunately, standardisation remains an issue and we look forward to further studies like those of Willems et al to alert the field to a problem that is often not adequately recognised especially if the IFA is called the ‘gold standard’. Despite decades of investigation, the field is not yet at the point where any one kit or assay is uniformly successful and has achieved the ‘gold standard’ status. Pending more technological advances, the use of a combination of formats (ie, IFA and solid phase) may be the best solution to fulfil the need to classify patients and identify those who are eligible for new treatments.