Statistics from Altmetric.com
The recent article by Pisetsky et al 1 showing data derived from a comparison of different antinuclear antibody (ANA) assays in a cohort of patients with established systemic lupus erythematosus highlighted the critical issue of ANA detection. With great interest, we read the correspondence by van Hoovels et al 2 describing variation in ANA detection by automated indirect immunofluorescence (IIF) analysis and the following letter from Mahler and Auza on the strong need for ANA testing standardisation.3 Nowadays, the IIF on human epithelial cells (HEp-2) is considered the gold standard for ANA testing, but biological and non-biological issues limiting the IIF test are not currently adequately clarified in study literature.4 5 In fact, despite the fact that the IIF method on HEp-2 cells has existed for about 40 years, there are little data on this topic, and studies are mainly focused on selected patient groups, rather than samples from ‘real life’ routine laboratory work. This bias could influence recommendations and have important clinical and diagnostic implications. Indeed, an improper clinical interpretation of the test can lead to a misdiagnosis, inappropriate therapies and unnecessary costs. Given the cost of a single ANA testing and the direct/indirect cost of an incorrect systemic rheumatic diseases diagnosis, the cost–benefit clearly weights the balance in favour of the test. Real-life laboratory work emphasises how important it is that international scientific guidelines/recommendations take the IIF interobserver and intraobserver reading variability into consideration. For the same reason, we find inappropriate and outdated comparing IIF sensitivity with solid phase assays (SPAs) regardless of IIF ANA testing interobserver variability evaluation.6 Recently, we conducted a large multicentre evaluation of the interobserver reading variability on 556 consecutive samples, for a total of 1679 images, collected in three laboratories with IIF expertise, using HEp-2 cell substrate at 1:80 screening dilution.7 Data have revealed a pairwise agreement little further away than substantial both for fluorescence intensity and for staining pattern recognition (k=0.602 and k=0.627, respectively). Moreover, since the readers involved in our study were very experienced in the field, the data presented may not represent a typical laboratory scenario.
Two additional critical issues concern the ANA cytoplasmic reporting and the IIF borderline zone (ranging from 1:80 to 1:160 titre). On one hand, there is no consensus as to whether cytoplasmic pattern is to be considered ANA positive or negative, which worsens the ANA variability problem. On the other hand, the weak positive samples carry most of the variability and are rarely represented in study literature; specifically, our study showed a poor specific agreement (k=0.430). The poor agreement on weak positive samples may impact the evaluation of computer-aided diagnosis (CAD) system application in the analysis of positive/negative screening. In light of this, the use of a CAD system may strongly help to overcome most of the IIF drawbacks and should be considered a reliable standardisation tool that could reduce interlaboratory variability.8 Additionally, it may help to reduce the nomenclature discrepancies between laboratories when describing IIF patterns, and it can represent an objective criterion for a titer assignment. Despite the research and industrial efforts directed towards the development of CAD systems in IIF, the performances observed during routine laboratory use are not fully satisfactory. We therefore deem that their use as second readers should be encouraged, but meanwhile firm and independent assessment is needed. Under real-life conditions, the double-blind reading method is not wholly respected today and the CAD system could represent a tool that supports laboratories but is increasingly widespread with no scientific recommendation ad hoc. Furthermore, we want to underline that even if the International Consensus on ANA Patterns have recommended a screening dilution of 1:160 for ANA testing,9 in real life many laboratories still continue working at 1:80 dilution because we know how difficult it is for both clinicians and laboratorists to accept and effectively integrate changes in this field. Additionally, most of the important current studies on cost-effectiveness of ANA with SPAs or ANA with CTD screen are done considering an ANA screening titre of 1:80, and should be reconsidered after the 1:160 ANA titre calculation to make the studies’ messages current. We thank the authors for opening the discussion and hope that future projects with specific and proper study designs will help scientific guidelines work in harmony with real-life laboratories.
Handling editor Josef S Smolen
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent Not required.
Provenance and peer review Not commissioned; internally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.