Antinuclear Antibodies Testing Method Variability: A Survey of Participants in the College of American Pathologists' Proficiency Testing Program

Recognition of rare antinuclear antibody patterns based on a novel attention-based enhancement framework

Rare antinuclear antibody (ANA) pattern recognition has been a widely applied technology for routine ANA screening in clinical laboratories. In recent years, the application of deep learning methods in recognizing ANA patterns has witnessed remarkable advancements. However, the majority of studies in this field have primarily focused on the classification of the most common ANA patterns, while another subset has concentrated on the detection of mitotic metaphase cells. To date, no prior research has been specifically dedicated to the identification of rare ANA patterns. In the present paper, we introduce a novel attention-based enhancement framework, which was designed for the recognition of rare ANA patterns in ANA-indirect immunofluorescence images. More specifically, we selected the algorithm with the best performance as our target detection network by conducting comparative experiments. We then further developed and enhanced the chosen algorithm through a series of optimizations. Then, attention mechanism was introduced to facilitate neural networks in expediting the learning process, extracting more essential and distinctive features for the target features that belong to the specific patterns. The proposed approach has helped to obtained high precision rate of 86.40%, 82.75% recall, 84.24% F1 score and 84.64% mean average precision for a 9-category rare ANA pattern detection task on our dataset. Finally, we evaluated the potential of the model as medical technologist assistant and observed that the technologist's performance improved after referring to the results of the model prediction. These promising results highlighted its potential as an efficient and reliable tool to assist medical technologists in their clinical practice.

Reliability and reproducibility of antinuclear antibody testing in pediatric rheumatology practice

Antinuclear antibody (ANA) testing is common practice among health care practitioners when evaluating children and adolescents with non-specific symptoms including fatigue and aches and pains. When positive, ANA results often lead to referrals to pediatric rheumatologists as these antibodies may be key indicators for specific pediatric rheumatologic diagnoses. The reliability and reproducibility of ANA tests varies with assay techniques and validation and interpretation of results. In the following article, review of ANA testing in pediatrics is provided along with case examples that demonstrate the reliability and reproducibility of these results in specific scenarios common in the practice of pediatric rheumatology. Guidelines for more accurate utilization of ANA testing are presented with the aim to improve testing and interpretation by ordering clinicians.

Anti-Nuclear Antibody Quantitation: Calibration and Harmonization Adjustment via Population Interrogation

BACKGROUND

The 2019 classification criteria for systemic lupus erythematosus (SLE) includes an initial criterion requiring the presence of an antinuclear antibody (ANA), positive at a titer of at least 1:80 on HEp-2 cells, or equivalent. However, results of ANA tests performed on HEp-2 cells vary when tested in different laboratories. Calibration of ANA assays by achieving a common specificity in healthy control populations offers the possibility of achieving harmonization via population interrogation, but the expected specificity in a healthy control population is not known.

METHODS

The studies used to determine the use of ANAs performed by immunofluorescence microscopy on HEp-2 cells as the entry criterion for classification of SLE were reanalyzed by a meta-analysis to determine the expected frequency of positive ANAs in healthy control populations at serum dilutions of 1:40 and 1:80.

RESULTS

Our meta-analysis demonstrated that the expected specificity in a healthy control population of ANA performed using serum diluted 1:80 is 91.3% (CI 86.1-94.7%). The expected specificity of ANA performed at 1:40 serum dilution is 79.2% (CI 72.3-84.8%).

CONCLUSION

One approach to achieving harmonization of ANA assays from different laboratories with each other and with expected performance would involve adjusting assays so that about 10% of a healthy control population has a positive ANA when tested at 1:80 dilution, and about 20% of the healthy control population has a positive ANA when tested at 1:40 dilution. This pragmatic approach to calibration and harmonization adjustment via population interrogation offers an opportunity for individual laboratories to be aligned with each other and with ANA performance expected for consistent categorization of patients with SLE.

In vitro diagnostics for the medical dermatologist. Part I: Autoimmune tests

Despite the expansion of available in vitro laboratory tests at a rate far exceeding that of dermatologic pharmaceuticals, the existing literature is dominated by discussion of the latter. With the advent of numerous new tests, it can be difficult for practicing dermatologists to stay up-to-date on the available options, methodologies, and recommendations for when to order one test over another. Understanding the inherent strengths and weaknesses of these options is necessary to inform appropriate ordering and proper interpretation of the results. The first article in this continuing medical education series summarizes information on methodology, test characteristics, and limitations of several in vitro laboratory tests used for the work up of undifferentiated patients suspected of having dermatologic autoimmune diseases and it provides a general guide to ordering these tests.

Application of Supervised Machine Learning to Recognize Competent Level and Mixed Antinuclear Antibody Patterns Based on ICAP International Consensus

Background: Antinuclear antibody pattern recognition is vital for autoimmune disease diagnosis but labor-intensive for manual interpretation. To develop an automated pattern recognition system, we established machine learning models based on the International Consensus on Antinuclear Antibody Patterns (ICAP) at a competent level, mixed patterns recognition, and evaluated their consistency with human reading. Methods: 51,694 human epithelial cells (HEp-2) cell images with patterns assigned by experienced medical technologists collected in a medical center were used to train six machine learning algorithms and were compared by their performance. Next, we choose the best performing model to test the consistency with five experienced readers and two beginners. Results: The mean F1 score in each classification of the best performing model was 0.86 evaluated by Testing Data 1. For the inter-observer agreement test on Testing Data 2, the average agreement was 0.849 (κ) among five experienced readers, 0.844 between the best performing model and experienced readers, 0.528 between experienced readers and beginners. The results indicate that the proposed model outperformed beginners and achieved an excellent agreement with experienced readers. Conclusions: This study demonstrated that the developed model could reach an excellent agreement with experienced human readers using machine learning methods.

Dr. Naides replies

We thank Dr. Russell for raising the issue of reporting the false positivity rate of antinuclear antibody (ANA) indirect immunofluorescent assay (IFA) testing.1 It is difficult, however, for a laboratory to state a false positive rate, per se, as the determination of “falseness” is dependent on clinical evaluation that is typically not available to most laboratories.

The Variability of Antinuclear Antibody Testing

The group from the College of American Pathologists1 have carried out an extensive survey of the approaches to antinuclear antibody testing by different laboratories, predominantly in the USA. They reviewed the techniques used and possible reasons for the variations in results.

Antinuclear Autoantibodies in Health: Autoimmunity Is Not a Synonym of Autoimmune Disease

The incidence of autoimmune diseases is increasing. Antinuclear antibody (ANA) testing is a critical tool for their diagnosis. However, ANA prevalence in healthy persons has increased over the last decades, especially among young people. ANA in health occurs in low concentrations, with a prevalence up to 50% in some populations, which demands a cutoff revision. This review deals with the origin and probable physiological or compensatory function of ANA in health, according to the concept of immunological clearance, theory of autoimmune regulation of cell functions, and the concept of functional autoantibodies. Considering ANA titers ≤1:320 as a serological marker of autoimmune diseases seems inappropriate. The role of anti-DFS70/LEDGFp75 autoantibodies is highlighted as a possible anti-risk biomarker for autoimmune rheumatic disorders. ANA prevalence in health is different in various regions due to several underlying causes discussed in the review, all influencing additive combinations according to the concept of the mosaic of autoimmunity. Not only are titers, but also HEp-2 IFA) staining patterns, such as AC-2, important. Accepting autoantibodies as a kind of bioregulator, not only the upper, but also the lower borders of their normal range should be determined; not only their excess, but also a lack of them or "autoimmunodeficiency" could be the reason for disorders.