Prevalence and serological profile of anti-DFS70 positive subjects from a routine ANA cohort

Anti-Dense Fine Speckled 70 (DFS70) antibodies are a common finding in clinical laboratory referrals. High prevalence of DFS70 autoantibodies in healthy population and usual negative association with Antinuclear Antibody (ANA)-associated autoimmune rheumatic diseases (AARD) were reported. The aim of this study was to evaluate the prevalence of DFS70 autoantibodies and their association with other autoantibodies in the context of a routine ANA referral cohort. Consecutive sera submitted for ANA screening were analyzed for anti-DFS70 antibodies by indirect immunofluorescence (IIF) (n = 3175, 1030 men and 2145 women) then confirmed by immunoblotting. Anti-DFS70 positive samples were also assayed for a large spectrum of other circulating autoantibodies. The prevalence of anti-DFS70 antibodies was 1.7% in the whole population and 4.6% in the ANA-positive samples. Comparison between DFS70 IIF and immunoblotting showed an excellent correlation between the two methods. The prevalence of anti-DFS70 positive was significantly higher in females (2.1%, 45/2145) than in males (1.0%, 10/1030). Of note, no concomitant autoantibodies were found in the DFS70-positive male group compared with DFS70-positive females group that showed other serum autoantibodies in the 51% of cases. Anti-DFS70 reactivity in male population may represent an useful biomarker predicting the absence of other autoantibodies. On the contrary, the serological profile of DFS70-positive females required further investigations in order to define the presence of concomitant disease-marker autoantibodies.

Anti-DFS70 antibodies were initially described in patients with interstitial cystitis and later in heterogeneous chronic inflammatory conditions, tumours and even in apparently healthy individuals, but their clinical impact is still unknown [25][26][27][28] . It was reported that none of anti-DFS70 positive subjects showed symptoms suggestive of an AARD after clinical follow-up of 4 years 18 .
Some previous studies suggested that the isolated anti-DFS70 reactivity could be taken as biomarker to exclude AARD (Antinuclear Antibody (ANA)-associated autoimmune rheumatic diseases) from ANA-positive healthy individuals 9,11,[29][30][31] . Understanding the serological and clinical profile of anti-DFS70 positive subjects thus could avoid inappropriate referral to care specialists and follow-up tests for healthy people. Accordingly, the aim of this study was to define the prevalence of anti-DFS70 antibodies in a routine diagnostic laboratory setting and the associated serum autoantibodies to support the clinical use of these markers.

Methods
Samples. Collected 30. Positive/negative interpretation based on a F.I. was evaluated during focusing and confirmed by expert-readers. Positive samples were also tested using the conventional serial dilution method (1:160, 1:320, 1:640, 1:1280 until to 1:2560) to assign end-point titer. The titer was defined as the reciprocal of the highest dilution of serum that still shows immunofluorescent staining.
Immunoblot Anti-DFS70 Assay. The serum samples positive for DFS70-like pattern in IIF were further processed for detection of human IgG autoantibodies against DFS70 specificity (truncated sequence of the DFS70 antigen (residues 349-435)) by Immunoblotting assay. In addition, confirmed anti-DFS70 positive samples, were also evaluated for the antibodies against Sm, U1-RNP, Sm/RNP, SSA/Ro60kD, SSB, Scl-70, PM-Scl 100, Ku, CENP-A/B, PCNA, Mi-2 antigens using immunoblotting kit (Alphadia, Wavre, Belgium) by automatic Blu Diver Instrument (BDI) (Alifax, Polverara (PD), Italy). During the automated test procedure, the BDI sequentially incubates the strips in the wells of ready-to use reagent cartridges. Human antibodies bind the corresponding specific antigen(s) on the membrane and enzyme activity leads to development of purple dots on the membrane pads. The intensity of the coloration is directly proportional to the amount of antibody present in the sample. Semi-quantitative interpretation was done by Dr Dot Software (Alifax) and scanning system using Arbitrary Unit (AU).
Statistical analysis. Statistical analyses were performed using the Graph Pad Prism statistical package (v5; Graph Pad Software, San Diego, CA). Statistical significance was calculated using the non-parametric Mann-Whitney-U test and correlations were analyzed by the Spearman's rank correlation test and R-squared analyses (R 2 ); categorical variables were compared using Fisher' Test. A p value < 0.05 was considered statistically significant.
Ethics approval and consent to participate. The study was approved by the Regional Ethics Committee    the typical DFS70 staining pattern. Titer distribution revealed no statistically differences when compared with other ANA patterns (p > 0.05). More than half of the patients presenting DFS70 monopattern (63% of the total cohort) showed high titers (≥1:640) (Fig. 3). A quantitative comparison between DFS70 IIF titers and AU immunoblotting showed an excellent correlation between two methods as expressed by a Spearman correlation (Fig. 4A). Regression analysis (R 2 = 0.99) also confirmed the relationship between IIF titers and immunoblotting AU (Fig. 4B).

Association of anti-DFS70 and other serum autoantibodies.
Collected DFS70-positive serum samples were assayed for a large spectrum of other circulating autoantibodies. By comparing DFS70-positive males with females, it emerged that isolated anti-DFS70 reactivity was found only in the male group. In contrast, in the 51% of DFS70-positive females, concomitant serum autoantibodies were found. Most common detected autoantibodies were anti-TPO (16.0%), anti-TG (11.0%), anti-tTg-IgA (9.0%) and ANCA (2.

Discussion
Previous studies using ANA-IIF test on HEp-2 substrates provide the detection of high-titer autoantibodies with typical DFS70 staining pattern in 1% to 16% of the studied cohorts. These heterogeneous results were obtained in different patient cohorts, including blood donors, healthy individuals, routine ANA cohorts, selected ANA positive healthy individuals and routine ANA positive subjects [9][10][11][12][13][14][15][16][17][18][19][20][21][22] . In details, a rate of positivity ranged from 0 to 5% was found in blood donors, healthy children and in routine ANA screening cohort. Higher frequencies were reported in ANA-positive populations. In addition, some studies examined serum samples only by IIF without specific confirmatory assays, resulting in higher rate of positivity. It has been showed that not all samples displaying DFS70 IIF pattern were then confirmed by specific CLIA or ELISA methods 23,32,33 . We analyzed 3175 consecutive unselected samples screened for ANA during the routine work-up by IIF at screening dilution of 1:160, then confirmed by immunoblotting (Fig. 1). We found 1.7% of anti-DFS70 positive sera, in line with some previous studies conducted on routine ANA screening population 6,10,13,16,19,21,34 . In fact, Mahler et al. 6 analyzed 3263 sera submitted for ANA screening for the presence of anti-DFS70 antibodies and observed 1.62% of positivity. Noteworthy, Bizzaro et al. 15 also reported a low frequency (0.8%) of sera displaying the anti-DFS70 reactivity on HEp-2 substrate in a large cohort of samples screened for ANA in clinical laboratory. However, because our study was performed with ANA-requested specimens from outpatients and inpatients, the overall frequency of anti-DFS70 antibodies in general population is unknown. Gender differences with greater prevalence of female (4:1) was observed in anti-DFS70 negative/ANA positive group and within the anti-DFS70 positive samples. This is an important finding since the increased immune reactivity in females predisposes them to developing AARD. A number of factors may underlie this striking gender difference and further studies are needed to better understand these findings.
Regarding referring sources of anti-DFS70 positive samples, a comparative analysis highlighted that greater prevalence of anti-DFS70 positive subjects referred from the outpatients, with value of 2.1% (Fig. 2).
We found high titers (≥1:640) in 64.3% of the DFS70 IIF-positive sera, with a titer greater than 1:2560 found only in one sample. Titer distribution revealed no statistically differences when compared with other ANA positivity (p > 0.05) (Fig. 3).
The use of specific methods to confirm DFS70 IIF suspicion was highly recommended. Recent data showed that there were no differences in diagnostic accuracy among methods using truncated (as Alphadia Dot) or full-length DFS70 antigenic sequence 34,35 . We perform a quantitative comparison among IIF titers and immunoblotting AU in anti-DFS70 positive sera resulting in excellent correlation (Fig. 4A,B).
We also analyzed the simultaneous presence of a broad spectrum of other circulating disease-markers autoantibodies. Muro et al., found a high prevalence of disease-related autoantibodies in anti-DFS70 positive patients, but of note, their cohort included only 22 anti-DFS70 positive females. Noteworthy, our data evidenced isolated reactivity of anti-DFS70 autoantibodies in male group, and high prevalence of disease-marker autoantibodies in females (51%). In particular, most common detected autoantibodies were anti-TPO, anti-TG, anti-tTg-IgA and ANCA (Fig. 5). It is hoped that future studies will compare the prevalence of concomitant antibodies in males versus females either in ANA positive/anti-DFS70 negative group.
Major strengths of our study are the use of anti-DFS70 confirmatory assay to avoid false positive results and the gender analysis in appraisal of the anti-DFS70 antibodies prevalence. The prevalence of anti-DFS70 was previously assessed in different cohorts mostly smaller than ours, therefore, we believe that our anti-DFS70 positive cohort should be considered suitable for a preliminary evaluation although a future larger confirmatory study is needed. In contrast, the scarce number of anti-DFS70 positive males could be considered inadequate to draw any firm conclusions. Therefore, based on low male prevalence, it is mandatory to screen larger cohort by future multicentre studies.

Conclusion
According to previous observations performed on routine ANA cohorts, our prevalence of anti-DFS70 positivity was 1.7% of adult screened population. Serological profile of anti-DFS70 positive females required further clinical and laboratory investigations in order to define the presence of associated disease-marker autoantibodies and their clinical significance. In contrast, isolated anti-DFS70 specificity in male population suggests that the DFS70 could be considered reliable screening indicator for absence of other circulating autoantibodies resulting in considerable cost-saving potential.

Data Availability
The dataset generated and analyzed during the present study are not publicly available, owing to restrictions by our organization, but they are available from the corresponding author on reasonable request.