Serum Anti-Ro52/Tripartite Motif-containing 21, A Novel Criterion of Interstitial Pneumonia With Autoimmune Features

Background: Antibodies to Ro52/tripartite motif-containing 21 (TRIM21), referred to as anti-Ro52, are found in patients diagnosed with diverse systemic autoimmune rheumatic disease and associated with interstitial lung diseases. However, little is known about the clinical characteristics of anti-Ro52 in patients with idiopathic interstitial pneumonias (IIPs). We aimed to analyze the prevalence, co-existent autoantibodies, and clinical characteristics of anti-Ro52 in patients with IIP. Methods: The study enrolled 288 patients diagnosed with IIP. Serum anti-Ro52 was detected using enzyme-linked immunosorbent assay (ELISA). Co-existent autoantibodies, including anti-aminoacyl-tRNA synthetases (anti-ARS), were analyzed by immunoprecipitation. ELISA using recombinant antigens was performed as needed. Clinical, laboratory, and radiographic ndings and survival of IIP patients were compared between anti-Ro52 positives and negatives. Results: Anti-Ro52 (20/288; 6.9%), anti-ARS (18/288; 6.3%), and anti-Ro60/SS-A (16/288; 5.6%) were the most common autoantibodies detected in IIP patients. Among 20 IIP patients who had anti-Ro52, anti-ARS was present in 8 (40%), and anti-Ro60/SS-A was found in 6 (30%) patients. The criteria for interstitial pneumonia with autoimmune features (IPAF) were signicantly better fullled by patients with anti-Ro52 than those without (P = 0.001). Meeting serological domain (P < 0.001) and Raynaud’s phenomenon (P = 0.009) were signicantly more common in the anti-Ro52-positive patients. There was no signicant difference in the overall survival rate of IIP patients with vs. without anti-Ro52 (log-rank P = 0.51). Conclusion: Anti-Ro52-positive IIP patients have clinical features consistent with IPAF. Adding anti-Ro52 in IPAF criteria may be considered in the future.

Anti-Ro52 is associated with the presence of ILD in SSc, PM/DM, and mixed connective tissue disease (MCTD) [13,15,16,20]. One study in PM/DM reported higher prevalence of ILD in patients with anti-Ro52 than without, however, it might be due to an association of anti-Ro52 and anti-ARS [14]. Another study on patients with anti-Ro52-positive ILD reported the absence of an established diagnosis of SARD in the majority (78%) of patients while nearly half (49.3%) ful lled the IPAF criteria [21]. Further studies are warranted to explore the prevalence, co-existent autoantibodies, and clinical characteristics associated with the presence of anti-Ro52 in IIP patients.
This study aims to analyze the clinical signi cance of anti-Ro52 in patients with IIPs and the associated clinical and immunological characteristics. These ndings may contribute to more accurate classi cation of IIP.

Study participants
A two-center retrospective cohort study was conducted by the Department of Respiratory Medicine, Nagasaki University Graduate School of Biomedical Sciences, Nagasaki, Japan, and the Department of Respiratory Medicine, University of Occupational and Environmental Health, Kitakyushu, Japan. Patients with IIP were enrolled in the study between March 2007 and October 2016 (n = 288). At the rst visit, IIP was diagnosed based on clinical, laboratory, and radiological ndings as per the de nition of the American Thoracic Society/European Respiratory Society (ATS/ERS) international multidisciplinary consensus classi cation [1,2]. The study was conducted in accordance with the amended Declaration of Helsinki. The Institutional Review Board of the Nagasaki University Hospital, Nagasaki, Japan (Approval No: 16042517), and the University of Occupational and Environmental Health, (Approval No: H27-238) approved the protocol. Informed consent was obtained from all subjects. Observation and follow-up of each patient was conducted on an annual basis and was censored on April 30, 2020. Patients who were lost to follow-up were censored at the date of last contact/follow-up and those alive as on April 30, 2020, were censored for overall survival analyses.

Detection of serum autoantibodies
Serum samples of patients were obtained during their rst visit and stored at −20 °C until further use. For the analyses of autoantibodies, 35 S-methionine radiolabeled K562 cell extract was immunoprecipitated with IgG puri ed from 8 µL of human serum samples. The immunoprecipitated proteins were electrophoresed on sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) as described previously [22]. Brie y, cells were labeled with 35 S-methionine and cysteine, lysed in 0.5 M NaCl, 2 mM ethylenediaminetetraacetic acid (EDTA), 50 mM Tris (pH 7.5), 0.3% octylphenyl polyethylene glycol (IGEPAL CA-630) buffer containing 0.5 mM phenylmethylsulfonyl uoride and 0.3 trypsin inhibitory units (TIU)/mL aprotinin, and immunoprecipitated using protein-A-Sepharose beads coated with IgG.
Immunoprecipitates were then washed with 0.5 M NaCl-NET/IGEPAL CA-630 and analyzed by SDS-PAGE and autoradiography. The speci city of autoantibodies was con rmed by the use of human reference sera [22]. Antibodies to Ro52/TRIM21, histidyl-tRNA synthetase (Jo-1), and melanoma differentiationassociated protein 5 (MDA5) were tested by enzyme-linked immunosorbent assay (ELISA) as described previously [22]. All recombinant proteins were purchased from Diarect (Freiburg, Germany). Brie y, 96-well microtiter plates (Immobilizer Amino; Nunc Naperville, IL, USA) were coated with 0.5 μg/ml of recombinant protein and blocked with 0.5% bovine serum albumin (BSA)-NET/ IGEPAL CA-630 for 1 h at room temperature. Patients' sera (1:250) and alkaline phosphatase-conjugated goat anti-human IgG (1:1000; γ-chain speci c; Jackson Immunoresearch, Hershey, PA, USA) diluted in 0.5% BSA-NET/ IGEPAL CA-630 were used as the sample and secondary antibodies, respectively. A standard curve was generated using serial 1:5 dilutions of a high-titer prototype serum. Optical density of samples measured at 405 nm was converted into units based on the standard curve.

Clinical data collection and analyses
Demographic data, clinical information, results of laboratory and pulmonary function tests, and analyses of bronchoalveolar lavage uid were obtained from medical records. Physical ndings were con rmed by rheumatologists when appropriate. Classi cation criteria of IPAF were based on the 2015 ERS/ATS Task Force research statement [5]. During the observation period, none of the patients developed clinical or laboratory features associated with SARD diagnosis. Although patient data (n = 281) used in this study partially overlapped with a retrospective study published earlier [6], our research data related to anti-Ro52 are unique.

Statistical analyses
Data are presented as the median [interquartile range] or frequency (%). Fisher's exact test was used to compare categorical variables. Comparisons between groups were made using the Mann-Whitney U test. Survival analyses were performed using the Kaplan-Meier method and the log-rank test. All analyses were conducted at a signi cance level of α = 0.05. All statistical analyses were performed using the STATA 16.1 software (StataCorp, College Station, TX, USA).

Discussion
This is the rst study investigating the frequency of serum anti-Ro52 antibodies in unselected patients with IIP. Similar to the prevalence of anti-ARS (6.3%), anti-Ro52 was detected in 6.9% of patients with IIP. Presence of serum anti-Ro52 was signi cantly associated with ful llment of IPAF criteria, particularly with respect to the serological domain and Raynaud's phenomenon, in IIP patients.
IPAF criteria (P = 0.001) related to the serological domain (P < 0.001) were more frequently ful lled by anti-Ro52-positive (50%) than anti-Ro52-negative patients (17%) in our IIP cohort. A previous retrospective study showed that 49.3% of the ILD patients who had anti-Ro52 met the IPAF criteria, similar to our result [21]. Anti-Ro52-positive patients could be negative in immuno uorescence ANA tests, however, anti-Ro52 was associated with IPAF serological domain, indicating that it frequently coexists with the other autoantibodies included in the IPAF serological domain (Table 2). Co-existence of anti-Ro52 with other speci c autoantibodies in various SARD have been reported [14,15,18]. Although anti-Ro52 is not speci c for a particular type of SARD diagnosis, a 14-fold increased risk of developing SARD was reported in IIP patients who met the IPAF criteria [25]. Thus, presence of anti-Ro52 might be considered as a useful clinical diagnostic tool for the early detection of IIP in patients who pose a higher risk of developing SARD in the future.
IPAF criteria are used for the identi cation of a subset of IIP patients exhibiting autoimmune features but lacking a de nitive diagnosis of SARD [5] The ATS/ERS task force has suggested the need for further validation and revision of IPAF criteria [5]. Accordingly, there has been a proposal for the inclusion of several myositis-speci c antibodies (anti-NXP-2, anti-TIF1γ) in the IPAF criteria [26]. In contrast, anti-double stranded DNA, anti-Sm, anti-topoisomerase I (Scl-70), and anti-MDA5 are diseasespeci c diagnostic antibodies that have a proven link to the diagnosis of SLE [27], SSc [28], and clinically amyopathic DM (CADM) [29]. These disease-speci c marker antibodies are produced prior to the clinical manifestation of the associated CTD and the association of IPAF with these antibodies might be an indication of early stage CTD. The appropriateness of the inclusion of these antibodies in the IPAF criteria is controversial. Nevertheless, we propose the addition of anti-Ro52 to the IPAF criteria because of its proven association with SARD and wide co-existence with other autoantibodies including anti-ARS in IIP patients, even prior to being diagnosed for SARD.
The frequency of Raynaud's phenomenon was signi cantly higher in anti-Ro52-positive patients than in anti-Ro52-negative patients in our IIP cohort (P = 0.009) ( Table 3). Nearly half of the IPAF patients exhibit at least one clinical domain with Raynaud's phenomenon as the most common symptom [30,31] In this study, all three patients with anti-Ro52 who had Raynaud's phenomenon were classi ed as IPAF (Supplementary Table S1 and S2). Thus, testing for serum anti-Ro52 might be helpful in classifying IIP patients with Raynaud's phenomenon as those meeting the IPAF criteria. Raynaud's phenomenon is associated with underlying or future development of SARD [32] but is not considered a predictor for its prognosis or development in IPAF patients [30,31] probably due to the low prevalence and short follow-up periods. Thus, the clinical signi cance of Raynaud's phenomenon in IPAF patients remains controversial.
Patients with anti-Ro52 have a higher frequency of rapidly progressive ILD and a higher rate of mortality than those without anti-Ro52 in SARD [13,15,16,20]. Herein, presence of anti-Ro52 was not signi cantly associated with overall mortality, possibly due to the heterogeneity of IIPs and the limited number of patients.
Recent research has reported the heterogeneity related to the prognosis and response to treatment of IIP patients with anti-ARS, wherein, certain patients were refractory to treatment with poor prognosis, while others responded well [7,33]. Patients with PM/DM positive for both anti-Ro52 and anti-ARS had severe myositis and joint impairment with a higher prevalence of ILD [14,16]. In this study, among 18 anti-ARS-positive patients, signi cant differences were not seen in symptoms, characteristics (Supplementary Table S3 -5), and prognosis (Supplementary Figure S1) related to SARD, between anti-Ro52-positive and -negative patients. However, these ndings might considerably be affected by the small number of IIP patients with anti-ARS and further research is thus required.
Several limitations of this study are acknowledged. First, this study was a retrospective study with variable follow-up intervals and periods. Second, the sample size was relatively small and comprised only of Japanese individuals from two university hospitals. Third, although none of the patients with IIP developed any autoimmune diseases during the follow-up period, the observation period was short. It is possible that some might develop SARD in the future because ILD could precede the development of SARD in certain patients [25].

Conclusions
The ful llment of IPAF criteria and presence of Raynaud's phenomenon were more frequently seen in the presence than in the absence of anti-Ro52 in patients with IIP. We propose the addition of anti-Ro52 to the IPAF criteria because of its association with different types of SARD and wide co-existence with other autoantibodies in the serological domain. Further prospective studies on a large cohort are needed to elucidate the clinical signi cance of anti-Ro52 in patients with IIP. The datasets used for the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no con icts of interest (COI). Authors' contributions MT designed the study, had full access to all the data in this study, performed statistical analyses, and wrote the initial draft. MS, TH, and ST tested autoantibodies in the sera by immunoprecipitation and ELISA. NS, MS, HI, KYatera, and HM made substantial contribution to the conception and design of the study. MT, NS, HI, HY, and TK acquired the data. NS, MS, KYamasaki, KYatera, and MH participated in drafting and critically revising the article for important intellectual content. YF gave advice on the statistical analyses. KYatera was attributable for the nal responsibility for the decision to submit the article for publication. All authors have read and approved the nal manuscript.    dehydrogenase; CK: creatine kinase; KL-6: Krebs von den Lungen-6; SP-A: surfactant protein-A; SP-D: surfactant protein-D; VC: vital capacity; FEV 1 : forced expiratory volume in one second; DL CO : diffusing capacity of the lung for carbon monoxide; TCC: total cell counts; CD: cluster of differentiation P value: anti-Ro52 positive vs. anti-Ro52 negative Table 5. Results and patterns of patients with/without anti-Ro52/tripartite motif-containing 21 antibodies (anti-Ro52)