SARS-CoV2 serology screening in spondyloarthritis patients in north-Eastern Italy: A pilot study

ABSTRACT

Background: Serology could help defining the real extent of Severe Acute Respiratory Syndrome CoronaVirus 2 (SARS-CoV2) diffusion in the population, especially in individuals considered at higher risk of SARS-CoV2 infection (COVID-19), such as Spondiloarthritis (SpA) patients undergoing immunosuppressive therapy or health care workers (HCW). In fact, COVID-19 detection is complicated by the fact that many patients can be asymptomatic. In these cases, it has also been suggested that a weaker immune response might be elicited. In this context, the role of anti-cytokine targeted therapy -commonly used as treatment in SpA-is uncertain, as it is not clear whether it is detrimental or protective towards severe disease forms. Objectives: The aim of the study was to explore the potential role of serology in detecting previous contact with SARS-CoV2 in SpA patients and HCW, and compare the frequency of positive findings with a control population. Methods: Consecutive patients affected by axial or peripheral SpA, classified according to Assessment of SpondyloArthritis international Society (ASAS) criteria and undergoing cytokine-targeted therapy, as well as HCW and controls from the pre-COVID-19 era (control group, 2015) were recruited. In SpA patients, disease activity was assessed by Ankylosing Spondylitis Disease Activity Score (ASDAS) and Disease Activity Score on 28-joint-count (DAS28). Sera from all patients were analysed through chemiluminescent analytical system (CLIA) for the presence of IgG and IgM anti-SARS-CoV2. Patients with a positive serological test (either IgM or IgG) additionally underwent real time Polymerase Chain Reaction (RT-PCR) in nasopharyngeal swabs in order to test for active infection. In SpA patients, serology was repeated after 3 months. Data across the 3 groups were compared by ANOVA or Chi-square, while comparison between 2 groups were conducted by Wilcoxon signed rank test or Chi-Square, for continuous and categorical data respectively. P ≤ 0.05 were considered as significant. Results: A total of 396 patients were recruited 200 SpA, 95 HCW and 101 healthy controls. SpA patients were mostly (54%) males, with mean age 49.6 ±14.7 years, and all were treated with anti-TNFα (78%), anti-IL-17 (9%) and anti-IL-23 drugs (7%), or small molecules (6%). Their disease activity level was moderate-low as assessed by ASDAS (1.95 ±0.98) and DAS28 (2.33 ±2.02). Among HCW and controls, 35% and 62% were male, with mean age 46.7 ±12.9 and 50.6±10.6 respectively. Positive serology (IgM or IgG, or both) was found in 12.5% SpA patients, 8.4% HCW, 0% controls (p=0.001). Among these, IgM titres were higher in the SpA group than in HCW (2.76±2.94 versus 0.80±0.67 KU/L, p= 0.016), while IgG mean titres were lower in the SpA group than in HCW (0.88±3.18 KU/L versus 1.05±0.88, p= 0.035). SpA patients with positive serology more frequently reported COVID-19 like symptoms than those with negative serology (20% vs 4%, p=0.009) and 2 had COVID-19 as confirmed by RT-PCR, none with a severe disease course. None of the HCW reported symptoms or tested positive by RT-PCR. In the SpA patients, at 3 months, the mean IgM titre decreased from 2.76±2.93 to 2.38±2.95 (p=0.001), while the IgG titres decreased from 0.89±3.25 to 0.31±0.87 (p=ns). Interestingly, the IgM or IgG titer at a single-patient level did not seem to change much in terms of absolute value (Figure 1), except in one patient, with documented COVID-19 (positive RT-PCR), in whom IgG level even decreased at 3 months. Conclusion: Serology revealed that exposure to COVID-19 in SpA patients, as well as HCW, was higher than expected based on reported symptoms. Targeted anti-cytokine therapy could act as a protective factor for a severe disease course in SpA patients. However, in this population, IgG and IgM titres did not change in a clinically significant manner at 3 months, and patient did not seem to develop an immune profile consistent with durable response. This result could be due to a weaker immune response in mild infections, but further studies are w rranted to clarify the pathophysiology beyond these observations.