AB0619 PROGNOSTIC FACTORS OF PATIENTS WITH ANTI-MDA5 ANTIBODY-POSITIVE DERMATOMYOSITIS COMPLICATED WITH INTERSTITIAL PNEUMONIA -A JAPANESE SINGLE CENTER STUDY-

Abstract

Background:Anti-melanoma differentiation-associated gene 5 antibody (anti-MDA5 Ab)-positive dermatomyositis (DM) is frequently associated with rapidly progressive interstitial pneumonia (RPIP), whose prognosis is assumed to be poor[1]. Although outcome of DM-RPIP has been reported to be improved by early immunosuppressive therapy, we still experience the cases with severe outcome. Only several reports mentioned the prognostic factors and they have not been fully elucidated.Objectives:To identify the predictors of prognosis in patients with anti-MDA5 Ab-positive DM associated with interstitial pneumonia (DM-IP).Methods:Anti-MDA5 Ab-positive DM-IP patients admitted to Fujita Health University Hospital between January 2010 and October 2019 were consecutively included and stratified into 2 groups, the survived and the deceased groups. DM was diagnosed according to the criteria proposed by Bohan and Peter[2]. Clinically amyopathic DM was diagnosed according to the criteria proposed by Sontheimer [3]. Diagnosis of IP was based on findings of high resolution CT scan (HRCT). The definition of RPIP was rapid exacerbation of hypoxemia or HRCT findings in a period of days to one month after the onset. Clinical features and prognosis of the patients were collected retrospectively and compared between groups. Candidates of predictors are extracted by the univariable analysis using Fisher’s exact test for dichotic parameters and Wilcoxon signed-rank test for continuous parameters and multivariable analysis using logistic regression analysis. Receiver operating characteristic (ROC) curve analysis was examined to obtain the cut-off level. Survival was examined using Kaplan-Meier method and Log-rank test.Results:Twenty-one patients were involved. Eight were deceased and 13 were survived. The deceased group had a higher ratio of male (75% versus 25%, p= 0.018). All deceased cases were with RPIP and 67 % in the survived cases. Levels of serum ferritin (4490 versus 646 ng/mL, p = 0.0026), CRP (2.1 versus 0.9 mg/dL, p = 0.0490), CK (1150 versus 290 U/L, p = 0.017), AST (194 versus 108 U/L, p = 0.025) and LDH (674 versus 368 U/L, p = 0.011) were higher in the deceased group. Interestingly, skin ulcers were tended to be more frequent (12.5% versus 87.5%, p= 0.0587), and anti-SS-A antibody was also more frequently detected (14.3% versus 85.7%, p=0.0072) in the survived group. Using ROC analysis cut-off values were 963 ng/mL for serum ferritin level (sensitivity 100%, specificity 83%), 0.7 mg/dL for CRP (sensitivity 75%, specificity 69%), 308 U/L for CK (sensitivity 88%, specificity 77%), 62 U/L for ALT (sensitivity 100%, specificity 62%), and 454 U/L for LDH (sensitivity 88%, specificity 77%). Patients were divided into two groups based on these cut-offs or based on dichotic parameters and survival was examined between 2 groups. Except CRP and anti-SS-A antibody, survival was significantly worse in parameter-positive or higher groups. Interestingly, anti-SS-A antibody-positive group had better outcome compared with those without.Conclusion:In our analysis, novel candidates such as serum CK, AST, and LDH levels were newly extracted and parameters previously reported was also included and those were also associated with the clinical outcome. In addition, anti-SS-A antibody was identified as a novel protective factor associated with a good outcome.