Effects of Autoimmune Disease on Non-High Risk MDS Outcomes: A Population Analysis

Introduction: Myelodysplastic syndromes (MDS) frequently occur in patients (pts) with autoimmune diseases (AD). Concomitant inflammatory or AD are found in 10-30% of MDS cases. Studies exploring the impact of AD on MDS outcomes show discordant results and small samples in these studies hinder definitive conclusions. We conducted a large epidemiological analysis aimed at defining the impact of concomitant AD on outcomes of non-high-risk MDS pts.

Methods: Pts ≥66 years with MDS were identified from 2007 - 2017 in the Surveillance, Epidemiology, and End Results (SEER) - Medicare database. Cases were excluded if they lacked continuous Medicare A and B, were enrolled in HMOs, had MDS diagnosed solely from death certificate or autopsy, or had MDS with excess blasts. Using claims from 12 months before MDS diagnosis, pre-existing AD was identified. MDS cohorts with and without concomitant AD were compared. Covariates included age, sex, race, and MDS risk, classified into low and intermediate using the International Classification of Diseases for Oncology (ICD-O-3) codes, rurality (using Rural-Urban Continuum Codes), and Charlson Comorbidity Index (CCI). Outcomes included transformation to acute myeloid leukemia (AML) during follow-up and overall survival, calculated from MDS diagnosis to date of death or end of follow-up. Logistic regression tested the association between AD and AML transformation. Cox proportional hazards regression estimated the association of AD with survival. Risk estimates are reported as odds ratio (OR) and hazard ratios (HR), with a corresponding 95% confidence interval (CI).

Results: A total of 13,344 pts diagnosed with MDS between 2007 - 2017 were analyzed, including 1,978 (14.8%) with baseline AD. Pts with concomitant AD were younger, proportion >80 years accounted for 60% vs. 65.2% in MDS cases without AD (p<0.001). They were also predominantly female (55.4% vs. 45.3%, p<0.001), non-Hispanic whites (NHW) (88.8% vs. 86.2%, p=0.045), and had a higher comorbidity burden, with CCI ≥2 accounting for 68.5% in the AD cohort compared to 51.2% (p<0.001). MDS risk distribution based on ICD-O-3 histology codes is similar between groups. In a multivariate logistic regression model adjusting for age, sex, race, MDS risk/histology, rurality, and CCI, baseline AD at MDS diagnosis was associated with 18% increased odds of AML transformation; OR = 1.18 (95% CI 1.01-1.39). Table 1 shows other covariates significantly associated with AML transformation. Increasing age, female sex, and higher CCI score were factors associated with decreased odds for AML transformation. As expected, MDS risk had the largest effect, the intermediate risk was associated with 2.3-fold increased odds of AML transformation (95% CI 1.8-2.8) compared to low risk. In a multivariate Cox-proportional hazard model adjusting for age, sex, race, MDS risk/histology, rurality, and CCI, baseline AD at MDS diagnosis was associated with a 6% decreased risk of overall death (HR 0.94, 95% CI 0.89 - 0.99). Table 2 shows other covariates significantly associated with overall survival. Non-Hispanic Black race and female sex were factors associated with decreased risk of death while rising age, rurality, and higher CCI were associated with increased hazard for death. MDS risk had the largest effect on overall survival, with the intermediate risk associated with a 52% increased risk of death (95% CI, 1.44-1.60) compared to low risk.

Conclusions: In a large population-based analysis, pts with AD and non-high risk MDS were predominantly female, and Non-Hispanic Whites had a higher comorbidity burden. Concomitant AD at MDS diagnosis was associated with increased risk of AML transformation but decreased risk of mortality. These seemingly contradicting results suggest that the pathophysiologic mechanisms linking systemic inflammation with MDS and their clinical consequences remain to be fully elucidated. Inflammatory bone marrow microenvironment in AD may promote clonal evolution while neoplastic clones driven by auto-inflammation could have a less aggressive phenotype. Future studies should assess molecular mutations and clonal dynamics involved in autoimmunity and MDS to build on these epidemiologic findings. Despite the limitations of a claim-based analysis, AD prevalence among MDS pts and factors associated with MDS outcomes in our cohort were consistent with previous reports.

Shastri:Janssen: Consultancy; Rigel Pharmaceutical: Membership on an entity's Board of Directors or advisory committees; NACE: Honoraria; Kymera Therapeutics: Membership on an entity's Board of Directors or advisory committees, Research Funding.