Poster Session I: Myelodysplastic syndromes - Clinical
Myelodysplastic syndromes (MDS) are clonal hemopathies with a relatively heterogeneous spectrum of presentation and a variable risk of transformation into acute myeloid leukemia (AML). Autoimmune diseases (AID) are present in 10 to 30% of MDS, with uncertain prognostic significance in this context. To date, no MDS-related factor has been clearly identified as a predictor for the occurrence of AIDs.
To study the association between MDS characteristics and the occurrence of an AID and its impact on overall survival and leukemia-free survival.
We retrospectively collected data from primary MDS adult patients followed-up at the University Hospital of Poitiers between January 2010 and December 2017. Immunosuppressive therapy secondary-MDS were excluded. The date of onset and type of AID as established by the standard criteria were collected. Steroid dependence was defined as a prednisone equivalent amount > 10 mg/day during at least 3 months. Simultaneous diagnosis of MDS and AID were considered if established in a 3-months interval from the MDS one. Baseline characteristics at diagnosis of the MDS were registered and compared between those with and without AIDs.
AIDs were identified in 44 of 179 MDS patients (24.6%). MDS-AID patients were more frequently male (29/44, p < 0.05). AIDs occurred before MDS in 63.6% of cases. Associated AIDs were: systemic vasculitis (23%), inflammatory rheumatism (18%), neutrophilic dermatosis (12.5%), connective tissue diseases (12.5%) and autoimmune cytopenia (11%). At the time of AID diagnosis, 48% of patients had general signs, 37.5% had cutaneous and/or joint signs. AID was steroid-dependent in 39% of cases (9/23). No association between AID and MDS subtypes was observed. A significant linear tendency was found between the R-IPSS and the presence of AID (p = 0.02): that is, the lower the IPSS-R score, the higher the risk of association with AID. The presence of AID did not affect survival or progression to acute leukemia.
The association of MDS to AIDs is frequent and of clinical heterogeneous expression. It occurs mainly in low-risk MDS and does not seem to impact overall survival, that seems logical. The pathogenic mechanisms are complex and remain still completely unknown, but the hypothesis of a common pathogenic pathway seems strongly supported. The association MDS-AID does not seem to impact overall survival in our cohort and occurs mainly in low-risk MDS with a significant linear augmentation of the risk by the creasing categories of R-IPSS. That raises the question of the role of different physio-pathological processes underlying low-risk and high-risk MDS in the predisposition of AIDs. Larger prospective ongoing studies will better clarify our results.