In a previous study, we identified somatic mutations of SF3B1, a gene encoding a core component of RNA splicing machinery, in patients with myelodysplastic syndrome (MDS). Here, we define the clinical significance of these mutations in MDS and myelodysplastic/myeloproliferative neoplasms (MDS/MPN). The coding exons of SF3B1 were screened using massively parallel pyrosequencing in patients with MDS, MDS/MPN, or acute myeloid leukemia (AML) evolving from MDS. Somatic mutations of SF3B1 were found in 150 of 533 (28.1%) patients with MDS, 16 of 83 (19.3%) with MDS/MPN, and 2 of 38 (5.3%) with AML. There was a significant association of SF3B1 mutations with the presence of ring sideroblasts (P < .001) and of mutant allele burden with their proportion (P = .002). The mutant gene had a positive predictive value for ring sideroblasts of 97.7% (95% confidence interval, 93.5%-99.5%). In multivariate analysis including established risk factors, SF3B1 mutations were found to be independently associated with better overall survival (hazard ratio = 0.15, P = .025) and lower risk of evolution into AML (hazard ratio = 0.33, P = .049). The close association between SF3B1 mutations and disease phenotype with ring sideroblasts across MDS and MDS/MPN is consistent with a causal relationship. Furthermore, SF3B1 mutations are independent predictors of favorable clinical outcome, and their incorporation into stratification systems might improve risk assessment in MDS.

Consideration of iron-chelation (IC) in transfusion-dependent patients is recommended in most clinicalpractice guidelines on myelodysplastic syndromes (MDS). The financial impact of IC on health-care systems is predicted through economic modeling, but an analysis based on actual prevalence is lacking. Here, we have investigated the potential drug-costs and need for IC in a cohort of 189 United Kingdom-based MDS patients diagnosed from 2000 to 2010. Patients with low or intermediate-1 IPSS scores were identified as eligible for IC if 24 red cell units (RCU) had been transfused over 12 consecutive months or the transfusion-intensity averaged 2 RCU per month. Drug-costs were calculated from the time patients qualified for IC until death or last follow-up. In 159 patients with low/intermediate-1 MDS, survival was superior with a low IPSS score (P 5 0.014), age <70 years (P 5 0.043), transfusion-independence at diagnosis (P 5 0.0056) and transfusion-intensity of <2 RCU per month (P 5 0.009). Reflecting the time elapsed since diagnosis, longer survival was observed with a cumulative red cell load of 75 U (P 5 0.046). By logistic-regression analysis, transfusion-intensity independently predicted survival (P 5 0.0035) in low and intermediate-1 risk MDS patients. Forty-one patients fulfilled criteria for consideration of IC. Of these, 6 patients died within 1 month; 35 patients survived for a median of 16 months (range 1-61). Had patients commenced IC, the anticipated drug-costs alone would have been $526,880-$2,064,800 over 10 years. The lack of association between cumulative transfusion-load and survival calls for a prospective evaluation of the cost-utility of IC in patients surviving long-term, to enable evidencebased recommendations in MDS management. Am. J. Hematol. 86:406-410, 2011. V

SummaryGenetic and epigenetic alterations contribute to the biological and clinical characteristics of myelodysplastic syndromes (MDS), but a role for socioeconomic environment remains unclear. Here, socioeconomic status (SES) for 283 MDS patients was estimated using the Scottish Index of Multiple Deprivation tool. Indices were assigned to quintile categorical indicators ranked from SES1 (lowest) to SES5 (highest). Clinicopathological features and outcomes between SES quintiles containing 15%, 20%, 19%, 30% and 16% of patients were compared. Prognostic scores identified lower-risk MDS in 82% of patients, with higher-risk disease in 18%.SES quintiles did not associate with age, gender, cytogenetics, IPSS, or in subanalysis (n=95), driver mutations. The odds ratio of a diagnosis of RA was greater than other MDS sub-types in SES5 (OR 1.9, p=0.024). Most patients (91%) exclusively received supportive care. SES did not associate with leukaemictransformation or the cause of death. Cox regression models confirmed male gender (p<0.05), disease-risk (p<0.0001) and age (p<0.01) as independent predictors of leukaemia-free survival, with leukaemic-transformation an additional determinant of overall survival (p=0.07). Thus, if access to healthcare is equitable, SES does not determine disease biology or survival in MDS patients receiving supportive treatment; whether outcomes following disease-modifying therapies are influenced by SES requires additional studies.

No abstract