Temporal trends in overall survival among patients with high-risk MDS: Insights from the spanish MDS registry (RESMD)

Abstract

Abstract Background: High-risk myelodysplastic syndromes (HR-MDS) are associated with poor survival outcomes. Although the introduction of hypomethylating agents (HMAs) and increased access to allogeneic stem cell transplantation (allo-SCT) represented therapeutic advances, no new drugs have been approved for HR-MDS since HMAs nearly two decades ago. Supportive care has improved, with broader use of antibiotics, antifungals, and transfusion support, but whether these changes have translated into survival gains remains uncertain. Aims To assess temporal trends in overall survival (OS) in patients with HR-MDS in real-world clinical practice using data from the Spanish MDS Registry (RESMD), and to evaluate whether prognostic shifts or demographic changes account for observed patterns. Methods We included all patients with high-risk MDS (HR-MDS) recorded in the Spanish Registry of Myelodysplastic Syndromes (RESMD) between 2000 and 2024. HR-MDS was defined as IPSS ≥1.5, IPSS-R >3, or IPSS-M >0. The RESMD is a nationwide, multicenter registry that collects prospective, real-world clinical data on patients with MDS across more than 30 hospitals in Spain. Overall survival (OS) was calculated from diagnosis to death or last follow-up. Patients were stratified by diagnosis period into quinquennia: 2000–2004, 2005–2009, 2010–2014, 2015–2019, and 2020–present. Kaplan–Meier curves and log-rank tests were used for univariate comparisons. Multivariable Cox regression models were constructed to evaluate temporal trends in OS, adjusting for age at diagnosis and IPSS-R risk group. Results A total of 4002 patients with high-risk MDS were included in the analysis. Median age at diagnosis increased from 66.6 years in 2000–2004 to 72.3 years in 2020–present. The proportion of patients receiving HMAs rose markedly from 1.3% to 35.6%, while allo-SCT rates declined from 14.1% to 10.4%. Risk profiles by IPSS-R remained stable across quinquennia. The proportion of therapy-related MDS significantly increased across quinquennia, rising from 6.6% in 2000–2004 to 16.5% in 2020–present (p < 0.001), consistent with growing numbers of patients surviving prior malignancies and receiving cytotoxic therapies. Median OS varied slightly across quinquennia: 18.7 months (2000–2004), 17.7 (2005–2009), 18.6 (2010–2014), 20.0 (2015–2019), and 22.9 months (2020–present), though differences did not reach statistical significance (log-rank p = 0.15). Similar non-significant trends were observed within IPSS-R subgroups (intermediate, high and very high risk). In Cox regression adjusting for age, only patients diagnosed in 2020–present showed significantly better survival compared to 2000–2004 (HR 0.79, 95% CI 0.64–0.97, p = 0.025), while other time periods did not differ significantly. After further adjustment for IPSS-R, survival improved progressively across quinquennia: patients diagnosed in 2010–2014 (HR 0.84, 95% CI 0.74–0.96, p = 0.008), 2015–2019 (HR 0.77, 95% CI 0.67–0.89, p < 0.001), and 2020–present (HR 0.69, 95% CI 0.56–0.85, p < 0.001) all had significantly lower mortality compared to the reference period. Age and IPSS-R remained strong independent predictors of survival in the multivariate analysis. The discrepancy between unadjusted and adjusted analyses may reflect a shift in patient characteristics over time. The increase in median age at diagnosis suggests that older adults are more frequently being diagnosed, possibly due to improved survival from other malignancies and broader access to diagnostic evaluations in elderly patients. Conclusions In this nationwide registry analysis of over 4000 patients with HR-MDS, unadjusted analyses showed no significant improvement in overall survival over two decades. However, in multivariable Cox regression adjusting for age and IPSS-R risk, more recent diagnosis was independently associated with better survival. These findings suggest that therapeutic and supportive care advances may have had a delayed but measurable impact on outcomes in high-risk patients. These results highlight the importance of continued efforts to translate therapeutic advances into meaningful clinical benefit, especially in an aging population with limited treatment options.