Are IPSS-R and IPSS Cytogenetic Risk Stratification Informative At the Time of Allogeneic Hematopoietic Cell Transplantation?

Abstract 1400

Myelodysplastic syndromes (MDS) are a heterogeneous group of hematopoietic stem cell disorders. Cytogenetic aberrations are one of the main elements of risk stratification among published scoring systems including the International Prognostic Scoring System (IPSS) or its recent revision (R-IPSS). R-IPSS cytogenetic risk stratification has 5 cytogenetic subgroups rather than 3 in the IPSS. We hypothesized that R-IPSS and/or IPSS cytogenetic risk stratification at the time of allogeneic hematopoietic cell transplantation (alloHCT) can better predict outcome afterwards. In patients who had cytogenetic results at both diagnosis and alloHCT, we evaluated the cytogenetic changes over time and compared how well the two cytogenetic risk stratification systems predicted outcome.

One hundred consecutive patients with MDS undergoing an alloHCT (52% related donor, 13% unrelated donor, and 35% umbilical cord blood) at the University of Minnesota between 1995 and 2011 were evaluated. Their median age was 52 years (range, 18–69 years). Seventy-eight percent of the patients had ≥90% Karnofsky performance score, and 60% had INT-2 or a high IPSS score. Half received reduced-intensity conditioning (RIC).

Patients with IPSS good-risk cytogenetics at diagnosis had a significantly better OS (62% at 5-year) compared to patients with intermediate or poor-risk cytogenetics. However, IPSS cytogenetic stratification at the time of transplantation predicted neither overall survival (OS) nor relapse (Table). Patients with IPSS-R very poor risk cytogenetics at either diagnosis or alloHCT had a dismal outcome; no patients in this category at alloHCT survived 5 years, and their relapse rate was also significantly higher (41%) compared to other cytogenetic risk subgroups (Table). In multivariate analysis, the only factor predicting relapse was the intensity of the conditioning regimen: those patients receiving RIC had a higher risk of relapse (HR, 2.5, 95% CI 1.1–5.4, p=0.03) than those who received myeloablative conditioning regimens.

This data indicate that IPSS (at diagnosis) and IPSS-R (at both diagnosis and alloHCT) cytogenetic classifications provide important prognostic information in the best and worst subgroups, respectively, but are not as helpful for patients in the intermediate subgroups.