Improving Revised International Prognostic Scoring System Pre-Allogeneic Stem Cell Transplantation Does Not Translate Into Better Post-Transplantation Outcomes for Patients with Myelodysplastic Syndromes: A Single-Center Experience.

ABSTRACT

The natural history of patients with myelodysplastic syndromes (MDS) is variable. The Revised International Prognostic Score (IPSS-R) is commonly used in practice to predict outcomes in patients with MDS at both diagnosis and before hematopoietic stem cell transplantation (HSCT). However, the effect of change in the IPSS-R before allogeneic HSCT with chemotherapy or hypomethylating agents on post-transplantation outcomes is currently unknown. We assessed whether improvement in IPSS-R prognostic score pre-HSCT would result in improvement in clinical outcomes post-HSCT. Secondary goals included studying the effect of prognostic factors on post-transplantation survival. All patients with MDS who underwent allogeneic HSCT at the Leukemia/BMT Program of British Columbia between February 1997 and April 2013 were included. Pertinent information was reviewed from the program database. IPSS-R was calculated based on data from the time of MDS diagnosis and before HSCT. Outcomes of patients who had improved IPSS-R pre-HSCT were compared with those with stable or worse IPSS-R. Overall survival (OS) and event-free survival (EFS) were estimated using the Kaplan-Meier method, with P values determined using the log-rank test. Hazard ratios were calculated using multivariable Cox proportional hazards regression models to study the effects of the prognostic variables on OS and EFS. A total of 138 consecutive patients were included. IPSS-R improved in 62 of these patients (45%), worsened in 23 (17%), remained stable in 41 (30%), and was unknown in 12 (9%). OS was not statistically different across the improved, worsened, and stable groups (30% versus 22% versus 40%, respectively; P = .63). The cumulative incidences of relapse and nonrelapse mortality at 5 years were 28.4% (95% confidence interval [CI], 21.1 to 36.1) and 31.6% (95% CI, 23.8 to 39.7), respectively. The rate of relapse was 23% in patients with <5% blasts at the time of HSCT, 69% in those with 5% to 20% blasts, and 66% in those with >20% blasts (P = .0004). In the entire cohort OS was 34% and EFS was 33%. There was no significant difference in outcomes between patients who received myeloablative conditioning and those who received nonmyeloablative conditioning before HSCT (OS, 34% and 39%, respectively; P = .63 and EFS, 34% and 32%, respectively; P = .86). OS was not statistically different among patients with improved, worsened, or stable IPSS-R. On multivariate analysis, only 3 factors were associated with OS cytogenetic risk group at diagnosis, blast count at transplantation, and the presence or absence of chronic graft-versus-host disease. Improving IPSS-R before HSCT does not translate into better survival outcomes. Blast count pretransplantation was highly predictive of post-transplantation outcomes.