Lille Scoring System Rather Than DIPSS Is a Better Predictive of Overall Mortality After Allogeneic Hematopoietic Cell Transplantation (HCT) for Primary Myelofibrosis Using Reduced Intensity Conditioning: A Report From the Center for International Blood and Marrow Transplant Research (CIBMTR)

Abstract 432

The Dynamic International prognostic scoring system (DIPSS) is increasingly being used as a prognostic tool for determining the risk of mortality for primary myelofibrosis (PMF), and has largely replaced Lille scoring system. However, it is unclear whether this scale can predict mortality after reduced intensity conditioning (RIC) allogeneic HCT, a procedure that is increasingly being utilized, as demonstrated by data from the CIBMTR.

Using the CIBMTR database, the impact of patient, disease and transplant related factors on outcomes of 222 patients, who underwent HCT for PMF using RIC was analyzed. Median follow-up of survivors was 50 months (range, 3–165). Median age at HCT was 55 years, and 56 (25%) were >60 years. Donors were matched related donor (MRD), well-matched unrelated (MUD), and partially/mismatched unrelated (MMUD) in 85 (38%), 94 (42%), and 43 (19%), respectively. Conditioning regimens were: Fludarabine (Flu) and Melphalan (Mel), 62 (28%); Flu and Busulphan (Bu), 81 (36%), Flu and total body irradiation (TBI), 49 (22%); and others 30 (14%). Disease-risk status at HCT according to Lille scoring system was: low 48 (22%), intermediate (Int) 105 (47%), and high 69 (31%); and according to DIPSS was: low 25 (11%), int-1 110 (50%), int-2 81 (36%), and high, 4 (2%).

The cumulative incidences of acute graft versus host disease (GvHD) at 100 days and chronic GvHD at 5-years were 48% (95% confidence intervals [CI] 41–54) and 50% (95% CI 43–57), respectively. The cumulative incidence of relapse/progression and non-relapse mortality (NRM) at 5-years was 28% (95% CI 22–34) and 38% (95% CI 31–44), respectively. The corresponding disease-free and overall survival was 34% (95% CI 28–40), and 37% (95% CI 31–44), respectively. In multivariate analysis, high-risk disease defined by Lille scoring system was associated with two-fold higher mortality compared to low-risk disease (Table). Higher risk disease status as defined by DIPSS was not associated with a significant increase in mortality when compared to lower-risk disease (Table). MUD and MMUD use were associated with higher mortality risk compared to MRD with relative risk (RR) of 1.59 (95% CI 1.00–2.52) and 2.6 (95% CI 1.56–4.35), respectively. A comparison of conditioning regimens demonstrated a trend towards reduced mortality with FluMel when compared to FluBu (RR 0.56, 95% CI 0.33–0.92; overall p=0.11), or other regimens (RR 0.51, 95% CI 0.26–0.99; overall p=0.11).

In conclusion, the current study highlights that the DIPSS was limited in predicting the mortality after RIC transplantation for PMF, while the Lille scoring system remained predictive of mortality in high risk patients. These findings underscore the need for transplant-specific scoring system. Compared to other conditioning regimens FluMel appears to be associated with a trend towards better survival, which needs to be confirmed in prospective randomized trials.