Comparison of dipss and mysec-pm for prediction of outcome in post-pv and et myelofibrosis after allogeneic stem-cell transplantation
Résumé
We aimed to validate the MYelofibrosis SECondary to PV and ET prognostic model (MYSEC-PM) in 159 patients with myelofibrosis secondary to polycythemia vera (PV) and essential thrombocythemia (ET) from the European Society for Blood and Marrow Transplantation registry undergoing transplantation from matched siblings or unrelated donors. Furthermore, we aimed to test its prognostic performance in comparison with the Dynamic International Prognostic Scoring System (DIPSS). Score performance was analyzed using the concordance index (C): the probability that a patient who experienced an event had a higher risk score than a patient who did not (C > .5 suggesting predictive ability). Median follow-up of the total cohort was 41 months (range, 34 to 54), 45 months in post-PV and 38 months in post-ET myelofibrosis. Survival at 1, 2, and 4 years was 70% (95% CI, 63% to 77%), 61% (95% CI, 53% to 69%), and 52% (95% CI, 43% to 61%) for the total cohort; 70% (95% CI, 59% to 80%), 61% (95% CI, 49% to 73%), and 51% (95% CI, 38% to 64%) for post-PV; and 71% (95% CI, 61% to 81%), 61% (95% CI, 50% to 72%), and 54% (95% CI, 42% to 66%) for post-ET myelofibrosis (P = .78). Overall, the DIPSS was not significantly predictive of outcome (P = .28). With respect to the MYSEC-PM, overall survival at 4 years was 69% for the low-risk, 55% for the intermediate 1-risk, 47% for the intermediate 2-risk, and 22% (0% to 45%) for the high-risk groups. The prognostic model was predictive of survival overall (P = .05), whereas groups with intermediate 2 and high risk showed no significant difference (P = .44). Assessment of prognostic utility yielded a C-index of .575 (95% CI, .502 to .648) for the DIPSS, whereas assessment of the MYSEC-PM resulted in a C-statistics of .636 (95% CI, .563 to .708), indicating improvement in prediction of post-transplant survival using the new MYSEC-PM. In addition, transplantations from an unrelated donor in comparison with an HLA-identical sibling showed worse outcome (P = .04), and transplant recipients seropositive for cytomegalovirus in comparison with seronegative recipients (P = .01) showed worse survival. In conclusion, incorporating transplant-specific and clinical and mutational information together with the MYSEC-PM may enhance risk stratification.