Pb2454: development of a prognostic model for survival of mixed phenotype acute leukaemia patients treated with allogeneic haematopoietic stem cell transplantation

Topic: 22. Stem cell transplantation - Clinical Background: MPAL is a rare and dangerous type of acute leukaemia. Past studies on MPAL therapy were primarily focused on chemotherapy but not transplantation. With the development of allo-HSCT, it has become possible for almost any patient to have a donor and undergo transplantation. However, there is no exact indication of who can be treated with allo-HSCT, and different hospitals may make different decisions. To select MPAL patients who might benefit the most from allo-HSCT, the medical practitioner needs an informative prognostic tool to weigh the risks and benefits. To our knowledge, this is the first study to focus on the prognostic model of MPAL patients with allo-HSCT. Our study helps to inform MPAL patients undergoing allo-HSCT about their risk for having a poor outcome, thus assisting in decisions on therapeutic strategies based on the 3-year survival of MPAL patients. Aims: The aim of this study was to develop a new prognostic model for predicting 3-year overall survival among MPAL patients treated with allo-HSCT so that a more evidence-based decision on allo-HSCT could be made as part of clinical practice guidance. Methods: We retrospectively evaluated 91 MPAL patients treated with their first allo-HSCT as consolidation therapy from Peking University People’s Hospital between 1995 and 2022 for prognostic model development. The primary outcome was 3-year overall survival. We included clinical characteristics, laboratory results, induction therapy and transplantation results as potential risk predictors. We used the multivariable logistic regression model to estimate the coefficients associated with each potential predictor and to produce the model. We assessed the prognostic model performance by evaluating the discrimination [area under the curve (AUC)], calibration (calibration plot), and net benefit [decision curve analysis (DCA)]. Results: In total, 91 MPAL patients were treated using transplantation. Among them, 44 (48.4%) had the B-lymphoid/myeloid phenotype (B + My), 41 (45.1%) had the T-lymphoid/myeloid phenotype (T + My), 1 (1.1%) had the B-lymphoid/T-lymphoid phenotype (B + T), 1 (1.1%) had the B-lymphoid/T-lymphoid/myeloid phenotype (B + T+ My), and 4 (4.4%) were unclear. The three-year overall survival (OS) rate was 83.5%. A prognostic model was established to predict the 3-year outcome. Stem cell engraftment and relapse were the strongest predictors of mortality. Other predictors included in the final model were the platelet count and glucose before transplantation. The discrimination and calibration were satisfactory, with C statistics 0.931 in the development dataset. These indices showed good discrimination in the MPAL patients treated with allo-HSCT. Summary/Conclusion: This prognostic model can be used to obtain predictions of survival in MPAL patients treated with allo-HSCT, assisting in pre-transplantation risk assessment and stratification and the decision of whom, how, and when to perform transplantation of great importance. Stem cell engraftment and relapse are important prognostic factors. Keywords: Acute leukemia, Prognostic factor, Allogeneic hematopoietic stem cell transplant