Dataset Information

Prediction of Hematopoietic Stem Cell Transplantation Related Mortality- Lessons Learned from the In-Silico Approach: A European Society for Blood and Marrow Transplantation Acute Leukemia Working Party Data Mining Study.

ABSTRACT: Models for prediction of allogeneic hematopoietic stem transplantation (HSCT) related mortality partially account for transplant risk. Improving predictive accuracy requires understating of prediction limiting factors, such as the statistical methodology used, number and quality of features collected, or simply the population size. Using an in-silico approach (i.e., iterative computerized simulations), based on machine learning (ML) algorithms, we set out to analyze these factors. A cohort of 25,923 adult acute leukemia patients from the European Society for Blood and Marrow Transplantation (EBMT) registry was analyzed. Predictive objective was non-relapse mortality (NRM) 100 days following HSCT. Thousands of prediction models were developed under varying conditions: increasing sample size, specific subpopulations and an increasing number of variables, which were selected and ranked by separate feature selection algorithms. Depending on the algorithm, predictive performance plateaued on a population size of 6,611-8,814 patients, reaching a maximal area under the receiver operator characteristic curve (AUC) of 0.67. AUCs' of models developed on specific subpopulation ranged from 0.59 to 0.67 for patients in second complete remission and receiving reduced intensity conditioning, respectively. Only 3-5 variables were necessary to achieve near maximal AUCs. The top 3 ranking variables, shared by all algorithms were disease stage, donor type, and conditioning regimen. Our findings empirically demonstrate that with regards to NRM prediction, few variables "carry the weight" and that traditional HSCT data has been "worn out". "Breaking through" the predictive boundaries will likely require additional types of inputs.

SUBMITTER: Shouval R

PROVIDER: S-EPMC4778768 | biostudies-literature | 2016

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Prediction of Hematopoietic Stem Cell Transplantation Related Mortality- Lessons Learned from the In-Silico Approach: A European Society for Blood and Marrow Transplantation Acute Leukemia Working Party Data Mining Study.

Shouval Roni R Labopin Myriam M Unger Ron R Giebel Sebastian S Ciceri Fabio F Schmid Christoph C Esteve Jordi J Baron Frederic F Gorin Norbert Claude NC Savani Bipin B Shimoni Avichai A Mohty Mohamad M Nagler Arnon A

PloS one 20160304 3

Models for prediction of allogeneic hematopoietic stem transplantation (HSCT) related mortality partially account for transplant risk. Improving predictive accuracy requires understating of prediction limiting factors, such as the statistical methodology used, number and quality of features collected, or simply the population size. Using an in-silico approach (i.e., iterative computerized simulations), based on machine learning (ML) algorithms, we set out to analyze these factors. A cohort of 25 ...[more]

PMID: 26942424

Similar Datasets

Project description:Three randomized controlled trials in early severe systemic sclerosis demonstrated that autologous hematopoietic stem cell transplantation was superior to standard cyclophosphamide therapy. This European Society for Blood and Marrow Transplantation multi-center prospective non-interventional study was designed to further decipher efficacy and safety of this procedure for severe systemic sclerosis patients in real-life practice and to search for prognostic factors. All consecutive adult systemic sclerosis patients undergoing a first autologous hematopoietic stem cell transplantation between December 2012 and February 2016 were prospectively included in the study. Primary endpoint was progression free survival. Secondary endpoints were overall survival, non-relapse mortality, response and incidence of progression. Eighty systemic sclerosis patients were included. Median follow-up duration was 24 (6-57) months after stem cell transplantation using cyclophosphamide plus antithymocyte globulins conditioning for all, with CD34+ selection in 35 patients. At 2 years, progression free survival was 81.8%, overall survival was 90%, response was 88.7% and incidence of progression was 11.9%. The 100 days non-relapse mortality was 6.25% (n=5) with four deaths from cardiac event, including three due to cyclophosphamide toxicity. Modified Rodnan skin score and forced vital capacity improved with time (p< 0.001). By multivariate analysis, baseline skin score >24 and older age at transplant were associated with lower progression free survival (Hazard ration 3.32) and 1.77, respectively). CD34+-selection was associated with better response (Hazard ration: 0.46). This study confirms the efficacy of autologous stem cell transplantation in real-life practice for severe systemic sclerosis using non myeloablative conditioning. Careful cardio-pulmonary assessment to identify organ involvement at patient referral, reduced cyclophosphamide doses and CD34+ selection may improve outcomes. The study was registered at ClinicalTrials.gov: NCT02516124.

Project description:In contrast to the upfront setting in which the role of high-dose therapy with autologous hematopoietic cell transplantation (HCT) as consolidation of a first remission in patients with multiple myeloma (MM) is well established, the role of high-dose therapy with autologous or allogeneic HCT has not been extensively studied in MM patients relapsing after primary therapy. The International Myeloma Working Group together with the Blood and Marrow Transplant Clinical Trials Network, the American Society of Blood and Marrow Transplantation, and the European Society of Blood and Marrow Transplantation convened a meeting of MM experts to: (1) summarize current knowledge regarding the role of autologous or allogeneic HCT in MM patients progressing after primary therapy, (2) propose guidelines for the use of salvage HCT in MM, (3) identify knowledge gaps, (4) propose a research agenda, and (5) develop a collaborative initiative to move the research agenda forward. After reviewing the available data, the expert committee came to the following consensus statement for salvage autologous HCT: (1) In transplantation-eligible patients relapsing after primary therapy that did NOT include an autologous HCT, high-dose therapy with HCT as part of salvage therapy should be considered standard; (2) High-dose therapy and autologous HCT should be considered appropriate therapy for any patients relapsing after primary therapy that includes an autologous HCT with initial remission duration of more than 18 months; (3) High-dose therapy and autologous HCT can be used as a bridging strategy to allogeneic HCT; (4) The role of postsalvage HCT maintenance needs to be explored in the context of well-designed prospective trials that should include new agents, such as monoclonal antibodies, immune-modulating agents, and oral proteasome inhibitors; (5) Autologous HCT consolidation should be explored as a strategy to develop novel conditioning regimens or post-HCT strategies in patients with short (less than 18 months remissions) after primary therapy; and (6) Prospective randomized trials need to be performed to define the role of salvage autologous HCT in patients with MM relapsing after primary therapy comparing it to "best non-HCT" therapy. The expert committee also underscored the importance of collecting enough hematopoietic stem cells to perform 2 transplantations early in the course of the disease. Regarding allogeneic HCT, the expert committee agreed on the following consensus statements: (1) Allogeneic HCT should be considered appropriate therapy for any eligible patient with early relapse (less than 24 months) after primary therapy that included an autologous HCT and/or high-risk features (ie, cytogenetics, extramedullary disease, plasma cell leukemia, or high lactate dehydrogenase); (2) Allogeneic HCT should be performed in the context of a clinical trial if possible; (3) The role of postallogeneic HCT maintenance therapy needs to be explored in the context of well-designed prospective trials; and (4) Prospective randomized trials need to be performed to define the role salvage allogeneic HCT in patients with MM relapsing after primary therapy.

Dataset Information

Prediction of Hematopoietic Stem Cell Transplantation Related Mortality- Lessons Learned from the In-Silico Approach: A European Society for Blood and Marrow Transplantation Acute Leukemia Working Party Data Mining Study.

Publications

Prediction of Hematopoietic Stem Cell Transplantation Related Mortality- Lessons Learned from the In-Silico Approach: A European Society for Blood and Marrow Transplantation Acute Leukemia Working Party Data Mining Study.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets