Search

Plain language summaryThere are various ways in which clinicians can predict the risk of disease progression in patients with leukemia, helping them to treat the patients accordingly. However, these approaches are usually designed by human experts and might not fully capture the complexity of a patient's disease. Here, with a large cohort of patients with acute myeloid leukemia, we design an unsupervised machine learning model - a type of computer model that learns from patterns in data without human input-to separate these patients into subgroups according to risk. We identify four distinct groups which differ with regards to patient genetics, laboratory values, and clinical characteristics. These groups have differences in response to treatment and patient survival, and we validate our findings in another dataset. Our approach might help clinicians to better predict outcomes in patients with leukemia and make decisions on treatment. Eckardt et al. identify four distinct risk clusters of patients with acute myeloid leukemia using unsupervised metaclustering, which is agnostic to current risk stratification models, in a large multicenter cohort. Clusters differ in terms of response to induction therapy and long-term survival, and are externally validated. BackgroundIncreasingly large and complex biomedical data sets challenge conventional hypothesis-driven analytical approaches, however, data-driven unsupervised learning can detect inherent patterns in such data sets.MethodsWhile unsupervised analysis in the medical literature commonly only utilizes a single clustering algorithm for a given data set, we developed a large-scale model with 605 different combinations of target dimensionalities as well as transformation and clustering algorithms and subsequent meta-clustering of individual results. With this model, we investigated a large cohort of 1383 patients from 59 centers in Germany with newly diagnosed acute myeloid leukemia for whom 212 clinical, laboratory, cytogenetic and molecular genetic parameters were available.ResultsUnsupervised learning identifies four distinct patient clusters, and statistical analysis shows significant differences in rate of complete remissions, event-free, relapse-free and overall survival between the four clusters. In comparison to the standard-of-care hypothesis-driven European Leukemia Net (ELN2017) risk stratification model, we find all three ELN2017 risk categories being represented in all four clusters in varying proportions indicating unappreciated complexity of AML biology in current established risk stratification models. Further, by using assigned clusters as labels we subsequently train a supervised model to validate cluster assignments on a large external multicenter cohort of 664 intensively treated AML patients.ConclusionsDynamic data-driven models are likely more suitable for risk stratification in the context of increasingly complex medical data than rigid hypothesis-driven models to allow for a more personalized treatment allocation and gain novel insights into disease biology. CA Eckardt

Immune thrombocytopenia (ITP) is the consequence of a complex, still not completely understood immunological dysregulation. Proposed mechanisms include autoantibody-induced platelet destruction, impaired platelet production as well as abnormalities in T-cell immunity, such as Th1 polarization, a high proportion of Th17 cells and a reduced number of regulatory T cells. Although the etiology of ITP is incompletely understood and considered multifactorial in most cases, genetic variants are thought to play a key role in susceptibility to ITP, especially in persistent or chronic ITP. Efforts are currently underway to uncover possible predisposing genetic factors for the development of ITP. Single nucleotide polymorphisms and copy number variations have been identified in several immune-related genes, such as cytokine genes, Fc gamma receptor genes or T-cell co-stimulation genes, and have been associated with patients' susceptibility to ITP. However, due to the clinical heterogeneity and low incidence of ITP it remains challenging to perform genetic analyses with sufficiently large sample size within informative patient populations, highlighting the need for collection of well annotated biomaterials in clinical trials or registry projects. Another significant challenge is to go beyond performing association studies alone and to establish genotype-phenotype associations, thus proving causality between a genetic alteration and ITP pathogenesis. This review summarizes our current knowledge on genetic alterations identified as potential predisposing factors for the development of ITP in adults, thereby addressing signaling pathways considered critical for ITP pathogenesis.

Biallelic mutations of the CEBPA gene (CEBPAbi) define a distinct entity associated with favorable prognosis; however, the role of monoallelic mutations (CEBPAsm) is poorly understood. We retrospectively analyzed 4708 adults with acute myeloid leukemia (AML) who had been recruited into the Study Alliance Leukemia trials, to investigate the prognostic impact of CEBPAsm. CEBPA mutations were identified in 240 patients (5.1%): 131 CEBPAbi and 109 CEBPAsm (60 affecting the N-terminal transactivation domains [CEBPAsmTAD] and 49 the C-terminal DNA-binding or basic leucine zipper region [CEBPAsmbZIP]). Interestingly, patients carrying CEBPAbi or CEBPAsmbZIP shared several clinical factors: they were significantly younger (median, 46 and 50 years, respectively) and had higher white blood cell (WBC) counts at diagnosis (median, 23.7 × 109/L and 35.7 × 109/L) than patients with CEBPAsmTAD (median age, 63 years, median WBC 13.1 × 109/L; P < .001). Co-mutations were similar in both groups: GATA2 mutations (35.1% CEBPAbi; 36.7% CEBPAsmbZIP vs 6.7% CEBPAsmTAD; P < .001) or NPM1 mutations (3.1% CEBPAbi; 8.2% CEBPAsmbZIP vs 38.3% CEBPAsmTAD; P < .001). CEBPAbi and CEBPAsmbZIP, but not CEBPAsmTAD were associated with significantly improved overall (OS; median 103 and 63 vs 13 months) and event-free survival (EFS; median, 20.7 and 17.1 months vs 5.7 months), in univariate and multivariable analyses. Additional analyses revealed that the clinical and molecular features as well as the favorable survival were confined to patients with in-frame mutations in bZIP (CEBPAbZIP-inf). When patients were classified according to CEBPAbZIP-inf and CEBPAother (including CEBPAsmTAD and non-CEBPAbZIP-inf), only patients bearing CEBPAbZIP-inf showed superior complete remission rates and the longest median OS and EFS, arguing for a previously undefined prognostic role of this type of mutation.

Allogeneic hematopoietic stem cell transplantation (alloHCT) represents the only potentially curative treatment in high-risk AML patients, but up to 40% of patients suffer from relapse after alloHCT. Treatment of overt relapse poses a major therapeutic challenge and long-term disease control is achieved only in a minority of patients. In order to avoid post-allograft relapse, maintenance as well as pre-emptive therapy strategies based on MRD-detection have been used. A prerequisite for the implementation of pre-emptive therapy is the accurate identification of patients at risk for imminent relapse. Detection of measurable residual disease (MRD) represents an effective tool for early relapse prediction in the post-transplant setting. However, using established MRD methods such as multicolor flow cytometry or quantitative PCR, sensitive MRD monitoring is only applicable in about half of the patients with AML and advanced MDS undergoing alloHCT. Donor chimerism analysis, in particular when performed on enriched leukemic stem and progenitor cells, e.g. CD34+ cells, is a sensitive method and has emerged as an alternative option in the post alloHCT setting. In this review, we will focus on the current strategies for lineage specific chimerism analysis, results of pre-emptive treatment using this technology as well as future developments in this field.

Achievement of complete remission (CR) signifies a crucial milestone in the therapy of acute myeloid leukemia (AML) while refractory disease is associated with dismal outcomes. Hence, accurately identifying patients at risk is essential to tailor treatment concepts individually to disease biology. Machine Learning (ML) is a branch of computer science that can process large data sets for a plethora of purposes. The underlying mechanism does not necessarily begin with a manually drafted hypothesis model. Rather the ML algorithms can detect patterns in pre-processed data and derive abstract information. We used ML to predict CR and 2-year overall survival (OS) in a large multi-center cohort of 1383 AML patients who received intensive induction therapy using clinical, laboratory, cytogenetic and molecular genetic data. To enable a customizable and reusable technological approach and achieve optimal results, we designed a data-driven platform with an embedded, automated ML pipeline integrating state-of-the-art software technology for data management and ML models. The platform consists of five scalable modules for data import and modelling, data transformation, model refinement, machine learning algorithms, feature support and performance feedback that are executed in an iterative manner to approach step-wisely the optimal configuration. To reduce dimensionality and the the risk of overfitting, dynamic feature selection was used, i.e. features were selected according to their support by feature selection algorithms. To be included in an ML model, a feature had to pass a pre-determined threshold of overall predictive power determined by summing the normalized scores of the feature selection algorithms. Features below the threshold were automatically excluded from the ML models for the respective iteration. In that way, features of high redundancy or low entropy were automatically filtered out. Our classification algorithms were completely agnostic of pre-existing risk classifications and autonomously selected predictive features both including established markers of favorable or adverse risk as well as identifying markers of so-far controversial relevance. De novo AML, extramedullary AML, double-mutated (dm) CEBPA, mutations of CEBPA-bZIP, NPM1, FLT3-ITD, ASXL1, RUNX1, SF3B1, IKZF1, TP53, U2AF1, t(8;21), inv(16)/t(16;16), del5/del5q, del17, normal or complex karyotypes, age and hemoglobin at initial diagnosis were statistically significant markers predictive of CR while t(8;21), del5/del5q, inv(16)/t(16;16), del17, dm CEBPA, CEBPA-bZIP, NPM1, FLT3-ITD , DNMT3A, SF3B1, U2AF1, TP53, age, white blood cell count, peripheral blast count, serum LDH and Hb at initial diagnosis as well as extramedullary manifestations were predictive for 2-year OS. For prediction of CR and 2-year OS, AUROCs ranged between 0.77 - 0.86 and 0.63 - 0.74, respectively. We provide a method to automatically select predictive features from different data types, cope with gaps and redundancies, apply and optimize different ML models, and evaluate optimal configurations in a scalable and reusable ML platform. In a proof-of-concept manner, our algorithms utilize both established markers of favorable or adverse risk and also provide further evidence for the roles of U2AF1, IKZF1, SF3B1, DNMT3A and bZIP mutations of CEBPA in AML risk prediction. Our study serves as a fundament for prospective validation and data-driven ML-guided risk assessment in AML at initial diagnosis for the individual patient. Image caption: Patient features were automatically selected by machine learning to predict complete remission (CR) and 2-year overall survival (OS) after intensive induction therapy. Based on a continuous feature support metric with a predefined cut-off of 0.5 (determined by optimal classification performance), 27 and 25 features were automatically selected for prediction of CR (A) and 2-year OS (C), respectively. For each of these features predicted by machine learning, odds ratios and 95% confidence intervals (CI) were calculated for CR (B) and 2 year OS (D). BMB: bone marrow blast count; FLT3h/low: FLT3-ITD ratio, h=high>0.5; Hb: hemoglobin; karyotype, c: complex aberrant karyotype (≥ 3 aberrations); karyotype, n: normal karyotype (no aberrations); LDH: lactate dehydrogenase; PBB: peripheral blood blast count; PLT: platelet count; WBC: white blood cell count. Figure 1 Figure 1. Disclosures Schetelig: Roche: Honoraria, Other: lecture fees; Novartis: Honoraria, Other: lecture fees; BMS: Honoraria, Other: lecture fees; Abbvie: Honoraria, Other: lecture fees; AstraZeneca: Honoraria, Other: lecture fees; Gilead: Honoraria, Other: lecture fees; Janssen: Honoraria, Other: lecture fees . Platzbecker: Janssen: Honoraria; Celgene/BMS: Honoraria; AbbVie: Honoraria; Novartis: Honoraria; Takeda: Honoraria; Geron: Honoraria. Müller-Tidow: Pfizer: Research Funding; Janssen: Consultancy, Research Funding; Bioline: Research Funding. Baldus: Celgene/BMS: Honoraria; Amgen: Honoraria; Novartis: Honoraria; Jazz: Honoraria. Krause: Siemens: Research Funding; Takeda: Honoraria; Pfizer: Honoraria; art-tempi: Honoraria; Kosmas: Honoraria; Gilead: Other: travel support; Abbvie: Other: travel support. Haenel: Bayer Vital: Honoraria; Jazz: Consultancy, Honoraria; GSK: Consultancy; Takeda: Consultancy, Honoraria; Novartis: Consultancy, Honoraria; Roche: Consultancy, Honoraria; Amgen: Consultancy; Celgene: Consultancy, Honoraria. Schliemann: Philogen S.p.A.: Consultancy, Honoraria, Research Funding; Abbvie: Consultancy, Other: travel grants; Astellas: Consultancy; AstraZeneca: Consultancy; Boehringer-Ingelheim: Research Funding; BMS: Consultancy, Other: travel grants; Jazz Pharmaceuticals: Consultancy, Research Funding; Novartis: Consultancy; Roche: Consultancy; Pfizer: Consultancy. Middeke: Roche: Consultancy, Honoraria; Janssen: Consultancy, Honoraria, Research Funding; Abbvie: Consultancy, Honoraria; Pfizer: Consultancy, Honoraria; Jazz: Consultancy; Astellas: Consultancy, Honoraria; Sanofi: Honoraria, Research Funding; Novartis: Consultancy; Gilead: Consultancy; Glycostem: Consultancy; UCB: Honoraria.

Mutations of the key myeloid transcription factor CCAAT/enhancer binding protein alpha (C/EBPa) are found in 5-10% of patients with acute myeloid leukemia (AML). Two mutational clusters exist, in the aminoterminal transcription activation domains (TAD1 or 2) and in the basic leucine zipper domain (bZIP) located at the carboxyterminal-part of the protein. Biallelic mutations (biCEBPA) have been found to be associated with improved outcome and are now included as an independent entity in the WHO-classification. In contrast, monoallelic CEBPA-mutations (moCEBPA) do not appear to provide prognostic information. We characterized a large cohort of AML patients for CEBPA mutations and further analyzed the mutational spectrum of mono- and biallelic CEBPA-mutant AML patients to better understand potential differences in the biology of these groups. Patients and Methods: Patients (including all age groups) analyzed had a newly diagnosed AML and were registered in clinical protocols of the Study Alliance Leukemia (SAL)(AML96, AML2003 or AML60+, SORAML) or the SAL-register. Screening for CEBPA mutations was done using PCR and capillary electrophoresis. All identified CEBPA mutations were confirmed using conventional Sanger sequencing and the samples were further analyzed using next generation sequencing (Trusight Myeloid Panel, Illumina) for the presence of associated alterations. Results: In the 4578 patients analyzed, 228 (5%) with CEBPA-mutations were identified. An initial analysis revealed substantial clinical differences between the different mutation subtypes. Patients with biCEBPA (n=111) were significantly younger (median age 46 yrs) than wt-CEBPA patients (median 57 yrs; p Disclosures Middeke: Sanofi: Honoraria. Platzbecker:Janssen-Cilag: Honoraria, Research Funding; Celgene Corporation: Honoraria, Research Funding; TEVA Pharmaceutical Industries: Honoraria, Research Funding; Amgen: Honoraria, Research Funding; Novartis: Honoraria, Research Funding. Thiede:AgenDix: Employment, Other: Ownership.