scholarly journals Prediction of Complete Remission and Survival in Acute Myeloid Leukemia Using Supervised Machine Learning

Blood ◽  
2021 ◽  
Vol 138 (Supplement 1) ◽  
pp. 108-108
Author(s):  
Jan-Niklas Eckardt ◽  
Christoph Rollig ◽  
Michael Kramer ◽  
Sebastian Stasik ◽  
Julia-Annabell Georgi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Complete Remission ◽  
Myeloid Leukemia ◽  
Research Funding ◽  
Initial Diagnosis ◽  
Data Driven ◽  
Blood Cell Count ◽  
Blast Count ◽  
Selection Algorithms

Abstract 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.

scholarly journals Clinical Outcomes Following Treatment with Gilteritinib or Quizartinib in Patients with Relapsed/Refractory FLT3-ITD+ Acute Myeloid Leukemia

Blood ◽  
2020 ◽  
Vol 136 (Supplement 1) ◽  
pp. 42-43
Author(s):  
Alexander E. Perl ◽  
Qiaoyang Lu ◽  
Alan Fan ◽  
Nahla Hasabou ◽  
Erhan Berrak ◽  
...  
Keyword(s):  
Acute Myeloid Leukemia ◽  
Complete Remission ◽  
Myeloid Leukemia ◽  
Research Funding ◽  
Phase 3 ◽  
Hematopoietic Stem ◽  
Travel Costs ◽  
Baseline Characteristics ◽  
Acute Myeloid

Background: Gilteritinib is approved for patients (pts) with relapsed/refractory (R/R) FLT3-mutated acute myeloid leukemia (AML), based on findings from the phase 3 ADMIRAL trial (Perl AE, et al. N Engl J Med. 2019). A phase 3 trial, QuANTUM-R, demonstrated the benefit of quizartinib in pts with R/R AML with FLT3 internal tandem duplication (FLT3-ITD) mutations (Cortes JE, et al. Lancet Oncol. 2019). Although eligibility criteria across both studies were similar, QuANTUM-R was more stringent as to prior therapy intensity and remission duration, which potentially enriched for higher-risk pts. We sought to describe outcomes from ADMIRAL among pts who otherwise met eligibility for QuANTUM-R. Methods: In this post-hoc analysis, a subset of pts from ADMIRAL were matched with R/R FLT3-ITD+ AML pts from QuANTUM-R on the basis of baseline characteristics and prior treatment criteria. Matched pts were either refractory to initial anthracycline-based chemotherapy or had relapsed ≤6 mos after achieving composite complete remission (CRc) with an anthracycline-based regimen. Results: Overall, 218 pts with R/R FLT3-ITD+ AML in the ADMIRAL trial (gilteritinib, n=140; salvage chemotherapy [SC], n=78) were matched with the QuANTUM-R intention-to treat (ITT) population (N=367; quizartinib, n=245; SC, n=122). Proportions of pts preselected for high-intensity SC were 66% (n=143/218) in the matched ADMIRAL ITT population and 77% (n=281/367) in the QuANTUM-R ITT populations. Demographic and baseline characteristics of the matched ADMIRAL ITT population and QuANTUM-R ITT population were similar. Median durations of exposure to gilteritinib and quizartinib were 3.8 mos and 3.2 mos, respectively, and median number of treatment cycles received were five and four, respectively. Rates of hematopoietic stem cell transplantation (HSCT) were similar in pts treated with gilteritinib (35%; n=49/140) or quizartinib (32%; n=78/245), as were the proportions of pts who resumed gilteritinib (23%; n=32/140) or quizartinib (20%; n=48/245) therapy post-HSCT. Median overall survival (OS) in pts treated with gilteritinib or quizartinib was longer than that observed with SC. After a median follow-up of 17.4 mos, median OS was 10.2 mos with gilteritinib versus 5.6 mos with SC (hazard ratio [HR]=0.573 [95% CI: 0.403, 0.814]; one-sided nominal P=0.0008). After a median follow-up of 23.5 mos, median OS with quizartinib was 6.2 mos versus 4.7 mos with SC (HR=0.76 [95% CI: 0.58-0.98]; one-sided P=0.02). After censoring for HSCT, median OS was 9.3 mos with gilteritinib versus 5.5 mos with SC (HR=0.525 [95% CI: 0.356-0.775]; nominal one-sided P=0.0005), and 5.7 mos versus 4.6 mos with quizartinib versus SC, respectively (HR=0.79 [95% CI: 0.59, 1.05]; one-sided P=0.05). In both QuANTUM-R and matched ADMIRAL populations, the survival benefits of quizartinib and gilteritinib compared with SC were maintained across multiple subgroups, including high FLT3-ITD allelic ratio subsets. Compared with SC, high CRc rates were observed in pts treated with either gilteritinib (57%; n=80/140) or quizartinib (48%; n=118/245). The complete remission (CR) rate with gilteritinib was 23% (n=32/140), whereas the CR rate with quizartinib was 4% (n=10/245) (Table). Median time to achieve CRc was 1.8 mos with gilteritinib and 1.1 mos with quizartinib, median duration of CRc was 5.5 mos with gilteritinib and 2.8 mos with quizartinib. The safety profiles of gilteritinib and quizartinib were generally similar, though aspartate or alanine aminotransferase elevations (any grade) were more frequent with gilteritinib (41-44%) than quizartinib (≤13%), whereas neutropenia (14% vs 34%, respectively), fatigue (24% vs 39%, respectively), and prolonged QT intervals (9% vs 27%, respectively) were more frequent with quizartinib. Conclusions: In pts with R/R FLT3-ITD+ AML and similar baseline characteristics, both gilteritinib and quizartinib were generally well tolerated and associated with improved survival and treatment response compared with SC. Responses to gilteritinib and quizartinib, as measured by CRc, were similar; blood count recovery varied between the two FLT3 inhibitors. Although cross-study comparisons have substantial limitations, the findings suggest that while remission is achieved faster with quizartinib, response may be more durable and survival potentially longer with gilteritinib. Disclosures Perl: Syndax: Consultancy, Honoraria; Leukemia & Lymphoma Society, Beat AML: Consultancy; Novartis: Honoraria, Other, Research Funding; Agios: Consultancy, Honoraria, Other; Jazz: Honoraria, Other; FORMA Therapeutics: Consultancy, Honoraria, Other; Daiichi Sankyo: Consultancy, Honoraria, Other: Writing/editorial support, travel costs for meetings, Research Funding; FUJIFILM Pharmaceuticals USA, Inc: Research Funding; New Link Genetics: Honoraria, Other; Arog Pharmaceuticals Inc: Other: uncompensated consulting, travel costs for meetings; Actinium Pharmaceuticals Inc: Consultancy, Honoraria, Research Funding; Biomed Valley Discoveries: Research Funding; Astellas: Consultancy, Honoraria, Other: writing/editorial support, travel costs for meeting presentations related to study, Research Funding; Bayer HealthCare Pharmaceuticals: Research Funding; AbbVie Inc: Consultancy, Honoraria, Other, Research Funding; Takeda: Honoraria, Other: Travel costs for meeting; Loxo Oncology Inc, a wholly owned subsidiary of Eli Lilly & Company: Consultancy, Honoraria, Other. Lu:Astellas: Current Employment. Fan:Astellas Pharma: Current Employment. Hasabou:Astellas Pharma: Current Employment. Berrak:Astellas: Current Employment. Tiu:Eli Lilly & Company: Current equity holder in publicly-traded company, Ended employment in the past 24 months; Astellas Pharma Global Development: Current Employment.

scholarly journals Increased Idarubicin Dosage during Consolidation Therapy for Adult Acute Myeloid Leukemia Improves Leukemia-Free Survival

Blood ◽  
2016 ◽  
Vol 128 (22) ◽  
pp. 338-338
Author(s):  
Bradstock Kenneth ◽  
Emma Link ◽  
Juliana Di Iulio ◽  
Jeff Szer ◽  
Paula Marlton ◽  
...  
Keyword(s):  
Acute Myeloid Leukemia ◽  
Complete Remission ◽  
Board Of Directors ◽  
Induction Therapy ◽  
Myeloid Leukemia ◽  
Research Funding ◽  
Consolidation Therapy ◽  
Advisory Committees ◽  
Free Survival ◽  
Acute Myeloid

Abstract Background: Anthracylines are one of the major classes of drugs active against acute myeloid leukemia (AML). Increased doses of daunorubicin during induction therapy for AML have been shown to improve remission rates and survival. The ALLG used idarubicin in induction therapy at a dose of 9 mg/m2 x 3 days (total dose 27 mg/m2) in combination with high-dose cytarabine and etoposide (Blood 2005, 105:481), but showed that a total idarubicin dose of 36 mg/m2 was too toxic in this context (Leukemia 2001, 15:1331). In order to further improve outcomes in adult AML by anthracycline dose escalation, we conducted a phase 3 trial comparing standard to an increased idarubicin dose during consolidation therapy. Methods: Patients achieving complete remission after 1 or 2 cycles of intensive induction therapy (idarubicin 9 mg/m2 daily x3, cytarabine 3 g/m2 twice daily on days 1,3,5 and 7, and etoposide 75 mg/m2 daily x7; ICE protocol) were randomized to receive 2 cycles of consolidation therapy with cytarabine 100 mg/m2 per day for 5 days, etoposide 75 mg/m2 for 5 days, and idarubicin 9mg/m2 daily for either 2 or 3 days (standard and intensive arms respectively). No further protocol therapy was given. The primary endpoint was leukemia-free survival from randomization to consolidation therapy (LFS) with overall survival (OS) as secondary endpoint. Results: A total of 422 patients with AML (excluding cases with CBF rearrangements or APL) aged 16 to 60 years were enrolled between 2003-10, with 345 (82%) achieving complete remission, and 293 being randomized to standard (n=146) or intensive (n=147) consolidation arms. The median age was 45 years in both arms (range 16- 60), and both groups were balanced for intermediate versus unfavorable karyotypes and for frequency of mutations involving FLT3-ITD and NPM1 genes. Of the randomized patients, 120 in the standard arm (82%) and 95 in the intensive arm (65%) received the second consolidation cycle (p<0.001). The median total dose of idarubicin received in the 2 consolidation courses was 36 mg/m2 (range 17-45), or 99% (47-125%) of the protocol dose in the standard arm, versus 53 mg/m2 (18-73), or 98% (33-136%) of the protocol dose in the intensive arm. The durations of grades 3-4 neutropenia and thrombocytopenia were significantly longer in the intensive arm, but there were no differences in grade 3 or 4 non-hematological toxicities. There were no non-relapse deaths during consolidation on the standard arm and 2 in the intensive (0% vs 1%; p =0.50). Subsequently, 41 patients in the standard arm and 37 in the intensive arm underwent elective allogeneic BMT during first remission. On intention to-treat analysis uncensored for transplant and with a median follow-up time of 5.3 years (range 0.6 - 9.9), there was improvement in LFS in the intensive arm compared with the standard arm (3 year LFS 47% (95% CI 40-56%) versus 35% (28-44%); HR 0.74 (95% CI 0.55-0.99); p=0.045) (Figure 1). The 3 year OS for the intensive arm was 61% (95% CI 54-70%) and 50% (95% CI 43-59%) for the standard arm; HR 0.75 (95% CI 0.54-1.05); p=0.092). Although adverse cytogenetics, presence of FLT3-ITD mutation, and absence of NPM1 mutation were all associated with poorer outcomes, there was no evidence of a benefit of intensive consolidation being confined to specific cytogenetic or gene mutation sub-groups. Conclusion: We conclude that in adult patients in complete remission after intensive induction chemotherapy an increased dose of idarubicin delivered during consolidation therapy results in improved LFS, without increased non-hematologic toxicity. Figure 1. Figure 1. Disclosures Szer: Ra Pharma: Honoraria, Membership on an entity's Board of Directors or advisory committees; Alexion Pharmaceuticals, Inc.: Honoraria, Membership on an entity's Board of Directors or advisory committees; Alnylam: Honoraria, Membership on an entity's Board of Directors or advisory committees. Marlton:Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees; AbbVie: Honoraria, Membership on an entity's Board of Directors or advisory committees; Janssen: Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Novartis: Honoraria, Membership on an entity's Board of Directors or advisory committees; Pfizer: Honoraria, Membership on an entity's Board of Directors or advisory committees; Roche: Honoraria, Membership on an entity's Board of Directors or advisory committees. Wei:Novartis: Consultancy, Honoraria, Research Funding; Roche: Consultancy, Honoraria; CTI: Consultancy, Honoraria; Abbvie: Honoraria, Research Funding; Servier: Consultancy, Honoraria, Research Funding; Celgene: Consultancy, Honoraria, Research Funding; Amgen: Consultancy, Honoraria, Research Funding. Cartwright:ROCHE: Consultancy, Membership on an entity's Board of Directors or advisory committees. Roberts:Servier: Research Funding; Janssen: Research Funding; Genentech: Research Funding; AbbVie: Research Funding. Mills:Novartis: Membership on an entity's Board of Directors or advisory committees, Other: Meeting attendance sponsorship. Gill:Janssen: Membership on an entity's Board of Directors or advisory committees. Seymour:Roche: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Janssen: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Research Funding, Speakers Bureau; Gilead: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Genentech: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees; Celgene: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Travel support, Speakers Bureau; AbbVie: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: Travel support, Research Funding, Speakers Bureau; Takeda: Consultancy, Honoraria, Membership on an entity's Board of Directors or advisory committees.

scholarly journals Predicting Complete Remission of Acute Myeloid Leukemia: Machine Learning Applied to Gene Expression

2019 ◽  
Vol 18 ◽  
pp. 117693511983554 ◽  
Author(s):  
Ophir Gal ◽  
Noam Auslander ◽  
Yu Fan ◽  
Daoud Meerzaman
Keyword(s):  
Gene Expression ◽  
Machine Learning ◽  
Acute Myeloid Leukemia ◽  
Complete Remission ◽  
Myeloid Leukemia ◽  
Statistical Significance ◽  
Expression Patterns ◽  
Pathway Enrichment Analysis ◽  
Data Set ◽  
Acute Myeloid

Machine learning (ML) is a useful tool for advancing our understanding of the patterns and significance of biomedical data. Given the growing trend on the application of ML techniques in precision medicine, here we present an ML technique which predicts the likelihood of complete remission (CR) in patients diagnosed with acute myeloid leukemia (AML). In this study, we explored the question of whether ML algorithms designed to analyze gene-expression patterns obtained through RNA sequencing (RNA-seq) can be used to accurately predict the likelihood of CR in pediatric AML patients who have received induction therapy. We employed tests of statistical significance to determine which genes were differentially expressed in the samples derived from patients who achieved CR after 2 courses of treatment and the samples taken from patients who did not benefit. We tuned classifier hyperparameters to optimize performance and used multiple methods to guide our feature selection as well as our assessment of algorithm performance. To identify the model which performed best within the context of this study, we plotted receiver operating characteristic (ROC) curves. Using the top 75 genes from the k-nearest neighbors algorithm (K-NN) model ( K = 27) yielded the best area-under-the-curve (AUC) score that we obtained: 0.84. When we finally tested the previously unseen test data set, the top 50 genes yielded the best AUC = 0.81. Pathway enrichment analysis for these 50 genes showed that the guanosine diphosphate fucose (GDP-fucose) biosynthesis pathway is the most significant with an adjusted P value = .0092, which may suggest the vital role of N-glycosylation in AML.

scholarly journals Streaming feature selection algorithms for big data: A survey

2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Noura AlNuaimi ◽  
Mohammad Mehedy Masud ◽  
Mohamed Adel Serhani ◽  
Nazar Zaki
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Big Data ◽  
Real Time ◽  
Relevant Literature ◽  
Heterogeneous Data ◽  
Proper Solution ◽  
Exact Figure ◽  
Selection Algorithms ◽  
Over Time

Organizations in many domains generate a considerable amount of heterogeneous data every day. Such data can be processed to enhance these organizations’ decisions in real time. However, storing and processing large and varied datasets (known as big data) is challenging to do in real time. In machine learning, streaming feature selection has always been considered a superior technique for selecting the relevant subset features from highly dimensional data and thus reducing learning complexity. In the relevant literature, streaming feature selection refers to the features that arrive consecutively over time; despite a lack of exact figure on the number of features, numbers of instances are well-established. Many scholars in the field have proposed streaming-feature-selection algorithms in attempts to find the proper solution to this problem. This paper presents an exhaustive and methodological introduction of these techniques. This study provides a review of the traditional feature-selection algorithms and then scrutinizes the current algorithms that use streaming feature selection to determine their strengths and weaknesses. The survey also sheds light on the ongoing challenges in big-data research.

scholarly journals Data driven feature selection and machine learning to detect misplaced V1 and V2 chest electrodes when recording the 12‑lead electrocardiogram

2019 ◽  
Vol 57 ◽  
pp. 39-43 ◽  
Author(s):  
Khaled Rjoob ◽  
Raymond Bond ◽  
Dewar Finlay ◽  
Victoria McGilligan ◽  
Stephen J. Leslie ◽  
...  
Keyword(s):  
Machine Learning ◽  
Feature Selection ◽  
Data Driven

A NOVEL FEATURE SELECTION ALGORITHM WITH SUPERVISED MUTUAL INFORMATION FOR CLASSIFICATION

2013 ◽  
Vol 22 (04) ◽  
pp. 1350027
Author(s):  
JAGANATHAN PALANICHAMY ◽  
KUPPUCHAMY RAMASAMY
Keyword(s):  
Machine Learning ◽  
Data Mining ◽  
Feature Selection ◽  
Mutual Information ◽  
Selection Algorithm ◽  
Feature Selection Algorithm ◽  
Class A ◽  
Selection Algorithms ◽  
The Relationship ◽  
Class Variable

Feature selection is essential in data mining and pattern recognition, especially for database classification. During past years, several feature selection algorithms have been proposed to measure the relevance of various features to each class. A suitable feature selection algorithm normally maximizes the relevancy and minimizes the redundancy of the selected features. The mutual information measure can successfully estimate the dependency of features on the entire sampling space, but it cannot exactly represent the redundancies among features. In this paper, a novel feature selection algorithm is proposed based on maximum relevance and minimum redundancy criterion. The mutual information is used to measure the relevancy of each feature with class variable and calculate the redundancy by utilizing the relationship between candidate features, selected features and class variables. The effectiveness is tested with ten benchmarked datasets available in UCI Machine Learning Repository. The experimental results show better performance when compared with some existing algorithms.

scholarly journals Hypomethylating Agent and Venetoclax Combination Yields Comparable Outcomes to CPX-351 in Newly Diagnosed Acute Myeloid Leukemia

Blood ◽  
2019 ◽  
Vol 134 (Supplement_1) ◽  
pp. 3895-3895
Author(s):  
Hannah Asghari ◽  
Dasom Lee ◽  
Yehuda E. Deutsch ◽  
Onyee Chan ◽  
Najla Al Ali ◽  
...  
Keyword(s):  
Acute Myeloid Leukemia ◽  
Complete Remission ◽  
Board Of Directors ◽  
Myeloid Leukemia ◽  
Research Funding ◽  
Risk Group ◽  
Newly Diagnosed ◽  
Advisory Committees ◽  
Acute Myeloid

Background The therapeutic landscape for acute myeloid leukemia (AML) has become complex with recent drug approvals. CPX-351 has become standard-of-care for patients (pts) with therapy-related AML and AML with myelodysplasia-related changes. Moreover, earlier phase studies combining hypomethylating agents (HMA) and Venetoclax (HMA+Ven) in the frontline setting for elderly patients have demonstrated high response rates and improved survival. Given the overlapping indications, yet lack of comparative outcome data between these therapeutic regimens, treatment decisions have become challenging in the frontline setting. Therefore, we compared the outcomes of newly diagnosed AML pts receiving HMA+Ven vs. CPX-351. Methods We retrospectively annotated 119 pts that received frontline treatment with HMA+Ven and CPX-351 at Moffitt Cancer Center and Memorial Healthcare System between 2013 and 2019. Pts were divided in two cohorts: HMA+Ven (Cohort A) or CPX-351(Cohort B). Via comprehensive chart review of each patient that received HMA+Ven, we further classified a subgroup of pts meeting criteria to receive CPX-351 as CPX-351eligible. Clinical and molecular data were abstracted for each patient in accordance with IRB requirements. Overall response rate (ORR) was the combined total of complete remission (CR), complete remission with incomplete count recovery (CRi), and morphologic leukemia free state (MLFS). Fisher's Exact method was used to determine significance. Kaplan-Meier analysis was performed to estimate median overall survival (mOS) with log-rank test to determine significance. All p-values are two-sided. Results Out of 119 total pts, 41 pts received HMA+Ven (Cohort A) and 78 pts received CPX-351 (Cohort B) with baseline characteristics outlined in Table 1. Among 111 response evaluable pts, ORR was 64.1% in Cohort A, including 28.2% with CR and 28.2% with CRi (Table 2). ORR was 50.0% in Cohort B, comprised of CR in 29.2% and CRi in 18.1%. There was no difference in ORR between Cohort A and Cohort B (64.1% vs. 50%, p 0.17). A significantly greater fraction of pts in Cohort B underwent allogeneic stem cell transplant (allo-SCT) (24.4% vs. 2.4%, p=0.004). ORR was higher in pts with European LeukemiaNet (ELN)-defined favorable/intermediate (fav/int) risk compared to adverse risk group in Cohort A (100% vs. 58.3%, p=0.03), however there was no difference in Cohort B (52.6% vs. 49.1%, p=1.0). ORR was similar among adverse risk groups in both cohorts (58.3% in Cohort A vs. 49.1% in Cohort B, p=0.47). Among responders, median time to best response was significantly longer in Cohort A (61.0 days vs. 40.5 days, p<0.0001). Median duration of response was not reached (NR) in both cohorts. Impact of somatic mutations on ORR is represented in Figure 3. Median follow-up was 6.5 months (mo) in Cohort A and 13.0mo in Cohort B. Median OS was similar in both cohorts (A vs. B, 13.8mo vs. 11.1mo, p=0.82) (Figure 1). Among responders, mOS was NR in Cohort A and 18.2mo in Cohort B (p=0.88) (Figure 2). Compared to Cohort B, mOS was superior for pts with fav/int risk disease in Cohort A (14.2mo (B) vs. NR (A), p=0.045) and not different for adverse risk group (11.1mo (B) vs. 7.3mo (A), p=0.2). Prior HMA exposure was 26.8% in Cohort A and 29.5% in Cohort B for an antecedent hematologic malignancy, however it did not impact mOS (p=0.86) or ORR (p=0.7). Early mortality rates for Cohort A and B were similar at day 30 (2.4% vs. 0%) and day 60 (4.9% vs. 3.8%). Rate of relapse was similar between cohorts A and B (16.0% vs. 30.6%, p=0.24). We then compared the outcomes of pts in Cohort B to CPX-351eligible arm from Cohort A (n=14). ORR and mOS were similar in Cohort B and CPX-351 eligible arm (ORR: 50% vs. 50%, p=1.0; mOS 11.1mo vs. 13.8mo, p=0.43). Only 1 patient (7.1%) of the CPX-351eligible arm underwent allo-SCT. Conclusion Our study demonstrates that HMA+Ven results in comparable response rates and survival outcomes to patients receiving CPX-351 when used as an initial remission therapy for patients with newly diagnosed AML, however the median follow up for patients receiving HMA+Ven was short. Survival did not appear to be impacted by a significantly greater proportion of patients proceeding to allo-SCT in the CPX-351 arm. Overall, HMA+Ven may represent a reasonable frontline remission therapeutic choice in patients with AML and a randomized trial would seem justified. Disclosures Kuykendall: Abbvie: Honoraria; Janssen: Consultancy; Incyte: Honoraria, Speakers Bureau; Celgene: Honoraria. List:Celgene: Membership on an entity's Board of Directors or advisory committees, Research Funding. Lancet:Pfizer: Consultancy, Research Funding; Agios, Biopath, Biosight, Boehringer Inglheim, Celator, Celgene, Janssen, Jazz Pharmaceuticals, Karyopharm, Novartis: Consultancy; Daiichi Sankyo: Consultancy, Other: fees for non-CME/CE services . Sallman:Celyad: Membership on an entity's Board of Directors or advisory committees. Komrokji:celgene: Consultancy; Agios: Consultancy; pfizer: Consultancy; DSI: Consultancy; JAZZ: Speakers Bureau; JAZZ: Consultancy; Novartis: Speakers Bureau; Incyte: Consultancy. Sweet:Abbvie: Membership on an entity's Board of Directors or advisory committees; Stemline: Consultancy; Agios: Membership on an entity's Board of Directors or advisory committees; Bristol Myers Squibb: Membership on an entity's Board of Directors or advisory committees; Incyte: Research Funding; Astellas: Membership on an entity's Board of Directors or advisory committees; Novartis: Membership on an entity's Board of Directors or advisory committees, Speakers Bureau; Pfizer: Consultancy; Celgene: Speakers Bureau; Jazz: Speakers Bureau. Talati:Agios: Honoraria; Jazz Pharmaceuticals: Honoraria, Speakers Bureau; Celgene: Honoraria; Daiichi-Sankyo: Honoraria; Astellas: Honoraria, Speakers Bureau; Pfizer: Honoraria.

scholarly journals Assessing Efficacy and Safety of Daunorubicin and Cytarabine Liposomal in Patients with Acute Myeloid Leukemia

Blood ◽  
2019 ◽  
Vol 134 (Supplement_1) ◽  
pp. 5134-5134
Author(s):  
Elizabeth Rogers ◽  
Maho Hibino ◽  
Rebecca Garcia Hunt ◽  
Leslie Renee Ellis ◽  
Rupali Bhave ◽  
...  
Keyword(s):  
Acute Myeloid Leukemia ◽  
Mortality Rate ◽  
Complete Remission ◽  
Myeloid Leukemia ◽  
Research Funding ◽  
Comprehensive Cancer Center ◽  
Report Generation ◽  
Free Survival ◽  
Younger Patients ◽  
Acute Myeloid

Background. Acute myeloid leukemia (AML) is the most common acute leukemia in adults. The median age of diagnosis is 67 years old, and it has unfavorable outcomes in older patients. Approximately one-third of patients are diagnosed after the age of 75. Thus, as the population continues to increase in age, the incidence of AML will continue to expand (NCCN guidelines: AML. Version 3.2019). The long term disease free survival (DFS) rates for patients > 60 years of age is 5-15% whereas younger patients < 60 years of age have a better DFS rate of up to 40% (Dohner H, et al. N Engl J Med. 2015). Recent advancements have been made in patients with AML, including the approval of daunorubicin and cytarabine liposomal (Vyxeos®) for the treatment of adults with 2 poor risk types of AML: newly diagnosed therapy related AML (t-AML) or AML with myelodysplasia-related changes (AML-MRC). Given the financial constraints of this new medication, our objective was to determine the safety and efficacy of daunorubicin and cytarabine liposomal in our adult patients with t-AML and AML-MRC at a single academic medical center. Methods. This was a single center, retrospective, chart review at Wake Forest Baptist Health (WFBH) Comprehensive Cancer Center from August 1, 2017 to November 1, 2018. Patients were selected via report generation if they had received at least one dose of daunorubicin and cytarabine liposomal during the study period. The initial induction dose of daunorubicin and cytarabine liposomal was 44 mg/m2 of daunorubicin and 100 mg/m2 cytarabine administered on days 1, 3, and 5 for up to 2 cycles to achieve remission. If a second induction was necessary, the same induction doses were given on days 1 and 3 only. The consolidation dose was 29 mg/m2 of daunorubicin and 65 mg/m2 of cytarabine on days 1 and 3 for up to 2 cycles. The primary endpoint was overall survival (OS). Secondary endpoints included event free survival (EFS), 30-day and 60-day mortality, complete remission (CR) and morphologic complete remission with incomplete blood count recovery (CRi), adverse drug reactions, and financial impact to the health system. Descriptive statistics were utilized for demographic data. Time to event data was analyzed using the Kaplan-Meier method. SPSS IBM and Microsoft Excel Software were utilized for data analysis. Results. A total of 37 AML patients were identified as receiving daunorubicin and cytarabine liposomal from August 2017 to November 2018. Of those 37 patients, 27 had AML-MRC and 10 had t-AML. The average patient was a 70 year old Caucasian male with an ECOG performance status of 1 and a Charlson Comorbidity Index of 6 (Table 1). The median OS was 10 months and EFS was 7 months. The 30-day mortality rate was 16% and 60-day mortality rate was 19%. Eighteen patients (49%) achieved a CR and 2 patients (5%) achieved a CRi. A subgroup analysis was conducted for prior hypomethylating agent (HMA) use, age > 75 years old, < 60 years old, molecular mutations including FLT3-ITD and TP53 mutations, and t-AML. Poorer outcomes were noted in patients > age 75, prior HMA use, and the t-AML subgroups. Table 3 highlights the OS, 60-day mortality rate, transplant received and CR+CRi for each subgroup. The median time to platelet and absolute neutrophil count (ANC) recovery was 32 and 33 days, respectively. Eight patients (21.6%) proceeded to transplant post administration of daunorubicin and cytarabine liposomal. All patients experienced at least one adverse event with hematologic being the most commonly observed toxicity (Table 4). Most patients received induction therapy with daunorubicin and cytarabine liposomal in the inpatient setting whereas consolidation was predominantly administered in an outpatient encounter. Conclusions. Daunorubicin and cytarabine liposomal was considered an effective treatment option for patients with t-AML and AML-MRC with a CR+CRi rate of 54%. Younger patients (< 60 years old) exhibited the greatest benefit with an OS of 12 months and 60 day mortality rate of 0%. However, poorer outcomes were demonstrated in elderly patients (> 75 years old), patients with FLT3-ITD positive mutations, and patients with previous HMA use, with an OS less than 2 months in each subgroup and mortality rates ranging up to 60%. Thus, additional studies are necessary to determine the role of daunorubicin and cytarabine liposomal in these higher risk patient subgroups > age 75, FLT3-ITD positive patients, and patients with previous HMA use. Disclosures Manuel: Novartis: Speakers Bureau; Jazz Pharmaceuticals: Speakers Bureau. Pardee:Rafael Pharmaceuticals: Consultancy, Research Funding; Karyopharm: Research Funding; Spherix Intellectual Property: Research Funding; Pharmacyclics/Janssen: Speakers Bureau; Celgene: Speakers Bureau; Amgen: Speakers Bureau; CBM Bipharma: Membership on an entity's Board of Directors or advisory committees. Powell:Janssen: Research Funding; Rafael Pharmaceuticals: Consultancy, Research Funding; Novartis: Consultancy, Speakers Bureau; Jazz Pharmaceuticals: Consultancy, Research Funding, Speakers Bureau; Pfizer: Consultancy, Research Funding.

scholarly journals Extramedullary Disease at Diagnosis of Acute Myeloid Leukemia Does Not Influence Outcome of Patients Undergoing Allogeneic Hematopoietic Cell Transplant in First Complete Remission

Blood ◽  
2015 ◽  
Vol 126 (23) ◽  
pp. 2020-2020
Author(s):  
Fotios V. Michelis ◽  
Hans A. Messner ◽  
Naheed Alam ◽  
Vikas Gupta ◽  
Dennis Dong Hwan Kim ◽  
...  
Keyword(s):  
Acute Myeloid Leukemia ◽  
Complete Remission ◽  
Cumulative Incidence ◽  
Myeloid Leukemia ◽  
Research Funding ◽  
Univariate Analysis ◽  
Multivariable Analysis ◽  
Hematopoietic Cell ◽  
First Complete Remission ◽  
Acute Myeloid

Abstract Occurrence of extramedullary (EM) disease at diagnosis of acute myeloid leukemia (AML) has been associated with increased risk of relapse and worse outcomes post-chemotherapy. There is minimal data in the literature concerning the association with outcomes following allogeneic hematopoietic cell transplantation (HCT). The purpose of this single-centre study was to retrospectively investigate the impact of EM disease at diagnosis on the outcome of 303 patients with AML in first complete remission (CR1) that underwent HCT during the time period 2000-2013. Median age at HCT was 51 years (range 18-71), 151 (50%) patients were female. Myeloablative conditioning (MAC) was used in 202 (67%) patients, reduced-intensity (RIC) in 101 (33%) patients. Donors were related for 194 (64%) patients, unrelated for 109 (36%) patients. Grafts were peripheral blood stem cells (PBSC) in 253 (83%) patients and bone marrow in 50 (17%) patients. Median follow-up of patients alive was 63 months (range 12-168). Cytogenetics at diagnosis were available for 263 (87%) of patients, of which 16 (5%) were favorable, 185 (61%) were intermediate and 62 (20%) were unfavorable risk (MRC classification). Primary induction failure prior to achievement of CR was seen in 67 (22%) patients. In vivo T-cell depletion was performed in 71 (23%) patients. A total of 124 patients (41%) underwent HCT during the years 2000-2006 and 179 patients (59%) during the years 2007-2013. EM disease at diagnosis was seen in 39 patients (13%), of whom 11 patients had CSF disease, 7 patients had gingival infiltration and 5 patients had leukemia cutis. Univariate analysis for overall survival (OS) demonstrated that EM disease at diagnosis had no influence (HR=0.96 for EM, 95%CI=0.60-1.51, p=0.85, Figure 1). Multivariable analysis for OS including the previously described variables verified this observation. EM disease did not influence cumulative incidence of relapse (CIR) on univariate analysis (HR=0.94 for EM, 95%CI=0.45-1.96, p=0.86, Figure 2), and this also was confirmed on multivariable analysis. Moreover, EM disease did not influence cumulative incidence of non-relapse mortality on both univariate (HR=0.94 for EM, 95%CI=0.53-1.66, p=0.83) and multivariable analysis. In conclusion, EM disease at diagnosis of AML in patients achieving CR1, does not seem to influence outcomes post allogeneic HCT. This is significant in the consideration of allogeneic HCT for the treatment of this unfavorable subtype of AML. We are unable to comment on whether a similar percentage of patients with EM disease versus without EM disease, achieve CR1. Figure 1. Figure 1. Figure 2. Figure 2. Disclosures Kim: Bristol-Myers Squibb: Consultancy, Research Funding; Novartis Pharmaceuticals: Consultancy, Research Funding.

Sign in / Sign up

Export Citation Format

Share Document