User login
NEWPORT BEACH, CALIF. – New research suggests patients with normal karyotype acute myeloid leukemia (NK-AML) can be divided into four risk groups associated with overall survival.
Investigators used machine learning algorithms to study the association between mutations and overall survival in 1,352 patients with NK-AML. The analysis revealed combinations of mutations that could be used to classify NK-AML patients into favorable, intermediate-1, intermediate-2, and unfavorable risk groups.
For example, patients who had NPM1 mutations but wild-type FLT3-ITD and DNMT3A, had a median overall survival of 99.1 months and could be classified as favorable risk. Conversely, patients who had NPM1, FLT3-ITD, and DNMT3A mutations, had a median overall survival of 13.4 months and could be classified as unfavorable risk.
Aziz Nazha, MD, of the Cleveland Clinic, and his colleagues conducted this research and presented the findings at the Acute Leukemia Forum of Hemedicus.
The investigators looked at genomic and clinical data from 1,352 patients with NK-AML. The patients were a median age of 55 years and had a median white blood cell count of 21.3 x 109/L, a median hemoglobin of 9.1 g/dL, and a median platelet count of 61 x 109/L. More than half of patients (57.3%) were male.
The patients were screened for 35 genes that are commonly mutated in AML and other myeloid malignancies. The investigators used machine learning algorithms, including random survival forest and recommender system algorithms, to study the association between mutations and overall survival in an “unbiased” way.
Dr. Nazha said there were a median of three mutations per patient sample, and “there are some competing interests between those mutations to impact the prognosis of the patient.”
The investigators used the mutations and their associations with overall survival to classify patients into the risk groups outlined in the table below.
These findings can improve the risk stratification of NK-AML and may aid physicians in making treatment decisions, according to Dr. Nazha and his colleagues. To move this work forward, the investigators are attempting to develop a personalized model that can make predictions specific to an individual patient based on that patient’s mutation information.
Dr. Nazha reported having no financial disclosures relevant to this research. Other investigators reported relationships with the Munich Leukemia Laboratory.
The Acute Leukemia Forum is held by Hemedicus, which is owned by the same company as this news organization.
NEWPORT BEACH, CALIF. – New research suggests patients with normal karyotype acute myeloid leukemia (NK-AML) can be divided into four risk groups associated with overall survival.
Investigators used machine learning algorithms to study the association between mutations and overall survival in 1,352 patients with NK-AML. The analysis revealed combinations of mutations that could be used to classify NK-AML patients into favorable, intermediate-1, intermediate-2, and unfavorable risk groups.
For example, patients who had NPM1 mutations but wild-type FLT3-ITD and DNMT3A, had a median overall survival of 99.1 months and could be classified as favorable risk. Conversely, patients who had NPM1, FLT3-ITD, and DNMT3A mutations, had a median overall survival of 13.4 months and could be classified as unfavorable risk.
Aziz Nazha, MD, of the Cleveland Clinic, and his colleagues conducted this research and presented the findings at the Acute Leukemia Forum of Hemedicus.
The investigators looked at genomic and clinical data from 1,352 patients with NK-AML. The patients were a median age of 55 years and had a median white blood cell count of 21.3 x 109/L, a median hemoglobin of 9.1 g/dL, and a median platelet count of 61 x 109/L. More than half of patients (57.3%) were male.
The patients were screened for 35 genes that are commonly mutated in AML and other myeloid malignancies. The investigators used machine learning algorithms, including random survival forest and recommender system algorithms, to study the association between mutations and overall survival in an “unbiased” way.
Dr. Nazha said there were a median of three mutations per patient sample, and “there are some competing interests between those mutations to impact the prognosis of the patient.”
The investigators used the mutations and their associations with overall survival to classify patients into the risk groups outlined in the table below.
These findings can improve the risk stratification of NK-AML and may aid physicians in making treatment decisions, according to Dr. Nazha and his colleagues. To move this work forward, the investigators are attempting to develop a personalized model that can make predictions specific to an individual patient based on that patient’s mutation information.
Dr. Nazha reported having no financial disclosures relevant to this research. Other investigators reported relationships with the Munich Leukemia Laboratory.
The Acute Leukemia Forum is held by Hemedicus, which is owned by the same company as this news organization.
NEWPORT BEACH, CALIF. – New research suggests patients with normal karyotype acute myeloid leukemia (NK-AML) can be divided into four risk groups associated with overall survival.
Investigators used machine learning algorithms to study the association between mutations and overall survival in 1,352 patients with NK-AML. The analysis revealed combinations of mutations that could be used to classify NK-AML patients into favorable, intermediate-1, intermediate-2, and unfavorable risk groups.
For example, patients who had NPM1 mutations but wild-type FLT3-ITD and DNMT3A, had a median overall survival of 99.1 months and could be classified as favorable risk. Conversely, patients who had NPM1, FLT3-ITD, and DNMT3A mutations, had a median overall survival of 13.4 months and could be classified as unfavorable risk.
Aziz Nazha, MD, of the Cleveland Clinic, and his colleagues conducted this research and presented the findings at the Acute Leukemia Forum of Hemedicus.
The investigators looked at genomic and clinical data from 1,352 patients with NK-AML. The patients were a median age of 55 years and had a median white blood cell count of 21.3 x 109/L, a median hemoglobin of 9.1 g/dL, and a median platelet count of 61 x 109/L. More than half of patients (57.3%) were male.
The patients were screened for 35 genes that are commonly mutated in AML and other myeloid malignancies. The investigators used machine learning algorithms, including random survival forest and recommender system algorithms, to study the association between mutations and overall survival in an “unbiased” way.
Dr. Nazha said there were a median of three mutations per patient sample, and “there are some competing interests between those mutations to impact the prognosis of the patient.”
The investigators used the mutations and their associations with overall survival to classify patients into the risk groups outlined in the table below.
These findings can improve the risk stratification of NK-AML and may aid physicians in making treatment decisions, according to Dr. Nazha and his colleagues. To move this work forward, the investigators are attempting to develop a personalized model that can make predictions specific to an individual patient based on that patient’s mutation information.
Dr. Nazha reported having no financial disclosures relevant to this research. Other investigators reported relationships with the Munich Leukemia Laboratory.
The Acute Leukemia Forum is held by Hemedicus, which is owned by the same company as this news organization.
REPORTING FROM ALF 2019