An international team of researchers has created statistical models to estimate the probability of death in individuals receiving non-intensive treatments for acute myeloid leukemia (AML).
Acute myeloid leukemia patients are generally treated with strong chemotherapy, and numerous statistical models have been created to predict whether a patient would benefit from this treatment.
However, some AML patients, notably those aged 75 and older and those with other illnesses, may be unable to endure the side effects of aggressive chemotherapy and are consequently treated with less intensive regimens. There are currently no methods that can properly anticipate which patients will benefit the most from these non-intensive therapy.
“To our knowledge, this is the first attempt to build tools to predict outcomes for patients receiving non-intensive AML therapy to inform decision-making and, ultimately, improve treatment outcomes,” says Dr. Megan Othus, senior author.
To addreTo address these limitations, the researchers created and evaluated a number of statistical models that predicted the probability of early death in AML patients receiving non-intensive treatments. The models were initially created using data from the MRC/LI-1 NCRI’s clinical trial, which included 796 patients (median age 75). The scientists next verified their models’ accuracy against data from another 540 patients (median age 77) who had joined in three SWOG Cancer Research Network leukemia clinical studies – S0432, S0703, and S1612.
The models take into account a patient’s age as well as a variety of health indicators such as performance status (how well the patient can perform everyday tasks), white blood cell and platelet counts, the presence or absence of a specific genetic mutation (NPM1), and several self-reported measures of the patient’s quality of life.
The researchers discovered that, while all of their models were only marginally accurate in predicting early death in both the MRC/NCRI trial patients and the SWOG trial patients, the most effective models were those that incorporated several quality-of-life scores from the QLQ-C30 instrument, a widely used patient-reported outcome survey.
According to the researchers, their findings emphasize the difficulty in predicting outcomes for these individuals using only commonly accessible clinical data.
“Making predictive models that are more accurate and useful,” adds Dr. Othus, “may require incorporating information from additional blood and toxicity biomarkers collected in the early stages of a patient’s treatment or information from additional patient-reported outcome measures.”
Source: SWOG Cancer Research Network
Image Credit: Getty
You were reading: Scientists find a new way to predict risk of early death in AML patients