Grant: Doctoral Research Award - PA - Bioinformatics and Computational Biology

Acute myeloid leukemia (AML) is a deadly blood cancer with a 5-year survival rate of only 32%. While most patients respond to treatment initially, the recurrence of AML is very common. In order to monitor the disease, clinicians often use advanced DNA sequencing methods that enable the identification of mutations in the patient's blood. However, the detection of mutations is usually insufficient to determine if a patient is likely to experience relapse. The need for better risk assessment methods is crucial because early identification of high-risk patients means that clinicians will have a greater window of opportunity to intervene with more aggressive, tailored therapeutic strategies. In this proposal, I aim to develop a new method for predicting relapse risk in AML patients who achieve remission. Instead of relying only on mutations, it will also account for another layer of information known as DNA methylation. I will use a specialized assay to detect both mutations and DNA methylation patterns in the blood of AML patients who received treatment. This data will be mined for biomarkers that are correlated with how long a patient is able to maintain remission. Finally, these molecular signatures will be used to develop a machine learning model that can predict if a patient has a high risk of experiencing relapse. Ultimately, the results from this project will lead to an optimized test that clinicians can use to rapidly identify high-risk patients and take preventive measures. This work will improve the dismal survival rates in AML and offer patients an increased quality of life for a longer period of time.