COVID-19 is an emerging, rapidly evolving siituation.

What people with cancer should know:

Guidance for cancer researchers:

Get the latest public health information from CDC:

Get the latest research information from NIH:

Advancing Personalized Treatment Strategies for Patients with Acute Myeloid Leukemia

Acute Myeloid Leukemia (AML) is a hematologic malignancy characterized by the rapid proliferation of abnormal myeloid cells in the bone marrow and blood. AML is known for its heterogeneity, presenting challenges in treatment and predicting patient outcomes. In a paper recently published in Cell Reports Medicine, CPTAC researchers describe a comprehensive proteogenomic approach to characterizing Acute Myeloid Leukemia (AML). As summarized by study leader Dr. Brian Druker, the goal of this work was “to better understand the pathways that contribute to drug sensitivity and resistance using samples from patients with AML.” Brian went on to say that “these studies may allow better matching of patients with more effective therapies or avoidance of therapies that would be predicted to be ineffective.”

Through non-negative matrix factorization (NMF) on a subset of samples with mRNA, protein, and phosphosite data, the researchers delineated four AML molecular subtypes. Importantly, these subtypes demonstrated biological relevance by correlating with specific mutation patterns and clinical annotations. A key breakthrough emerged with the development of a proteomic-only classification model; by leveraging 147 proteins and phosphosites, this model successfully classified all 210 patient samples into distinct subtypes.

The study delved into differences in drug response patterns between AML subtypes. By exploring responses to 46 drugs across various subtypes and mutations, researchers identified cases wherein subtypes and mutation status jointly predicted drug response. One unique aspect of this study is the integration of multiple types of omic data to uncover common patterns of response to multiple different targeted therapies, resulting in the ability to classify potential therapeutic agents into complementary groups. 

"We were able to look at patterns of drug responses in hundreds of people by including protein and gene measurements together, and this gave us a level of detail that has not been possible in prior studies,” said lead author Dr. Sara Gosline.. “This is a great example where we are able to put our growing knowledge of protein signaling and machine learning models to benefit patients in the future.”

The exploration extended to drug combinations as well, uncovering both additive and antagonist effects. Network analysis offered a deeper understanding of how certain combinations act on different AML subtypes, and linear modeling techniques were used to decipher the molecular mechanisms behind drug responses. Distinct proteomic signatures associated with responses to specific drugs, such as venetoclax and panobinostat, were identified. In conversation with these findings, lead author Dr. Karin Rodland commented, “the identification of complementarity between drugs has the potential to improve combination therapies for AML and counteract the development of resistance to single agents.”

In order to showcase the potential of proteomic subtyping in predicting responses to treatment, the team validated their predictions through in vitro assays on AML cell lines undergoing resistance development.

This pioneering effort demonstrated the power of a comprehensive proteogenomic approach in guiding personalized treatment strategies and provides a foundation for future research with publicly accessible data (ProteomePhosphoproteome). In the words of study leader Dr. Jeffery Tyner, “This study provides a comprehensive proteogenomic dataset that enables us to understand the ways in which AML tumors respond to or evade therapeutics. The biological insights from this study can help to design more effective regimens for the treatment of AML.”