Data Offer New Insights into AML

A recently released dataset from a large cohort of patients with acute myeloid leukemia (AML) offers new insights into the disease and provides information that will drive additional research efforts. The dataset, part of which was recently published, integrates genomic information about tumor samples with data on how cancer cells respond to drugs (Nature 2018;562:526–31). It was developed as an early iteration of the ongoing Beat AML Master Trial.

“AML is the most commonly diagnosed leukemia, and it's also the one with the poorest prognosis,” explains Brian Druker, MD, director of the Knight Cancer Institute at Oregon Health & Science University in Portland and a senior author on the study. Although several therapies have been approved for AML in recent years, he says the standard of care has changed little, in part because “AML is an incredibly heterogeneous leukemia with a fair number of driver mutations that can occur in lots of different combinations.” Previous research has defined at least 11 distinct genetic subtypes of the disease (N Engl J Med 2016;374:2209–21).

To investigate the relationship between genomics and drug responses in AML, Druker and his colleagues collected 672 tumor samples from 562 patients. They used whole-exome and RNA sequencing to identify mutations and gene expression profiles, and performed ex vivo drug sensitivity testing to determine how malignant cells responded to 122 different drugs. Then, they linked genomic data with drug response in each sample.

For example, the researchers found that tumors with combinations of FLT3, NPM1, and DNMT3A mutations were particularly sensitive to the Bruton tyrosine kinase inhibitor ibrutinib (Imbruvica; Johnson & Johnson and Pharmacyclics). Additionally, mutations in TP53 or ASXL1 resulted in a broad pattern of drug resistance, whereas mutations in both BCOR and RUNX1 were associated with increased sensitivity to selective JAK inhibitors. The entire dataset is available via the Vizome platform (http://www.vizome.org/).

Shannon McWeeney, PhD, also of Oregon Health & Science University and a senior author on the study, notes that although massively parallel sequencing has allowed researchers to fully characterize the genomic architecture of AML, pairing genomics with drug response in the same patient samples provides functional context that has been missing in previous large-scale cohort studies. “That has huge implications for therapeutic stratification with regard to clinical trials and ultimately patient care,” she says.

One of those trials is the Beat AML Master Trial, a large umbrella trial spearheaded by the Leukemia and Lymphoma Society (LLS) that matches patients with first-line therapies based on genomic sequencing of tumors. “The goal is to take the data that we've just published and use that as a guide for some of the upcoming treatment arms that we'll be adding,” explains Druker, a lead investigator of Beat AML.

Monica Guzman, PhD, of Weill Cornell Medicine in New York, NY, who was not involved in the study, calls it “a wonderful resource for other researchers that are interested in identifying drugs that [might] target tumors that are difficult to eliminate.” Her lab, for example, is already exploring data relevant to their work on drug resistance in patients with a PTPN11 mutation.

Ravi Majeti, MD, PhD, of Stanford University in California, who was also not connected to the study but has received research support from LLS, plans to integrate the data into his research on IDH mutations and drug sensitivity. For him, the study, along with the Beat AML Master Trial, supports a broader goal of quickly evaluating different drugs and drug combinations for AML.

He says that although several agents have recently been approved for AML and many more are in development, “we're going to likely see that only a subset of patients will respond to any individual agent—the real hope is to be able to figure out which drug is right for which patient so that the entire population affected with this disease will have better outcomes.” –Catherine Caruso

For more news on cancer research, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/CDNews.