Whole-Genome Sequencing Could Help Guide AML Treatment
, by Elia Ben-Ari
For people with acute myeloid leukemia (AML) or myelodysplastic syndromes (MDS), results from a new study suggest that using whole-genome sequencing to identify the genomic changes in their tumor cells could simplify and improve how doctors select the best treatment regimen for each patient. Whole-genome sequencing can determine the complete genetic makeup of a patient’s sample in a single test, whereas the current approach to identify genomic changes involves several different tests.
The tests are used to assess the risk that patients’ cancer will come back, or relapse, after chemotherapy, and to recommend treatments based on that risk.
In the new study, when compared head to head with conventional testing on the same samples, whole-genome sequencing–based testing detected all of the abnormalities that were detected by conventional chromosome analysis and yielded new findings in about 25% of patients with AML or MDS. And more than half of those patients had a finding that could have potentially changed their treatment, said David Spencer, M.D., Ph.D., of the Washington University School of Medicine in St. Louis, an investigator on the study.
The streamlined whole-genome sequencing approach used in the study provided results within a few days and cost about the same as conventional genetic testing, making it more feasible for routine use, Dr. Spencer said.
“This study is important because it shows that a single test could potentially replace the many tests we currently use to assign a patient to a particular risk group” and help make decisions about treatment, said Christopher Hourigan, D.M., D.Phil., chief of the Laboratory of Myeloid Malignancies at NIH’s National Heart, Lung, and Blood Institute, who was not involved with the new study.
“In acute myeloid leukemia, the big treatment decision is whether a patient should get a stem cell transplant,” Dr. Spencer said. Stem cell transplantation is often recommended for patients whose cancer cells contain genetic changes that point to a high risk of relapse.
The new findings, which were reported March 11 in the New England Journal of Medicine, represent an important proof of principle but will need to be validated in larger studies and at other centers, Dr. Hourigan said.
Advantages of Whole-Genome Sequencing
One important type of genomic test used to help guide treatment for people with AML or MDS, known as cytogenetic analysis, involves analyzing cells from a sample of bone marrow or blood to look for changes in chromosomes, including broken, missing, rearranged, or extra chromosomes that are often found in these diseases.
Cytogenetic analysis requires a sample that contains living cells, Dr. Spencer explained. And in people with AML or MDS, that sample usually comes from a bone marrow biopsy—a procedure that is standard when diagnosing such cancers. The cells obtained from the sample are then grown in the lab, and their chromosomes are examined under a microscope.
However, Dr. Spencer said, “In up to 20% of patients the cells don’t grow, or you aren’t able to look at enough chromosomes to make a determination.” One reason this happens is because samples may have to be shipped to a company or academic medical center for testing, and the cells may not survive shipping.
By contrast, whole-genome sequencing requires only DNA and not living cells. However, it has generally been done only in research studies because of the time and cost involved.
Other genetic tests currently used to help guide treatment for people with AML or MDS include targeted sequencing of a limited set of genes in a sample of tumor cell DNA. These tests look for genetic changes that are known to occur in these cancers and are too small to see with cytogenetic analysis.
Streamlined Approach Helped Speed Up Results
For their study, the Washington University team developed a streamlined approach to sequencing the entire genome and analyzing the data from patient samples. By focusing the data analysis on the parts of the genome and types of genetic alterations already known to be important in AML and MDS, the approach “dramatically simplifies the analysis and streamlines the whole process,” Dr. Spencer said.
First, the team compared results from whole-genome sequencing with those of conventional tests for 235 patients with a known or suspected blood cancer who had previously undergone successful cytogenetic analysis. These patient samples were selected to represent a broad spectrum of cytogenetic and other genetic features of AML and MDS.
The team found that whole-genome sequencing detected all of the chromosome changes that had been detected by cytogenetic analysis and, importantly, identified additional chromosome changes in 17% of the 235 patients, some of whom had had inconclusive results with conventional testing.
Next, to assess the feasibility of using whole-genome sequencing for routine clinical testing, the team analyzed samples from 117 patients newly diagnosed with AML or MDS who were seen at their hospital. They tracked how long it took to get the results and compared them with those of conventional genetic tests.
The results of whole-genome sequencing were available in a median of 5 days and as little as 3 days—a speed that Dr. Hourigan called “impressive” and at least as fast as conventional testing. Whole-genome sequencing yielded new genetic information in 24.8% of the newly diagnosed patients and put 16.2% of those 117 patients into a new risk category.
“They were able to correctly assign more patients with AML to the adverse-risk, or highest-risk, group, which would be offered a bone marrow transplant,” and that could have implications for improving treatment outcomes, Dr. Hourigan said.
In addition, Dr. Spencer said, “in our study the technical costs of whole-genome sequencing were similar to the [commercial] costs of conventional testing. Although the actual cost for clinical testing using whole-genome sequencing may still be greater than that of conventional tests today, the time when the costs are the same will be here very shortly.”
Approach Not Yet Ready for Everyday Care
One limitation of the whole-genome sequencing approach used in the study is that it did not identify as many smaller-scale genetic changes as targeted sequencing did.
The genetic changes that were missed were present in only a small fraction of tumor cells, Dr. Spencer explained. His team showed that this limitation could be overcome by sequencing more DNA from a given sample, which would cost more but should become feasible as the cost of whole-genome sequencing continues to drop.
Another limitation of the study is that it doesn’t show whether using whole-genome sequencing for diagnostic testing will improve how long patients live, Dr. Hourigan said.
“We need a large precision medicine trial to validate this approach,” Dr. Hourigan continued.
One trial that may be able to do that is MyeloMATCH, a large NCI-sponsored precision medicine trial that will test existing and investigational treatments for AML and MDS and is due to launch later this year. Dr. Hourigan is one of the physician–scientists involved in planning that trial.
From Unthinkable to Realistic
The cost of sequencing an entire human genome has fallen from millions of dollars per genome to about $1,000 in the last 15 years, “and that cost is predicted to continue to fall, whereas standard clinical testing has been stable or increasing in cost,” Dr. Hourigan said.
In addition, he said, “The capacity for sequencing is increasing at a phenomenal rate, so that what was previously unthinkable just a decade ago is now realistic.”
In the not-too-distant future, both he and Dr. Spencer said, whole-genome sequencing could be used more routinely to better monitor patients for a recurrence of their cancer after treatment, investigate whether AML and MDS run in families, and help scientists learn about genetic factors involved in causing these cancers to develop.
Whole-genome sequencing could also be used to help diagnose, monitor, and plan treatment for other cancer types, including solid tumors, and would be especially advantageous for cancers that have multiple known genetic abnormalities, Dr. Spencer said.
The new study, he noted, builds on the past 20 years of genomics research, including the sequencing of the first complete human genome and large studies like The Cancer Genome Atlas (TCGA), funded in part by NCI, which provided information about genetic mutations in leukemia.
“The next phase of [our] research, which we will start in a month or so, is to use this approach to return results to patients and their doctors and understand how many patients actually had their treatments changed,” Dr. Spencer said.