Genomic Sequencing Helps Scientists Trace High-risk Leukemia Relapse

Apr 20 2015

Genomic Sequencing Helps Scientists Trace High-risk Leukemia Relapse

leukemiaThe ability to sequence all the genetic abnormalities in a tumor cell was unimaginable when Stephen Hunger, MD, was a pediatric oncology fellow. Twenty-five years later, he and a team of researchers from throughout the country used highly sensitive deep sequencing techniques to see how genetic changes in acute lymphoblastic leukemia (ALL) cells evolve from diagnosis to remission and relapse.

ALL, a fast-growing form of cancer that affects immature white blood cells, accounts for 30 percent of all pediatric cancers. While most patients respond well to current chemotherapy, 15 percent will relapse, and ALL remains a leading cause of pediatric cancer death. Once relapse occurs, it is much, much harder to cure patients.

In order to better understand high-risk pediatric ALL, Dr. Hunger, chief of the Division of Oncology and the director of the Center for Childhood Cancer Research at The Children’s Hospital of Philadelphia, and his co-investigators wanted to trace the founder mutations that appear in every leukemia cell and see how other mutations persist, expand, or are eradicated between the time of initial diagnosis and relapse. They focused on 20 trios of samples from patients: ALL cells from the time of initial diagnosis, blood samples obtained while patients were in remission, and leukemia samples obtained at the time of relapse.

The researchers performed whole exome sequencing, which identifies all the coding regions of genes, and they also performed whole genome sequencing, which identifies the noncoding regions as well. They also performed “deep sequencing” to confirm the presence of mutations and determine their frequency in samples. This sophisticated analysis gave the researchers high coverage of every piece of the leukemia cells’ genetic material.

“In many cases, we were reading the same sequences 500 to 1,000 times,” Dr. Hunger said. “It is very useful because it allows us to get exact fractions of how often specific changes are present in the whole population of leukemia cells.”

The study’s results published in Nature Communications will help scientists to decipher the process of how ALL and other cancers relapse. These insights are only “the tip of the iceberg” of what is to come from the Therapeutically Applicable Research to Generate Effective Therapies (TARGET) study, said Dr. Hunger, who is the principal investigator of the TARGET ALL Project. TARGET is collaborative effort of a large, diverse consortium of investigators devoted to determining the genetic changes that drive the initiation and progression of hard-to-treat childhood cancers.

When the study team looked at leukemia cells from the time of initial diagnosis, they recognized the founder cells that all have specific abnormalities that kick-start the ALL disease process. The researchers also identified two to five subclones, which are descendants of the founder clone that have additional mutations present. Typically, one subclone was dominant in 90 to 95 percent of cells.

“One of the most striking things we found is that in almost every case, the clone that was most dominant at diagnosis was not present at relapse,” Dr. Hunger said, which suggests that those cells were effectively extinguished with current chemotherapy.

However, some rare subclones were able to survive, populate, and create the relapse. At the time of initial diagnosis, these subclones were often detected in only 1 or 2 percent of cells.

“One of the key points is that in every case, it was clearly the same leukemia that persisted because these founder mutations were present both at diagnosis and relapse,” Dr. Hunger said. “So it’s not like you developed a new leukemia. It also tells us that we have to look at the rare changes that are present at diagnosis because those are the cells that are likely to come back and acquire more changes.”

When the researchers examined blood samples taken at the end of the first month of treatment, in some cases they identified a low frequency of mutations that also appeared at the relapse stage. These findings raise the question of whether it would be valuable for scientists to perform highly sensitive genomic analysis following the first few weeks of chemotherapy to provide early detection of mutations that might drive relapse.

To help shed light on this answer, the study team already is performing deep sequencing of hundreds of pediatric ALL cases that relapsed and comparing them to those that never relapsed. They will search for any differences in the genetic basis of the leukemia cells from the two groups.

“Those sorts of studies will help us learn whether there are particular mutations that help to predict who will and won’t be cured,” Dr. Hunger said.

Another interesting finding from the current study, Dr. Hunger pointed out, is that certain mutations emerged at relapse and caused resistance to common chemotherapy drugs, yet those mutations were never found at any level when the patient was first diagnosed.

“Could we test patients during treatment to see if those mutations are starting to be detected, and if they were, could we change treatment to use different drugs that these mutations don’t cause resistance to?” Dr. Hunger suggested as a future research question.

Pediatric ALL was the first disease to be piloted for the TARGET initiative, which launched in 2006. The project expanded to include research efforts focused on acute myeloid leukemia, neuroblastoma, osteosarcoma, and Wilms’ Tumor. The TARGET ALL team includes investigators from the Children’s Oncology Group , the National Cancer Institute, University of New Mexico Cancer Center, and St. Jude Children's Research Hospital.