For the first time, oncology researchers using genomics techniques have identified genetic mutations directly linked to early relapse in children with acute lymphoblastic leukemia (ALL), and have shown that mutations in a particular gene confer resistance to the most commonly used ALL treatment and maintenance therapies.

Most of the 20 percent of childhood ALL patients who relapse after treatment and maintenance under current protocols have a poor prognosis, and those who relapse early are least likely to survive.

In two simultaneously published papers, two independent groups using different sequencing approaches and targeting different cell types found a number of mutations associated with relapse. Both studies zeroed in further on mutations in the gene NT5C2 as a culprit in resistance to specific nucleoside analog chemotherapies that are currently the “backbone” of treatment and maintenance therapy of ALL: 6-mercaptopurine (6-MP) and 6-thioguanine.

NT5C2 codes for an enzyme that metabolizes nucleoside analogs. After finding the association, the researchers showed in vitro that populations of cells with this mutation were resistant to 6-MP and 6-thioguanine, but that some structurally different nucleoside analogs, including nelarabine and cytarabine, are apparently not affected.

This, the researchers say, suggests immediate implications for patients found to harbor these mutations.

NYU Cancer Institute’s William L. Carroll, whose team published its findings in Nature Genetics, as well as Adolfo Ferrando's team at Columbia University, which published in Nature Medicine, both accessed “trios” of samples from the same patients at diagnosis, remission, and relapse. Previous studies, they say, have had the disadvantage of comparing samples from different patients.

Carroll says his lab specializes in “the most common form of ALL, called B-precursor or B-ALL, [which accounts for] 85 percent of ALL,” and investigated 10 cases of early relapse using RNA sequencing—“looking for things that would change the protein structure.” Ferrando’s group studies T-cell ALL—responsible for 15-percent of pediatric ALL—and followed five boys who relapsed early using high-throughput parallel exome sequencing.

Carroll and Ferrando say their studies validate each other, and both propose that tiny clones harboring NT5C2 mutations may be present at diagnosis, and are then selected for by the medications themselves.

Says Ferrando, “Basically, this study has highlighted that there is heterogeneity in the leukemic clones, and that chemotherapy imposes a Darwinian pressure that selects for clones that are resistant to the activity of these drugs.” Those clones are the ones that result in relapse, he says.

“Relapsed ALL remains one of the leading causes of death due to cancer in childhood. So, finding a smoking gun like this can really help us design these therapies,” Carroll says.

“If we could detect this clone before frank relapse occurs, before the patient presents with the low blood counts and when you can see the leukemic cells by microscopy, then maybe we will switch treatment into combinations that no longer rely on the 6-MP,” says Carroll. “So, we want to see if this is a biomarker we can use to detect impending relapse and thereby switch therapies before the patients develop bona fide relapse.”

Ferrando says the discovery also highlights the importance of compliance with ALL maintenance therapy. He points out that a recent study found that, “If you are below 95 percent compliance in the dose of 6-MP that you take, if you skip just a few doses here and there, you actually have a higher risk of relapse.”

Mount Sinai Hospital pediatric ALL oncologist Birte Wistinghausen, who was not involved in this research, finds the papers “both nicely done and exciting.”

“Both papers validated these genome sequencing techniques,” she says. “It shows that it’s not just ‘passenger’ mutations we can pick up—there are somatic mutations we didn’t pick up before, that we could target to prevent relapse.”

Secondly, says Wistinghausen, “Both found a mutation we didn’t know about that explains why some patients relapse early, and gives the hope that one day, maybe we will be able to do deep sequencing and detect this mutation at diagnosis so we can modify those patients’ therapy.”

She adds that this will take not only a great deal of research, including incorporating assays for NT5C2 mutations into future clinical trials, but also, investigating mutations that “may account for the other 80 percent of patients who relapse.”

Both Carroll and Ferrando say they are moving in that direction by extending their studies to other genes, as well as to adult ALL patients, whose prognosis at relapse is far poorer than children’s, at about 50 percent.

“I think for those of us who treat leukemia, it is a really dramatic finding,” Carroll says. “And there will be others. With the access to both the samples and, now, the new next-generation sequencing, it’s going to provide an advantage to us to really understand the blueprint of relapse and to develop specific therapies. As opposed to just throwing all chemotherapies at it, we can individualize treatment.”

Joyce Gramza is an award-winning journalist based in Central New York. A graduate of New York University's Science, Health & Environmental Reporting Program, she has 20 years' experience in local and national print, Web, radio/podcast and TV/video. Her beat specialties are genetics & genomics, and neuroscience & mental health. She is a die-hard dog lover.