Researchers believe they have identified cells that are responsible for relapse of acute myeloid leukemia (AML).
These “leukemic-regenerating cells” (LRCs), which are distinct from leukemic stem cells (LSCs), seem to arise in response to chemotherapy.
Experiments in mouse models of AML suggested that targeting LRCs could reduce the risk of relapse, and analyses of AML patient samples suggested LRCs might be used to predict relapse.
Allison Boyd, PhD, of McMaster University in Hamilton, Ont., and her colleagues reported these findings in Cancer Cell.
The researchers evaluated the leukemic populations that persist after chemotherapy by analyzing AML patient samples and xenograft AML models. The team found that LSCs were depleted by chemotherapy, and a different cell population, LRCs, appeared to arise in response to treatment.
LRCs are “molecularly distinct from therapy-naive LSCs,” the researchers said. In fact, the team identified 19 genes that are preferentially expressed by LRCs and could be treated with drugs.
One of these genes is DRD2, and the researchers found they could target LRCs using a small-molecule antagonist of DRD2.
Targeting LRCs
Dr. Boyd and her colleagues compared the effects of treatment with a DRD2 antagonist in AML xenografts populated with therapy-naive LSCs and AML xenografts that harbored LRCs following exposure to cytarabine.
The researchers said DRD2 antagonist therapy “moderately” affected AML progenitors in the LSC model but “had profound effects on regenerating LRCs.”
Treatment with the DRD2 antagonist also improved the efficacy of chemotherapy.
In xenografts derived from one AML patient, treatment with cytarabine alone left 50% of mice with residual disease. However, the addition of the DRD2 antagonist enabled 100% of the mice to achieve disease-free status.
In xenografts derived from a patient with more aggressive AML, all recipient mice had residual disease after receiving cytarabine. Treatment with the DRD2 antagonist slowed leukemic regrowth and nearly doubled the time to relapse.
Targeting LRCs also reduced disease regeneration potential in samples from other AML patients.
“This is a major clinical opportunity because this type of leukemia is very diverse and responds differently across patients,” Dr. Boyd said. “It has been a challenge in a clinical setting to find a commonality for therapeutic targeting across the wide array of patients, and these regenerative cells provide that similarity.”
Predicting relapse
Dr. Boyd and her colleagues also analyzed bone marrow samples collected from AML patients approximately 3 weeks after they completed standard induction chemotherapy.
The team found that progenitor activity was enriched among residual leukemic cells. However, patient cells lacked gene expression signatures related to therapy-naive LSCs.
“Instead, these highly regenerative AML cells preferentially expressed our LRC signature,” the researchers said.
The team also found evidence to suggest that LRC molecular profiles arise temporarily after chemotherapy. The LRC signature was not observed at diagnosis or once AML was reestablished at relapse.
“We think there are opportunities here because now we have a window where we can kick the cancer while it’s down,” Dr. Boyd said.
She and her colleagues also found the LRC signature might be useful for predicting relapse in AML patients.
The team assessed expression of SLC2A2, an LRC marker that has overlapping expression with DRD2, in seven patients who were in remission after induction.
Chemotherapy increased expression of SLC2A2 only in the four patients who had residual disease – not in the three patients who remained in disease-free remission for at least 5 years. “These results suggest that LRC populations represent reservoirs of residual disease, and LRC marker expression levels can be linked to clinical outcomes of AML relapse,” the researchers said.
This study was supported by the Canadian Cancer Society, the Canadian Institutes of Health Research, the Ontario Institute for Cancer Research, and other organizations.
jensmith@mdedge.com
SOURCE: Boyd AL et al. Cancer Cell. 2018 Sep 10. doi: 10.1016/j.ccell.2018.08.007.