NEW YORK (GenomeWeb News) – Two new studies appearing in today’s issue of the New England Journal of Medicine are uncovering genetic variants and miRNA signatures that may eventually help diagnose and guide treatment for a type of adult leukemia.
Acute myeloid leukemia or AML is a cancer of white-blood cell, red-blood cell, or platelet-making cell precursors. It typically affects older adults, starting in the bone marrow and often spreading to the blood and other parts of the body. As cancerous cells proliferate and replace other blood cells, AML leads to symptoms such as infections, fatigue, dizziness, bruising, and paleness.
The underlying genetics of AML can vary tremendously, though. Some forms of AML are characterized by chromosomal changes that are detectable under a microscope. These are considered high risk and are generally treated most aggressively. But nearly half of individuals with AML appear cytogenetically normal. For these individuals, diagnosis and treatment decisions are more difficult.
Although cytogenetically normal patients are usually classified as intermediate risk, recent work has shown that certain genetic mutations and gene expression profiles are associated with better or worse outcomes in these individuals. This, in turn, can influence treatment options, since aggressive treatments such as allogeneic stem-cell transplantation — transplants involving donor bone marrow — also carry risks.
In the first of two AML papers published in NEJM today, lead author Richard Schlenk, a physician at the University Hospital of Ulm in Germany, and his colleagues looked at the frequency of mutations in five genes in hundreds of patients with cytogenetically normal AML. Their goal: to determine how different genetic variants influence post-remission treatment response.
Specifically, they focused on the nucleophosmin gene NPM1, the fms-related tyrosine kinase gene FLT3, the CCAAT/enhancer binding protein alpha gene CEBPA, the mixed-lineage leukemia gene MLL, and the neuroblastoma RAS viral oncogene homolog gene NRAS.
The genes have all been linked to AML, but some mutations in these genes are negative and others are positive. For instance, internal tandem repeats in FLT3, termed FLT3-ITD, are generally associated with poor outcomes, as are partial tandem repeats in MLL. On the other hand, mutations in CEBPA and NPM1 are associated with better outcomes.
To look at these genes in relation to each other and to treatment outcomes, Schlenk and his team did genetic testing on 872 cytogenetically normal AML patients during four different multi-center prospective trials carried out by the German-Austrian Acute Myeloid Leukemia Study Group over about 11 years.
The individuals, aged 16 to 60 years old, got the same drug therapy in the first treatment stage. Then, if an appropriate donor was available, they received donated bone marrow transplants. If not, they got either chemotherapy or were transplanted with their own bone marrow.
Mutations in two genes — CEBPA and NPM1 — were associated with complete remission after therapy, though the positive NPM1 effect was restricted to those who did not also have the FLT3-ITD mutation.
In general, those with the beneficial NPM1 mutation (and lacking the FLT3-ITD mutation) did just as well with chemotherapy or transplantation with their own stem cells as they did with treatments of donated bone marrow. On the other hand, those who did not have the beneficial NPM1 genotype generally had much better outcomes when they received donated bone marrow than when they didn’t. Younger age was also associated with better outcomes.
“The benefit of the transplant was limited to the subgroup of patients with the prognostically adverse genotype FLT3-ITD or the genotype consisting pf wild-type NPM1 and CEBPA without FLT3-ITD,” Schlenk and his colleagues wrote. “We recommend that screening for NPM1, FLT3, and CEBPA mutations be part of the initial workup for newly diagnosed AML.”
In a second paper, also appearing online in NEJM today, researchers at Ohio State University and elsewhere focused on those individuals who had high-risk genes — FLT3-ITD, wild type NPM1, or both, evaluating the changes in microRNA expression associated with better AML outcomes.
Lead author Guido Marcucci, a hematologist and oncologist at the Ohio State University’s Comprehensive Cancer Center, and his team profiled miRNA expression in these high-risk, cytogenetically normal AML patients and found a dozen miRNAs made up an miRNA signature linked to relapse-free survival.
For that study, done within a national clinical cooperative group called Cancer and Leukemia Group B, researchers analyzed the miRNA expression for 64 cytogenetically normal AML patients under 60 years old. From the 305 candidate miRNAs detected in the initial phase of the study, the team discovered an miRNA signature consisting of 12 miRNAs.
Increased expression of five miRNAs, belonging to the miRNA-181 family, was linked to event-free survival in cytogenetically normal AML patients. On the other hand, increased expression of the remaining miRNAs was associated with greater AML recurrence and poorer outcomes. From these results, the team developed an miRNA summary value and validated its predictive power in 55 AML patients who had FLT3-ITD, wild type NPM1, or combinations of the two.
When they analyzed miRNA summary value in combination with gene expression microarray data, Marcucci and his team found that the miRNAs tended to influence the expression of genes involved in inflammation and in the innate immune system. This suggests there is, indeed, a functional relationship between the miRNA signature, gene expression, and AML outcomes in the AML patient group tested.
In an accompanying editorial in the same issue of NEJM, Bob Lowenberg, a hematologist at Erasmus University Medical Center in the Netherlands, expressed enthusiasm about the idea of finding markers to guide AML therapy, particularly in terms of allogeneic stem-cell transplantation. Still, he noted, more research is necessary before the genetic markers found in the two studies can be applied to AML prognosis and treatment.
“With the intense ongoing scientific activity in many centers, there is an emerging list of relevant genetic abnormalities that warrant consideration,” he wrote. “We will probably not have a single, universally accepted system for stratification of AML prognosis for some time.”