Molecular signatures of acute myeloid leukemia vary considerably between younger, older patients

Soheil Meshinchi

The genetic signature of acute myeloid leukemia among pediatric and young adult patients appeared distinctly different from that observed among adults, according to study results.

Soheil Meshinchi, MD, PhD, a pediatric AML specialist at Fred Hutchinson Cancer Research Center, and colleagues assessed genetic information for nearly 1,000 patients from Children’s Oncology Group AML clinical trials.

“The design of the study was to do the most comprehensive genetic profiling by conducting whole-genome and whole-transcriptome — mRNA and miRNA — sequencing in as large a cohort as possible,” Meshinchi told HemOnc Today. “We also performed methylation profiling to determine the epigenetic signature for leukemia in younger patients.”

Results showed fewer than 40 genes were mutated in more than 2% of cases in the pediatric cohort. However, pediatric and young adult patients showed novel fusions and focal deletions in MBNL1, ZEB2 and ELF1, whereas adults did not demonstrate these features.

Pediatric cases also were marked by mutations in GATA2, FLT3, CBL, MYC-ITD, NRAS, KRAS and WT1. However, they did not demonstrate mutations in DNMT3A and TP53, which are commonly among adult patients.

HemOnc Today spoke with Meshinchi about the differences in genetic signatures based on age, whether these findings suggest the ‘trickle-down’ approach to pediatric AML is over, and the potential that a ‘trickle-up’ approach may be more effective.

 

Question: How did this study come about?

Answer: AML is considered a disease of older adults. Around 80% of cases occur among individuals aged older than 60 years. Pediatric and young adult patients comprise a small proportion of those with AML. It has been assumed that AML is the same disease whether it’s in older or younger patients, and that you could take the data discovered or treatments developed for the elderly and use them for younger patients.

In the last 20 years, we have found that younger patients have an entirely different genetic makeup than older patients. We had a hint that the disease among the young and the old was not the same. The fact two mutations — IDH1 and DNMT3A — were discovered to be highly prevalent in AML among older patients but were extremely rare or not observed among those aged younger than 40 years provided momentum to approach the NCI for funding for definitive profiling of AML in children and young adults.

We started to expand our interest in defining the true genetic makeup of AML based on age. In evaluation of the incidence of AML across age group, it becomes clear that AML is quite rare until about sixth decade of life, when there is a sharp rise in AML incidence. This might link evolution of AML among older adults to other events, such as myelodysplastic syndrome, requiring more comprehensive genomic evaluation of younger patients with de novo AML.

We evaluated the genetic makeup of AML in younger patients and contrasted the results with those of The Cancer Genome Atlas study performed in older patients. We found a remarkable difference in the makeup of AML by age.

The most dramatic difference is that, among older adults, AML is defined by mutations in DNA sequence. For example, a base pair change from a C to a G. However, among younger patients, genetic changes are much larger structural alterations that lead to generation of new gene fusions and novel cancer-causing proteins. These structural alterations are highly prevalent among younger patients. For example, virtually 100% of infants with AML have highly penetrant, catastrophic structural events, or translocations that are sufficient for malignant transformation. Such structural events are far less common among older adults.

 

Q: Historically, a “trickle - down” approach to AML has been used, where findings in studies of adults have been applied to younger populations. Do your results contradict that approach?

A: The way AML has been studied is to go to where it is most common. Most studies are done in the older patient population. Treatments and therapies have been developed in that cohort with the hope that discoveries for patients aged 60 years or 70 years simply would be incorporated into the care of younger patients. Even now, treatments are being developed for mutations found in adults. The problem is that mutations discovered and targeted in older patients simply do not occur in younger patients, so such adult-centric targeted therapies would be useless for those aged younger than 40 years or 50 years. We need to design treatments based on appropriate mutations and genetic variants. That is why our database is so valuable. We can develop treatments that target age-appropriate mutations, structural variants and translocations.

Q: Given the era of limited resources and research dollars, is it feasible or realistic to develop these types of treatments?

A: That is a good question. Until now, all efforts have been put toward developing therapies for AML in the elderly, virtually ignoring any meaningful efforts to study or develop therapeutics for AML among children and young adults. From a social and economic standpoint, life years gained by curing AML in an 80- or 90-year-old would be modest. However, curing a child or a young adult with AML would have a more meaningful impact, as such young AML survivors may have 60 or more productive years remaining. The challenge of developing therapies for younger patients is that the mutations that are used for therapeutic targeting are less common. However, other targets that can be exploited have been virtually ignored. Studying the transcriptome has identified a large number of abnormally expressed genes and proteins that should be evaluated as potential targets for therapeutic development. One example is a protein called mesothelin, which we discovered in our study. It is normally expressed in lung cancer, but we saw it expressed at high levels in a large subset of younger patients with AML. This protein can be easily targeted by an available drug called anetumab ravtansine (BAY 94-9343, Bayer) developed for lung cancer. This drug is being considered for use in pediatric AML. There are a number of abnormally expressed proteins like this that can be similarly targeted.

 

Q: You make a strong argument for treating younger patients, and that it is possible to find therapies for these targets. But is that happening now?

A: It is not happening. But at some point, one has to think that — if we are going to treat AML effectively — we have to identify biomarkers and develop therapies that ‘trickle up’ from younger patients to older patients. We have to find targets that are expressed in younger and older patients. Right now, we are looking underneath the lamppost because that is where the light is. Perhaps we should have a better strategy for identifying fundamental pathways that can be useful for older and younger patients. The other question is whether AML in older patients is the same disease as in younger patients. Just because they look the same under the microscope does not make it the same disease. There is evidence that a majority of cases of AML among the elderly may have arisen from myelodysplastic syndrome, which may have an entirely different genetic makeup and clinical response profile than de novo AML seen in younger patients.

 

Q: With the ‘trickle up’ concept, you are suggesting a paradigm shift. Is there resistance to this idea, or are other clinicians and researchers open to it?

A: I think people are very open to the idea. Frankly, we have not been able to do anything new in AML in 40 years. There have been no new drugs in decades, and we are still using the same therapy developed in 1970s. Frustrations with current approaches is reaching a maximum. Not only is there no resistance, there is a great sense of hopefulness that some of these novel targets actually can move the field forward. We hope this leads to fundamental changes in the way people think about AML. Going after protein products of some of these altered genes is something we have not been able to do in the past.

 

Q: What are the next steps for research?

A: The next step is to expand our studies to maximize our understanding of the disease. We are starting the next phase of our discovery efforts by sequencing DNA and RNA from leukemic cells of 1,200 children and young adults to identify targets toward developing targeted therapies. We will look to identify immunotherapeutic targets, as well as targets of small molecule inhibitors. – by Rob Volansky

 

Reference:

Bolouri H, et al. Nature Medicine. 2018;doi:10.1038/nm.4439.

 

For more information:

Soheil Meshinchi, MD, PhD, can be reached at 1100 Fairview Ave. N., Seattle, WA 98109; email: smeshinc@fredhutch.org.

 

Disclosure: Meshinchi reports no relevant financial disclosures.

Soheil Meshinchi

The genetic signature of acute myeloid leukemia among pediatric and young adult patients appeared distinctly different from that observed among adults, according to study results.

Soheil Meshinchi, MD, PhD, a pediatric AML specialist at Fred Hutchinson Cancer Research Center, and colleagues assessed genetic information for nearly 1,000 patients from Children’s Oncology Group AML clinical trials.

“The design of the study was to do the most comprehensive genetic profiling by conducting whole-genome and whole-transcriptome — mRNA and miRNA — sequencing in as large a cohort as possible,” Meshinchi told HemOnc Today. “We also performed methylation profiling to determine the epigenetic signature for leukemia in younger patients.”

Results showed fewer than 40 genes were mutated in more than 2% of cases in the pediatric cohort. However, pediatric and young adult patients showed novel fusions and focal deletions in MBNL1, ZEB2 and ELF1, whereas adults did not demonstrate these features.

Pediatric cases also were marked by mutations in GATA2, FLT3, CBL, MYC-ITD, NRAS, KRAS and WT1. However, they did not demonstrate mutations in DNMT3A and TP53, which are commonly among adult patients.

HemOnc Today spoke with Meshinchi about the differences in genetic signatures based on age, whether these findings suggest the ‘trickle-down’ approach to pediatric AML is over, and the potential that a ‘trickle-up’ approach may be more effective.

 

Question: How did this study come about?

Answer: AML is considered a disease of older adults. Around 80% of cases occur among individuals aged older than 60 years. Pediatric and young adult patients comprise a small proportion of those with AML. It has been assumed that AML is the same disease whether it’s in older or younger patients, and that you could take the data discovered or treatments developed for the elderly and use them for younger patients.

In the last 20 years, we have found that younger patients have an entirely different genetic makeup than older patients. We had a hint that the disease among the young and the old was not the same. The fact two mutations — IDH1 and DNMT3A — were discovered to be highly prevalent in AML among older patients but were extremely rare or not observed among those aged younger than 40 years provided momentum to approach the NCI for funding for definitive profiling of AML in children and young adults.

We started to expand our interest in defining the true genetic makeup of AML based on age. In evaluation of the incidence of AML across age group, it becomes clear that AML is quite rare until about sixth decade of life, when there is a sharp rise in AML incidence. This might link evolution of AML among older adults to other events, such as myelodysplastic syndrome, requiring more comprehensive genomic evaluation of younger patients with de novo AML.

We evaluated the genetic makeup of AML in younger patients and contrasted the results with those of The Cancer Genome Atlas study performed in older patients. We found a remarkable difference in the makeup of AML by age.

The most dramatic difference is that, among older adults, AML is defined by mutations in DNA sequence. For example, a base pair change from a C to a G. However, among younger patients, genetic changes are much larger structural alterations that lead to generation of new gene fusions and novel cancer-causing proteins. These structural alterations are highly prevalent among younger patients. For example, virtually 100% of infants with AML have highly penetrant, catastrophic structural events, or translocations that are sufficient for malignant transformation. Such structural events are far less common among older adults.

 

Q: Historically, a “trickle - down” approach to AML has been used, where findings in studies of adults have been applied to younger populations. Do your results contradict that approach?

A: The way AML has been studied is to go to where it is most common. Most studies are done in the older patient population. Treatments and therapies have been developed in that cohort with the hope that discoveries for patients aged 60 years or 70 years simply would be incorporated into the care of younger patients. Even now, treatments are being developed for mutations found in adults. The problem is that mutations discovered and targeted in older patients simply do not occur in younger patients, so such adult-centric targeted therapies would be useless for those aged younger than 40 years or 50 years. We need to design treatments based on appropriate mutations and genetic variants. That is why our database is so valuable. We can develop treatments that target age-appropriate mutations, structural variants and translocations.

 

PAGE BREAK

Q: Given the era of limited resources and research dollars, is it feasible or realistic to develop these types of treatments?

A: That is a good question. Until now, all efforts have been put toward developing therapies for AML in the elderly, virtually ignoring any meaningful efforts to study or develop therapeutics for AML among children and young adults. From a social and economic standpoint, life years gained by curing AML in an 80- or 90-year-old would be modest. However, curing a child or a young adult with AML would have a more meaningful impact, as such young AML survivors may have 60 or more productive years remaining. The challenge of developing therapies for younger patients is that the mutations that are used for therapeutic targeting are less common. However, other targets that can be exploited have been virtually ignored. Studying the transcriptome has identified a large number of abnormally expressed genes and proteins that should be evaluated as potential targets for therapeutic development. One example is a protein called mesothelin, which we discovered in our study. It is normally expressed in lung cancer, but we saw it expressed at high levels in a large subset of younger patients with AML. This protein can be easily targeted by an available drug called anetumab ravtansine (BAY 94-9343, Bayer) developed for lung cancer. This drug is being considered for use in pediatric AML. There are a number of abnormally expressed proteins like this that can be similarly targeted.

 

Q: You make a strong argument for treating younger patients, and that it is possible to find therapies for these targets. But is that happening now?

A: It is not happening. But at some point, one has to think that — if we are going to treat AML effectively — we have to identify biomarkers and develop therapies that ‘trickle up’ from younger patients to older patients. We have to find targets that are expressed in younger and older patients. Right now, we are looking underneath the lamppost because that is where the light is. Perhaps we should have a better strategy for identifying fundamental pathways that can be useful for older and younger patients. The other question is whether AML in older patients is the same disease as in younger patients. Just because they look the same under the microscope does not make it the same disease. There is evidence that a majority of cases of AML among the elderly may have arisen from myelodysplastic syndrome, which may have an entirely different genetic makeup and clinical response profile than de novo AML seen in younger patients.

 

Q: With the ‘trickle up’ concept, you are suggesting a paradigm shift. Is there resistance to this idea, or are other clinicians and researchers open to it?

A: I think people are very open to the idea. Frankly, we have not been able to do anything new in AML in 40 years. There have been no new drugs in decades, and we are still using the same therapy developed in 1970s. Frustrations with current approaches is reaching a maximum. Not only is there no resistance, there is a great sense of hopefulness that some of these novel targets actually can move the field forward. We hope this leads to fundamental changes in the way people think about AML. Going after protein products of some of these altered genes is something we have not been able to do in the past.

 

Q: What are the next steps for research?

A: The next step is to expand our studies to maximize our understanding of the disease. We are starting the next phase of our discovery efforts by sequencing DNA and RNA from leukemic cells of 1,200 children and young adults to identify targets toward developing targeted therapies. We will look to identify immunotherapeutic targets, as well as targets of small molecule inhibitors. – by Rob Volansky

 

Reference:

Bolouri H, et al. Nature Medicine. 2018;doi:10.1038/nm.4439.

 

For more information:

Soheil Meshinchi, MD, PhD, can be reached at 1100 Fairview Ave. N., Seattle, WA 98109; email: smeshinc@fredhutch.org.

 

Disclosure: Meshinchi reports no relevant financial disclosures.