Matthew Collin (Newcastle University, Newcastle upon Tyne, UK)
Erdheim Chester Disease (ECD) is caused by mutations in the DNA of blood cells. Blood cells are formed in the bone marrow from stem cells that can live for many years producing many generations of new stem cells. The frequent questions that patients ask about ECD include: where did it come from? how long have I had it? couldn’t anything have been done sooner? why does my disease affect a particular site? Our research aims to shed new light on these important questions. The technique we will use is called ‘phylogenetic mapping’. This approach allows us to go back in time and ‘date-stamp’ the origin of mutations that cause ECD, to within a few years of a patient’s past life. The way that this works is by growing many clones of single stem cells in the laboratory and sequencing the whole genome of each clone. Each clone differs from the next by a few mutations in its DNA. Some of these mutations arose a very long time ago in the ancestors of the stem cell when the patient was younger. By sequencing about one hundred clones it is possible to reconstruct the life history of the stem cells within a person and so create a timeline of mutations as they appear over the years. Among these mutations will be the mutation that caused ECD in the patient. If we know the timeline of all of the mutations, we can ‘datestamp’ the ECD mutation. We can then estimate how long the ECD mutation lay dormant in the body, how quickly it grew to a size that could cause disease, and whether it was assisted by any other mutations on the way. These are fundamental issues. In other related diseases called myeloproliferative neoplasms, mutations arise in childhood and evolve into different types of disease depending on other events in the life of a patient, over decades. When we apply this analysis to ECD, we should be able to answer the questions about its origin. The potential benefits also include the possibility to detect ECD at an early stage before it has evolved to cause disease; to determine why there is a spectrum of affected organs in different patients; and, why some patients have higher risk disease that progresses more quickly. Finally, unraveling the ‘personal life history’ of ECD may help us in the future with personalized therapy for better outcomes.
Amount: 200,000 USD in partnership with the Leukemia & Lymphoma Society
Interim Report
Lay summary of the research
Erdheim Chester Disease (ECD) is caused by mutations in the DNA of blood cells. Blood cells are formed in the bone marrow from stem cells that can live for many years producing many generations of new stem cells. The frequent questions that patients ask about ECD include: where did it come from? how long have I had it? couldn’t anything have been done sooner? why does my disease affect a particular site? Our research aims to shed new light on these important questions. The technique we will use is called ‘phylogenetic mapping’. This approach allows us to go back in time and ‘date-stamp’ the origin of mutations that cause ECD, to within a few years of a patient’s past life. The way that this works is by growing many clones of single stem cells in the laboratory and sequencing the whole genome of each clone. Each clone differs from the next by a few mutations in its DNA. Some of these mutations arose a very long time ago in the ancestors of the stem cell, when the patient was younger. By sequencing about one hundred clones it is possible to reconstruct the life history of the stem cells within a person and so create a time line of mutations as they appear over the years. This is like drawing a family tree of how all of the cells are related to one another and is known as the ‘phylogeny’. Among these mutations will be the mutation that caused ECD in the patient. If we know the timeline of all of the mutations, we can ‘date-stamp’ the ECD mutation. We can then estimate how long the ECD mutation lay dormant in the body, how quickly it grew to a size that could cause disease and whether it was assisted by any other mutations on the way. These are fundamental issues. In other related diseases called myeloproliferative neoplasms, mutations arise in childhood and evolve into different types of disease depending on other events in the life of a patient, over decades. When we apply this analysis to ECD, we should be able to answer the questions about its origin. The potential benefits also include: a possibility to detect ECD at an early stage before it has evolved to cause disease; to determine why there is a spectrum of affected organs in different patients; and, why some patients have higher risk disease that progresses more quickly. Finally, unravelling the ‘personal life history’ of ECD may help us in the future to personalized therapy for better outcomes.
Progress
We have successfully expanded clones from 5 patients, 3 with ECD, 1 with ECD/LCH cross- over and 1 with LCH. In 2 of these cases, we captured BRAFV600E mutated clones but unexpectedly, there were no BRAFV600E mutated clones in 3 patients, even though the mutation was detectable in their blood. We are further exploring where the mutation lies in these patients to confirm that it is in the stem cell population.
In the patients where we obtained expanded clones containing BRAFV600E mutated clones, we have sequenced nearly 300 whole genomes and reconstructed the family trees (phylogenies). We observed, in both patients, that at least three mutations had arisen almost simultaneously in the genes KRAS, NRAS and BRAF. Surprisingly, these appeared to be independent events as each mutation was found on a separate branch of the phylogeny (see figure). We did not expect this, because most cancers evolve, with one mutation leading to another mutation and so on with all mutations contributing to a single branch. What we have observed in ECD is a parallel process with multiple branches all evolving together. The next step is to find out if these parallel branches are all present together lesions of ECD in the tissues. If this is confirmed it will suggest that they somehow co-
operate together. This pattern is unlike that observed in any other cancer and might be a very rare event. Potentially, this might explain why ECD is so uncommon.
How far have we answered the research questions?
The questions that we set out to answer are listed below with our progress to date in italics
- At what age in life do mutations driving ECD occur?
How quickly do clones expand to cause disease?
What is the latency between mutation and disease onset?
In both patients who were over 50 years of age, it appears that mutations arose at least two to three decades earlier. These features are similar to myeloproliferative neoplasms. Theoretically from this, it should be possible to track and identify a risk of ECD earlier in life. - How does BRAFV600E interact with TET2 mutated clonal hematopoiesis?
What is the sequence of mutations?
Does TET2 mutation promote ECD through a cell-autonomous effect?
Does TET2 mutation enhance the fitness/growth rate of ECD clones?
TET2 is still under analysis in these two patients. The other genes, KRAS and NRAS are known to be associated with ECD. Surprisingly we found that each gene was operating independently of BRAF in the phylogeny – the opposite of what we expected. We know need to confirm if this is true in lesional tissue. - Are novel driver mutations found in patients lacking known mutations of clonal hematopoiesis?
We did not find any novel driver mutations yet but we are still analyzing the data. We found a very unusual pattern of the driver mutations, as described above.
Plan for the second year
We aim to solve the following questions in the next 12 months
- We need to check in which cells the BRAF mutation is found in patients whose expanded clones did not include mutated BRAF. Either the mutation is not where we expect, or there may be an issue with in vitro expansion of cells that contain mutated BRAF (point 2).
- We are also exploring a new way of generating DNA for sequencing that does not involve growing clones. We anticipate that this will help us to map mutations in BRAF-mutated cells that do not expand into clones in vitro.
- We need to confirm that the unusual parallel pattern that we have observed in the phylogenies so far, is also present in the lesions of the patients
Explanation of how the pattern in histiocytosis has never been observed in other cancers.