EHA 2016
Sequencing multiple myeloma reveals a complex landscape of genetic lesions
Dr Niccolo Bolli - University of Milan, Milan, Italy
In the latest study I will be presenting we sequenced myeloma at diagnosis from 418 patients. We have both technological and clinical messages coming out of this project. First of all, we’re trying to develop a one-stop package where we can really get to know relevant prognostic information in multiple myeloma from this one sequencing technology. So that means not only categorised gene mutations but also move on to copy number changes in chromosomes and chromosome segments as well as balanced translocations which have a clear prognostic impact in multiple myeloma. So this is a field that’s fairly novel and needs standardisation but the field is quite ripe now to start exploiting it for such clinical purposes. Clearly we do this because we want our technical and laboratory efforts to have a clinical impact and we’re now in a position where we can interrogate the vast array of prognostic markers in multiple myeloma in an unbiased and high throughput way through higher dimensional statistics. We’re trying to do that and we’re trying to deliver a message to the clinicians that may improve how we treat multiple myeloma in the future.
Could you tell us more about the mutations you’ve been working with in detail?
We found a number of mutations, we included in our targeted panel 246 genes which are either myeloma genes or cancer genes in general. Our effort was a mixed confirmation and discovery effort where we were trying to establish the prevalence of known myeloma genes as well as finding novel ones. So we find that aside from the known myeloma genes there are few that come out as striking in the extended gene panel we applied. Despite we sequenced 418 patients, which is not a small number for sure, I think we need to increase that size by at least a log before we can clearly identify novel genes which are significantly recurrent in multiple myeloma and even then the clinical significance of those will be probably not much given we would find those mutated at a probably 2-3% recurrence rate. So the main players gene mutation wise are for sure genes feeding to the phosphor-ERK pathway, or the MAP kinase pathway if you want, so those are KRAS, NRAS, BRAF, those are certainly the three most commonly mutated genes. Aside from those we have p53 which is a common player in many haematological and solid cancers and we have other genes which are more specific to multiple myeloma such as FAM46C and DIS3 which are genes involved in RNA processing. Aside from that we have a less frequently mutated set of mutations feeding to the NF-kappa B pathway such as TRAF3, CYLD, LTB, which are also specific to multiple myeloma.
From all the sampling of them did you find any concordance between them?
We know that, for example, mutations feeding to the MAP kinase pathway tend to be mutually exclusive which means that if you have one mutated gene feeding to that pathway you have less selective pressure to mutate another one because you already have achieved the proliferative effect you wanted for that cell to proliferate more. We’re also starting to find a set of genes which are more commonly associated with the hyperdiploid myeloma subgroup which is mostly, I would say, FAM46C. On the other side we have mutations such as those of DIS3 or IRF4 or ZFHX4 which are more commonly associated with IGH translocated subtypes of multiple myeloma.
You mentioned some of this earlier but do you think any of these have an immediate prognostic value?
Interestingly of the many genes we studied we could only find p53 mutations as having a clear prognostic impact for both progression free and overall survival which survives statistical cut-offs such as p-value and false discovery rate cut-offs. So the common message is that cytogenetic alterations are by far leading the pool of prognostic markers in multiple myeloma whereas gene mutations may have a role in subgroup analysis or in specific subsets which we are trying to investigate.
