There is really no uniform definition of so-called high-risk acute lymphoblastic leukaemia. So one cannot define high-risk leukaemia just based on immuno-phenotype or genotype or, for that matter, based on the early treatment response as measured by minimal residual disease. So we found that you really need to take into account multiple factors including presenting clinical features, like age, presenting white blood cell count; you need to take account of the immuno-phenotype, whether T-cell or early T-cell precursor ALL or B-ALL. On top of that you need to take into account early treatment response as measured by minimal residual disease. This is still not enough, you really have to account for the available treatment because some very high-risk leukaemias in the past have now become highly curable if you have the right treatment. So therefore to define high-risk leukaemia we have to combine all these factors.
On top of that we need to account for the germline genotype. The genes that we inherited can affect not only the development of leukaemia in the first place and also will determine whether patients are more likely to develop drug resistance or develop long-term side effects. Therefore, to define high-risk leukaemia we need to combine all these factors together. These factors play a different role in different subtypes of leukaemia so it is a complicated matter. Therefore, in my talk I try to give some examples to the audience that we have to combine all these factors to define the high-risk leukaemias which have important prognostic and therapeutic implications.
What evidence is there from recent studies that supports ongoing research in this area?
There are many studies showing there are many… genotypes are actually very heterogeneous. For example, Philadelphia chromosome-like ALL is a totally heterogeneous disease so you really need to know the cooperating mutations and then you need to know the available treatment, you need to know how they respond to treatment. So you need to take all these factors to decide how to treat the individual patients.
How do you see this progressing in the near future?
In the next five years I’m sure everybody will use clinical features plus genotype and minimal residual disease as measured by flow cytometry or PCR and then to use it for the rent of the therapy. On top of that I hope people start measuring the germline variants or the polymorphisms and use this information to further refine the treatment.
Is there anything you would like to add?
First of all, to define genotype we should not just use conventional chromosomal analysis, we really need to do RNA sequencing, not only to identify driver mutations and also important co-operative mutations, to precisely identify the genotypes. Also everybody must have minimal residual disease measurement because just by doing that you will automatically improve the outcome of the overall patients. Finally, we need to know the germline of our patients in order to use the optimal treatment for the patients.