EHA 2015

Biomarkers for childhood relapsed acute lymphoblastic leukaemia

Prof Anthony Moorman - Newcastle University, Newcastle, UK

Now, in childhood relapsed acute lymphoblastic leukaemia what are the compelling reasons for looking at gene biomarkers to look at things like risk prediction?

The compelling reason is that it’s not a fatal disease any more so relapsed ALL can be cured. If you look at all patients with relapsed childhood ALL around 50% of them can still be cured of their disease. The trick is how do we identify those patients that can be cured with chemotherapy after relapse versus those patients that need to go on to a more intensive protocol such as a stem cell transplantation or those that need a completely new type of therapy? What we’ve done is we’ve incorporated clinical risk factors with genetic risk factors to come up with an index that can divide patients into three groups that can be treated accordingly.

You’ve come up with an index, I’d like to know what it is but also how you’ve applied it and vindicated it.

What’s in the index is the clinical risk categories which is the time to first relapse and also the site of relapse. What we've added in are chromosomal abnormalities, these are acquired chromosomal abnormalities, and also gene mutations in four key genes. We have validated it in terms of within a single protocol but it hasn’t been validated yet in an independent study and we’re currently doing that in collaboration with the German study group.

Acute lymphoblastic leukaemia falls into a number of categories, does this work for all of them or are the tests different depending on whether it’s T or B-cells?

There will be differences, there are differences between B and T. The one we’ve applied it to at the moment is B-cell precursor ALL but we are currently working on T ALL. But different genes are involved in T ALL so we’ll have to compose a separate index for that particular disease.

So could you name just one or two of the genes to give me a flavour of this and explain how they mix in with the clinical factors to give you what you believe is a reliable score?

The key genes that we’ve been looking at are TP53, which of course is mutated in many of these cancers. TP53 is not mutated very often at initial diagnosis of ALL but its incidence of mutation at relapse is very high. That strongly predicts an adverse outcome after first relapse. Another one is Ras mutations. These are very common again in plenty of cancers but importantly they can be targeted. So there are plenty of drugs at the moment called MEK inhibitors which can target patients with Ras mutations. So identifying those as a risk factor gives an alternative therapeutic option.

Now, with the study you did, were you able to tell with your new score how much more carefully you could distinguish patients falling into one treatment group as compared with those who require another treatment?

Yes, so what we did is we moved away from just having a good risk and a poor risk subgroup to having a good, intermediate and very poor risk subgroup. So the very poor risk subgroup have an overall survival after first relapse of less than 20% whereas in the good risk group it’s around 70-75% so it’s a real advantage.

What are the clinical applications of those new scores?

The clinical applications can be that the patients that we know are going to have a good response post-relapse chemotherapy can just receive chemotherapy, we don’t need to consider them necessarily for a stem cell transplantation and they don’t need to be thought of in terms of new therapeutic targets. But equally the patients that do very poorly, irrespective of what kind of treatment you can give them, they can be offered more targeted chemotherapies that are being produced and are on the horizon at the moment.

So what message would you give cancer doctors to remember, briefly, coming out of this?

The key thing to do is whenever you have a patient with acute lymphoblastic leukaemia you need to ensure that they get their genetic work-up done at diagnosis and also at relapse. And wherever possible treat them according to a… on a clinical trial so that we can assess the full implications of everything that we’re finding in the genetic arena.