Predicting the relapse of acute myeloid leukaemia with genetic screening

Share :
Published: 2 Dec 2014
Views: 2557
Rating:
Save
Dr Elaine Mardis - University of Washington, Washington, USA

Dr Elaine Mardis speaks to ecancertv at 26th EORTC-NCI-AACR Symposium about the use of bone marrow banking to identify genetic variations in acute myeloid leukaemia. This screening can help doctors understand and predict which patients are at risk of relapse. 

 

The study that I presented really is a study that we are preparing to do during the next year in our AML clinics at Washington University in St Louis. It’s really focussed on acute myeloid  leukaemia which is a typically deadly disease that occurs in older adults and really builds on a lot of the discovery science that we’ve had going on in AML since about the mid-2000s; so the genomics of AML, to be clear. What we want to do is with incoming patients use a bone marrow that is typically examined by pathology based methods to diagnose their cancer and extract DNA and RNA from those tumour cells. Then do direct genomic sequencing and also RNA sequencing, that’s called another omics, transcriptomics. Really the idea is to paint a picture of the mutations and other events in the cancer cells that are driving that individual patient’s tumour.

Not so much just within that initial or diagnostic marrow but then also using the marrow that’s taken 30 days after the patient has gone through extensive chemotherapy and really comparing the day 0 to the day 30 marrow. What we found by sort of prototyping this, if you will, in testing out our ideas in already banked marrow, so patients for which they have been through the 30 day regimen, we understand whether they’ve had a relapse in their disease or whether they remain in remission, we’ve been able to show that in those patients that have relapse we can actually go to that day 30 marrow and still detect mutations that are in genes that are known drivers in the disease. We think that over time we’ll be able to shape the landscape of patients that will relapse and may relapse even more quickly than we anticipate or want. We may be able to use those genetic mutations to identify additional therapies that they might be able to have access to after their chemotherapy regimen to ensure that they stay in remission which would have to be tested, of course, but that’s the hypothesis. In testing that hypothesis we’ll need to have this foundational clinical data to really build towards a secondary study.

What are the next steps?

The current practice in AML really plays into the scenario that I laid out, mainly because patients routinely have these bone marrow biopsies as they go through their treatment regimen. So we won’t have to design anything new, if you will, or pay for anything that’s outside of the standard of care, what patients normally experience, but mainly we’re just taking a portion of that sample, adding the genomic information, and trying to help the oncologist understand better those patients who will relapse and relapse quickly, as I’ve said.

So over the next year we really plan to do this. We typically see in the order of about 120 or so new AML patients per year in our clinics so this will be an early start with a relatively small number of patients. But, as I said, it’s this foundation of clinical translation, collecting that data, studying it, really understanding which patients relapse in real time rather than that retrospective look that I presented yesterday. And building the case, if you will, that in some patients where we’re detecting using genomics that they will relapse and we may have to do something extra.

I think it really plays in nicely with this current era that we’re in, even in the blood cancers, much like solid tumours, where we’re now seeing pharmaceutical companies really taking the genomic mutations that have been identified in specific genes and either repurposing drugs for those new targets or developing new drugs for those targets as well. And really developing our capabilities to have a precision solution for patients, depending upon what we find when we examine the tumour DNA. So it’s building towards that over the next two years but really this next year will be very critical in terms of collecting the data and seeing what we get. In a way it’s an experiment.

When do you intend on presenting your next data?

I give talks quite often and we work in the genomics of many cancer types. So AML we’ve had the most history in but I do a lot of work in breast cancer and other solid tissue malignancies as well. It’s a wait and see; we usually try to get the results out as quickly as we can, not waiting for a publication or anything formal like that, but once we really feel like we have a definitive result. So I would hope that by the middle of 2015 when we’ve accrued some data and we’re really seeing how this plays out in real time, that’s the challenge, we should be able to present the next steps. Or whether the plan has been a crazy success and we need to do it in other clinics, distribute it out to other patients in other sites or, equally probably, it has been some sort of a disaster and we have to change plans or completely abandon it.

I don’t think the latter will happen but we always have to examine that as a formal possibility. But we’re very hopeful that it will usher in new thoughts and ideas about what to do for these tricky patients that either are going to relapse quickly or, I didn’t mention this initially, but approximately 10-20% of patients essentially never respond at all to the conventional chemotherapy, they’re called primary refractory disease. We don’t really understand the reasons for that but we think we will see some of those patients, undoubtedly given their percentages, in our sample set. There also it will be interesting to think about targeted therapies that may be appropriate for a second go-round, if you will, to really try and get a response out of those patients.

What’s the take-home message?

The thing to say is that what we heard in the session that I spoke in were several examples of solid tumour clinical trials that go in the same vein as what I presented for the leukaemia. So in some ways we’re really paralleling solid tissue but in the haem malignancies we’re certainly not the only people doing this. We’re really trying to apply that genomic information that’s been accruing over the last several years to begin to change the reality for patients and that’s really the plan, if you will, that we hope to accomplish next year and then in the years to come.