EHA 2016

Pre-leukemic clones that survive chemo linked to higher risk of AML recurrence

Dr Klaus Metzeler - Ludwig-Maximilians-Universität, Munich, Germany

Good morning, thanks for giving me the opportunity to present our data. The presentation in the plenary session will actually be given by my post-doc Maja Rothenberg-Thurley and I’m happy to talk here about our data on pre-leukemic clones that survive chemotherapy and that are linked to a higher risk of recurrence in acute myeloid leukaemia.

Acute myeloid leukaemia is an aggressive blood cancer which primarily affects older patients. From previous studies we know today that AML can originate from a clone of pre-leukemic stem cells. These pre-leukemic cells still look normal under the microscope but they already carry some somatic mutations that are later found in the full-blown leukaemia. So when AML is treated with chemotherapy and the patient is in a remission we know that a pre-leukemic clone can sometimes persist and can be detected with molecular methods. So the question that we were asking in our study is does the persistence of a pre-leukemic clone after chemotherapy affect treatment outcomes and patient survival?

To do this we studied 107 patients who were treated on a German multicentre clinical trial by the AMLCG cooperative group. We performed mutational analysis of 68 genes known to be mutated in leukaemias. We started by studying samples obtained at the time of AML diagnosis and we identified leukaemia driver mutations. Then we went to study samples obtained during remission and looked if we can re-identify any of those mutations in the remission sample, indicating persisting mutations and potentially a persisting pre-leukemic clone. Our results – we indeed found persisting mutations in 36% of our patients. You can see here in this graph that these mutations were not randomly distributed but they affected a very distinct set of genes. The grey bars show you all these mutations that were present in the diagnosis sample but lost in the remission sample and the bars indicate their frequency in our patient cohort. The orange bars show you those mutations that persisted in the relapsed samples. So you see that mutation persistence was particularly common in DNMT3A, TAT2, ASXL1 and SRSF2. Those are precisely those mutations that are known to occur in these pre-leukemic clones.

Mutation persistence after chemotherapy was way more common in older patients and importantly patients with persisting mutations had a significantly higher risk of AML recurrence as shown in the red curve in this slide. Importantly, this higher relapse risk persisted even in a multi-variant analysis where we adjusted for patient age, cytogenetics and other risk factors.

So the summary is persistence of pre-leukemic mutations is quite common in AML patients who are disease free after induction chemotherapy. This phenomenon is more frequent in older patients and it’s linked to a higher risk of disease recurrence. It is well known that older AML patients have a poor prognosis but the reasons for that aren’t fully understood yet. So the hypothesis that would come from our data is that frequent persistence of pre-leukemic clones may be one potential explanation for the higher relapse risk that we observe in older patients.