ESH CML/MPN 2025 | Studying blast phase CMML uncovers two distinct subgroups and significant heterogeneity

You know, blast-phase CMML represents when CMML transforms into acute myeloid leukemia based on the standard definition of 20% blasts in blood or bone marrow. It happens in probably 20 to 30% of CMML patients at some point in their course, and it results in a really devastating form of AML that has a particularly dismal prognosis, I think many viewers will probably have experienced that. It’s a very difficult AML to treat and really very few things work, even transplant has very poor outcomes in that group. 

So our recent work was to try and better understand blast-phase CMML. It’s not been very well studied. And so what we did was we took… well we confirmed the poor prognosis in a very large cohort of patients, and we confirmed, as expected, the median survival was less than six months for these patients. And in our lab we have access to a large biobank of samples, which included 40, I think it was 44 cases, where we had cells stored in the bank from the chronic phase, but also after transformation to blast phase. And we had a look at various parameters to try to sort of better understand whether these are all the same or whether there’s any heterogeneity amongst these cases and see if there’s anything there that could be exploited to improve therapy. And the upshot was by looking at various factors: we looked at the immunophenotype of the blast phase CMML, we looked at the mutations, and we also did RNA sequencing on all of the blasts from the blast phase CMML. We did some machine learning to try and sort of separate these out into distinct subcategories, and we found some interesting things, particularly that it is very heterogeneous, and there did seem to broadly be two subgroups. So whereas most people think of CMML transforming to a monoblastic or monocytic subtype of AML, that’s only the case in just around half of cases and quite often the transformation occurs in a much more primitive stem cell, so you get a much more sort of N0, N1 type AML than one would expect from CMML. 

And we showed important differences in therapeutic sensitivity between these two subgroups. We show particularly poor responses to venetoclax or ven-aza, more so in the more mature, what we call the mature subtype transformation. And we also looked at drug sensitivity profiles and tested a range of existing novel compounds and showed quite different sensitivity patterns across this sort of spectrum of maturation. 

So it’s discovery work at this stage, but it could serve as a framework for future clinical trials whereby we might be better off defining the subtype more than just saying that they’ve transformed to AML and we might start to treat them based on the type of AML, particularly focusing the existing drugs differently. We may be looking for other things in the more mature subtype AMLs than standard ven-aza, which would probably be the standard of care, but we know doesn’t work very well. And also as a framework for future clinical trials where we could potentially try and target specific novel therapies to specific transformation subtypes.

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.