ASH 2020 | Decitabine vs hydroxyurea for advanced proliferative CMML

This was an academic trial that was conducted in three European countries: Germany, Italy, and France. And the rationale is that there is no available disease modifying treatment options in patients with proliferative CMML. Most of these patients are not eligible for STEM cell transplantation. And actually there was a nulled randomized trials carried in the 1990s showing that hydroxyurea was beneficial in this patient compared to other cytoreductive therapies. Nevertheless, over the last 10 to 15 years, we’ve been extensively using hypomethylating agents in these patients, mostly in retrospective studies or phase II studies. And thus, we aim to carry a prospective randomized comparison of the standard cytoreductive therapy with hydroxyurea, with a single agent hypomethylating agents and namely decitabine. This was a randomized trial focusing on proliferative CMML patients with advanced disease based on myeloproliferative features or cytopenias or bone marrow blast success. So in total, there was 170 patients randomized one-to-one to either arm. The treatment was well balanced between the groups.

And the most, the first thing we noticed was that as expected, there was a higher response rate with the decitabine compared to hydroxyurea. Nevertheless, the primary endpoint of the study was event-free survival defined as a death, transformation to leukemia or progression of CMML. There was no difference between patients treated with decitabine and hydroxyurea with respect to the primary endpoint of event-free survival, neither was there a difference in terms of overall survival. So we’ve inspected several reasons why this could be, why patients that have a higher response rate with decitabine do not have a superior event-free or overall survival.

The first explanation we sought was that roughly one third of patients with hydroxyurea received hypomethylating agents after exiting the study. However, censoring for that did not change our conclusion. So this is probably not the explanation. Another explanation could be that the benefit of decitabine is confined to a subset of patients.

So we have a molecular analysis ongoing, but using conventional scoring system suggests CPSS, there could be a slightly higher benefit of decitabine in patients with intermediate to or high-risk CPSS. So patients with higher risk disease, that were roughly two thirds of the included patients, but this did not reach statistical significance.

So the last hypothesis that we are currently exploring is that the better disease control with decitabine was offset by more frequent adverse events and toxicity. There are signals for that, that we are currently investigating. So, I think the main conclusion from the Dakota trial is that it is feasible to conduct randomized trial in CMML that has been an orphan [inaudible] in neoplasms, in that respect over the last decades, though, we must coordinate our efforts. We are convinced that hypomethylating agents are potentially a disease modifying agents, but they need to be combined with other agents because they have a superior response rate that should be translated into a survival benefit for these high risk patients.