ASH 2020 | Magrolimab in combination with azacitidine in untreated AML

So magrolimab is a very unique drug. This works through a mechanism of action called CD47 inhibition. CD47 is expressed on the surface of tumor cells, and this interacts with SIRP-alpha, which is present on the surface of macrophages. And this interaction basically shuts down or exhausts macrophages, and then the macrophages are not able to do their routine activity, which is to phagocytize the tumor cell.

So about 15, 20 years ago, researchers at Stanford started looking at potential ways to harness the macrophages as an attacker to the tumor cells. And what they found is that if you blocked the CD47/SIRP-alpha interaction, you could reactivate macrophages, and then they would phagocytize tumor cells. And this activity pre-clinically was most profound in acute myeloid leukemia, and then in lymphoma. And so now almost a decade later, we are starting to see those clinical trials translating into good clinical activity in those two disease subsets that were predicted, which are acute myeloid leukemia, also myelodysplastic syndrome, and then in lymphoma. So at this year’s ASH meeting, we’ll be presenting the data specifically in the acute myeloid leukemia frontline patients.

Now to kind of give a background, there has been a lot of progress in frontline acute myeloid leukemia therapy. We have, of course, azacitidine venetoclax treatment, which is giving us unprecedented CR/CRi rates of almost 70%, with the median overall survival between 15 to 18 months, and even two-year survival rates of around 40%, which is major improvement compared to what we had historically with azacitidine. But at this year’s ASH, you will see, and also there is published data showing, that there are certain subsets that are not responding as well to azacitidine and venetoclax, and there is still room for significant improvement.

One of those is TP53. So now there have been three data sets, including the Phase IB study of azacitidine and venetoclax, as well as the VIALE-A study published in New England Journal recently, as well as a oral presentation by Dr Kim from MD Anderson, that will look at decitabine, 10-days venetoclax. In all these three publications, there’s about 80 or so TP53-mutated patients, and what we’re seeing consistently is that the CR/CRi rate is around 50 to 55%, which is not bad. But the median duration of response is only four to five months, which is not good, and the median overall survival is only five to seven months.

In fact, in our MD Anderson analysis, Dr Kim looked at dec/ven versus historical studies with dec alone in TP53-mutated AML, and we actually find that there was no benefit. Median overall survival is five months versus 4.5 months with the dec/ven versus dec alone. So now we’re even starting to wonder if the ven is adding much at all. It does improve response rate. That may be a benefit.

So now with the aza/magrolimab, what we have started seeing over the last year and a half, is that not only are the response rates high in the general frontline AML, the overall response rate is about 64% with the CR/CRi rate of about 60%, but specifically in this high risk group of TP53-mutated AML, we started seeing an enrichment of higher response and improved survival.

So at this year’s ASH meeting, we will be presenting… Dr Sallman, my colleague, will be presenting data looking at this combination of the azacitidine/magrolimab in all frontline AML, but focusing on TP53 AML. So we have a total of about 50 patients that will be presented here, frontline, older AML, not considered suitable for induction chemotherapy. 34 of those patients are TP53-mutated AML, where we expect with aza/ven, CR/CRi rate of 50, 55%, median overall survival five to seven months, what we’re seeing with the aza/magrolimab is a CR/CRi rate of about 70%, so better than the 50/55%, but really more important, you were seeing the duration of response seems to be quite improved, almost nine and a half to 10 months. This is better than the four to five months published consistently with aza/ven, and more importantly, the median overall survival is coming close to 13 months, which is almost double of what we have seen, five to seven months, with aza/ven in the Phase I, Phase II, Phase III, and our MD Anderson study.

So we believe there is a real path for the aza/magrolimab, clearly in TP53 AML. But what’s interesting is that we did have a cohort of 16 frontline non-TP53-mutated AML patients… The enrollment, as you can imagine, was hard for this because aza/ven was doing so well in the phase three, and people were using that extensively. But what we actually see is that the CR/CRi rates are close to 60%, so a little lower than aza/ven, but not bad, but importantly, the median overall survival is 18.9 months, which is actually very similar to aza/ven, even in the TP53 wild-type.

So I think that the magrolimab has, hopefully, a very bright future. Initially, the approach is going to be to focus on TP-53 AML with randomized studies, looking at that population to get the drug approved for AML. But I think that there is an important role for this drug also in wild-type. And maybe there doesn’t have to be competition of aza/magro, aza/ven, but a triplet could be the solution to really improve survival and long-term outcomes, three and five-year survival. And so we have now started a triplet at MD Anderson of azacitidine, venetoclax, magrolimab for all frontline AML patient, including TP53 wild-type and mutated. But I think that this drug will be an important one with a unique mechanism of action that we can combine with many of our other backbone therapies in the future.