Vadastuximab talirine in combination for newly diagnosed AML

Share :
Published: 3 Dec 2016
Views: 2488
Rating:
Save
Prof Harry Erba - UAB Comprehensive Cancer Center, Birmingham, USA

Prof Erba presents results at ASH 2016 from a phase 1b study of vadastuximab talirine, or 33A, administered alongside standard induction therapy for acute myeloid leukaemia.

Prof Erba discussed these results further in his interview with ecancertv here, and they were reviewed by Dr Mikkael Sekeres in his summary of the conference session.

ecancer's filming at ASH 2016 has been kindly supported by Amgen through the ECMS Foundation. ecancer is editorially independent and there is no influence over content.

Thanks for asking me into speak for my fellow investigators about this phase Ib study in vadastuximab talirine, I’ll call it 33A, in combination with 7 3 induction chemotherapy for newly diagnosed AML patients. So let’s put this in perspective with some background. In 1973, 43 years ago, the first paper was published on what we still continue to use as the initial therapy for a very aggressive cancer, acute myeloid leukaemia. It was a combination of three days of an anthracycline drug and seven days of cytarabine and we call it 3 7 or 7 3, depending on which institution trained you. Nothing has been shown yet to be superior to that, despite four decades of clinical research. We can cure some people with this regimen followed by post-remission treatment.

What we’ve learned more recently is the depth of the remission after that initial treatment may predict for a better outcome or better survival. So not only not seeing the disease under the microscope but using very sensitive tests for minimal residual disease, these have not been standardised yet, they’re not used commercially in practice, these are research tools, but things like flow cytometry, next gen sequencing, this kind of data has shown us that the deeper the remission after the induction the better the outcome for patients which will lead to the hypothesis that if we can change in some way the induction therapy to get a deeper remission that we’ll call an MRD negative, or minimal residual disease negative, state that this may ultimately lead to an improved outcome for patients.

Now, before we did this study of vadastuximab talirine there had been a prior drug on the market that you may have heard of – gemtuzumab ozogamacin, which had the trade name Mylotarg. It is also an antibody-drug conjugate, it is targeted to CD33. CD33 is an antigen, it’s a marker that’s found on the outside, on the cell membrane of leukemic blasts in about 90% of AML patients, so it’s expressed in the majority of patients with AML. There have been randomised clinical trials from Europe, the MRC in the United Kingdom, and France, that have shown improved outcomes with the addition of gemtuzumab ozogamacin to intensive chemotherapy. Recently the Children’s Oncology Group here in the United States showed a lower risk of relapse, a better relapse free survival, for children who had gemtuzumab ozogamacin incorporated into their regimen.

So there’s some historic precedent in the literature that suggests that adding an antibody-drug conjugate may improve outcomes and we hypothesised that this might be by giving a drug that can lower the amount of residual disease after the initial induction.

So we conducted a phase Ib study and these slides are not going to talk much about the dose escalation, but I’m happy to answer those questions. We treated 42 patients in the study and these were adults up to the age of 65, AML excluding acute promyelocytic leukaemia which you know is treated with arsenic and all-trans retinoic acid. They couldn’t have had prior therapy for AML but secondary acute myeloid leukaemia, AML arising from myelodysplastic syndrome, was allowed. The patients received this antibody-drug conjugate 33A in combination with 7 3 on the first day and the fourth day of that seven day regimen in a 28 day cycle and then responses were evaluated by bone marrow aspirates and biopsies at day 15 and 28, the way we normally do with this intensive chemotherapy regimen. A second induction regimen and post-remission therapy were according to the investigators’ choice and did not include CD33A. About half of the patients, in fact exactly half, 21 out of the 42, did go on to get an allogeneic stem cell transplant. The key study objectives were to evaluate the safety and tolerability, to establish a maximally tolerated dose of this antibody-drug conjugate in combination with our standard therapy and, in a preliminary way, to assess anti-leukemic activity.

So there were 42 patients, the median age was 45.5 years old but we treated people up to the age of 65. You can see they either had to have an ECOG score of 0 to 1, which means that they spent very little time in bed or sitting around, they can do most of their normal daily activities. 17% had secondary AML, 50% had intermediate risk karyotypes and 36% adverse karyotypes; we know this is a very important prognostic factor for remission rates but more importantly long-term survival. Only 10% had nucleophosmin mutated disease and 14% had FLT3 mutated – a little bit lower than what we would expect. About half of the patients with a normal karyotype would have a nucleophosmin mutation, so if you say 50% had normal karyotype roughly, 50% of them would have nucleophosmin, that should be about 25%, a little bit lower, and the FLT3 mutated, again that’s about half of what we would expect in a previously untreated population of patients. We don’t have, obviously, an explanation for this except for the possibility that at the same time this study is running at the fourteen centres that conducted this study we also had competing trials that specifically were looking at FLT3 inhibitors. So it’s an interesting topic for the future of how, as we pass out this disease, how do we study the efficacy of these different regimens.

Now every patient treated with intensive chemotherapy for AML develops grade 4 neutropenia, that’s a neutrophil count under 500, at which they’re at risk of infection, and a platelet count under 25,000, grade 4 thrombocytopenia. But some patients started the study without those things and the top part of the panel there shows you the incidence of developing grade 3 or 4 cytopenias while on the study if they weren’t present at the beginning. And there’s nothing really surprising there, the entire regimen as a whole is myelosuppressive, what we need to talk about is is it more myelosuppressive than just 3 7 alone. In terms of the non-hematologic toxicity they’re mostly grade 1 and 2 milder toxicities and very similar to what we would have seen if the patient was receiving 3 7 alone. Now because gemtuzumab ozogamacin was associated with a risk of liver toxicity I’m pointing out here that only one of our 42 patients had a grade 3 elevation in ALT or AST, that’s five times the upper limit of normal.

So here’s the efficacy results. 76% of our patients achieved a response, most of those were complete remission meaning the neutrophils recovered over 1,000, very important for preventing infection, and platelets recovered over 100,000. 17% of the total had a CRi which were all due to the platelets not recovering to over 100,000, very similar to what had been seen with gemtuzumab ozogamacin. Then you see the breakdown according to the cytogenetic risk group – a lower overall response rate in the adverse, the 15 patients with adverse karyotype. However, it compares favourably with what we might expect from historical controls for that cytogenetic group.

So overall a response rate of 76%, not too dissimilar to what we might expect for a well-chosen population of patients who are fit for a clinical trial, not really different, so why are we getting excited about this? Well, as I said, there’s the historical data with gemtuzumab suggesting that this really may lead to something but there was a hint from our study that this may add benefit. The first hint was that 30 out of the 32 patients to achieve that remission only required one round of chemotherapy. Now that might not be something that leads to FDA approval, just that fact, or a registration, but it’s really important for our patient who really cares about the difference between being in the hospital for four weeks versus six to eight weeks if they get two cycles. It also suggested, though, that we might be getting deeper remissions so we did an exploratory analysis, a correlative study, here. We used a very sensitive flow cytometric assay done by one of the international experts on using flow cytometry for looking for minimal residual disease, Brent Wood at the University of Washington. What we found was that 25 out of those 32 complete remissions, so just over three-quarters, were actually complete remissions that were MRD negative. So going back to our hypothesis, if we can get patients into an MRD negative state maybe the outcomes would improve and later today I’ll present some comparison with another dataset suggesting that that rate of minimal residual disease may actually be better.

So, in conclusion, what we felt we showed was that we were able to combine active doses of 33A with 7 3. The doses here were less than the dose used as a single agent but all doses in our phase Ib study, including lower doses than what we actually used here, showed complete remissions as a single agent. So they are effective doses of 33A and it added acceptable on-target myelosuppression. We saw platelet counts recovering over 100,000 and neutrophils over 1,000 by about 4½ - 5 weeks, which we felt was reasonable and patients were able to go on and get post-remission therapy. The non-hematologic toxicity was similar to 7 3 alone and, again harkening back to gemtuzumab, we didn’t see any veno-occlusive disease or sinusoidal obstructive syndrome and only one patient with reversible grade 3 hepatic toxicity. And it was a very low 30 day mortality of 2%.

There’s also preliminary evidence of added activity, not just the response rate of 76%, you could argue that that’s not really dissimilar to what we might expect, but the responses were rapid, achieved in most patients with only one cycle, and three-quarters, 78%, were deep remissions that were MRD negative which we are hoping will translate into a real benefit for patients which obviously we’ll only be able to test in the setting of a randomised trial which is planned to start in the first quarter of 2017 comparing 7 3 versus 33A with 7 3. We’re also still exploring in this phase Ib study just a single dose on day one. Thank you.