Elias Jabbour, MD; Harry Paul Erba, MD, PhD; Anjali Advani, MD; and Sameem Abedin, MD, discuss key updates pertaining to the leukemia treatment landscape that were presented at the 64th American Society of Hematology (ASH) Annual Meeting and Exposition.
Clinicians tasked with treating patients with leukemia have steadily been gaining more effective treatment options, including new therapeutic approaches for patients who historically have had poor prognoses, including those with acute lymphocytic leukemia (ALL), chronic myeloid leukemia (CML), and acute myeloid leukemia (AML).
In a recent OncLive The Talk video program, Elias Jabbour, MD; Harry Paul Erba, MD, PhD; and Anjali Advani, MD; joined moderator Sameem Abedin, MD, to discuss key updates pertaining to the leukemia treatment landscape that were presented at the 64th American Society of Hematology (ASH) Annual Meeting and Exposition. The panel touched on updated findings that could potentially affect the treatment paradigms for several different patient populations. Treatment regimens included chemotherapy-containing combinations, bispecific T-cell engagers, kinase inhibitors, and novel agents.
In ALL, Jabbour talked through key findings from the following studies:
The group then moved on to discuss updates in CML, with Advani taking the lead and outlining results from the following trials:
Finally, Erba rounded out the discussion by focusing on data from a couple trials in the AML space:
Jabbour: One of the major abstracts was a late-breaking abstract presented by Mark R. Litzow, MD, in frontline ALL. This is the ECOG-E1910 study, which randomly assigned patients with newly diagnosed BCR::ABL1-negative B-lineage [disease] who were minimal residual disease [MRD] negative in remission to either blinatumomab or blinatumomab and chemotherapy followed by maintenance.
The primary end point was overall survival [OS]. Interestingly, investigators showed that the combination with blinatumomab did improve OS where the median has not been reached. The survival for the SOC arm, which is chemotherapy, is 71 months. This is a breakthrough and will change the practice of ALL today. Blinatumomab is approved for only MRD-positive disease; the data from the abstract here would open the door for front line [for use in MRD-negative disease].
[An] abstract I presented in older patients [reported the phase 1] results of the GOLDEN GATE study for patients with Ph-negative B-cell precursor ALL. It is a randomized trial testing low-dose chemotherapy plus blinatumomab induction, consolidation, and maintenance vs a SOC arm. SOC could be HyperCVAD [cyclophosphamide, vincristine sulfate, doxorubicin hydrochloride, and dexamethasone] or the [German multicenter ALL] chemotherapy regimen in patients who are aged 55 years and older, [those who are aged 40 to 55 years with severe predefined comorbidities], [or those who are] unfit for intensive chemotherapy.
Data were reported in 10 patients and showed that the combination of blinatumomab and low-dose chemotherapy is safe and effective. In the safety run-in phase, patients were pretreated and most had a lot of disease to start, with a bone marrow blast of approximately 50%. Ten patients received therapy [and] responded after 1 cycle having a complete remission [CR].
We have 2 randomized trials, GOLDEN GATE and ECOG-E1910, hopefully establishing blinatumomab as SOC in the frontline setting.
In a [third] abstract of a phase 1b study [NCT04521231], patients who were enrolled in the escalation phase were given weekly subcutaneous [blinatumomab] the first week, and then 3 times per week. Among patients in the [dose expansion] cohorts 1, 2, 3, and 4, [investigators] showed that 9 of 14 patients evaluable for MRD were able to achieve MRD negativity, which is better than with IV [intravenous] blinatumomab, and some patients did already fail blinatumomab. Where are we going with this data? We’re expanding 2 cohorts—cohorts 3 and 4—to be able to define the recommended phase 2 dose with those.
This will be joining the GOLDEN GATE study, where we explore blinatumomab [IV] in the front line and blinatumomab in the subcutaneous formulation. I’m very excited about these data, as this will also be practice changing as well.
[In] Ph-positive ALL, we’re going from intensive chemotherapy and transplant to a chemotherapy-free regimen. We know Ph-positive ALL will heavily rely on TKIs [tyrosine kinase inhibitors] and transplant.
We explored a combination of blinatumomab and ponatinib [in a phase 1 study]. Data from 40 patients [showed that] CR is universal, complete molecular response is 87%, and the 2-year OS is 95%, which is the best I’ve seen. The follow-up was 18 months, so very short, but if confirmed in the long run, this is the new SOC. In this study, 1 patient received transplantation—this is really good.
The outcomes are changing drastically in oral therapies. I look forward [to] shorter therapy and a better outcome with the integration of immune therapy, especially [chimeric antigen receptor] T-cell [therapies] in the front line.
Advani: In [OPTIC], they looked at 3 doses of ponatinib: 15 mg, 30 mg, and 45 mg [Table 1].4 If patients reached a BCR::ABL1 less than or equal to 1%, they then de-escalated to 15 mg in all cases, regardless of what dose they had been assigned to.
Data presented at ASH was a 3-year update. The primary end point of the trial was a BCR::ABL1 less than or equal to 1% at 12 months; a secondary end point was safety.
[The trial] randomly assigned 283 patients [across doses]. Looking at responses at the 12-month point, the patients [who] seemed to do the best were on the 45-mg arm. They seemed to have the highest rate of response, and there did not seem to be increased adverse effects [AEs]—at least not significantly—when you looked at the 3 arms.
This study’s focus was [finding an optimal] dose [to] start patients on…[and it was] looking at AEs, which were manageable. Most of the AEs were thrombocytopenia, neutropenia, hypertension, and anemia. Grade 3 or higher AEs were not very high, and the rate of arterial occlusive events was also not high in this study. [This showed] that ponatinib has overall a fairly good efficacy and manageable safety profile.
Abedin: Based on this dose-optimization study, are we comfortable with the way they’ve prescribed ponatinib 45 mg in the beginning, then [reduced] the dose to 15 mg after significant reduction? Is that what we should consider now?
Advani: That’s a great question. [It] definitely looked like there was a difference in the 3 doses, but it’s a little hard to know in terms of OS. [Although] there was a difference looking at the [survival] numbers, I didn’t feel very strongly that you would necessarily need to [reduce]. I was a little surprised that there weren’t more AEs in the 45-mg arm, the general age and cardiovascular profile of these patients.
Jabbour: To Dr Advani’s point, there is no PFS [progression-free survival] difference between the 2 arms, so it can essentially go either way. Yes, we can look at higher response rates, but the advantage is mainly in the T315I-mutant population where we see a big difference in PFS. Practically speaking, 45 mg to 15 mg is the best way to go for a T315I mutation. For the rest, you can go with 30 mg and 15 mg. Numerically, there is no difference for 30 mg and 45 mg yet, but we may see a difference in the long run. In my practice today, I use 45 mg for T315I mutations and patients with multiple TKIs, and I reduce to 15 mg once response is obtained. Otherwise, I use 30 mg and reduce to 15 mg.
Advani: [Another] abstract looked at molecular responses within OPTIC. This [research] tried to see [whether] less than or equal to the cut point of BCR::ABL1 was predictive of a positive outcome for patients. They looked at clinical responses molecularly and showed that if [a patient] did not achieve less than or equal to 10% BCR::ABL1 by 12 months, PFS was decreased. That was the cut point for showing adverse outcomes.
[DASCERN] looked at dasatinib or continuation of imatinib after 3 months of imatinib in patients [who] had a suboptimal response or early molecular response. This looked at patients who had not achieved less than or equal to 10% BCR::ABL1 at 3 months.
The patients in the trial who were in the imatinib arm were allowed to switch over to dasatinib if they met the ELM [extreme learning machine] criteria for crossing over. The interesting thing in this study was [that] approximately half of the patients with imatinib ended up crossing over anyway [after meeting] ELM criteria. Looking at the numbers, there was no difference in OS. There was a numerically small difference in PFS, but it’s a little hard [to interpret] because many of the patients also ended up crossing over—at least half of them.
Abedin: The takeaway that I had with either a crossover to a secondgeneration TKI or not, perhaps it gets patients to a stop point sooner but outside of that there’s no benefit. Is that enough to kind of have that consideration?
Advani: I think especially with a lot of young patients you’re really trying to get them to have a deeper molecular remission so that they’re not on lifelong TKI treatment— that’s the main thing.
Erba: AGILE was a study of previously untreated patients who were unfit for intensive chemotherapy, with IDH1 mutations. Patients were randomly assigned [to] azacitidine and placebo or azacitidine and ivosidenib. A criticism of the study was that the control arm did not have venetoclax, but this study was done in countries where venetoclax had not yet been approved—it could not be done or completed in the United States. The experimental arm was superior, with an improvement in median OS with ivosidenib of 24 months compared with less than a year with azacitidine.
There are several reasons to consider [azacitidine plus ivosidenib] as an option, even though it’s only 5% to 10% of our [patients with] AML. One thing I’ve seen in my own [practice] is that the neutrophil count tends to recover very quickly in the first cycle, [which is] likely due to differentiation, so patients may get out of harm’s way in terms of neutropenia more quickly.
In the analysis of the VIALE-A, if you look at the IDH1-mutant subset of patients, the median survival with azacitidine/venetoclax was 15 months, not 24 months. That’s the median survival for the combination of patients with IDH1- and IDH2-mutations. It’s really the IDH2 that’s driving that. Especially for my older patients, once they get into a remission with ivosidenib and azacitidine there’s not a lot of playing with the doses or the duration of ivosidenib. It’s not myelosuppressive, so you don’t have to worry about how many days [or doses] of ivosidenib, [like] you do with venetoclax.
Are these remissions true, deep remissions? [Another] paper looked at the rate of variants of allelic frequency clearance of IDH1. They used a myeloid gene panel, so the VAF [variant allele frequency] cutoff is 2%. They found [that] in 90% of patients, the IDH1 mutation in the responders was clear. In the concomitant mutations that were not the DTA mutations, there was also a high rate of clearance of other mutations, such as RUNX1.
They [also] looked at patients [with relapsed disease], and there were no patients with an IDH1 mutation. They had 2 patients develop IDH2 mutations, [and] they were taken off for other reasons. Other mutations that were present in genes [included] ASXL1, FLT3, RUNX1, and TET2, in patients who relapsed. [This is] a very nice analysis, especially the deep remissions that were seen in the AGILE trial [Table 26].
Abedin: With the emergence of an IDH2 mutation in patients who are treated with the IDH1/HMA combination, would you consider this a potential second target as an important feature or not in terms of selecting this over venetoclax/HMA?
Erba: The fact that IDH2 mutations come out doesn’t bother me at all. I would still be able to switch to HMA/venetoclax or able to give them enasidenib. The other mutations are also problematic since we don’t have targets for others except for the FLT3.
Now I’m going to move on to a discussion of the magrolimab data, [which] is a first-in-class anti-CD47 antibody target. It binds to CD47, prevents interaction with SIRPα macrophage, then activates the innate immune system to get rid of cancer cells. It may have broad applicability in all of cancer medicine. What was seen in initial trials with azacitidine/venetoclax [and] azacitidine/magrolimab was a very high response rate in patients with TP53-mutated disease, giving us some excitement about that combination in a group of patients who tend to be resistant to chemotherapy.
Data from [The University of Texas] MD Anderson [Cancer Center in Houston were] presented on the triplet combination of azacitidine and venetoclax with magrolimab. The magrolimab dose is ramped up during cycles 1 and 2 until you get to the final dose of 30 mg/kg on days 1 and 15. It was given with azacitidine and venetoclax in the standard dosing regimen of both of those drugs.
In patients with relapsed or refractory disease, especially those who had prior exposure to venetoclax, the activity was low [with] an OS [that] was quite poor. Looking at the data for the [patients with] TP53-mutated and wild-type [disease] with the combination, the response rates were quite high. If you look at the responses that were seen in [the patients with] TP53-mutated and wild-type [disease], they were pretty similar in terms of CR with incomplete hematological recovery rate: 64% of 22 patients with TP53 mutations and 91% in the 11 patients with TP53 wild-type [disease].
More than half of these remissions were MRD negative, and many [patients] had cytogenetic responses. They presented data on this very small cohort of patients, [which showed] encouraging OS compared with what we might expect, but it’s not a randomized trial. We can look at historical controls, and they did exactly that from the database at MD Anderson. They showed that the azacitidine/venetoclax/magrolimab combination compared with hypomethylating agent [HMA]/venetoclax did appear to improve median survival. Their prior data with HMA/ venetoclax in [patients with a] TP53 [mutation] was 5.7 months. That’s what was reported in VIALE-A, but in this trial, the median survival was 10 months. They did a propensity matching that also showed a greater median survival compared with their historical controls with HMA/venetoclax.
[It was] not a randomized trial [and there were] a lot of caveats, including how many patients are brought to transplant and that not all [patients with a] TP53 [mutation] are the same. The VAF may matter—more importantly, whether the patient has a biallelic mutation. A lot of detail is yet to be found, and randomized trials are underway to try to see [whether] we have something that might benefit this poor-risk group of patients with TP53 mutations.
Advani: It certainly sounds like adaptive response may have some pay out with the magrolimab triplet particularly in TP53-mutant disease. There’s previously been data where the doublet certainly hasn’t seemed to improve any survival outcome compared with HMA alone. It does seem to be that this triplet has some benefit in terms of survival as well. What are your impressions in the TP53 wild-type population in this study?
Erba: I think it’s hard to comment right now. I don’t think that magrolimab and CD47 should be restricted to just TP53 mutations. Response rates were very high but it was 11 patients, 91% response rate. You could get that easily in a small subset with azacitidine/ venetoclax so we need a lot more data in terms of the duration of these remissions. The data looked encouraging but until you do randomized trials you really don’t know and those are being done.
Jabbour: Another point with magrolimab is that anemia is seen early on. Originally there were deaths from anemia, and so clinicians should be very careful not to have an individual who is anemic, especially early on, treat them and support them in order to avoid complications. This is something we don’t see often with the triplets if you use HMA, for example the IDH inhibitor is not as myelosuppressive as it is with the FLT3 inhibitors.
Erba: That’s true. The second cycle sometimes had to be delayed a little bit more too for recovery but I agree with you not as myelosuppressive as gilteritinib.
Advani: You really have to work carefully with your blood bank. Logistically, it takes time and this is another thing to think of when expanding this treatment into the community—it may be harder given some of those blood banking issues.
Erba: The next abstract [concerns] a triplet combination [and looks] at an all-oral formulation in AML, where we use a backbone of oral decitabine and cedazuridine with venetoclax in combination with an IDH1 or IDH2 inhibitor. Triplets are the way we’re going.
As we analyze the data being presented, it’s important to keep in mind that in AML, we are also fighting with the myelotoxicity that we see with these regimens. In these small cohorts of patients [receiving] oral decitabine [and] cedazuridine with venetoclax and either ivosidenib or enasidenib, you see high response rates that you might expect because these are [patients with] IDH [mutations] who often respond to HMA and venetoclax.
Does using a triplet do something more than just getting a deeper remission? Does it extend the remission? Does it improve survival? Is the toxicity that we’ll see in terms of myelosuppression going to be worth it for responses or duration of responses?
These triplets may have their greatest benefit in clearing MRD negativity prior to an allogeneic stem cell transplantation, but we have a lot of work to be done. [Based on] data presented in the abstract, we need to hear more about what kind of myelotoxicity was seen with this regimen and what dosing adjustments had to be done.
I’ll finish with a novel agent we haven’t talked about: HM43239, now called tuspetinib. This oral kinase inhibitor has broad activity against various myeloid kinases, including FLT3, SIK, JAK1, JAK2, and the mutant forms of KIP.
In [APTIVATE, which was] presented in a poster at ASH, investigators looked at 50 patients [who were] almost equally divided between [those with] FLT3 wild-type, FLT3-mutated, [and] relapsed or refractory [disease]. The drug was tolerable, and they didn’t see any drug-related QT prolongation or rhabdomyolysis creatine kinase elevations.
In this phase 1 study with dose escalation, in patients with [a] FLT3 mutation, they saw remissions. Among the 22 patients with FLT3 mutations, 18% had a clinical response, and 3 of those 7 patients had prior treatment with gilteritinib and/or midostaurin. In [patients with FLT3] wild-type [disease], 3 of the 28 responded. It’s hypothesized that this inhibitor has activity against other kinases that are important in driving resistance to the agents we have for AML, specifically FLT3. We need to learn who is responding, what pathways are being inhibited, [and whether] it’s a certain mutational profile. In the analysis, they saw these responses with mutations in [several] high-risk genes, including partial tannin duplication of MLL, RAS, RUNX, and TP53.
We have a lot more to learn about this, but it’s exciting data and it gets to what we were talking about in CML. We all remember the debates over having a more targeted ABL kinase inhibitor or a less targeted kinase inhibitor that can overcome resistance pathways. Which one is better? I don’t think we’ve figured it out definitively in CML, and we’re a long way from doing that in AML.
Sameem Abedin, MD, is an assistant professor of hematology in the Department of Medical Oncology at Froedtert & Medical College of Wisconsin in Milwaukee, WI.
Elias Jabbour, MD, is a professor of medicine in the Department of Leukemia at The University of Texas MD Anderson Cancer Center in Houston, TX.
Harry Paul Erba, MD, PhD, is a professor of medicine and a clinical investigator in the Division of Hematologic Malignancies and Cellular Therapy in the Department of Medicine, as well as director and Leukemia Program Director of Phase I Development in Hematologic Malignancies at Duke Cancer Institute and Duke University School of Medicine in Durham, NC.
Anjali Advani, MD, is director of the Inpatient Leukemia Program in the Department of Hematologic Oncology and Blood Disorders at the Taussig Cancer Institute, as well as an associate professor at the Cleveland Clinic Lerner College of Medicine of Case Western Reserve University in Cleveland, OH.