CME
Physicians: Maximum of 1.00 AMA PRA Category 1 Credit™
Released: March 21, 2023
Expiration: March 20, 2024
Eunice Wang, MD:
Leukemia stem cells have been shown to overexpress BCL2, which helps promote their survival. This understanding led to the development of venetoclax, a BCL2 inhibitor, for treating AML.1,2 VIALE-A was a double-blind, randomized phase III trial comparing azacitidine plus venetoclax vs azacitidine plus placebo in previously untreated patients with AML who are ineligible for standard intensive induction treatment.2 Results from this trial showing improved OS with venetoclax plus azaciditine resulted in FDA approval and established this combination as the standard of care for older and/or unfit patients with newly diagnosed AML.3
Eunice Wang, MD:
VIALE-A enrolled 433 patients who were 75 years of age or older or who had significant comorbidities that precluded them from receiving intensive induction chemotherapy. A key eligibility criterion was no prior treatment with hypomethylating agents (HMAs), and patients had to have intermediate-risk or poor-risk AML. Patients were randomized to receive either venetoclax 400 mg/day or placebo daily plus azacitidine 75 mg/m2 daily (SC or IV) for Days 1-7 of each cycle in both arms.
The primary endpoint was OS, with secondary endpoints including responses and MRD negativity. At ASH 2022, Pratz and colleagues4 presented updated results after 2 further years of follow-up from VIALE-A, for a median follow-up duration of longer than 42 months.
Eunice Wang, MD:
The median age of patients in both groups was 76 years. As one would expect in an older population of patients with AML, many had secondary AML (25%); of this group, more than one half had prior myelodysplastic syndrome (MDS). Approximately one third of patients had AML with myelodysplasia-related changes. The Eastern Cooperative Oncology Group (ECOG) performance status (PS) ranged from 0 to 3, and treatment was initiated regardless of initial presenting mutation status.
At 42-43 months of follow-up, no patients remained on treatment in the placebo/azacitidine group, and only 25 patients (9%) remained on treatment in the venetoclax/azacitidine group.
Eunice Wang, MD:
Review of the safety data demonstrated no new significant safety signals in long-term follow-up, with an overall similar frequency of adverse events (AEs) between both arms. Of note, nearly all patients (>95%) experienced grade ≥3 AEs. As previously reported, hematologic events were increased in patients receiving venetoclax plus azacitidine.
Eunice Wang, MD:
OS was similar to earlier results reported from this study: Patients receiving venetoclax/azacitidine achieved a median OS of 14.7 months vs 9.6 months with placebo/azacitidine (HR: 0.58; P <.001). Patients with IDH1/2 mutations had improved outcomes overall, with a median OS of nearly 20 months with venetoclax/azacitidine vs 6.2 months with azacitidine/placebo (HR: 0.314; P <.001). Patients in the venetoclax/azacitidine group who achieved CR and had evidence of MRD negativity (MRD <10-3) also had improved outcomes, with a median OS of 34.2 months vs 18.7 months in patients who achieved CR with venetoclax/azacitidine but did not reach the MRD threshold.
Eunice Wang, MD:
In the venetoclax/azacitidine arm, the duration of CR and CRi at the 100% OS analysis was slightly longer at 18.2 months than the previously reported 17.5 months. The duration of CR also was slightly longer at 22.1 months vs 17.5 months in the previous report.
Eunice Wang, MD:
In summary, the long-term results from the VIALE-A phase III trial continue to confirm the OS benefit of venetoclax/azacitidine compared with azacitidine monotherapy in this population of older and/or unfit patients with newly diagnosed AML. The OS benefit was confirmed for patients who had previously been reported to have good risk, including those with IDH1/2 mutations and those who achieved MRD-negative status as defined on this protocol. No new safety signals were identified.
Dr Fathi, what do you think about these outcomes?
Amir Fathi, MD:
I thought the presentation was reassuring, as it confirms what we already suspect: Patients who are older and not sufficiently robust for intensive therapy do well with this combination and do much better than with an HMA alone; patients with certain molecular subtypes do particularly well, including those with IDH mutations; and patients who achieve very deep remissions have better outcomes. Most of this is not surprising.
Traditionally, older patients with AML have very poor outcomes. What affected me the most from this presentation was the small proportion of patients with extended survival. I have now begun to wonder whether long-term survival may be possible in patients who receive azacitidine and venetoclax. I suspect the combination of HMA/venetoclax eventually will be evaluated in a younger patient population. It would be wonderful if a larger proportion of patients could have extended survival with this combination.
Eunice Wang, MD:
I agree. In this study, 9% of patients remained on azacitidine/venetoclax at 43 months of follow-up. The fact that we have 5-year long-term data in this patient population is remarkable. This study is important in setting the groundwork for future studies where we can further improve on this impressive data, either with triplets or moving into a younger patient population.
Eunice Wang, MD:
As discussed, the combination of azacitidine and venetoclax is approved for patients with previously untreated AML who are ineligible for intensive chemotherapy.3 In the approved indication, azacitidine is given for 7 days and venetoclax is given once daily in a continuous fashion as part of a 28-day cycle. Willekens and colleagues5 examined whether there is a role for azacitidine in combination with truncated venetoclax administration in older, unfit patients who perhaps did not meet the eligibility criteria for the original VIALE-A trial.
In their retrospective analysis, the investigators evaluated charts of 82 patients with newly diagnosed AML who were ineligible for intensive chemotherapy and had received a “7+7” regimen of azacitidine plus venetoclax for Days 1-7 of a 28-day treatment cycle (no monotherapy venetoclax between Days 8 and 28).
Objectives included response rates, survival, and determining whether a truncated venetoclax schedule might mitigate some of the myelosuppressive toxicities that can prevent long-term use of this combination.
Eunice Wang, MD:
The median patient age was 75.2 years (range: 50.5-89), which is similar to the 76-year median age in VIALE-A.2 More than one half of the patients in this retrospective analysis were 75 years of age or older. Approximately one third of patients had medical comorbidities that would have excluded them from participation in VIALE-A (eg, active solid/lymphoid neoplasm), and some had an ECOG PS as high as 4; in the VIALE-A trial, the PS was a maximum of 3. In this analysis of the 7+7 regimen, the ECOG PS was between 2 and 4 in approximately 40% of patients.
Approximately 70% of patients were of adverse risk per European LeukemiaNet 2022 criteria. TP53 mutations were found in 21% of patients, and 33% had poor-risk cytogenetics. Overall, this was a very frail cohort with biologically poor-risk disease.
Eunice Wang, MD:
In this report, the follow up on the 7+7 regimen was very short (only 4.8 months), with a median of just 4 cycles. Nevertheless, toxicity led to discontinuation of the truncated venetoclax/azacitidine regimen in 28% of patients after the third cycle.
Despite this, results showed an ORR (CR/CRi) of 68.3%, which is very similar to the 66.4% ORR in the original VIALE-A trial. There was a slightly longer time to achieve a response (2 cycles vs 1.2-1.4 cycles in VIALE-A). Despite this, patients with normal karyotype, mutated NPM1, and mutated IDH2 continued to do well, with ORRs of 80% to 100%. Patients with TP53 mutations or complex karyotype had ORRs of 44% and 47%, respectively, and those with adverse European LeukemiaNet risk also did relatively well, with an ORR of 61%.
Eunice Wang, MD:
The 60-day mortality rate was 6.1%. Rates of febrile neutropenia (49%) and platelet or red blood cell transfusions (88%) were similar to results from VIALE-A. Patients required a median of 36 days to neutrophil recovery and 31 days to platelet recovery with the 7+7 regimen.
Eunice Wang, MD:
Median OS in this study was slightly less than in VIALE-A (12.8 vs 14.7 months). Median event-free survival (EFS) was again very short at 7.5 months. It is possible that the shorter OS and EFS in this study were due to worse PS, older age, and more comorbidities in this population.
Among the 58 patients in this study who would have been eligible for VIALE-A (ie, no exclusion criteria), the median OS was similar at 13.8 months, and median EFS was 11.4 months. In the 63 patients without a TP53 mutation, both median OS and median EFS were 17.15 months. In patients with NPM1 and/or IDH2 mutations, both median OS and median EFS were much longer than in other groups at 25.76 months.
Eunice Wang, MD:
Despite venetoclax being limited to only 7 days per cycle, 61% of patients in CR/CRi still required further dose reduction, mostly due to myelosuppression. To mitigate myelosuppression, the venetoclax duration was reduced from 7 days to 5 days in 97% of patients, the venetoclax daily dose was reduced to 200 mg in 50% of patients, and the interval between cycles was extended to 35 days in 29% of patients. Despite this, 71% of patients in CR/CRi remain on treatment with this truncated regimen, and the estimated median OS for patients continuing on therapy is 25.8 months.
Eunice Wang, MD:
In this retrospective analysis of reduced venetoclax plus azacitidine (7+7) in difficult-to-treat patients with AML, there was similar early toxicity after cycle 1, and response rates (68%) were similar to the conventional 28-day venetoclax schedule.
As Dr Fathi mentioned, we are very impressed to see such variations on the standard VIALE-A regimen and appear to achieve long-term survival even in older, frail patients with AML with very adverse risk. This 7+7 regimen looks to be almost equally efficacious in this patient population as the standard of care, suggesting that patients may be overtreated with 28 days of venetoclax. Surprisingly, the hematologic toxicity was not significantly mitigated with the 7+7 regimen, so this still needs to be closely monitored and will remain a therapeutic challenge.
Moving forward, looking at backbones for further clinical development, more regimens may incorporate less venetoclax given these intriguing retrospective data. This clearly warrants validation in prospective clinical trials.
Amir Fathi, MD:
I expect these results will trigger more controversy and conversation. In my opinion, the traditional 28 days of venetoclax is probably too much. It makes sense to decrease the duration of venetoclax over the course of treatment with subsequent cycles to allow patients to tolerate treatment better longitudinally and benefit more over the long term.
The challenge I have with these data is that it does not seem to make a lot of intuitive sense to use less venetoclax and still see significant toxicity, which the investigators report. After reducing the venetoclax duration to decrease the intensity, why does much of the toxicity still remain? And why is the duration of count recovery still so slow?
Also, although the ORR seems to be basically the same, it is far less durable—by several months. This may be due in part to the patient population in this analysis, which was older, had more comorbidities, and had a poorer PS than other patients with newly diagnosed AML. I am not fully convinced on whether 7 days is generally better than 28. Going forward, I am likely to start with the continuous dosing but then do an early marrow biopsy to make a determination about whether to reduce the duration of venetoclax in subsequent cycles.
In my opinion, going from 7 to 5 days of venetoclax is unlikely to make a significant difference. In my clinical practice, when patients with AML are receiving only 7 days of venetoclax, I prolong the duration of cycles of HMA treatment, which seems to have been the third most frequent management strategy the investigators took here.
Eunice Wang, MD:
I agree with your assessment. I think the intriguing thing is, we often have patients who are likely too sick and too frail to consider traditional venetoclax plus azacitidine simply because of the duration and depth of the myelosuppression. The results from this retrospective analysis suggest that some of those older, more frail patients might still be eligible for therapy. I noticed this analysis did not include MRD data, and I am curious if depth of MRD might be more associated with the duration of venetoclax than we think. Overall, these results open the door to treating selected older, frail patients with AML who have a PS of 3 or 4 and concomitant cancers.
It is intriguing that shortened dosing of venetoclax could have very impressive results, but I do not think these results mean we should start to use this truncated regimen in routine clinical practice. Like you, in many of my older patients I am routinely doing a bone marrow biopsy on Day 21 to determine if I should shorten venetoclax therapy. The question now is whether I should get the biopsy at Day 14 instead and consider stopping after 14 days.
Amir Fathi, MD:
As we’ve mentioned previously, the field of AML therapy is investigating triplets based on the backbone of venetoclax plus azacitidine. Potential third agents include IDH inhibitors, FLT3 inhibitors, and monoclonal antibodies.
At ASH 2022, Daver and colleagues6 presented results from a phase I/II trial of azacitidine and venetoclax plus the anti-CD47 antibody magrolimab in patients with AML.
The study included a frontline cohort (n = 43), a cohort of patients with R/R AML who were venetoclax naive (n = 18), and a cohort of patients with R/R AML with prior venetoclax exposure (n = 18). Patients received azacitidine 75 mg/m2 on Days 1-7 and venetoclax 400 mg on Days 1-21. Magrolimab was dose escalated to 30 mg/kg during cycle 1 and then administered weekly in cycle 2, followed by administration every 2 weeks in subsequent cycles. The primary endpoints were the maximum tolerated dose (MTD), the recommended phase II dose (RP2D), and the rate of CR/CRi.
The patient population was somewhat younger in the venetoclax-naive group, but otherwise patient characteristics were well balanced between arms. Of note, more than 60% of patients in all 3 arms had TP53 mutations, a population of particular interest with magrolimab due to responses seen in previous studies of azacitidine/magrolimab doublets.7,8
Amir Fathi, MD:
The efficacy data for the 43 patients who were treated in the frontline setting are shown here. The CR/CRi rate is quite promising, despite relatively small patient numbers. Among patients with de novo AML, the rate is lower in those with TP53 mutations (64% vs 91% with wild-type TP53). Of note, although only 5 patients with secondary AML had TP53 mutations, 3 of them achieved a CR/CRi. Rates of MRD-negative CR/CRi also were promising in the de novo AML group at 60% with wild-type TP53 and 64% with mutated TP53.
Amir Fathi, MD:
The OS curves for patients in the de novo frontline cohort (n = 33) showed 12-month OS rates of 83% with wild-type TP53 vs 53% with mutated TP53. Among the 10 patients with secondary AML, the median OS was <10 months with or without mutated TP53.
Amir Fathi, MD:
In the frontline cohort, 9 patients with TP53-mutated AML underwent hematopoietic stem cell transplant (8 de novo, 1 secondary). With the caveat of small patient numbers, those who underwent transplant did better, particularly in terms of relapse-free survival (RFS), with a median RFS of 16.3 months vs 4.2 months with no transplant. Median OS also was improved in the transplant group at 16.6 months vs 9.8 months with no transplant.
Amir Fathi, MD:
In the R/R cohorts, the 18 patients who had not received prior venetoclax did significantly better, with a 44% CR/CRi rate, which is quite good. By contrast, only 2 of the 18 patients (11%) with prior venetoclax exposure achieved remission. This is disheartening, but perhaps not surprising, as patients who previously received venetoclax likely developed resistance that the addition of magrolimab was not able to overcome.
Amir Fathi, MD:
At a median follow up of 17.9 months, it is clear, despite the small number of patients, that patients who are venetoclax naive had better outcomes compared with those with prior venetoclax exposure. The median RFS was 7.5 months in the venetoclax-naive group vs 3.1 months with prior venetoclax exposure. Likewise, median OS was 5.6 months vs 3.1 months, respectively.
Amir Fathi, MD:
There was no significant signal in terms of grade 3 or higher AEs, except perhaps febrile neutropenia (44%), which is commonly seen with the venetoclax/azacitidine doublet and lung infection (35%). Adding magrolimab increases the potential for infusion reactions, but in this study only 3 (4%) patients had grade 3 infusion reactions. These were mitigated with steroid pretreatment for subsequent doses. No patients discontinued treatment due to TRAEs.
Amir Fathi, MD:
In this phase I/II study of the novel combination of magrolimab with azacitidine and venetoclax, the triplet demonstrated encouraging activity as frontline therapy for AML. In patients with TP53-mutated de novo AML, the CR/CRi rate was 64%, and in those with wild-type TP53 de novo AML, it was 91%. In my opinion, the promising data in the newly diagnosed patients with TP53 mutations are worthy of further study. The R/R cohorts are a bit of a mixed bag. It appears that individuals with prior exposure to venetoclax likely do not benefit from triplet therapy, but venetoclax-naive patients seem to have a promising rate of composite response.
Dr Wang, what are your thoughts about Dr Daver’s presentation of this study?
Eunice Wang, MD:
The lower degree of benefit in patients with R/R AML may represent different mechanisms of resistance at the time of recurrence. In any case, those data showed disappointing DoR, particularly, as you mentioned, in patients who had prior venetoclax exposure.
I think this triplet is intriguing. Results from previous studies of the magrolimab/azacitidine doublet suggested higher response rates and longer DoR in patients with TP53 mutations.8 I am just not sure that making it a triplet regimen with the addition of venetoclax significantly increases benefit. It will be important to see larger trials of how the triplet fares in the upfront AML setting, particularly for patients with mutated TP53, where there definitely could be a role. I do not expect there will be a meaningful clinical role in patients with wild-type TP53.
Amir Fathi, MD:
Let’s move on to another class of agents, menin inhibitors. Menin inhibitors are small molecules that interfere with the interaction of the menin protein with KMT2A. By inhibiting this interaction, the block of myeloid differentiation is released, which leads to normal differentiation and maturation of myeloid precursors and, hopefully, therapeutic response.9
At ASH 2022, Dr. Erba and colleagues10 presented initial results from the phase I/II KOMET-001 study of ziftomenib in patients with R/R AML. Both Dr Wang and I are involved in this particular study. The data presented at ASH focused on the phase Ia and Ib portions of the trial. This study will eventually enroll a planned 60 adult patients with R/R AML. Patients with or without KMT2A or NPM1 aberrations were included in the dose-escalation phase Ia portion of the study, where ziftomenib was administered at varying dose levels (50-1000 mg/day). The primary endpoints for the phase Ia portion of the trial were MTD and RP2D.
In the subsequent dose-validation/dose-expansion phase Ib portion, ziftomenib was administered at 2 dose levels that showed promise in phase Ia: 200 mg/day and 600 mg/day. The primary endpoints in phase Ib were safety and the minimal effective dose. Of note, phase Ib focused only on patients with AML with KMT2A rearrangements or NPM1 mutations because, mechanistically, menin inhibitors are thought to be most effective in these subtypes.
Amir Fathi, MD:
Most of the baseline characteristics were typical of this population. I would like to highlight the mutation status of these patients: Approximately one third of patients had KMT2A alterations, 13% of patients had NPM1 mutations, and more than one half of patients had neither of these target alterations. Of note, several patients had undergone heavy pretreatment and transplant: 73% had prior venetoclax, and 23% had a prior stem cell transplant.
Amir Fathi, MD:
The Ia safety data showed no significant safety signals. Many of the AEs were cytopenias, and these were mostly mild. Treatment-related QT prolongation, which was of potential concern, was not detected. Two patients had dose-limiting toxicities: At 400 mg, 1 patient had pneumonitis and postaspiration syndrome, and the 1 patient who received 1000 mg had treatment deescalated to 800 mg because of differentiation syndrome.
Amir Fathi, MD:
An efficacy signal was noted in phase Ia: 2 cases of decreased leukemic burden in the group with NPM1 mutations, ultimately leading to 1 CR. Stable disease was noted in 3 patients with KMT2A rearrangements.
Amir Fathi, MD:
Phase Ib of this study focused only on patients with KMT2A rearrangements and NPM1 mutations and evaluated 2 doses: 200 mg (n = 17) and 600 mg (n = 36). Patients were heavily pretreated with a median of 3 prior therapies.
Amir Fathi, MD:
An AE of particular interest with menin inhibitors is differentiation syndrome, which can be severe with this class of therapy. In KOMET-001, DS was more common in patients who had KMT2A alterations (38.5% with 200 mg and 37.5% with 600 mg), including those with grade ≥3 toxicity (30.8% with 200 mg and 25% with 600 mg). Of interest, the incidence of DS was lower in studied patients with NPM1 mutations. In the 600-mg cohort, 20% of patients with NPM1 mutations had DS, with only 1 patient with grade ≥3.
In total, 5 patients developed DS of any grade with the 200-mg dose vs 10 patients with the 600-mg dose. Despite numerical differences between the 2 ziftomenib doses, the rates of DS were similar between the 200-mg and 600-mg doses (29% and 28%, respectively).
In my opinion, the incidence of DS likely affected both response and tolerability. At 600 mg, the ORR in patients with NPM1 mutations was 75% vs only 16.7% for those with KMT2A rearrangements.
Amir Fathi, MD:
In the group with NPM1 mutations, those receiving ziftomenib 600 mg had a CR rate of 30%, which is excellent in R/R patient populations. The composite CR rate was 35%, and the ORR at this dose was 40%.
In patients with KMT2A rearrangements at the 600-mg dose, the response rates were significantly lower. The composite CR rate was 11% (2 patients), and the ORR was 16.7%. There were no CR/CR with partial hematologic recovery (CRh) responses with the 200-mg dose for this cohort of patients.
Amir Fathi, MD:
The investigators analyzed responses more thoroughly in the 600-mg group. In those with NPM1 mutations, 30% achieved CR. Another 5% of patients each achieved CRi/CR with incomplete platelet recovery (CRp) and morphologic leukemia-free survival (MLFS); 35% of patients had stable disease. The proportion of responses was lower in patients with KMT2A rearrangements, but a substantial proportion of patients (44%) had stable disease.
Amir Fathi, MD:
Menin inhibitors are an exciting new class of drug with a real efficacy signal and, for the most part, are reasonably tolerable in these early-phase data.11 That said, patients receiving ziftomenib are at risk of DS and require vigilance, close monitoring, and active treatment to help them better tolerate this agent.
I thought the response data in the patients with NPM1 mutations, who were heavily pretreated for their AML, were quite remarkable. Patients with KMT2A rearrangements had lower rates of remission, but I wonder if some of that was affected by the tolerability in that group. Evaluation in larger numbers of patients will be needed to truly understand the efficacy.
What are your thoughts, Dr Wang?
Eunice Wang, MD:
I too was impressed with these findings. In the phase Ib portion, approximately 30% of the patients with NPM1-mutated disease had evidence of a CR, so clearly ziftomenib has clinical efficacy. Even in the patients with KMT2A rearrangements, many had extramedullary disease or stable disease that did not quite meet the criteria for CR but nevertheless clearly benefited from treatment. This is a very exciting result, and I am looking forward to more studies of ziftomenib in these and other patient subsets of R/R AML.
Amir Fathi, MD:
Dr Wang, what is your assessment of the risk of DS with ziftomenib?
Eunice Wang, MD:
DS is a common AE because of the mechanism of action of some targeted therapies in AML, for example, IDH1/2 inhibitors and FLT3 inhibitors, as well as retinoic acid and arsenic for treatment of acute promyelocytic leukemia. The differentiation associated with menin inhibitors appears to be somewhat more aggressive. I think this represents evidence of biological activity, which could translate into clinical efficacy if the patient can tolerate the differentiation effect.
In terms of management, for some patients, cytoreduction to lower the white blood cell count before starting treatment would be essential to minimize the risk of DS. I think mitigation strategies need to be built into clinical trial regimens for treatment with menin inhibitors and other potent differentiation agents. As you know from your experience and expertise with IDH inhibitors, guidelines have been developed for monitoring and managing DS. IDH inhibitors have been successfully developed despite black box warnings for DS. A series of recommendations for DS associated with menin inhibitors will likely be warranted, similar to what you and your colleagues have done for IDH inhibitors.
Amir Fathi, MD:
Another menin inhibitor, revumenib, is currently in development. Updated results from the dose-escalation phase I AUGMENT-101 study of revumenib in R/R acute leukemias with KMT2A or NPM1 alterations were presented at ASH 2022.
The study population consisted of both adult and pediatric patients and had 2 cohorts based on concurrent receipt of strong CYP3A4 inhibitors. The primary endpoints were safety, tolerability, pharmacokinetics, and determination of the RP2D.
The patient population included AML (n = 56), ALL (n = 11), and 1 patient with mixed-phenotype acute leukemia. Most patients were heavily pretreated with a median of 4 prior therapies, including prior stem cell transplant in 46% of patients. In total, 68% had KMT2A rearrangements, and 21% had NPM1 mutations.
Amir Fathi, MD:
One of the specific AE signals for this agent is QTc prolongation, which occurred in 53% of patients but was grade ≥3 in only 13%. No patients discontinued therapy due to QTc prolongation or arrythmias. DS was seen in 16% of patients and was managed with the traditional approaches of steroids and hydroxyurea for cytoreduction. No grade ≥3 DS was noted.
Amir Fathi, MD:
Revumenib showed antileukemic activity with an ORR of 53% in the efficacy cohort and 52% in those who received the RP2D. Patients with KMT2A rearrangements and NPM1 mutations had good response proportions, with an ORR of 59% for those with KMT2A rearrangements and 36% with NPM1 mutations. In those who achieved CR/CRh, a large proportion were MRD negative: 73% and 100% in those with KMT2A and NPM1 aberrations, respectively.
The duration of response is quite promising at a median of 9 months, particularly in this heavily pretreated patient population.
Amir Fathi, MD:
Revumenib showed antileukemic activity with an ORR of 53% in the efficacy cohort and 52% in those who received the RP2D. Patients with KMT2A rearrangements and NPM1 mutations had good response proportions, with an ORR of 59% for those with KMT2A rearrangements and 36% with NPM1 mutations. In those who achieved CR/CRh, a large proportion were MRD negative: 73% and 100% in those with KMT2A and NPM1 aberrations, respectively.
The duration of response is quite promising at a median of 9 months, particularly in this heavily pretreated patient population.
Amir Fathi, MD:
Again, keeping in mind that these 60 patients were heavily pretreated, a median OS of 7 months is very promising.
Amir Fathi, MD:
Revumenib appears to exhibit antileukemic activity and fairly good tolerability in this early study. The risk of QT prolongation means patients need to be monitored closely, as does the potential for DS. Patients who had a KMT2A alteration seemed to have good response rates, as did patients with NPM1 mutations. The durability of response and survival needs to be studied in larger populations, but these are promising results, and I hope revumenib and other menin inhibitors will become an option for patients who have NPM1 or KMT2A alterations.
Eunice Wang, MD:
Some key differences between these 2 menin inhibitor trials is that AUGMENT-101 is a little further along, has enrolled more patients than KOMET-001, and included patients with ALL, as well as pediatric patients. In both studies, the CR/CRh rate was approximately 30%, again with the caveat of small patient numbers. DS was more common with ziftomenib in KOMET-001, whereas QTc prolongation was more common with revumenib in AUGMENT-101.
The results from these 2 studies validate the potential benefit from this drug class, and I anticipate that menin inhibitors will become part of the standard of care in this setting in the next few years. I look forward to seeing if combination regimens may enhance the efficacy of menin inhibitors, for example, combining with venetoclax plus azacitidine or an intensive backbone.
Eunice Wang, MD:
Now that FLT3 inhibitors are approved in AML as both monotherapy and in combination with intensive chemotherapy, efforts are underway to determine if FLT3 inhibitors can be combined with a less-intensive backbone, such as azacitidine and venetoclax.
This is a phase I/II trial of gilteritinib in combination with venetoclax/azacitidine in patients with FLT3-mutant AML who were unfit for intensive chemotherapy, high-risk MDS, or chronic myelomonocytic leukemia.12 This study enrolled 20 patients with R/R disease and 27 newly diagnosed patients. Patients received induction therapy comprising daily venetoclax for 28 days, azacitidine 75 mg/m2 for 7 days, and gilteritinib 80-120 mg on Days 1-28. The patients received up to 24 cycles of consolidation therapy with truncated dosing of the same regimen. The primary endpoints were the MTD of gilteritinib in phase I and the CR/CRi rate in phase II, with secondary endpoints including CR rate, MRD negativity, DoR, OS, and safety.
The patient population was mostly older, unfit patients who were ineligible for intensive chemotherapy. In the frontline cohort, 70% had a FLT3 ITD mutation, and 30% had a TKD mutation. The majority (67%) had a normal karyotype. Of note, 30% of patients in the R/R cohort had previously received a FLT3 inhibitor (10% had received gilteritinib), and 40% had prior treatment with an HMA/venetoclax.
Eunice Wang, MD:
For the overall responses in this study, we saw nearly universal CRs (92%) in patients treated in the frontline setting and a 100% modified composite CR rate (CR/CRi/MLFS). In the R/R cohort, 70% of patients had evidence of some clinical activity (CR/CRi/MLFS), with 35% of these responses being MLFS, 20% CR, and 15% CRi.
Of note, although 25% of patients had no response in the R/R setting, every patient in the frontline setting had some evidence of clinical efficacy.
Eunice Wang, MD:
Looking deeper at the response data, even in the R/R cohort—where one quarter of patients did not respond—the median OS was 5.8 months, with a median follow-up of 27 months. In the frontline cohort, the median RFS and OS have not yet been reached at a median follow-up of 12 months.
Patients who had no prior HMA/venetoclax and/or no prior gilteritinib in the R/R cohort did better than those who had been exposed to those agents, as one would expect. Those who had no prior exposure had a median OS of 10.5 months vs 4.8 months with prior exposure to these agents. Also, this regimen appeared effective in patients with both ITD and TKD mutations.
Eunice Wang, MD:
Overall, this combination was relatively well tolerated. In the R/R cohort, there was evidence of grade 3 febrile neutropenia (25%), grade 3-4 hypotension (15%), and grade 3-5 infection (55%). By contrast, in the frontline cohort, where patients likely had a better functional status, the AE profile was much milder. Only 1 patient had a grade 5 event (infection) compared with 4 patients in the R/R cohort.
Eunice Wang, MD:
There was evidence of myelosuppression with this triplet combination. In the frontline cohort, the absolute neutrophil count took a median of 39-43 days to recover to ≥1000 cells/µL in cycles 1 and 2 of consolidation. Likewise, platelets took a median of 30-33 days to recover to ≥100,000 cells/µL. This is only slightly longer than what is typically seen with standard venetoclax/azacitidine; 54% of patients in the frontline cohort required dose reductions vs 36% in the R/R cohort.
Eunice Wang, MD:
Overall, I think this triplet combination is extremely promising with good tolerability, safety, and universal response rates in the frontline cohort. Early data showed that the presence of a FLT3 ITD mutation could be a mechanism of resistance and a poor prognostic marker in patients treated with venetoclax/azacitidine alone. We also have seen that the combination of gilteritinib/azacitidine led to improved response rates but not necessarily OS. The triplet of a FLT3 inhibitor to overcome mechanisms of resistance added to the venetoclax/azacitidine backbone appears to be highly active for this particular mutation subset. I am looking forward to further clinical trials that will investigate exactly which dose strategies and durations of these 3 agents would be optimal for this patient population.
Dr Fathi, what are your thoughts on these data?
Amir Fathi, MD:
As you mentioned, there theoretically are 2 current combination regimens available to use in patients with FLT3 mutations, neither of which is ideal. For venetoclax/azacitidine, it seems that patients with FLT3 ITD mutations do not do as well as those with TKD mutations or those who are FLT3 wild-type.13 For gilteritinib/azacitidine, that study did not reach its endpoint, and the results will remain unclear, as this was not a placebo-controlled study, and patients in the azacitidine-alone arm could receive gilteritinib off protocol after stopping azacitidine.14
Now, the triplet of gilteritinib, venetoclax, and azacitidine has been studied, with the major concern to date being tolerability. As discussed, this can be managed by dose adjustments and decreasing the duration of venetoclax. With tolerability management, the triplet gives us the chance to incorporate gilteritinib into upfront therapy.
Amir Fathi, MD:
V-FAST is an ongoing multicenter, nonrandomized, multiarm phase Ib master protocol study of various combinations of liposomal cytarabine/daunorubicin (CPX-351) with targeted agents (venetoclax and midostaurin) for the treatment of newly diagnosed AML.15 At ASH 2022, McCloskey and colleagues15 presented results from a subgroup analysis of CPX-351 plus midostaurin in patients with FLT3 ITD or TKD mutations. The primary endpoints are determining the RP2D, safety, and dose-limiting toxicities. Secondary endpoints include responses and MRD negativity.
The patients in this study had a median age in the 60s and were predominantly male, and most had an ECOG PS of 0-1. In addition, the majority of patients had intermediate- or poor-risk disease, as well as de novo AML.
Amir Fathi, MD:
In general, the AEs seen on this study were as expected, with the majority of AEs being hematologic due to marrow suppression. For example, grade ≥3 febrile neutropenia was seen in 70% of patients, grade ≥3 leukopenia and thrombocytopenia in >50%, and neutropenia and anemia each in approximately 40%.
Other treatment-emergent AEs of interest included 1 case each of elevated bilirubin and decreased ejection fraction; 4 of 18 patients (22.2%) with FLT3 ITD mutations and 2 of 6 (33.3%) with FLT3 TKD mutations died, but none were considered related to treatment.
Amir Fathi, MD:
In terms of efficacy, the CR rate was quite high at 82%, and 50% of those with FLT3 ITD and 33.3% with FLT3 TKD achieved MRD-negative CR. Many patients, particularly in the ITD group (55%), were able to be bridged to stem cell transplant.
Amir Fathi, MD:
These interesting data suggest that the addition of midostaurin did not seem to worsen the AE profile of CPX-351. However, in terms of the efficacy, I’m not sure if this moves the needle that much. In general, the upfront treatment of FLT3-positive AML with combination strategies including FLT3 inhibitors does lead to very high levels of remission. I think additional advanced-phase studies are needed.
Dr Wang, what are your thoughts on these results?
Eunice Wang, MD:
The CR rates with the V-FAST backbone of CPX-351 added to midostaurin are approximately 83%, which is similar to the approximately 80% CR/CRi rates seen with the addition of a second-generation FLT3 inhibitor such as gilteritinib or crenolanib to standard 7+3 chemotherapy. I agree that enhancing the cytarabine/anthracycline backbone can increase CR rates, but it is at the expense of longer time for count recovery and myelosuppression. So, the question is how to improve upon the RATIFY data with 7+3 and midostaurin: Would it be preferable to intensify the cytarabine/anthracycline component or the FLT3 inhibition? I will be very curious to see the long-term outcomes of this combination of CPX-351 and midostaurin beyond just CR rates.
Back
