During an OncLive Peer Exchange® a panel of AML experts discussed these newly approved agents and provided insight on how they are rapidly changing the standard of care for various AML subgroups.
Harry Erba, MD, PhD
An estimated 19,520 new cases of acute myeloid leukemia (AML) are expected in the United States in 2018, representing approximately 1.1% of all new US cancer cases.1 Until 2017, there was little drug development in AML—treatment largely revolved around standard chemotherapy regimens, as it had for the past 40 years. By the end of 2017, however, there were 4 new drug approvals by the FDA , as well as many other agents showing promise in clinical trials (Table2-9). These treatment advances have been attributed to increased understanding of molecular drivers of AML. During an OncLive Peer Exchange® moderated by Harry Erba, MD, PhD, a panel of AML experts discussed these newly approved agents and provided insight on how they are rapidly changing the standard of care for various AML subgroups. They also provided insight on several promising agents on the horizon.All 4 newly approved AML drugs are targeted agents, but each targets the disease in different ways. “CPX-351 [Vyxeos] is the liposomal daunorubicin/cytarabine formulation that’s targeted to a clinically defined group of patients: those with treatment-related AML and AML with myelodysplasia-related changes. Then there are 2 drugs targeted to mutations: midostaurin [Rydapt], for patients with previously untreated AML with FLT3-ITD or FLT3-TKD mutations, and enasidenib [Idhifa], which is the oral IDH2 inhibitor approved for IDH2-mutated AML in relapsed or refractory disease. Finally, we have a different kind of targeted therapy, an antibody—drug conjugate against CD33: gemtuzumab ozogamicin [Mylotarg],” Erba said.
Midostaurin was approved in combination with chemotherapy based on data from the phase III RATIFY trial.2,10 In the study, 717 patients were randomly assigned to midostaurin (n = 360) or placebo (n = 357). Randomization was stratified by FLT3 mutation subtype: point mutation in the tyrosine kinase domain (TKD) or internal tandem duplication (ITD) mutation with a high ratio (>0.7) or a low ratio (0.05-0.7) of mutant to wild-type alleles. The FLT3-ITD subtype was ITD high in 214 patients, ITD low in 341 patients, and TKD in 162 patients. Compared with patients in the placebo group, those in the midostaurin group had significantly longer overall survival (OS) (HR for death, 0.78; one-sided P = .009) and event-free survival (HR for event or death, 0.78; one-sided P = .002). The benefit was consistent across all FLT3 subtypes.10
“This is the first agent, when added to chemotherapy, to really show a benefit,” Eunice Wang, MD, said, “and that was a 4-year overall survival benefit. Survival curves were distinct, and they remain distinct. In fact, when you look carefully at the curves, the difference in the overall survival curve seems to separate out shortly after induction and continue along the therapy during induction and consolidation."
Based on RATIFY findings, adding midostaurin has become the standard of care for newly diagnosed AML patients with FLT3 mutations who are receiving induction chemotherapy with anthracycline and cytarabine and in combination with consolidation. Midostaurin's benefit in a larger subgroup of AML patients has yet to be investigated, but it may be helpful because it also targets multiple tyrosine kinases. “[Midostaurin] was developed as a pan-tyrosine inhibitor,” Wang explained.
The panelists discussed 2 key considerations with use of midostaurin in clinical practice. First, the companion diagnostic test for FLT3-ITD and TKD mutations, which is a polymerase chain reaction (PCR) assay, might show a positive result, whereas next-generation sequencing might show a negative result. This is because technical challenges with sequencing can cause FLT3 mutations to be missed. “That’s a practical lesson here: A PCR-based assay should be done for FLT3 at the time of diagnosis because the next-generation sequencing panel may miss it,” Erba said.Second, the panelists warned about midostaurin toxicities and drug interactions. “One of the issues we’ve had practically is that it can cause QTc prolongation. It interacts with azoles. Part of that may be because it’s such a broad spectrum in pan-kinase inhibition,” Wang said. Dose reduction is rarely necessary, she noted, and interval assessments of QT by electrocardiogram are often sufficient in patients taking midostaurin concurrently with medications that can prolong the QT interval.
Enasidenib was approved based on Study AG221- C-001, a phase I/II, open-label, single-arm, multicenter trial that included 199 adults with relapsed or refractory AML who had an IDH2 mutation detected on the RealTime IDH2 Assay.3,11 Unlike FLT3 mutations, IDH mutations are not AML-defining. “We see them in MDS [myelodysplastic syndromes], we see them in MPNs [myeloproliferative neoplasms], and we see them in AML. That being said, there’s a significant fraction of patients with AML who have these mutations,” Alexander E. Perl, MD, said. There are 2 types of these mutations, he explained. One causes alpha-ketoglutarate to be converted to a metabolite called 2-hydroxyglutarate, or 2HG, which is known to contribute to leukemogenesis.
In Study AG221-C-001, patients with IDH2-mutated AML received enasidenib 100 mg daily.11 After a median follow-up of 6.6 months, 23% of patients achieved a complete response (CR) or CR with partial hematologic recovery (CRh). Of these patients, 19% had a CR lasting a median of 8.2 months and 4% had a CRh lasting a median of 9.6 months. The median time-tofirst response was 1.9 months (range, 0.5-7.5 months) and the median time-to-best response of CR/CRh was 3.7 months (range, 0.6-11.2 months). Additionally, of the 157 patients who required transfusions at the initiation of the trial, 34% were able to stop receiving transfusions during at least one 56-day time period on enasidenib. Of the 42 patients who did not require transfusions at the start, 76% maintained transfusion independence.11
A key concern with enasidenib is its toxicity profile, particularly regarding differentiation syndrome. “It occurred in about 10% or 20% of patients on the study that led to the approval of enasidenib—which, mind you, is a phase I/II study, so across all these doses, that’s how often they saw this,” Perl said. “It led to leukocytosis, it led to patients developing fevers, and it led to pulmonary infiltrates. In some patients, there was concern for cardiotoxicity in terms of pericardial effusions or pleural effusions, and it was steroid responsive.” It is important for clinicians to recognize these risks and be cautious when initiating enasidenib in patients with pre-existing lung conditions, he noted.
“We need to recognize that [such symptoms] may be an expected effect and may be signaling response,” Jorge E. Cortes, MD, said, noting that it may take some time for patients to show response to treatment. “This is not chemotherapy where if you’re not in remission in the first cycle, you’re not going to respond. Some responses take months before you start seeing something, and even then, [patients] may continue to improve. It may look [like just] a stable disease or a partial response, but they will get into a complete remission.”
CPX-351 was approved based on an improvement in OS observed in a phase III study (NCT01696084) that randomly assigned 309 patients with newly diagnosed therapy- related AML (t-AML) or AML with myelodysplasia-related changes (AML-MRC) to receive CPX-351 (n = 153) or separately administered treatments of daunorubicin and cytarabine (ie, 7 + 3 chemotherapy; n = 156).4,12 Both t-AML and AML-MRC are high-risk forms of AML that are generally associated with a poor prognosis. After a minimum follow-up of 13.7 months, CPX-351 showed superior OS (median, 9.56 vs 5.95 months; HR, 0.69; P = .005). The 60-day mortality also favored CPX-351 (13.7% vs 21.2%).
The panelists noted that it can be difficult to identify some patients with AML-MRC who would truly benefit from CPX-351 versus standard chemotherapy—the pathology report is subjective and the 2016 revisions to the World Health Organization classification of myeloid neoplasms and acute leukemia added nuances that further confound the issue.
Nevertheless, the panelists agreed that there would be no harm in giving CPX-351 to patients who meet current label requirements, even if their next-generation panel ultimately shows them to have favorable risk. “You’re not going to do a disservice to these patients. The randomized phase II studies—both the frontline and the salvage settings—show that there was a trend for survival for the overall population, and it was significant for secondary AML. So, you can say that, at the very least, it is equivalent to 7 + 3 chemotherapy,” Cortes said.
CPX-351 has shown a good toxicity profile versus chemotherapy. “If you looked at the rates of infections during aplasia, they were higher in the CPX-351 arm, and the duration in years was longer,” Perl said, "and yet, other toxicities were lesser. These patients seemed to, at least in our limited experience, get less mucositis. They don’t lose their hair. It’s a different pattern of toxicity.” It remains unclear, however, whether there is less cardiotoxicity with CPX-351, which is an important consideration as patients go into transplant.
Gemtuzumab ozogamicin originally received accelerated approval from the FDA in 2000 for CD33-positive AML in first relapse in patients aged ≥60 years who were not candidates for cytotoxic chemotherapy.13 However, it was voluntarily withdrawn from the US market in 2010 after a phase III confirmatory trial did not show clinical benefit and reported a significantly higher rate of treatment-related toxicity, including a high number of early deaths.5 The drug’s current FDA approval is based on studies that have shown clinical benefit and less toxicity with lower doses.
“One of the first studies that came out was a French study that looked at 3 doses of 3 mg/m2 given on days 1, 4, and 7 of 7-and -3 chemotherapy induction. That study did not show an increased induction mortality but showed better protection against relapse, ultimately leading to an improvement in event-free survival, which is the primary endpoint of the study and actually, overall survival was a secondary endpoint, but it was still met,” Perl said.14
He also discussed a meta-analysis by the UK National Cancer Research Institute that showed a statistically significant reduction in relapse rates and improved OS for gemtuzumab ozogamicin—treated patients across all available phase III data in adults, without appreciable increases in toxicity.15 “The adverse karyotype patients did not show a benefit, and most of the benefit was seen in the patients with core-binding factor disease, but a small benefit was also seen in the intermediate- risk group. Overall, it was better than not getting it, so that’s what led to the approval of [gemtuzumab ozogamicin],” Perl explained.
When Erba asked the panelists if they were using gemtuzumab ozogamicin for their patients, which at the time of the Peer Exchange was about 3 months post approval, they all said yes. “I’ve used it 3 times,” Perl said, noting he had used it for a patient with acute promyelocytic leukemia and for 2 patients as salvage therapy.
“We are definitely recommending that favorable karyotype patients—the young and fit patients—get gemtuzumab ozogamicin added on to standard 7-plus-3 chemotherapy,” Wang said. “We’ve also talked about using it in intermediate- cytogenetic young and fit people, if they are not FLT3-positive, to try to give them an advantage.” Wang has also offered it as a single agent to elderly patients who do not want to come in 7 or 10 days in a row to receive treatment. Like Perl, Cortes said he was using it predominantly for salvage. “Our frontline [core-binding factor] study has always included [gemtuzumab ozogamicin] because on a clinical trial, it was available even when it was taken out of the market,” he said. “Outside of a clinical trial setting, I do use it for salvage in the same patient population and for older patients in particular.”The panelists discussed several promising agents in clinical trials, including 3 very potent FLT3-targeted agents that are more selective than midostaurin. These include quizartinib, gilteritinib, and crenolanib. Among these, quizartinib is the most selective and not as protein bound, which could be beneficial, Cortes said. “Particularly in phase II, it showed very nice activity. Patients with FLT3 mutations showed around 50% response. Most of these were CRis [complete remission with incomplete hematologic recoveries]. There were a few CRs in that, but most of these were CRis, so there was an incomplete recovery of platelets and/or neutrophils,” Cortes said.16
Following the Peer Exchange, quizartinib showed promising results in the phase III QuANTUM-R study, where it is being assessed in the salvage setting.17 In the study, singleagent quizartinib significantly prolonged OS versus salvage chemotherapy and enabled some patients to reach the transplant stage.
Gilteritinib and crenolanib, both of which are type 1 tyrosine kinase inhibitors, were developed to overcome mutations in FLT3 that emerged during treatment with other FLT3-directed therapies and conferred resistance. “They have slightly different mutations that they have specificity for, and in this case, gilteritinib and crenolanib both have activity against ITD and D835 resistance mutations, which is potentially an advantage over quizartinib,” Perl said.
Wang proceeded to discuss the results of a small phase II study, which she presented at the 2017 American Society of Hematology Annual Meeting that showed high response rates with the combination of crenolanib and cytarabine anthracycline-based chemotherapy.18 “Overall, there were CR/CRi rates of about over 80% after 1 or 2 cycles. Overall survival was at a median of about 17-plus months at about 79%. There was a cumulative rate of relapse of about 16%. Now, there was a very small number of patients—we’re talking about 29 patients—but those numbers, at least initially, compare favorably with some of the RATIFY data,” Wang said. In December 2017, the FDA granted fast track designation for crenolanib for patients with FLT3 mutation-positive relapsed or refractory AML.19
The panelists did not discuss related studies, but in April 2018 gilteritinib’s manufacturer, Astellas Pharma, announced that it submitted a new drug application to the FDA seeking approval as a treatment for adult patients with FLT3 mutation—positive relapsed or refractory AML.20 Submission was based on data from the ongoing phase III ADMIRAL study,21 which have yet to be published.
Ivosidenib (AG-120) has shown activity against IDH1 mutations, which are observed in 6% to 10% of patients with AML. In phase I study results published in early June, ivosidenib monotherapy induced an overall response rate of 41.6% (95% CI, 32.9-50.8) in 125 patients with IDH1-positive relapsed/refractory AML.22 The rate of CR or CRh was 30.4% (95% CI, 22.5-39.3), including 21.6% CR. The FDA is evaluating ivosidenib in this setting under its priority review program, with a decision expected by August 21, 2018.The panelists expressed considerable optimism about the potential for improved outcomes because of the number of agents recently approved as well as many others poised for approval based on favorable clinical trial data. Having multiple targeted agents and other treatments available finally enables a personalized approach to care in AML, they noted.
“I think that we have struggled with the complexity and the heterogeneity, both of our patients and the biology of the disease, for many years. We’re finally beginning to crack that,” Wang said. “I’m excited to see that we actually have active agents we can play with, layering them together and really looking at the patient, looking at the disease, and figuring out which thing to take from each category to make that patient live the longest.”
Courtes added, “It is much more hopeful for our patients because we are starting to see some improvement in outcomes in areas where we had not before.” However, he noted, considerable work still needs to be done across all AML subgroups, even those for whom multiple recently approved therapies are now available. He encouraged clinicians to continue enrolling patients in clinical trials to keep the field moving forward.