How I treat acute myeloid leukemia with differentiation therapy

Visual Abstract

Abstract

An increasing number of acute myeloid leukemia (AML) therapeutics have been developed, not as cytotoxic therapies but rather as targeted agents able to restore the aberrant and leukemogenic “block” in normal differentiation. All-trans retinoic acid and arsenic trioxide are classic examples of differentiating agents for treatment of acute promyelocytic leukemia (APL); newer therapies functioning through differentiation include isocitrate dehydrogenase 1 and 2 inhibitors, FMS-like tyrosine kinase 3 inhibitors, and menin inhibitors. The terminal differentiation of leukemic blasts via differentiating-agent therapy can lead to a constellation of signs and symptoms, originally referred to as “retinoic acid syndrome” and now termed “differentiation syndrome” (DS), characterized predominantly by systemic inflammatory response system–like features of dyspnea, pulmonary infiltrates, pleural and pericardial effusions, unexplained fevers, hypotension, edema, and renal insufficiency. DS in patients with newly diagnosed APL is generally straightforward to identify; however, DS in patients with multiply relapsed AML can be more challenging to diagnose, due to nonspecific signs and symptoms that can be mistakenly attributed to infectious etiologies or the underlying refractory leukemia itself. Prompt consideration of DS, rapid initiation of systemic corticosteroids, and early cytoreduction in the setting of concomitant hyperleukocytosis are essential for optimal management.

Treatment for patients with acute myeloid leukemia (AML) is being transformed for some patients, as targeted agents increase complete remission rates and survival when appropriately applied. Although the efficacy of newer regimens has been established in major clinical trials, use in routine practice is often challenging, and careful choices need to be made. Associate Editor Selina M. Luger introduces this case-based How I Treat series designed to provide guidance in those circumstances. Wei et al discuss how to maximize the benefits when the venetoclax-azacitidine regimen was indicated, while Issa and colleagues discuss differentiation therapy, highlighting how the clinical course for patients varies from what is seen with intensive chemotherapy-based induction regimens. Green and Wang tackle the challenge of managing patients with secondary AML. Finally, Roboz and Canaani outline how maintenance can benefit selected patients not undergoing allogeneic transplant, while exploring the many unanswered questions that arise in clinical practice.

Introduction

Over the past decade, there have been multiple new drug approvals to treat acute myeloid leukemia (AML).¹ Although the original nonselective cytotoxic agents such as cytarabine and anthracyclines maintain an important role in AML therapy, an improved understanding of the biological basis of myeloid malignancies has led to rationally designed AML therapeutics targeting specific molecular drivers of disease.^{2, 3, 4, 5, 6, 7, 8} Many of these newer agents exert their antileukemic effects through direct and selective inhibition of leukemia-promoting aberration(s), resulting in the release of the pathogenic “differentiation block” and allowing for myeloid maturation of leukemic blasts.⁹ These agents, generally, have less myelosuppression and fewer standard chemotherapeutic side effects such as mucositis and alopecia. However, an increasingly recognized adverse event associated with these therapies is differentiation syndrome (DS).

DS, initially termed all-trans retinoic acid (ATRA) syndrome or retinoic acid syndrome, was originally described in patients with acute promyelocytic leukemia (APL) receiving therapy with ATRA or later arsenic trioxide (ATO).^10,11 It was recognized to occur in the setting of neutrophil recovery, with characteristic findings of leukocytosis, culture-negative fevers, hypotension, dyspnea, pleural and pericardial effusions, weight gain, peripheral edema, and acute renal failure.^{11, 12, 13} Retinoic acid syndrome typically occurs 1 to 2 weeks into ATRA and/or ATO therapy, with an incidence reported anywhere from 3% to 37%.^{10,14, 15, 16} This broad range of incidence results from inconsistent diagnostic criteria and nonspecific signs and symptoms of this syndrome, as well as variable use of concurrent cytotoxic chemotherapy and prophylactic corticosteroids, which likely serve to decrease DS incidence and severity (Table 1).

Table 1.

Corticosteroid dosing strategies for DS prophylaxis in APL

APL treatment	DS prophylaxis	Indication	Source
ATRA + ATO vs ATRA + Ida (APL0406)	Prednisone 0.5 mg/kg daily during induction	All patients	Lo-Coco et al13
ATRA + Ida (LPA99)	Prednisone 0.5 mg/kg daily for 15 d	All patients	Sanz et al17
AIDA (LPA2005)	Prednisone 0.5 mg/kg daily for 15 d	WBC >5 × 109/L	Sanz et al18
ATRA + DNR (mitoxantrone + ATRA second cycle) (IC-APL)	Dexamethasone (2.5 mg/m2 per 12 h IV for 15 d)	WBC >5 × 109/L	Rego et al19
ATRA + DNR ± AraC	High-dose dexamethasone	WBC >10 × 109/L	Adès et al20
ATRA + Ida + ATO (APML4)	Prednisone 1 mg/kg daily for at least 10 d	All patients	Iland et al21
ATRA + ATO ± GO	Methylprednisolone 50 mg daily for 5 d followed by rapid taper	All patients	Ravandi et al22,23
ATRA + Ida (LPA96)	Dexamethasone 10 mg twice daily	WBC >5 × 109/L	Sanz et al15
ATRA + Ida (ICC-APL-01)	Dexamethasone 5 mg/m2 for 5 d	WBC >10 × 109/L	Testi et al24
ATRA + ATO ± Ida (AAML1331)	Dexamethasone 2.5 mg/m2 for 14 d	WBC >10 × 109/L	Kutny et al25

AIDA, ATRA with idarubicin; AraC, cytarabine; DNR, daunorubicin; Ida, idarubicin.

Case 1

A 50-year-old Hispanic woman, with no significant past medical history, presented to the emergency department with severe vaginal bleeding. She had been having increasing fatigue with bruising on her arms and legs. Her white blood cell (WBC) count was 1.6 × 10⁹/L, with 85% blasts, a hemoglobin of 11.4 g/dL, and platelet count of 23 × 10⁹/L. Her fibrinogen level was 135 mg/dL, and a diagnosis of APL was suspected based on evidence of disseminated intravascular coagulation (DIC) and after review of the peripheral smear based on morphology of the blasts with presence of Auer rods. ATRA was initiated and fluorescence in situ hybridization confirmed the presence of PML::RARA rearrangement. She enrolled on protocol (NCT01409161) and was started on the combination of ATRA and ATO for low-risk APL. For DS prophylaxis, she was given methylprednisolone 50 mg IV daily for 5 days, followed by tapering doses starting on day 6. During the second week of the cycle, she developed worsening peripheral edema and dyspnea, along with an increase in WBC to 15 × 10⁹/L. A diagnosis of DS was made, and she received dexamethasone 10 mg IV twice daily with 1 dose of gemtuzumab ozogamicin (GO) at 6 mg/m², along with supportive measures (supplemental 02 and diuresis) leading to resolution of the signs and symptoms of DS within 2 to 3 days. Afterward, steroid doses were tapered.

Mechanisms

The underlying pathogenesis of DS remains incompletely characterized but is thought to stem from exuberant cytokine production in differentiating and maturing leukemic cells, leading to a hyperinflammatory state analogous to systemic inflammatory response system, associated with vascular capillary leak, and increased organ infiltration by maturing cells^12,14 (Figure 1A). Most of our knowledge on the mechanisms underlying DS stems from studies of ATRA and/or ATO therapy in APL. Implicated inflammatory cytokines include interleukin-1 (IL-1), IL-6, IL-8, tumor necrosis factor alpha, and chemokine (C-C motif) ligand 2 (CCL2) or release of the serum protease cathepsin G.^{26, 27, 28} Increased expression of beta-2 integrins has additionally been implicated in DS pathophysiology, leading to increased adhesion of leukemic blasts to the endothelium and subendothelial matrix, and facilitating subsequent migration and extravasation into various tissues.^{29, 30, 31} Lung histology sections in patients with DS confirm interstitial infiltration with maturing granulocytic cells, small vessel inflammation, and diffuse alveolar damage with intra-alveolar hemorrhage.³²

Figure 1.

Mechanisms of DS. (A) In some cases, differentiation of leukemia cells is associated with a cytokine imbalance leading to a clinically evident DS. (B) Various phenotypes of differentiation in leukemia after treatment have been identified, evident in peripheral blood counts or posttreatment bone marrow analyses, with terminal differentiation to mature cells being the best characterized. Other phenotypes are less understood. SIRS, systemic inflammatory response system.

Diagnosis

The accurate diagnosis of DS remains a significant challenge, because signs and symptoms of DS are nonspecific. The most common manifestations include pulmonary findings (dyspnea, hypoxia, pulmonary infiltrates, and pleural and/or pericardial effusions), weight gain, culture-negative fevers, hypotension, and renal insufficiency, initially described by Frankel et al (Figure 2).¹² Although these manifestations are certainly the most recognizable based on multiple publications describing DS in APL, other less common manifestation of DS have been described, and characterization of DS associated with the latest differentiating agents such as menin inhibitors has not been fully elucidated to date (Figure 3B). Laboratory findings include leukocytosis, often with a rise in neutrophil count and concurrent decrease in blast percentage, although this is not always seen (Figure 1B), and therefore, leukocytosis is not considered diagnostic of DS by most criteria. In some DS cases associated with isocitrate dehydrogenase (IDH) or menin inhibitors, leukocytosis can be predominantly monocytic, making the distinction of DS vs leukemia progression even more challenging (Figure 1B). Imaging findings are similarly nonspecific, with radiographic findings on chest radiograph showing effusions and infiltrates in most patients (Figure 3B).

Figure 2.

DS criteria for diagnosis and for grading of severity.

Figure 3.

Characteristics of DS by therapy. (A) Reported frequency and severity of DS by targeted therapy. (B) Reported manifestations of DS. bili, bilirubin; Ena, enasidenib; Gilt, gilteritinib; I, inhibitor; Ivo, ivosidenib; manif, manifestations; Olu, olutasidenib; Quiz, quizartinib; resp, respiratory; Rev, revumenib. Data sources for the following drugs: ATRA/ATO (Montesinos et al,¹⁰ Lo-Coco et al,¹³ and Woods and Norsworthy⁵⁷); enasidenib (Stein et al,⁵ Fathi et al,³⁴ Norsworthy et al,³⁸ and FDA package insert³⁵); ivosidenib (DiNardo et al,⁴ Norsworthy et al,³⁸ and FDA package insert³⁶); olutasidenib (de Botton et al³³ and FDA package insert³⁷); gilteritnib and quizartinib (FDA package insert^42,43); and revumenib (Issa et al⁴⁹ and Aldoss et al⁵⁰).

When the above signs and symptoms occur 10 days into a newly diagnosed patient with APL receiving ATRA and ATO therapy, an accurate diagnosis of DS is relatively straightforward (as illustrated in case 1), so long as the provider is trained to recognize the nonspecific constellation of signs and symptoms accurately as DS based on APL diagnosis and time line of differentiating-agent treatment initiation. However, when these manifestations develop 6 weeks into salvage therapy of a multiply refractory AML with FMS-like tyrosine kinase 3 (FLT3) and IDH1 mutations receiving an IDH1 or FLT3 targeted agent, they can often be mistakenly assumed to be due to an undiagnosed infection or persistent/progressive disease, both of which are equally plausible etiologies (and may even be occurring in tandem; see “Case 2”). Thus, accurate diagnosis requires both an awareness of DS and a high index of suspicion when these nonspecific findings are seen. Any true confirmation of DS can only be established in retrospect: after observing the rapid resolution of signs and symptoms with corticosteroid therapy and supportive care, highlighting that attention to diagnosing and treating other potential underlying etiologies such as occult infections or congestive heart failure exacerbations (especially common in leukemia patients due to age, cardiac comorbidities, malnutrition, and frequency of fluid shifts with frequent transfusions) is particularly imperative.

The most recognized and used diagnostic algorithm for DS includes a grading system proposed by Montesinos et al based on the Programa de Estudio y Tratamiento de las Hemopatías Malignas (PETHEMA) studies of ATRA and anthracycline therapy for APL (Figure 2).¹⁰ In this scoring system, patients with ≥4 signs or symptoms of DS (dyspnea, unexplained fever, weight gain ≥5 kg, unexplained hypotension, acute renal failure, and chest radiograph with pulmonary infiltrates or pleuropericardial effusion) are considered to have severe DS, and ≤3 constitute moderate DS. Although this algorithm has not yet been formally validated with DS associated with other targeted AML therapeutics, it is reasonable to use this grading system for consistency across cohorts and trials. Future studies could focus on establishing diagnostic criteria specific to the most recent differentiating agents with applicable severity scoring systems.

Case 2

A 72-year-old Caucasian man with a past medical history of hypertension and diabetes presented to clinic for an evaluation of AML that progressed after treatment with azacitidine and venetoclax. A bone marrow biopsy revealed AML with myelodysplastic-related changes, with 50% blasts and mutations in DNMT3A, IDH1, and KRAS. His WBC was 1.6 × 10⁹/L, with 15% peripheral blasts, a hemoglobin of 6.8 g/dL, and a platelet count of 43 × 10⁹/L. He was started on treatment with ivosidenib at a dose of 500 mg once daily. Three weeks after start of ivosidenib, he presented to the emergency department with shortness of breath and a pleuritic chest pain. His laboratory studies revealed a WBC of 11.2 × 10⁹/L, with 47% peripheral blasts, 35% monocytes, 9% lymphocytes, 3% neutrophils, and a platelet count of 25 × 10⁹/L. His oxygen saturation on pulse oximetry was 88%, requiring oxygen supplementation with a nasal cannula. A computed tomography of the chest showed diffuse bilateral ground glass airspace opacities predominantly in the dependent portions of the lungs, with a small right pleural effusion. He was admitted to the hospital, received red blood cell transfusion and supplemental 02, and was started on empiric antibiotics for pneumonia, diuretics, and dexamethasone 10 mg IV twice daily for suspicion of DS. Ivosidenib was held, and his breathing and oxygenation improved in the following days, after which he was discharged from the hospital with tapering doses of steroids and ongoing ivosidenib at 500 mg once daily. His end of cycle bone marrow biopsy showed decrease in blasts to 23% but now predominantly monocytic, with improvement in his platelet count to 56 × 10⁹/L, a WBC of 3.4 × 10⁹/L, and a hemoglobin of 9.1 g/dL. Treatment with ivosidenib was continued.

IDH inhibitor–associated DS

Targeted mutant IDH1 and IDH2 inhibitors are rationally designed small molecules that bind within the IDH enzymatic active site, blocking aberrant 2-hydroxyglutarate production and inducing myeloid differentiation.^4,5,33 DS has been reported to occur in 12% to 19% of patients with IDH1 or IDH2 mutations receiving ivosidenib, olutasidenib, or enasidenib, with most cases moderate in severity and responsive to corticosteroids, hydroxyurea, and supportive care (Figure 3A). However, fatal cases of DS have been reported with all 3 inhibitors, and all 3 carry the US Food and Drug Administration (FDA) black box warnings due to the risk of DS.^{4,33, 34, 35, 36, 37}

The median time to development of IDH inhibitor–related DS is longer than with APL, with a median time of DS occurrence of 17 to 20 days, although some cases have been reported to develop as late as 150 days with enasidenib and 561 days into therapy with olutasidenib (Figure 3A).^33,38 It is of particular interest to recognize that the development of DS is not clearly associated with clinical response to IDH inhibitors, as illustrated in case 2.^10,34 This unpredicted lack of association between DS and clinical benefit may be due to the presence of multiple leukemia clones at relapse with nondifferentiating/nonresponding clones preventing response or possibly an unsuccessful terminal differentiation process leading to a more monocytic but still dysfunctional leukemia clone (Figure 1B).

The half-life of IDH inhibitors is long: 4 days, 2.8 days, and 8 days for ivosidenib, olutasidenib, and enasidenib, respectively. For this reason, although treatment discontinuation is recommended in cases of severe DS, as discussed in more detail below, it is especially important to realize that washout time lines will take several days with IDH-directed therapy, and corticosteroids remain the mainstay of immediate therapy (Figure 4).

Figure 4.

Guidelines for monitoring and treatment of DS. Ara-C, cytarabine; CBC, complete blood count; EKG,electrocardiogram; LDH, lactate dehydrogenase; LFTs, liver function tests; PT, prothrombin time; PTT partial thromboplastin time; TLS, tumor lysis syndrome; TTE, transthoracic echocardiogram.

FLT3 inhibitor–associated DS

The incidence of DS with FLT3 inhibitors is less frequent than with ATRA or IDH inhibitors, estimated at 1% to 5% across the various FDA–approved FLT3 inhibitors.³⁹ Cases of DS occurring in the setting of FLT3 inhibitors also may be more likely to present with rash, with several reports of Sweet syndrome (neutrophilic dermatoses) with these therapies in the literature.^40,41 Similar to IDH-DS, symptom onset of DS related to FLT3 inhibitors is often weeks to months from treatment initiation (Figure 3A). There have been no reported cases of DS occurring with midostaurin, likely related to the use of midostaurin in combination with intensive chemotherapy, in which DS is minimized. Reported cases of DS related to gilteritinib and quizartinib occur relatively infrequently, described in ∼3% to 5% of patients receiving monotherapy for relapsed or refractory disease (Figure 3A).^42,43 In a review of the 11 patients who experienced DS in clinical studies of gilteritinib, time to DS onset was 2 to 75 days, and 9 patients recovered; 1 fatal episode of DS with gilteritinib was reported in the pivotal ADMIRAL trial, leading to an FDA black box warning (Figure 3A).^3,42,44

Of interest, an analysis of patients treated with quizartinib monotherapy identified 2 distinctly different categories of response, via either cytotoxicity or differentiation, and a similar dichotomous pattern of response was also identified with gilteritinib.^44,45 Differentiation responses were seen more frequently in patients with NPM1 and/or DNMT3A comutations.⁴⁵

Menin inhibitor–associated DS

Small molecule inhibitors of the protein-protein interaction between menin and lysine methyltransferase 2A (KMT2A) induce myeloid differentiation in acute leukemias dependent on this interaction, with initial preclinical and then clinical development focusing on AML harboring KMT2A rearrangements (KMT2Ar) or NPM1 mutations.^{46, 47, 48} It is likely that the first of this new class of small molecules will be approved for the treatment of relapsed and refractory KMT2Ar acute leukemias as early as the fall of 2024. Similar to IDH and FLT3 inhibitors, the promotion of myeloid differentiation in patients treated with menin inhibitors can lead to DS.⁴⁹ In a pivotal phase 2 study of the menin inhibitor revumenib (formerly SNDX-5613), all-grade DS was 28% and grade ≥3 DS was 16%; no grade 5 DS events were recorded.⁵⁰ Phase 1 studies of ziftomenib (formerly KO-539) and JNJ-75276617 (Janssen) had all-grade DS incidences of 12% and 20%, respectively. Of note, 1 grade 5 DS event has been reported with both ziftomenib and JNJ-75276617.^51,52 Because of the difficulty of crosstrial comparisons, especially with phase 1 studies, it is unknown whether the true rates and severity of DS differs between the various menin inhibitors in clinical trials. Similarly, whether the severity of DS in patients with menin inhibitors will be attenuated in combination therapies, and whether the severity differs from other differentiating agents, or by the targeted leukemia genotype, remain unknown. Additional research is needed to better understand and predict the patterns of response and differentiation after treatment with menin inhibitors to accurately distinguish response with differentiation, residual disease, and leukemia progression (Figure 1B).

Others

Many additional small molecules and targeted therapeutics in clinical development involve a mechanism of action including differentiation of leukemic blasts, in which the development of DS is to be expected. One such agent, FHD-286, a chromatin modifier and a first-in-class Brahma-related gene 1/Brahma homologue (BRG1/BRM) inhibitor, has reported an incidence of DS of 15% and, notably, was placed on a full clinical hold by the FDA in 2022 due to reported fatal cases of DS.⁵³ With adequate safety and monitoring plans in place, this clinical trial has reopened with FHD-286 in combination with hypomethylating agent therapy.⁵³ A second agent, iadademstat (formerly ORY-1001), a first-in-class LSD1 inhibitor in clinical development, has also been associated with DS, with fatal DS described in 1 of 27 treated patients.⁵⁴

Additionally, the hypomethylating agents (ie, azacitidine and decitabine) have themselves been reported to cause DS.^39,55 Of interest, in the randomized AGILE study of azacitidine and ivosidenib combination vs azacitidine and placebo, there was an 8% reported incidence rate of DS in the azacitidine and placebo arm (compared with 15% in the combination arm).⁵⁶

Treatment

The mainstay and most effective treatment for DS is systemic corticosteroids (Figure 4). Prompt initiation of IV or orally administered dexamethasone at a dose of 10 mg twice daily, for at least 3 days and then tapered once symptoms resolve, is recommended upon any suspected diagnosis of DS. Although DS can prove rapidly fatal without therapy, prompt initiation of corticosteroids is effective and associated with mortality rates of <1%.⁵⁷

Leukocytosis is common with DS, occurring in >50% of patients, and corticosteroids can further elevate the WBC count. We strongly recommend concurrent cytoreductive management with hydroxyurea, GO, and/or cytarabine whenever the WBC count is >25 × 10⁹/L (or >10 × 10⁹/L for APL), to minimize additional complications of leukostasis including DIC. We note that although the use of GO is often prioritized in APL for this situation, the role and effectiveness of GO in DS with leukocytosis occurring with novel targeted agents is not well characterized.^58,59

We recommend treatment discontinuation of the targeted therapy in cases of severe DS (ie, hospitalization, need for supplemental oxygen, hemodynamic instability, or leukocytosis >50.0 × 10⁹/L) until DS symptoms resolve. As mentioned above, it is important to recognize that many of the targeted therapies have a long half-life, especially ivosidenib, gilteritinib, and enasidenib at 4, 5, and 8 days, respectively, and thus, treatment discontinuation is not anticipated to lead to timely clinical improvement, and again the importance of frontline corticosteroids must be stressed.

Accumulating evidence demonstrates that treatment with the DS-associated targeted therapy can generally be reinitiated at the original dose once DS has improved. This is consistent with the literature in APL suggesting the risk of DS disappears once the patient has obtained a complete remission; however, this can be more complicated in relapsed patients with various leukemic clones receiving a targeted inhibitor as monotherapy, because some clones may differentiate and respond faster or more successfully than other more resistant subclones. Indeed, DS recurrence in IDH-mutated patients has been reported in a minority of patients experiencing IDH-DS, with a median of ∼3 weeks between the original and secondary episodes, suggesting slower steroid tapers may be of benefit in patients with persistent or severe symptoms (Figure 3A).³⁸

Additionally, and as mentioned in detail previously, simultaneously treating any possible or suspected concurrent conditions such as infection or congestive heart failure exacerbation is essential, because many patients with relapsed or refractory AML may have multiple competing etiologies underlying their clinical picture.

DS prophylaxis

Prophylactic administration of corticosteroids to prevent or minimize DS is often used in APL treatment regimens, based on retrospective analyses demonstrating decreased DS-related incidence, severity, and complications, despite a lack of clear impact on DS-related mortality.^{11,13,15,22,60} Whether to administer prophylactic corticosteroids for the first 1 to 2 weeks or the entirety of the first month and whether to reserve prophylactic steroids only for patients experiencing leukocytosis or administer to all patients presenting with APL are debated. Current expert opinion for APL management recommends the use of prophylactic corticosteroids (prednisone 0.5-1 mg/kg daily or dexamethasone 10 mg twice daily) in patients at high risk of DS, defined as patients with APL with a WBC count >5 × 10⁹/L to 10 × 10⁹/L or elevated creatinine/renal insufficiency (>1.4 mg/dL; Table 1).^10,18

Due to the extended time frame of DS development in patients with relapsed leukemia receiving targeted therapies (ie, months, not days or weeks), and the increased prior duration of neutropenia and heightened risk of fungal and other infections, prophylactic corticosteroids are not currently recommended outside of APL. We recommend that leukocytosis, however, be treated with cytoreduction before treatment initiation with IDH, FLT3, menin inhibitors, or other differentiating targeted agents to minimize the incidence and severity of DS and leukostasis complications including DIC, given the association of DS with burden of disease.^10,34

In summary, anticipation and early recognition of DS associated with targeted therapies in AML is critical for the successful delivery of these agents. DS prevention should also include informing patients of the possible manifestations of DS, with heightened awareness during the first 2 months of differentiating-agent therapy, and the early initiation of corticosteroids for DS-related signs and symptoms. Carefully designed clinical and translational studies are needed to better illuminate the underlying mechanisms of DS and determine optimal future diagnostic and therapeutic strategies to further decrease DS-related morbidity and mortality.

Conflict-of-interest disclosure: G.C.I. reports research funding from Celgene, Merck, Kura Oncology, Syndax, Astex, and Novartis; and received consultancy or advisory board fees from NuProbe, AbbVie, Novartis, Sanofi, AstraZeneca, Syndax, and Kura Oncology. E.M.S. reports grants from Eisai and Bristol Myers Squibb (BMS); consulting fees from Novartis, PinotBio, Janssen, BMS, Agios Pharmaceuticals, Jazz Pharmaceuticals, Menarini, Genentech, Genesis, AbbVie, Neoleukin Corporation, Gilead Sciences Inc, Syndax Pharmaceuticals Inc, OnCusp Therapeutics, CTI BioPharma, Foghorn Therapeutics, Servier Laboratories, Calithera Biosciences, Daiichi Sankyo, Aptose Biosciences, Syros, Syndax Pharmaceuticals, Inc, Astellas Pharma, Ono Pharma, and Blueprint Medicines; and participation in advisory boards for Epizyme Inc and Cellectis. C.D.D. reports research funding from AbbVie, Astex, Immune-Onc, BMS, Cleave, Foghorn, Loxo, Rigel, and Servier; consulting fees from Amgen, AbbVie, Astellas, BMS, Genmab, GlaxoSmithKline, Gilead, Jazz, Schrodinger, Servier, and Stemline Therapeutics; honoraria for educational events from AbbVie, Astellas, BMS, Jazz, and Servier; meeting support from Servier; and has participated on a Genmab data safety board.