Profile of Glasdegib for the Treatment of Newly Diagnosed Acute Myeloid Leukemia (AML): Evidence to Date

Introduction

Acute myeloid leukemia (AML) is an aggressive and biologically heterogeneous hematologic malignancy primarily affecting older adults. Standard of care intensive induction chemotherapy consists of a 7-day continuous infusion of cytarabine at 100–200 mg/m² per day on days 1 to 7 and daunorubicin at 60–90 mg/m² per day on days 1 to 3 (“7+3”), and can induce complete response (CR) rates as high as ≥80% with a 5-year overall survival (OS) of ~40–50% in younger patients without adverse cytogenetic or molecular risk factors.^1–5 Despite encouraging response rates in young and fit patients, AML is typically diagnosed at a median age of 68 years in the United States, with 1/3 of newly diagnosed patients being ≥75 years of age.^6,7 Although response rates in the elderly and unfit patient population have improved in the era of the hypomethylating agents (HMAs), 5-year survival remains <10% in patients over 65 years old and rates of cure remain low.^1,8–11 The reason for poorer outcomes in older and unfit patients is multifactorial, owing to inability to tolerate intensive therapy, more deleterious genetic changes leading to reduced response rates and increased incidence of relapse, increased comorbidities, and ineligibility for curative allogeneic hematopoietic stem cell transplant (HSCT).^1,5,8,12 Prior to the advent of the HMAs, the established non-intensive agent most widely used in clinical practice was low-dose cytarabine (LDAC). Although toxicity was low, response rates were a modest 7–18% with median OS of 5 months. In 2018 based on the preliminary results of the Phase II Bright AML 1003 trial, the Smoothened (SMO) inhibitor glasdegib was approved for the treatment of this patient population based on increased response rates and OS in combination with low-dose cytarabine (LDAC) vs LDAC alone.^13,14 In recent years, the use of HMAs alone or in combination with the BCL2 inhibitor venetoclax has established a dominant role in the treatment of older/unfit adults with AML; however, ongoing studies may identify novel roles for glasdegib in the treatment of newly diagnosed AML, as part of both lower intensity and intensive approaches.

The Hedgehog Signaling Pathway

The canonical Hedgehog (Hh) signaling pathway, first discovered in Drosophila, is a highly conserved signaling pathway of key importance in embryological development, with roles including but not limited to primitive hematopoiesis and organogenesis. The majority of these roles appear to be epigenetically silenced in most human tissues after early development; roles in adults include extrathymic T cell development and pro-survival signaling in germinal center B cells, however the existence of an active role in adult hematopoiesis is unclear.^15–17

The human Hh signaling cascade, illustrated in Figure 1, begins with three ligands: Sonic Hedgehog (SHH), Indian Hedgehog (IHH), and Desert Hedgehog (DHH). These ligands bind to the transmembrane protein Patched (PTCH), releasing its inhibition of the 7-transmembrane G-like protein-couples receptor SMO. SMO, when uninhibited, can release transcription factors including GLI-1, GLI-2, and GLI-3 from their repressor complex Sufu, activating them and enabling transcription of target genes including cyclin-dependent kinases (eg, CCND1, CCND2) and pro-survival proteins (eg, BCL-2, BCL-XL). This leads to a myriad of downstream effects including pro-survival and anti-apoptotic signaling, as well as self-renewal and differentiation of hematopoietic stem cells.^16–20

Figure 1 The Hedgehog signaling pathway. (A) PTCH inhibits SMO, suppressing Hh signaling. (B) Hh ligands release inhibition of PTCH on SMO, allowing Hh signaling via release of transcription factors and promotion of downstream gene expression. (C) Glasdegib inhibits SMO, suppressing Hh signaling. Abbreviations: Hh, Hedgehog; PTCH, Patched; SMO, Smoothened.

In 1987, the gli gene, which encodes for the transcription factors GLI-1/2/3, was found to be highly expressed in human glioma.²¹ Later, in 1996, aberrant Hh signaling was linked with the formation of basal cell carcinoma in the inherited basal cell nevus syndrome.²² Over time, deregulated Hh signaling and GLI transcription factor activation were implicated in a wide range of hematologic malignancies including AML, acute lymphoblastic leukemia (ALL), chronic myeloid leukemia (CML), primary myelofibrosis, and multiple myeloma.^17,23–26 Evidence began to mount that aberrant Hh signaling played a role in the survival, renewal, and expansion of the leukemic stem cell (LSC).^25,27

Targeting Aberrant Hh Signaling and the Approval of Glasdegib

In an effort to modulate aberrant Hh signaling in hematologic malignancies, compounds with the ability to modulate or target Hh signaling were developed. In 2000, the plant-derived compound cyclopamine showed inhibition of the Hh pathway in mouse embryonic fibroblasts.²⁸ Subsequently, the small molecule SMO inhibitor PF-04449913 (PF-913), now known as glasdegib, was developed. Preclinical studies, including in patient-derived xenografts (PDX), showed an ability to reduce tumor burden, sensitize quiescent malignant stem cells to chemotherapy such as cytarabine, and decrease chemoresistance mediated by the bone marrow microenvironment.^29–31 In a Phase I study conducted in 2015, glasdegib showed efficacy in the treatment of multiple hematologic malignancies including AML: of 28 patients treated, some evidence of biological response (CR, complete or partial remission with incomplete hematological recovery, partial response, stable disease, and minor response) was noted in 16 patients (57%), including one patient achieving morphological CR with incomplete hematological recovery (CRi).³² Although clinical activity in AML with glasdegib monotherapy was modest, the landmark Phase 2 Bright AML 1003 trial demonstrated improved OS (8.8 vs 5.5 months) and rate of CR (19.2% vs 2.6%) in intensive chemotherapy-ineligible patients with AML or high-risk myelodysplastic syndrome (MDS) with the combination of Glasdegib/LDAC vs LDAC alone.^13,14 Toxicity was similar in both arms without an increase in grade 3–4 adverse effects in the combination arm, with alopecia, dysgeusia, QTc prolongation, and muscle spasms thought to be linked to SMO inhibition.¹⁴ 36-month post-hoc analysis confirmed these findings: Improved OS occurred across all cytogenetic risk groups, and a survival trend with glasdegib/LDAC was observed in patients with both de novo AML (hazard ratio 0.72) and even more pronounced in patients with secondary AML (hazard ratio 0.287).¹³ Additional post-hoc analysis revealed benefit with glasdegib/LDAC vs LDAC alone even in patients who did not attain CR, including improved rates of blood product transfusion independence (15% vs 2.9%) and durable recovery in the absolute neutrophil count (ANC) ≥1000/μL (45.6% vs 35.5%).³³ Based on the preliminary results of the Bright AML 1003 trial, The Pfizer-developed glasdegib (brand name Daurismo) at the dose of 100 milligrams daily was approved by the United States Food and Drug Administration (FDA) in November 2018 in the USA for use in combination with low-dose cytarabine for the treatment of newly diagnosed AML in patients aged ≥75 years or those who have comorbidities that preclude use of intensive induction chemotherapy.^34,35

The HMAs azacitidine and decitabine were approved by the FDA in 2004 and 2005, respectively, in the treatment of MDS. In 2008, azacitidine was approved in the use of AML with 20–30% blasts in patients ineligible for intensive therapy, and in clinical practice they became commonly utilized in this patient population regardless of blast count. Response rates including hematologic improvement were modest among trials (10–50%), a median of 3.5–4.3 months of therapy was needed to achieve best response, and median OS was under one year.^36–40 After the FDA approval of glasdegib, certain comparative analyses found a possible survival advantage with glasdegib/LDAC vs HMA monotherapy.^41,42 The treatment landscape changed in 2018 when the oral BCL2 inhibitor venetoclax was approved in combination with an HMA or LDAC in the treatment of newly diagnosed in AML in patients aged 75 years or older or who have comorbidities that preclude use of intensive induction chemotherapy. The Phase 1b trial leading to approval and the confirmatory Phase 3 VIALE-A trial, which compared azacitidine/venetoclax to azacitidine/placebo, showed previously unprecedented CR/Cri rates of 66–74% and median OS 16.9 months with nonintensive therapy.^43–45 Although glasdegib/LDAC may have been preferred over HMA monotherapy by some providers, the novel combination of venetoclax/HMA soon became the leading lower intensity therapy for the older and unfit population. As a result of the timing of this approval, glasdegib saw limited uptake in clinical practice.

Novel Trials and Future Directions

Although the glasdegib/LDAC combination therapy has largely fallen out of favor due to the widespread adoption of venetoclax/HMA, numerous clinical trials (summarized in Table 1) are currently underway in an effort to harness glasdegib’s potential to eliminate LSCs and expose synergy with currently available therapies, both non-intensive and intensive. In 2015, an ex vivo study showed synergistic potential with the combined use of azacitidine and the SMO inhibitor erismodegib.¹⁸ Next, in 2017 a subsequent ex vivo study demonstrated that GLI3 signaling appeared to be abnormally methylated and silenced in most AML, independent of SMO activation, and that HMAs could restore this activity and sensitize AML cells to glasdegib.⁴⁶ These preclinical discoveries have led to new clinical trials evaluating the use of glasdegib in novel combination regimens. The recently published phase 1b Bright AML 1012 (NCT02367456) studying glasdegib in combination with azacitidine has reported a median OS of 9.2 months and a CR and overall response rate of 20% and 30%, respectively, in patients with ND-AML, with a relatively low incidence of cytopenias and delayed marrow recovery. These results, although early with a median duration of follow-up of 8.5 months, appear to be at least comparable if not superior to those of glasdegib/LDAC. A signal of increased OS was noted for patients with FLT3 mutations with median OS not reached, which may expose a niche for this combination therapy should this trend be confirmed.^47,48 The phase 2 GLAD-AML trial studying glasdegib in combination with decitabine enrolled one patient before being terminated due to failure to accrue patients in the setting of the COVID-19 pandemic.⁴⁹

Table 1 Recent and Ongoing Clinical Trials Examining Novel Glasdegib-Based Combinations in Newly Diagnosed AML

The phase 2 trial NCT01546038 published in 2018 examined glasdegib in combination with intensive induction with 7+3 (intravenous cytarabine 100 mg/m2 on days 1–7 and daunorubicin 60 mg/m2 on days 1–3) in untreated AML or MDS with ≥10% blasts. CR rates were similar to historical controls; however, post-hoc review suggested a potential benefit in OS with a median OS of 14.7 months in AML patients ≥55 years old vs 8.7 months with historical control. The OS curve plateaued between 24 and 36 months with ~40% of patients alive at 36 months.⁵⁰ An improvement in OS and possible cure, despite CR rates similar to historical control, can in theory be mechanistically explained by purported action of glasdegib on LSCs rather than tumor bulk cells. Sample size was limited; however, results were encouraging. The randomized, double-blinded phase 3 Bright AML 1019 trials evaluate the combination of glasdegib with intensive and lower intensity induction in newly diagnosed AML in two separate arms. The intensive arm combines glasdegib/placebo with 7+3 while the lower intensity trial combines glasdegib/placebo with azacitidine. The primary end point of both arms is OS, although long-term relapse-free survival will be another point of interest. The trail has been completed and is pending publication.⁵¹ The phase 2 NCT04231851 (CPX-351 and Glasdegib for Newly Diagnosed Acute Myelogenous Leukemia With MDS Related Changes or Therapy-related Acute Myeloid Leukemia) is currently enrolling.

Many exciting trials are currently underway; however, glasdegib continues to have a potential niche in the landscape of currently approved therapies. Although venetoclax/HMA produces considerably higher CR rates than glasdegib/LDAC, the toxicity is also significantly higher when venetoclax is added to an HMA: 98–100% of patients in the phase 1b study experienced an adverse event (AE) grade 3 or higher, with a 31–42% rate of febrile neutropenia and 4–7% incidence of sepsis.^44,52 In the confirmatory Phase III VIALE-A trial, the incidence of grade 3–4 thrombocytopenia, neutropenia, and febrile neutropenia in the azacitidine/venetoclax vs azacitidine/placebo groups were 45% vs 38%, 42% vs 28%, and 42% vs 19%, respectively.⁴⁵ 30-day mortality with azacitidine/venetoclax and Glasdegib/LDAC were comparable at 7% vs 6%, respectively; however, the fatal adverse event rate was considerably higher with azacitidine/venetoclax at 23% vs 7%.^14,44 The safety profile of glasdegib/LDAC in the Bright AML 1003 study compares favorably to venetoclax/azacitidine, with a 28.6% incidence of febrile neutropenia and a 31% incidence of grade 3–4 thrombocytopenia.¹⁴ The combination is better tolerated, less myelosuppressive, and results in fewer treatment-related hospitalizations. In elderly and unfit patients with less physiological reserve or value quality of life over treatment intensity, glasdegib/LDAC may be a preferable treatment option. Additionally, it can be utilized off-label in the second line or later after the failure of venetoclax/HMA and targeted therapies.⁵³

Conclusions

Initially developed as a non-intensive and relatively nontoxic treatment of newly diagnosed AML in older and unfit adults, glasdegib has experienced limited utilization in the era of the HMAs and venetoclax combinations; however, in combination with LDAC, it remains an FDA approved and NCCN guideline-based option for first or subsequent line non-intensive therapy, including in the setting of HMA ± venetoclax failure. The promise of inhibition of the SMO protein and Hh pathway lies in the potential to suppress LSCs and prevent relapse, rather than a direct cytotoxic effect and an improvement in initial CR rate. Based on these preclinical findings with early clinical evidence, glasdegib is being tested in novel therapeutic combinations such as with HMAs and intensive induction chemotherapy. Success in ongoing clinical trials and clinical evidence of synergy in combination with currently approved treatments may renew and expand the role of glasdegib in the modern armamentarium of therapies to treat newly diagnosed AML.