Real-World Observational Study of Outcomes for Acute Myeloid Leukemia (AML) in Patients Treated with Glasdegib or Venetoclax in US Community Oncology Practices

Introduction: Glasdegib (GLAS) and venetoclax (VEN) were approved by the FDA in Nov 2018 for the treatment of AML in patients (pts) who are ≥75 years old or who have comorbidities that preclude intensive induction chemotherapy. Data related to the clinical outcomes of pts who receive these agents in the community oncology setting are limited. We conducted a retrospective, observational study to assess US real-word (RW) patient characteristics and clinical outcomes for pts who received a GLAS- or VEN-based regimen as first line (1L) therapy for AML. This study was observational and not designed or powered to compare characteristics or outcomes between pts treated with GLAS and those treated with VEN. We present RW findings comparing to the respective pivotal clinical trial populations (the BRIGHT AML 1003 trial which evaluated GLAS and the VIALE-A trial which evaluated VEN), including an exploratory analysis to match the characteristics of the RW cohorts to the clinical trials.

Methods: A retrospective, observational, US multi-site, cohort chart review of adult pts who initiated 1L therapy for AML between 01 Dec 2018 and 31 Dec 2019 was conducted. Eligible pts had at least 6 months (mo) of follow-up since initiation of 1L (unless the pt died within this follow-up time, with known date of death), known cytogenetic risk profile, and had at least 1 bone marrow biopsy completed following index therapy initiation. Index therapy was either a GLAS- or VEN-based regimen (with low-dose cytarabine [LDAC], azacytidine [AZA), or decitabine). Pts treated in a clinical trial were excluded. The treating physician extracted all data for a given patient, including demographics, clinical characteristics, treatment patterns and dosing, transfusions, follow-up, disease response, and clinical outcomes.

Results:

50 pts treated with GLAS as 1L therapy for AML, most of whom (82.0%) received GLAS in combination with LDAC, were followed for a median of nearly 7 mo, 28% of whom were still alive at data collection. The median age at initiation of 1L GLAS was 77 (47-89) years. Approximately half of 1L GLAS pts had an ECOG-PS of 0 or 1 at 1L therapy initiation. The median overall survival (mOS) for pts who received 1L GLAS was 6.9 mo (95% CI: 5.4-8.9); for comparison, the mOS for 88 GLAS- treated pts in the BRIGHT AML 1003 clinical trial was 8.8 mo. The median duration of follow-up was shorter in the RW cohort versus the trial (6.9 vs 27.1 mo). ECOG-PS 3 and poor cytogenetic risk score was noted in a greater proportion of RW 1L GLAS cohort vs BRIGHT AML 1003 trial (14% vs 0% and 58% vs 41%, respectively). The mOS for a subset of RW pts who matched all eligibility criteria for BRIGHT AML 1003 was 7.7 mo.

83 RW pts treated with VEN as 1L therapy for AML, the majority of whom (61.5%) received VEN in combination with AZA, were followed for a median of 8.4 mo and 38.6% of these pts were reported to be alive at data collection. The median age at initiation of VEN was 73 years and ECOG-PS of 0 or 1 was reported for 44.6% pts, 2 for 47.0% and 3 for 8.4%. The mOS of pts treated with 1L VEN was 8.4 mo (95% CI: 5.7-16.2); this compared to 286 VEN-treated pts in the VIALE-A trial with a mOS of 14.1 mo. The VEN RW cohort has shorter follow up compared to VIALE-A (8.4 mo vs 20.5). In comparison to pts treated with VEN in the VIALE-A trial, a greater proportion of pts in the RW VEN cohort had ECOG-PS scores of 2 or 3 (55.4% vs 45%) and poor cytogenetic risk (49.4% vs 36%). When the RW VEN cohort was matched to the eligibility criteria for VIALE-A, the RW mOS was not reached (n=23).

Conclusions:

In this RW observational study, the pt clinical and cytogenetic characteristics in the RW cohorts were less favorable for both GLAS and VEN than those observed in their respective clinical trials. In the RW, mOS was 1.9 mo lower for GLAS and 5.7 mo lower for VEN versus their respective clinical trials. This study was not comparative between GLAS and VEN, and indeed the pt populations were different with a higher median age and a higher proportion of ECOG-PS 3 in the GLAS RW cohort than the VEN RW cohort; the main combination agent also differed (LDAC or AZA). The settings and populations evaluated in these studies were unique and offer novel insights into the RW treatment patterns and effectiveness of both therapeutic options. Outcomes may be explained by differences in the characteristics of the RW cohorts and the eligibility criteria between the clinical trials.

Gajra:Cardinal Health: Consultancy, Current equity holder in publicly-traded company, Ended employment in the past 24 months; Cellectar: Honoraria; AstraZeneca: Honoraria. Kish:EVERSANA: Current Employment. Russell-Smith:Pfizer: Current Employment, Current holder of stock options in a privately-held company. Savill:Astra Zeneca: Consultancy; Janssen: Consultancy; Morphosys: Consultancy; Bayer: Consultancy; Daiichi Sankyo: Consultancy; Incyte: Consultancy; Pfizer: Consultancy; Genentech: Ended employment in the past 24 months; Cardinal Health: Current Employment, Current holder of stock options in a privately-held company.