Real-World Treatment Patterns and Transfusion Burden Among Newly Diagnosed Older Adults with Acute Myeloid Leukemia

Abstract

Background: The approval of Venetoclax (VEN; a BCL-2 inhibitor) has changed the treatment paradigm for older/unfit newly-diagnosed AML patients (ND AML) and improved outcomes in the landmark randomized phase 3 clinical trial, VIALE-A. However, real world evidence of population-level clinical benefit is still limited, especially on the impact of transfusion burden and achievement of transfusion independence (TI). Repeated transfusions are associated with significant risks and costs and impair quality of life for patients. This study examined real-world treatment patterns and clinical outcomes focusing on transfusion independence in a contemporary cohort of Medicare beneficiaries with newly diagnosed AML. Methods: This cohort study examined newly diagnosed Medicare beneficiaries (age ≥65 years) with AML using claims from the Centers for Medicare & Medicaid Services, representing 100% of Medicare fee-for-service beneficiaries. The identification period was July 2016-June 2020, with an observational period from January 2016 to December 2020. The first active AML treatment identified in the data was categorized into four different treatment categories; 1) Venetoclax combination (VEN; Venetoclax in combination with hypomethylating agents (HMA) or low-dose cytarabine (LDAC)), 2) low-intensity treatment (LOW; HMA monotherapy or LDAC), 3) high-intensity chemotherapy (HIGH; patients with an inpatient stay lasting ≥22 days or ≥7 days where a non-specific antineoplastic treatment or cytotoxic chemotherapy was documented), and 4) other targeted therapies (OTHER; ivosidenib, glasdegib, gilteritinib, enasidenib, midostaurin, gemtuzumab ozogamicin). Patients who received any AML therapy prior to the index date except for treatments for myelodysplastic syndromes (e.g., HMAs and lenalidomide) were excluded. Transfusion independence (TI) was defined as a rolling period of no red blood cell (RBC) or platelet (PLT) transfusions for a consecutive 56-days. Results: We identified 5,855 ND AML Medicare beneficiaries, with 734 patients who met eligibility criteria and were categorized as VEN, 2,562 patients as LOW, 1,318 patients as HIGH, and 146 patients as OTHER. The median follow-up duration was 8 months and the median age at diagnosis was 76 years old in all patients. Patients had a mean Charlson Comorbidity Index score of 3, including common comorbidities: 23.4% with congestive heart failure and 27.6% with chronic pulmonary disease. The VEN cohort initiated their first line (1L) treatment on an average of 30 days from their diagnosis. Median duration of 1L therapy was 4.8 months (IQR: 1.9-6.6) in the VEN cohort, 3.1 months (IQR: 1.1-8.1) for the LOW cohort, 1.7 months (IQR: 0.9-1.8) for the HIGH cohort, and 2.1 months (IQR: 1.0-6.1) in the OTHER cohort. The mean daily dose in the VEN cohort over the 1L treatment period was 287 mg (median: 332). 15.4% of VEN patients had overlapping hospitalizations, indicating that patients may initiate Venetoclax at inpatient setting. Within 56-days prior to treatment initiation, 86.4% of VEN cohort had PLT-TI and 54.5% had RBC-TI, and similar proportions of TI were observed in other treatment groups. When evaluating the transfusion status every 56-days on a rolling weekly basis (Figures 1 & 2): a trend of decreased proportion of TI were observed from 56-days prior to the treatment initiation across treatment groups. By weeks 0-7 post-treatment initiation, 49.5% in VEN cohort achieved TI in PLT and 18.0% in RBC, which was lower than the LOW cohort (58.2% were TI in PLT vs. 29.4% in RBC). However, there was an apparent increasing trend of TI rate in both RBC and PLT since the first week after the patients initiating VEN, while the TI rate only slightly increased over time since the treatment initiation in the other treatment group. At weeks 8-15, a higher proportion of patients achieved TI in PLT and RBC in the VEN cohort, compared with the other treatment groups (PLT-TI: 74.2% in VEN, 70.1% in LOW, 40.0% in HIGH, and 57.6% in OTHER; RBC-TI: 52.2% in VEN, 41.9% in LOW, 35.6% in HIGH, and 41.3% in OTHER). At week 12-19, 62.1% in VEN achieved RBC-TI, versus 47.7% in LOW, 44.2% in HIGH, and 48.1% in OTHER. Conclusions: In the Medicare elderly population, ND AML patients treated with VEN had a longer median duration of therapy compared with other treatment groups. A larger proportion of VEN patients were able to achieve transfusion independence over time compared to the other treatment groups. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal