Treatment Modalities and the Impact of Granulocyte Colony-Stimulating Factor in AML Patients Treated With Venetoclax...

Azacitidine-venetoclax (AZA-VEN) has become a standard treatment in older or chemotherapy-ineligible AML patients. While the registration trial showed a median overall survival (OS) of 14.7 months, most real-world studies have not reproduced this result.

We analyzed treatment patterns, adverse events, responses and outcomes in 199 patients treated with AZA-VEN in the DATAML registry.

The median age was 75.6 years, 48.7% had secondary AML (11% post-MPN); 11.7% had another cancer, and 9% had received AZA for prior myelodysplastic syndrome. Cytogenetic risk was intermediate (55.8%) or adverse (43.7%). The most frequent gene mutations were ASXL1 (33%), TET2 (27%), TP53 (26%), RUNX1 (24%) and SRSF2 (24%). The first cycle was performed on an outpatient basis in 39.4% of patients. The median number of cycles was 4 and 37% received > 6 cycles. Beyond cycle 6, reductions in treatment dose or duration were attributed to VEN in 55% of patients and to AZA for 43%. The complete remission (CR) plus CR with incomplete hematologic recovery (CRi) rate was 58.5%, and day-30 death rate was 3%. Median OS was 8.9 months. mPRS and refined-ELN2024 classifications were significantly associated with response and OS. In CR/CRi patients, G-CSF use was significantly and independently associated with improved OS (median, 10.1 months without versus 20.3 with G-CSF; P = .001). In an external cohort, G-CSF was also associated with a trend towards improved OS.

In real-world clinical practice, patients with more severe characteristics are typically selected for AZA-VEN, which could explain suboptimal outcomes. G-CSF should be prospectively explored in AZA-VEN treated patients.