Impact of Molecular Ontogeny on Hematological Recovery in AML Patients Treated with Hypomethylating Agent Plus Venetoclax Therapy

Introduction: Therapy with hypomethylating agents (HMA) plus venetoclax (VEN) is approved for patients (pts) unfit to receive intensive chemotherapy in newly diagnosed acute myeloid leukemia (AML). However, many pts experience treatment delays due to hematologic toxicity, necessitating shortened days of VEN or HMA administration and extending cycle lengths. AML ontogeny can be classified by molecular characteristics into de-novo, secondary and TP53 mutated AML (Lindsley Blood 2015). Secondary ontogeny and TP53 mutated AML are associated with prior history of myeloid malignancy or prior cytotoxic therapy. We sought to determine the impact of AML ontogeny on treatment and hematologic outcomes in pts treated with HMA+VEN.

Methods: We conducted a retrospective study of pts with newly diagnosed AML treated with at least one cycle of HMA+VEN at Dana-Farber Cancer Institute between 2016 and 2023. To exclude the impact of persistent disease on blood counts, we only included those who achieved clearance of leukemia blasts including complete remission (CR), complete remission with incomplete count recovery (CRi), or morphological leukemia-free state (MLFS). We defined three ontogeny groups in a hierarchical pattern based on presence of TP53 mutation (“TP53 mt”), >1 secondary ontogeny mutation (ASXL1, BCOR, EZH2, SF3B1, SRSF2, STAG2, U2AF1 or ZRSR2) without TP53 mutation (“secondary”), and all others (“de novo”). We collected data for a maximum of five cycles of treatment including cycle lengths, days of venetoclax administration, hematologic counts and recovery times, and transfusion requirements.

Results: A total of 135 pts were included: 37 (27%) were in the TP53 mt group, 61 (45%) in the secondary group, and 37 (27%) in the de novo group. The median age differed between de novo (77 years), secondary (75 years), and TP53 mt (63 years) groups (p < 0.001). The median baseline hemoglobin (hgb) was higher in the de novo group (9.0 g/dL) than secondary (8.2 g/dL, p=0.054) and TP53 pts (8.0 g/dL, p=0.003). Similarly, the median baseline platelet (plt) count was higher in de novo pts (59 K/µL) compared to secondary (48 K/µL, p=0.2) and TP53 mt pts (23 K/µL, p < 0.001). The median number of cycles of HMA+VEN received did not differ (p=0.08) between pts with de novo (5 cycles, range [R] 1-13), secondary ontogeny (3.5, R 1-17) and TP53 mt pts (3, R 1-8). During cycle 1, VEN duration (28 days for all three ontogenies) and cycle length (35 vs 37 vs 34 days, p=0.46) was similar across groups. During cycle 2, VEN duration was shorter in secondary compared to de novo pts (21 vs 25 days, p=0.006).

During cycle 1, the median number of red blood cell (6 vs 3, p=0.006) and plt (3 vs 0, p=0.027) transfusions was higher in secondary compared to de novo AML. Similarly, the median number of red blood cell (8 vs 3, p=0.001) and plt (5 vs 3, p<0.001) transfusions was higher in TP53 mt compared to de novo AML. Several treatment-emergent differences surfaced during later cycles of HMA+VEN between ontogeny groups. Hgb was lower in the TP53 mt group compared with de novo at the start of cycle 2 (p=0.024) and cycle 3 (p=0.008). Although absolute neutrophil count (ANC) did not differ during cycles 1 and 2, at start of cycle 3, median ANC was lower in TP53 mt compared to de novo pts (300 vs 1150/µL, p<0.001). Median ANC recovery time in cycle 3 was longer in secondary compared to de novo pts (42 vs 32 days, p=0.047), as was hgb recovery (31 vs 14 days, p=0.018). Furthermore, the median cycle 3 length (days) was also longer in secondary compared to de novo and TP53 mt pts (42 vs 35 vs 30 days, p=0.002). There were no significant differences in cycle length and time to count recovery in cycles 4 and 5.

Conclusions: Overall, pts with either TP53 mt or secondary ontogeny had lower hgb and plt at baseline, required more transfusion support during cycle 1 and later developed more pronounced myelosuppression compared to de novo pts when treated with HMA+VEN. Prior studies in AML have demonstrated similar efficacy in 14-day or even 7-day dosing of VEN compared to 28-day dosing (Aiba 2023, Willekens 2024). In addition, the VERONA trial showed HMA+VEN was safe and had high response rates in high-risk myelodysplastic syndromes when given for 14 days (Garcia ASH 2023). Hence, upfront dose adjustment of VEN based on AML ontogeny should be considered and studied prospectively. This may help reduce the need for blood and platelet transfusions during cycle 1 and minimize delays in subsequent treatment cycles.