AUTHORS
Alex Bataller, MD, PhD, Omer Karrar, MD2, Abhishek Maiti, MBBS3, Jayastu Senapati, MBBS, MD, DM4, Ian Bouligny, M.D.5, Ghayas C. Issa, MD6, Musa Yilmaz, MD7, Guillermo Montalban-Bravo, MD8, Kelly Sharon Chien, MD9, Manuel M. Maroun, MD10, Gautam Borthakur, MD11, Farhad Ravandi, MD12, Naval Daver, MD13, Courtney DiNardo, MD, MSCE14, Tapan Kadia I, MD15, Hagop Kantarjian, MD16, Guillermo Garcia-Manero, MD17, Nicholas Short, MD18
AFFILIATIONS
1 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
2 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
3 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
4 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
5 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
6 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
7 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
8 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
9 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
10 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
11 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
12 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
13 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
14 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
15 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
16 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
17 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
18 The University of Texas MD Anderson Cancer Center, Houston, Texas, United States
ABSTRACT
Introduction: Treated secondary AML (tsAML) is defined as AML progressing from a previously treated myeloid malignancy. Patients with tsAML are often excluded from both frontline and relapsed/refractory (R/R) trials, due to their classification as "newly diagnosed" AML but with history of prior therapy exposure. Prior studies—mostly in patients treated with hypomethylating agents (HMA) for antecedent myeloid malignancy—show poor outcomes of tsAML, with median OS <6 months, similar to R/R AML. HMA+venetoclax (Ven) is increasingly used in myelodysplastic syndromes (MDS) and chronic myelomonocytic leukemia (CMML), but outcomes of tsAML after this combination remain unevaluated. Here, we compare outcomes among patients with tsAML previously exposed to HMA alone (HMA tsAML), HMA+Ven (HMA+Ven tsAML), and R/R AML after frontline HMA+Ven.
Methods: We retrospectively analyzed pts with newly diagnosed AML from 1/2014 to 2/2023, divided into: de novo AML (no MDS/CMML history), untreated sAML (AML after untreated MDS/CMML), HMA tsAML (AML after MDS/CMML with HMA exposure), and HMA+Ven tsAML (AML after MDS/CMML with HMA+Ven exposure). A second cohort of pts who received frontline HMA+Ven for AML (9/2017–12/2024) and relapsed or were refractory (R/R AML cohort) was also analyzed.
Results: Among pts with newly diagnosed AML (n=2,310), 1,415 (61%) had de novo AML, 194 (8%) had previously untreated sAML from MDS/CMML, 569 (25%) had tsAML, and 132 (6%) had AML secondary to other myeloid neoplasms. Among tsAML pts, 475 had received HMA without Ven (HMA tsAML) and 77 had received HMA+Ven (HMA+Ven tsAML) for prior MDS/CMML. Median age was 70 in both cohorts (P=ns). Diploid and adverse-risk cytogenetics were present in 22%/51% (HMA tsAML) and 30%/53% (HMA+Ven tsAML) (P=ns for both). Mutations at AML diagnosis were similarly distributed between cohorts, except IDH2 (8% HMA tsAML vs 0% HMA+Ven tsAML, P=0.03). Prior alloSCT for MDS/CMML occurred in 11% (HMA tsAML) vs 25% (HMA+Ven tsAML) (P=0.001).
A cohort of pts with R/R AML after frontline HMA+Ven (n=147; 116 relapsed and 31 refractory) was analyzed for direct comparison with the tsAML cohorts. The median age was 73 years. Diploid and adverse-risk cytogenetics were present in 23% and 50% of pts, respectively. No significant differences in baseline features were observed between the tsAML and R/R AML cohorts, except the for the incidence of NPM1 mutations (4% in tsAML vs 16% in R/R AML, P=0.02).
Treatment for AML included intensive chemotherapy (32% HMA tsAML, 19% HMA+Ven tsAML, 16% R/R AML), low-intensity therapy (59%, 69%, and 57%, respectively), and early-phase clinical trials (9%, 12%, and 27%, respectively). The composite remission rate (complete remission [CR]/CR with incomplete count recovery) was 30% for HMA tsAML, with similar rates for HMA+Ven tsAML and R/R AML (17% and 16%, respectively) (P=0.04 for HMA vs. HMA+Ven tsAML, P=ns for HMA+Ven tsAML vs. R/R AML). A similar pattern was observed with CR rates, with CR rates of 16%, 4%, and 6%, respectively (P=0.009 for HMA vs. HMA+Ven tsAML, P=ns for HMA+Ven tsAML vs. R/R AML). Among pts who achieved remission, alloSCT was performed in 20% of pts with HMA tsAML, 26% with HMA+Ven tsAML, and 8% with R/R AML.
The median overall survival (OS) and 2-year OS were 17 months and 42% in de novo AML, 10 months and 29% in untreated sAML, and 5 months and 8% in tsAML, respectively (P<0.001). Among pts with tsAML, outcomes were significantly worse in those with prior HMA+Ven vs. HMA only exposure (median OS and 2-year OS: 3 months and 3% vs. 5 months and 9%, respectively; P=0.003). The median OS and 2-year OS for the R/R AML cohort were 4 months and 6%, respectively, which was significantly worse than pts with HMA tsAML (P=0.003) but similar to those with HMA+Ven tsAML (P=ns).
Conclusions: Pts with HMA+Ven-exposed tsAML have comparably dismal OS to pts with R/R AML after frontline HMA+Ven failure, and worse OS than non-Ven-exposed tsAML pts. tsAML outcomes are extremely poor (median OS 5 months), representing an ultra-high-risk group with response and OS significantly inferior to de novo and untreated sAML. These findings support including pts with HMA or HMA+Ven tsAML in early-phase clinical trials, particularly those with HMA+Ven tsAML in whom the clinical course closely mirrors R/R AML after HMA+Ven failure.
DISCLOSURES
Alex Bataller, MD, PhD: No relevant conflicts of interest to declare.