AUTHORS
Bret R Miller, MD, Jaime Haro-Silerio, MD2, Noah Balestra, MD3, Andy Li, MD4, Johnathan Fraebel, MD5, Chase West, MD6, Eiline Cai, MD7, Lewis Nasr, MD8, Harrison Turner, MD9, Trung Nguyen, MD10, Jason Lu, MD11, Tareq Abuasab, MD12
AFFILIATIONS
1 Baylor College of Medicine, Houston, Texas, United States
2 Baylor College of Medicine, Houston, Texas, United States
3 Baylor College of Medicine, Houston, Texas, United States
4 Baylor College of Medicine, Houston, Texas, United States
5 Baylor College of Medicine, Houston, Texas, United States
6 Baylor College of Medicine, Houston, Texas, United States
7 Baylor College of Medicine, Houston, Texas, United States
8 Baylor College of Medicine, Houston, Texas, United States
9 Baylor College of Medicine, Houston, Texas, United States
10 Baylor College of Medicine, Houston, Texas, United States
11 Baylor College of Medicine, Houston, Texas, United States
12 Baylor College of Medicine, Houston, Texas, United States
ABSTRACT
Introduction: Intensive chemotherapy (IC) has long been the standard induction for fit adults with acute myeloid leukemia (AML); with the results of PARADIGM, the role of lower intensity chemotherapy (LIT) is increasingly explored as the standard induction regimen. Whether the two approaches differ in depth of response, as measured by measurable residual disease (MRD), is less well established, and toxicity comparisons are rarely stratified by response depth. We compared CR/CRi and minimal residual disease (MRD)-negativity rates between IC and LIT, and among responders assessed overall survival (OS) and treatment-associated toxicity by MRD status.
Methods: Retrospective analysis of adults with newly diagnosed AML treated at Baylor St. Luke’s Medical Center between March 2017 and July 2025. Patients with APL or non-AML myeloid neoplasms were excluded. IC comprised 7+3 based chemotherapy; LIT comprised venetoclax-based regimens (n=71); and non-venetoclax based (n=5). Best response was categorized as CR/CRi versus other; patients without adequate response documentation were non-evaluable. MRD was assessed in CR/CRi patients by flow cytometry at best response with sensitivity 0.01%. ELN 2022 was applied to the IC arm and ELN 2024 to the LIT arm; risk distributions are therefore reported descriptively and not compared across arms. Categorical comparisons used chi-square or Fisher's exact tests. OS among responders was estimated by Kaplan-Meier from a landmark at date of documented response and compared by log-rank test. Toxicity comparisons were exploratory and uncorrected for multiplicity.
Results: Of 256 patients screened, 163 met criteria (IC n=87, LIT n=76). Median age was 52.7 (IC) versus 72.3 years (LIT) (p<0.001). ELN 2022 risks were 31.0% favorable, 21.8% intermediate, and 47.1% adverse and ELN 2024 risks were 26.3% favorable, 35.5% intermediate, 36.8% adverse, and 1.3% unknown.
CR/CRi was significantly higher with IC than LIT (n=67, 77% vs n=44, 57.9%, p=0.009). Among the 111 responders, MRD-negativity was more frequent after IC (n=46, 68.7% vs n=19, 43.2%, p=0.008). Allogeneic transplantation was performed in 25.4% of IC and 6.8% of LIT responders (p=0.013).
In landmark analysis from date of response, median OS was 53.3 months for IC/MRD-negative responders, not reached for IC/MRD-positive responders, 24.7 months for LIT/MRD-negative responders, and 10.6 months for LIT/MRD-positive responders. MRD status showed numerically favorable OS trends within each arm (IC p=0.10; LIT p=0.22); however, differences did not reach statistical significance. Median follow up time was 31.7 months in the LIC group and 50.2 months in the IC group (P<0.001).
Among MRD-negative responders, toxicity did not differ significantly during induction with LIT vs IC, though rates were numerically lower with LIT for neutropenic fever (52.6% vs 69.6%, p=0.26), transaminitis (10.5% vs 23.9%, p=0.31), and pneumonia (10.5% vs 19.6%, p=0.49). Among MRD-positive responders, neutropenic fever was less frequent with LIT (54.2% vs 81%, p=0.07), although did not reach statistical significance; all other toxicities analyzed were similar (all p≥0.17). Thirty-day mortality was 1.1% (IC) versus 1.3% (LIT).
Conclusions: IC produced both higher CR/CRi rates and deeper remissions, with MRD-negativity achieved in roughly two-thirds of IC responders versus two-fifths after LIT. Notably, this occurred despite a higher proportion of adverse-risk disease in the IC arm. MRD-negativity tracked with longer OS in both arms, supporting its prognostic value irrespective of induction intensity, although it did not reach statistical significance in our study, likely due to the small sample size. Toxicity was broadly comparable within the response strata. Interpretation is limited by the 19-year median age difference between arms, nonrandomized treatment assignment, and the limited sample size.
DISCLOSURES
Bret R Miller, MD: No relevant conflicts of interest to declare.