AUTHORS
Noah Balestra, MD, Jaime Haro-Silerio, MD2, Bret Miller, MD3, Johnathan Fraebel, MD4, Eiline Cai, MD5, Trung Nguyen, MD6, Lewis Nasr, MD7, Chase West, MD8, Andy Li, MD9, Harrison Turner, 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
AML is often refractory to induction chemotherapy and has a high risk of relapse after remission. Relapsed/refractory (R/R) AML carries a poor prognosis, with no consensus on optimal salvage intensity. While intensive chemotherapy (IC) has been the traditional salvage approach for fit patients, recent studies have challenged the superiority of IC over modern low intensive therapy (LIT) regimens in the frontline setting, where LIT has shown comparable survival with significantly reduced toxicity. Whether LIT as salvage therapy performs similar to intensive salvage regimens remains unclear. We compared survival and toxicity outcomes by salvage intensity (stratified by prior induction regimen) at a single academic center.
Methods
A retrospective analysis of patients at Baylor St Luke's Medical Center from October 2013 to March 2026 was performed. Patients with R/R AML were stratified into four groups based on induction → salvage intensity: IC→IC (n=30), IC→LIT (n=19), LIT→IC (n=12), and LIT→LIT (n=23). IC regimens included 7+3, CPX-351 (liposomal cytarabine-daunorubicin), CLIA, and MEC; LIT included Azacitidine, Decitabine, or Cladribine/low-dose cytarabine ± venetoclax. Primary end points were overall survival (OS) and progression free survival (PFS). Secondary endpoints were rates of adverse effects, CR/CRi, negative Neogenomics minimal residual disease (MRD) flow cytometry (sensitivity 0.01%), and post-salvage bone marrow transplants. Kaplan-Meier analysis with log-rank testing was performed to compare OS and PFS and chi square tests of independence were used.
Results
Of 256 patients with AML, 84 received salvage therapy. Patients treated with IC→IC were younger than IC→LIT (mean age 44.6 vs 57.8 years; p=0.003) with a similar proportion of ELN 2022 adverse-risk disease (40.0% vs 52.6%, p=0.39). Despite these favorable characteristics, IC→IC did not improve PFS (4.9 vs 12.0 months; p=0.76) or OS (9.4 vs 16.0 months; p=0.85) compared with IC→LIT. CR/CRi rates were 62.5% vs 60% (p=0.88), and MRD-negative responses were 50.0% vs 27.2% (p=0.24), respectively. Post-salvage transplant rates were 23.3% (IC→IC) vs 31.6% (IC→LIT) (p=0.52).
LIT→IC patients were younger than those who received LIT→LIT on average (62.2 vs 70.0 years; P=0.03) with similar rates of ELN 2024 adverse-risk rates (25.0% vs 30.4%; p=0.73), and similar PFS (1.2 vs 3.5 months; p=0.25), OS (4.1 vs 5.1 months; p=0.65), rates of CR/CRi (55.6% vs 45.5%; p=0.65), MRD-negativity (20.0% vs 11.1%, p=0.60) and transplantation (16.7% vs. 13.0%; p=0.77).
Toxicity differed by salvage intensity. Patients receiving IC salvage had significantly higher rates of neutropenic fever (92.9% vs. 38.1%, p<0.0001), sepsis (76.2% vs. 28.6%, p<0.0001), bacteremia (45.2% vs. 21.4%, p=0.02), ICU admission (33.3% vs. 11.9%, p=0.02), and transaminase elevations (45.2% vs. 23.8%, p=0.04). The rates of UTI, pneumonia, and AKI did not differ (p=0.08, 0.61, and 0.80, respectively). Thirty-day mortality after salvage initiation was not significantly different at 4.7% for IC salvage and 11.9% for LIT salvage (p=0.23).
Conclusions
We detected no statistically significant advantage for intensive salvage in either induction stratum in terms of survival and rates of CR/CRi, negative MRD status, or post-salvage bone marrow transplants. However, numerically longer survival was observed with LIT salvage compared to IC salvage after both IC and LIT induction. This is despite the numerically lower rates of adverse disease and significantly younger average age of patients who received IC→IC and LIT→IC, suggesting that intensive salvage failed to convert prognostic advantages into improved outcomes. Meanwhile, intensive salvage carried a markedly higher toxicity burden, with more than twice the rate of neutropenic fever and sepsis and nearly three times the rate of ICU admission, a burden that may offset any cytoreductive advantage in this population. Subgroup sizes were small and these analyses cannot exclude a clinically meaningful survival difference; the retrospective design, pooling of relapsed and primary refractory disease, and absence of mutation-level stratification (e.g., TP53, FLT3) further limit interpretation. Prospective studies comparing salvage intensity with molecular risk-adjusted endpoints are warranted.
DISCLOSURES
Noah Balestra, MD: No relevant conflicts of interest to declare.