Outcomes of low-intensity AML therapy based on treatment facility Free

Background

Historically, AML management required prolonged inpatient care, typically at academic centers. With the advent of low-intensity therapies—such as hypomethylating agents (HMAs), venetoclax, and targeted agents—outpatient treatment has increased across various settings. Prior studies reported better outcomes for patients with AML treated at high-volume academic institutions, attributed to earlier transplant referrals, multidisciplinary support, and more frequent use of intensive regimens. This study provides a contemporary comparison of outcomes in AML treated with low-intensity therapy based on treatment facility, focusing on key molecular mutations.

Methods

We conducted a retrospective single-center analysis of patients with AML who received low-intensity therapy at any treatment line. The study categorized patients into two groups: those treated at the academic center (University of Iowa Holden Comprehensive Cancer Center) and a collaborative/community center managed in community settings, either independently or with academic center consultation. Low-intensity regimens included HMAs (azacitidine or decitabine) or low-dose cytarabine (LDAC), alone or with venetoclax. The “other” group included patients who received single-agent targeted therapies, either as standard-of-care or in a clinical trial. Differences in categorical and continuous variables across treatment settings were analyzed using chi-square/Fisher's exact tests and ANOVA, respectively. Overall survival (OS) was estimated by the Kaplan-Meier methods; Cox regression and logistic regression were used to evaluate OS and the odds of CR/CRi/MLFS, respectively. Effects were reported as hazard (HR) and odds (OR) ratios, along with 95% confidence intervals (CI). Covariates significant on univariable analysis were included in multivariable models. All tests were two-sided, with significance at p<0.05 using SAS v9.4.

Results

Among 135 patients with AML treated with low-intensity therapy, 84 (62.2%) were in the academic center group, while 51 (37.8%) were in the collaborative/community group. Venetoclax with HMAs or LDAC was used in 84 patients (63.6%), while 48 (36.4%) received single-agent therapies, including HMAs or targeted agents. Most patients (85.9%) had ≥1 genes of interest (IDH1, IDH2, TP53, KMT2A rearrangement, or NPM1). A total of 23 patients (17.2%) were treated with low-intensity regimens before allogeneic stem cell transplantation.

Median OS for the entire cohort was 1.2 years (95% CI: 1.0–1.4), with no statistically significant OS difference between the academic and collaborative/community groups. Treatment regimens varied significantly by facility type (p=0.04): academic centers more frequently used venetoclax + azacitidine (37.3% vs. 24%) and “other” therapies (18.1% vs. 6.0%), while the collaborative/community group more commonly used venetoclax + decitabine (24.0% vs. 20.5%) and single-agent HMAs (decitabine: 34.0% vs. 18.1%; azacitidine: 12.0% vs. 6.0%).

Multivariable analysis identified TP53 mutation (p=0.03) and age (p<0.01) as significant predictors of OS. Patients with TP53 mutations had a 73% increased risk of death (HR: 1.73; 95% CI: 1.04–2.89), although their odds of CR/CRi/MLFS were not significantly different from those without the mutation. Each additional year of age was associated with a 4% increased risk of death (HR: 1.04; 95% CI: 1.02–1.06). Lines of treatment (p=0.02) and NPM1 mutations (p<0.01) were significantly associated with response. Patients treated in the second-line setting had a 73% lower likelihood of achieving CR/CRi/MLFS (OR: 0.27; 95% CI: 0.10–0.73), while those with NPM1 mutations had almost four times the odds of achieving CR/CRi/MLFS (OR: 3.93; 95% CI: 1.62–9.56). Although response rates were higher, patients with NPM1 mutations had a comparable OS to the remainder of the AML population.

Conclusion

This analysis found no significant OS differences between patients with AML in the academic group and the collaborative/community clinic group. However, treatment patterns varied significantly: academic centers more often used targeted agents and less frequently used single-agent HMAs. TP53 mutations were associated with worse OS despite comparable response rates. Conversely, NPM1-mutated patients achieved better response rates but had comparable OS to the general AML population. Updated data including genes of interest, cytogenetic abnormalities, and targeted therapies will be reported at the meeting.