# Azacitidine monotherapy versus combination regimens as post-HSCT maintenance therapy in high-risk myeloid malignancies: a retrospective cohort study

## TL;DR

Azacitidine alone works as well as combination treatments for preventing cancer relapse after stem cell transplants in high-risk blood cancer patients.

## Contribution

First direct comparison of azacitidine monotherapy and combination regimens for post-transplant maintenance in myeloid malignancies.

## Key findings

- Azacitidine monotherapy had a 9.1% relapse rate, similar to combination regimens.
- Relapse-free and overall survival rates were comparable across treatment groups.
- Azacitidine monotherapy had a better toxicity profile than combination regimens.

## Abstract

Azacitidine (AZA) monotherapy demonstrates efficacy for post-transplant maintenance in patients with high-risk myeloid malignancies. However, no study has directly compared combination regimens. In this retrospective cohort study, 59 patients (AML = 56, MDS = 3) received AZA-based maintenance post-allogeneic hematopoietic stem cell transplantation (allo-HSCT), as monotherapy (n = 33), AZA plus interferon-α (IFN-α) (n = 15), or AZA plus targeted agents (n = 11). At the median 31-month follow-up, the overall relapse rate was 10.2% (6/59), with comparable rates across groups (AZA: 9.1%, AZA + IFN-α: 13.3%, AZA + targeted: 9.1%) (P = 0.850). Three 3-year relapse-free survival (89.4%; 95%CI 84.7–94.1%) and overall survival (84.4%; 95%CI 78.9–89.9%) rates did not differ significantly between monotherapy and combination regimens (RFS P = 0.975; OS P = 0.770). Overall adverse event rates showed no statistical difference (P > 0.05), although febrile reactions were more common with IFN-α combination regimens (P < 0.001). These findings demonstrate that AZA monotherapy has non-inferior efficacy compared with combination regimens and has a favorable toxicity profile, establishing it as a viable backbone for maintenance therapy in MRD-negative patients. Biomarker-driven combinations warrant prospective validation.