2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Saturday, December 12, 2026
04:00 p.m. - 05:30 p.m.
MCCNO - 295-296
Treatment Considerations, Combination Therapy, Drug Development, Biological Therapies, Antibody Therapy, Adverse Events, Clinical Practice (Health Services and Quality), Patient Safety, Diseases, Bone Marrow Failure Syndromes, Acquired Marrow Failure Syndromes , Aplastic Anemia, Paroxysmal Nocturnal Hemoglobinuria, Research, Clinical Research, Clinical Trials, Epidemiology, Real-World Evidence, Study Population, Human, Adult
PAPER

349 A randomized trial of sirolimus for relapse prevention in patients with severe aplastic anemia responsive to immunosuppressive therapy

Fieke W Hoff, MD, PhD, Luiz Fernando Bazzo Catto, MD, PhD2, Ruba Shalhoub, MS3, Xingmin Feng, PhD4, Jennifer Lotter, MS5, Nicholas Chien-Juei Lee, MD6, Ivana Darden, RN7, Colin Wu, PhD8, Emma Groarke, MD9, Neal Stuart Young, MD10, Bhavisha A Patel, MD11
1 National Institutes of Health, Bethesda, United States
2 National Institutes of Health, Bethesda, United States
3 National Institutes of Health, Bethesda, United States
4 National Institutes of Health, Bethesda, United States
5 National Institutes of Health, Bethesda, United States
6 National Institutes of Health, Bethesda, United States
7 National Institutes of Health, Bethesda, United States
8 National Institutes of Health, Bethesda, United States
9 National Institutes of Health, Bethesda, United States
10 National Institutes of Health, Bethesda, United States
11 National Institutes of Health, Bethesda, United States

Background: Initial response to standard immunosuppressive therapy (IST) with horse antithymocyte globulin (h-ATG), cyclosporine (CSA) and eltrombopag (EPAG) in severe aplastic anemia (SAA) is high, with overall response rates achieved in 60-80% of cases. However, about 30-40% of patients relapse after discontinuation of CSA. This randomized, controlled, phase 2 clinical trial investigated the safety and efficacy of sirolimus (SRL) for preventing relapse in patients previously treated with IST who remained on CSA. The trial was registered at as #NCT02979873.

Methods: Patients aged ≥ 2 years (yrs) with a history of SAA, who were treated with h-ATG+CSA+EPAG (93%), h-ATG+CSA (6%), or cyclophosphamide (1%), achieved hematologic response and were on maintenance CSA for >6 months were eligible. Patients were randomized 1:1 to receive SRL or no treatment (observation (OBS)). After CSA discontinuation (Day 0), SRL was initiated on Day 1 at 6 mg, followed by 3 months of maintenance dose of 2 mg (therapeutic goal: 5-15 ng/mL). The primary endpoint was rate of relapse at 2 yrs following conversion from CSA to SRL compared to stopping CSA. Secondary endpoints included safety and tolerability of SRL, time to relapse, overall relapse rate, blood cell counts, clonal evolution to myeloid neoplasm, and overall survival (OS). Due to slow accrual and futility analysis, the study closed after reaching 80% planned enrollment (n=84).

Results: A total of 84 patients were enrolled between December 2016 and September 2024; 42 patients were assigned to receive SRL, and 42 patients OBS. The 2 cohorts were well-balanced for demographics and disease characteristics; median age was 38 yrs (range, 7-80 yrs). Ninety percent (n=38) of patients completed 3 months of SRL treatment. Three patients discontinued SRL because of toxicities; headache (grade 3 (n=2), grade 2 (n=1)), nausea (grade 2 (n=2), mucositis (grade 2 (n=1), fatigue (grade 2 (n=1)), all possibly drug-related. Other grades 3 and 4 adverse events were similar between cohorts and were disease related.

The 2-yrs relapse rate was 17% (n=8) in the SRL cohort vs 26% (n=11) in the OBS cohort (p=0.390). The median time to relapse was longer in the SRL cohort (7.2 vs 2.8 months, p=0.15). An additional 5 relapses occurred at median time 43.7 months (range 25.6 to 81.5), 4/5 were in OBS cohort. Most relapses were associated with a decline in all hematopoietic lineages. Univariate cox regression analysis identified baseline Hb (hazard ratio (HR, 0.56 [0.40-0.79]) and older age, a known risk factor for relapse, as significant predictors of relapse.  

Baseline blood counts at study enrollment were similar between cohorts. Hemoglobin (Hb) and platelet count (PLT) decreased at 3 months in the SRL cohort (both p=<0.001) but returned to baseline by 1 year, with no changes observed in the OBS cohort. EPO and TPO levels showed reciprocal transient increases, suggesting compensation. Cytokine levels and lymphocyte phenotyping were not affected by SRL. However, we found that relapsed patients (n=23) had significantly higher baseline FLT3L (p=0.012) and PDL1 (p=0.016) levels compared to non-relapsed patients, but there was no difference between SRL and OBS arms. FLT3L (p<0.001) and PDL1 (p=0.007) increased further at 3 months, and at relapse in relapsed patients.

Six patients developed either clonal evolution (n=1) or clinically significant paroxysmal nocturnal hemoglobinemia (n=5); all occurred in the OBS group (p=0.009). One patient developed myelodysplastic syndrome with increased blasts after 23 months. At a median follow-up of 5.4 yrs (IQR, 3.4-7.8), the 5-yrs OS was 95.7% and did not differ between cohorts (p=0.55)

Conclusion: This phase 2 randomized trial investigated short duration SRL as a relapse prevention strategy in patients who remain on CSA after achieving a response to definitive IST. SRL resulted in a nonsignificant reduction in 2-yrs relapse rate and delayed time to relapse. All clinically significant PNH expansion and clonal evolution events occurred in the OBS cohort despite balanced baseline characteristics, suggesting a possible biologic effect of mTOR inhibition on clonal selection warranting further investigation. We identified residual anemia after successful IST as a predictor of relapse and higher levels of FLT3L and PDL1 in patients who relapsed; these variables could help risk stratify patients for longer duration of IST in future relapse prevention strategies.

Fieke W Hoff, MD, PhD: No relevant conflicts of interest to declare.