Impacts of IPSS-m and conditioning intensity in allogeneic hematopoietic cell transplant for patients with myelodysplasia neoplasms

Abstract

Abstract Introduction: Allogeneic hematopoietic cell transplantation (allo-HCT) is the only curative treatment for patients with myelodysplastic syndrome (MDS), and multiple factors are reported to associate with the prognosis of MDS patients underwent allo-HCT, including disease burden before allo-HCT, age, comorbidity, and risk stratification. However, Whether the molecular International Prognostic Scoring System (IPSS-M) score calculated at allo-HCT and conditioning intensity have prognostic values for transplant outcomes are still controversial. Methods: We initiated a retrospective, multicenter study of MDS patients aged ≥ 14 years who underwent the first allo-HCT between March, 2019 to March, 2025. Results: 572 subjects were enrolled. The median age was 48 (range, 14-77) years, with a median follow-up of 21.6 (range, 5-76) months after allo-HCT. 374 patients received grafts from haploidentical donors (HID, 65.38%), 140 from matched sibling donors (MSD, 24.48%), and 58 from matched unrelated donors (MUD, 10.9%). 379 patients received myeloablative conditioning (MAC), majorly including busulfan plus with cyclophosphamide (BuCy) based and busulfan plus with fludarabine. 193 patients received reduced intensity conditioning (RIC), including busulfan plus fludarabine and melphalan plus fludarabine. The cumulative incidence of grade II–IV and III-IV acute graft versus host disease (GvHD) by day100 was 22.4% (95% CI: 19.0%-26.0%) and 14.1% (95% CI: 11.3%-17.2%). The 2-year cumulative incidence of total and moderate/severe chronic GvHD was 52.8% (95% CI: 47.9%-57.3%) and 21.4% (95% CI: 17.6%-25.6%). The 3-year probabilities of overall survival (OS), relapses and non-relapse mortality (NRM) were 77.8% (95% CI: 74.1%-81.7%), 18.6% (95% CI: 15.1%-22.5%), and 11.4% (95% CI: 8.94%-14.3%), respectively. In the multivariable model, age ≥ 55 years, the presence of multi-hit TP53mut and/or CK, IPSS-M very high subgroup, transplant conditioning intensity (TCI) [4-6], and HCT comorbidity index (HCT-CI) ≥ 3 are independent risk factors for OS. Age ≥ 55 years, TCI [4-6], and HCT-CI ≥ 3 are risk factors for NRM. The presence of multi-hit TP53mut and/or CK, very high IPSS-M category and bridging allo-HCT are risk factors for relapses. The IPSS-M very high subgroup was associated with inferior survival compared with IPSS-M lower-risk patients (59.0% [95% CI: 48.2%-72.1%] vs. 82.2% [95% CI: 78.2%- 86.3%], P < 0.001). The molecular profile in IPSS-M very high subgroup was heterogeneous, with 23.4% having a TP53mu (92% were TP53multi) and the remaining 77.6% with TP53 wild-type disease commonly had ASXL1, U2AF1, and RUNX1 mutation. IPSS-M very high without multi-hit TP53mut or CK had significant improved survival (OS: 72.0% [95% CI: 61.4%- 88.4%]; relapses: 21.4% [95% CI: 11.4%-33.5%] compared to those with multi-hit TP53mu and/or CK, but inferior to those in other IPSS-M categories. IPSS-M score change from diagnosis to allo-HCT was available for 480 patients. Transplant outcomes were unaffected by IPSS-M score change either in untreated patients or patients receiving pretransplant cytoreductive therapy. No significant differences were observed in OS, relapse and NRM between patients with MAC and those with RIC. We divided the conditioning intensity into low transplant conditioning intensity (TCI [1-2], n = 155, 27.1%), intermediate (TCI [2.5-3.5], n = 262, 45.8%) and high intensity (TCI [4-6], n = 155, 27.1%). The 3-year probabilities of OS were 84.3% (95% CI: 78.2% - 90.9%) vs. 80.6% (95% CI: 75.5% - 86.1%) vs. 66.8% (95% CI: 58.8% -75.9%), respectively, for patients with TCI [1-2], TCI [2.5-3.5], and TCI [4-6]. The 3-year relapse risk showed no difference between the three groups. The 3-year probability of NRM was 5.35% (95% CI: 2.5% -9.8%), 7.98% (95% CI: 5.04%-11.8%), and 23.1% (95% CI: 16.7% - 30.1%), respectively, among the three groups. When stratified by IPSS-M category, the effect of conditioning regiment TCI (TCI [4-6] vs. TCI [1-2]) on OS and NRM was only significant in subgroup with IPSS-M very high (OS: HR, 3.66, 95% CI, 1.39-9.65; NRM: HR, 12.19, 95% CI, 1.56-95.37). Conclusions: Our study showed that upfront allo-HCT conferred superior OS and disease control in MDS patients compared with bridging allo-HCT. The IPSS-M very high subgroup was associated with inferior survival. Low to intermediate TCI might be optimal in MDS patients eligible for allo-HCT with reduced NRM while sparing non-relapse mortality.