In the present study, we observed that the 2-year probabilities of relapse, NRM, LFS, and OS after allo-HSCT were 17.2%, 4.6%, 78.2%, and 86.3%, respectively. To our knowledge, this was the first and largest study identifying the efficacy and safety of allo-HSCT in adult AML patients with NUP98 rearrangements in the real-world setting.
NUP98-rearranged AML is categorized as intermediate unless adverse cytogenetic and/or molecular abnormalities were concurrently identified 7 in 2022 ELN risk stratification system; however, early case studies 34,35 and later larger cohorts study 36–38 both reported that the clinical outcomes of NUP98-rearranged AML patients were unsatisfactory. This disparity may be due to the fact that AML-NUP98 was rarely reported in Western countries. Recently, a cohort of 809 de novo, non-APL, younger (ages 18–65 years) AML patients receiving standard chemotherapy was used to validate the prognostic significance of ELN-2022, AML with NUP98::HOXA9 rearrangement seemed to have a trend of a higher relapse rate as well as a worse relapse-free survival and a worse OS rates among the ELN-2022 intermediate-risk group, there was no statistically significant just because of the limited sample size (n = 10) 39. Considering that NUP98 rearrangement is cryptic in a substantial portion of cases which might be missed by routine karyotype analysis, fluorescence in situ hybridization (FISH), polymerase chain reaction (PCR), and even RNA sequencing for NUP98 rearrangement screening should be considered in newly diagnosed AML patients.
AML-NUP98 showed a poor response to conventional chemotherapy. In our cohort, all 26 patients showed MRD positive before allo-HSCT, which also supported that standard chemotherapy could not help to achieve deep molecular remission in these patients. Previous studies reported that pre-transplant MRD is the independent prognostic factor for relapse and survival in AML patients 40–42. However, nearly 90% of the patients could achieve persistent molecular remission after HSCT and the relapse rate was only 17.2% in the present study, suggested that pre-transplant NUP98 transcript levels may not significantly impact the post-transplant relapse. This may suggest that the graft-versus-leukemia (GVL) may help to further clear the residual leukemia after allo-HSCT.
Most of the patients (24/26, 92.3%) received haploidentical related donor (HID) HSCT in the present study. This also suggested that HID HSCT is safe and effective for these patients. On the other hand, HID HSCT may show a stronger GVL effect than matched sibling donor (MSD) HSCT 43, and several studies showed that HID HSCT could achieve a lower relapse rate and a better survival rate compared with MSD HSCT in high-risk AML including those with MRD positive before HSCT 44–46. Because only 2 patients received non-HID HSCT, we could not further compare the clinical outcomes among different donor types.
The incidences of total aGVHD and cGVHD were 58.0% and 47.5%, respectively, in our study, which seemed to be comparable with other studies enrolled HID HSCT recipients 45,47. However, the incidences of grade III to IV aGVHD and severe cGVHD were only 11.5% and 27.2%, respectively. Because of the great progress in GVHD treatment, no one died of GVHD. Considering the concomitant GVL effects, the occurrence of manageable GVHD is acceptable in these high-risk AML.
Previous studies observed FLT3-ITD mutations were not rare in patients with NUP98 rearrangements 17,48. FLT3 inhibitor have demonstrated promising efficacy in NUP98-rearranged laboratory models, including mouse model with HSPCs bearing NUP98-HOXD13 and FLT3-ITD49. AML patients with NUP98::NSD1 who showed poor response initial treatment could benefit from FLT3 inhibitors in some reports 50. FLT3 inhibitor (e.g., sorafenib) has been extensively studied as post-transplant maintenance treatment in FLT3-mutated AML, showed significant benefit in relapse reduction and survival improvement, and has been recommended by international guidelines 51–53. In our cohort, ten patients (38.5%) with FLT3-ITD mutations simultaneous received sorafenib or a more selected FLT-3 inhibitor, gilteritinib, as the maintenance therapies. The relapse rate of maintenance therapy group was 10 percent lower than those without maintenance therapy in the present study, and without statistical difference may only because of the small sample size.
Our study is limited by the retrospective designed, small sample size, and a relatively short follow-up. Future prospective studies could further identify the clinical outcomes of allo-HSCT in adult AML patients with NUP98 rearrangement in larger cohorts with a longer follow-up.