预处理强度对MDS患者异基因移植结局的影响:多中心回顾性队列研究
该多中心回顾性队列纳入337例接受allo-HSCT的MDS患者,按TCI评分分为低、中、高强度三组。高强度组3年NRM显著高于低、中强度组,3年OS率最低,而三组3年累积复发率无显著差异。结论提示高强度预处理未降低复发,反而增加NRM并损害OS。
要点
- 高强度预处理(TCI-3)显著增加MDS患者allo-HSCT后非复发死亡风险。
- TCI-3组3年OS率最低(61.2%),三组间3年累积复发率无显著差异。
- 低、中强度预处理在控制毒性与维持疗效间取得较优平衡。
核心结论
- 高强度预处理未能降低复发率,反而显著增加allo-HSCT后NRM风险并损害OS。
- 低、中强度预处理在控制毒性与维持疗效之间取得较优平衡。
研究设计与样本
- 多中心回顾性队列研究,纳入2019年2月至2025年4月国内4家移植中心接受allo-HSCT的337例MDS患者。
- 依据TCI评分分为TCI-1(1~2分,低强度)、TCI-2(2.5~3.5分,中等强度)和TCI-3(4~6分,高强度)三组。
- 采用Fine-Gray竞争风险模型分析NRM和复发,Kaplan-Meier法计算OS并用Log-rank比较,IPTW均衡协变量,DR模型结合Fine-Gray评估TCI对NRM的独立影响。
主要发现
- 未加权分析:TCI-1、TCI-2和TCI-3组移植后3年NRM分别为7.7%(95% CI:0.4%~15.1%)、8.3%(95% CI:4.3%~12.3%)和27.1%(95% CI:18.0%~36.3%)(P=0.008)。
- TCI-1与TCI-2组NRM风险差异无统计学意义(sHR=1.106,95% CI:0.371~3.298,P=0.857);TCI-3组NRM风险显著高于TCI-1组(sHR=4.023,95% CI:1.414~11.452,P=0.009)和TCI-2组(sHR=3.673,95% CI:1.951~6.912,P<0.001)。
- 多因素分析显示TCI-3、年龄≥50岁和HCT-CI评分>2分是NRM的独立危险因素。
- 经IPTW校正后,TCI-3组NRM风险仍显著高于TCI-1组(sHR=6.090)和TCI-2组(sHR=4.562);DR校正后TCI-3仍为NRM独立危险因素(sHR=5.011,95% CI:1.118~22.452,P=0.035)。
- TCI-3组3年OS率最低(61.2%,95% CI:51.5%~72.7%);三组间3年累积复发率差异无统计学意义;移植后3年NRM对移植失败的贡献率为39.0%。
关键数值与推荐
- TCI评分分组:TCI-1为1~2分(低强度),TCI-2为2.5~3.5分(中等强度),TCI-3为4~6分(高强度)。
- NRM独立危险因素:TCI-3、年龄≥50岁、HCT-CI评分>2分。
临床怎么用
- MDS患者allo-HSCT预处理选择时,应避免高强度预处理(TCI-3),因其增加NRM并损害OS而未降低复发。
- 低、中强度预处理(TCI-1或TCI-2)在控制毒性与维持疗效间平衡更优,可作为优先考虑。
- 对年龄≥50岁或HCT-CI评分>2分的高危患者,更应谨慎评估预处理强度。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
Objective: To investigate the effect of transplant conditioning intensity (TCI) on transplant outcomes in patients with myelodysplastic neoplasms (MDS) undergoing allogeneic hematopoietic stem cell transplantation (allo-HSCT) . Methods: This multicenter retrospective cohort study included 337 patients with MDS undergoing allo-HSCT between February 2019 and April 2025 in four transplant centers in China. Patients were categorized based on TCI scores into TCI-1 (1-2 points, low intensity), TCI-2 (2.5-3.5 points, moderate intensity), and TCI-3 (4-6 points, high intensity) groups. The Fine-Gray competing risk model was applied to analyze nonrelapse mortality (NRM) and relapse. Overall survival (OS) was estimated employing the Kaplan-Meier method and compared using the log-rank test. Inverse probability of treatment weighting (IPTW) was performed to balance covariates among groups to eliminate baseline confounding bias. Furthermore, a doubly robust (DR) model combined with the Fine-Gray competing risk model was applied to assess the independent effect of TCI on NRM. Results: Unweighted analysis revealed that the 3-year NRM post-transplantation in the TCI-1, TCI-2, and TCI-3 groups was 7.7% (95% CI: 0.4%-15.1%), 8.3% (95% CI: 4.3%-12.3%), and 27.1% (95% CI: 18.0%-36.3%), respectively (P=0.008). No significant difference in the risk of NRM was observed between the TCI-1 and TCI-2 groups (sHR=1.106, 95% CI: 0.371-3.298, P=0.857), whereas the risk of NRM in the TCI-3 group was significantly higher than that in the TCI-1 (sHR=4.023, 95% CI: 1.414-11.452, P=0.009) and TCI-2 groups (sHR=3.673, 95% CI: 1.951-6.912, P<0.001). The multivariable analysis identified TCI-3, age of ≥50 years, and hematopoietic cell transplantation-comorbidity index score of >2 points as independent risk factors for NRM. After IPTW adjustment, the results remained consistent: the risk of NRM in the TCI-3 group remained significantly higher than that in the TCI-1 (sHR=6.090) and TCI-2 groups (sHR=4.562). The comparison of the contribution of NRM to transplant failure across groups at different time points revealed a significantly increased contribution of NRM to transplant failure with increasing conditioning intensity, especially in the TCI-3 group (sHR=5.011, 95% CI: 1.118-22.452, P=0.035). NRM within 3 years posttransplantation accounted for 39.0% of transplant failures. The Kaplan-Meier curve revealed that the 3-year OS rate was the lowest in the TCI-3 group (61.2%, 95% CI: 51.5%-72.7%). No statistically significant difference in the 3-year cumulative incidence of relapse was observed across the three groups. Conclusion: High-intensity conditioning failed to translate into a reduction in relapse rate but rather significantly increased the risk of NRM and compromised OS after allo-HSCT. Low- and moderate-intensity conditioning achieves a better balance between controlling toxicity and maintaining efficacy. 目的: 探讨不同移植预处理强度(TCI)对骨髓增生异常肿瘤(MDS)患者异基因造血干细胞移植(allo-HSCT)结局的影响。 方法: 本研究为多中心回顾性队列研究,纳入2019年2月至2025年4月国内4家移植中心接受allo-HSCT的337例MDS患者。依据TCI评分将预处理强度分为TCI-1(1~2分,低强度)、TCI-2(2.5~3.5分,中等强度)、TCI-3(4~6分,高强度)三组。采用Fine-Gray竞争风险模型分析非复发死亡和复发,Kaplan-Meier法计算总生存(OS)率,并用Log-rank进行比较。为消除基线混杂偏倚,采用逆概率加权法(IPTW)均衡组间协变量;在此基础上,进一步采用双重稳健模型(DR)结合Fine-Gray竞争风险模型,评估TCI对非复发死亡的独立影响。 结果: 未加权分析显示,TCI-1、TCI-2和TCI-3三组移植后3年非复发死亡率(NRM)分别为7.7%(95% CI:0.4%~15.1%)、8.3%(95% CI:4.3%~12.3%)和27.1%(95% CI:18.0%~36.3%)(P=0.008)。TCI-1和TCI-2两组发生非复发死亡的风险之间差异无统计学意义(sHR=1.106,95% CI:0.371~3.298,P=0.857),TCI-3组发生非复发死亡的风险显著高于TCI-1组(sHR=4.023,95% CI:1.414~11.452,P=0.009)和TCI-2组(sHR=3.673,95%CI:1.951~6.912,P<0.001)。多因素分析显示TCI-3、年龄≥50岁和造血干细胞移植合并症指数(HCT-CI)评分>2分是非复发死亡的独立危险因素。经IPTW校正后,结果保持一致:TCI-3组患者的非复发死亡风险仍显著高于TCI-1组(sHR=6.090)和TCI-2组(sHR=4.562)。进一步采用DR校正潜在残余混杂后,TCI-3仍是非复发死亡的独立危险因素(sHR=5.011,95%CI:1.118~22.452,P=0.035)。比较不同时间点各组非复发死亡对移植失败的贡献率,随着预处理强度增加,非复发死亡对移植失败的贡献率呈明显上升趋势,尤其在TCI-3组中表现最为突出。移植后3年非复发死亡对移植失败的贡献率为39.0%。Kaplan-Meier曲线显示TCI-3组3年OS率最低(61.2%,95% CI:51.5%~72.7%)。三组间3年累积复发率差异无统计学意义。 结论: 高强度预处理未能转化为复发率的下降,反而显著增加allo-HSCT后非复发死亡风险并损害OS。低、中强度预处理能在控制毒性与维持疗效之间取得较优平衡。.
引用
- Wang L, Zhang ZL, Zhang RX, et al. [Impact of conditioning intensity on transplant outcomes following allogeneic hematopoietic stem cell transplantation in patients with myelodysplastic neoplasms: a multicenter retrospective cohort study]. Zhonghua Xue Ye Xue Za Zhi. 2026. PMID: 41991307. doi:10.3760/cma.j.cn121090-20251029-00490
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。