AFFILIATIONS
1 1. Peking University People’s Hospital, Peking University Institute of Hematology, Beijing, China 2. National Clinical Research Center for Hematologic Disease, Beijing, China 3. Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China; 4. Collaborative Innovation Center of Hematology, Peking University Institute of Hematology, Beijing, China, Beijing, China
2 1. Peking University People’s Hospital, Peking University Institute of Hematology, Beijing, China 2. National Clinical Research Center for Hematologic Disease, Beijing, China 3. Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China; 4. Collaborative Innovation Center of Hematology, Peking University Institute of Hematology, Beijing, China, Beijing, China
3 1. Peking University People’s Hospital, Peking University Institute of Hematology, Beijing, China 2. National Clinical Research Center for Hematologic Disease, Beijing, China 3. Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China; 4. Collaborative Innovation Center of Hematology, Peking University Institute of Hematology, Beijing, China, Beijing, China
4 1. Peking University People’s Hospital, Peking University Institute of Hematology, Beijing, China 2. National Clinical Research Center for Hematologic Disease, Beijing, China 3. Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China; 4. Collaborative Innovation Center of Hematology, Peking University Institute of Hematology, Beijing, China, Beijing, China
5 The First Affiliated Hospital of Harbin Medical University, Harbin, China
6 Xiangya Hospital of Central South University, Changsha, China
7 The Third Xiangya Hospital of Central South University, Changsha, China
8 The First Affiliated Hospital of Zhengzhou University, Zhengzhou, China
9 First Hospital of Jilin University, Jilin, China
10 Chinese Academy of Medical Sciences & Peking Union Medical College, Tianjin, China
11 Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China
12 Sichuan Provincial People's Hospital, Chengdu, China
13 Guangzhou First People's Hospital, Guangzhou, China
14 Guangdong Provincial Hospital of Chinese Medicine, Guangzhou, China
15 1. Peking University People’s Hospital, Peking University Institute of Hematology, Beijing, China 2. National Clinical Research Center for Hematologic Disease, Beijing, China 3. Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China; 4. Collaborative Innovation Center of Hematology, Peking University Institute of Hematology, Beijing, China, Beijing, China
16 1. Peking University People’s Hospital, Peking University Institute of Hematology, Beijing, China 2. National Clinical Research Center for Hematologic Disease, Beijing, China 3. Beijing Key Laboratory of Cell and Gene Therapy for Hematologic Malignancies, Peking University, Beijing, China; 4. Collaborative Innovation Center of Hematology, Peking University Institute of Hematology, Beijing, China, Beijing, China
ABSTRACT
Introduction: Engraftment syndrome (ES) is a peri-engraftment inflammatory complication after allogeneic hematopoietic stem cell transplantation (allo-HSCT), and it occurs around the time of neutrophil recovery (J Hematol Oncol, 2021; Eur J Hematol, 2016; Bone Marrow Transplant, 2001). The clinical heterogeneity and prognostic significance of ES after allo-HSCT remain unclear (Bone Marrow Transplant, 2022; Bone Marrow Transplant, 2023). In addition, the clinical features of ES overlap with those of early acute graft-versus-host disease (aGVHD). However, the temporal relationship between ES and aGVHD remains uncertain. We conducted a multicenter retrospective study to characterize the clinical presentation and prognostic factors of ES after allo-HSCT, evaluate the temporal association between ES and grade II–IV aGVHD, and identify the unsupervised ES phenotypes in relation to overall survival (OS) and non-relapse mortality (NRM).
Methods: A total of 680 ES patients who underwent allo-HSCT at 11 large transplant centers in China between January 2016 and October 2025 and 2040 matched controls who did not develop ES after allo-HSCT were enrolled in this study. Patients with ES were divided into three cohorts for unsupervised phenotyping development and validation. The derivation cohort included 305 patients treated at Peking University People’s Hospital from 2020 to 2025. The same-center temporal validation cohort included 191 patients treated between 2016 and 2019, and the external validation cohort included 184 patients treated at the remaining 10 centers. Gaussian mixture modeling (GMM) was used to identify unsupervised ES phenotypes, and the ES-GMM model incorporated eight continuous variables, including laboratory markers during conditioning and hematopoietic recovery kinetics. The prognostic relevance of ES-GMM phenotypes was assessed for OS and NRM. Time-dependent Cox regression models were constructed to evaluate whether ES was associated with an increased risk of subsequent grade II–IV aGVHD, and dynamic-state analyses were conducted to assess the prognostic impact of the coexistence of ES and grade II–IV aGVHD.
Results: In the multivariate Cox regression models, delayed platelet engraftment was independently associated with worse OS among patients who developed ES after allo-HSCT (hazard ratio [HR] = 1.014; 95% confidence interval [CI]: 1.004–1.025; P = 0.007). The ES-GMM model revealed two clinically interpretable phenotypes in the derivation cohort, namely a delayed-engraftment high-risk phenotype and an earlier-engraftment lower-risk phenotype. The high-risk group had inferior OS (HR = 3.07, 95% CI: 1.53–6.17; log-rank P < 0.001) and higher NRM (subdistribution HR [SHR] = 3.72, 95% CI: 1.58–8.79; Gray’s test P = 0.001) in the derivation cohort. Similar adverse associations were observed in the same-center temporal validation cohort for OS (HR = 2.39, 95% CI: 1.09–5.23; log-rank P = 0.024) and NRM (SHR = 2.54, 95% CI: 1.09–5.95; Gray’s test P = 0.029) and in the external validation cohort for OS (HR = 2.50, 95% CI: 1.20–5.18; log-rank P = 0.011) and NRM (SHR = 2.89, 95% CI: 1.22–6.83; Gray’s test P = 0.013). In the time-dependent analyses, ES was associated with an increased hazard of subsequent grade II–IV aGVHD during the first 30 days after allo-HSCT in the fully adjusted model (HR = 2.95, 95% CI: 2.12–4.10; P < 0.001). Dynamic-state analyses further revealed that ES alone was not significantly associated with increased mortality, and patients who developed both ES and grade II–IV aGVHD had the worst OS (HR = 1.68, 95% CI: 1.00–2.81; P = 0.049) and NRM (HR = 4.49, 95% CI: 2.80–7.22; P < 0.001) outcomes in the fully adjusted model.
Conclusions: ES after allo-HSCT is clinically heterogeneous and prognostically relevant. Delayed platelet engraftment was independently associated with worse OS and constituted a key feature of the delayed-engraftment high-risk ES phenotype identified through unsupervised ES-GMM analysis. This phenotype was associated with worse OS and higher NRM, with reproducible prognostic patterns across the validation cohorts. ES was also linked to an increased early hazard of subsequent grade II–IV aGVHD, and the combined ES + aGVHD state identified patients with the poorest outcomes. These findings highlight hematopoietic recovery kinetics as a clinically meaningful basis of ES risk stratification and support a phenotype-based approach to understanding ES after allo-HSCT.