AUTHORS
Jintao Lin, Zhen Cai, MD,PhD2, Donghua He, MD,PHD3, Jingsong He, PhD4, Gaofeng Zheng5, Jing Chen6, Li Yang, MD,PHD7
AFFILIATIONS
1 Bone Marrow Transplantation Center, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
2 Bone Marrow Transplantation Center, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
3 Bone Marrow Transplantation Center, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
4 Bone Marrow Transplantation Center, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
5 Bone Marrow Transplantation Center, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
6 Bone Marrow Transplantation Center, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
7 Bone Marrow Transplantation Center, The First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, China
ABSTRACT
Background
In the era of novel anti-myeloma agents, autologous hematopoietic stem cell transplantation (ASCT) remains the first-line standard of care for eligible patients with multiple myeloma (MM), yet post-transplant relapse persists as a major clinical challenge. Over the past two decades, the prognostic potential of graft minimal residual disease (gMRD) has been extensively explored, but conclusions remain controversial and its clinical utility has not been fully defined. Importantly, autologous grafts contain diverse hematopoietic and immune cell subsets implicated in post-transplant outcomes, yet whether graft cellular composition confounds the prognostic effect of gMRD remains largely uncharacterized. This study aimed to investigate the association of gMRD status with response depth and long-term prognosis in MM patients undergoing ASCT, confirm its independent prognostic significance via systematic assessment of the potential confounding role of graft cellular composition, and further evaluate its prognostic stratification performance and clinical utility when integrated with conventional risk stratification systems.
Methods
This study is a prospective, multi-center cohort study that enrolled 176 MM patients who received their first ASCT. 10-color flow cytometry (with a sensitivity of 10?6) was used to detect gMRD. Receiver operating characteristic (ROC) curve analysis was used to evaluate the discriminative performance of eight core graft cellular subsets for gMRD positivity at three predefined cutoffs (1×10?4, 1×10?5, 1×10?6). Univariate and multivariate Cox proportional hazards regression models were constructed to identify independent prognostic factors for progression-free survival (PFS) and overall survival (OS). Firth’s penalized partial likelihood approach was applied for OS analysis to mitigate small-sample bias. Subgroup analyses and combined risk stratification with conventional prognostic indicators were also performed.
Results
The cutoff of 1×10?6 was determined as optimal for prognostic stratification. Using this threshold, 117 patients (66.5%) were classified as gMRD-negative and 59 (33.5%) as gMRD-positive, with well-balanced baseline characteristics across groups. Survival analyses further revealed that gMRD-negative patients had significantly better PFS (p=0.004), and the OS curves of the two groups also showed an obvious separation trend. Three months after transplantation, the rate of complete response/strict complete response (CR/sCR) in gMRD-negative patients was significantly higher than that in gMRD-positive patients, while gMRD-positive patients were more likely to achieve very good partial response (VGPR) (p<0.001). At the 1×10?6 cutoff, none of the eight graft cellular subsets exhibited statistically significant discriminative capacity for gMRD status. After multivariate adjustment for baseline characteristics, conventional risk factors, and graft cellular components, gMRD positivity remained an independent adverse prognostic factor for both PFS (HR = 4.82, P = 0.004) and OS (HR = 6.36, P = 0.033). High-risk cytogenetic abnormalities (HRCA) and elevated graft monocyte proportion were independently associated with shorter PFS, while CD34? cell infusion > 2.5 × 106 cells/kg and higher graft dendritic cell proportion were independent protective factors for PFS and OS, respectively. Subgroup analysis further confirmed that gMRD maintains stable prognostic prediction value in most clinical subgroups including HRCA, R-ISS stage I II, R ISS stage III, CD34? cell infusion > 2.5 × 106/kg and age ≥60 years, and can achieve refined risk re-stratification for high-risk populations.
Conclusion
gMRD serves as a robust, independent prognostic biomarker in MM patients undergoing first-line ASCT, with prognostic value that is independent of graft cellular composition. Integration of gMRD into conventional risk stratification frameworks substantially enhances prognostic accuracy, providing a valuable basis for individualized risk stratification and post-transplant treatment optimization.
DISCLOSURES
Jintao Lin: No relevant conflicts of interest to declare.