Leukemia & Lymphoma, 01/04/2026
Background
Acute Graft-Versus-Host Disease (aGVHD) is a severe complication following allogeneic hematopoietic stem cell transplantation, commonly affecting the skin, liver, and gastrointestinal tract. Glucocorticoids (prednisone or methylprednisolone 1–2 mg/kg/day) are the first-line treatment; however, approximately 40–50% of patients progress to steroid-refractory aGVHD (SR-aGVHD), with an overall survival rate of only 5–30%. Currently, there is no standard second-line therapy; ruxolitinib shows efficacy but with limited response rates.
Mesenchymal stem cells (MSCs) are a promising immunomodulatory therapy, with varying efficacy depending on the source: bone marrow-derived MSCs (BM-MSCs) are widely used (ORR 70.4% at day 28, increasing to 82.8% when combined with basiliximab), iMSCs remain in early-stage development, and placenta-derived MSCs have shown high efficacy in small studies but lack confirmation. Umbilical cord-derived MSCs (UC-MSCs) are a non-invasive source with fewer ethical concerns and strong immunomodulatory capacity; preclinical and early clinical data suggest improved responses, particularly in gastrointestinal involvement, although real-world evidence remains limited.
This study aims to evaluate the efficacy and safety of UC-MSCs as an adjunct to standard therapy in patients with SR-aGVHD, with the hypothesis that they improve early response compared to standard therapy alone.
Patients and Methods
Study design and objectives
This was a retrospective, observational, single-center study. The primary objective was to compare the early complete response (CR) rate at 4 weeks in patients with steroid-refractory aGVHD (SR-aGVHD) between two groups: those treated with umbilical cord-derived mesenchymal stem cells (UC-MSCs) in combination with standard second-line therapy and those receiving standard second-line therapy alone.
Secondary objectives included comparisons between the two groups in overall response rate (ORR), overall survival (OS), progression-free survival (PFS), cumulative relapse rate, incidence of chronic GVHD (cGVHD), and safety.
Patients and definitions
A total of 95 patients diagnosed with steroid-refractory acute GVHD (SR-aGVHD) after allogeneic hematopoietic stem cell transplantation (allo-HSCT) from December 2019 to January 2024 were included in the study. All patients received grafts from HLA-matched related donors (siblings or other family members); HLA-matched unrelated donors were not used due to limitations in regional registry availability.
SR-aGVHD was defined as: disease progression within 3–5 days after starting methylprednisolone at 2 mg/kg/day; failure to achieve partial response (PR) after 7 days; or failure to achieve complete response (CR) or very good partial response (VGPR) after 14 days. aGVHD was graded according to the MAGIC criteria, and risk assessment was performed based on the Minnesota standard.
Treatment groups and UC-MSC administration
Patients were divided into two groups based on treatment approach: the MSC group (n = 44) received intravenous UC-MSCs in combination with second-line standard therapy, while the non-MSC group (n = 51) received only second-line standard therapy. Group assignment was based on actual clinical decisions, including physician assessment, patient characteristics, and treatment accessibility.
UC-MSCs were manufactured at a GMP-compliant facility. Cells were isolated from a single umbilical cord donor using explant culture under serum-free conditions; all infusions for each patient used cells from the same donor.
UC-MSCs were administered intravenously at a dose of 1–3 × 10⁶ cells/kg per infusion. The treatment regimen was individualized; the frequency and total number of infusions were determined by the treating physician based on initial clinical response and GVHD progression. The median cumulative number of infusions was 3 (range 1–8), with a median cumulative dose of 2.71 × 10⁶ cells/kg.
Outcome measures and safety assessment
Treatment response (CR and ORR) was assessed at week 4 after the initiation of SR-aGVHD therapy. Overall survival (OS) was calculated from the time of the first UC-MSC infusion (MSC group) or from the start of second-line therapy (non-MSC group) until death from any cause or the last follow-up.
Progression-free survival (PFS) was defined from the same time points to the first occurrence of an event, including disease progression (GVHD progression or hematologic relapse) or death from any cause.
Safety was evaluated by monitoring vital signs for 24 hours post-infusion and recording adverse events, including infections and cytopenias, throughout the follow-up period.
Data analysis
Statistical analyses were performed using SPSS and R. Appropriate tests were applied for continuous and categorical variables. Kaplan–Meier and log-rank tests were used for OS and PFS analyses, and Cox regression was employed to identify prognostic factors (p ≤ 0.05).
Results
Patient Characteristics
Between December 2019 and January 2024, a total of 95 patients developed SR-aGVHD, including 50 males and 45 females, with a median age of 32 years (range, 6–64). The underlying diseases were mainly leukemia, with AML accounting for 46.3% (n = 44) and ALL for 28.4% (n = 27).
Fifty-three patients underwent haploidentical HSCT (HID-HSCT), with the graft source primarily being a combination of bone marrow and peripheral blood stem cells (BM + PBSCs, 54.7%).
No statistically significant differences were observed between the two groups regarding age, sex, underlying disease, or transplant-related characteristics.
GVHD characteristics
The median onset of aGVHD was 33 days (range, 7–100 days), with the MSC group experiencing earlier onset compared to the non-MSC group (25.5 vs. 40.0 days); however, this difference was not statistically significant (p = 0.347).
Most patients had involvement of two organs (50.5%), with a similar distribution between the two groups (p = 0.199). Overall, t
Response
The median number of MSC infusions in the MSC group was 3 (range, 1–8). The MSC group achieved significantly higher complete response (CR) rates compared with the non-MSC group: for grade II–IV aGVHD, 84.1% vs. 58.8% (p = 0.007), and for severe grade III–IV aGVHD, 54.5% vs. 31.4% (p = 0.034). Overall response rates (ORR) were also significantly higher in the MSC group.
Organ-specific analysis showed markedly higher CR rates in the MSC group compared with the non-MSC group for patients with liver involvement (45.5% vs. 13.7%, p = 0.009) and gastrointestinal involvement (61.4% vs. 33.3%, p = 0.003). Among patients with multi-organ aGVHD (≥2 organs), the MSC group also achieved higher CR rates (59.1% vs. 29.4%, p = 0.017).
Influencing factors
Multivariate Cox regression analysis showed that age, aGVHD severity, and hemoglobin levels were significantly associated with poor prognosis. Specifically, patients aged ≥35 years had a higher risk of adverse outcomes compared with those aged 18–35 years (HR = 5.02; 95% CI: 1.01–24.93; p = 0.049).
Regarding aGVHD grade, patients with grade IV disease had a higher risk of poor prognosis compared with grade III (HR = 5.63; 95% CI: 1.09–29.09; p = 0.039). Additionally, patients with hemoglobin ≤90 g/L had a higher risk than those with >90 g/L (HR = 5.13; 95% CI: 1.57–16.77; p = 0.007).
These results indicate that older age, severe aGVHD, and anemia are independent risk factors affecting patient prognosis.
Incidence of cGVHD
In the MSC group, 41 patients survived more than 100 d post-transplant. Among them, 11 cases developed cGVHD, including 2 extensive types and 9 limited types. In the non-MSC group, 47 patients survived more than 100 d post-transplant, with 19 cases of cGVHD, including 5 extensive types and 14 limited types.
Incidence of viral infections
In the MSC group, there were 9 cases of CMV infection, 21 cases of EBV infection, and 4 cases of co-infection with both viruses.
In the non-MSC group, there were 7 cases of CMV infection, 21 cases of EBV infection, and 3 cases of co-infection. The incidence rates of CMV were 20.1% and 13.7%, while the incidence rates of EBV were 47.7% and 41.2%, with no statistically significant differences between the groups (p > 0.05).
Relapse and survival
In the MSC group, there was 1 case of bone marrow relapse and 1 case of extramedullary relapse. In contrast, the non-MSC group had 9 cases of bone marrow relapse. The relapse rates were 3.5% and 12.8%, respectively (p = 0.019), indicating that the MSC group had a lower relapse rate compared with the non-MSC group.
Analysis of causes of death
The causes of death differed significantly between the two groups. In the MSC group (n = 16), the main causes were severe pneumonia (56.3%) and severe aGVHD (43.8%).
In the non-MSC group (n = 17), deaths were primarily related to bone marrow relapse (52.9%) (p = 0.007). Notably, the rate of death due to severe pneumonia was significantly higher in the MSC group compared with the non-MSC group (56.3% vs. 17.6%, p = 0.032). No significant differences were observed between the two groups for other causes of death, such as organ failure or CMV encephalitis.
Progression-free survival and overall survival
At the end of follow-up, the PFS rate was 65.9% in the MSC group and 62.7% in the non-MSC group (HR = 0.90; 95% CI: 0.47–1.74; p = 0.760). The OS rates were 63.6% and 66.7%, respectively (HR = 1.04; 95% CI: 0.52–2.06; p = 0.911).
These results indicate that although UC-MSCs significantly improved early response in SR-aGVHD, they did not confer a significant long-term benefit in OS or PFS.
Discussion
This real-world retrospective study suggests that adding umbilical cord-derived mesenchymal stem cells (UC-MSCs) to standard therapy can significantly improve early response in patients with SR-aGVHD, particularly in those with severe disease (grade III–IV) and liver or gastrointestinal involvement. However, this benefit did not translate into improved long-term overall survival (OS).
The lower relapse rate in the MSC group (3.5% vs. 12.8%) suggests that UC-MSCs do not impair the graft-versus-leukemia (GVL) effect as previously feared. Proposed mechanisms include selective immunomodulation, transfer of tumor-suppressive miRNAs via extracellular vesicles, and improvement of the bone marrow microenvironment, although these have not been directly demonstrated. The lack of OS improvement may reflect a “risk balance”: reduced mortality from relapse but increased mortality from severe infections (notably pneumonia) related to immunosuppression.
Independent adverse prognostic factors included age ≥35, grade IV aGVHD, and hemoglobin ≤90 g/L. Overall, UC-MSCs appear to be a promising option for early disease control, but their long-term efficacy remains limited. Prospective, multicenter, large-scale trials with biomarker monitoring are needed to clarify their role and underlying mechanisms.
References
Niu, Y., Chen, B., Zhao, Y., Yu, X., Wang, T., & Ma, L. (2026). Umbilical cord-derived mesenchymal stem cells for steroid-refractory acute graft-versus-host disease: a real-world study. Leukemia & Lymphoma, 1–8.
Source: Leukemia & Lymphoma
Link: https://www.tandfonline.com/doi/full/10.1080/10428194.2026.2634172#abstract
