Introduction

In combination with a calcineurin inhibitor, mycophenolate mofetil (MMF) is used for the prophylaxis of graft-versus-host disease (GVHD) after allogeneic hematopoietic cell transplantation (HCT). Compared to intravenous methotrexate (MTX), MMF is associated with a lower incidence of mucositis and shorter time to hematopoietic engraftment but comparable incidence of acute GVHD^1–3, resulting in the preferred use of MMF for GVHD prophylaxis in elderly patients^4–6 or those undergoing cord blood transplantation (CBT)^5–8. In contrast to other countries, MMF has only been approved as an oral formulation in Japan. However, some patients are unable to take oral MMF mainly because of regimen-related toxicity (RRT). Although several studies have evaluated the clinical impact of omitting planned MTX due to toxicity after allogeneic HCT^9–15, no study has evaluated the clinical impact of interrupting planned oral MMF on GVHD prophylaxis. Therefore, in this study, we investigated the clinical influence of oral MMF interruption on CBT outcomes at our hospital.

Materials and Methods

Patients and transplant procedures

We retrospectively analyzed the consecutive data of 53 adult patients who underwent single-unit unrelated CBT and received planned oral MMF in combination with cyclosporine for GVHD prophylaxis between November, 2013 and March, 2023 at our hospital. GVHD prophylaxis consisted of intravenous cyclosporine (3 mg/kg/day from day -1) and oral MMF (30 mg/kg/day from days 0 to 27)¹⁶. Unrelated cord blood was supplied by cord blood banks in Japan. The cord blood unit, conditioning regimen, GVHD prophylaxis, and supportive care were determined by the treating physicians^16–20. The Institutional Review Board of the Institute of Medical Science, the University of Tokyo approved this retrospective study (2023-33-0810) and the adoption of an opt-out consent mechanism.

Definitions

Neutrophil recovery was defined as the recovery achieved on the first three consecutive days when the absolute neutrophil count was higher than 0.5 × 10⁹/L. Platelet recovery was defined as that achieved on the first seven consecutive days when the platelet count was higher than 20 or 50 × 10⁹/L without platelet transfusion support. Diagnosis and grading of acute and chronic GVHD were based on the standard criteria^{21, 22}. Overall survival (OS) was defined as the time from CBT to death, subsequent allogeneic HCT, or the date of last contact with patients who were lost to follow-up. Non-relapse mortality (NRM) was defined as the time from CBT to death without disease relapse. The number of human leukocyte antigen (HLA) disparities between the cord blood grafts and recipients was defined as low resolution for HLA-A, HLA-B, and HLA-DR in the graft-versus-host direction. HCT-specific comorbidity index (HCT-CI)²³ and refined disease risk index (rDRI)²⁴ were classified according to published criteria.

Statistical analyses

Groups were compared using the Mann-Whitney U test for continuous variables and Fisher's exact test for categorical variables. The effects of MMF interruption on OS and NRM were graphically illustrated using the Simon-Makuch plots. Multivariate analysis was conducted using the Cox proportional hazards model for overall mortality, and the Fine and Gray model for NRM, neutrophil recovery, platelet recovery, grades II-IV acute GVHD, grades III-IV acute GVHD, overall chronic GVHD, and extensive chronic GVHD. Multivariate analysis involved the following factors as covariates: MMF interruption (yes vs. no), which was treated as a time-dependent covariate, age (<65 vs. ≥65 years), gender (male vs. female), HCT-CI (<3 vs. ≥3), rDRI (low/intermediate vs. high/very high), cryopreserved cord blood total nucleated cell (TNC) dose (<2.5 × 10⁷/kg vs. ≥2.5 × 10⁷/kg), and low-resolution HLA disparities in the graft-versus-host direction (0, 1 vs. 2). P-values < 0.05 were considered to be statistically significant. Statistical analyses were conducted using EZR version 1.61²⁵.

Results

Patient characteristics

All patient and CBT characteristics are presented in Table 1. The median age of the entire cohort was 63 years (interquartile range [IQR], 60-66 years). The most common disease was acute myeloid leukemia (57%). Disease risk defined by rDRI was high or very high in 75% of the patients. The majority of conditioning regimens included 180 mg/m² fludarabine, 9.6 mg/kg intravenous busulfan, 4 Gy total body irradiation, and 12 g/m² high-dose cytarabine (83%)¹⁶. The median cryopreserved cord blood TNC dose was 2.52 × 10⁷/kg (IQR, 2.21-3.24 × 10⁷/kg), and the median cryopreserved cord blood CD34⁺ cell dose was 0.99 × 10⁵/kg (IQR, 0.79-1.23 × 10⁵/kg). Ten patients (19%) had previously undergone allogeneic HCT.

Among the 53 patients, the planned dose of MMF was interrupted in 14 with a median of 19.5 d (range, 3-27 d) of CBT. The patients in whom the planned dose of MMF was interrupted were young (P=0.031) and had previously undergone allogeneic HCT (P=0.014). The main causes of MMF interruption were mucositis and vomiting due to RRT (n=8), general malaise due to organ failure or infection (n=3), alveolar hemorrhage (n=1), encephalitis (n=1), and engraftment failure (n=1). No additional immunosuppressants were administered to the patients during MMF interruption.

Association of MMF interruption with hematopoietic recovery and GVHD

In Fisher's exact test, MMF interruption was associated with lower platelet recovery rate, which was defined as ≥ 50,000/μL (P=0.024), but not the rates of neutrophil recovery (P=0.220), grades II-IV acute GVHD (P=0.140), grades III-IV acute GVHD (P=0.181), overall chronic GVHD (P=0.315), and extensive chronic GVHD (P=1.000; Table 2). In the multivariate analysis, MMF interruption, which was treated as a time-dependent covariate, was significantly associated with lower platelet recovery, which was defined as ≥ 20,000/μL (hazard ratio [HR], 0.26; 95% confidence interval [CI], 0.12-0.58; P = 0.001) and ≥ 50,000/μL (HR, 0.27; 95% CI, 0.10-0.70; P = 0.006; Table 3), but not neutrophil recovery (Table 3), acute GVHD, and chronic GVHD (Table 4).

Impact of MMF interruption on OS and NRM

In univariate analysis, MMF interruption, which was treated as a time-dependent covariate, was significantly associated with poorer OS and higher NRM (Figure 1). Multivariate analysis also revealed that MMF interruption was significantly associated with poorer OS (HR, 5.41; 95% CI, 2.03-14.43; P < 0.001) and higher NRM (HR, 7.56; 95% CI, 1.99-28.79; P = 0.002; Table 5).

Among the 14 patients in whom the planned dose of MMF was discontinued, 10 died during the last follow-up. The causes of death were pneumonia in 3 patients, relapse in 2, alveolar hemorrhage in 1, gastrointestinal hemorrhage in 1, acute GVHD in 1, multiple organ failure in 1, and sepsis in 1.

Discussion

Previous studies have demonstrated the clinical impact of MTX omission day 11 due to toxicity after allogeneic HCT; however, their results are controversial, mainly because of the small sample size^9–15. A recent meta-analysis by Kharfan-Dabaja et al. demonstrated that day 11 omission of MTX was associated with poor OS, but not NRM, in acute or chronic GVHD¹⁵. For GVHD prophylaxis, MMF is started at 15-45 mg/kg orally or intravenously twice or thrice a day starting on day 0 and continued for 27-40 d until termination or tapered down through days 96-180^26–30. However, the ideal MMF concentration, dosage schedule, and treatment duration for GVHD prevention remain unclear^{31, 32}. Our study is the first to evaluate the clinical impact of interrupting planned oral MMF treatment on transplant outcomes. We found that MMF interruption led to poor platelet recovery, poor OS, and high NRM after CBT, but did not affect the incidence of acute and chronic GVHD. However, our results should be interpreted cautiously as most patients with interrupted MMF did not take any other oral drugs or diet. Poor oral intake alters the microbiota diversity and composition, resulting in a high incidence of gastrointestinal GVHD and poor clinical outcomes^{33, 34}. Therefore, although intravenous MMF is safe and effective for GVHD prophylaxis^{35, 36}, whether the use of intravenous MMF overcomes the negative effects of oral MMF interruption remains unclear.

Here, our data showed that MMF interruption did not affect the incidence of acute or chronic GVHD, which is consistent with a meta-analysis evaluating the clinical impact of day 11 MTX omission due to toxicity after allogeneic HCT¹⁵. Only one patient in whom the planned dose of MMF was interrupted died due to acute GVHD. Interruption of planned oral MMF for GVHD prophylaxis may be associated with the development of severe RRT. Indeed, most patients in whom the planned dose of oral MMF was administered exhibited significant organ damage, which may have contributed to a higher NRM apart from GVHD. Previous studies have shown the significant impact of early complications on subsequent complications after allogeneic HCT^{37, 38}. Grade 2 or higher gastrointestinal RRT is frequently observed with the most common conditioning regimen¹⁶. Early severe gastrointestinal RRT may be associated with the interruption of planned oral MMF for GVHD prophylaxis, possible contributing to subsequent complications and poor outcomes after CBT.

Here, we found that the interruption of oral MMF was significantly associated with poor platelet recovery after CBT. The exact mechanisms underlying the association between the interruption of oral MMF and lower platelet recovery remain unknown. However, a previous study reported that tacrolimus combined with MMF is superior to tacrolimus alone in neutrophil engraftment after CBT, but not in platelet recovery³⁹. Addition of MMF may promote engraftment by suppressing hyperimmune reactions in cord blood cells³⁹. These effects may be responsible for the poor platelet recovery in patients in whom the planned dose of MMF was interrupted. Several studies have shown that delayed platelet recovery is associated with poor outcomes after HCT^{40, 41}, which is consistent with our CBT data. Overall, our findings suggest an association between MMF interruption and poor platelet recovery and mortality after CBT.

In summary, we found that MMF interruption was significantly associated with poor OS and high NRM after CBT. However, further studies with larger cohorts are required to confirm the clinical significance of oral MMF interruption in GVHD prophylaxis.

Acknowledgments

The authors thank all of the physicians and staff at our hospital.

Author Contributions

K.F. collected the data and analyzed the data. T.K. designed the research, collected the data, analyzed the data, performed the statistical analysis, and wrote the manuscript. Y.N. contributed to the critical review of the manuscript. All the other authors contributed to data collection. All authors approved the final version.

Conflicts of Interest

The author declares no conflict of interest. Disclosure form provided by the author are available on the website.

Satoshi Takahashi is one of the Editors of Blood Cell Therapy. He was not involved in the editorial evaluation or decision to accept this article for publication.

Acknowledgments

The authors thank all of the physicians and staff at our hospital.

Data Sharing Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Acknowledgments

The authors thank all of the physicians and staff at our hospital.

Statements

This study was performed in accordance with the ethical standards laid down in the 1964 Declaration of Helsinki. The Institutional Review Board of the Institute of Medical Science, the University of Tokyo approved this retrospective study (2023-33-0810), and the use of an opt-out consent mechanism in this retrospective study.