Incidence and Clinical Characteristics of Oral Mucositis in Allogeneic Hematopoietic Stem Cell Transplantation with Post-transplant Cyclophosphamide Treatment

Post-transplant cyclophosphamide (PTCy) treatment has been increasingly used in allogeneic hematopoietic stem cell transplantation (allo-HSCT). However, its effects on the burden and severity of oral mucositis (OM) remain unclear. A total of 177 patients received allo-HSCT with PTCy treatment at Xiangya Hospital between 2015 and 2020. Among them, 140 patients whose OM was prospectively graded using the World Health Organization (WHO) oral toxicity scale were included in this retrospective study. The grafts were peripheral stem cells from matched (28/140, 20.0%) and mismatched (112/140, 80.0%) donors. Conditioning intensity was categorized as myeloablative conditioning (MAC; 82/140, 58.6%) or reduced intensity conditioning (RIC; 58/140, 41.4%). The overall incidence of any OM was high (116/140, 82.9%) but the incidence of severe OM was relatively low (40/140, 28.6%). The median duration of OM was 10 days (2–22 days post-transplantation) from day –2 to day +15 (median day +8). Earlier onset of OM was correlated with greater severity. Multivariate analysis showed that conditioning intensity (MAC vs RIC, odds ratio [OR] 6.128, 95% condence interval [CI] 2.319–16.198) and donor type (mismatched vs matched, OR 3.252, 95% CI 1.089–9.717) were associated with increased risk of severe OM. No signicant implications of severe OM were observed on acute or chronic GVHD. Patients with severe OM had slightly worse overall survival, but the difference was not statistically signicant (p = 0.078). Therefore, severe OM does not appear to lead to a worse transplant outcome if an intensied oral care protocol is adopted.


Introduction
Oral complications following hematopoietic stem cell transplantation (HSCT) typically manifest as a wide range of lesions with varying degrees of severity, ranging from mild, disabling, to life-threatening.
Oral mucositis (OM) before trilineage engraftment is one of the most debilitating complications of HSCT.
Up to 47-100% of transplant recipients may experience early OM, based on a recent systematic review. [1] OM primarily arises from the direct toxicities of conditioning regimens and a graft-versus-host disease (GVHD) prophylaxis regimen or indirectly from systemic toxicities and the immune status. Furthermore, infections may have a greater contribution to post-transplant OM compared with previous estimations.
[ [2][3][4][5] Post-transplant cyclophosphamide (PTCy), which was initially developed in the setting of haploidentical transplantation [6], has proven to be a promising treatment for transplant recipients with various donor types [7][8][9], bone marrow or peripheral grafts [8,10,11], myeloablative conditioning (MAC) or reduced intensity condition (RIC) [10], and benign or malignant diseases [12][13][14][15][16][17]. However, the burden and severity of OM remain unclear when a high dose of a cytotoxic drug such as cyclophosphamide is used shortly after the conditioning. To facilitate the management and prevention of OM for transplant recipients receiving PTCy treatment, we performed a retrospective analysis to evaluate the incidence, clinical patterns, risk factors, and clinical signi cance of OM in a cohort of patients who received allo-HSCT with PTCy treatment.

Study design and population
This was a single-center study of patients who presented at the stem cell transplant department of Central South University Xiangya Hospital, Changsha, China, between 2015 and 2020. The patients underwent allogeneic HSCT after providing informed consent. The Institutional Review Board (IRB) approved the study in accordance with the Declaration of Helsinki. Data and associated electronic materials were independently retrospectively reviewed by two researchers. A patient was included if the HSCT platform was based on PTCy treatment and the grade of OM was based on the World Health Organization (WHO) scale. This cohort was not restricted according to transplant indication and donor and conditioning types. All patients received mobilized peripheral blood stem cells. Severe OM was de ned as grades 3 to 4.

Transplant protocols
Based on high-resolution typing of class I (A, B, and C) and class II (DR and DQ) human leukocyte antigen (HLA), donors were classi ed as either matched or mismatched (less than a 10/10 HLA match).
Transplantation from mismatched unrelated donors and haploidentical donors was either based on a clinical trial or real-world situation with approval of the clinical IRB. The conditioning regimen was either MAC or RIC de ned according to the Bacigalupo criteria. [18] Total body irradiation was not included in our conditioning regimen. Prophylaxis against GVHD was based on a uniform PTCy regimen, combining a high dose of cyclophosphamide with a short course of methotrexate (MTX; 15 mg/m 2 on days 1 and 10 mg/m 2 on days 2 and 5) treatment after transplant. Per the institutional protocol, leucovorin was not administered following minimal MTX administration. Cyclosporine A was intravenously administered from day 6 after transplant and switched to oral administration after 3 weeks if the patient was able to take oral medications. In the case of matched or mismatched donors, cyclosporine prophylaxis was administered for 6 or 9 months and subsequently tapered in patients with no active GVHD. Mycophenolate mofetil is not regularly used in our transplant protocol. Rabbit anti-thymocyte globulin (ATG; Genzye, France) at a total dosage of 8 mg/kg was administered over 4 days to patients with inherited benign disorders such as severe thalassemia.
Prevention and management of mucositis According to our protocol, an intensive OM care plan was adopted. In the phase prior to conditioning, all patients were advised to consult a dental team for oral examination and elimination of dental, periodontal, endodontic, and oral mucosal lesions or active infections. From the initiation of conditioning therapy, all patients received a prophylaxis regimen including systematic infection prevention for fungi, viruses, and bacteria. Local oral care involved treatment with a nystatin suspension of 4 wU/ml (3 ml) and 0.2% chlorhexidine (5 ml) in addition to basic oral hygiene. The mouthwashes were prescribed to be used every 4 hours. Upon a diagnosis of mucositis, a suspension of granulocyte macrophage-colony stimulating factor was added to the previous oral care regimen. Opiates were used if necessary, to control the pain associated with mucositis.

Statistical analysis
Comparison of baseline characteristics of patients with and without severe OM was performed using the Chi-square test, Fisher exact test, or Mann-Whitney U test. Continuous variables were compared using the non-parametric Kruskal-Wallis test. Multivariate logistic regression analysis was performed to identify independent predictors of mucositis. Transplantation outcomes such as the incidence of acute GVHD (aGVHD), chronic GVHD (cGVHD), and engraftment of donor blood cells were also compared between patients with and without severe OM. Overall survival was estimated using the Kaplan-Meir calculator and compared between groups using the log-rank test. All statistical analyses were performed using SPSS version 26.0 (IBM, Armonk, NY, USA).

Population characteristics
The characteristics of the 140 patients and their donors included in this study are summarized in Table 1. The patients represented a wide age range with a median of 15 years. The indication of HSCT was benign disorders for the majority of patients, including severe aplastic anemia (55, 39.3%), paroxysmal nocturnal hemoglobinuria (PNH) (2, 1.4%), inherited bone marrow failure (2, 1.4%), severe thalassemia (42, 30.0%), inherited immune de ciency (3, 2.1%), and leukodystrophy (3, 2.1%). Among the remaining 34 patients who underwent HSCT for treatment of a malignancy, 10 had acute leukemia (7.1%) and 24 had myelodysplastic syndrome (17.1%). Less than 30% of the donors were matched (13.6% related and 14.3% unrelated), and among the mismatched donors, 65.7% were haploidentical related donors and 6.4% were unrelated donors. With respect to the conditioning regimens, MAC was more frequently used than RIC. Ninety of the 140 (64.3%) transplants included ATG in the GVHD prophylactic regimen along with PTCy.

Outcomes associated with OM
We did not nd an association between severe OM and aGVHD (p = 0.5556), cGVHD (p = 0.798), engraftment of neutrophils (p = 0.509), and engraftment of platelets (p = 0.703). Moreover, there was no signi cant difference in overall survival according to the severity of OM with a median follow-up of 561 days (95% CI 538-737 days) (Fig. 2).

Discussion
OM is one of the most common complications of HSCT. However, to the best of our knowledge, the incidence and severity of OM among patients with PTCy-based allogeneic HSCT have been unclear to date. [19] In our cohort of 140 cases who received transplants with PTCy treatment, the incidences of any OM (82.9%) and severe OM (28.6%) were comparable to those in previous reports, in cases using a traditional transplant regimen other than PTCy (47.2-100% and 2.5-60.9%, respectively). [1, 20-25, 19, 26] PTCybased prophylaxis was reported to show a better safety pro le compared with ATG and was associated with reduced severity of mucositis. [27] In our study, more severe OM was observed in patients after an intensive conditioning regimen (MAC) compared with that after a less intensive regimen (RIC). In a recent study using RIC and PTCy, severe OM was observed at a similar frequency (15%) in patients with leukemia or myelodysplastic syndrome. [28] Using the PTCy platform, OM typically began about 1 week after completing the conditioning regimen and lasted for 6-10 days. An early onset of OM would suggest a greater possibility of subsequent severe OM. This information is important for a nursing team to establish a personalized protocol for an oral care regimen.
Because severe OM requires special medical attention such as parenteral nutrition, infection prophylaxis, and uid replacement, we focused on the risk factors associated with severe OM (above grade 3 according to the WHO criteria) as opposed to cases of mild or moderate OM. MAC was associated with increased risk of severe OM in our cohort, based on both univariate and multivariate analyses. This result was in line with those of previous studies. [19,[29][30][31] However, the incidence of severe OM in this study was lower than that in previous reports. [1] This might be re ective of the relatively younger patients, more frequent use of udarabine-based reduced toxicity conditioning regimens, [19] and the standardized and intensi ed oral care protocol. Notably, mismatched donors and lack of ATG in the GVHD prophylaxis regimen were signi cantly related to severe OM in multivariate analysis. The underlying reason for these protective effects is currently unclear. Therefore, studies on whether and how the immune status plays a role in the mechanism of early OM post-transplant are warranted.
In line with previous studies, severe OM was not associated with higher incidence of aGVHD and cGVHD. We also did not observe a detrimental in uence of severe OM on engraftment of neutrophils and platelets. Although the occurrence of severe OM did not lead to signi cantly worse overall survival, the clinical implications of severe OM should not be ignored. Indeed, OM has been reported to be the single most debilitating complication of a transplant, causing severe pain, affecting oral function, increasing the risk for bleeding and systemic infection, and having the potential to compromise the upper airway.
References are needed to support this statement. Accordingly, oral care was intensively implemented from the beginning of HSCT according to our institutional protocol, including prevention and symptomatic management of OM.
There are several limitations to this study. First, although OM was graded according to the WHO scale, the full picture of oral changes such as the number, location, and stage of the lesions was not systematically recorded and analyzed. Including the speci c symptoms and ndings from the oral examination may offer a better de nition of the mucositis burden and phenotype. Second, although OM was prospectively documented daily following the local protocol, this was not a prospectively designed study. Therefore, the grading of OM would have likely differed according to the individual primary nurses. Finally, this was a single-cohort study, which might have limited the power to detect meaningful associations.
In summary, OM remains a high-burden complication among allogeneic-HSCT recipients receiving PTCy treatment. However, the incidence of severe OM does not appear to be increased on this platform. Patients who receive MAC and cells from a mismatched donor are at greater risk for developing severe OM. ATG included in the GVHD prophylaxis may reduce the risk of severe OM via an as-yet unknown mechanism. The occurrence of severe OM did not lead to a worse transplant outcome if intensi ed oral care was adopted. Understanding the pathophysiology of OM development could facilitate the development of novel strategies to manage oral complications in allogeneic-HSCT recipients.

Declarations
Funding (information that explains whether and by whom the research was supported): Not applicable Con icts of interest/Competing interests (include appropriate disclosures):