**1. Introduction**

Allogeneic hematopoietic cell transplantation (allo-HCT) is protocol treatment for hematological cancers and also for non-malignant disorders [1]. Graft versus host disease (GVHD), which is an important complication of allo-HCT is induced by the interactions of the host's immune system and transplanted donor cells. According to the time of appearance and clinical signs, GVHD is classified as acute or chronic [2,3]. Chronic GVHD (cGVHD), most frequently occurs ≥100 days after the transplant, affecting 25–40% of long-term HCT survivors [4]. cGVHD is classified as limited when single organs

are involved, such as the skin or liver, and extensive when multiple affection occurs, in organs such as the skin, liver, eyes, salivary glands, oral mucosa and others [5–7].

Oral manifestations occur in about 80% of patients with extensive cGVHD. Most frequent oral lesions are erythema, atrophy of the mucosa, oral lichen planus, oral mucositis, xerostomia and oral infections [6,8,9]. Oral mucositis is an inflammatory reaction of the oral mucosa, often occurring after high doses of chemotherapy, radiation therapy and/or stem cell transplantation. In hematopoietic cell transplantation (HCT) patients, mucositis may often occur along the entire orodigestive tract. The prevalence of oral mucositis is stated to be at 30–70% after chemotherapy, up to 90% after HCT [10]. Oral lichen planus aspects may vary from white lacey patches to open sores, involving the tongue and inner surface of cheeks [10,11].

Oral involvement could represent the only manifestation of the cGVHD or could be comprised in a multitude of chronic symptoms. cGVHD may appear on the oral mucosa (e.g., oral verruciform xanthoma, erytroplakia), at the salivary glands (e.g., multiple mucoceles on the soft palate, hyposalivation, xerostomia, impairment in the quality of saliva, gland swelling) and in musculoskeletal apparatus disfunction [11–13].

Dysphagia is one of the most debilitating symptoms and is induced by the pain associated with oral mucositis [9]. Local palliation of the oral symptoms is achieved either by systemic therapy, or topical treatment, or both. In this light, the use of topical agents determines the reduction of the oral symptoms, determines the reduction of the systemic immunosuppressant doses, minimizes their side effects and increases the healing process [2,10].

Although topical corticosteroids are not specifically approved for treatment of oral cGVHD, they are used to treat these symptoms, based on previous experiences that prove their well accepted use in other mucosal conditions [13–15].

Given the availability of limited data, the present research attempts to provide a systematic approach of literature including randomized clinical trials (RCTs) concerning the efficacy of topical dexamethasone, clobetasol and budesonide in oral GVHD.

#### **2. Materials and Methods**

We conducted a systematic review using RCTs to compare topical dexamethasone, clobetasol and budesonide for oral GVHD. This study was in accordance with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines and the Cochrane Collaboration format [16,17].

Three databases (PubMed, Web of Science, Scopus) were searched to identify eligible articles from inception to February 2020, using keywords ("dexamethasone", "clobetasol", "budesonide", "topical corticosteroids", "corticosteroids", "oral graft versus host disease", "oral GVHD") combined with a Boolean term ("AND") as follows: "dexamethasone AND oral graft versus host disease"; "dexamethasone AND oral GVHD"; "clobetasol AND oral graft versus host disease"; "clobetasol AND oral GVHD"; "budesonide AND oral graft versus host disease"; "budesonide AND oral GVHD"; "topical corticosteroids AND oral graft versus host disease"; "topical corticosteroids AND oral GVHD"; "corticosteroids AND oral graft versus host disease"; and "corticosteroids AND oral GVHD". Articles were evaluated by their titles and abstracts. The contents of the articles were assessed in order to determine if the studies met the inclusion/exclusion criteria. The full texts of the potentially relevant studies were retrieved and assessed. The reference lists of the chosen articles were manually searched to identify any other relevant studies that have been missed out using the search strategy.

The inclusion criteria used in the article selection were adult (≥18 years) oral GVHD; treatment with topical corticosteroids (dexamethasone, clobetasol, budesonide); RCT; and human studies, published in English. All other articles that did not complete the upper criteria were excluded from our research.

Two independent reviewers assessed the articles for eligibility and extracted the data using a standardized data extraction form. All lack of concordance was solved by a third reviewer. The following data were taken out: author, year, country, study type, sample size, mean age, male: female ratio, oral GVHD at baseline, treatment design, clinical response, side effects, outcome.

The Cochrane Collaboration's "Risk of Bias" tool 2.0 was used to assess the quality of these studies [17]. For every RCT included, a risk of bias was provided for the following domains: random sequence generation, allocation concealment, blinding of outcome assessment, incomplete outcome data, selective reporting and other bias. These domains were judged by two reviewers and were evaluated as low, unclear or high, and a third reviewer was invited to solve all unclear results.
