The association between cumulative radiation dose and the incidence of severe oral mucositis in head and neck cancers during radiotherapy

Abstract Background Quality of life can be influenced by oral mucositis (OM), and it is necessary to implement OM management strategies before the initiation of radiotherapy (RT) in patients with head and neck cancer (HNC). Aims To examine the association between the cumulative radiation dose and the incidence of severe OM in HNC patients receiving RT. Methods and results A retrospective observational cohort study was conducted in a Showa University Fujigaoka Hospital, in Japan. We retrospectively analyzed 94 patients with HNC who developed OM during RT. We defined OM as a more than grade 2 OM. The cumulative incidence of OM curves of the two categories was estimated using the Kaplan–Meier method and compared using the log‐rank test. We estimated the hazard ratio (HR) for OM after the adjustment of factors for covariates using Cox's regression analysis. Patients with smoking history had a significantly later development of OM than those with no smoking history (20 Gy‐incidence OM 68.7% vs 39.7%, P = .003). In contrast, patients undergoing concurrent chemotherapy had an earlier development of OM than those undergoing RT alone (20 Gy‐incidence OM 24.2% vs 55.7%, P < .001). Multivariate analysis revealed that no smoking history and concurrent chemotherapy were independent predictive factors, with a HR of 0.526 (P = .025) and 2.690 (P < .001), respectively. Conclusion We demonstrated that no smoking history and concurrent chemotherapy may be predictive of OM in HNC patients.


| INTRODUCTION
Head and neck cancers (HNC) represent 5% of all cancers. In 2018, they accounted for an estimated 887 649 new cancer cases and 453 307 cancer-related deaths globally. 1 The head and neck are also closely related to swallowing, voice respiration, articulation, and mastication, and the loss of these functions can dramatically lower patients' quality of life (QOL). Kam et al reported that the incidence of suicide in patients with HNC is more than thrice that of the general U.S. population. 2 They also indicated that this may be linked to the relationship between anatomic sites and the ability to speak and/or swallow. The standard treatment of HNC is radiotherapy (RT), to keep these functions. In advanced HNC, chemotherapy is concurrently administered with RT. Sarraf et al indicated that chemoradiotherapy (CRT) is superior to RT alone for patients with advanced nasopharyngeal cancers for progression-free survival and overall survival (OS). 3 Radiation-induced oral mucositis (OM) is common among patients with HNC and is the most debilitating side effect of RT. 4 CRT increases the incidence of side effects compared to RT alone; Hata et al reported that OM risk of more than grade 2 increases by 5.6 times compared to RT alone. 5 OM leads to reduced oral intake and increases dysphagia due to pain, which can dramatically lower the patients' QOL. Chen et al revealed that OM was the most common oral dysfunction. 6 Therefore, it is important to complete treatment to maintain the QOL of patients while properly managing pain control.
It is necessary to implement OM management strategies before the initiation of RT in patients with HNC. 7 Some studies have reported on the risk factors for OM in HNC patients. [8][9][10] The initial clinical signs of OM include mucosal erythema and superficial sloughing that may occur with a cumulative radiation dose of 20 to 30 Gy, which is accompanied by the beginning of the breakdown of the intact mucosa followed by ulceration. 11 Vera et al, also reported that HNC patients with nasopharyngeal or oropharyngeal tumors who receive cumulative radiation doses >50 Gy are more likely to develop OM. 10 However, there is no evidence on the predictive factors for OM in relation to cumulative radiation dose. It is important to predict OM before the initiation of RT in patients with HNC. Therefore, we conducted this retrospective analysis of patients with HNC who were treated with RT to investigate the relationship between cumulative radiation dose and the incidence of severe OM.

| METHODS
We conducted a retrospective cohort study using data obtained from medical records. This retrospective study included HNC patients who were admitted or attended to the Showa University Fujigaoka Hospital between January 2005 and March 2015. The inclusion criteria were as follows: participants experienced OM during RT with HNC patients. Participants who experienced no OM were excluded due to investigation of the relationship between cumulative radiation dose and the incidence of severe OM. We were defined incident of OM as a more than grade 2 OM due to severe OM lower the patients' QOL.
Therefore, 94 patients were eligible for the analysis. We evaluated the association between cumulative radiation dose and the incidence of OM in HNC patients.
The tumors were histologically diagnosed and staged according to the TNM classification; they were confirmed by neck and chest computed tomography, bone scintigraphy, endoscopy, and histological diagnosis by biopsy. We extracted the population using the diagnosis code in our original system. The ethics committee of our institution approved the study (approval number: 201516).

| Treatment
Patients were irradiated with standard radiation (total 35 counts, 2 Gy once a day) or hyperfractionation (total about 58 counts, 1.2 Gy twice a day). The treatment period was about 6 to 8 weeks. Patients who received concurrent chemotherapy were administered with oral tegafur/ gimeracil/oteracil (S-1) (80 mg/m 2 ), 12

| Evaluation of OM
OM was graded as 1-4 according to the National Cancer Institute Common Toxicity Criteria for Adverse Events, version 4.0. 15 The most severe grade of OM was based on the worst OM from the first to the last day of RT. We also investigated the time of onset of OM and the time of the worst severity of OM.

| Statistical methods
We evaluated the association between cumulative radiation dose and the incidence of OM. When we analyzed factors with a frequency of 65% in 94 patients, we could evaluate a hazard ratio of 1.85, with a power of 80%.
We classified OM into three grades: grades 1, 2, and 3. We observed the following factors and classified them into two categories: age (<65 years vs ≥65 years), sex (male vs female), alcohol history (absent vs present), smoking history (absent vs present), type of RT (standard vs hyperfractionation), stage (1 and 2 vs 3 and 4), chemotherapy (absent vs concurrent), prior WBC level (<4000/μL vs ≥4000/μL), prior ALT level (<50 IU/L vs ≥50 IU/L), prior Cr level (<1.00 mg/dL vs ≥1.00 mg/dL), and prior Alb level (<3.5 g/dL vs ≥3.5 g/dL). The cumulative incidence of OM curves using two or three categories was estimated using the Kaplan-Meier method and compared by the log-rank test. The cumulative incidence of OM was defined as the incidence of more than grade 2 OM in this study. We estimated the hazard ratios (HRs) for the incidence of OM after the adjustment for covariates using Cox's regression analysis proportional hazards model (stepwise methods). The time of onset of OM (day) and the cumulative radiation dose (Gy) were compared using the Student's t test. P-values less than .05 were considered statistically significant.
The statistical analyses were performed using IBM SPSS Statistics 23.

| RESULTS
Baseline characteristics are summarized in  Figure 1A).
Patients undergoing concurrent chemotherapy also had an earlier F I G U R E 1 Kaplan-Meier the incidence of oral mucositis curves according to the grade of oral mucositis, smoking history, and chemotherapy development of OM compared with those undergoing RT (20 Gy incidence OM 24.2% vs 55.7%, P < .001, Table 2, Figure 1C). In contrast, patients with a history of smoking also had significantly delayed OM compared with patients with no history (20 Gy incidence OM 68.7% vs 39.7%, P = .003, Table 2, Figure 1B). Stage 3/4 and hyperfractionation of RT were also significantly associated with the development of OM (Table 2). However, there was no significant association between the development of OM and age, sex, and alcohol history.
Smoking history and concurrent chemotherapy were predictive of the development of OM on multivariate analysis ( We previously demonstrated that concurrent chemotherapy was identified as a significant, independent risk factor for the severity of OM. 16 Hata et al reported that 5-FU was related to worse OM. 5 This study also revealed a relationship between cumulative radiation dose, even lower doses, and the development of OM in patients with HNC during RT. Therefore, healthcare providers should be strategic with OM management during concurrent chemotherapy, even when the cumulative radiation dose is low.
Jyoti et al reported that former and active tobacco smoking during RT for cervical cancer is associated with unfavorable disease-free survival and OS outcomes. 17 Chen et al also reported that tobacco smoking during RT for HNC is associated with unfavorable outcomes. 18  Patients treated with hyperfractionation and accelerated fractionation with concomitant boost had significantly better local-regional control than those treated with standard fractionation. 19 However, the standard fractionation resulted in fewer adverse effects compared to hyperfractionation. In our study, although it was consistent that hyperfractionation was more at risk of OM than standard fractionation, this finding was not observed in multivariate analysis. We hypothesized that patients who received concurrent chemotherapy had a higher hyperfractionation rate than that in patients receiving radiation therapy. In addition, although the stage 3 or 4 group had a higher OM risk than those in the stage 1 or 2 groups in univariate analysis, we did not observe this finding in multivariate analysis. Stage may have been a confounding factor as more patients in the stage 1 and 2 groups received radiation therapy compared to those in the stage 3 or 4 groups.
This study had limitations. Our sample size was small and from a single institution. This study was also retrospective, and it was difficult to investigate clinical parameters in detail. In particular, the details of smoking history are important, including the number of cigarettes per day, smoking time, and passive smoking. We did not consider other risks such as preventive oral care and dry mouth, which are related to OM. Although the follow-up ended with the last RT in this study, we did not consider prognosis in patients with HNC.
In conclusion, we demonstrated that no smoking history and concurrent chemotherapy are predictive factors of severe OM related to cumulative radiation dose. However, smoking history needs to be considered in more detail in the future. OM reduces oral intake and increases dysphagia due to the associated pain and can dramatically lower patients' QOL.
Therefore, healthcare providers need to strategize OM management with these considerations. We recommend stringent management and monitoring of patients receiving concurrent chemotherapy, even when the cumulative radiation dose is low because early OM may progress to a severe state. Prospective studies on QOL that investigate QOL benefits related to management strategies may underscore the significance of concurrent chemotherapy and negative smoking history as markers of severe OM that need to be monitored closely by healthcare providers.

ACKNOWLEDGEMENT
There are no funding sources from any author.

CONFLICT OF INTEREST
The authors have no conflicts of interest to report.

ETHICAL STATEMENT
The study was approved by the Ethics Committee at Showa University Fujigaoka Hospital, Japan (approval number 201516). In this retrospective cohort study, informed consent from the patients was not needed according to the ethical approval.

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