The Long-term Recovery of Parotid Glands and Oral QOL in Nasopharyngeal Cancer Treated by Intensity-modulated Radiotherapy

Background of The purpose of this study was to evaluate longitudinal volume the change of parotid gland after radiotherapy for nasopharyngeal cancer and the relationship between parotid irradiation dose and xerostomia symptoms. Methods We retrospectively analyzed longitudinal changing of parotid gland volumes in 20 patients treated by intensity-modulated radiotherapy (IMRT). We assessed xerostomia 4 years or more after IMRT by measuring the degree of oral moisture and oral QOL evaluation. Results relative at parotid volume had signicantly at 31-42 after IMRT, in parotid receiving less than 40 Gy. The mean irradiated dose for bilateral parotids showed negative correlations with oral QOL score and oral moisture.


Abstract Background
Xerostomia is one of the most common adverse events of radiotherapy in head and neck cancer patients.
The purpose of this study was to evaluate longitudinal volume the change of parotid gland after radiotherapy for nasopharyngeal cancer and the relationship between parotid irradiation dose and xerostomia symptoms.

Methods
We retrospectively analyzed longitudinal changing of parotid gland volumes in 20 patients treated by intensity-modulated radiotherapy (IMRT). We assessed xerostomia 4 years or more after IMRT by measuring the degree of oral moisture and oral QOL evaluation.

Conclusion
The mean irradiated dose for the parotid should be reduce as much as possible to improve oral QOL long after IMRT.

Background
Xerostomia is one of the most common adverse events of radiotherapy in head and neck cancer patients [1]. Salivary glands are highly sensitive to radiation and they can be damaged by radiation, leading to xerostomia, which causes a decrease in patients' quality of life after radiotherapy [2]. The mechanism of radiation-induced damage of salivary glands, which has been mainly studied in animal models, is thought to be selective damage of the plasma membrane of secretory cells immediately after radiation exposure, followed by damage of DNA, death of acinar progenitor cells and nally lysis of acinar cells [3][4][5][6][7]. Saliva production is reduced after delivery of 10-15 Gy to the parotid gland [8,9] and although recovery of parotid gland function is possible with the lapse of time after irradiation with 40-50 Gy, higher doses cause irreversible and permanent xerostomia [10][11][12][13]. To spare parotid gland function, the recommended mean irradiated dose for the parotid gland is less than 25-30 Gy, and it was reported that saliva production recovers to the same level as that before treatment [14,15]. Hey et al. observed saliva ow for 36 months after radiation therapy in patients with head and neck cancer and reported that most of the recovery processes were considered complete within that period [16]. In a multicenter randomized study (PASSRORT trial), the clinical outcomes of intensity-modulated radiotherapy (IMRT) and 3Dcomformal radiotherapy (3D-CRT) in terms of parotid sparing were investigated. It was shown in that study that 39% of the IMRT patients and 74% of the 3D-CRT patients suffered from xerostomia at 12 months after treatment [11]. There have been many reports on functional changes of the parotid gland after radiation therapy, but there have been few reports on the volume of the parotid gland and its relationship with oral quality of life (QOL) and even fewer reports on longitudinal change of the parotid gland volume. The purposes of this study were to evaluate the longitudinal volume change of the parotid gland after de nitive IMRT in patients with nasopharyngeal cancer and to investigate the relationships among irradiated dose for the parotid gland, parotid gland volume and xerostomia symptoms.

Methods
This research was carried out in compliance with the content of the Declaration of Helsinki, and it was approved by institutional review board in our institution.

Patients selection
We retrospectively analyzed patients with nasopharyngeal cancer who were treated by de nitive IMRT combined with chemotherapy in our hospital between 2009 and 2014. Eligibility criteria were as follows: (a) pathologically proven nasopharyngeal carcinoma, (b) longitudinally and regularly evaluation by computed tomography (CT) in our hospital and availability of longitudinal data for 3 years or more,(c) no previously surgery for the head and neck area, (d) no history of diseases causing a salivary secretion disorder such as Sjogren's syndrome, (e) treatment by de nitive chemoradiotherapy, and (f) treatment by IMRT.

Radiotherapy
All patients received irradiation to bilateral prophylactic cervical lymph node regions in addition to the primary tumor and region of lymph node metastasis by IMRT. Clinical target volume (CTV) 1 was de ned as the primary tumor and region of lymph node metastasis with an appropriate margin, and CTV 2 was assigned to bilateral prophylactic cervical lymph node regions. Planning target volume (PTV) was set with a margin of 5 mm to the CTV. PTV 1 was irradiated with 70 Gy in 35 fractions (fr.) and PTV 2 was irradiated with 40-50 Gy in 20-25 fr. using a 6-15 MV X-ray linear accelerator. Re-planning was conducted in the third or fourth week of IMRT for all patients.

Parotid gland volumes
Longitudinal head-and-neck CT data were transferred to a 3D-radiation therapy planning system (Eclipse, Varian Medical System Inc.) using the digital imaging and communication in medical format (DICOM). For longitudinal evaluation of parotid gland volumes, the CT scans were categorized into 6 periods, periods 0, 1, 2, 3, 4 and 5, representing the timing of CT scans at 0-6 months, 7-18 months, 19-30 months, 31-42 months, 43-54 months, and 55-66 months after completion of IMRT.
Two radiation oncologists with 5 years and 15 years of clinical experience contoured the parotid glands on a three-dimensional radiation therapy planning system and measured and recorded the parotid gland volumes.

Measurement of the degree of moisture in the mouth and QOL evaluation
We assessed xerostomia 3 years or more after IMRT by measuring the degree of moisture in the mouth using a moisture-checking device (Mucus, Life Co., Ltd.) and oral QOL evaluation by GOHAI (General Oral Health Assessment Index).
The degree of moisture in the mouth was measured 3 times continuously by placing the moisturechecking device at the center of the tongue, and the average value of the 3 measurements was used as the measured value. Patients were instructed not to eat, drink or gargle for at least 30 minutes before the measurements and to rest for at least 5 minutes before the measurements. Measured values of the moisture-checking device were de ned as follows according to a previous report [17]: <25, advanced drying; 25-28, moderate drying; 28-30, mild drying; 30 <, normal.
The GOHAI questionnaire is a questionnaire about the frequency of problems caused by a bad condition in the mouth during the past 3 months and consists of 12 questions. Five scales are used for answers and QOL is evaluated by the total score of 12 questions (with the lowest total score 12 and highest total score 60 and a higher the total score indicating higher quality of life) [17].

Statistical analysis
Statistical analyses were performed with SPSS 26.0. Since there were considerable individual differences in parotid gland volumes, the parotid gland volumes in each period were converted to the ratio to parotid gland volumes before radiotherapy (relative parotid volume). The Dunnett's test was used to evaluate the longitudinal changes in relative parotid volumes for periods 0-5. Using the Student's t-test, we compared the relative parotid volumes in each period for patients receiving less than 40 Gy with those for patients receiving 40 Gy or more and those for patients aged 50 years or younger with those for patients aged more than 50 years. Correlation analysis was performed for mean irradiated dose of the parotid, relative parotid volumes, measured values of the moisture-checking device and GOHAI score at the last observation date. For all analyses, p < 0.05 was considered statistically signi cant.

Oral moisture and QOL score
In 16 of the 20 patients, the degree of moisture in the mouth was measured by a moisture-checking device and oral QOL was assessed by GOHAI at 62.1 ± 3.6 months (range, 55-66 months) after IMRT.
Measurements could not be performed in the other 4 patients because they did not visit our hospital or they refused to undergo an examination. The mean measured value of the moisture-checking device was 26.9 ± 5.7 (range, 6.2-30.7). The mean GOHAI score was 52.1 ± 7.3 (range, 38-60). A correlation was found between GOHAI score and the values of moisture-checking device (r = 0.566, p = 0.022) (Fig. 3). A negative correlation was found between mean irradiated dose for bilateral parotids and the values of moisture-checking device (r=-0.585, p = 0.017) (Fig. 4). A negative correlation was also found between mean irradiated dose for bilateral parotids and GOHAI score (r=-0.570, p = 0.021) (Fig. 5). No signi cant correlation was found between relative parotid volume and measured values of the moisture-checking device (r = 0.122, p = 0.665). There was also no signi cant correlation between relative parotid volume and GOHAI score (r = 0.343, p = 0.211).

Discussion
In this study, longitudinal changes in parotid gland volume were analyzed in patients with nasopharyngeal cancer who were treated by IMRT with chemotherapy, and the relationships of mean irradiated dose for the parotid with parotid gland volume and xerostomia symptoms were assessed. It was found that the parotid volume decreased rapidly during and immediately after radiotherapy and then recovered gradually and reached a plateau about 2 years after radiotherapy. We also found that the mean irradiated dose for bilateral parotids dose showed negative correlations with objective and subjective xerostomia symptoms after a long period. There have been few reports on long-term changes in the parotid after radiotherapy for head and neck cancer. To the best of our knowledge, there have only been two reports on long-term changes in parotid volume after radiotherapy for head and neck cancer. Tomitaka et al. evaluated parotid volumes at 2 weeks and 6, 12, and 24 months after radiotherapy in 15 patients who received 30 Gy /15 fr. as preoperative radiotherapy with opposed lateral elds for advanced oral cancer, and they reported that the parotid volumes decreased, reached a nadir at 6 months after radiotherapy, and had recovered at 24 months after radiotherapy [18]. They also reported that there was a correlation between decreased parotid volumes and decreased saliva production immediately after preoperative radiotherapy for patients with advanced oral squamous cell carcinoma [19]. Their results are consistent with our results; however, patients treated with radical radiotherapy or with IMRT were not included in their study. Chun et al. evaluated parotid volumes within 100 days, between 100 days and 1 year, and more than 1 year after radiotherapy in 11 patients with nasopharyngeal cancer who were treated by IMRT with 70 Gy / 35 fr., and they reported that the nadir of parotid volume was within 100 days and that the parotid volume was recovered more than 1 year after radiotherapy [20]. Their results are was also consistent with our results; however, the observation period in their study was not su cient. In our study, we analyzed longitudinal changes in parotid volumes for 3-5 years after radiotherapy. The present study is the rst study in which long-term changes in parotid volume after radiotherapy for patients with head and neck cancer and the relationships of parotid volumes with subjective and objective xerostomia symptoms were investigated. Hey et al. reported that most of the recovery process from radiation-induced salivary gland injury was completed within 36 months after radiotherapy [16], and their results are almost consistent with the results of our study. Our data for parotid volumes 2 years and more after radiotherapy suggest that there is a relationship between radiationinduced salivary gland injury and parotid volumes. A relationship between parotid volumes and salivary ow has been reported. However, the correlation between oral moisture and QOL has not been shown, because observer-based monitoring of xerostomia symptoms could underestimate the actual xerostomia symptoms compared with patient-reported symptoms [21]. The present study is the rst study showing the relation between parotid volume and oral QOL score 3 years or more after IMRT. And we showed the signi cant correlation between oral QOL score and measured values of the moisture-checking device; however, the correlation was not so strong. It might be because GOHAI score did not show only wetness in the oral cavity.
The parotid volumes decreased after radiotherapy and then recovered gradually and reached a plateau about 2 years after IMRT and there was no signi cant recovery thereafter. These results suggested that recovery from parotid volume reduction is mostly completed in about 2 years, as is recovery from radiation-induced salivary gland injury. Although the mechanism of radiation-induced decrease in parotid volumes has not been elucidated, it was reported that the decrease in parotid volumes might be due to the loss of acinar cells or brosis and that the recovery of the parotid volumes might be due to regeneration of acinar cells [6,20].
In this study, the relative volumes of only parotids receiving less than 40 Gy had recovered signi cantly after IMRT. Furthermore, the relative volumes of parotids receiving less than 40 Gy tended to be larger than those of parotids receiving more than 40 Gy in any periods except for immediately after IMRT, although there was no statistically signi cant difference. It suggested that mean irradiated dose for parotid grand should be suppressed less than 40 Gy. It is consistent with the past reports [22].
Both measured values of the moisture-checking device and GOHAI score were correlated with mean irradiated dose for bilateral parotids but were not correlated with relative parotid volume. This may also be due to the small number of patients. The functions and responses of the parotid glands are typically assumed to be uniform throughout the parotid gland, but they may not actually be uniform, and this may be the reason why there is no correlation between parotid volumes and function. Actually, there are reports that radiotherapy to the cranial half caused more functional loss than did radiotherapy to the caudal half in rats [23]. This is because saliva from the caudal part ows through the cranial part and is therefore affected by damage to the cranial part.
This study had some limitations. The number of eligible patients was only 20 and the number of patients in whom measurements by a moisture-checking device and evaluation of oral QOL were performed was only 16. Due to the small number of cases, there might have been some factors that were not clari ed in the analysis. Since this study was a retrospective study and was performed only for patients for whom long-term follow-up was possible, the possibility of bias in selection of patients cannot be ruled out. In addition, although only parotid volumes were evaluated in this study, the irradiation eld also included the submandibular gland, sublingual gland, and other salivary glands, and radiation-induced damage of salivary glands cannot be neglected for oral moisture or QOL. However, total saliva is mainly produced by the parotid gland, and radiation-induced injury to salivary glands mainly results from damage to parotid glands [1]. Therefore, we evaluated only parotid volumes in this study. Furthermore, we did not consider fatty change of the parotid after radiotherapy in this study.

Conclusions
The mean irradiated dose for the parotid should be reduce as much as possible to improve oral QOL long after IMRT.
Abbreviations IMRT intensity-modulated radiotherapy 3D-CRT 3D-comformal radiotherapy QOL quality of life CTV clinical target volume PTV planning target volume GOHAI General Oral Health Assessment Index Declarations Ethics approval and consent to participate This research was approved by institutional review board in our institution.

Consent for publication
Not applicable Availability of data and materials Data sharing is not applicable to this article as no datasets were generated or analysed during the current study.

Competing interests
The authors declare that they have no competing interests