Oral Oncology 49 (2013) 42–48

Contents lists available at SciVerse ScienceDirect

Oral Oncology

journal homepage: www.elsevier .com/locate /ora loncology

Scintigraphic assessment of salivary function after intensity-modulatedradiotherapy for head and neck cancer: Correlations with parotid doseand quality of life

Wen-Cheng Chen a,b, Chia-Hsuan Lai a, Tsair-Fwu Lee c, Chao-Hsiung Hung a, Kuo-Chi Liu a,Ming-Fong Tsai d, Wen-Hung Wang e, Hungcheng Chen f, Fu-Ming Fang g, Miao-Fen Chen a,b,h,⇑a Department of Radiation Oncology, Chang Gung Memorial Hospital, Chiayi, Taiwanb Chang Gung University, College of Medicine, Tao-yuan, Taiwanc Medical Physics & Informatics Lab., Department of Electronics Engineering, National Kaohsiung University of Applied Sciences, Kaohsiung, Taiwand Department of Nuclear Medicine, Chang Gung Memorial Hospital, Chiayi, Taiwane Department of Otolaryngology-Head and Neck Surgery, Chang Gung Memorial Hospital, Chiayi, Taiwanf Department of Radiation Oncology, Helen F. Graham Cancer Center, Newark, DE, United Statesg Department of Radiation Oncology, Chang Gung Memorial Hospital-Kaohsiung, Taiwanh Graduate Institute of Clinical Medicine, College of Medicine, Chang Gung University, Taoyuan, Taiwan

a r t i c l e i n f o s u m m a r y

Article history:Received 27 April 2012Received in revised form 5 July 2012Accepted 6 July 2012Available online 31 July 2012

Keywords:Head and neck cancerXerostomiaParotid gland functionIntensity-modulated radiotherapyQuality of life

1368-8375/$ - see front matter � 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.oraloncology.2012.07.004

⇑ Corresponding author. Address: Department ofGung Memorial Hospital, Chia-Yi, #6, Chia-Pu Rd., PutTel.: +886 5 362 1000x2011; fax: +886 5 362 1000x2

E-mail address: miaofen@adm.cgmh.org.tw (M.-F.

Objective: We investigated salivary function using quantitative scintigraphy and sought to identify func-tional correlations between parotid dose and quality of life (QoL) for head and neck cancer (HNC) patientsreceiving intensity-modulated radiotherapy (IMRT).Materials and methods: Between August, 2007 and June, 2008, 31 patients treated IMRT for HNC wereenrolled in this prospective study. Salivary excretion function (SEF) was previously measured by salivaryscintigraphy at annual intervals for 2 years after IMRT. A dose-volume histogram of each parotid glandwas calculated, and the normal tissue complication probability (NTCP) was used to determine the toler-ance dose. QoL was longitudinally assessed by the EORTC QLQ-C30 and H&N35 questionnaires prior to RT,and at one, three, 12 and 24 months after RT.Results: A significant correlation was found between the reduction of SEF and the mean parotid dosemeasured at 1 year (correlation coefficient, R2 = 0.651) and 2 years (R2 = 0.310) after IMRT (p < 0.001).The TD50 of the parotid gland at 1 year after IMRT is 43.6 Gy, comparable to results from western coun-tries. We further found that contralateral parotid and submandibular gland function preservation wascorrelated with reduced sticky saliva and a better QoL compared to the functional preservation of bothparotid glands, as determined by the EORTC QLQ-H&N35 questionnaire.Conclusion: A significant correlation was found between the reduction of SEF and the mean parotid dose.Preservation of contralateral parotid and submandibular gland function predicts a better QoL comparedto preservation of the function of both parotid glands.

� 2012 Elsevier Ltd. All rights reserved.

Introduction

Xerostomia is a common side-effect that occurs after radiother-apy (RT) in patients with head and neck cancer (HNC). Modern RTtechniques, such as three-dimensional conformal RT, or intensity-modulated RT (IMRT), can spare salivary glands, thus preservingsalivary flow rates and improving observer-accessed xerostomiacompared to conventional RT.1–3 The parotid gland dose-volume

ll rights reserved.

Radiation Oncology, Changz City, Chia-Yi, Hsien, Taiwan.067.Chen).

response following RT has been investigated in several large pro-spective studies.4–8 A wide range (26–43 Gy) of threshold meandose (TD50), the dose resulting in a 50% probability of a complica-tion for the whole organ irradiated uniformly, for parotidgland-stimulated salivary flow, has been reported. The mean dosereported to preserve parotid gland function in Asian studies variesfrom 31 to 43.9 Gy9–11; yet, none reported the TD50.

Although the apparent improvement in objective salivary func-tion and observed xerostomia was achieved by 3-D conformal orIMRT, the result does not necessarily improve the patient-reportedxerostomia.12 Since xerostomia is mainly a quality of life issue, apatient-reported quality of life questionnaire is more useful inassessing salivary function. Therefore, in the present prospective

W.-C. Chen et al. / Oral Oncology 49 (2013) 42–48 43

study, we longitudinally recorded the recovery of parotid glandfunction using salivary scintigraphy in patients receiving IMRT.The NTCP model was used to determine the TD50 of the parotidgland. The patient-reported quality of life (QoL) questionnaire,EORTC QLQ-C30, and the Head and Neck module (H&N35), were gi-ven to patients prior to, and periodically following, RT to assessperceived salivary function over time. Patient-, tumor-, and ther-apy-related factors were simultaneously assessed as predictors ofparotid gland function and QoL.

Materials and methods

Patient and disease characteristics

The study population consisted of patients with HNC treatedwith IMRT at our hospital. All patients provided written informedconsent approved by the institutional review board. Patients whosuffer from Sjögren’s syndrome or any other medical cause of xero-stomia were excluded. Patient use of any medications known to af-fect salivary gland function was prohibited. During the period fromAugust, 2007 to June, 2008, a total of 31 patients with primary(n = 15) or post-operative RT (n = 16) for HNC were entered intothis prospective study. Patient and tumor characteristics are listedin Table 1. IMRT was delivered by the computer-controlled auto-sequencing segmented or dynamic multileaf collimator of a Varianlinear accelerator (Linac 21 EX), according to methods describedpreviously.13 IMRT delivery directed at sparing the parotid glands(predominantly contralateral side), while treating the primary tar-gets and lymph nodes at risk was conducted. The prescribed doseswere 67.4–70.8 Gy (mean, 69.8 Gy) to the macroscopic tumor plan-ning target volume, 54.8–70.8 Gy (mean, 62.0 Gy) to the surgicaltumor bed planning target volume, and 46.8 Gy to the subclinicaldisease planning target volume, at 1.8–2 Gy per fraction. Nineteen

Table 1Patients and tumor characteristics of 31 patients.

Characteristic Value

Age (y)Mean 53Range 28–78

Gender (n)Female 1 (3)Male 30 (97)

Tumor siteNPC 11 (35)Oral cavity 14 (45)Oropharynx 4 (13)Larynx 1 (3)Parotid 1 (3)

Stage (TNM staging system)T1 3 (10)T2 12 (39)T3 6 (19)T4 7 (22)Not applicable/recurrent 3 (10)N0 16 (52)N1 5 (16)N2 7 (22)N3 0 (0)Not applicable/Recurrent 3 (10)

Surgery before RTYes 16 (52)No 15 (48)

ChemotherapyYes 19 (61)No 12 (39)

Data in parentheses are percentage.

patients received concurrent chemotherapy. Of these, five patientsreceived additional adjuvant chemotherapy. Nineteen patients re-ceived concurrent chemotherapy with weekly CDDP (40 mg/m2)for 5–7 courses (n = 18) or PF regimen (CDDP 80 mg/m2 on day1 + 5-FU 800 mg/m2, days 1–5, every 21 days) for two courses(n = 1). Of these, five patients received additional adjuvant chemo-therapy with PF regimen for 2–3 courses (n = 4) or TEF regimen(Taxol 60 mg/m2 on day 1 + CDDP 20 mg/m2 on day 1 + 5-FU800 mg/m2 on days 1–2) for one course (n = 1).

Salivary scintigraphy

All patients received salivary scintigraphy prior to IMRT deliv-ery. Patients fasted for 4 h prior to beginning the study. The studycommenced with patients receiving 10 mCi 99mTc pertechnetateintravenously. Sequential imaging of 1 frame/min was acquiredfor 30 min over the left- and right-anterior views of the head andneck. Salivary gland function was represented by saliva excretionfollowing sialogogue stimulation with acidic material (lemon juiceor 200 mg ascorbic acid tablet on the dorsal tongue). Salivaryexcretion factor (SEF) was quantified by determining the maximalexcretion activity per gland as a function of maximal uptake, as de-scribed previously by Roesink et al.6 The salivary scintigraphies arescheduled prior to IMRT and then 1 and 2 years after its comple-tion. The excretion response was analyzed per patient and subse-quently per individual gland. All patients received scintigraphies1 year after RT, whereas only 25 patients (25/31, 81%) receivedexaminations 2 years following RT. The six patients who did not re-ceive scintigraphies were due to tumor recurrence (n = 2) or fol-low-up refusal (n = 4).

Normal tissue complication probability model

The data were applied to the NTCP model proposed by Lyman.14

The NTCP model quantitatively assesses the effects of both theradiation dose and the volume of the gland irradiated on the prob-ability of radiation-induced changes in parotid gland function.Three parameters are presented in the sigmoid dose–responsecurve: n, m, and TD50, where n accounts for the volume effect ofan organ, m describes the slope of the dose–response curve andthe TD50 is the dose resulting in a 50% probability of a complicationfor uniform irradiation of the whole partial volume.

EORTC quality of life questionnaires (QLQ-C30 & H&N35)

Subjective salivary function was evaluated by the EORTC QLQ-C30 and H&N35 questionnaires. Patients were asked to answerthe questions prior to receiving IMRT, and then 1, 3, 12, and24 months later. The Traditional Chinese versions of the EORTCQLQ-C30 and H&N35 questionnaires were obtained from the Qual-ity of Life Unit, EORTC Data Center in Brussels, Belgium.15 All scalespertaining to the EORTC QLQ-C30 and H&N35 range from 0 to 100.A high score for a functional or global QoL scale represents a rela-tively high/healthy level of functioning or global quality of life,whereas a high score for a symptom scale represents the presenceof a symptom or problem.

Statistical methods

A logistic regression statistical method was used to study thedose–response relationship and volume effects in parotid glands.Spearman’s correlation was used to evaluate the factors associatedwith recovery of parotid gland function. Multivariate analysis wasperformed using a multiple regression model. The level of signifi-cance was set at p < 0.05. A paired sample t-test was used to com-pare the mean scores at each time point, with significance

Figure 1 Recovery of parotid gland excretion function as a function or mean parotid gland dose (A) 1 year (R2 = 0.651) and (B) 2 year (R2 = 0.310) after RT. The recovery ofparotid gland excretion function is expressed as the ratio of post- and pre-treatment salivary excretion factor (SEF).

Table 2Factors associated with recovery of parotid gland function at 1 year after RT.

Factors Spearman’s correlation b value 95% confidence interval Univariate analysis(p value)

Multivariate analysis(p value)

Age (P50 vs <50) 0.146 0.141 �3.28 to 24.92 0.26 0.13Tumor site (oral cavity vs non-oral cavity) �0.116 �0.101 �40.55 to 25.12 0.37 0.64Parotid mean dose �0.807 �0.759 �0.25 to 0.15 <0.001 <0.001Surgery (yes vs no) �0.052 �0.092 �40.27 to 26.17 0.69 0.67Chemotherapy (yes vs no) 0.017 0.166 �1.37 to 27.24 0.90 0.08

44 W.-C. Chen et al. / Oral Oncology 49 (2013) 42–48

determined by p < 0.05. The influence of patient-, treatment- orsalivary function variables on these changes were studied using atwo-sample t-test that compared the 1-year mean scores of swal-lowing, opening mouth, dry mouth and sticky saliva. The SPSSver.17.0 statistical software was used for data processing (IBMCorp., Armonk, NY).

Results

Patient characteristics

After a median patient follow-up of 46.8 months (range, 34.9–62.3), 30 (97%) of the 31 patients were alive at the time of manu-script preparation. Only one patient died of an unrelated disease(lung cancer). Among the surviving patients, 25 remain diseasefree, while the other five were alive with distant metastasis. Allscintigraphic examinations and QoL assessments were performedduring disease-free times. On dosimetry analysis, the ipsilateralparotid gland received a mean dose of 26.9–74.8 Gy (mean,51.7 Gy), and the contralateral lobe, 7.6–57.6 Gy (mean, 36.7 Gy).

Initial salivary function

The SEF before RT had a normal distribution, with a mean of48.1 ± 18.2%. A considerable amount of variability in parotid gland

output was found, from 19.2% to 72.8%. Five oral cancer patientshad no excretion after stimulation in one (ipsilateral) parotid glanddue to previous radical surgery; these measurements were notused in the final data analysis. No differences were found in uptakeand SEF between the left and right parotid glands. The basalscintigraphic measurements were independent of age, tumor siteor tumor stage, surgery or chemotherapy.

Effects of irradiation on salivary function

The mean SEF at 1 year post-RT of 57 glands in 31 patients was31.4 ± 20.3%. The SEF ratio was used to evaluate the percentage ofSEF recovery 1 and 2 years post-RT with respect to baseline SEF.The mean value of recovery of excretion function was 70% at 1 year(range: 0–163%, SD: 32%), and 68% at 2 years (range 0–134%, SD26%). For parotid glands that received doses equal to or greaterthan the median dose (44.69 Gy), the value of recovery of excretionfunction was 41 ± 27% at 1 year, and 41 ± 32% at 2 years of recov-ery. In contrast, for glands that received less than the median dose,the recovery function could be as high as 83 ± 32% at 1 year and78 ± 22% at 2 years (p < 0.001, and p < 0.001, respectively; two-sample t-test). Using the Spearman’s correlation, we found a strongcorrelation between parotid gland excretion recovery and meanparotid gland dose at 1 year (correlation coefficient, R2 = 0.651,p < 0.001) and 2 years after RT (R2 = 0.310, p < 0.001) (Fig. 1). In

Figure 2 Complication probability curves as a function of mean parotid gland dose. Solid and dashed lines represent normal tissue complication probability 1 and 2 yearsafter RT, respectively.

W.-C. Chen et al. / Oral Oncology 49 (2013) 42–48 45

order to identify if the patient- or treatment-related factors im-pacted parotid function recovery, we analyzed age, tumor site, par-otid mean dose, surgery and chemotherapy. Table 2 shows that theparotid mean dose was the only significant factor correlated withthe recovery of parotid gland function (R = �0.807, R2 = 0.651).

Normal tissue complication model

Fig. 2 shows the estimated TD50 and m for data at 1 and 2 yearsafter RT; no threshold dose was observed. Using the post-treat-ment SEF ratio of <45% proposed by Roesink et al.6 the TD50 andm values of the parotid gland were found to be 43.6 Gy, 0.18 and44.5 Gy, 0.30 1 and 2 years post-RT, respectively.

Patient-reported xerostomia

All questionnaires were filled out at each of the four time points(prior to treatment and 1, 3 and 12 months after IMRT); however,only 17 patients (54.8%) completed the questionnaires 24 monthsafter IMRT. As shown in Table 3, there were significant changesin several scales over this period in the surviving patients. A gen-eral trend of maximal deterioration in most QoL scales was ob-served at ti2, followed by a gradual recovery thereafter. Somesymptomatic problems, such as nausea/vomiting, appetite loss,pain, swallowing, taste/smell, and feeling ill, improved betweenti3 and ti4. Improvements in physical functioning, fatigue, socialeating, social contact, and sexuality also predominated betweenti3 and ti4. Compared with those at ti1, some head-and-necksymptomatic problems, such as swallowing, taste/smell, openingmouth, dry mouth, and sticky saliva, became significantly worseat ti4. For problems with swallowing, taste/smell and teeth, thescores are significantly different between ti1 and ti5. In order toclarify the effects of salivary gland function preservation on the re-lated QoL questionnaires, we compared patient-, treatment- or sal-ivary function (Table 4). By two-sample t-test, we found thatsurgery was associated with problems in opening the mouth. Par-otid gland functional preservation did not significantly improveswallowing, mouth opening, dry mouth or sticky saliva scores.On the contrary, contralateral submandibular and parotid glandfunctional preservation was better correlated with minimizing ofsticky saliva. By Spearman’s correlation, submandibular gland flow

recovery was the only factor correlated with sticky saliva (Table 5).Two examples of uptake and excretion curves are shown in Fig. 3.

Discussion

The results of this longitudinal study suggest that parotid glandexcretion recovery is strongly correlated with the mean parotidgland dose, as determined by scintigraphy analysis. The TD50 ofthe parotid gland was 43.6 Gy 1 year post-RT. Using the Spear-man’s correlation, we found a strong correlation between parotidgland excretion recovery and mean parotid gland dose at 1 and 2years post-RT. We further found that preservation of function ofthe contralateral parotid and submandibular glands predicts a bet-ter QoL, compared to preservation of the function of both parotidglands.

Several large prospective studies conducted over the past dec-ade have reported the dose constraints required to preserve paro-tid gland function, based on either salivary flow measurements orsalivary gland scintigraphy. In a pilot study conducted by Eisbruchet al. from the University of Michigan4, a TD50 of 26 Gy for the par-otid gland was reported by directly measuring stimulated salivaflow rates. Similar TD50 of 25.5 Gy and of 25.8 Gy were reportedby Saarilahti et al.7, and Blanco et al.8 using the same method.However, Roesink and her colleague from the University MedicalCenter Utrecht in the Netherlands reported quite different results.TD50 were 39 Gy and 43 Gy at 1 year post-RT, as measured by sal-ivary flow5 and salivary gland scintigraphy6, respectively. Althoughthis difference is likely attributable to the use of different RT tech-niques (IMRT by Eisbruch and conventional RT by Roesink), the dis-crepancy is large. In order to arrive at a definitive NTCP curve forparotid gland function, Dijkema et al. analyzed the combined andupdated results from both institutions, and found that the TD50

for parotid gland function 1 year after RT was 39.9 Gy.16 The cur-rent study shows a comparable TD50 (43.6 Gy 1 year post-RT). Inseveral studies of Asian patients, the mean dose reported to pre-serve parotid gland functions varied from 31 to 43.9 Gy9–11, yetno TD50 has been reported. Racial differences may not exist interms of the parotid gland response to irradiation.

We further found that parotid gland functional recovery was notcorrelated with other patient- or treatment- variables, such as age,tumor site, surgery or chemotherapy. Indeed, chemotherapy has

Table 3Change of the quality of life of 31 patients at 5 different time points after radiotherapy.

ti1 ti2 ti3 ti4 ti5 p value

EORTC QLC-C30Functional scales

Global quality of life 44 47 47 49 48 NSPhysical functioning 94 77a 77b 85c,f,h 82d <0.01Emotional functioning 80 77 77 81 83 NSCognitive functioning 87 79 78 81 79 NSSocial functioning 75 68 69 72 78 NSRole functioning– 87 80 76 88 86 NS

Symptom scalesFatigue 20 37a 33b 29f 30d,g <0.05Nausea/vomit 8 14a 13 3h 8g NSPain 17 28a 25 14h 18g NSDyspnea 5 13 17b 11 17 NSInsomnia 27 32 40 33 31 NSAppetite loss 15 27a 29b 23 24 NSConstipation 19 23 27 22 24 NSDiarrhea 13 14 13 6 7 NSFinancial problems 29 33 33 33 26 NS

EORTC QLQ-HN35Pain 17 32a 26b 13f,h 17g <0.05Swallowing 18 35a 36b 32c 34d <0.05Senses (taste/smell) 4 29a 27b 28c 28d <0.05Speech 18 25 27 21 19 NSSocial eating 18 34a 33b 26 28 NSSocial contact 13 28a 19e 17f 13g <0.05Sexuality 14 31a 32b 12f,h 19 NSTeeth 17 29a 31b 25 41d,i,j NSOpening mouth 20 27a 35b 33c 33 NSDry mouth 29 49a 48b 53c 48 <0.05Stick saliva 27 35 39 41c 39 NSCoughing 27 28 29 20h 24 NSFeeling ill 25 35a 31 22f,h 20g <0.05

Analysis by multivariate ANOVA for repeated measure.ti1 = before RT; ti2 = 1 month after RT; ti3 = 3 months after RT; ti4 = 12 months after RT; ti5 = 24 months after RT; NS = not significant.

a p < 0.05, ti2 compared with ti1.b p < 0.05, ti3 compared with ti1.c p < 0.05, ti4 compared with ti1.d p < 0.05, ti5 compared with ti1.e p < 0.05, ti3 compared with ti2.f p < 0.05, ti4 compared with ti2.g p < 0.05, ti5 compared with ti2.h p < 0.05, ti4 compared with ti3.i p < 0.05, ti5 compared with ti3.j p < 0.05, ti5 compared with ti4.

46 W.-C. Chen et al. / Oral Oncology 49 (2013) 42–48

been reported to reduce salivary flow17, albeit in a transient man-ner. In a large prospective study conducted by Marzi et al.18, a

Table 4Factors associated with the EORTC QLQ-H&N35 scores at 1 year after RT.

Factors 1-Year EORTC QLQ-HN35 score

Swallowing Openingmouth

Drymouth

Stickysaliva

Surgery Yes 35 48a 37 29b

No 28 18a 49 53b

Chemotherapy Yes 33 35 47 45No 30 30 36 33

Preservation of bothparotid glandsfunction

Yes 31 28 38 49No 32 37 46 35

Preservation ofcontralateral parotidand submandibularglands function

Yes 29 42 33 19c

No 33 31 46 47c

a p = 0.003.b p = 0.040.c p = 0.045.

similar TD50 of 41.6 Gy was found, yet 75% (44/59) of patients hadreceived chemotherapy as part of their treatment. In the presentstudy, chemotherapy was not found to be predictive of parotidgland functional recovery, as determined in a multivariate analysis.

Preservation of parotid gland function by IMRT significantly im-proved xerostomia-related symptoms and corresponded to a betterQoL compared to conventional RT.1 In the current study, most ofthe QoL results are superior to the patients receiving conventionalRT in the study conducted by Graff et al.3 The QoL ratings of our

Table 5Factors associated with sticky saliva and dry mouth.

Factors Sticky saliva Dry mouthSpearman’scorrelation

pvalue

Spearman’scorrelation

pvalue

Sum of salivary glandsflow recovery (%)

�0.1 0.956 �0.206 0.267

Sum of parotid glandsflow recovery (%)

0.174 0.348 �0.026 0.892

Submandibular glandflow recovery (%)

�0.43 0.016* �0.245 0.184

* p < 0.05, statistical significance.

Figure 3 The axial (1a, 1b) and coronal (2a, 2b) isodose and time activity curves on quantitative salivary scintigraphy of the four major salivary glands prior to intensity-modulated radiotherapy (IMRT) (3a, 3b) and 12 months post-IMRT (4a, 4b) of two patients (A and B). The recovery rates of R’t parotid (green line), L’t parotid (red line), R’tsubmandibular (dark green line) and L’t submandibular (orange line) gland function for patient A were 72, 64, 0 and 0% respectively. For patient B, only the L’t parotid and L’tsubmandibular gland functions were preserved (recovery rates, 115 and 59%, respectively). In the patient-reported quality of life questionnaire, the dry mouth score was 33.3for both patients; however, the sticky saliva scores were 66.7 and 0, respectively, indicating a better quality of life in patient B. (For interpretation of the references to color inthis figure legend, the reader is referred to the web version of this article.)

W.-C. Chen et al. / Oral Oncology 49 (2013) 42–48 47

patient groups were inferior to those reported by patients receiv-ing IMRT in the study of Pow et al.1, but are comparable to thoseof Graff et al.3 This may be because of the different patient cohorts(100% NPC patients in the study by Pow et al. vs. 16% NPC in Graffet al.). In our study, 45% of patients were treated with IMRT fortheir oral cavity cancer. Leung et al. confirmed that oral cancer sur-vivors have significantly worse QoL scales (social eating, social con-tact, open mouth, appetite loss, pain, swallowing, speech, socialeating and social contact) compared to NPC survivors.19 This couldbe attributed to that most of the oral cancer patients also receivedradical surgery as part of their initial treatment.

In the H&N35 module, swallowing, mouth opening, dry mouthand sticky saliva were reported as the four worst symptoms. In asubgroup analysis, we found that surgery, which might cause se-vere trismus (interincisor distance (IID)<2.5 cm), is an importantcontributor to poor QoL scores. In the current study, seven of eightpatients (87.5%) with IID < 2.5 cm received surgery as part of theirinitial treatment. Another important factor that may affect patient-reported xerostomia is submandibular gland function. Althoughonly 20–30% of stimulated saliva flow is produced by the subman-dibular glands, submandibular saliva is responsible for the major-ity of saliva production in the non-stimulated state. Further, itsmucin component could serve as the mucosal lubricant that con-tributes to a patient’s subjective sense of moisture. In a study con-ducted by Eisbruch et al.20, the major salivary gland flow rateswere found to have only a weak correlation with xerostomiascores. Factors found to be independently associated with xerosto-mia scores included the pre-RT baseline scores, the time since RT,and the mean doses delivered to the major salivary glands (notablyto the submandibular glands) and to the oral cavity. These findingsare comparable to those of the present study. We found that pres-ervation of one submandibular and one parotid gland, rather thanof both parotid glands, produced better xerostomia-related QoLscores.

In conclusion, recovery of parotid gland excretion was stronglycorrelated with the mean parotid gland dose, as determined byscintigraphy analysis. The TD50 of the parotid gland (43.6 Gy)inferred in our study is comparable to reports from westerncountries. Preservation of the function of at least one parotid and

one submandibular gland predicts a better QoL and minimal stickysaliva, as determined by self-reporting on the EORTC QLQ-H&N35questionnaire.

Funding sources

The study was supported by the National Science Council,Taiwan (Grant 96-2321-B-182A-004) and Chang Gung MemorialHospital (Grant CMRPG670371-2).

Conflict of interest statement

None declared.

References

1. Pow EH, Kwong DL, McMillan AS, Wong MC, Sham JS, Leung LH, et al.Xerostomia and quality of life after intensity-modulated radiotherapy vs.conventional radiotherapy for early-stage nasopharyngeal carcinoma: initialreport on a randomized controlled clinical trial. Int J Radiat Oncol Biol Phys2006;66(4):981–91.

2. Jabbari S, Kim HM, Feng M, Lin A, Tsien C, Elshaikh M, et al. Matched case-control study of quality of life and xerostomia after intensity-modulatedradiotherapy or standard radiotherapy for head-and-neck cancer: initial report.Int J Radiat Oncol Biol Phys 2005;63(3):725–31.

3. Graff P, Lapeyre M, Desandes E, Ortholan C, Bensadoun RJ, Alfonsi M, et al.Impact of intensity-modulated radiotherapy on health-related quality of life forhead and neck cancer patients: matched-pair comparison with conventionalradiotherapy. Int J Radiat Oncol Biol Phys 2007;67(5):1309–17.

4. Eisbruch A, Ten Haken RK, Kim HM, Marsh LH, Ship JA. Dose, volume, andfunction relationships in parotid salivary glands following conformal andintensity-modulated irradiation of head and neck cancer. Int J Radiat Oncol BiolPhys 1999;45(3):577–87.

5. Roesink JM, Moerland MA, Battermann JJ, Hordijk GJ, Terhaard CH. Quantitativedose-volume response analysis of changes in parotid gland function afterradiotherapy in the head-and-neck region. Int J Radiat Oncol Biol Phys2001;51(4):938–46.

6. Roesink JM, Moerland MA, Hoekstra A, Van Rijk PP, Terhaard CH. Scintigraphicassessment of early and late parotid gland function after radiotherapy for head-and-neck cancer: a prospective study of dose-volume response relationships.Int J Radiat Oncol Biol Phys 2004;58(5):1451–60.

7. Saarilahti K, Kouri M, Collan J, Hamalainen T, Atula T, Joensuu H, et al. Intensitymodulated radiotherapy for head and neck cancer: evidence for preservedsalivary gland function. Radiother Oncol 2005;74(3):251–8.

48 W.-C. Chen et al. / Oral Oncology 49 (2013) 42–48

8. Blanco AI, Chao KS, El Naqa I, Franklin GE, Zakarian K, Vicic M, et al. Dose-volume modeling of salivary function in patients with head-and-neck cancerreceiving radiotherapy. Int J Radiat Oncol Biol Phys 2005;62(4):1055–69.

9. Kam MK, Teo PM, Chau RM, Cheung KY, Choi PH, Kwan WH, et al. Treatment ofnasopharyngeal carcinoma with intensity-modulated radiotherapy: the HongKong experience. Int J Radiat Oncol Biol Phys 2004;60(5):1440–50.

10. Kwong DL, Pow EH, Sham JS, McMillan AS, Leung LH, Leung WK, et al. Intensity-modulated radiotherapy for early-stage nasopharyngeal carcinoma: aprospective study on disease control and preservation of salivary function.Cancer 2004;101(7):1584–93.

11. Hsiung CY, Ting HM, Huang HY, Lee CH, Huang EY, Hsu HC. Parotid-sparingintensity-modulated radiotherapy (IMRT) for nasopharyngeal carcinoma:preserved parotid function after IMRT on quantitative salivary scintigraphy,and comparison with historical data after conventional radiotherapy. Int JRadiat Oncol Biol Phys 2006;66(2):454–61.

12. Kam MK, Leung SF, Zee B, Chau RM, Suen JJ, Mo F, et al. Prospective randomizedstudy of intensity-modulated radiotherapy on salivary gland function in early-stage nasopharyngeal carcinoma patients. J Clin Oncol 2007;25(31):4873–9.

13. Chen WC, Hwang TZ, Wang WH, Lu CH, Chen CC, Chen CM, et al. Comparisonbetween conventional and intensity-modulated post-operative radiotherapyfor stage III and IV oral cavity cancer in terms of treatment results and toxicity.Oral Oncol 2009;45(6):505–10.

14. Lyman JT. Complication probability as assessed from dose-volume histograms.Radiat Res Suppl 1985;8:S13–9.

15. Bjordal K, Hammerlid E, Ahlner-Elmqvist M, de Graeff A, Boysen M, Evensen JF,et al. Quality of life in head and neck cancer patients: validation of theEuropean Organization for Research and Treatment of Cancer Quality of LifeQuestionnaire-H&N35. J Clin Oncol 1999;17(3):1008–19.

16. Dijkema T, Raaijmakers CP, Ten Haken RK, Roesink JM, Braam PM, HouwelingAC, et al. Parotid gland function after radiotherapy: the combined michigan andutrecht experience. Int J Radiat Oncol Biol Phys 2010;78(2):449–53.

17. Jensen SB, Mouridsen HT, Reibel J, Brunner N, Nauntofte B. Adjuvantchemotherapy in breast cancer patients induces temporary salivary glandhypofunction. Oral Oncol 2008;44(2):162–73.

18. Marzi S, Iaccarino G, Pasciuti K, Soriani A, Benassi M, Arcangeli G, et al. Analysisof salivary flow and dose-volume modeling of complication incidence inpatients with head-and-neck cancer receiving intensity-modulatedradiotherapy. Int J Radiat Oncol Biol Phys 2009;73(4):1252–9.

19. Wan Leung S, Lee TF, Chien CY, Chao PJ, Tsai WL, Fang FM. Health-relatedquality of life in 640 head and neck cancer survivors after radiotherapy usingEORTC QLQ-C30 and QLQ-H&N35 questionnaires. BMC Cancer 2011;11:128.

20. Eisbruch A, Kim HM, Terrell JE, Marsh LH, Dawson LA, Ship JA. Xerostomia andits predictors following parotid-sparing irradiation of head-and-neck cancer.Int J Radiat Oncol Biol Phys 2001;50(3):695–704.