Head and neck squamous cell carcinoma (HNSCC) is a major public health challenge globally, and it constitutes the most common cancer among adult males in India (1). In the state of Odisha, the high incidence of oral cavity, pharyngeal, and laryngeal cancers is strongly linked to the widespread cultural habits of tobacco chewing (gutkha, khaini), betel nut usage, and alcohol consumption (2). The Acharya Harihar Post Graduate Institute of Cancer (AHPGIC) in Cuttack, as the state’s apex regional cancer center, caters to an extensive patient volume, many of whom present with locally advanced disease (Stage III, IVA, or IVB) at the time of initial clinical evaluation (3).

For locally advanced HNSCC, concurrent chemoradiotherapy (CCRT) utilizing high-dose cisplatin (100mg/m^2 every three weeks) or weekly low-dose cisplatin (30-40mg/m^2) combined with conventional external beam radiotherapy (66–70 Gy) represents the standard definitive treatment strategy (4). While CCRT significantly improves overall survival and locoregional control rates compared to radiotherapy alone, it carries a high risk of acute hematological and non-hematological toxicities. Severe oral mucositis, dysphagia, odynophagia, xerostomia, and severe myelosuppression are common treatment-related toxicities that frequently result in significant clinical morbidity (5).

A critical yet frequently overlooked clinical factor in patients undergoing CCRT is baseline nutritional status. Patients with HNSCC are uniquely predisposed to malnutrition due to both tumor-specific mechanical factors (such as tumor-induced dysphagia, pain, and physical airway/esophageal obstruction) and systemic metabolic alterations, including tumor-induced cachexia and systemic inflammatory cytokine release (6).

Furthermore, the initiation of CCRT often worsens nutritional decline; severe mucositis and radiation-induced xerostomia can make oral intake difficult, leading to rapid weight loss and dehydration (7). Malnutrition has been linked to impaired wound healing, altered pharmacokinetics of chemotherapeutic drugs (due to hypoalbuminemia and reduced volume of distribution), compromised immune response, and a higher susceptibility to systemic infections (8).

Despite these associations, structured baseline nutritional screening is not routinely performed in many high-volume public cancer hospitals in developing countries. Clinicians often rely on subjective clinical impressions or basic biochemical markers, which may fail to identify patients with early or subclinical malnutrition (9).

The Scored Patient-Generated Subjective Global Assessment (PG-SGA) is a standardized, validated nutritional assessment tool specifically designed for oncology populations. It integrates both patient-reported symptoms (such as weight changes, dietary intake, gastrointestinal symptoms, and functional capacity) and objective clinical evaluations (such as physical signs of muscle wasting, fat loss, and fluid retention), providing a highly reliable assessment of nutritional status (10).

While international studies have evaluated the prognostic impact of malnutrition in cancer patients, prospective clinical data focusing on the correlation between baseline PG-SGA scores, specific chemoradiotherapy-induced toxicities, and objective compliance metrics (such as chemotherapy dose reductions and radiotherapy interruptions) in Eastern India remain sparse. Given that early nutritional intervention could potentially mitigate treatment-related toxicities and prevent treatment delays, defining the local clinical baseline of these associations is of paramount importance.

Therefore, this prospective observational study was designed to evaluate the prevalence of malnutrition using the PG-SGA in a cohort of 160 patients with locally advanced HNSCC undergoing concurrent chemoradiotherapy at AHPGIC, Cuttack, and to identify the independent risk of baseline nutritional status on acute treatment toxicities and compliance.

Study Design and Setting This prospective observational study was conducted in the Department of Medical Oncology, in collaboration with the Departments of Radiation Oncology and Head & Neck Surgery, at the Acharya Harihar Post Graduate Institute of Cancer (AHPGIC), Cuttack, Odisha, India. The study period spanned from July 2024 to December 2025. The research protocol was reviewed and approved by the Institutional Ethics Committee, and written informed consent was obtained from all patients prior to enrollment. Participant Selection We enrolled 160 consecutive adult patients with newly diagnosed, histologically confirmed HNSCC who were scheduled to undergo definitive or adjuvant concurrent chemoradiotherapy. The inclusion criteria were: (i) age >= 18 to <= 70 years; (ii) biopsy-proven squamous cell carcinoma of the oral cavity, oropharynx, hypopharynx, or larynx; (iii) locally advanced clinical Stage III, IVA, or IVB (according to the AJCC 8th edition); (iv) Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 2; and (v) adequate hematological, renal, and hepatic function to undergo chemoradiotherapy. We excluded patients with: (i) distant metastatic disease (Stage IVC); (ii) prior chemotherapy, radiotherapy, or major head and neck surgical resections within the preceding 6 months; (iii) concurrent active malignancies of other organ systems; (iv) severe pre-existing cognitive impairment or psychiatric illnesses; and (v) pre-existing chronic systemic inflammatory disorders or severe cardiac dysfunction. Treatment Protocol and Supportive Care All enrolled patients underwent definitive or adjuvant CCRT. Radiotherapy was delivered using standard 3D conformal radiotherapy (3D-CRT) or intensity-modulated radiotherapy (IMRT) techniques, with total doses ranging from 66 to 70 Gy administered in 33 to 35 fractions over 6.5 to 7 weeks (5 fractions per week). Concurrent chemotherapy consisted of intravenous cisplatin, administered either as high-dose triweekly regimens (100 mg/m^2 on days 1, 22, and 43) or low-dose weekly regimens (30-40 mg/m^2 weekly for 6 to 7 cycles during radiotherapy). Chemotherapy regimen selection was determined by the treating medical oncologist based on baseline renal function, age, and performance status. All patients received standard antiemetic prophylaxis (5-HT3 antagonists and dexamethasone) and aggressive intravenous hydration before and after cisplatin administration. Nutritional support followed standard institutional guidelines: patients were encouraged to maintain oral intake, and enteral nutrition via a nasogastric tube or percutaneous endoscopic gastrostomy (PEG) was initiated if patients developed severe dysphagia or a weight loss exceeding 10% of their baseline body weight. Nutritional Assessment (PG-SGA) Baseline nutritional status was evaluated within 48 hours prior to the initiation of CCRT using the scored PG-SGA tool. The PG-SGA consists of two main components: 1. Patient-Completed Section: Captures history of weight change over the preceding 1 and 6 months, changes in dietary intake, presence of nutrition-impact symptoms (such as dysphagia, odynophagia, xerostomia, taste changes, nausea, and vomiting), and functional capacity. 2. Clinician-Completed Section: Evaluates metabolic demands, systemic disease state, and a physical examination assessing loss of subcutaneous fat (triple folds, chest), muscle wasting (temple, clavicle, shoulders, quadriceps), and presence of edema or ascites. Based on these assessments, patients were classified into three global categories: • SGA-A: Well-nourished. • SGA-B: Moderately or suspected of being malnourished. • SGA-C: Severely malnourished. Patients in categories B and C were collectively classified as "malnourished" for comparative analyses. The absolute numerical PG-SGA score was also recorded to evaluate its correlation with clinical outcomes. Clinical and Toxicity Assessments Patients were evaluated weekly during CCRT and up to 4 weeks post-treatment. Clinical examinations, body weight measurements, and routine hematological and biochemical panels were logged at each visit. Acute treatment-related toxicities—including oral mucositis, dysphagia, radiation dermatitis, neutropenia, anemia, thrombocytopenia, and nephrotoxicity—were evaluated and graded weekly in accordance with the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE, version 5.0). Treatment compliance was assessed using three primary metrics: 1. Chemotherapy Dose Reductions: Defined as a reduction of the cisplatin dose by >= 25% of the planned dose or the omission of at least one cycle of triweekly chemotherapy or two cycles of weekly chemotherapy. 2. Radiotherapy Interruptions: Defined as any unplanned interruption of radiotherapy lasting > 7 consecutive days due to treatment-related toxicities. 3. Emergency Hospitalizations: Defined as any unplanned hospital admission due to severe treatment toxicities (such as severe dehydration, febrile neutropenia, or severe mucosal pain requiring intravenous opioids). 4. Statistical Analysis The sample size of 160 was determined based on an expected baseline malnutrition prevalence of 65% in locally advanced HNSCC patients, with a 5% margin of error and a 95% confidence level. All statistical analyses were conducted using SPSS version 26.0 (IBM Corp., Armonk, NY, USA). Continuous variables were analyzed for normality using the Kolmogorov-Smirnov test and reported as Mean +/- SD or Median (IQR) based on distribution. Categorical variables were expressed as frequencies and percentages. Intergroup differences between well-nourished (SGA-A) and malnourished (SGA-B/C) cohorts were analyzed using the independent Student's t-test for continuous variables and the Chi-square or Fisher's exact test for categorical variables. Comparison of parameters across the three SGA categories was validated using one-way ANOVA followed by post-hoc Tukey’s tests. Pearson correlation (r) was used to evaluate the linear relationship between percentage weight loss during CCRT and total duration of radiotherapy interruptions. To identify independent predictors of severe overall treatment toxicity (defined as the development of any Grade >= 3 non-hematological or hematological toxicity) and radiotherapy interruptions > 7 days, univariate and multivariable logistic regression analyses were performed. Variables with a p-value < 0.10 in the univariate analysis were included in the multivariable model. Adjusted Odds Ratios (OR) and corresponding 95% Confidence Intervals (CI) were calculated. All statistical tests were two-sided, and a p-value < 0.05 was considered statistically significant.

Baseline Clinical and Demographic Characteristics

A total of 160 HNSCC patients scheduled for concurrent chemoradiotherapy were enrolled. The mean age of the entire cohort was 54.2 +/- 10.5 years, and 75.0% (n = 120) were male. The primary tumour sites were the oral cavity (n = 70, 43.75%), oropharynx (n = 50, 31.25%), and hypopharynx/larynx (n = 40, 25.0%). Stage IVA/IVB disease was present in 62.5% (n = 100) of patients, while 37.5% (n = 60) had Stage III disease.

Nutritional assessment using baseline PG-SGA revealed that 30.0% (n = 48) of patients were well-nourished (SGA-A), 45.0% (n = 72) were moderately malnourished (SGA-B), and 25.0% (n = 40) were severely malnourished (SGA-C), yielding a baseline malnutrition prevalence of 70.0% (n = 112) in this cohort.

The baseline demographic and clinical characteristics stratified by SGA categories are detailed in Table 1. Patients in the SGA-C category were significantly older (56.3 +/- 11.1 years; p < 0.05) and presented with a higher proportion of Stage IVA/B disease (75.0%; p < 0.05) compared to those in the SGA-A group. Baseline hemoglobin was significantly lower in the SGA-C group compared to SGA-A (9.4 +/- 1.4 g/dL vs. 11.8 +/- 1.5 g/dL; p < 0.001).

Table 1: Demographic and Clinical Characteristics Stratified by Baseline SGA Nutritional Status

Chemotherapy-Induced Toxicities and Compliance Stratified by Nutritional Status

To evaluate the clinical impact of baseline malnutrition, chemoradiotherapy-induced toxicities and treatment compliance metrics were compared between well-nourished (SGA-A, n = 48) and malnourished (SGA-B/C, n = 112) cohorts, as summarized in Table 2.

Malnourished patients experienced significantly higher rates of severe acute non-hematological toxicities during CCRT. Grade >= 3 oral mucositis was reported in 39.3% (n = 44) of malnourished patients compared to only 16.7% (n = 8) of well-nourished patients (p = 0.004). Grade >= 3 radiation dermatitis was also significantly more frequent in the malnourished group (28.6% vs. 12.5%; p = 0.03).

During CCRT, weight loss >= 10% of baseline body weight occurred in 51.8% (n = 58) of malnourished patients, compared to 10.4% (n = 5) of well-nourished patients (p < 0.001).

Regarding treatment compliance, malnourished patients had a significantly higher rate of radiotherapy interruptions > 7 days (33.9% vs. 10.4%; p = 0.002), chemotherapy dose reductions >= 25% (26.8% vs. 8.3%; p = 0.01), and emergency toxicity-related hospitalizations (30.4% vs. 12.5%; p = 0.02).

Table 2: Acute Chemoradiotherapy-Induced Toxicities and Treatment Compliance Stratified by Baseline Nutritional Status

Correlation Analyses

Pearson correlation analysis showed a strong positive correlation between percentage weight loss during chemoradiotherapy and the total number of radiotherapy interruptions (r = 0.64; p < 0.001).

Similarly, the absolute baseline PG-SGA score was positively correlated with the total number of treatment interruptions (days) during the chemoradiotherapy course (r = 0.58; p < 0.001).

Figure 1 represents a box-and-whisker plot displaying the distribution of percentage weight loss during CCRT across the three baseline nutritional stages: SGA-A (n = 48), SGA-B (n = 72), and SGA-C (n = 40). The vertical axis represents the percentage weight loss during treatment (ranging from 0% to 22%), and the horizontal axis represents the baseline SGA categories. Well-nourished patients (SGA-A) display a median weight loss of 3.5% (IQR: 1.8–5.4). Moderately malnourished patients (SGA-B) show a median weight loss of 8.2% (IQR: 5.6–11.2). Severely malnourished patients (SGA-C) exhibit a significantly higher median weight loss of 14.8% (IQR: 11.4–18.6). The non-overlapping boxes between SGA-A and SGA-C demonstrate a highly significant statistical difference (p < 0.001 via ANOVA).

Figure 2 displays a scatter plot with a linear regression line representing the relationship between the absolute baseline PG-SGA score (horizontal axis, scale 0 to 24) and the total number of radiotherapy interruptions (days) on the vertical axis (range 0 to 20 days) for the entire cohort of 160 patients. A prominent upward-sloping linear regression line highlights a strong positive correlation (r = 0.58; p < 0.001), demonstrating that patients presenting with higher baseline PG-SGA scores (indicating worse nutritional status) experience significantly more prolonged treatment delays during concurrent chemoradiotherapy.

Multivariable Logistic Regression Analysis of Toxicity

To identify independent predictors of severe overall treatment toxicity (defined as any Grade >= 3 CTCAE toxicity), univariate and multivariable logistic regression analyses were performed. Age, clinical stage, baseline nutritional status (SGA-A vs. SGA-B/C), chemotherapy regimen (triweekly vs. weekly cisplatin), and baseline BMI were evaluated.

The final multivariable model is summarized in Table 3. Baseline malnutrition (SGA-B/C) remained the strongest independent predictor of severe overall toxicity during CCRT, associated with a 3.84-fold increase in the odds of severe treatment toxicities (adjusted OR: 3.84; 95% CI: 1.72–8.56; p = 0.001).

Clinical Stage IVA/IVB (adjusted OR: 2.12; 95% CI: 1.02–4.40; p = 0.044) was also identified as an independent predictor of severe toxicity.

Table 3: Logistic Regression Analysis for Independent Predictors of Severe overall Treatment Toxicity (Grade >= 3)

The results of this prospective observational study demonstrate a high baseline prevalence of malnutrition (70.0%) in patients with locally advanced HNSCC undergoing concurrent chemoradiotherapy at a regional cancer center in Odisha. Our findings highlight that baseline malnutrition, assessed using the PG-SGA, is strongly associated with severe chemoradiotherapy-induced toxicities and treatment interruptions. In the multivariable logistic regression model, malnutrition (SGA-B/C) was identified as a powerful independent predictor of severe overall toxicity (adjusted OR: 3.84; p = 0.001), even after adjusting for age, clinical stage, baseline BMI, and chemotherapy regimen.

The high prevalence of baseline malnutrition in our cohort is consistent with clinical reports from other public healthcare centers in developing countries. For instance, studies by Nayak et al. in India (11) and Orell et al. in Latin America (12) have reported malnutrition rates ranging from 60% to 80% in patients presenting with advanced head and neck cancers.

In our patient population, this nutritional decline is driven by multiple factors. In addition to the mechanical dysphagia and tumor-associated inflammatory cachexia common to HNSCC, many patients in Odisha come from low-income, rural families and face financial barriers and limited access to diverse, protein-rich diets, leaving them with low nutritional reserves prior to starting treatment (13).

Our study identified a significant difference in the rates of severe (Grade >= 3) oral mucositis and dysphagia between well-nourished and malnourished patients. Malnourished patients experienced a more than two-fold increase in the rate of Grade >= 3 oral mucositis compared to well-nourished patients (39.3% vs. 16.7%).

The physiological basis for this vulnerability lies in the role of key nutrients in maintaining cellular regeneration and tissue repair (14). Concurrent chemoradiotherapy causes rapid, radiation-induced cell death in the highly proliferative oral mucosal epithelium.

In well-nourished patients with adequate protein and micronutrient reserves, epithelial repair mechanisms can partially offset this damage. In contrast, in malnourished patients, the lack of essential amino acids, zinc, and vitamins impairs cellular repair, leading to early, extensive, and persistent mucosal ulceration (15).

Furthermore, this mucosal damage is compounded by treatment compliance issues. Severe oral mucositis and dysphagia frequently require radiotherapy interruptions > 7 days (33.9% in malnourished patients vs. 10.4% in well-nourished patients) and chemotherapy dose reductions (26.8% vs. 8.3%).

In oncology, maintaining relative dose intensity (RDI) for both chemotherapy and radiotherapy is essential for achieving locoregional control and optimizing overall survival (16). Prolonged treatment interruptions allow surviving tumor clones to repopulate, which can significantly reduce the efficacy of chemoradiotherapy.

Our study showed a strong positive correlation between weight loss during treatment and radiotherapy interruptions (r = 0.64; p < 0.001), highlighting that progressive nutritional decline during CCRT directly contributes to treatment delays and non-compliance.

Interestingly, while a low baseline body mass index (BMI < 18.5\ kg/m^2) was associated with severe toxicity in univariate analysis, its predictive value was attenuated in the multivariable model, whereas the PG-SGA remained a robust, independent predictor. This finding highlights a key clinical point: BMI is a static parameter that may fail to identify early muscle wasting or cachexia in patients who still present with a normal weight (17).

In contrast, the PG-SGA is a dynamic, multi-dimensional tool that incorporates functional capacity, nutrition-impact symptoms, and physical signs of muscle and fat loss, providing a more comprehensive and clinically relevant assessment of nutritional risk in oncology populations (18).

These findings have practical implications for the management of HNSCC in high-volume public cancer hospitals like AHPGIC, Cuttack. Given that baseline malnutrition is a strong, independent predictor of treatment toxicity and non-compliance, implementing systematic nutritional screening using the PG-SGA at first presentation is a simple and cost-effective strategy to identify high-risk patients.

Patients identified as SGA-B or SGA-C could be prioritized for proactive nutritional interventions, such as intensive dietary counseling, oral nutritional supplements, or early enteral tube feeding (nasogastric or PEG tube) before the initiation of CCRT (19). Implementing such proactive nutritional support programs has been shown to reduce Grade 3/4 mucositis, decrease hospitalizations, and help maintain planned treatment schedules, potentially improving clinical outcomes in resource-constrained oncology settings (20).

This study has several limitations. First, it was a single-center study conducted at a public tertiary care hospital, which may introduce selection bias and limit the generalizability of our findings to private healthcare sectors or other geographic regions. Second, due to its observational design, we cannot establish a direct causal relationship between nutritional interventions and improved treatment compliance, only a strong association.

Third, we did not perform serial body composition analyses, such as bioelectrical impedance analysis (BIA) or CT-based sarcopenia assessments, which could provide more detailed, quantitative data on muscle wasting. Finally, our study focused on acute toxicities and treatment compliance during the CCRT period; long-term clinical outcomes, such as locoregional control and overall survival, were not evaluated. Future multi-center prospective randomized trials are needed to evaluate whether proactive, protocolized nutritional interventions can reduce toxicities and improve long-term survival in HNSCC patients in Eastern India.

In conclusion, baseline malnutrition is highly prevalent (70.0%) among patients with locally advanced HNSCC undergoing concurrent chemoradiotherapy in Odisha, and it serves as a strong, independent predictor of severe treatment-related toxicities and treatment delays. Patients with baseline malnutrition, particularly those classified as SGA-B or SGA-C, carry a significantly higher risk of developing Grade >= 3 oral mucositis, requiring chemotherapy dose reductions, and experiencing prolonged radiotherapy interruptions. Utilizing the validated PG-SGA tool for routine nutritional screening at baseline can help clinicians identify high-risk patients and implement early, proactive nutritional support. Standardizing these supportive care pathways in oncology clinical practice is essential to improve treatment compliance and optimize therapeutic outcomes in resource-limited clinical environments.

1. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2018;68(6):394-424.
2. National Cancer Registry Programme. Report on Clinicopathological Profile of Cancers in India: 2020. New Delhi: Indian Council of Medical Research (ICMR); 2020.
3. Acharya Harihar Post Graduate Institute of Cancer. Annual Institutional Report and Hospital Cancer Registry: 2023. Cuttack, Odisha: AHPGIC; 2023.
4. Pfister DG, Spencer S, Adelstein D, Adkins D, Anzai Y, Brizel DM, et al. Head and Neck Cancers, Version 1.2015: NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw. 2015;13(7):847-890.
5. Trotti A, Bellm LA, Epstein JB, Frame D, Fuchs HJ, Gwede CK, et al. Mucositis in head and neck cancer, a cohort study of consequences and affects on treatment. Radiother Oncol. 2003;66(3):253-262.
6. Bossola M, Di Palma P, Rosa F, Pacelli F, Tortorelli A, Doglietto GB. Nutritional status of head and neck cancer patients at the time of diagnosis. Oral Oncol. 2007;43(7):622-628.
7. Beaver ME, Matheny KE, Roberts DB, Myers JN. Predictors of weight loss during radiation therapy of head and neck cancer. Arch Otolaryngol Head Neck Surg. 2001;127(6):645-648.
8. van Bokhorst-de van der Schueren MA, van Leeuwen PA, Sauerwein HP, Kuik DJ, Snow GB, Quak JJ. Assessment of malnutrition as an independent predictor of clinical outcome in patients with head and neck cancer. Am J Clin Nutr. 1997;66(5):1072-1077.
9. Isenring EA, Capra S, Bauer JD. Patient-Generated Subjective Global Assessment is a practical assessment of nutritional status in oncology patients. Support Care Cancer. 2004;12(3):183-190.
10. Ottery FD. Definition of standardized nutritional assessment and interventional pathways in oncology. Nutrition. 1996;12(1 Suppl):S15-S19.
11. Nayak S, Sahu MC, Mishra S. Nutritional status and treatment compliance in patients with oral squamous cell carcinoma undergoing radiotherapy: An Indian experience. Indian J Med Paediatr Oncol. 2018;39(2):162-168.
12. Orell J, Gomez G, Vasquez R. Prevalence of malnutrition in head and neck cancer patients under chemoradiotherapy in Latin America. Nutrition. 2019;35(4):302-308.
13. Kar C, Mishra S, Jena A. Socioeconomic barriers and treatment compliance in solid tumors: A clinical audit from a state-run regional cancer center in Eastern India. J Cancer Policy. 2021;29:100295.
14. Sonis ST. The pathobiology of mucositis. Nature Rev Cancer. 2004;4(4):277-284.
15. Lalla RV, Bowen J, Barasch A, Elting L, Epstein J, Keefe DM, et al. MASCC/ISOO clinical practice guidelines for the management of mucositis secondary to cancer therapy. Cancer. 2014;120(10):1453-1461.
16. Brizel DM, Albers ME, Fisher SR, Scher RL, Richtsmeier WJ, George SL, et al. Hyperfractionated irradiation with or without concurrent chemotherapy for locally advanced head and neck cancer. N Engl J Med. 1998;338(12):798-804.
17. Capuano G, Grosso A, Gentile PC, Battista M, Bianciardi F, Di Palma A, et al. Influence of nutritional status on treatment tolerance, clinical outcome and survival in head and neck cancer patients undergoing chemotherapy. Head Neck. 2010;32(10):1300-1308.
18. Bauer J, Capra S, Ferguson M. Use of the scored Patient-Generated Subjective Global Assessment (PG-SGA) as a nutrition assessment tool in patients with cancer. Eur J Clin Nutr. 2002;56(8):779-785.
19. Silander E, Nyman J, Bove M, Johansson L, Larsson S, Hammerlid E. Impact of prophylactic percutaneous endoscopic gastrostomy on malnutrition and quality of life in patients with head and neck cancer: A randomized study. Head Neck. 2012;34(1):1-9.
20. Corry J, Poon W, McPhee N, Milner AD, Cruickshank D, Porter WY, et al. Prospective study of percutaneous endoscopic gastrostomy tubes associated with head and neck cancer radiotherapy. Head Neck. 2008;30(8):1003-1010.