The Laryngoscope C 2015 The American Laryngological, V

Rhinological and Otological Society, Inc.

Long-term Functional Outcomes in Surgically Treated Patients With Oropharyngeal Cancer Oliver T. Dale, BMBS, MRCS; Cheng Han, PhD; Christopher A. Burgess, FRCS, MEd; Susannah Eves, MRCS; Carol E. Harris, BSc; Penny L. White, BSc; Rupali T. Shah, BSc; Alison Howard, BSc; Stuart C. Winter, MD, FRCS Objectives/Hypothesis: As survival rates in oropharyngeal cancer improve, long-term functional outcomes are increasingly important to understand. We report long-term functional outcomes in a cohort of surviving patients with oropharyngeal squamous cell carcinoma treated with primary surgery 6 radiotherapy. Study Design: Cross-sectional study. Methods: Patients undergoing primary surgery for oropharyngeal cancer in Oxford, United Kingdom, between 2000 and 2010 were identified. The University of Washington Quality-of- Life and MD Anderson Dysphagia Inventory questionnaires were sent to all patients. Multivariate analysis was performed to determine the relationship between clinical factors and swallowing outcomes. Results: Twenty percent of patients required gastrostomy-tube placement (mean feed duration, 114 days). On multivariate analysis, increased age, advanced T stage, and an open surgical approach were associated with significantly reduced quality-of-life scores. Conclusions: Mean functional scores were comparable to previously published series of patients treated with primary surgery. Gastrostomy insertion rate was lower than in many previously published studies. Furthermore, specific variables have been identified that are associated with adverse functional outcome. Key Words: Quality of life, oropharynx, head and neck. Level of Evidence: 4 Laryngoscope, 125:1637–1643, 2015

INTRODUCTION In the United Kingdom, the incidence of oropharyngeal squamous cell carcinoma (OPSCC) has increased over the last 2 decades from 2.9 per 100,000 males in 1992 to 3.5 per 100,000 in 2010, in common with other countries across the world.1–4 The rising incidence of OPSCC is attributed to an increasing role of human papillomavirus (HPV). Evidence suggests that HPV-positive OPSCC represents a distinct disease entity, tending to occur in a younger patient group, with an improved treatment response and survival when compared to HPV-negative disease.5 The younger patient demographic and improved survival outcomes mean that long-term functional outcomes are becoming increasingly important. Treatment for OPSCC remains controversial, and although there has been a significant increase in the use

From the Head and Neck Department, John Radcliffe/Churchill Hospitals, Oxford, United Kingdom Editor’s Note: This Manuscript was accepted for publication January 30, 2015. Heads Up registered charity number 1057295 (fund number: 2119). The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Mr. Oliver Dale, Research Fellow, John Radcliffe Hospital, Oxford, OX3 9DU, UK. E-mail: [email protected] DOI: 10.1002/lary.25226

Laryngoscope 125: July 2015

of primary chemoradiotherapy, surgical resection, reconstruction 6 adjuvant therapy remains an accepted treatment modality. The short-term toxicity of treatment and functional outcomes have been well described; however, there are limited data on long-term functional outcomes, in particular with regard to primary surgery and the impact on swallowing.6–9 The aim of this study was to describe the long-term functional outcomes in a series of patients undergoing primary surgery for OPSCC using established and validated head and neck–specific quality-of-life outcome tools, with a secondary aim of identifying surgical variables, that may influence long-term function.

MATERIALS AND METHODS The Oxford Head and Neck Cancer Database was used to identify all patients undergoing primary surgery for oropharyngeal cancer in Oxford, United Kingdom, between 2000 and 2010. Surgery comprised either “lip-split” mandibulotomy or transoral CO2 laser resection. Lip-split mandibulotomy was performed using a preplated, paramedian osteotomy, with division of the mylohyoid to gain access to the posterior oropharynx. Neither mandibulectomy nor osteocutaneous free flap reconstruction of the mandible were performed in any cases. Laser resection was performed using a CO2 laser at 4 W, mounted on a Zeiss operating microscope (Carl Zeiss AG, Jena, Germany) and laser micromanipulator. Oropharyngeal subsites included the tonsil, tongue base, soft palate, and posterior pharyngeal wall. Case note review was performed on all patients to confirm

Dale et al.: Functional Outcomes in OPSCC Patients

1637

the database findings, and only patients with a histological diagnosis of squamous cell carcinoma were included. Patient demographics, tumor site, and stage (Union for International Cancer Control, 7th Edition) were confirmed. Patients with previous head and neck cancer and those receiving primary radiotherapy or chemoradiotherapy were excluded from the study. Swallowing and quality-of-life questionnaires were sent to all surviving patients. Questionnaires included the University of Washington Quality-of-Life Questionnaire (UWQoL) version 4 and the MD Anderson Dysphagia Inventory (MDADI). Primary outcome measures were the rate of gastrostomytube insertion both during treatment and at 1 year, and questionnaire outcome scores. Scores for the UWQoL and MDADI questionnaires range from 0 to 100 and 20 to 100, respectively, with a score of 100 indicating maximal function. UWQoL outcomes for individual symptoms and composite physical and social/emotional scores were analyzed as previously described.10 Mean MDADI scores were calculated for each domain (global, emotional, physical, functional), and mean total scores were also determined.11 Questionnaire scores (MDADI mean domain scores and mean total score, UWQoL composite scores) and gastrostomy data were entered into contingency tables and were analyzed using the Pearson v2 test or Fisher exact test. Significant associations from the bivariate analysis were entered into multivariate analysis using age, T stage, American Joint Committee on Cancer stage, resection type, reconstruction type, chemotherapy, and time from treatment as independent variables. All statistics were performed using the Stata package release 11.0 (StataCorp, College Station, Texas). Mean duration of follow-up was 93 months (range, 53–165 months).

RESULTS In total, 168 patients with OPSCC were treated with primary surgery 6 adjuvant therapy. At the time of writing there were 107 patients alive and disease free. Seventy-two patients (65%) responded to the postal questionnaires and were included in the study. Of these patients, 56 (78%) were male, and 65 (90%) presented with advanced disease (stage 3 or 4). The primary site was tonsil in 49 cases (68%) and base of tongue in 19 cases (26%). Twenty-six patients (36%) underwent transoral surgery, and 46 (64%) had open surgery (lip-split mandibulotomy and oropharyngeal resection). Three patients (4%) were treated with no adjuvant treatment, 69 (96%) received external beam radiotherapy, and 16 patients (22%) received adjuvant chemoradiotherapy. During this study period, no patient received intensitymodulated radiation therapy (IMRT). Full demographic details for all patients are shown in Table I. In this study, 16 patients (22%) required gastrostomytube placement. Gastrostomy insertion rates were higher in patients undergoing chemoradiotherapy than those receiving radiotherapy alone (P 5.04). The mean duration of gastrostomy feeding was 129 days (range, 7–484 days). One patient had a gastrostomy tube in situ at 1 year following completion of treatment, none were tube dependent at final follow-up, and tube removal was achieved in all patients. Responses to the UWQoL and MDADI questionnaires are shown in Tables II and III. For the UWQoL questionnaire, the mean composite physical function score was 72 (standard deviation [SD] 15), and the mean Laryngoscope 125: July 2015

1638

TABLE I. Demographic Details. No. of Patients (%)

Total no. Gender Male

72 56 (78%)

Female T stage 1

16 (22%) 23 (32%)

2

35 (49%)

3 4

5 (7%) 9 (13%)

N stage 0 1

12 (17%) 9 (13%)

2a

13 (18%)

2b 2c

31 (43%) 4 (6%)

3

3 (4%)

Stage group 1

3 (4%)

2

4 (6%)

3 4

7 (10%) 58 (81%)

Tumor site Tonsillar fossa Tonsil

1 (1%) 49 (68%)

Base of tongue

19 (26%)

Posterior pharyngeal wall Soft palate

1 (1%) 1 (1%)

Overlapping lesion of oropharynx

1 (1%)

Resection type Laser resection Lip split mandibulotomy Reconstruction type No reconstruction Free flap Radiotherapy type None Adjuvant Chemotherapy type None Adjuvant chemoradiotherapy PEG status No Yes

26 (36%) 46 (64%) 27 (38%) 45 (63%) 3 (4%) 69 (96%) 56 (78%) 16 (22%) 56 (78%) 16 (22%)

PEG 5 percutaneous endoscopic gastrostomy.

social/emotional score was 80 (SD 16). Mean overall quality-of-life score from the UWQoL questionnaire was 68 (SD 20). Mean MDADI domain scores were as follows: Global 72 (SD 26), Emotional 74 (SD 20), Functional 81 (SD 19) and Physical 68 (SD 20). The mean total MDADI score was 74 (SD 19). Dale et al.: Functional Outcomes in OPSCC Patients

TABLE II. Mean Outcome Scores for the MD Anderson Dysphagia Inventory (n 5 72). MDADI domain

Mean Score (SD)

TABLE III. Mean Outcome Scores for the University of Washington Qualityof-Life Questionnaire (n 5 72). UWQoL question

Mean (SD)

Global

72.11 (25.63)

Taste

75.28 (28.68)

Emotional Functional

73.51 (19.68) 81.13 (19.39)

Saliva Appearance

53.61 (30.41) 71.88 (15.66)

Physical

68.42 (19.86)

Chewing

73.89 (28.90)

Mean total MDADI score

73.89 (18.77)

Speech Swallowing

82.50 (18.44) 75.21 (18.07)

Composite Physical Score

72.18 (15.08)

Activity Shoulder

78.52 (19.50) 72.78 (26.60)

MDADI 5 MD deviation.

Anderson

Dysphagia

Inventory;

SD 5 standard

The results from bivariate analysis are shown in Table IV. In bivariate analysis, increasing age, lip-split mandibulotomy, free flap reconstruction, and advanced tumor T stage were associated with lower functional scores in the “physical composite” score of the UWQoL questionnaire and the “functional,” “physical,” and mean MDADI scores. Also in bivariate analysis, an association was observed between adjuvant chemotherapy and higher scores in the “physical” and “social/emotional” UWQoL and the “global,” “emotional,” “functional,” and mean MDADI scores. Following multivariate analysis, increasing age was associated with lower scores in both the “physical” (coefficient 20.59, P 5.001) and “social/ emotional” (coefficient 20.61, P 5.002) composite scores of the UWQoL as well as lower scores in the “functional” (coefficient 20.56, P 5.015), “physical” (coefficient 20.50, P 5.022), and mean MDADI scores (coefficient 20.54, P 5.014) (Table V). Larger tumors (higher T stage) were independently associated with lower scores in MDADI “global” (coefficient 216.02, P 5.037), “emotional,” (coefficient 216.66, P 5.004), and “functional” (coefficient 214.24, P 5.012) domains, whereas lip-split mandibulotomy and free flap reconstruction were independently associated with significantly lower “physical” MDADI scores (coefficient 219.08, P