Proceedings of the 52nd Annual ASTRO Meeting S453

Author Disclosure: J.O. Thariat, None; L. Ramus, None; P. Maingon, None; V. Gregoire, None; S. Marcie, None; G. Malandain,None.

2558 Salivary Gland Tumors: Retrospective Review of the BCCA Experience from 1999 to 2008

R. Carlson J. Wu

British Columbia Cancer Agency, Vancouver, BC, Canada

Purpose/Objective(s): To undertake an exploratory review of the BCCA experience with malignant salivary gland tumors (SGTs)diagnosed between 1999 and 2008.

Materials/Methods: The Cancer Agency Information System (CAIS) was queried for malignant SGTs of the major salivaryglands diagnosed between Jan. 1, 1999 and Dec. 31, 2008. A retrospective chart review of all cases referred to the BCCA wasalso performed. The data that was collected included histological subtype, stage at diagnosis, surgery type, pathology (grade,LVI, perineural invasion and margin status) and radiation treatment.

Results: A total of 372 cases of SGTs were diagnosed in BC from Jan.1, 1999 to Dec. 31, 2008. There were a total of 27 differenthistological subtypes coded in CAIS. The three most common subtypes were acinic cell carcinoma (21.0%), mucoepidermoidcarcinoma (15.5%) and adenoid cystic carcinoma (11.0%). 325 (87%) cases were referred to a cancer centre within BC. Of thereferred cases, over 80% of patients had surgery, 70% received radiation (55% adjuvant and 15% were radiation alone) andonly 5% received immediate chemotherapy. 65% of the cases were male and the average age was 60 yrs. The distribution of casesby stage was 33% with stage I, 23% stage II, 13% stage III and 27% stage IV. A preliminary analysis at a median follow-up of 30months showed a median survival for all patients referred to be 75 (± 5.05 95% CI) months. Overall survival at 5 years was 60%.For each stage it was 91% for stage I, 97% for stage II, 52% for stage III and 26% for stage IV.

Conclusions: The main subtypes of malignant SGT in our series were acinic cell carcinoma, mucoepidermoid carcinoma andadenoid cystic. The main treatment is surgery with or without adjuvant radiation. Survival for patients with early stage diseaseis excellent and this decreases with increasing stage. Further analysis to look at the effect of primary treatment, histological subtype,grade, margin status and other factors is ongoing.

Author Disclosure: R. Carlson, None; J. Wu, None.

2559 Factors Associated with Distant Metastasis in Squamous Cell Carcinoma of the Head and Neck Treated

with Definitive Radiation Therapy: The UAB Experience

A. F. Dragovic1, J. J. Caudell2, J. A. Bonner1

1University of Alabama Medical Center, Birmingham, AL, 2University of Mississippi Medical Center, Jackson, MS

Purpose/Objective(s): We report on our experience and identify risk factors for distant metastasis (DM) in a large series of patientstreated with definitive radiation therapy for squamous cell carcinoma of the head and neck (SCCHN).

Materials/Methods: The records of patients treated at the University of Alabama at Birmingham between 1995 and 2007 whoreceived definitive radiotherapy (RT) with or without concurrent systemic therapy for SCCHN were reviewed. After excludingnasopharyngeal, sino-nasal, and unknown primary carcinomas, patients \18 years of age, or those with prior history of cancer,560 patients were available for analysis. Development of distant metastasis was recorded, and distant metastasis-free survival(DMFS) was calculated using the Kaplan-Meier method. Clinicopathologic, demographic, and treatment factors were assessedfor association with DMFS via univariate analysis. Potential prognostic factors from the univariate analysis were then furtheranalyzed in a Cox regression multivariate model.

Results: With a median follow-up of 26 months, 56 patients (10.0%) developed distant metastasis. Two-year actuarial DMFS was89%. Factors significantly associated with lower DMFS on univariate analysis included hypopharyngeal primary site (p = 0.03),higher T-stage (p = 0.006), higher N-stage (p \ 0.001), and locoregional recurrence (LRR; p \ 0.001). On multivariate analysis,only N-stage and LRR remained significant predictors (p\0.001 for each). No subgroup identified had a higher rate of DM thanLRR. The majority of initial metastases were to lung, while other sites (bone, skin, liver, brain, gastrostomy site) accounted for\25% of cases. Median survival after diagnosis of metastasis was 5 months (range: 0 - 86 months).

Conclusions: This large, retrospective series highlights the importance of locoregional control in DMFS in SCCHN. Efforts toimprove local therapy should thus be emphasized. Lung is the dominant site of metastasis, but other sites may warrant clinicalattention during follow-up. Survival after metastasis is generally poor.

Author Disclosure: A.F. Dragovic, None; J.J. Caudell, None; J.A. Bonner, None.

2560 Risk of Cervical Esophageal Stricture with Intensity Modulated vs. Conventional Radiation Therapy for

Head and Neck Cancer: A Systemic Review

T. Goldsmith, J. Wang, A. Holman, A. W. Chan

Massachusetts General Hospital, Boston, MA

Purpose/Objective(s): Intensity-modulated radiation therapy (IMRT) has become an established treatment for head and neck can-cer. Concerns remain regarding late toxicity, particularly treatment-induced cervical esophageal strictures with IMRT. The purposeof our study was to review the risk of esophageal stricture in patients with head and neck cancer after treatment with IMRT orconventional radiation therapy (CRT) and to compare the reported esophageal stricture incidence of prospective and retrospectivestudies.

Materials/Methods: An electronic PubMed search of cervical esophageal strictures after head and neck cancer radiation therapywas performed. The search was restricted to human clinical studies published in English language. Full text papers were examined

S454 I. J. Radiation Oncology d Biology d Physics Volume 78, Number 3, Supplement, 2010

and included if they met the following criteria: clinical studies published in peer reviewed journals and documentation of theabsence or presence of esophageal stricture with barium swallow or endoscopy after IMRT or CRT for head and neck cancers.Studies on strictures after re-irradiation or radiation for cervical or thoracic esophageal cancers were excluded.

Results: A total of 26 studies (10 IMRT, 16 CRT) were eligible for analysis. A total of 4,690 patients (611 IMRT, 4079 CRT)treated between 1989 and 2008 were available for analysis. Only 1 of the 10 IMRT studies (10%) was a prospective study while5 of the 16 CRT studies (31%) were prospective. There was a three-fold increased risk of reported esophageal stricture in patientstreated with IMRT compared to CRT (odds ratio [OR] 3.0, 95% confidence interval [CI] 2.4 - 3.8, p\0.001). The rates of esoph-ageal stricture after IMRT and CRT were 17.3% (95% CI 14.5% - 20.5%) and 6.5% (95% CI 5.8% - 7.3%), respectively. The risk ofreported stricture among the prospective studies was about 3 times higher compared to the retrospective studies (OR 2.9, 95% CI2.0 - 4.3, p\0.001). The rates of stricture among patients in the prospective and retrospective studies were 19.2% (95% CI 13.8% -25.6%) and 7.5% (95% CI 6.8% - 8.3%), respectively. The role of IMRT technique, chemotherapy, surgery, and the site of primarytumor in the risk of stricture could not be assessed due to insufficient data.

Conclusions: Our current study suggests that IMRT is associated with increased risk of cervical esophageal stricture compared toCRT. Prospective study design is associated with higher reported incidence of cervical esophageal stricture compared to retrospec-tive study design. Prospective assessment of swallowing function before and after radiation therapy with videofluoroscopic swal-low study is necessary to determine the long-term toxicity after treatment for head and neck cancer.

Author Disclosure: T. Goldsmith, None; J. Wang, None; A. Holman, None; A.W. Chan, None.

2561 Psychological Aspects and Detailed Quality of Life Evaluation in Head and Neck (HN) Cancer Patients

Treated with Definitive Accelerated Radiotherapy

A. Heyda K. Skladowski

Maria Skłodowska-Curie Memorial Cancer Center and Institute of Oncology, Gliwice Branch, Radiation Oncology Department,Gliwice, Poland

Purpose/Objective(s): Evaluation of psychological factors and detailed quality of life in relapse-free HN cancer patients treatedwith definitive accelerated radiotherapy (DAR).

Materials/Methods: Seventy-six relapse-free patients (23 women and 53 men) treated in the past (1995-2007) with DAR havecompleted modified Hospital Anxiety and Depression Scale (HADS-M with additional aggression subscale), quality of life tests- QLQ C-30 with H&N35 module, Satisfaction with Life Scale (SWLS), revised Life Orientation Test (LOTR), Visual-AnalogScales (VAS) of pain and discomfort in HN area.

Results: Depression and anxiety were low in overall (6 and 7 points from 24, respectively). Patients reported an average level ofaggression, optimism and satisfaction with life. Intensity of pain and discomfort in HN area was generally rather low - 3.4 and 3.3points from 10, respectively, however 50% of patients used painkillers (10% of pts reported very strong pain) and only 13% didn’treported any pain. Functional scales of QLQ C-30 showed that patients are doing quite well: global health status (QL2) was 58/100,physical functioning (PF2) 62/100, role functioning (RF2) 85/100, emotional functioning (EF2) 62/100, cognitive functioning(CF2) 73/100 and social functioning (SF) 79/100. Mouth dryness (HNDR 65/100), sticky saliva (HNSS 61/100) and painkillersusage (HNPK 50/100) were the most severe symptoms. In inter-sex analysis women had significantly lower level of aggression,lower pain frequency of HN area, less swallowing problems, less problems with eating in public, less problems with sticky salivaand better appetite. The subgroup of 24 patients with the middle length of follow-up (i.e. 5-8 years post DAR) had significantlyhigher results in depression, lower global health status and more frequently reported weight loss compared to other two subgroups(i.e. 14pts with 9-13 years and 39pts with 1-4 years of follow-up).

Conclusions: The relapse-free HN cancer patients treated by DAR have functioned quite well in their lives although almosteveryone has experienced some kind a pain in HN area. It looks women have functioned much better emotionally and had lesssymptoms directly related to late radiation toxicity than men. Patients treated 5 to 8 years from the survey timeline had significantlyhigher depression level, lower global health status and more frequently reported weight loss. It suggests that after 5 years of follow-up patients might be emotionally and physically exhausted because of chronic distress caused by post radiation effects andpossibility of cancer relapse.

Author Disclosure: A. Heyda, None; K. Skladowski, None.

2562 Positive Postradiotherapy Planned Neck Dissection is Strongly Associated with Increased Distant

Metastasis Rather Than Regional Relapse

S. Huang, D. Goldstein, I. Weinreb, B. Perez-Ordonez, S. Fung, J. Irish, J. Waldron, J. Kim, B. Cummings, B. O’Sullivan

Princess Margaret Hospital, Toronto, ON, Canada

Purpose/Objective(s): To evaluate the clinicopathological predictors for planned neck dissection (PND) positivity and outcomefollowing definitive radiotherapy (RT) ± concurrent chemotherapy (CRT) in head and neck cancer (HNC) patients.

Materials/Methods: A retrospective review was conducted of all newly diagnosed patients with squamous cell carcinomas of oro-pharynx, hypopharynx, larynx, and unknown primary that underwent a PND following RT/CRT between 1998 and 2007. PND wasdefined as neck dissection (ND) performed within 15 weeks after RT/CRT without clinical evidence of persistent primary disease.All positive (PND+) specimens were confirmed by pathology review. Univariate and multivariate analyses were performed forpredictors of PND positivity and outcomes including 5-year overall survival (OS), cause specific survival (CSS), disease-free sur-vival (DFS), local control (LC), regional control (RC) and distant metastasis rate (DM). In the PND+ cohort, we further investigatedthe predictive value of positive nodal resection margin (Margin+), level 4/5 nodal involvement (LN4/5+), extracapsular extension/soft-tissue deposit (ECE/STD) and lympho-vascular involvement (LVI) in the neck specimen for OS, CSS, and DFS. All time-to-event outcomes were calculated from the date of PND.

Results: A total of 194 PNDs (50 PND+ and 144 PND-) were performed in 1930 consecutive HNC patients. Median follow-up was4.8 (0.2 - 10.7) years. Primary tumor sites were 139 (72%) oropharynx, 19 (10%) hypopharynx, 10 (5%) larynx, and 26 (13%)