ESTRO 36 Abstract Book

S289 ESTRO 36 _______________________________________________________________________________________________

N.G. Burnet 2 1 Addenbrooke's Hospital - Oncology Centre University of Cambridge, Cambridge Cancer Trials Center, Cambridge, United Kingdom 2 Addenbrooke's Hospital - Oncology Centre University of Cambridge, Department of Oncology, Cambridge, United Kingdom Purpose or Objective Most patients undergoing radical radiotherapy (RT) for head and neck cancer (HNC) develop severe acute toxicity, and treating physicians aim to spare uninvolved neck nodal levels where possible. This study aimed to determine rates of gr. 2+ and 3+ acute oral toxicities in patients with HNC receiving helical IMRT with daily image guidance. We compared acute oral toxicities in patients who received unilateral nodal irradiation (UNI) and bilateral nodal irradiation (BNI). Material and Methods This was a sub-study of the Cancer Research UK VoxTox programme, that prospectively accrued high resolution acute and late toxicity data in patients undergoing radical RT. Selection criteria for this sub-study were: Prescription dose 65Gy/30#, concurrent weekly cisplatin (min. 4 cycles), sub-sites including oropharynx, oral cavity, larynx and hypopharynx. Patients were treated according to our local 3 dose-level protocol: macroscopic disease CTV 65Gy, high risk CTV 60Gy, lower risk CTV 54Gy (contralateral neck). Patients were reviewed at baseline, weekly during treatment, and at weeks 4 and 8 post- treatment. All data were collected using electronic clinical report forms by a specialist HNC oncologist, or experienced RTT. We collected data on 4 key oral toxicity endpoints using CTCAE v4.03: oral mucositis, dry mouth, salivary duct inflammation and dysphagia. Maximum toxicity rates of these four were also considered. Results 73 patients were selected; 11 had UNI, and 62 BNI. There was no reported toxicity or differences between groups at baseline. Plots of gr. 2+ and 3+ toxicity for all 4 endpoints are shown in Figure 1A-D. Maximum weekly toxicities are shown in Table 1. Overall, there appears little difference in gr. 2+ toxicity rates between groups. Furthermore, maximum or specific toxicity rates are similar (max toxicity 83% BNI vs 80% UNI). However, table 1 and figure 1 suggest that gr. 3+ toxicity developed more quickly in BNI patients. Using 2x2 tables and Fisher’s exact test, we compared rates at week 4. Overall maximum toxicity at week 4 was 33% for UNI, 56% for BNI, but this was not significant (p=0.33). This procedure was repeated for all 4 endpoints. Rates of mucositis were 8% (UNI) vs 33% (BNI); this trended towards significance (p=0.08). Week 4 gr. 3+ toxicity rates were also higher in the BNI group for dry mouth, ductal inflammation and dysphagia (0% vs 16.4%, 16.7% vs 23%, 33% vs 41% respectively). None of these differences showed statistical significance. Recovery rates post treatment appeared similar between cohorts. Conclusion This study shows that gr. 2+ toxicity rates appear to be similar between patients undergoing UNI or BNI throughout treatment and recovery. Peak toxicity (ie highest proportion of patients reporting gr. 3+ toxicity) was also similar. However, our data indicate that BNI patients may develop some acute toxicity earlier in treatment and we will re-test this hypothesis once a larger cohort has been recruited. OC-0544 Stereotactic radiotherapy in elderly patients: age, survival and performance status J.L. Monroy Anton 1 , L. Tejedor Pedrosa 2 , M. Soler Tortosa 1 , M. López Muñoz 1 , A. Soler Rodríguez 1 , A. Navarro Bergadá 1 , M. Estornell Gualde 1 1 Hospital universitario de la ribera, radiation oncology, Madrid, Spain

emerged after the first week of RT, patients who were not followed for at least 7 days were excluded from the analysis. The incidence of DOG for patients with at least one treated vertebra at Th8 or above was compared to the incidence for patients treated below Th8 using Fisher’s exact test. The relationships between the mean (D mean ) and maximum (D max ) esophageal doses and incidence of DOG was examined using the Wilcoxon rank sum test, comparing distributions of dose metrics in the two groups. Results 30 patients (12 women / 18 men) participated in the study, with prostate (8), breast (5) and lung (4) cancer as the most common primary diagnoses. Median number of vertebrae treated was 2 (range 1-9), with 9 patients treated to more than one site. 4 patients were excluded from this analysis due to withdrawal or inability to complete questionnaires prior to the one week mark. Out of the 26 patients, 11 reported DOG (Fig 1, Fig 2). For all 11 patients, the most cranially treated vertebra was Th8 or above; 4 of the 15 patients not reporting DOG were treated to Th8 or above (p<0.001). The median D mean was 7.8 Gy (range 0.0-15.9 Gy) for patients reporting DOG, and 2.2 Gy (0-10.8 Gy) for patients not reporting DOG, p=0.0043. Corresponding values for D max were 24.7 Gy (0- 31.4 Gy) with DOG and 9.8 Gy (0-31.4 Gy) without DOG, p=0.0043.

Conclusion There is a high incidence of patient-reported DOG in patients treated for SCC. This incidence correlates with mean and maximum esophageal dose, but also appears to depend strongly on irradiation of cranially located vertebras. Our results indicate that it may be possible to reduce the incidence of DOG in patients with SCC by reducing the esophageal dose. This provides an argument for the use of intensity-modulated radiotherapy optimized to limit dose to the esophagus for the treatment of SCC. OC-0543 Acute toxicity with helical IGIMRT for head and neck cancer: Unilateral vs bilateral nodal irradiation A.M. Bates 1 , D.J. Noble 2 , O. Young 1 , E. Wong 1 , J. Gemmill 2 , R.J. Benson 2 , S.J. Jefferies 2 , G.C. Barnett 2 ,