ESTRO 37 Abstract book

ESTRO 37

S596

Purpose or Objective Esophageal cancer has an increasing incidence and is rarely curable. When treated with radiotherapy, the anatomical location often results in toxicity to surrounding normal tissues such as mucosal tissue and lungs. Most esophageal cancers have hypoxic areas causing resistance to conventional anti-cancer therapies, making them susceptible for treatment with hypoxia- activated prodrugs (HAP). These agents are reduced in the absence of oxygen and form active cytotoxins that kill hypoxic cells. We investigated in vivo whether the HAP evofosfamide (TH-302) sensitizes esophageal carcinomas to radiotherapy. Additionally, normal tissue toxicity of this treatment combination was assessed in short-term (gut mucosa) and long-term (lung fibrosis) toxicity models. Material and Methods To assess therapeutic efficacy, esophageal squamous cell (OE21) or adeno (OE19) carcinoma cells were injected into the flank of immunocompromised mice and treated upon an average volume of 240 mm 3 with 5 consecutive daily injections (QD5) of TH-302 (50mg/kg) or vehicle and irradiated (sham or single dose 10Gy) at day 5. Tumor volumes were measured until four times the start volume was reached. For the normal mucosal short term tissue toxicity, a gut irradiation model was used, while a lung fibrosis model was applied to assess long term toxicity. Tumor or non-tumor bearing mice were injected with TH- 302 (50mg/kg) or vehicle (QD5) and subsequently irradiated using a precision image-guided small animal irradiator (XRAD225Cx, PXI). The abdominal area (sham, single dose 8 or 10 Gy) and the upper part of the right lung (sham, single dose 20 Gy) were irradiated with 40- mm or 5-mm square parallel-opposed fields respectively. Damage was assessed 72 hours later by histology and blood plasma citrullin levels (gut) and after 1 year by histology and non-invasive microCT imaging (lung). Results The combination treatment of TH-302 with irradiation resulted in a significant tumor growth delay in OE19 (P=0.02) and OE21 (P=0.03) carcinomas, compared to irradiation only with an enhancement ratio of 1.4 for both models. TH-302 monotherapy induced a slight increase in tumor growth delay in line with the histological evaluation of the hypoxic fraction using pimonidazole hypoxia marker. OE19 tumors appeared to be very radioresistant, which was confirmed by in vitro crystal violet cell viability assays. Irradiation resulted in a dose-dependent decrease of crypt survival (P<0.001), mucosal surface area (P<0.01) and citrullin levels (P<0.001) in both tumor and non-tumor bearing animals. TH-302 did not influence the radiation-induced short term toxicity. Results of the long-term toxicity will be presented during the meeting. Conclusion TH-302 in combination with radiotherapy induced significant growth delay without enhanced normal tissue toxicity. This indicates that the combination therapy is a promising approach to improve the therapeutic index for patients suffering from esophageal cancer, resulting in better tumor control and quality of life.

PO-1062 Can an open faced shell replace a closed face shell for cranial radiotherapy? Initial results. M. Brewer 1 , L. Welsh 1 , L. Corsini 1 , H. McNair 1 1 Royal Mardsen NHS Trust, Radiotherapy, London, United Kingdom Purpose or Objective Thermoplastic immobilisation shells in radiotherapy can present problems for patients who experience claustrophobia and related anxieties, resulting in session disruption and a poor patient experience in a significant number of cases (1). Closed-face shells (CFS) which cover the face are commonly used, however open-face shells (OFS) are emerging as an alternative for patients with claustrophobia-related anxieties (2). This study aims to compare the OFS with the CFS, and investigate if the interfraction reproducibility of the OFS is not significantly inferior to the CFS. Material and Methods Patients requiring virtually-simulated palliative cranial radiotherapy were randomised between the CFS and OFS. Treatment planning and delivery were carried out within standard hospital protocols with orthogonal electronic portal image pairs acquired on at least five fractions. Acquired images were compared to digitally reconstructed radiographs for both translation (lateral (x), longitudinal (y), and vertical (z)) and rotation (pitch and yaw). Mean population displacements for each parameter were compared using independent sample t- tests. Patient experience data and intrafraction stability were also measured but are not reported here due to ongoing recruitment. Results Following ethical approval, 22 patients were consented with 11 patients per shell type. The mean (range) population translational displacements and the mean (range) rotational displacements were similar (Table 1). There were no significant differences between groups.

Z (vertical ) mmMea n (range) -0.46(- 2.9 to +0.6) -0.670 (- 2.8 to +1.4)

X (lateral) mmMea n (range) -0.05(- 1.4 to +1.2) 0.06 (- 2.2 to +2.6)

Y (longitudinal ) mmMean (range)

YawMea n (range)

PitchMea n (range)

0.41° (- 0.9° to +2.4°) 0.34° (- 0.4° to +2.2°)

0.55° (- 0.7° to +1.7°)

CF S

0.63 (-2.4 to +1.8)

OF S 0.34° (- 0.2° to +0.9°) Table 1. Mean population translational and rotational displacements Conclusion Discussion and Conclusion The results suggest that the CFS and OFS have comparable reproducibility for both translation and rotation (pitch and yaw). The OFS has the potential to provide an alternative for patients with claustrophobia- related anxieties. As the study continues to recruit we will establish if this trend continues. 1. Clover K, Oultram S, Adams C, et al. Psycho-oncology; 20: 1334–1341. 2. Li G, Lovelock DM, Mechalakos J, et al. J Appl Clin Med Phys; 14: 243–254. -0.22 (-1.2 to +0.9)

Poster: RTT track: Patient preparation, positioning and immobilisation