Abstract Book

S292

ESTRO 37

degree of inter-patient and intra-tumour heterogeneity and there is a great need for predictive biomarkers, that can be used for individualized treatment protocols, or PRO (precision radiation oncology). Data from some of the previous trials on hypoxic modification have been used retrospectively to develop and validate such predictive biomarkers for future trials. One class of biomarkers is based on gene expression signatures that measure various genes upregulated by hypoxia in pre-treatment biopsies. Different signatures have been applied successfully to previous trials in head and neck cancer on the radiosensitizer nimorazole, DAHANCA 5 [1] and IAEA-HypoX [2], and a trial on accelerated radiotherapy with carbogen and nicotinamide, ARCON [3]. A similar approach has been successful in a bladder cancer trial on radiotherapy, carbogen, and nicotinamide, BCON [4] . Other classes of biomarkers successfully applied retrospectively to clinical trials include the circulating protein osteopontin, DAHANCA 5 [5]. In the ARCON trial, epidermal growth factor receptor (EGFR) expression [6], the pattern of carbonic anhydrase 9 (CAIX) expression [7], and expression of the proliferation marker Ki-67 [8] have been successfully applied. In a head and neck cancer trial on the hypoxic cytotoxin tirapazamine (TROG 02.02), combinations of circulating levels of interleukin 8 (IL-8) and hepatocyte growth factor (HGF) [9], and expression of the flavoprotein P450 (cytochrome) oxidoreductase (POR) [10] have been successfully applied. In the BCON trial, presence of necrosis has also been applied successfully [11]. Recently, a number of clinical trials on hypoxic modifications have been initiated which integrate genomic or imaging biomarkers. The NIMRAD study (ClinicalTrials.gov ID: NCT01950689) is placebo-controlled randomized trial of nimorazole vs radiotherapy alone in locally advanced head and neck cancer not suitable for chemotherapy or cetuximab, and will retrospectively analyse for two gene expression signatures [1,3]. In two other head and neck cancer trials on nimorazole, a gene expression signature is being measured prospectively during the trials. The randomized multicentre study EORTC-1219-ROG-HNCG/DAHANCA-29 (NCT01880359) has two primary aims, to test chemo-radiotherapy with or without nimorazole, and to test a 15-gene signature. In this trial, the signature is evaluated before randomization and included as parameter for stratification. DAHANCA 30 (NCT02661152) is a randomized non-inferiority trial testing chemo-radiotherapy with or without nimorazole in patients identified as less hypoxic by the 15-gene signature. Finally, DAHANCA 33 (NCT02976051) is a trial on image guided dose-escalated radiotherapy to patients with hypoxic tumours, based on FAZA-PET. [1] Toustrup. Cancer Res 2011;71:5923–31. [2] Metwally. Radiother Oncol 2015;116:15–20. [3] Eustace. Clin Cancer Res 2013;19:4879–88. [4] Yang. Clin Cancer Res 2017;23:4761–8. [5] Overgaard. Lancet Oncol 2005;6:757–64. [6] Nijkamp. Eur J Cancer 2013;49:3202–9. [7] Rademakers. Radiother Oncol 2013;108:517–2 [8] Rademakers. Head Neck 2015;37:171–6 [9] Le. Clin Cancer Res 2012;18:1798–807. [10] Hunter. Cancer Res 2015;75:4211–23. [11] Eustace. Radiother Oncol 2013;108:40–7.

Debate: This house believes that treatment intensification in stage III is a dream rather than a reality?

SP-0551 For the motion C. Faivre-Finn Institute of Cancer Sciences, Manchester, the United Kingdom

Abstract not received

SP-0552 Against the motion: Treatment intensification in stage III NSCLC is a dream rather than a reality? J. Belderbos 1 1 Netherlands Cancer Institute, Radiation Oncology, Amsterdam, The Netherlands Abstract text While new systemic treatment options for stage III non- small-cell lung cancer (NSCLC) patients are being developed, locoregional control remains essential. PACIFIC is the first phase III trial [1] that tested an immune checkpoint inhibitor (durvalumab) as consolidation therapy in patients with stage III NSCLC who had not progressed following concurrent chemoradiation (CCRT). Durvalumab decreased the probability of disease progression with 48%. With this improved systemic control probability in NSCLC patients, the locoregional control rates will even be more important, and treatment intensification is crucial. Efforts to improve control of locally-advanced NSCLC by escalating radiation dose while increasing the overall treatment time (OTT) in unselected patients however, have been unsuccessful. Several new techniques have recently been developed to either select patients and/or tumor sub-volumes that might benefit from dose escalation. Treatment techniques to deliver IMRT are widely available nowadays, and IMRT has been proven to reduce the chance to develop normal tissue toxicity (heart, esophagus) and alow dose escalation. Many studies have identified the tumor volume to be a significant predictor for local control. The primary tumor is often more voluminous than the involved lymph nodes and IMRT techniques help us to deliver inhomogene dose distributions.The recently reported TROG 09.02 trial CHISEL [2] showed improved freedom from local failure for SBRT versus conventional fractionation in patients with inoperable peripheral stage I NSCLC. SBRT (54 Gy in 3, or 48 Gy in 4 fractions) resulted in superior freedom from local failure and was associated with an improvement in overall survival compared to fractionated RT (66 Gy in 33 fractions or 50 Gy in 20 fractions). Several studies reported feasibility and tolerability of sequential SBRT after CCRT in locally advanced NSCLC. The smaller margins generally used in SBRT may even decrease toxicity. Another strategy is boosting a biological defined tumor sub-volume on the pre- treatment FDG-PET scan. This has been pursued in the PET-Boost trial which is currently being analyzed. Defining a tumor sub-volume during treatment, using FDG-PET scan was subject of several trials. All these designs have in common that they deliver individualized IMRT and escalate the dose while the OTT is not prolonged. Another strategy is to improve patient selection. Using circulating serum -microRNA signature and clinical factors, a dose response score has been reported that identified a subset of patients with an overall survival benefit from high-dose radiotherapy [3]. A mathematical model used to analyze 23 early stage patient cohorts[4] concluded that hypofractionation regimens overcome hypoxia (and cell-cycle radiosensitivity variations) by the pure impact of high