ESTRO 38 Abstract book

S27 ESTRO 38

Purpose or Objective Radiotherapy for locally advanced non-small lung cancer (LA-NSCLC) causes oesophageal and pulmonary toxicity, which might be severe especially when concurrent chemotherapy is applied. A possible solution to decrease severe toxicities, is dose reduction to the involved lymph nodes, thereby reducing the dose to the mediastinal structures. The reported low incidence of regional (lymph node) failures compared to the primary tumour (6% versus 16%) supports this policy [1]. The goal of this study was to assess the safety and efficacy of a dose reduction to the involved lymph nodes in LA-NSCLC patients receiving An observational study was performed with a sequential design cohort including 328 patients with LA-NSCLC. Both cohorts received hypofractionated radiotherapy of 24 x 2.75 Gy to the primary tumour. The standard cohort included 191 patients treated between June 2013 and June 2015 with a 24 x 2.75 Gy (EQD 2 =70Gy (α/β=10)) dose scheme to the involved lymph nodes. The dose-reduction cohort (treated from June 2015 until June 2017) consisted of 137 patients who received a reduced dose of 24 x 2.42 Gy (EQD 2 =60Gy (α/β=10)) to the involved lymph nodes. The effect of dose reduction on toxicity and OS was assessed using independent samples t-test, chi-square tests, Logrank and cox regression analyses. Additionally, we assessed whether patient, tumour and treatment characteristics significantly influenced the association between dose reduction and outcomes. Results The median follow-up for the standard cohort and dose reduction cohort was 48 (2-85) and 29 (3-35) months, respectively. Patient and tumor characteristics were comparable between the 2 cohorts. Oes_V 50 (the volume of the oesophagus receiving ≥50 Gy), mean lung dose (MLD) and mean heart dose were significantly lower in the dose-reduction cohort. The incidence of regional failures was non-significantly lower in the dose-reduction group, 12% versus 7%, respectively (P=0.430). Furthermore, a significantly improved OS was observed in the dose- reduction group, 28 versus 35 months respectively (P=0.016) (Fig 1). Acute toxicity grade ≥2 , (dysphagia, pneumonitis, dyspnoea and cough, P=0.001) and late toxicity grade ≥2 (dysphagia, vertebrae fracture and cough, P=0.041) were significantly lower in the dose reduction cohort. MLD significantly influenced the association between dose reduction and OS. Nonetheless, after correction for MLD the independent association remained borderline significant (HR 0.733; 95% CI 0.526- 1.021; P= 0.067). Figure 1. Kaplan-Meijer curve for overall survival between the standard and dose reduction cohort. (chemo)radiotherapy. Material and Methods

Results Figure1 shows the final performance, the area under the receiver operating characteristic curve (AUC). To the best of our knowledge, this is the first trial investigating radiomic strategy to predict tumor shrinkage during RCT and our data suggests that a specific signature can be identified (AUC 0.82).

Conclusion The initial results of CREO Project obtained are an original and innovative topic that opens up new research in the field of personalized medicine in radiation therapy. The identification of the external validation dataset is actually ongoing. OC-0064 Reducing radiotherapy dose to involved lymph nodes in locally advanced NSCLC: efficacy and toxicity. J. Van Diessen 1 , M. Kwint 1 , J. Sonke 1 , I. Walraven 1 , B. Stam 1 , J. De Langen 2 , J. Belderbos 1 1 Netherlands Cancer Institute, Radiation Oncology, Amsterdam, The Netherlands; 2 Netherlands Cancer Institute, Thoracic Oncology, Amsterdam, The Netherlands