Margriet Kwint

Safety and efficacy of reduced dose and margins 27 2 Introduction Concurrent chemoradiotherapy (cCRT) for locally advanced non-small cell lung cancer (LA-NSCLC) results in a 5-year overall survival (OS) of 32% [1, 2]. Local and regional failures as well as severe acute and late toxicities adversely affect OS [1, 3, 4]. Determining the balance between optimal treatment outcomes and low toxicity rates is challenging. The risk of severe pulmonary, oesophageal and cardiac toxicity is mainly determined by the involvement of mediastinal lymph nodes, the size and location of the primary tumour and the total radiation dose [3, 5-7]. Dose-limiting toxicities include radiation pneumonitis, associated with mean lung dose (MLD), and severe acute and late dysphagia (grade 3 and higher), associated with the volume of the oesophagus receiving >50-60 Gy [8, 9]. Moreover, the RTOG-0617 trial demonstrated an association of heart dose and OS [1]. Similar findings were found in various other cohorts [10, 11]. The reported incidence of regional failures (RF) after radiotherapy for LA-NSCLC was generally lower than local failures (LF), around 10% versus 30% after 2 years [12-14]. We reported on prognostic factors predicting LF and RF after cCRT in detail, revealing that volume was the only significant factor [12]. Since involved mediastinal lymph nodes have a smaller volume compared to the primary tumour in the majority of patients, we hypothesized that the dose needed to control lymph node metastases might be lower than the dose needed to control the primary tumour. A consequence of a lower dose to the mediastinum might also induce an efficient reduction of the pulmonary, oesophageal and cardiac toxicity rates. Additionally, further decrease of the toxicity rates might be obtained by a margin reduction. Previously, Schaake et al. demonstrated that the planning target volume (PTV) margins for the tumour and the lymph nodes might be reduced due to a daily online carina based correction strategy [15]. Since June 2015, patients with LA-NSCLC were treated in our institute to a lower radiotherapy dose to the involved mediastinal lymph nodes of 58 Gy (24x2.42 Gy; EQD2 10 =60 Gy), while the primary tumour was treated with 66 Gy (24x2.75 Gy; EQD2 10 =70 Gy) by using a simultaneous integrated boost technique. Simultaneously, the planning margins for both the primary tumour and the lymph nodes were reduced. The aim of this study is to investigate the effects of this reduction of dose to the involved lymph nodes and PTV-margins on the incidence of toxicities and outcomes.