Abstract Book

S256

ESTRO 37

was < 50%, the risk of all cause death was 14%, 29% and 42% at 1, 2 and 5 years respectively.

28 Bone RT

1 2

0 (0) 0 (0) 0 (0)

7 (28)

10 (40)

≥ 3

1 (4)

Respiratory disease 15 Lung RT

0

15 (100)

0 (0)

NA

Conclusion TKI is a recent opportunity to treat stage IV NSCLC with EGFR mutations or ALK translocation but scarce data are available on the effects of combined treatment with RT. In this study RT concomitant to TKI appears feasible and safe with satisfying OS. Ablative RT positively affects OS with low toxicities and, when possibile, should be recommended OC-0497 Dose-effect relation of heart dose and Overall Survival in stage I-III NSCLC patients B. Stam 1 , M. Rossi 1 , J. Belderbos 1 , J.J. Sonke 1 1 Netherlands Cancer Institute, Radiation Therapy, Amsterdam, The Netherlands Purpose or Objective Possible heart toxicity was previously determined from the association between heart dose and survival in early stage and in locally advanced stage non-small cell lung cancer (NSCLC) patients, treated with stereotactic radiation body therapy (SBRT) or chemoradiotherapy (CCRT). For the combination of these groups, we aim to determine a dose-effect relation between heart dose and 937 NSCLC patients were included in this study; 468 early stage NSCLC patients treated with SBRT (median 54 Gy in 3 fractions) between 2006-2013, and 469 locally advanced NSCLC patients treated with IMRT and daily low-dose hypofractionated CCRT (66 Gy in 24 fractions + daily cisplatin 6 mg/m2) between 2009-2014. Physical doses were converted to EQD2-doses using α/β ratio =3Gy. An average anatomy with the heart and its substructures (left and right ventricles, left and right atria, ascending aorta, left pulmonary artery, superior vena cava) contoured was employed to obtain consistent heart contours in all patients. For each (sub)structure, dosimetric parameters D V (V: 0-100%), V D (D: 0-max Gy), EUD n (n: 0.1-10) were obtained. Associations of these dosimetric parameters with OS were evaluated using univariate Cox regression. For each (sub)structure, the dosimetric parameter with the lowest Akaike information criterion (AIC) was used to build 3 dose-effect models; a Lyman-Kutcher-Burman (LKB) model at 1, 2 and 5 years. Using maximum-likelihood estimation parameters, median toxic dose (TD50) and steepness parameter m were optimized. We present the dose-effect models for the whole heart and the substructure with the lowest AIC. Results With a median OS of 28 months and a median follow-up (FU) of 32 months, 46% of patients were alive at median FU. Univariate analyses showed significant associations between various dosimetric parameters of all (sub)structures and all cause death. The low V parameters showed the lowest AICs for all substructures, lowest for the V 0.5Gy of the right atrium; hazard ratio (CI): 1.011 % -1 (1.008, 1.014). For the whole heart, the V 0.5Gy showed the lowest AIC; hazard ratio (CI): 1.014 % -1 (1.011- 1.017). The optimal LKB parameters to predict OS in this dataset are described in table 1, and shown for the whole heart_V 0.5Gy in figure 1. If the whole heart_V 0.5Gy was <50%, the risk of all cause death was 12%, 28% and 40% at 1, 2 and 5 years respectively. If the right atrium_V 0.5Gy overall survival (OS). Material and Methods

Conclusion For the combined stage I-III NSCLC patients, treated with SBRT or CCRT, the risk of all cause death was significantly correlated to heart dose, and was 40% at 5 years for patients with a whole heart_V 0.5Gy of 50% or a right atrium_V 0.5Gy of 44%. OC-0498 Randomized phase III trial of PCI with or without hippocampal avoidance for SCLC. Goecpseor- gicor N. Rodriguez de dios 1 , F. Couñago 2 , J. López 3 , P. Calvo 4 , M. Murcia 5 , M. Rico 6 , C. Vallejo 7 , J. Luna 8 , C. Cigarral 9 , I. Trueba 10 , N. Farre 11 , R. Manero 12 , X. Duran 13 , P. Samper 14 1 hospital del mar, radiation oncology, barcelona, spain 2 hospital quirón madrid, radiation oncology, madrid, spain 3 hospital virgen del rocío, radiation oncology, sevilla, spain 4 hospital universitario santiago de compostela, radiation oncology, santiago de compostela, spain 5 hospital sant joan de reus, radiation oncology, reus, spain 6 complejo hospitalario navarra, radiation oncology, pamplona, spain 7 hospital universitario ramón y cajal, radiation oncology, madrid, spain 8 hospital fundación jiménez díaz, radiation oncology, madrid, spain 9 complejo hospitalario salamanca, radiation oncology, salamanca, spain 10 hospital txagorritxu, radiation oncology, vitoria, spain 11 hospital santa creu i sant pau, radiation oncology, barcelona, spain 12 hospital del mar, neuropsychology, barcelona, spain 13 hospital del mar medical research institute, statistics, barcelona, spain 14 hospital rey juan carlos, radiation oncology, móstoles, spain Purpose or Objective Clinical evidence suggests that radiation dose received by the hippocampus during whole brain radiotherapy may play a role in radiation-induced neurocognitive decline. To prospectively evaluate the neurocognitive (NC) benefit of hippocampal sparing (PCI-HA), we have developed a phase III clinical trial (PREMER) to test hippocampal sparing during PCI. Here in, we report our experience with the first patients included in this trial. Material and Methods 60 patients undergoing PCI were randomized to receive PCI (n=30) or PCI-HA (n=30). The hippocampus was contoured, and hippocampal avoidance regions were created using a 5-mm volumetric expansion around the hippocampus. Linear accelerator –based intensity-