ESTRO 2020 Abstract Book

S462 ESTRO 2020

The incidence of LMD at 1 year is 12% (4/34 cavities) with a neurological cause of death in 3/4 patients. 2 of 4 patients with LMD had large cell neuroendocrine lung cancer. 4/44 patients received repeat SRT and 10/44 received WBRT for multiple new BM/LMD. Median OS of all patients was 15 months (3-83 mths). There has been 1 case of histologically confirmed radionecrosis without viable melanoma cells after 4 years of immunotherapy. Conclusion Our cohort confirms the safety and efficacy of postop hfSRT with 5 x 6 Gy. The 12% incidence of LMD approximates postop SRS (7-28%) and surgery (16%) which suggests cavity irradiation has little impact on disseminated tumour cells. There was no apparent dose relationship. Preoperative SRS is compelling (delineation of intact metastasis, 20% dose reduction, resection of irradiated volume, sterilisation of disseminated cells) with promising reports of efficacy with minimal toxicity. Preoperative SRS will be evaluated in a prospective trial aiming to further improve outcomes after resection of brain metastases. PO-0858 Stereotactic Radiotherapy for Treatment of Brain Metastases: a monocentric experience A. Gonnelli 1 , S. Montrone 2 , A. Molinari 2 , F. Pasqualetti 2 , F. Paiar 2 1 Azienda Ospedaliera Universitaria Pisa, Radiation Oncology, Pisa, Italy ; 2 Azienda Ospedaliero Universitaria Pisana, Radiation Oncology, Pisa, Italy Purpose or Objective Brain metastases are a well-established cause of morbidity and mortality, affecting 20%-40% of patients (pts) with cancer. In this retrospective study, we investigate the local control and radiation-induced brain necrosis in pts with brain metastases who received single-fraction or multifraction (SRS). Material and Methods Between September 2010 and January 2019, a total of 103 pts(mean age 66 years, range 39-86)presenting with 154 metastases (provenance: 101 lung, 22 breast, 9 melanoma, 9 colonrectum, 5 kidney, 4 thyroid, 2 bile duct, 2 endometrium) were submitted to SRS at Department of Radiation Oncology in Pisa. All of them had KPS ≥70. For all pts, we calculated the Graded Prognostic Assessment (GPA), a prognostic index for pts with brain metastases.Gross tumor volume (GTV) was defined as macroscopic contrast enhancing lesion on T1-MRI. Planning tumor volume was obtained by adding to GTV an isotropic margin of 3 mm in all directions.The RT treatment was performed with True Beam LINAC VARIAN System with VMAT technique, using 6MV photons. We utilized a head thermoplastic mask as immobilization system.In this analysis we evaluated results in terms of local control and radiation-induced brain necrosis, trying to find a correlation with the delivered dose and the planning volume.In order to conform the delivered dose, we calculated the equivalent dose in 2 Gy (EQD2) considering an alfa-beta ratio of 5 (colonrectum), 4.6 (breast), 3 (lung and endometrium) and 1(cholangiocarcinoma and melanoma). Results Median overall survival (OS) was 13 months. The GPA index was significantly associated with OS (p = 0.014).Radiological response on MRI was assessed by a neuroradiologist according to the RANO-BM criteria. Local control was defined as the absence of new radiographic enhancing abnormality in the irradiated areas on MR imaging. After a mean follow-up of 9,8 months (range: 2,2- 79,2), 29 metastases (18%) were in progression. The local control (complete/partial response/stable vs progression disease) was statistically related to a value of EQD2 ≥100 (p=0,049).Nineteen pts (12.3%) showed a radionecrosis at MRI performed during follow-up. This reaction was not

statistically related to GTV-volume(p=0,350) but was strongly related to a value of EQD2 ≥150 (p=0,054). Conclusion Local control appears to be dose dependent with significantly better control observed in tumors receiving EQD2 ≥100. The risk of developing radionecrosis increases with EQD2 ≥150. PO-0859 Postoperative stereotactic brain radiation therapy: a single centre retrospective analysis S. Takanen 1 , T. Frederic-Moreau 1 , J. Otz 1 , M. Fawzi 1 , H. Aich 1 , H. Albert-Dufrois 1 , A. Beddock 1 , A. Clement-Zhao 1 , H. Mammar 1 , E. Musat 1 , M.H. Muresan 1 , G. Crehange 1 , M. Minsat 1 , P. Verrelle 1 , P. Poortmans 1 1 Institut Curie Ensemble Hospitalier, Department of Radiation Oncology, site Saint-Cloud- Paris, France Purpose or Objective To evaluate and benchmark the local outcome of our clinical practice for a group of breast and lung cancer patients with brain metastases treated by postoperative fractionated stereotactic radiation therapy (fSRT). Material and Methods We retrospectively evaluated 56 patients with brain metastases from lung (42%) and breast (58%) cancer treated from 2015 to 2018. Sixty-two brain surgical cavities were treated with fSRT performed by a linear accelerator NovalisTx ™ (Varian), using either 3 fractions of 7.7Gy/f (37.5%) or 3 fractions of 6 Gy/f (62.5%) at prescription isodose level (PIL) of 85%. The clinical target volume (CTV) included the surgical cavity with the eventual postoperative residual tissue identified by magnetic resonance imaging (MRI). The planning target volume (PTV) was defined as CTV+1 mm. Differences between the two groups in terms of local control (LC) and distant brain failure (DBF) were assessed using the Kaplan- Meier method and log-rank test (p value<0.05). A multivariate analysis was done using the Cox proportional hazard model for most important prognostic factors including age, tumour type, Karnofsky performance status (KPS), dose/fraction, tumour size at diagnosis, Her2 status, postoperative residual enhancing tissue at MRI, interval time between surgery and start of fSRT. Radionecrosis (RN) was assessed by perfusion MRI. Results Median age at diagnosis was 64 years (range 32-80 years). Median follow up was 16.5 months (range 1-48). One-year LC rate was 68,5% and 63,4% for the 7.7Gy and 6Gy group, respectively. One-year DBF rate was 47% and 41% for the 7.7Gy and 6Gy group, respectively. The Kaplan Meier analysis didn’t show any difference between the 2 groups of fractionation regimen for LC (p=0.9) and DBF (p=0.2). Cox regression analysis showed that significant factors were tumour dimension >3 cm (p=0.027) for LC and timing between surgery and start of fSRT (>30 days) for both LC (p=0.037) and DBF (p=0.001). Conclusion The one-year local failure rate might be explained by the low prescription schedules used. On multivariate analysis, tumour dimension (>3 cm) and a delay >30 days between surgery and start of fSRT have a significant impact on LC and DBF. A larger number of patients and a longer follow up are necessary to define the dose and PIL adequate to obtain a good LC in postoperative treatment of brain metastases with fSRT. PO-0860 Hippocampal dose reduction while treating brain gliomas using 3DCRT S. ElsharkawY 1 , A. Abo Gabal 1 , A. Hassouna 1 , M. Mokhtar 2 , M. Hassan 3 1 National Cancer Institute- Cairo University, Radiation Oncology, Cairo, Egypt ; 2 National Cancer Institute- Cairo University, Radiation Oncology Physics, Cairo, Egypt ;