Abstract Book

S684

ESTRO 37

Results In areas with 60Gy overall radiation dose K rad , K ax , and MK demonstrated declining trend. In areas with 40Gy overall radiation dose, K rad showed significant decrease after both 16 and 33 fractions, MK demonstrated increasing trend bellow the significance trashold, K ax showed no significant changes. In areas with 20Gy overall delivered dose, significant changes of measured DKI parameters were not found. Conclusion DKI can detect complex early microstructure changes dynamics within irradiated white matter in areas with higher overall delivered doses. DKI can be a potential radiobiological biomarker for early irradiation injury of brain parenchym. Further study and examinations are necessary in order to evaluate potential benefit of DKI. EP-1234 Subventricular zone involvement determines the efficacy of salvage SRS for recurrent glioblastoma. M. Harat 1 , S. Dzierzęcki 2 , K. Dyttus 3 , M. Ząbek 4 1 The Franciszek Lukaszczyk Oncology Center, Radiotherapy, Bydgoszcz, Poland 2 Center of Postgraduate Medical Education, Neurosurgery, Warsaw, Poland 3 Centre of Oncology- Maria Sklodowska-Curie Memorial Institute-, Radiotherapy, Warsaw, Poland 4 Gamma Knife Center, Neurosurgery, Warsaw, Poland Purpose or Objective Radiosurgery (SRS) is a non-invasive modality allowing for one-day treatment of small volume recurrences. Due to lack of evidence whether this strategy may have a positive impact on wide spectrum of glioblastoma (GBM) types there are controversies regarding salvage SRS. Highly invasive nature of disease and possibility of edema occurrence after treatment results a major concerns in recurrent GBM. Recently different GBM types were described according to cortex (CTX) and subventricular zone (SVZ) invasion of primary tumor. Primary tumors that invade SVZ are known as radioresistant, associated with multifocal or distant progression and decreased survival. Our aim was to study the results of SRS in relation to SVZ invasion in recurrent setting. Material and Methods Between July 2012 and December 2016, 19 consecutive patients with 23 lesions were treated with Gamma Knife SRS as salvage treatment for recurrent glioblastoma. A neurosurgeon and a radiation oncologist were involved in treatment planning and target volume determinations for all 19 patients. All available MR images were reviewed according to SVZ invasion of primary tumor. Recurrent tumors in time of SRS planning were categorized as contacting or non- contacting CTX and/or SVZ. Any areas of new or progressive enhancement after SRS were also defined in relation to pre-SRS images as local, distant or multifocal progression, with or without SVZ invasion. The Kaplan– Meier method was used to estimate progression free survival, overall survival and survival from SRS. A complete resection was obtained in 13 patients (68.4%), a subtotal resection was performed in 4 patients (21%), and a biopsy was conducted in 2 patients (10.6%). All patients included in this study received a standard course of radiotherapy. 14 patients had subsequently received temozolomide. The mean time from the initial diagnosis to salvage radiosurgery was 15.6 months (range, 1– 66 months). The median tumor volume was 4.39 cm3 (range, 0.09– 12.25 cm3). The median marginal dose of 18 Gy (range, 15–20 Gy) was always given to the 50% isodose line. Results An actuarial survival analysis for the all patients revealed a median overall survival duration of 32.2 months after the initial diagnosis (95% CI 24.8–39.5 months) and 16.6 months after the date of SRS (95% CI 12.5–20.6 months).

The median survival time after salvage SRS was significantly higher in patient presenting no SVZ involvement at time of Gamma Knife irradiation comparing with patient showing SVZ or both SVZ and cortical infiltration (20.6 months vs 8 month vs 15 months, P=.019). Conclusion In conclusion our analysis supports that SRS Gamma Knife is a safe and effective salvage modality in selected patients with recurrent GBM. Additionally, obtained results suggest that involvement of SVZ could be used as independent poor prognostic factor thus targeting regions with SVZ involvement might serve as new promising strategy for prolonging overall survival in GBM. EP-1235 Clinicopathological and prognostic factors of recurrence for atypical and anaplastic meningiomas K. Ytuza Charahua de Kirsch 1 , M. Martín Sánchez 1 , H. Pian Arias 2 , E. Fernández Lizarbe 1 , R. Hernanz de Lucas 1 , S. Sancho García 1 1 hospital Universitario Ramon Y Cajal, Radiation Oncologist, Madrid, Spain 2 hospital Universitario Ramon Y Cajal, Pathologist, Madrid, Spain Purpose or Objective Atypical and anaplastic meningiomas are tumors with a tendency to grow fast and recurrence. After surgical treatment with complete resection recurred approximately 30-50%. The aim of this study was to evaluate clinicopathological and prognostic factors of recurrence for atypical and anaplastic meningiomas treated in our institution. Material and Methods We retrospectively reviewed 56 patients treated in our hospital between 1998 and 2016 diagnosed with Grade II and III meningiomas according to the 2016 WHO classification of tumors of central nervous systems. We analyzed the following prognostic factors of recurrence : age, gender, symptoms at diagnosis, location of tumor, extent of surgical resection (Simpson Grading System), necrosis, bone and brain invasion. Patients’ characteristics were described with mean and absolute frequencies. To univariate analyses we used the Chi- Square test. Statistical analyses were performed using SPSS 20 software. Results 54 patients were classified as atypical meningioma and 2 patients as anaplastic meningioma. The mean follow-up was 41 months. The mean age was 60 years (ranging: 14– 86). Of the 56 patients, 32 (57%) were male and 24 (43%) were female. The most frequent signs and symptoms at diagnosis were headache and seizures (50%). The most common tumor location was brain convexity in 19 patients (34 %). Simpson resection Grade I was performed in 20% of the patients, Grade II in 59%, Grade III in 4% and Grade IV in 14%. Bone invasion was present in 10 patients (18%), brain invasion in 11 (19%) and necrosis in 23 (41%). A total of 18 patients (32%) suffered a relapse. The median time of failure was 35 months. Of patients with Simpson resection grade I-II relapsed 33 % vs 44% with Simpson resection grade III-IV; with bone invasion relapsed 50% vs 28% without bone invasion; with brain invasion relapsed 36% vs 27% without brain invasion; and with necrosis 39 % vs 53% without necrosis.The treatment after failure was: surgery 83%, radiotherapy 50%, both treatments 50% and no treatments 11%. 6 relapsed patients suffered a progression disease during follow up. Univariate analysis did not show a statistically significant relationship between prognostic factors and brain recurrence (Simpson resection 95% CI 0.39-6.98 with p=0.53; bone invasion 95% CI 0.59-11.31 with p=0.195; brain invasion 95% CI 0.34-6.21 with p=0.61; and necrosis 95% CI 0.61-6.81 with p=0.24).