Abstract Book

S1280

ESTRO 37

Established (LN229 and U87) as well as primary Glioblastoma cell lines (T76, H2, H5, H9) were used for invasion studies. Primary cells were generated from patient derived tumor tissue. Low-LET irradiation (~1.7 keV / µm) was performed using an x-ray irradiation device (RS225, Gulmay Medical, UK; 200kV) and high-LET alpha particle irradiation (LET=164 keV / µm) was performed using an Americium-241 source. The invasive potential of glioblastoma cells was examined 24h after irradiation by applying 8 µm pore sized transwell chambers (Corning TM ). Gene expression was measured using quantitative RT- PCR. Tumor cell adhesion potential, size and other morphological features were measured with the CanQ system (Venneos GmbH). Results GBM cell lines were tested for their invasion after x-ray and alpha irradiation. At a dose of 2 Gy x-ray irradiation results showed no changes in the invasion. However, 4 Gy x-ray irradiation led to a significant increase (p≤0.05) in the invasive property of all tested cell lines (LN229: 1.4 fold, U87: 1.6 fold, H2 1.9 fold, H5 1.6 fold and H9 1.2 fold). Likewise we could show that tumor size and adhesion is decreased after 4 Gy x-ray irradiation. Studies using high-LET irradiation with a dose of 1.3 Gy, which was determined as the equivalent dose to 4 Gy x- ray irradiation in the cell survival assay, did not show any influence on the invasion potential. qPCR analysis revealed up-regulations of several proteases (MMP-2, MMP3, CTSA, CSTD) and protease inhibitors (TIMP-1, TIMP-2, SerpinB8) as possible molecular reasons for the observed invasive increase. Conclusion Our results demonstrate that high-LET irradiation could be a better treatment option for GBM tumours and may lead to improved patient survival in future. EP-2321 Evolution of OLIG2 expression during radio(chemo)therapy has a prognostic value in glioblastoma. C. Bouchart 1 , A.L. Trépant 2 , M. Hein 3 , D. Van Gestel 1 , I. Salmon 4 , P. Demetter 2 1 Institut Jules Bordet- Université Libre de Bruxelles, Department of Radiation-Oncology, Brussels, Belgium 2 Erasme University Hospital- Université Libre de Bruxelles, Department of Pathology, Brussels, Belgium 3 Erasme University Hospital- Université Libre de Bruxelles, Department of Psychiatry and Sleep Laboratory, Brussels, Belgium 4 Erasme University Hospital- Université Libre de Bruxelles, Department of Pathology and DIAPATH- Center of Microscopy and Molecular Imaging CMMI, Brussels, Belgium Purpose or Objective Glioblastomas (GBM) are highly radio- and chemoresistant tumors associated with poor outcome. Oligodendrocyte lineage transcription factor 2 (OLIG2) is known to play a pivotal role in GBM particularly in proneural subtypes and for glioma stem cells (GSCs). The aim of our study is to better understand the evolution of OLIG2 expression during treatments and to investigate its clinical significance in GBM. Material and Methods Immunohistochemical expression analysis of OLIG2 was carried out on paired GBM samples of 72 patients, comparing initial resections with recurrent tumors after radiation therapy alone (RT) (n=36) or radio-

chemotherapy with temozolomide according to the Stupp regimen (RT-TMZ) (n=36). Semi-quantitative analysis of OLIG2 nuclear staining in tumor block was performed by two observers. Every tumor was scored according to the number of stained cells (no expression 0, low expression ≤ 30% and high expression > 30%). (Figure 1)

Results Compared to the primary tumor, OLIG2 expression was decreased in GBM recurrences after treatment with RT (44.00 ± 22.00 vs 32.00 ± 23.00%, p = 0.002) or RT-TMZ (32.00 ± 23.00 vs 29.00 ± 22.00%, p = 0.044). The decrease in OLIG2 did not differ significantly according to the treatment used (p = 0.606). Among the 44 patients (61%) with a high initial OLIG2 expression (> 30%), the results were similar for RT (57.00 ± 11.00 vs 40.00 ± 22.00%; p <0.001) and RT-TMZ (62.00 ± 13.00 vs 30.00 ± 21.00%; p <0.001). Compared to individuals with reduced expression (≤ 30% / n = 25) in post-treatment (RT or RT- TMZ), those with high conserved expression of OLIG2 (> 30% / n = 19) presented later recurrence (13.54 ± 9.94 vs 5.87 ± 2.44 months, p <0.001) and improved survival (19.57 ± 6.97 vs 14.41 ± 5.17 months, p = 0.014). The distribution between the two treatments was similar in the groups with high or reduced expression (RT 57.89 vs 56.00%, p = 0.900). (Table 1)

Conclusion Treatments globally reduce expression of OLIG2. Patients with conserved high expression of OLIG2 after RT or RT-