Abstract Book

S251

ESTRO 37

Material and Methods The methodical approach was to confirm retrospectively a clinical relevance of CD44v6 followed by an in silico fine tuning to identify transcript variants. Subsequently, in vitro analyses of the radiation response and CD44v6 in established HNSCC cell lines were performed in order to characterize the underlying mechanisms and to identify druggable targets. siRNA interference technique and assays to analyze cell death and DNA damage response were used to elucidate the role of CD44v6 in radiation- induced cell inactivation. Finally, an orthotopic mouse xenograft model was established to validate the findings red in 96 surgical specimen of the LMU-KKG cohort, a database that includes tumor samples and clinical information of HNSCC tumor patients undergoing adjuvant radiochemotherapy. Elevated expression levels were associated with impaired overall survival. The results could be validated using a publicly available dataset of HNSCC tumors from TCGA. TCGA data allowed identification of the transcript variant CD44v3-10 to be mainly responsible for poor outcome in patients. In vitro there was significant positive correlation between clonogenic survival at 2Gy, and CD44v6 gene and protein expression in a panel of 7 HNSCC cell lines. Functionally, RNAi silencing of CD44v6 expression sensitized cells to irradiation. This was mainly due to an impaired proliferation rate and an increase in irradiation-induced senescence. Moreover, gene expression of factors of the senescence-associated secretory phenotype was altered in CD44v6 depleted compared to control cells. In an orthotopic xenograft model, CD44v6 silenced and non- silenced cells were inoculated into the mylohyoideus muscle of nude mice. Upon tumor engraftment, mice were treated with 2x 6 Gy or left untreated, using a CT- based small animal irradiation platform. First results indicate more pronounced tumor growth delay in CD44v6 silenced tumors, and thus a better therapeutic response towards radiation therapy. Conclusion Our study identifies CD44v6 as crucial driver of therapy resistance in a subset of HNSCC. We are currently investigating the underlying mechanisms in depth in order to evaluate CD44v6’s potential as prognostic marker and therapeutic target in personalized radiotherapy of HNSCC in greater detail. OC-0489 TAM and HLA class I expression in relation to HPV and clinical outcome in head and neck cancer D. Ou 1,2,3,4,5 , J. Adam 6 , I. Garberis 6,7 , P. Blanchard 1 , F. Nguyen 1 , A. Levy 1,3 , O. Casiraghi 6 , P. Gorphe 8 , I. Breuskin 8 , F. Janot 8 , S. Temam 8 , J. Scoazec 2,6,7 , E. Deutsch 1,2,3 , Y. Tao 1,2,3 1 Gustave Roussy Cancer Campus, Department of Radiation Oncology, Villejuif, France 2 Université Paris Sud, Faculté de médecine du Kremlin- Bicetre, Le Kremlin-Bicetre, France 3 INSERM1030, molecular radiotherapy, Villejuif, France 4 Shanghai Medical College- Fudan University, Department of Oncology, Shanghai, China 5 Fudan University Shanghai Cancer Center, Department of Radiation Oncology, Shanghai, China 6 Gustave Roussy Cancer Campus, Department of Pathology, Villejuif, France 7 INSERM US23/CNRS UMS3655, Molecular analysis- modelling and imaging of cancer disease- Experimental and Translational Pathology, Villejuif, France 8 Gustave Roussy Cancer Campus, Department of Head and Neck Oncology, Villejuif, France Purpose or Objective To investigate the prognostic value of tumor associated macrophages (TAM) and HLA class I expression in patients in vivo. Results CD44v6 mRNA expression levels were measu

other two signatures do also show increased HR values (albeit not significant) further supporting the role of EMT in HNSCC prognosis. Remarkably, these signatures only reached consensus for 55 out of 95 samples. Sub-analyses show that the prognostic value for the signatures was reduced in these remaining discordantly classified samples. In an attempt to optimize the EMT signatures for HNSCC, we therefore used the consensus samples to train a new classifier and tested it on the subgroup of discordantly classified patients. We show that this classifier was prognostic for distant metastasis-free survival in these samples (HR 10.6 ; p=0.004).

Fig 1. The prognostic value of expression based EMT signatures in our HPV negative HNSCC cohort (n=96) . Signatures are indicated by the first author of their respective publications.

*: p= <0.05 **: p= <0.01 Conclusion

HPV negative HNSCC patients with tumors that score high for EMT features showed increased metastasis rates, independent of known clinical risk factors. This suggests that biomarkers for EMT would be a valuable addition to estimate risks of metastasis in the clinic. OC-0488 Prognostic biomarkers and targets for personalization of radiotherapy of HNSCC: CD44v6 U. Schötz 1 , M. Orth 2 , M. Selmansberger 3 , J. Schuster 2 , B. Stegen 2 , J. Hess 3,4 , K. Unger 3 , H. Zitzelsberger 3,4 , C. Belka 2,4 , R. Engenhart-Cabillic 1 , K. Lauber 2,4 1 Phillips-University Marburg, Department of Radiotherapy and Radiooncology, Marburg, Germany 2 Ludwig-Maximilians-University, Department of Radiotherapy and Radiation Oncology, Munich, Germany 3 Helmholtz Zentrum München – German Research Center for Environmental Health GmbH, Research Unit of Radiation Cytogenetics, Neuherberg-Munich, Germany 4 Helmholtz Zentrum München- German Research Center for Environmental Health GmbH, Clinical Cooperation Group 'Personalized Radiotherapy in Head and Neck Cancer', Munich, Germany Purpose or Objective Standard treatment for locally advanced head-and-neck squamous cell carcinoma (HNSCC) is radiochemotherapy combined with surgery. Despite intense treatment plans with radiation doses above 70 Gy, 5-year overall survival rates are still limited to approximately 50%, and therapeutic failure is mainly attributed to therapy resistance. Apart from HPV status there are no suitable markers available to predict treatment outcome. Aim of this study was to identify in a preclinical setting a prognostic marker and potential therapeutic target. A promising group of candidates are transcript variants of the surface receptor CD44 containing variant exon v6 (CD44v6).