ESTRO 38 Abstract book

S307 ESTRO 38

evaluable for 31 patients. Overall survival (OS) and progression-free survival (PFS) of the four phenotypes were significantly different (5-year OS: PL 95.2%, PH 80%, NL 53.7%, NH 20.8%, p < 0.001; 5-year PFS: PL 79.6%, PH 50%, NL 48.7%, NH 7.4%, p < 0.001). We found a significant correlation of the p16-radiomic phenotype with density of CD8+ T cells (r = 0.39, p = 0.001), FoxP3+ T cells (r = 0.45, p < 0.001) and CD68+ cells (r = 0.40, p < 0.001), but not with PD-L1 expression. Specifically, all patients with high density of CD8+ T cell, high FoxP3 and high CD68+ macrophages simultaneously, which suggests a favorable immune activated state, exhibit as p16+/radio low phenotype (Figure 1). Legend: Figure 1. Heatmap of distribution of density of CD8+ T cells, FoxP3+ T cells and CD68+ cells according to the four p16-radiomic phenotype (N=74). The columns represent patients and the rows indicate the 3 immune biomarkers. In the bottom color bars each color represents a p16- radiomic phenotype.

Force for cancer treatments and cardiovascular toxicity of the European Society of Cardiology (ESC), European Heart Journal , Volume 37, Issue 36, 21 September 2016, Pages 2768 2801, https://doi.org/10.1093/eurheartj/ehw211 2. Hamo CE, Bloom MW et al, Cancer-therapy-related cardiac dysfunction and heart failure.Part 2 Prevention, treatment, guidelines and future directions, Circ Heart Fail , 2016 Volume 9, issue 2 https://doi10.1161/Circheartfailure.115.002843 3. Kapoor,A et al Monitoring risk factors of cardiovascular disease in cancer survivors, Clinical Medicine, 2017 Volume 17, no4 293 4. Faithfull, Kirby M et al. Mitigating risk of Cardiovascular disease Cancer Nursing Practice. 16, 1,18- 23. https://doi: 10.7748/cnp.2017.e1352 OC-0586 Immunological contexture basis of a prognostic radiomics signature in head and neck cancers D. Ou 1,2,3,4 , J. Adam 5 , I. Garberis 5,6 , P. Blanchard 2 , F. Nguyen 2 , A. Levy 2,3,4 , O. Casiraghi 5 , R.T.H. Leijenaar 7 , P. Gorphe 8 , I. Breuskin 8 , F. Janot 8 , C. Robert 2,3,4 , P. Lambin 7 , S. Temam 8 , J. Scoazec 3,5,6 , E. Deutsch 2,3,4 , Y. Tao 2,3,4 1 Fudan University Shanghai Cancer Center, Department of Radiation Oncology, Shanghai, China ; 2 Gustave Roussy Cancer Campus, Department of Radiation Oncology, Villejuif, France ; 3 Université Paris Sud, Université Paris Sud, Paris, France; 4 INSERM, INSERM1030 molecular radiotherapy, Villejuif, France ; 5 Gustave Roussy Cancer Campus, Department of Pathology, Villejuif, France ; 6 INSERM, INSERM US23/CNRS UMS3655- Molecular analysis- modelling and imaging of cancer disease- Experimental and Translational Pathology, Villejuif, France ; 7 Maastricht University Medical Centre+, The D- Lab: Decision Support for Precision Medicine- GROW - School for Oncology and Developmental Biology, Maastricht, The Netherlands ; 8 Gustave Roussy Cancer Campus, Department of Head and Neck Surgical and Medical Oncology, Villejuif, France Purpose or Objective While in recent years radiomics has been increasingly studied and often associated with clinical endpoints, the relationships of radiomics and tumor biology are largely unknown. In this study, we sought to explore the immunological contexture basis of a previously developed prognostic radiomic signature in head and neck cancers. Material and Methods Ninety-five patients were included in the analysis. Evaluation for density of CD8+ T cells, FoxP3+ T cells, CD68+ cells, PD-L1 expression and p16 expression was performed on pretreatment biopsy tissue samples with immunohistochemistry methods. A total of 544 radiomics features of the primary tumor were extracted from radiotherapy planning computed tomography scans. We categorized patients into four phenotypes based on p16 expression and a previously developed 24-feature based prognostic radiomic signature [1]: p16+/radio low (PL), p16+/radio high (PH) p16-/radio low (NL),p16-/radio high (NH). The correlations between the four phenotypes and biomarker expressions in the tumor micro-environment were analyzed using Spearman’s rank correlation test. Survival rates were calculated with the Kaplan–Meier method and compared using the log-rank test. Results The median follow-up was 50 months (range: 4–104). IHC results of CD68 were evaluable for all the 95 patients, CD8 and FoxP3 were evaluable for 74 patients, and PD-L1 was Proffered Papers: RB 6: Pre-clinical research in particle therapy

Conclusion We demonstrate that radiomic approaches permit noninvasive assessment of immunological characteristic of tumors in head and neck cancers. Further validation in external cohort is required. OC-0587 Preclinical studies of MRI guided BNCT at Torino and Pavia Universities N. Protti 1 , D. Alberti 2 , A. Toppino 3 , S. Bortolussi 4 , S. Altieri 4 , A. Deagostino 3 , S. Aime 5 , S. Geninatti-Crich 2 1 National Institute of Nuclear Physics INFN, Pavia Unit, Pavia, Italy ; 2 University of Torino, Department of Molecular Biotechnology and Health Sciences, Torino, Italy ; 3 University of Torino, Department of Chemistry, Torino, Italy ; 4 National Institute of Nuclear Physics INFN / University of Pavia, Pavia Unit / Department of Physics, Pavia, Italy ; 5 University of Torino / IBB-CNR, Department of Molecular Biotechnology and Health Sciences / MBC, Torino, Italy Purpose or Objective Boron Neutron Capture Therapy (BNCT) is a binary hadrontherapy performed on head and neck recurrent and primary cancers, skin melanomas and highly malignant brain tumours. BNCT is based on the capture reaction induced by low energy neutrons on 10 B selectively delivered by tumour-targeting drugs. A couple of high LET hadrons triggers cell death through the local energy deposition along densely ionizing tracks with ranges of 5- 9 μm, which destroy tumour cells without affecting adjacent healthy tissues. This makes BNCT a promising treatment for disseminated and infiltrating tumours that cannot be handled by surgery, conventional radiotherapy or heavy ion therapy, which require a precise localisation of the pathology.