Abstract Book

S178

ESTRO 37

SP-0335 Clinical data informing dose/dose per fraction and scheduling strategies D. De Ruysscher 1 1 MAASTRO Clinic, Radiation Oncology, Maastricht, The Netherlands Abstract text It is well-known that radiotherapy modulates the immune system in such a way that together with immune therapy it may provoke a local and systemic anti-cancer immune response. As there is substantial heterogeneity between patients, tumours, the micro-environment and intra- tumour-heterogeneity, it would be surprising that the radiation-immune response would not be affected by classical radiotherapy parameters such as dose, dose per fraction, dose rate, overall treatment time and the timing of radiotherapy with a certain immune therapy. Pre-clinical models may give a clue to all these questions, but as human tumours and the immune system are fundamentally different from that of rodents, a simple extrapolation from lab results to patients is inappropriate. Clinical trials remain therefore essential. To the best of my knowledge, at the time of writing, no prospective studies that specifically were designed to address dose/ fractionation questions have been published. Moreover, in the absence of established biomarkers, the only endpoints that can be used are related to clinical outcome, such as response rates, progression-free survival (PFS) and overall survival (OS). Prospective non-randomized studies that combined interferon, IL-2, GM-CSF or ipilimumab with radiotherapy ranged from a single fraction of 8 Gy to 60 Gy in 2 Gy fractions, with no clear differences in outcome. Abscopal responses were observed with radiotherapy in patients who progressed after ipilimumab and in whom IO was continued after 30 Gy/ 10 fractions, 20/5 and 24/1. In NSCLC, abscopal respons was seen after 30 Gy/ 10 fractions. In conclusion, the optimal dose/ fractionation to induce immune activation is far from being elucidated and may occur over a wide range of doses and fractionations. SP-0336 Ongoing and upcoming clinical trials evaluating different RT schedules in combination with immunotherapy F. McDonald The Institute of Cancer Research and The Royal Marsden NHS Foundation Trust SP-0337 Understanding and targeting the underlying drivers of tumor hypoxia B. Wouters 1 , D. Cojocari 2 , J. Zhang 2 , T. McKee 2 , M. Zaidi 2 , M. Koritzinksy 2 1 Princess Margaret Cancer Centre, Ontario Cancer Institute, Toronto, Canada 2 Princess Margaret Cancer Centre, Princess Margaret Cancer Centre, Toronto, Canada Abstract text Low oxygenation, or hypoxia, is prevalent in human solid tumours and is a known contributor to therapy resistance. Pancreatic cancer is characterized by a poorly perfused, hypoxic microenvironment which contributes to poor treatment response and dismal patient survival rates. Hypoxic stress leads to the activation of several adaptive pathways including the unfolded protein response (UPR) and autophagy. Autophagy is a pro- Abstract not received Symposium: New strategies to target hypoxia in combination with radiotherapy

survival mechanism that degrades cellular components and provides nutrients to cells exposed to stressful environments. Hypoxia and the UPR promote autophagy activation through increased expression of the autophagy initiator kinase ULK1. The importance of autophagy activation through ULK1 and its contribution to hypoxia tolerance and the hypoxic tumour microenvironment in pancreatic cancer is unclear. The purpose of this study was to identify the underlying mechanisms through which ULK1 may promote hypoxia tolerance and consequently influence the tumour microenvironment. We used pancreatic cancer cell lines and patient-derived organoid models to investigate this question in vitro and in vivo. ULK1 was repressed using inducible lentiviral-shRNAs or the small molecule ULK1 kinase inhibitor 6965. First, we found that ULK1 is particularly important for the survival of pancreatic cancer cell lines and patient-derived organoids, and that its depletion led to increased endoplasmic reticulum (ER) mass, ER stress, and UPR activation. Inhibition of ULK1 in patient-derived pancreatic tumour organoids, using genetic and pharmacologic means, significantly decreased the organoid formation efficiency in hypoxia. Second, we uncovered a mechanism through which ULK1 can modify the oxygenation levels of a whole tumour by regulating the mitochondrial content and subsequent oxygen consumption. We found that loss or inhibition of ULK1 leads to dramatic increases in both mitochondrial content and oxygen consumption. In vivo immunohistochemical analysis demonstrated that this resulted in steeper oxygen gradients and more rapid development of tumour hypoxia. However, ULK1 loss also sensitized hypoxic cells to ER stress and cell death, resulting in increased areas of necrosis. The dual effects of ULK1 inhibition on the mitochondria and the ER resulting in more hypoxia, but decreased hypoxia tolerance resulted in dramatic tumour regression of pancreatic cancer xenografts. Together, our data suggest that ULK1 and autophagy promote hypoxia tolerance through two distinct mechanisms. Upon ULK1 knockdown, loss of mitochondrial homeostasis leads to increased oxygen consumption, and enhanced severity of hypoxia. At the same time, loss of ER homeostasis exacerbates hypoxia induced ER stress, reducing the tolerance of cells to hypoxia and causing a substantial tumour growth delay. This new mechanistic understanding of the importance of autophagy provides new opportunities for development of hypoxia directed therapies. SP-0338 Targeting oxygen consumption G. Higgins 1 1 Higgins Geoff, Department of Radiation Oncology, Oxford, United Kingdom Abstract text Previous attempts to improve clinical outcomes of patients treated with radiotherapy by overcoming tumour hypoxia have met with limited success. To date, no such treatment is in widespread clinical use. Most hypoxia modification strategies such as the use of oxygen mimetics, hyperbaric oxygen, and carbogen have sought to increase oxygen ‘supply’ to tumours. Pre- clinical and mathematical modelling data suggests that decreasing oxygen ‘demand’ may be a more effective way to reduce tumour hypoxia. Drugs that reduce mitochondrial oxygen consumption at clinically relevant concentrations might therefore be able to significantly reduce tumour hypoxia, leading to increased tumour radiosensitivity. I will discuss results from two drugs that reduce OCR: 1) We have identified that atovaquone, a safe, and widely used antimalarial drug reduces the oxygen consumption rate (OCR) of tumour cells via inhibition of complex III of the electron transport chain. Atovaquone causes rapid reduction of hypoxia in both 3D spheroid,