invited speakers

Session 1

Amanda Chaplin, University of Leicester

Redefining Non-Homologous End Joining DNA-repair using Cryo-electron Microscopy

Cellular DNA is exposed to multiple sources of damaging agents, including endogenous sources such as oxidation and exogenous sources such as radiation.  DNA repair mechanisms are vital as DNA double-strand breaks (DSBs) can cause cell death and eventually cancer if left unrepaired. Non-homologous end joining (NHEJ) is one of the two mechanisms required for DSB repair. NHEJ is dependent on several canonical proteins, namely DNA-PKcs, Ku70/80, DNA Ligase IV, XRCC4 and XLF, in addition to several regulatory proteins. Traditionally, NHEJ was thought to consist of three simple linear steps. However, recent cryo-EM data has provided an unexpected glimpse of alternate complex protein arrangements, leading us to propose that the mechanism of NHEJ is more complicated that originally believed. We have identified two alternate long-range DNA-PK dimers, one mediated by Ku80 and the other by XLF. These dimers are essential for efficient DNA repair. We have also recently shown that the accessory protein, PAXX can stabilise specifically the Ku80 DNA-PK dimer and how this has overlapping roles with XLF. Furthermore, we have used cryo-EM to visualise small molecules such as IP6 binding and DNA-PKcs inhibitors, which will aid in future therapeutic development.


Tom Clarke, Boston, USA

ZNF280A links DNA double-strand break repair to human 22q11.2 distal deletion syndrome

DNA double-strand breaks (DSB) are one of the most deleterious forms of DNA damage, and if unresolved result in DNA mutations and chromosomal aberrations that can cause disease, including cancer. Repair of DSBs by homologous recombination (HR) requires extensive nucleolytic digestion of DNA ends in a process known as DNA end-resection. In recent years, progress has been made in understanding how this process is initiated, however the later stages of this process – long range DNA end-resection, is not well understood. Indeed, many questions remain as to how the DNA helicases and endonucleases that catalyze this process are regulated, a key step to avoid spurious activity in the absence of breaks. The importance of DNA end-resection in human disease is highlighted by several human genetic syndromes which are caused by mutations or deficiencies in key proteins involved in this process. In this study, using high throughput microscopy (HTM) coupled with a cDNA “chromORFeome” library, we have identified ZNF280A as a novel chromatin factor that is essential for DNA double-strand break repair. Mechanistically, we demonstrate that ZNF280A promotes long-range DNA end resection by facilitating the recruitment of the BLM-DNA2 helicase-nuclease complex to DNA double-strand break sites, enhancing efficiency of the enzymatic activity of this complex at DNA damage sites. ZNF280A is therefore a key accessory factor for DNA end-resection and DNA repair by homologous recombination. Importantly, ZNF280A is hemizygously deleted in a human genetic condition, 22q11.2 distal deletion syndrome. Features of this condition include congenital heart disease, microcephaly, immune deficiency, developmental delay, and cognitive deficits – features that are associated with other human syndromes caused by defects in genes involved in DNA repair. Remarkably, cells from individuals with a 22q11.2 distal deletion have defects in homologous recombination and increased incidence of genome instability, providing the first evidence of defective DNA repair as a potential mechanistic explanation for several clinical features associated with this human condition.


Session 2

Rita Pedrosa, Queen Mary University of London

Targeting radiation induced vascular endothelial-cell dysfunction to modulate response to therapy

30% of all cancer patients, over 90,000 new cancer cases, will receive radiotherapy as part of their curative treatment in the UK.  However, resistance to radiotherapy is still a major challenge especially in non-small cell lung cancer (NSCLC) patients. Tumour vasculature-derived angiocrine signals (chemokines, cytokines, and growth factors secreted by vascular endothelial cells) have important roles in modulating responses to DNA-damaging agents, however their role in resistance to RT remains unexplored. In this study, we investigate how RT-induced vascular inflammation/dysfunction and derived angiocrine signals mediate resistance to RT.

Using a published scRNA seq dataset (Nolan et al, Nature Cancer, 2022), we have uncovered an enrichment upon radiation of a subset of vascular endothelial cells in the lung that possess immune modulatory, antigen presenting functions and unique expression of PD-L1, named iMECs (immune modulatory endothelial cells). This subcluster of ECs presents upregulation of inflammatory associated signatures such as TNF-a/NF-Kb and Jak/STAT3, among others. Furthermore, the existence of this iMEC subcluster was confirmed in human scRNA seq datasets and at protein level using image mass cytometry of early-stage lung cancer samples. Moreover, using both immunofluorescence and western blot nuclear fractionations, we have confirmed that RT induces activation of the canonical inflammatory pathway- NF-KB in mouse lung endothelial cells in vitro. We have also analysed the secretory phenotype (angiocrine signalling) of human pulmonary microvascular endothelial cells upon radiation in vitro, confirming secretion of several cytokines and chemokines associated with immune cell regulation.

Using 2D co-culture models, we have also demonstrated that vascular endothelial cells provide radioprotective signals to some tumour cell lines, and this effect seems to be, at least partially, contact-independent, as demonstrated with transwell experiments.

Initial in vivo data, using multifocal adenoviral-cre induced KP (KrasG12D p53LOF) NSCLC mouse model, suggests that tumour growth is resistant to hemithorax RT of 5 fractions of 2Gy, without major changes in blood vessel numbers or vascular growth patterns. Furthermore, immune characterisation of this model by flow cytometry revealed a more immunosuppressive response. Unifocal orthotopic NSCLC mouse models were also used to interrogate targeted RT (small animal radiation research platform, SARRP)-derived vascular responses and their effects on the TME. Using KP cells and CMT cells unifocal lung tumours, has revealed distinct responses to RT, with KP cells model being less sensitive to treatment compared with CMT. Interestingly, distinct vascular remodelling responses are observed within the two models.

Future studies will investigate the differential RT-derived vascular responses (and proportions of iMECs in the tumour vasculature) and associated immune infiltrate profiles comparing resistant (KP multifocal and unifocal models) vs sensitive (KL- KrasG12D Lkb1fl/fl; CMT167) mouse models, and how these might correlate with responses to RT. Additional in vitro studies will aim to address the ability of endothelial cells upon RT to recruit and possibly change immune cell phenotypes.

In conclusion, we have shown that RT induces NF-KB dependent vascular endothelial-cell inflammation, and the derived angiocrine signals contribute to protection of adjacent TC, while also possibly regulating immune cell responses. Therefore, modulating vascular EC responses to radiation might prove beneficial in improving responses to treatment of RT resistant NSCLC.

Hala Estephan, University of Oxford

Hypoxia inhibits MHC I expression and antigen presentation to escape immune surveillance

Hypoxia is a common feature of solid tumors that has previously been linked to resistance to radiotherapy and chemotherapy, and more recently to immunotherapy. Hypoxic tumors exclude T cells and inhibit their activity, suggesting that tumor cells acquire a mechanism to evade T cell recognition and killing. Using an unbiased proteomic approach to determine what mechanisms contribute to tumor immune evasion by hypoxia, we found that hypoxia inhibited MHC I at the protein level and in consequence induces a significant change in antigen presentation. Hypoxia decreases MHC I expression in an oxygen-dependent manner, mediated by the activation of autophagy through the PERK arm of the unfolded protein response. Furthermore, using an immunopeptidomics-based LC-MS approach, we found a significant reduction in presented antigens under hypoxia. Inhibition of autophagy under hypoxia rescued MHC I expression and enhanced antigen presentation. In experimental tumors, reducing mitochondrial metabolism through a complex I inhibitor increases tumor oxygenation and both MHC I levels as well as the immunopeptidome. These data provide the molecular mechanism governing tumor immune evasion in hypoxic conditions, offering novel insights for therapeutic interventions targeting hypoxia-induced alterations in antigen presentation.

Session 3

Stephen McMahon, Queen’s University Belfast

Modelling intrinsic radiosensitivity - How far does DNA repair get us?

It is well-established that genetic differences between cancers significantly affect their radiosensitivity, which in turn plays a major role in determining treatment response in clinical settings. However, despite this knowledge, there has been limited application of radiosensitivity models to personalise radiotherapy doses based on these differences, due to challenges in building robust predictions of responses across different biological systems.

The role of DNA repair role in radiation response is well-established, with decades of literature supporting its critical influence on cell fate. Extensive research has been undertaken to understand these processes, both to better characterise the response of different cells to radiation, and to identify potential targets for radiosensitisation.

This talk will review our work on modelling DNA repair in response to radiation-induced damage. This includes simulating the initial distributions of damage, its interaction and (mis)repair, and its subsequent consequences for cell fate. Importantly, this approach also considers the function of different genetic pathways in these systems, and the impact that dysregulation of key DNA repair processes has on radiosensitivity.

This modelling approach effectively captures numerous aspects of biological responses, including initial DNA damage and its repair over time, as well as biological consequences such as mutation and chromosome aberration formation, and overall clonogenic survival. This is applicable across different radiation qualities, and in cells of different DNA repair capacities, validated in both cell line data and CRISPR-Cas9 knockout screens.  

However, while this work highlights the critical impact of DNA repair dysregulation on radiosensitivity, it also offers the opportunity to explore the prevalence of such factors in clinical cohorts. Analysis of both cell line and patient population databases shows that only a small fraction of samples – on the order of a few percent – exhibit mutational profiles associated with DNA repair defects which materially affect radiosensitivity.

This suggests that while these defects can serve as a significant radiosensitivity marker when present, they cannot explain the entire range of radiosensitivity observed in patient populations. This highlights the need for exploration of other regulators of response. Some possible pathways which may be driving these effects will be discussed, to highlight areas to underpin future radiobiological modelling.

Andreas Kyprianou, Warwick

Proton beam de-energisation and the Bragg Peak for cancer therapy via jump stochastic differential equations.

Proton beam therapy is an approach to treating certain cancers, that has been in operation in the UK for less than a decade. It consists of firing a high energy protons beam towards cancerous tissue. The physics of proton deceleration ensures that the beam energy can be focused into unhealthy tissue. The Bragg Peak describes the rate of energy loss per unit length along the beam and is used as a calibration tool for treatment preparation. Bortfeld (1997) proposed a parametric family of curves that can be accurately calibrated to data. The Bortfeld curve is strictly a one-dimensional profile and there are currently no known mathematical models in higher dimensions. We build from first principles the first mathematical model for the de-energisation of protons using stochastic differential equations. Our approach affords us the luxury defining the natural analogue of the Bragg Peak curve in two or three dimensions. This work is purely theoretical and a first step providing the foundations for future work in which we will develop comparative studies with existing methods.

Session 4

Richard Amos, University College London

Development of clinical and pre-clinical light-ion beam therapy in the UK


Interest in the application of protons and other light-ions for radiotherapy continues to grow globally due to the favourable dose-deposition characteristics compared to x-ray based techniques. The potential to reduce radiation-related toxicities for cancer patients indicated for radiotherapy drives this interest. In recent years NHS England (NHSE) has developed a national proton beam therapy (PBT) service located at the Christie Hospital in Manchester and at University College Hospital in London. This service provides treatment to those patients indicated for PBT along with participation in clinical trials to test the efficacy of PBT for disease sites not yet indicated.

There is also a growing need to have greater access to facilities for pre-clinical research in the field of light-ion beam therapy. Innovative treatment techniques such as ultra-high dose rate (UHDR) and spatially-fractionated ion-beam delivery have shown some evidence of improved efficacy in early pre-clinical investigations, however the underlying mechanisms are yet to be understood. Furthermore, radiobiological studies of various light-ion species are desirable to ascertain their relative effectiveness. To meet this need there are plans to develop an ion-therapy research facility (ITRF) in the UK. Such an undertaking requires a cost-effective solution to be viable. Research is ongoing to design a laser-hybrid accelerator for radiobiological applications (LhARA), and for this system to be the source for the ITRF.

This presentation will provide an update on the current status of the PBT clinical service and summarize ongoing research for the development of the LhARA and ITRF.

Mark Hill, Oxford

Radiation track structure: how does their spatial and temporal properties drive the radiobiological response.

Ionising radiation is far more biologically effective than might be expected from the limited amount of energy deposited or the comparatively small amount of DNA damage induced, compared to the vast amount of endogenous damage arising from normal metabolism of the cell. This is due to the unique way energy is deposited along highly structured tracks of ionisation and excitation events, resulting in the correlation of DNA damage sites from the nanometre to the micrometre scale.  Correlation of these events along the track on the nanometre scale results in clustered damage, which not only includes DNA double-strand breaks (DSB) and the more difficult to repair complex DSB (which includes additional damage within a few base pairs) but also non-DSB clusters. Track structure varies significantly with radiation quality and the increase in relative biological effectiveness (RBE) observed with increasing linear energy transfer (LET) in part corresponds to an increase in the probability and complexity of clustered DNA damage produced. Likewise, with increasing LET there is an increase probability of correlation over larger scales, associated with packing of DNA and associated chromosomes within the cell nucleus. This can also have a major impact on biological response, with difference becoming more pronounced with low doses associated with radiation protection exposures. The proximity of the correlated damage along the track increases the probability of miss-repair through pairwise interactions resulting in an increase in probability and complexity of DNA fragments/deletions, mutations and chromosomal rearrangements. The temporal properties radiation can also have a major impact on the resulting biological effectiveness.

Understanding the mechanisms underlying the biological effectiveness of ionising radiation can provide an important insight into the resulting radiation biology, improving the efficacy of radiotherapy, as well as the risks associated with exposure. This requires a multi-scale approach for modelling, considering the physics of the track structure from the millimetre to the nanometre scale, temporal aspects of exposure, the structural packing of the DNA within the nucleus, the resulting chemistry, along with the subsequent biological response. In addition to an overview of the link between physical interactions, associated chemistry and biological response, the presentation will also highlight some of the common misconceptions.

Session 5

Marianne Azner, University of Manchester

Radiotherapy “big data”: the role of AI and advanced image analysis

This presentation will explore the impact of advanced technologies such as deep learning (DL) and large-scale image analysis (e.g. voxel-wise analysis, radiomics) on treatment planning, delivery, and outcomes in radiotherapy. We will review clinical applications, as well as developments for research, e.g. use large cohorts of real-world data to learn from every patient and enhance our understanding of dose-response relationships.

Uwe Oelfke, Institute of Cancer Research

SFRT/FLASH irradiators for pre-clinical research:  Microbeams, FLASH SARRP and LFXT

Reviving the paradigms of spatially fractionated radiotherapy (SFRT) and dose delivery at Ultra-High dose rates (> 40 -100 Gy/s, FLASH) has inspired a wealth of pre-clinical research in the community of radiation physics and biology for the last decade.

The main aim of these studies was i) to elucidate the biological mechanisms of the observed dose sparing effects in normal tissues and ii) to study the tumoricidal properties of SFRT and FLASH. While these efforts led to several exciting hypotheses and potential explanations of the collective ‘in-vivo’ set of data, we are still lacking a satisfying understanding of their underpinning radiobiological mechanisms.

 One severe bottleneck for rapid progress of the respective research is the lack of suitable experimental irradiation facilities. For FLASH RT, the most common irradiation modality are electron beams, either generated by dedicated LINACs or modified clinical accelerators with energies ranging from 6 - 10 MeV. Another prominent modality for FLASH RT are high energy protons. For photons at pre-clinical energies of 150 – 300 kV, a dedicated irradiator – the FLASH SARRP – has recently been introduced.

For SFRT, utilizing spatial beam widths of microbeams (< 250 microns), minibeams (1-5 mm) and larger grids of radiation (> 5mm), there is a scarce spectrum of facilities available. The most pronounced bottleneck is the availability of pre-clinical microbeam irradiators, especially since the ESRF in Grenoble has recently decided to put the microbeam-mode of operation of beamline ID17 on hold.

We will report on the development and commissioning of our micro-beam and FLASH irradiators at the Centre for Cancer imaging and will also cover the development of the line-focussed X-ray source (LFXT), currently ongoing at the Technical University of Munich. The concept of the LFXT, originating from work completed at ICR in 2017, promises to deliver microbeams at flash dose rates of up to 100 Gy/s with an unprecedented geometrical accuracy.

Our adaptation of a conventional SARRP platform for the delivery of microbeams is based on the integration of an electronically controlled slit collimator, allowing varying beam widths ranging from 50 - 170 microns. We will describe the process of dosimetric commissioning, report on the achievable dose patterns and describe the developed workflow for the irradiation of several ‘in-vivo’ tumour models including a set of initial results.

 This section will be followed by a brief report on the initial dosimetric calibration of the FLASH SARRP platform, which consists of two rotatable x-ray tubes operating at a maximal output at 150 kV and a current of 630 mA resulting in dose rates between 80 and 90 Gy/s. Finally, the talk will introduce the concept of the LFXT and its technical realisation with its first prototype at TUM.

Session 6

  • Christophe Badie, UKHSA

New insights in radiation leukaemogenesis

To improve health risk estimates and radiological protection for low dose and dose-rate exposures of ionising radiation (IR) encountered in occupational, medical, and public/emergency situations, further research is required. Epidemiological studies provided clear evidence for increased leukaemia incidence following IR exposure even at low doses with acute myeloid leukaemia (AML) being the most prevalent. Animal studies greatly contribute to improve the understanding of radiation-induced AML (rAML). Murine rAML feature both hemizygous chromosome 2 deletions and point mutations (R235) within the haematopoietic regulatory gene Spi1. Analysing 123 rAML, we identified new pathways without Spi1 R235 where the decrease in Spi1 gene expression is negatively correlated with Spi1 promoter DNA methylation at specific CpG sites. Moreover, we generated mouse models to track preleukemic cells in vivo to reveal the sequence of molecular events and identify the cells of origin and we confirmed the ‘’two-hit’’ mechanism using a hemizygous Spi1 R235C point mutation conferring hypersensitivity to rAML. Similar SPI1/PU.1 polymorphisms in humans could lead to IR enhanced susceptibility following medical or environmental exposure. Increased rAML sensitivity and shortened dose-dependent latency of this model allow to quantify rAML risk at low doses/dose-rates, otherwise prohibited by the high numbers of animals to reach statistical significance. Data generated are being used to assemble biologically based risk projection models to evaluate low dose rAML incidence and the role of hyper-radiosensitivity (HRS) in rAML by altering the probability of Spi1 mutations occurrence/persistence relevant for human populations. Last, we identified an epigenetic signature of IR in therapy-related AML patients.

  • Amy Berrington, Institute of Cancer Research

Cancer Risks from Pediatric CT Scans 

CT scans save lives by improving diagnosis, limiting unneeded medical procedures, and enhancing treatment.  The rapid growth in use over the last few decades has raised concerns, however, about the associated levels of radiation exposure especially in children.  This led to several large-scale epidemiological studies of cancer risks in cohorts of children who underwent CT scans.  The results from the UK, Australian and European CT scan cohorts will be presented and compared with a focus on leukemia and brain tumours.   Methodological issues will be described and the results put into context with other low-dose epidemiological studies.  Finally the implications for radiation protection will be discussed.    

Session 7

  • Ananya Choudhury, University of Manchester
Translating hypoxia basic science into clinical practice

Scientists have been aware of the importance of hypoxia for decades culminating in the Nobel prize for Physiology in 2019. Hypoxic cancers are more likely to metastasise, be treatment resistant and have a poor prognosis. This talk will bridge the gap between laboratory discoveries and patient care. I will discuss discoveries and interventions that allow targeting hypoxia to improve patient outcomes in the clinic.

  • Monica Olcina, University of Oxford
Improving radiotherapy in immunosuppressive microenvironments.

Colorectal cancer subtypes with the worst prognostic outcome are stromal-rich, have poor CD8+ T-cell infiltration and high complement gene expression levels. The impact of high complement expression on treatment outcome in these tumours, however, is still unclear. When grown subcutaneously, tumour organoids originally derived from villinCreER Apcfl/fl KrasG12D/+ Trp53fl/fl TrgfbrIfl/fl mice, display features resembling those of colorectal cancers associated with poor outcome; including stromal-rich regions and poor CD8+ T-cell infiltration. Following RNA-sequencing of these tumours we have found that the complement system is the most significantly upregulated immune system pathway to be upregulated shortly following radiotherapy. In this model, we also find that C5aR1 is robustly expressed on malignant colorectal epithelial cells, suggesting tumour-cell specific functions. Indeed, targeting C5aR1 results in increased radiation-induced cell death specifically in tumours but not normal tissues and this results in improved tumour control following radiotherapy. Collectively, these data suggest that upregulation of complement genes may be part of the stress response mounted by irradiated tumours and that targeting C5aR1 could be targeted to improve tumour radiation response without increased toxicity in the normal tissue.

Session 8

  • Sam Terry (Kings College London) 
Preclinical efforts in molecular radiotherapy

Molecular radiotherapy (MRT), where primary tumour and metastases are irradiated, is an exciting research area for radiobiologists to apply their expertise from X-ray radiation to. With increased numbers of radiopharmaceuticals being tested preclinically and becoming available in the clinic, and a great investment of pharmaceutical companies in this area, there is now a need to better understand the biological effects of radionuclides. In this presentation, we will describe where the field is at preclinically and where future efforts are best placed to truly make the most of the potential of these radiopharmaceuticals.

  • Jon Wadsley, Sheffield Teaching Hospitals NHS Trust
Molecular radiotherapy- current status and unanswered questions

Molecular radiotherapy (MRT) refers to the delivery of radiation to malignant tissue via the interaction of a radiopharmaceutical with molecular sites. This may be administered orally, intravenously or more directly, for example by infusion into the hepatic artery. Whilst historically this treatment modality has been restricted to a small number of rare tumour sites, more recently a rapidly growing number of new therapies covering a wider range of cancers have emerged.

In this session we will review the current clinical indications for MRT and the evidence supporting these. We will consider where the major gaps are in our knowledge, and what further research is required to allow us to optimise treatments for each individual patient. We will review recent clinical trials which have attempted to address some of these issues and the lessons learned from these.

Finally we will consider current initiatives seeking to promote MRT research in the UK and further afield.