Academic research community (GECIP)

DNA and RNA are the genetic material carrying the instructions used in human development, normal functioning and reproduction. RNA plays a vital role in controlling when and where particular genes are expressed. Our research laboratory discovers and characterises new RNA molecules in cancer and how clinical intervention of these RNAs may represent new therapeutic strategies.

Solitary fibrous tumors (SFTs) can affect any region of the body and have a recurrent NAB2-STAT6 gene fusion (a specific genomic alteration). The diagnosis of SFTs can be challenging and the (epi)genetic changes driving aggressive behavior are unknown. The WGS results of the 100.000 Genomes project have also shown that genetic aberrations which are thought to be important as malignant drivers of these lesions, are not always identified in malignant SFTs. The WGS results thus indicate that epigenetic factors (factors which do not involve a change in the genome but do influence gene expression) might be driving malignant behavior in SFTs. Furthermore, few studies have also looked at the influence of the tumour micro-environment on the methylation profiles of SFTs. The latter is important as the heterogeneous (epi)genetic profiles in other cancers might be microenvironment driven and this could also possibly correlate with biological aggressiveness of these tumours. Through the work outlined in this project genetic and epigenetic correlations with histopathological predictors of malignant behaviour will be determined. Elucidating the gaps and inconsistencies in our (epi)genetic and tumour micro-environment knowledge of SFTs will help to improve patient stratification and prognostication.

A variant, rs2305089, is found in nearly all patients with a rare tumour called chordoma. This inherited variant is also found in roughly half of the general European population implying that further variants are needed to cause a chordoma to develop. The rs2305089 variant resides in the DNA sequence that codes for a gene called brachyury or TBXT. TBXT is normally only expressed during embryonic development and is absent after birth. However TBXT is expressed in chordomas and also a minority of commoner cancers such as colon and breast. It is not known whether the rs2305089 variant is also commoner in these cancers than in the healthy population. Knowing this may shed light on why TBXT is expressed in these cancers. The Genomics England data presents a unique opportunity to find this out for the first time across several cancers.

Cancer variants run the gamut from small alterations involving few DNA bases to complex variants involving large genomic regions. Small alterations are easy to detect using current DNA sequencing technologies (i.e. short-read or Illumina sequencing). However, complex DNA alterations cannot be fully characterized using mainstream DNA sequencing techniques. Therefore, the functional consequences of these variants and their clinical implications remain largely uncharacterized, although increasing evidence suggests that they play a crucial role in cancer evolution and drug resistance. In this project, we will develop computational tools to analyse data from sequencing machines whose unique technology permits to resolve these complex alterations by continuously reading long stretches of DNA; hence these methods are collectively known as "long-read sequencing". These emerging technologies have enormous potential to resolve complex DNA alterations, and thus, help us better understand their clinical implications, and more broadly, the genomic architecture of human tumours. We will compare short and long-read sequencing data generated for the same tumour specimens to assess and compare the diagnostic power and potential to resolve DNA alterations of both technologies. Although we will primarily focus on the analysis of brain tumours, sarcomas, and hypermutated tumours, the tools we will develop will be applicable across diverse cancer types, and will lay out the foundation for the application of long-read sequencing methods in clinical settings, as well as the development and validation of additional algorithms.

Gene-fusions involving NTRK have been identified as therapeutic targets for Larotractinib and Entrectinib in rare cancer types (secretory breast carcinoma, mammary analogue secretory carcinoma and infantile fibrosarcoma. Analysis of around 600 cases within the sarcoma GeCIP domain have shown these NTRK fusions were detected WGS in about 2% of cases (n=13), including osteosarcomas and soft tissue sarcomas of different subtypes. Preliminary results from RNA-seq data from UCL Cancer Institute and a targeted panel at the Royal Marsden Hospital have shown these fusions are transcribed in two cases. We propose a pan-cancer study to determine the prevalence of NTRK rearrangements in other cancer types, and to determine if these are transcribed and therefore targetable.

Sarcoma is a rare type of cancer with approximately 70 subtypes. Although classification of sarcoma subtypes has improved over the last 20 years, pathologists still not infrequently find it challenging to classify a tumour, that is, provide a firm diagnosis. Without a robust diagnosis, it is difficult to provide the patient with a standard of care treatment. We consider that through the DNA methylation profiling of sarcomas from approximately 1000 patients we will be able to classify these tumours more robustly. We will also correlate the profiles with clinical outcome and response to therapies. It is possible that the new information could lead to new treatments, as there are encouraging research results coming through with respect to the development of new drugs targeting abnormal methylation.

Chordoma is a rare cancer that arises in the bone, affecting 1 in 800,000 people. It occurs along the spinal column and grows into surrounding tissues causing different symptoms depending on the site of origin. Surgery is the most common form of treatment, and some patients receive radiation therapy including proton beam therapy - a specialised form of radiation. Chordomas often recurs locally, despite optimal surgery and radiation. The mean survival is 7 years from diagnosis, and this has not improved for decades. By looking at blood and tissue samples from patients with chordoma who have been recruited to the 100,000 Genomes Project we aim to undertake a comprehensive study of this rare disease. Our comprehensive molecular profiling of chordomas will give us insight into how this tumour develops and spreads. We will then link these molecular findings to disease outcome with the aim of developing biomarkers in the tumours and in patients’ blood which will predict behaviour and potentially the development of new treatments. We will also use digital imaging to modernise how diagnoses can be made and allow pathologists to use their skills to address more complex questions about disease. This proposal extends, complements and adds value to a current NIHR England-wide study - Guiding Chordoma Treatment Through Molecular Profiling - in which all patients with chordoma are being invited to enrol in a study which allows longitudinal analysis of multiple aspects of their disease and how it impacts on their lives.

Chondrosarcoma is the most common primary bone tumour in adults for which surgery is the only effective treatment. There is a spectrum of changes seen down the microscope and these changes (grading) are used to inform doctors how a tumour is likely to behave, and also dictates the treatment. There is significant evidence that there is poor inter- and intra-reproducibility between pathologists when they grade cartilaginous tumours. We propose to use the genomic analysis of cartilaginous tumours generated from the 100,000 Genomics Project for patient benefit. We will correlate the genomic / epigenomic data with the histological imaging of the same tumours, and the clinical outcome. We will undertake machine learning and employ artificial intelligence to analyse and grade the histological images of the cartilaginous tumour diagnoses with the aim of improving the reproducibility of grading the disease and providing diagnoses that are more informative – that is - providing patients with a more reliable prognosis. By linking these data with circulating tumour DNA it will allow us predict with even greater accuracy the risk of disease progression. Furthermore, the ctDNA should allow better surveillance of patients and early detection of relapse. This proposal extends, complements and adds value to a current NIHR England-wide study in which plasma is being collected on all patients with cartilaginous tumours

Adamantinoma is a rare cancer arising in bone, predominantly occurring in young people. It is difficult to select the best treatment for those affected because we cannot distinguish between tumours that maybe cured by limited surgery alone and tumours that are at risk of either recurring or spreading to other parts of the body, and therefore would benefit from more radical treatment. To date, there is virtually nothing known about either the aetiology or the genetic events that cause these tumours, as they have not been the focus of significant research. We have biobanked this tumour type over the last 15 years, a significant collection in view of its rarity. We now want to use the most modern technologies to understand what makes these tumours grow, and to learn what factors allow them to spread. To this end, we will study their genetic code compared to the normal genetic code in the blood of those patients with this disease. Our results will allow doctors to improve diagnostic accuracy, direct treatment and provide information about outcomes and prognosis for patients.

Sarcomas have complex genomes and unlike other cancers most of this in sarcoma complexity is in the form of gains and losses of DNA (copy number). It is difficult to interpret this type of data especially when it affects multiple regions of the genome at once. Our group has recently refined a method to unravel this complexity using a mathematical algorithm. We have applied this method to the most complex of sarcomas and been able to derive information such as the processes that cause these copy number changes. This work has been recently published. We have called these copy number signatures and now aim to apply it across the GEL sarcoma dataset to learn more about the processes that cause these changes in sarcoma samples. Moreover we believe that this can be used as a prognostic marker in some sarcoma subtypes as has been shown in ovarian cancer for instance.

Some sarcoma can have a strong familial component, and are more common in people with recognised hereditary cancer syndromes, such as retinoblastoma, and Li-Fruameni syndrome. Up to 33% of paediatric sarcomas are associated with a significant family history of cancers. The risk of sarcomas in relatives of all children with sarcoma is increased compared to age-matched controls. Surgery is still the most effective single treatment available to patients with sarcoma. Therefore, we will use the data provided from the 100,000 Genomes Project to identify genetic markers that could be used in the future to detect individuals with an increased risk for sarcoma. This could allow for early detection of sarcoma, for instance by detecting low-grade sarcomas before they progress to a more aggressive form, and thereby improving overall survival. Importantly, many studies of asymptomatic individuals with retinoblastoma, and Li-Fraumeni syndrome provide strong support for this approach.

There is emerging evidence to suggest that in multiple cancer types, the immune system plays an important role in determining outcome. In multiple cancer types, the number of variants in the cancer is associated large numbers of immune cells in tumours and with a good responses to drugs that invigorate the immune system. Detailed studies investigating this relationship in sarcoma have not been conducted to date. However, there is evidence for some sarcoma subtypes showing good responses to immune based therapies including alveolar soft part sarcoma, angiosarcoma and undifferentiated pleomorphic sarcoma. The genetic basis for these responses is largely unexplored. The aim of this study is to investigate the genomic aberrations that are associated with various immune signatures present in soft tissue sarcomas. We believe this could provide evidence upon which clinical drug trials can be initiated.

Genomic medicine is transforming clinical practice in the health care system in the UK. This has been largely initiated and accelerated through the Genomics England 100,000 Genomes Project to which recruitment has now closed (31st December 2019). Building on this project NHSE has introduced a single national testing network sited across seven Genomic Laboratory Hubs, to consolidate and enhance the existing laboratory provision. The ambition is for all cancers to be subjected to in depth molecular profiling so that patients can receive optimal treatment. To achieve this the testing must be done at large scale not only for cost and logistic reasons but also because we still have a great dal to learn about the cause the treatment of cancer. This can be accelerated though centralisation of sample testing and analysis. The challenge ahead is significant but much can be learnt from the experience of the 100,000 Genomes Project. At the Royal National Orthopaedic Hospital, we consented just over 1000 patients to the 100,000 Genomes Project. We propose to study in depth what we achieved so that we can improve the patient pathway at every level: we will analyse how patients who require the new molecular tests are identified, how the samples are obtained, through to the interpretation of the new tests, and their delivery to patients.

The clinical outcome for patients diagnosed with osteosarcoma, the most common primary bone tumour in young adults in the UK, has not improved significantly in the last 40 years. This is despite the standard of care being an intense regime of chemotherapy followed by surgery. Furthermore, we still nearly 40 years on cannot predict the 40% of patients who do not respond to chemotherapy. Using the newest technologies, we will investigate how osteosarcoma evolve and progress, and how they develop resistance to chemotherapies. This study has been made possible through patient involvement in the 100,000 Genomes Project.

The clinical outcome for patients diagnosed with osteosarcoma, the most common primary bone tumour in young adults in the UK, has not improved significantly in the last 40 years. This is despite improvements in the way in which we classify these tumours done the microscope and into different genetic subtypes. The primary treatment for osteosarcoma is chemotherapy followed by surgical excision, but nearly 40 years on we still cannot predict which patients benefit from receiving chemotherapy. The 100,000 Genomes Project has allowed the biobanking and genomic sequencing of the largest clinically annotated sample collection of osteosarcoma in the world. We are particularly interested in epigenetics of osteosarcoma, meaning we will look at how expression of different genes is controlled within human tissue cells, rather than looking at abnormalities in the genetic code or DNA itself. Little research into epigenetics osteosarcoma has been done but in other cancers it has been informative. Indeed, there are new drugs which are being developed which block epigenetic driven growth of tumours and which are showing encouraging results. We will study the methylomes and transcriptomes of osteosarcomas and correlate the findings with the clinical outcome of patients. The data will be integrated with the genomic data from the same samples thereby providing a comprehensive understanding of the molecular biology of this disease. Our aim is to learn why this disease develops, and to learn how to predict which patients are likely to respond to standard of care therapies. Ultimately we wish to identify new targets for treatments.

Sarcoma is a rare type of cancer with approximately 70 subtypes. Although classification of sarcoma subtypes has improved over the last 20 years, pathologists still not infrequently find it challenging to classify a tumour – that is – provide a firm diagnosis. Without a robust diagnosis, it is difficult to provide the patient with a standard of care treatment. We consider that through the genomic analysis of the sarcoma samples from approximately 1000 patients we will identify new recurrent driver variants which could be exploited as biomarkers for classifying sarcoma subtypes. This would allow better stratification for treatment and potentially the development for new treatments.

Sarcoma is a rare type of cancer with approximately 70 subtypes. Despite the more accurate histological classification and the intense multimodal chemo- and radio-therapeutic regimes, the clinical outcome for most patients with sarcoma has not improved over the last 40 years. Furthermore, prognoses are not based on objective criteria, and the ability to predict response to therapies is poor. To change this, there is a need to understand the molecular basis of sarcoma in the context of disease behaviour. We are confident that the sarcoma samples from at least 1000 patients, of which approximately 30% have multiple samples collected and analysed, will play a major role in improving patients’ care and treatment plans. Using the newest technologies, we will study the changes to the cancer genome that drive sarcoma development, which could be used in the future as markers for diagnosing sarcoma subtypes, and may be used to guide clinical interventions. We will also study what causes variants in these tumours, which may lead to preventative strategies in the future. This is made possible through patient involvement in the 100,000 Genomes Project, meaning we have access to many different tumours.

Sarcoma is a rare type of cancer with approximately 70 subtypes. Despite the more accurate histological classification and the intense multimodal chemo- and radio-therapeutic regimes the clinical outcome for most patients with sarcoma has not improved over the last 40 years. Furthermore, prognoses are not based on objective criteria, and the ability to predict response to therapies is poor. To change this, there is a need to understand the molecular basis of sarcoma in the context of disease behaviour. We are confident that the sarcoma samples from at least 1000 patients, of which approximately 30% have multiple samples collected and analysed, will play a major role in improving patients’ care and treatment plans. Using the newest technologies, we will investigate whether different areas in the same tumour have different genetic abnormalities. These analyses can tell us how tumours develop, evolve and progress over time and space. This is made possible through patient involvement in the 100,000 Genomes Project, meaning we have access to many different tumours. Our study aims to stratify patients for treatment, improve early detection rates and the success rates of therapies, improving survival and future outcome for sarcoma patients.

We shall seek to identify the contribution of known and new gene variants carried in the germline (constitutionally, in the blood) to development of sarcoma

Sarcoma represents only 2% of all cancers and is composed of over 100 subtypes many requiring specific treatments and conversations with patients to explain their prognoses. However, in many cases the accurate information cannot be provided to individuals about their prognosis or how they may respond to therapies. Computer algorithms which use the digitised images of tumours together with molecular data and clinical outcome of 100s – 1000s of patients it should be possible to help pathologists provide more accurate diagnoses, prognoses, and predictions of responsiveness to therapies.

This computerised approach to studying sarcoma will benefit patients affected by sarcoma.