Projects using CellBank samples

A genome wide screen of SNPs to identify susceptibility alleles for childhood acute lymphoblastic leukaemia

Acute lymphoblastic leukaemia (ALL) is a common cancer of childhood in developed countries. Current evidence indicates it is likely a priori that all risk affected by any exposures will operate in the context of a genetic background of variable susceptibility. The United Kingdom Childhood Cancer Study (UKCCS) has established a unique clinical resource for studying inherited predisposition to ALL. Using DNAs from patients with ALL available through UKCCS we shall conduct a genome-wide association study of tagging SNPs to identify common variants influencing the risk of developing the disease. This information should greatly assist in identifying those at increased risk. In addition associations identified may prove extremely valuable via the functional links they reveal and either endorse current aetiological hypotheses or suggest new ones that merit testing via gene/environment specific hypotheses.

An investigation of the molecular basis for B lineage acute lymphoblastic leukaemia cell migration to extramedullary sites

Acute lymphoblastic leukaemia (ALL) is the commonest childhood cancer. Although massive strides have been made in treating this disease a significant minority of children still suffer disease recurrence (relapse) which may lead on to significant long-term health problems or death. Although leukaemia begins in cells from the bone marrow, most children who relapse have evidence of leukaemia at distant sites round the body, particularly the brain. This study sets out to understand how leukaemic cells might enter the brain and other distant sites and how they survive when they get there. This knowledge may allow identification of new tests and treatments for leukaemia.

Asparaginase-associated pancreatitis during treatment of childhood acute lymphoblastic leukaemia; characteristics and risk factors - a PdL/IBFM phenotype-genotype study

Acute lymphoblastic leukemia (ALL) is the most common cancer in children aged 1-14.9 years, with an incidence of approximately 3.5 cases per 100,000 children in Europe and the U.S. Survival rates have increased to 85% after first line treatment, but many patients are burdened by life-long toxicities. Asparaginase (ASP) is a crucially important antileukaemic drug. Studies have shown that patients receiving prolonged administration of this drug have a superior survival rate, compared to patients who have their therapy truncated due to toxicity. ASP is associated with several toxicities, pancreatitis being one of the most common. Asparaginase Associated Pancreatitis (AAP) has an incidence of 4-10% in contemporary protocols using ASP and is among the most common reasons for truncating ASP treatment. However, risk factors, including genetic predisposition to AAP, are not well known. This study provides deep phenotyping of more than 600 AAP patients based on data from 15 large international childhood ALL groups. Furthermore, we will explore the host genome variants associated with developing AAP. The latter is examined in more than 300 childhood ALL patients. The retrospective AAP phenotype study is the first of its kind in collecting data on a specific toxicity from multiple international childhood ALL groups, i.e. across many treatment protocols. This will provide sufficient power in describing in detail the phenotype of AAP. This large, and unique, dataset will allow exploration of risk factors, long term outcome including risk of chronic pancreatitis, and establish guidelines for ASP re-exposure after an episode of AAP. The Genome Wide Association (GWA) Study investigates ~2,500,000 Single Nucleotide Polymorphisms (SNPs) spread across the genome with emphasis on the exome. We will link individual SNPs as well as biological pathways to the risk of developing AAP. This is done based on SNP analysis of >300 of the AAP cases with available deep phenotype data. This study will provide new knowledge on AAP, enabling a more personalized cancer treatment strategy with ASP. Thereby improving treatment of childhood ALL in patients with high risk of AAP.

Biology of the TEL/ABL (ETV6/ABL)-positive leukaemia

CAR-T cell immunotherapy for relapsed/refractory T-cell acute lymphoblastic leukaemia (T-ALL)

T-cells are cells of our immune system that look for and destroy infected cells. Since cancer cells arise from our own tissues, T-cells do not usually recognise and attack them. Medical science has long sought to find a way to bring our immune system into the fight against cancer. Recently, a technology called 'CAR T-cell therapy' has achieved this - T-cells are taken from a patient's blood, 're-programmed' to recognise cancer by genetic engineering and infused back.

CAR T-cell therapy has been extremely successful in treating children with a type of leukaemia called B-acute lymphoblastic leukaemia (B-ALL) who have failed to respond to chemotherapy. In studies in the US, CAR T-cell therapy has resulted in almost 100% response, with about 60% of patients staying in remission. At UCL, we have a study called CARPALL open at GOSH, which is showing similar results.

About 15% of children and 25% of adults with ALL have T-ALL rather than B-ALL. Although these leukaemias are related, T-cells have different proteins on their surface and this means current CAR therapy cannot be used. T-ALL is a relatively rare disease and there are no new treatments on the horizon. Patients who have failed standard therapy have a poor prognosis despite very intensive chemotherapy and bone-marrow transplants.

We have found a target protein on T-ALL cancer cells against which we propose to develop a CAR therapy so that children and adults who have failed standard treatment can receive CAR T-cells. Since B-ALL and T-ALL are related, we believe there is a good chance that this CAR therapy will work in T-ALL. Given our experience - nine open CAR studies at UCL, we are confident this funding will lead to a phase I study.

Cell/Gene therapy for HSCT: Gene edited anti-CD3 chimeric antigen receptor T-cells

Cell therapy is a therapy in which cellular material is introduced into a patient. Generally, intact, living cells are introduced to the patient. For example, T-cells that are engineered to target disease-causing cells, such as cancer cells, may be introduced to a patient. In this way, cancer cells may be destroyed. We have engineered a T-cell that can target and kill other T-cells (which may cause disease), but importantly would not be able to recognise other engineered T-cells.

Characterisation of altered epigenetic cell states following loss of CREBBP function and their role in the induction and treatment of ALL

Acute lymphoblastic leukaemia (B-ALL) is an aggressive malignancy of immature immune B-cells and remains the leading cause of cancer-related death in children. Mutation of the CREBBP gene has been implicated in high-risk and relapsed B-ALL and drives several other cancers, indicating a fundamental role in cancer progression and treatment resistance.

CREBBP controls which genes a cell can turn on and off. We propose that CREBBP mutation “re-wires” immature B-cells into a pre-leukaemia that is fertile ground for the secondary mutations that drive B-ALL. We are developing a mouse model of CREBBP-mutated B-ALL to understand how this re-programming contributes to BALL initiation and evolution, and are using tractable human B-ALL cell lines to screen and test for new drugs that counteract CREBBP mutation. We hope to identify a number of compounds for preclinical testing in high risk B-ALL, and perhaps more broadly in other B lineage malignancies.

The patient samples requested in this application are key to pre-clinically proving the efficacy of the candidate compounds we identify before use in human patients. Given the long lead-time time required to fully characterise these patient samples and to generate the patient-derived xenografts (PDXs) models from them, we are approaching the Cell Bank now. We expect our drug and genetic screening approaches to identify candidate compounds by autumn 2020 and we will aim to test these drugs in the PDX models generated herein by early 2021.

Characterisation of sub-set of patients

Understanding the genetics of leukaemia has led to improvements in treatment for certain groups of patients. One set of genes, the immunoglobulin genes, are important in lymphoma and multiple myeloma, where they have been studied in detail. A unique group of patients with acute lymphoblastic leukaemia have been identified with abnormalities of these genes. We plan to characterise these abnormalities and relate them to the other genetic changes and the clinical features in ALL compared with other blood cancers. Understanding how these genes function may define a new group of patients in which modified treatment may improve survival.

Characterisation of the gene expression profiles in IGH translocated B-cell precursor acute lymphoblastic leukaemia

Leukaemia is a type of cancer that leads to the uncontrolled accumulation of blood cells. Acute lymphoblastic leukaemia (ALL) is a subtype of leukaemia and is caused by genetic defects in the patient's DNA. How the alterations in our genes lead to the development of leukaemia is not fully understood. We identified genetic defects in patients with ALL that result in the swapping of material between two chromosomes (translocation). We believe that the genes involved in these translocations contribute to the development of leukaemia and propose to study their role. We will assess the similarities and differences that may be present by looking at the levels of gene expression amongst these patients. We will investigate whether we can begin to subgroup these patients according to which signalling pathways may be disrupted. The data generated from this analysis will inform future experiments investigating specific drug combinations that may benefit these patients.

Characterisation of the genomic landscape of cytogenetically undefined B-cell precursor acute lymphoblastic leukaemia (BCP-ALL)

Clonal transcription and leukaemia stem cell heterogeneity in paediatric MLL rearranged and CBF acute myeloid leukaemia (AML)

Leukaemia is a type of blood cancer. One of its common types is called acute myeloid leukaemia (AML). This type of leukaemia occurs both in adults and children and causes most of the leukaemia related deaths. Although this disease has different biology in adults compared to children, everyone is treated similarly. However, the disease comes back in 30-40% of children with AML. We know a lot more about how to better treat the disease in adults compared to childern. It was shown in many studies that cells that start the disease and continue to exist following treatment are called 'Leukaemia Stem Cells' of LSCs. These cells are considered the mother of all cancer cells in AML. They don't disappear following treatment and makethe disease come back. LSCs have specific tags on them, and inside them, that make them different to the rest of the cells in the patient, making them more dangerous. So, it is important to know these marks so that we can make more effective treatments. in our lab we have found a new marker on LSCs in children with AML called CD180. This is a protein, and we want to find out more about it, and see if we can use it to kill AML cells.

So we aim to study CD180 on the LSCs in children who suffer from AML. As LSCs vary greatly among each other, we will focus on studying the differences in their biology. Next, we will study the ways the LSCs survive after chemotherapy and how they make the disease come back. We aim to study CD180 on LSCs and find out if it marks cells that survive treatment. Our study will provide biological information on LSCs and CD180 in AML in children. We plan to use this information to design better therapies that will be relevant and important for children suffering from AML.

Contribution of mitotic defects to the origin of aneuploidy in childhood B-cell acute lymphoblastic leukemia

B-cell acute lymphoblastic leukemia (B-ALL) is the most frequent cancer in children. It is characterized by an uncontrolled proliferation of blood cells in the bone marrow. One third of childhood B-ALL show an abnormal number of chromosomes, a phenomenon known as aneuploidy. Aneuploid B-ALL cases with can show abnormally higher or lower chromosome numbers than normal cells, and are classified as hyperdiploidy or hypodiploidy, respectively. Aneuploidy is a consequence of an abnormal cell division in the blood cells. However, next to nothing is known about its causes and consequences. We have recently reported that hyperdiploid leukemic cells have defects in the cell division process, called mitosis. Moreover, artificial disruption of these processes in cells with normal chromosome numbers, generated hyperdiploid cells resembling those in B-ALL samples, together with hypodiploid cells, thus suggesting that both abnormalities may arise by a similar mechanism.

Here, we plan to study the cell division process in hypodiploid B-ALL, which has a particularly poor prognosis, in order to assess whether it shares a common pathogenic mechanism with hyperdiploid B-ALL. We plan to perform a detailed characterization of the cell division process in these leukemic cells using the most advanced cell biology and biochemistry techniques, and high-resolution microscopy of living cells to record the defects of these cells when they divide. Finally, this project aims to generate the first in vivo models of aneuploid leukemias in mice that will be crucial to understand its origin and development, thus facilitating the development of new treatment strategies.

Creation of CEBPE/IGH cell line as a means to understand the role of CEBPE overexpression in acute lymphoblastic leukaemia

Acute lymphoblastic leukaemia (ALL) has its origins in the early stages of immune B-cell development. Genetic studies of ALL have identified regions of DNA harbouring risk factors that inherited from parent to child, including on chromosome 14 at the gene CEBPE. Some ALL tumour cells also possess a large defect in their DNA on chromosome 14 also involving CEBPE  that occurs spontaneously in immune cells but is not inherited. However, the role of CEBPE in ALL is currently unknown. The objective of this project is to take tumour cells from an ALL patient who possessed such a large DNA defect influencing CEBPE and use this as a model system to investigate how CEBPE promotes cancer development.

Deciphering chemotherapy-assisted neurotoxicity in childhood ALL using a deep phenotyping-genotyping approach

Acute lymphoblastic leukaemia is a type of blood cancer and is the commonest cancer to affect children. It is an increasingly curable disease with overall survival exceeding 90%. Unfortunately, the chemotherapy required to achieve this cure is very toxic. Currently the risk of death from treatment side-effects is nearly equivalent to the risk of death from leukaemic relapse (recurrence) for the 50% of children with “good-risk” disease. This has led to increasing focus on understanding the causes of treatment-related toxicity. In this project we comprehensively address one such side-effect - acute neurotoxicity - using a worldwide cohort of cases to achieve in-depth analysis of the clinical features, risk factors and any genetic associations. We are using Cellbank samples collected from UK patients to look for any underlying genes that might increase a child's risk of getting neurotoxicity during treatment. In this way we hope to understand why only some children are affected and develop ways to identify and implement preventative strategies for those at risk.

Defects in foetal B-lymphoid development and in utero origins of Down syndrome associated childhood acute lymphoblastic leukaemia

There is growing evidence that many childhood leukaemias start to develop in foetal life. Immature blood cells in the foetus, called progenitor cells, may develop abnormalities that make them more susceptible to events after birth that transform them into rapidly proliferating leukaemic cells. This secondary event or ‘second hit’ required to develop childhood leukaemia may happen several months or even years after birth. Children with Down syndrome (DS) have an increased risk of developing acute lymphoblastic leukaemia (ALL) and we believe that this is driven by the presence of an extra copy of chromosome 21 (trisomy 21; T21) in all the cells (including blood progenitor cells). We also know that many children without DS who develop leukaemia have T21 in the leukaemic cells. Therefore I would also like to examine DS fetal BM progenitors to try and understand how T21 may render progenitors more susceptible to leukaemic transformation. Indeed T21 may be one of the ‘first hits’ in childhood leukaemia. The leukaemia that children with DS develop is biologically different from other childhood leukaemias that occur in children without DS; for example infant ALL (ALL developing in a child <1 year of age) is very rare in the DS cohort. We therefore want to compare the properties of fetal DS progenitors with progenitors from DS-ALL and non-DS ALL. We expect DS fetal progenitors to be similar to those from DS-ALL and the characteristics that differentiate them from non-DS ALL may help us understand the origins of other childhood leukaemias, especially infant ALL.

Detailed investigation of DNMT3 mutation in specific patient

Detailed genetic studies on samples from a specific patient with a very rare mutation - DNMT3 - will be carried out.

Determining inherited genetic predisposition in late relapse of T-ALL

Acute lymphoblastic leukaemia (ALL) is the commonest cancer in children, occurring in approximately 400 children each year in the UK. Although most children can be cured, approximately 15% will experience relapse of the disease. Relapse occurs slightly more often in a subtype of ALL called T-ALL. Usually relapse recurs during treatment but a small number of relapses occur much later. Recent work indicates that some of these late relapses may actually be a completely new leukaemia with no relation to the previous leukaemia. We believe that this suggests the child has an inherited risk factor that makes them particularly prone to getting T-ALL. We therefore plan to study cases of late relapse to see if patients carry an underlying genetic risk.

Developing leukaemic biomarkers to enable personalised CNS-directed therapy

In this project we hope to develop highly accurate tests for leukaemia that has spread to the brain. Ultimately, this work should produce better, gentler treatments with a personalised amount of chemotherapy tailored to each child's specific risk of leukaemia recurrence in the brain.

Acute lymphoblastic leukaemia is the commonest childhood cancer. Great treatment advances mean that more than 90% of children now survive, although many suffer significant side-effects from treatment. One particularly challenging area is how best to treat leukaemia that has spread to the brain (the central nervous system, CNS). Current tests for CNS leukaemia are crude and do not predict CNS recurrence (which is often incurable), so all children receive large amounts of CNS-targeted chemotherapy to try and prevent this. This treatment is unpleasant and has serious side-effects in some children, including seizures and reduced IQ. Better predictive test are urgently needed in order to identify children that can safely receive less intensive treatment and/or identify those at high risk of leukaemia recurrence in the CNS. Our current tests for CNS leukaemia rely on detecting leukaemia cells free-floating in the fluid around the brain (cerebrospinal fluid, CSF) using a microscope. The problem with these tests is that the majority of leukaemic cells are stuck to the protective cell layers that surround the brain (the meninges), so the free floating cells are likely to represent only a very small proportion of the overall amount of leukaemia in the CNS. New technologies can detect tiny amounts of genetic material released from cancers into body fluids (cell-free DNA) and measuring this cell-free DNA can indicate the presence of cancers before they become visible. In addition, it has been shown that levels of cell-free DNA rise prior to clinical relapse of tumours providing an early warning system for impending recurrence. We will investigate whether we can use cell-free DNA, measured in CSF, to determine the level of CNS leukaemia and also to track how rapidly it responds to treatment. In addition, work in our laboratory has identified a number of small molecules (metabolites and proteins) in CSF that seem to indicate the presence of leukaemia, we now want to test how accurate these are on larger numbers of samples. These new tests have a lot of theoretical advantages over existing tests and are likely to be much better at detecting submicroscopic amounts of leukaemia. Knowing how much CNS disease is present, and how quickly it responds to therapy, should allow us to determine exactly how much treatment each child needs to completely eradicate all the leukaemia in their brain.

Development of an Artificial Intelligence tool for assessing the prognosis of children affected by Chronic Myeloid Leukaemia

Chronic myeloid leukaemia (or CML) is a type of blood cancer that is characterised by the presence of the aberrant protein BCR-ABL1. As most of the cancer cells need BCR-ABL1 to survive, using targetted drugs against this protein called tyrosine kinase inhibitors (TKI) is very effective at controlling the disease. However, not all the cancer cells die with the treatment and when the treatment stops, these cells have the capacity to start the leukaemia all over again. Thus, CML patients need to get TKI treatment for the rest of their lives. While TKIs are relatively safe for adult patients, for whom they were developed, they can affect normal growth and delay puberty in children, which can have a severe effect on the children's lives. Also, CML is usually more aggressive in children adults but current stratification methods do not take this into account. Thus, we propose to study paediatric CML cell-by-cell in order to develop a stratification algorithm using artificial intelligence to help clinicians choose the best treatment for children with CML, which could range between a lower dose of TKI to reduce treatment side effects or an early curative bone marrow transplant to improve prognosis.

Development of magnetic micro and nanostructures for the application of cell treatment and phenotypic profiling of cell interaction

This project aims to identify a novel signature able to distinguish leukaemic cells from the normal ones. Cancer cells will be specifically targeted with magnetic nanoparticles able to disrupt their membrane, provoking cell death without harming the healthy cells.

Development of pluronic micelle nanocarriers for the treatment of childhood acute lymphoblastic leukaemia

Treatment for children with acute lymphoblastic leukaemia (ALL) is very toxic and can be life threatening, so intensifying current therapies is not a realistic method of improving survival. A naturally occurring drug, Parthenolide, is particularly effective at killing leukaemia cells without harming normal blood cells. Unfortunately, Parthenolide is not soluble in water and so cannot be used to treat children. We will develop new preparations to improve Parthenolide delivery and make it more effective against ALL cells. We will also investigate whether these delivery methods can be used to reduce drug doses to alleviate problems associated with children being over-treated.

Dissecting heterogeneity in paediatric Acute Myeloid Leukaemia stem cells toward the development of effective CAR-T cell strategies.

Current therapeutic approaches for paediatric Acute Myeloid Leukaemia (AML) result in treatment failure or disease recurrence in about 30% of patients, who will unfortunately face a very poor prognosis. New therapeutic approaches are therefore urgently needed for these patients. One recently developed therapeutic approach takes advantage of patients' own immune cells which are isolated from the pateints' blood, genetically engineered to better recognise and fight leukaemia cells and then re-infused back in patients. Such approach has provided unprecedented results in the treatment of other types of blood cancers but has not been successfully applied to the treatment of AML, yet. AML represents a bigger challenge for the application of such an approach as AML cells present very similar characteristics to their physiological counterparts in normal blood. To avoid the risk of unwanted toxicity and unspecific targeting of normal haematopoietic cells, specific therapeutic agents (able to discriminate between normal cells and leukaemic cells) will need to be identified. I here propose to apply 'single-cell' approaches, which allow to evaluate 'one cell at the time' to dissect normal and malignant cells isolated from the same individuals. Such unprecedented resolution on leukaemic cells will be applied to identify specific features of the disease, including novel candidate therapeutic targets that might be applied for the successful treatment of resistant paediatric AML.

Dissecting regulatory functions of transposable elements in paediatric acute myeloid leukaemia

The genome is the complete set of our genetic material, consisting of DNA. Genes make up only a very small fraction of our genome. The majority of the genome is a vast landscape of repetitive sequences, some of which have the extraordinary capacity to autonomously self-replicate and move within the genome. These ‘mobile elements’ are called transposons. Transposons were initially dismissed as being “junk DNA”, without any function. However, we now know that transposons can control gene activity helping to define which genes are turned on or off and therefore have the ability to modulate biological processes.

Through their potential to control gene activity, transposons may have a role in cancer development, including in Acute Myeloid Leukemia (AML), which affect both adults and children. Studies in AML to date have been limited to the non-repetitive parts of the genome. The contribution of transposons to childhood AML development remains unexplored.

This project will test the roles of transposons in a particular type of childhood AML, which presents a poor clinical outcome in patients. We will test whether transposon’s ability to switch genes on and off contributes to the properties of AML as it presents in patients. Importantly, for the first time we will explore whether transposons could potentially be deactivated as a potential tumor suppressive strategy in cells taken directly from AML patients.

Exploring the genomic landscape of patients with acute lymphoblastic leukaemia who have a poor response to induction therapy.

Leukaemia is caused by defects in genes that control the way blood cells behave and operate. Over the past five decades scientists have been using a variety of techniques to investigate specific genes or sets of genes. As a results many gene defects are used to help diagnosis and treat children with leukaemia. A new technique called whole genome sequencing allows scientists to look not just at a few genes, tens of genes or even hundreds of genes but at all genes (~25,000) in one step. In this project, we will be looking at how whole genome sequencing can be used in real time to help diagnosis children with leukemia and also to identify gene defects that cause some leukaemias to be resistant to standard therapy. 

Exploring the role of the tumour suppressor ATRX in MLL leukaemia

For a cancer to occur it is known that certain changes, or mutations have to take place in a cell's DNA. In certain leukaemias a key change that leads to the development of cancer is a rearrangement of a particular gene called MLL (MLLr). This subset of leukaemias is difficult to treat and often relapse. To date little is known about how MLLr occurs. We have recently identified a protein (ATRX) that is absent in MLLr leukaemias, and this project will aim to explore whether loss of ATRX could be important in the formation of MLL rearrangements and thereby leukaemia initiation and/or maintenance. Understanding this process may give important clues that can be exploited for the develpment of future therapies. We will aim to determine whether ATRX normally prevents the formation of fusions at the MLL gene, suggesting its loss is required for the initiation of the cancer.

Expression of Rho GTPase family members in T-acute lymphoblastic leukaemia

Generation of B-ALL PDXs for the in vivo validation of LGR5-targeting immunotherapy

Acute lymphoblastic leukaemia (B-ALL) is an aggressive malignancy of immature immune B-cells and remains the leading cause of cancer-related death in children. LGR5 is a stem cell related surface protein that is specifically and highly expressed on preB cells, including in a proportion of B-ALL cases. We have been approached to test a novel antibody-drug treatment that targets leukaemia cells expressing LGR5, with the potential for specifically eradicating the most resistant cells. Given the success of other immune therapies in B-ALL we are hopeful that this could rapidly translate to clinical benefit for our patients. 

Genetic abnormalities in predicting relapse and post relapse survival in acute lymphoblastic leukaemia

Leukaemia, like other cancers, is a genetic disease. The discovery and characterisation of abnormal genes in the leukaemic cells of children with leukaemia has provided scientists and clinicians with a wealth of information about this disease. Currently many of these abnormal genes help in the diagnosis and treatment of patients. In particular, they can be useful in identifying patients who are at the highest or lowest risk of relapse. However, we do not fully understand the biological or clinical impact of many abnormal genes and there are likely to be abnormal genes that have yet to be discovered. The purpose of this study is to use novel and high-throughput technologies to comprehensively characterise a large cohort of patients for the presence of known abnormal genes and to identify novel abnormal genes. The data produced by this project will be analysed in the context of a modern clinical trial so that the results can be rapidly translated to optimise the treatment and outcome of future children with leukaemia.

Genetic characterisation and therapeutic targeting of paediatric mixed phenotype acute leukaemia

Genetic definition of the BCR-ABL1-like subgroup in acute lymphoblastic leukaemia

Genetic predisposition for venous thromboembolism (VTE) in patients with ALL

The outcome for patients with ALL has improved with intensive chemotherapy. However, a plethora of treatment related side effects is observed. Current treatment protocols are adapted to reduce toxicity whilst preserving survival. One significant, life-threatening toxicity is venous thrombo-embolism (VTE). We are working together with our Australian collaborators to develop an accurate risk model to predict VTE in patients undergoing treatment for ALL. 

The proposed risk model includes clinical, laboratory and genetic risk factors. This application is seeking funding for validation of 4 potential genetic risk factors identified in the Australian study. In addition, we will repeat the statistical analysis across the combined Australian, NOPHO and UK cohort to look for new genetic risk factors.

A validated risk prediction model can in the immediate future be used to identify patients at high risk for VTE to be able to offer thromboprophylaxis and prevent VTE. The next Australian study aims to test the validity of the current model and offer thromboprophylaxis to high risk patients. We are keen to help increase the accuracy of this model by validating genetic risk factors, with a view to applying this model in UK ALL trials soon.

Genetic polymorphism in the PAI-1 gene and risk of osteonecrosis in patients with acute lymphoblastic leaukaemia treated on ALL2003

Some children who are treated for leukaemia develop a side-effect of treatment called osteonecrosis; this affects their bones and in its most severe form can cause very painful joint problems. We are investigating whether some children may have a genetic predisposition to developing this side-effect. If so, we may be able to offer them better, more effective treatment in future.

Genetic regulation of normal and aberrant haematopoiesis

Haematopoiesis is the formation of blood cells. All cellular components of the blood are derived from stem cells that have the ability to become any type of cell in the blood system. Leukaemia is a cancer of the blood and bone marrow, characterised by having too many white blood cells. This is thought to arise when there is a change in the behaviour of genes that normally regulate the development of blood cells. We and others have identified a number of such genes. To understand  exactly what they do, we wish to examine their behaviour in blood stem cells. The umbilical cord blood is a rich source of these stem cells and that is why we wish to obtain this from the cord blood bank. The planned research will increase our understanding of how leukaemia occurs and progresses and to facilitate the development of new anti-leukaemic drugs.

Genetics of acute lymphoblastic leukaemia

Examining DNA from chidren with leukaemia will enable us to identify new risk factors for the disease providing fresh insights into why the disease develops. Furthermore, this information may also show why some children have different outcomes from their disease.

Genomic characterisation of IGH@translocations in B-precursor acute lymphoblastic leukaemia

Understanding the genetics of leukaemia has led to improvements in treatment for certain groups of patients. One set of genes, the immunoglobulin genes, are important in lymphoma and multiple myeloma, where they have been studied in detail. A unique group of patients with acute lymphoblastic leukaemia have been identified with abnormalities of these genes. We plan to characterise these abnormalities using technologies that sequence the patient's genome and allow us to find common abnormalities amongst this group of leukaemia patients. Understanding the genomes of these patients may define a new subgroup in which modified treatment may improve survival.

Genomic landscape of relapsed t(1;19) positive acute lymphoblastic leukaemia

Global methylation patterns in resistant and relapsed childhood acute lymphoblastic leukaemia and effect this may have on resistance

There has been considerable progress in the treatment of childhood leukaemia and in the UK approximately 80% of children with acute lymphoblastic leukaemia (ALL) can be cured with current chemotherapy. Leukaemia is treated with a combination of drugs designed to kill the leukaemia cells. Unfortunately some leukaemia cells are resistant to these drugs. This means that a significant number of children with ALL have disease which is resistant to treatment. One way to overcome this resistance is to study the biology of resistant leukaemia cells and use this information to develop new drugs specifically targeted to overcome the mechanism of resistance. The genetic material in resistant leukaemia cells is known to have some important differences compared to chemotherapy sensitive leukaemia cells. The expression of particular genes (whether they are turned on or off) in leukaemic cells can be related to how the disease responds to chemotherapy. The expression of some genes in leukaemia is controlled by a process called 'DNA methylation.' Our research is investigating the relationship between DNA methylation and the development of resistant disease in childhood acute leukaemia. We know that DNA methylation is a reversible process; therefore understanding how it is involved in resistant disease may lead to the development of new treatments for children with leukaemia in the future.

Hypoxia inducible genes in Acute Myeloid Leukaemia: Implications in chemotherapy and relapse

Currently, it is believed that the evolution of acute myeloid leukaemia (AML) is driven by stem cells that are 'dormant' in the bone marrow and because of this they are more resistant to drugs used in chemotherapy treatments leading later on to relapse of patients. This resting state in which the stem cells are found is believed to be due, at least in part, to the low levels of oxygen (or hypoxia) in the bone marrow activating various signals inside these stem cells. In this study we will determine: 1) If these signals of hypoxia are active in stem cells of the different types of AML; 2) If in relapsed patients these signals are more intense; 3) If modifying these hypoxia signals benefits the treatment and decreases the likelihood of relapse. This will allow us to understand the mechanisms by which the AML stem cells depend on these hypoxia signals and to design a new and more efficient therapeutic strategy to prevent relapse.

Identification of genetic variation influencing acute lymphoblastic leukaemia

Identification of progression genes in t(8;21) and t(12;22) leukaemias

Identification of recurrent patterns of gene expression in relapsed ALLR3 patients

Although approximately 85% children with acute lymphoblastic leukaemia are now cured, the disease will return (relapse) in around 15% of children, with only 30-40% surviving. This makes relapsed ALL the fourth most common childhood cancer. As treatments are now given at maximum doses, with toxicity-related deaths matching the number of deaths from disease, it is important that new drugs are identified. Outcomes for relapsed ALL patients could be greatly improved by the characterisation of their DNA to identify possible new drug targets. This would enable better treatment at diagnosis to include the use of targeted therapies, together with conventional chemotherapy, in those patients considered to be high risk. A new relapsed trial (IntReALL) aims to test several novel drugs for relapsed ALL and will be the largest ever clinical trial for ALL, involving 19 countries and more than 1500 children. The trial associated research project will perform key laboratory studies that include techniques to measure very low levels of disease and also monitor levels of proteins that may predict how well a patient will respond to a drug. In addition, relapsed patient DNA will be analysed to identify possible new targets for drugs. In addition, generate of a bank of primary material from IntReALL patients which will be pivotal in the testing of other novel drugs for this disease. Similar applications are underway in other IntReALL partners, which will result in a global laboratory effort to optimise the treatment for relapsed ALL.

Identifying candidate prognostic mutations in relapsed childhood acute lymphoblastic leukaemia

Acute lymphoblastic leukaemia (ALL) is the commonest childhood cancer. With modern treatment, over 80% of children are cured of this disease. In the remainder, the disease recurs or fails to respond to treatment. In the latter group of children, additional chemotherapy treatment is adminstered and sensitive bone marrow monitoring tests are used to decide which of these children should receive bone marrow transplantation. Using these sensitive tests, children who despite chemotherapy continue to have leukaemia in the bone marrow, are offered bone marrow transplantation. This study proposes to examine the DNA of leukaemic cells from children in the chemotherapy and bone marrow transplant groups to identify specific gene mutations that enable leukaemic cells to resist chemotherapy. Two methods of such testing will be compared. A conventional technique called DHPLC will be compared against a novel technology called 'DNA tagging for pooled populations'. The latter approach is potentially a more streamlined cost-effective technique for the detection of these gene mutations. In the longer-term, these disease-resistance mutations can potentially be tackled using novel targeted therapy that is rapidly becoming available. The findings from this study will be integrated into the forthcoming protocol for treatment of relapse in childhood ALL.

Identifying methylation markers in good risk acute lymphoblastic leukaemia

Acute lymphocblastic leukaemia (ALL) is the most common type of cancer seen in children. Treatment of ALL has improved dramatically in recent times and survival rates are now well above 80%. However, because it is more common than other cancer types in children, it is still one of the main causes of cancer related morbidity and mortality in children. One way in which current treatments can be improved is by identifying patients most likely not to be cured by standard treatment, who can then be given more intensive treatment which can lead to improved survival. However, it is not currently possible to identify all patients at high risk and in a significant number of patients regarded as low risk the disease can come back (known as relapse). Thus new methods are required to allow identification of these patients.

Identifying novel clinically relevant molecular aberrations in pediatric Down syndrome acute lymphoblastic leukemia using next-generation sequencing methods

Previous studies suggest that the genetic landscape of pediatric Down Syndrome Acute Lymphoblastic Leukemia (DS-ALL) patients is dissimilar from that of pediatric ALL patients without Down syndrome and that other genetic lesions may underlie DS-ALL and/or explain the adverse prognosis associated with DS-ALL. We therefore would like to propose to identify genetic alterations in DS-ALL patients with new sequencing techniques, including DNA and RNA sequencing, in order to identify more or other genetic aberrations in DS-ALL patients which may point to new ways to stratify and more effectively treat these patients. Functional studies will address the contribution of aberrant genes or genomic lesions to the survival capacity of leukemic cells.

Identifying the hierarchical lineage of leukaemia initiation in high-hyperdiploid pre-B acute lymphoblastic leukaemia

Acute leukaemias are the most well studied of all cancers mainly due to the ease of access and the relatively limited number of genetic alterations. Investigating monozygotic twins with concordant leukaemia provides a unique opportunity to analyze the early genetic steps of leukaemogenesis. Previous studies by Greaves and his colleagues revealed that several recurrent chromosomal aberrations, associated with specific clinical subtypes of acute lymphoblastic leukaemia (ALL) are present in both co-twins confirming their early (in utero) occurrence and implicating their role in the pathogenesis. However, the stage of the cell maturation process in which these aberrations emerge is poorly understood. Immunoglobulin (IG) and T-cell receptor (TCR) gene arrangements are hallmarks of B- and T-cell development thus can serve as points of reference at determination of the timing of initiation. Very limited data are available in this field with reported studies each including only twin pair or triplet. Our aim is to reveal the IG and TCR gene rearrangements in a group of monozygotic twin children with concordant high-hyperdiploid pre B-ALL and determine in which phase of the B-cell ontogeny the common pre-natal, pre-leukaemic cell occurred. We predict that twin children with concordant pre-B-ALL share some early but not all clonal IG and TCR rearrangements which allows us to determine the cell of origin precisely.

Identifying the T681I gatekeeper mutation in EBF1-PDGFRB Ph-like acute lymphoblastic leukaemia using droplet digital polymerase chain reaction (ddPCR)

Understanding the underlying mechanisms of resistance in Ph-like ALL is critical to find novel therapeutic strategies in the relapse setting. We first focussed on identifying resistance mechanisms in Ph-like ALL patients harboring the EBF1-PDGFRB rearrangement by using well-validated in-vitro saturation mutagenesis screens to predict the spectrum of drug resistant mutations. Our screens suggest that kinase domain (KD) point mutations may represent the primary mechanism of acquired resistance in EBF1-PDGFRB. The most common KD mutation identified was the T681I mutation, which is analogous to BCR-ABL1's T315I gatekeeper mutation. We further hypothesize that these KD mutations may pre-exist at diagnosis at a frequency that cannot otherwise be detected by Sanger, or even next generation sequencing. The primary objective of this study is to determine the presence of the T681I gatekeeper mutation in patients with EBF1-PDGFRB at diagnosis, end of induction, and at relapse by using a droplet digital PCR designed for rare mutation detection.

Immunophenotypic and genetic characterisation of leukaemic stem cells in childhood acute myeloid leukaemia

Investigating proteome re-wiring in leukaemic stem cells to reveal new targets in paediatric AML

Acute myeloid leukaemia (AML) is an aggressive cancer of the blood system, affecting a wide demographic of patients from the very young to the elderly. AML is the second most frequent type of leukaemia in childhood, accounting for almost 20% of childhood leukaemia. Despite improvements in treatment, the outcomes still remains poor with the average five-year survival rate ~44% in paediatric cases and adult cases have been estimated as low as 15% for two-year survival rates. The key reason behind the poor overall prognosis in AML is due to the failure of intensive chemotherapy to deplete cells at the origin of the disease, so called Leukemic Stem Cells (LSCs), which are responsible for relapsed disease. Whilst paediatric cases are eligible for intensive chemotherapy, and it can lead to a durable cure, there are also significant consequences associated with the toxic side affects, which are harsh and may lead to health complications in the future. The ability to discriminate between a healthy and malignant stem cell is key to eradicating cancer, whilst preserving healthy tissue. In recent years we have learnt much about the genetic elements governing cancer stem cell biology, but our understanding of the most functional molecules in cells, proteins, is still in it’s infancy. This work will study the repertoire of proteins, how they are regulated and how they maintain the cancer stem cells in paediatric AML, at the onset of disease, during chemotherapy and when patients relapse. By studying the protein biology of AML, we will shine a light on new therapeutic targets and begin to design new, more effective therapies aimed at targeting paediatric cancer with minimal off-target effects.

Investigating the role of genetic variation in risk of childhood acute lymphoblastic leukaemia in Down syndrome

Investigating the role of RUNX1-mediated Polycomb recruitment in RUNX1-RUNXIT1 leukaemias

RUNX1 is a protein crucial for haematopoiesis during development and is one of the most common targets of genetic alteration in human leukaemia; thus, understanding the molecular mechanisms of how RUNX1 functions is important. We have discovered that a key function of RUNX1 in leukaemia is to recruit Polycomb proteins. Polycomb proteins are repressors of genes. Inappropriate activation or repression of genes is a common driving force for many leukaemias. Mutations in Polycomb proteins are relatively enriched in leukaemias with mutations of RUNX1, and leukaemias with this combination of mutations remain an unmet clinical need. Understanding the interplay between RUNX1 and Polycomb proteins can give us insight into inappropriate gene regulation in these leukaemias, and may help us predict why some patients respond better to treatment than others.

Investigation of the effects of IKZ1 deletions on expression of Gpr132 in Philadelphia positive acute lymphoblastic leukaemia

Sometimes leukaemia cells have deletions of important genes, and this may affect the way the cells respond to chemotherapy. We are interested in the effects of deletions of a gene called Ikaros, which appears to make leukaemias harder to treat. Following several years’ work in our lab, we now believe that loss of Ikaros causes increased levels of a protein called G2A, which causes the leukaemia cell to divide abnormally. We will test whether the gene which makes the G2A protein is more active in leukaemia cells which don’t have Ikaros. If this is true, it will help us to understand how we might improve the treatment for children with these types of leukaemia.

Investigation of the role of JAK2 mutations in the pathogenesis of acute lymphoblastic leukaemia in children with Down syndrome

Recent advances in our understanding of how leukaemia develops have shown that most cases of childhood leukaemia are initiated by chromosome changes before birth, but require additional mutational alterations to progress to full clinical leukaemia. Several acquired genetic mutations are now known to be present in the leukaemic cells of a significant proportion of Down syndrome (DS) children with ALL. The same genetic aberrations are also found in non-DS ALL, although at a lower incidence. Our aim is to determine which of these genetic events occur early and have the potential to initiate or cause progression of the disease. To do this we are using new technologies to look at individual, single cells for the presence of multiple genetic rearrangements. We are also using neonatal blood spots (Guthrie cards) of DS-ALL children to see whether we can ‘backtrack’ specific genetic abnormalities to birth, several years before they develop leukaemia.

Investigation of the role of the non-coding genome in relapsed T-acute lymphoblastic leukaemia.

Acute lymphoblastic leukaemia (ALL) is the commonest cancer in children, occurring in approximately 400 children each year in the UK. Although most children can be cured, appoximately 15% will experience relapse of the disease. Relapse is extremely difficult to treat, particularly in those with a subtype called T-ALL, and new treatments are urgently required.  

We plan to use state-of-the-art next generation sequencing techniques to analyse samples taken from patients with relapsed T-ALL to identify the genetic changes that cause this type of leukaemia. This will allow us to design novel therapies that target these key genetic drivers, which will hopefully result in better outcomes for these patients. 

Legacy thiopurine study samples

This study will attempt to validate recent results involving prediction of risk of relapse.

Leukaemogenesis by the CALM-AF10 fusion protein: Development of a NOD-SCID mouse model

Leukaemias carrying CALM-AF10 fusions are associated with a poor outcome on current therapy. Ths project has the potential for unravelling the mechanisms by which this rearrangement causes leukaemia. In addition, this project seeks to identify pathways that may be amenable for targeted treatment, as well as create suitable in vivo models where such therapeutic molecules can be tested.

Lymphocyte development in Down syndrome

Down syndrome is the most common genetic condition in the UK affecting 750 newborn babies every year. Children with Down syndrome have an increased risk of developing acute lymphoblastic leukaemia (ALL) which originates from transformation of the progenitor cells which give rise to B lymphocytes. The sequence of B lymphocyte alterations and the exact events which cause it in Down syndrome remain largely unclear. We are therefore investigating this by studying the effects of genes important for normal B-cell development in healthy children and children with Down syndrome.

Molecular and functional characteristics of leukaemia-initiating cells in MLL-AF4 positive infant acute lymphoblastic leukaemia

There is growing evidence that many childhood leukaemias start to develop in fetal life. Immature blood cells in the fetus may develop abnormalities that make them more susceptible to transform into rapidly proliferating leukaemic cells. One such primary hit is a re-arranged gene called MLL (MLLr). Some MLLr leukaemias develop very early (<12 months of age - infant leukaemia), and these children tend to do poorly even with intensive treatment. They differ from MLLr leukaemias that present later in life by being inherently more aggressive and difficult to treat. I would like to investigate why infant MLLr leukaemia is different biologically and whether this is because of unique fetal cells that undergo the 'first hit'. To do this I shall examine the progenitor cells from fetal bone marrow samples and compare their characteristics and function with that of infant and non-infant MLLr leukaemia cells. This will help us identify unique pathways that initiate the development of infant leukaemia and ultimately design novel therapies to treat this refractory leukaemia.

Molecular characterisation of intrachromosomal amplification of chromosome 21 (iAMP21)

iAMP is arguably the most important molecular prognostic marker for patients with B-lineage acute lymphoblastic leukaemia (ALL) identified in recent years. In 2003, we were the first to define a subset of patients with this abnormality. Since then we have been working towards its full characterisation which will ultimately expand our understanding of the genetic mechanisms driving leukaemogenesis and provide accurate methods of detection. We propose further molecular profiling of these patients to identify recurrent genomic regions. Sophisiticated molecular methods will be used to assess the gene content of aberrant genomic regions. The effect of these changes on gene and protein expression will be investigated and will initiate experiments to assess the potential role of these changes in leukaemogenesis. Retrospective analysis of a large patient cohort will allow correlations with prognosis to be determined.

Molecular characterisation of intrachromosomal amplification of chromosome 21 (iAMP21) - further studies

In childhood acute lymphoblastic leukaemia, chromosomal abnormalities are often used to predict prognosis and in risk stratification for treatment. About 10 years ago, we identified an abnormality known as iAMP21 which was associated with a very high risk of relapse on standard treatment. Now, with intensive therapy, the outcome has improved dramatically. In order to consider less toxic personalised therapy for these patients, we need to understand the way in which this abnormality drives leukaemia. We have made significant progress in this area but data are based on historical samples. With samples from a new cohort of patients using state-of-the-art techniques, we are confident that those genes responsible for the development of leukaemia in these patients will be identified to provide more accurate diagnosis to identify those who require intensive treatment.

Molecular characterisation of the deregulation of CRLF2 in childhood B-cell precursor acute lymphoblastic leukaemia

Overexpression of genes is a frequent and important molecular marker in patients with B-lineage acute lymphoblastic leukaemia. We have identified overexpression of a gene in patients where material is exchanged between the sex chromosomes and chromosome 14, bringing this target gene under the control of a potent enhancing element leading to gene dysregulation. We have identified a deletion of sequence to the target gene which also leads to increased expression. We are in the process of characterising these aberrations, which will ultimately expand our understanding of the genetic mechanisms driving leukaemogenesis and provide accurate methods of detection for what may be a poor-risk genetic abnormality.

MRD monitoring of children with Philadelphia positive acute lymphoblastic leukaemia receiving Dasatinib

A rare subset of childhood acute lymphoblastic leukaemia is characterised by the presence of an aberrant chromosomal translocation. This is called the Philadelphia Chromosome and patients with this subtype have a poor outcome and have been usually transplanted. The translocation results in the increased expression of the ABL gene. The advent of drugs such as Imatinib, which suppress ABL has dramatically improved outcome of these patients. In this trial, we propose to investigate the benefit of an even more potent ABL inhibitor called Dasatinib. We propose to monitor the level of disease using molecular tools that can detect 1 malignant cell in 10,000-100,000 cells. We think almost half will have disease levels lower than this and can be cured with chemotherapy i.e without transplantation.

Mulitplex screening for interacting compounds in acute leukaemia (MuSICALstudy)

Leukaemia is the most commonly occurring cancer among children and young people (30%). Acute myeloid leukaemia (AML) accounts for 20% of cases, with a poor overall survival of 60% and a treatment landscape that has remained largely unchanged for four decades for paediatric and adult AML. Data now shows that there is a need for age-specific treatments. We have used a novel bioinformatics approach combined with large-scale drug-screening to identify all possible pair-wise combinations from 80 drugs. One of the novel candidate drug combinations identified appears to only target paediatric AML harbouring a particular mutation associated with a poor prognosis in paediatric patients. The next steps involve testing this combination in primary paediatric AML patient material alongside normal bone marrow samples to ensure this combination only reduces cell viability of the AML blast cells.

Novel therapies for TP53 mutated paediatric malignancies

The tumour suppressor gene, TP53 is the guardian of the human genome, providing protection from the development of cancer by correcting any faults that occur in the body during cell division.  The TP53 gene is one of the most frequently inactivated/mutated genes in childhood cancer and is associated with a poor prognosis in many paediatric tumour types, such as leukaemia, neuroblastoma and the brain tumour, medulloblastoma.  Current therapeutic approaches are failing to successfully treat this population of children with TP53-mutant tumours and new approaches are now urgently required.

This project will establish a biologically defined cohort of TP53 mutant tumours across a spectrum of childhood malignancies (acute lymphoid and myeloid leukaemia, lymphoma, neuroblastoma and medulloblastoma). This unique cohort will be utilised firstly to describe the nature and spectrum of TP53 defects and secondly, collaborative matched datasets, investigating chromosomal changes, whole genome aberrations and gene expression levels, will be interrogated to identify any common biology amongst these TP53-mutant tumours.  Importantly these investigations will advance our knowledge and have the potential to identify future therapeutic targets which will ultimately improve outcomes in this difficult to treat group of tumours.

Optimisation of dexamethasone therapy in childhood acute lymphoblastic leukaemia

This study is investigating whether small changes in DNA may alter the way that children with leukaemia process the medicine dexamethasone. We would like to see whether these changes in DNA may mean some children get a higher exposure to dexamethasone, and whether this, in turn, means they experience more side effects.

Optimising novel therapies in the ALLtogether trial using Mass Cytometry

The aim of our project is to investigate different drug targets within childhood acute lymphoblastic leukaemia (ALL) samples at different time points. These will include newly diagnosed samples but also samples from children whose leukaemia is not responding well to conventional therapy and are at high risk of relapse. The aim is to identify new drugs that are not used as standard but may enable improved treatment of ALL.

Pilot project to study the expression of the meiotic regulator PRDM9 by childhood leukaemias

Abnormal numbers and rearrangements of chomosomes are a typical feature of childhood leukaemia and contribute to disease causation; understanding their causes could aid disease treatment and prevention. PRDM9 is a protein important for correct chromosome exchanges during gamete production. Rare forms of PRDM9 are more common among children with leukaemia and there is evidence that it is expressed in non-gamete cells. It is unknown whether PRDM9 is expressed in leukaemia cells, or if it contributes to chromosomal abnormalities causing leukaemia. We will determine which forms of PRDM9 are expressed in leukaemias and whether expression is absent from normal blood cells.

Pilot study to assess the feasibility of utilising samples from CellBank for whole genome sequencing as part of the 100,000 genome project

Genomics England (GEL) are responsible for delivering the 100,000 genome project on behalf of the Department of Health. This is an NHS transformation programme aimed at developing whole genome sequencing (WGS) capacity for clinical diagnostics. GEL Clinical Interpretation Partnerships (GECIPs) have been set up in order to coordinate sequencing activity with different disease subgroups. The Haematology Oncology GECIP domain covers all activity within this field and includes a childhood leukaemia sub-domain which is led by Professor Moorman. The Haematology Oncology GECIP is proposing to undertake WGS on approximately 4000 patients with the overall goal of deriving new prognostic and predictive biomarkers and ultimately driving the development of new therapies.

Resolving the genomic architecture of childhood and adolescent acute lymphoblastic leukaemia using long-read whole genome sequencing

Blood cancer (leukaemia) cells in patients with this disease have alterations in their DNA (known as genetic abnormalities) which make them different to normal blood cells. Over the past 30 years, scientists have discovered and characterised many genetic abnormalities. Patients newly diagnosed with leukaemia will undergo testing to determine the genetic subtype of their leukaemia. Importantly, doctors will use this information to decide the optimal treatment pathway for each patient. Current genetic testing is limited and does not cover all DNA (which is known as the genome). Recently whole genome sequencing has given scientists a much better view of the genetic abnormalities causing the disease. There are several different methods that can be used to perform whole genome sequencing and the current gold standard is called short-read sequencing but it only covers 90% of the genome. The focus of this study is to evaluate a new type of sequencing methodology called long-read sequencing. Long-read sequencing will enable scientists (a) to read the parts of the genome that short-read sequencing fails to cover, and (b) to characterise complex genetic abnormalities more precisely. Once evaluated, long-read sequencing may be a more useful diagnostic tool than short-read sequencing. In this study, we will perform longread sequencing on a cohort of ~400 children, adolescents and young adults with a specific type of blood cancer: acute lymphoblastic leukaemia treated on national clinical trials. This unique cohort has been carefully selected to address a number of important scientific questions about key high-risk genetic subtypes of acute lymphoblastic leukaemia that are difficult to cure. As the cohort is focussed on patients recruited to national clinical trials, we will be able to examine the clinical relevance of our findings. Finally, the scale of the project will enable us to assess long-read sequencing as a diagnostic test.

Retrospective investigation of tumour genetics in children with ETV6-RUNX1 positive ALL:  Ponte di Legno International Collaborative Leukaemia Project

ETV6-RUNX1 translocation (t(12;21)) is the most common genetic abnormality that occurs in childhood acute lymphoblastic leukaemia. In general, the presence of ETV6-RUNX1 translocation is a favorable factor and very few children relapse. However, much uncertainty remains on why these few children do not respond as well, which is why this study explores possible differences in tumor genetics between the two groups of children. If additional genetic abnormalities are found, future therapy can be adjusted earlier in the treatment process and lead to low relapse rates for all children.

Sensitivity of T-acute lymphoblastic leukaemia cells to JAK kinase inhibitors

Sequential acquisition and clonal architecture of genetic aberrations in paediatric T-acute lymphoblastic leukaemia

This project concentrates on identifying how genetic changes already known to be important in a particular type of childhood leukaemia – T-cell acute lymphoblastic leukaemia (T-ALL) - evolve and in what order. Earlier work from our group has shown that genetic evolution of cancer cells in other types of ALL (B-cell type rather than T-cell type) is non-linear and adopts a branching tree or ‘Darwinian’ evolutionary structure. Understanding of ‘cancer initiating’ events in this disease would allow the development of therapies specific to these early events which are likely to be more successful than therapies targeting events further down the evolving leukaemic branching tree. Our project is summarised in three phases. In phase 1 we will apply techniques we have already used in other subtypes of leukaemia to confirm the ‘branching tree’ model also applies in T cell ALL. We will aim to support our hypothesis that all leukaemic cells within each sample share a large genetic rearrangement within (intra) or between (inter) the chromosomes. These rearrangements are our hypothesised initiating event and are hypothesised to occur in all cells as they happen prior to the ‘branching’ of the tree. In phase 2 we will address a similar question for other genetic changes – subtle changes to the sequence of DNA, the so called ‘genetic code’ – to determine in what order these occur and what their relative importance is as a drug target. In phase 3 we will consider how to apply our new knowledge to the practical screening of diagnostic patient samples and extend it to exploring the ongoing evolution of the leukaemia cell branching tree in relapsed patients and what genetic alterations are important at this stage in the disease. If we have identified a key drug target we will liase within our organisation with the drug development team.

Studies to determine whether molecular markers predict response to therapy and can be used to stratify treatment

The majority of children with acute lymphoblastic leukaemia (ALL) can now be cured of their disease but certain subgroups of patients still have a relatively poor prognosis. One such subgroup is those patients with T-ALL, particularly those that have persistence of detectable disease (Minimal Residual Disease or MRD) at the end of induction therapy. In the last few years mutations have been identified in specific genes of some patients with T-cell ALL that may be associated with response to treatment and long-term outcome. We aim to screen samples from children with T-ALL who have been entered into the current national ALL trial for mutations in 3 genes called NOTCH1, FBXW7 and PTEN that are all involved in the production of T-cells. Presence of mutations will be related to early response to treatment and long-term outcome. We will also ascertain the outcome of patients with mutations in the NOTCH pathway who remain MRD-positive after induction therapy to try and refine the indications for intensification of treatment and thus reduce the number of children subjected to a bone marrow transplant procedure.

The aminotransferase system and nutrition in Acute Myeloid Leukaemia and Acute Lymphoblastic Leukaemia as a predictor of cytogenetics and potential therapeutic avenue.

Leukaemia is the 12th most common cancer in the UK, with patients often suffering severe side effects from therapy. Unlike some other forms of cancer, such as breast cancer, it is impossible to identify how a patient will respond to therapy. This makes treatment decision making challenging; for example, patients who are unresponsive may be advised to undergo aggressive therapy that is unlikely to work, or potentially even cause therapy related cancers. New avenues of research need to provide markers that accurately predict patient outcome and allow informed treatment regimes. Previous work demonstrated that a key protein, from a family of enzymes known as aminotransferases, was found in highly invasive tumours of the brain, with greater quantities of this protein associated with a much poorer prognosis for patients.

This study will build upon this research by looking at many different aminotransferase enzymes within leukaemic cells - correlating this with diagnosis subtype and genetic findings. We have already conducted preliminary investigations that have found that more aminotransferase is produced in chemotherapy-resistant leukaemic cell lines and that these cell lines have altered needs for amino acids. We will look more closely at how many different enzymes within this family are associated with leukaemic cells and how their presence is correlated with patient outcome and treatment response. These findings will contribute to an evidence base for informed treatment decision making, and novel diagnostic and treatment outcomes.

The characterization of previously undetected MLL gene abnormalities in infant ALL

Infant acute lymphoblastic leukemia (ALL) is a rare disease with a very poor prognosis. This is especially pronounced for infants with specific high risk features, most importantly rearrangements in a gene called mixed lineage leukemia (MLL) gene. Currently, testing for MLL rearrangements (MLL-r) is performed through fluorescence in-situ hybridization (FISH) analysis and karyotyping. We discovered two patients with high risk clinical features for whom traditional MLL investigations showed wild type MLL (MLL-WT), but RNA sequencing of these patient samples revealed abnormalities in the MLL gene, specifically MLL partial tandem duplications (PTD). MLL-PTD has not previously been described in infant ALL, but has been shown to be a poor prognostic feature in adult acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS), as well as in pediatric AML. False negatives in the detection of MLL events may have implications for these infants in regards to their prognosis, risk stratification, and in some cases, their treatment plan. Given our index cases, we hypothesize that there are other patients with infant ALL with MLL events that have not been detected by FISH and cytogenetic analysis. Using diagnostic samples from infants registered on the international Children’s Oncology Group (COG) AALL0631 trial, we will use next generation sequencing to determine whether other such cases exist. We aim to describe the specific MLL gene abnormalities discovered and establish their prevalence. Using the available outcome data for these patients, we will explore whether these previously undetected MLL abnormalities carry prognostic significance. In addition, we will aim to identify and compare their associated gene expression signatures. The existence of a cohort of infants with MLL abnormalities not detected by traditional methods but with similar poor prognosis would have important implications, demonstrating the need to change the diagnostic approach used in infant ALL for patients that are deemed to have absent MLL-r by current methods.

The CXCL8-producing T-cell: function in health and carcinogenesis

We have identified a novel type of T-cell - one of the white blood cells that help protect the body from infection - in human neonates. These cells are rarely found in adults but, if present, have markers suggesting they are naïve, newly produced cells from the thymus (the organ from which T-cells emerge). These cells are novel as they make a mediator called CXCL8. T-cell acute lymphoblastic leukaemia cells (T-ALL) arise from transformation of cells as they develop in the thymus. We have looked at a few T-ALL to date and these also seem to make a lot of CXCL8. Previously, it has been shown that T-ALL that make CXC8 seem to have a poor response to therapy. We want to test different T-ALL samples to see if those T-ALL that make CXCL8 have a particular profile and whether by blocking CXCL8 this may have an effect on the ability of the T-ALL to continue to grow.

The prognostic significance of genetic abnormalities in T-cell acute lymphoblastic leukaemia

In childhood B-lineage ALL, chromosomal abnormalities are linked to prognosis and are used in risk stratification for treatment. Novel high resolution techniques have uncovered new genetic abnormalities in T-ALL, particularly those involved in T-cell development, for which the prognostic significance is unclear. We propose to use two recently established techniques of MLPA and targeted next generation sequencing on the cohorts of childhood T-ALL from ALL2003 and ALL97 to conclusively determine whether these new abnormalities are associated with treatment response and evaluate their role in risk stratification of future treatment trials.

The prognostic significance of IZF1, other B-cell differentiation gene abnormalities and related genetic changes in childhood B cell precursor acute lymphoblastic leukaemia (BCP-ALL)

The role of CD33 and anti-CD33 therapy in childhood acute myeloid leukaemia

Relapse is the commonest cause of death in childhood acute myeloid leukaemia (AML). Multiple courses of chemotherapy is the mainstay of treatment but a ceiling of benefit has been reached and toxicity is significant. Relapse is due to the persistence of cells that grow slowly and can evade death by conventional chemotherapy, called leukaemic stem cells. Novel treatments that can target these cells are required. The aim of our project is to increase understanding of a protein called CD33 and novel agents that target CD33. This protein is present on the surface of AML cells and the leukaemia stem cells. We aim to identify new targets that will work in combination with the current anti-CD33 treatments available to create novel combination therapies.

The role of miRNAs in developing haematopoetic cells and infant leukaemia

Cancer development is in many cases caused by a dysregulation of gene expression due to the aberrant action of transcription factors, which are proteins responsible for the correct expression of genes. Recently, a further level of gene regulation has been described following the discovery of small RNA molecules known as microRNAs (miRNAs) which act as repressors of gene expression. Subsequent work has demonstrated that miRNAs are also key players in cancer development and progression, including cancers of the blood system. A number of leukaemias occurring in infants are known to have a pre-natal origin; however their precise cellular source and mechanism of emergence remain undefined. The aim of this project is to establish the repertoire of miRNAs expressed in cells of the developing blood system and in paediatric leukaemia samples. The function of miRNAs will then be further studied to uncover their role in normal blood development and how these processes become dysregulated in cancer development. The results obtained from these studies are likely to shed light on leukaemia pathogenesis and may highlight novel strategies for therapeutic interventions.

The use of phenotypic high content screening for the prediction of clinical response to chemotherapy in paediatric acute myeloid leukaemia (AML)

While 80% of all children's cancers are curable, Acute Myeloid Leukaemia is one of the exceptions with only 60% of children responding to standard chemotherapy for this disease. The challenge is to provide help to the doctors to treat the 40% of children who progress to second line therapy. With so many anti-cancer drugs now available, it would be highly desirable if there was a way to determine which ones in advance might be most effective for a particular child. Imagen Therapeutics has a test which may be able to address this challenge. However, the first step to exploring whether our test will work is to see if it can predict, in a set of banked AML samples, the clinical outcome of the children concerned. If it can, then this would be strong evidence that the assay can be used to guide future second and third line therapy in paediatric AML.

Tracking the juvenile myelomonocytic leukaemia stem cell

This research project will study a rare and fatal form of childhood blood cancer that is called Juvenile Myelomonocytic Leukaemia (JMML). This type of blood cancer usually presents within the first 2 years of life and it carries very poor prognosis. To date, the only available treatment is transplantation of haematopoietic cells from a healthy donor to the patient (stem cell transplant). Unfortunately, JMML is very difficult to treat and even after treatment, the disease often returns. The aim of this project is to identify the exact types of cells where the disease originates and study their characteristics in great detail. We believe that these cells are also most likely to be responsible for the high rate of disease relapse and the disappointing response of JMML to conventional leukaemia therapy. We hope that understanding the pathways which have led to the abnormal growth and development of the disease will help to develop much better treatment targeted to the abnormal cells responsible for generating and propagating JMML. Furthermore, with this project we aim to use the latest technologies available in order to detect all the genetic abnormalities that can lead to the disease development; and understand how the different genetic alterations affect the natural course of the disease. This will enable doctors in the future to identify more effectively the high risk patients and apply the most appropriate treatments for each patient, with the overall aim to improve the outcome of the children with this disease. Lastly, our detailed genetic analysis will enable us to identify patients that carry inherited mutations and provide the families of these children with the appropriate follow up for the family members that are likely to be affected.

Transcriptional regulation of LMO2 in T-acute lymphoblastic leukaemia

Acute lymphoblastic leukaemia is a cancer of the white blood cells and is the commonest cancer to affect children. Although cure rates for some subtypes of leukaemia have improved rapidly over recent decades, these advances are not shared by all sub-types. T-acute lymphoblastic leukaemia is one such subtype and has been shown to be associated with mis-regulation of genes associated with the control of transcription, the first step on the way to producing active proteins. LMO2 is misregulated in over 60% of patients with T-acute lymphoblastic leukaemia and the cause for this is not currently clear. By using state of the art DNA chip technology, only available since the completion of the genome project, I have identified sequences of DNA that are mis-regulated in T-acute lymphoblastic leukaemia cell lines and, if they are also active in patient samples, we will have gained greater understanding of how this gene is inappropriately switched on in this disease, which may suggest ways to identify possible new and more targeted treatments.

Unravelling clinical heterogeneity in Philadelphia positive ALL

Defects in genes that control blood cell development are common causes of blood cancers, such as acute lymphoblastic leukaemia. One well-recognised example involves a switch of genes between two chromosomes to produce an abnormal chromosome called the Philadelphia chromosome. This single genetic change can produce at least two types of leukaemia - chronic myeloid leukaemia (CML) and Philadelphia positive acute lymphoblastic leukaemia (Ph+ALL).

These two leukaemias are usually thought of as very distinct diseases and have different treatments. However, new scientific research indicates that some (but not all) cases of Ph+ALL share a lot in common with CML. Ph+ALL cases with CML-like features (called CML-like Ph+ALL) appear to be a new sub-type of leukaemia and may need different treatment approaches to enable long-term cure.

In this project we will investigate this newly discovered leukaemia sub-type - CML-like Ph+ALL. We have three experimental aims:

1. To discover whether CML and CML-like Ph+ALL originate in the same types of primitive blood cells located in the bone marrow. This will help us understand how CML-like Ph+ALL develops.
2. To measure the levels of gene expression in the different leukaemia sub-types to see if we can identify a genetic 'signature' that allows us to distinguish between Ph+ALL, CML and CML-like Ph+ALL. This will help devise new tests to use in the clinic to identify patients with CML-like Ph+ALL at the time of diagnosis.
3. To test CML-like Ph+ALL cells in the laboratory to see if they will respond to new less-toxic therapies that our research team have recently developed to increase long-term cure for patients with CML.

In this way, we aim to improve the outlook for patients diagnosed with Ph+ALL, by making sure they get the right diagnosis and treatment

Using computational modelling to identify effective drug combinations in high-risk acute leukaemia

Many aggressive blood cancers have mutations that cause overactivity in enzymes called kinases, which are usually linked to poor patient outcomes. These kinases can be directly targeted with new non-chemotherapy drugs called kinase inhibitors, which could be more effective and have fewer side-effects than chemotherapy. However, these drugs are often ineffective when used alone, with many working better in combination, and it is very difficult to predict which combinations work best in patients.

Systems biology approaches might provide a fast-track to finding the best combinations. We will first use computer models to predict which inhibitor combinations work best in practice, and test these directly in the laboratory. To ensure that combinations will be effective and safe in patients, we will test them rigorously in mouse models and human cancer samples. Our hope is that these approaches will find better and less toxic treatments for patients with aggressive blood cancers.

Whole genome sequencing of genetically defined subtypes of childhood acute lymphoblastic leukaemia

The cancer cells of patients with leukaemia carry genetic abnormalities which are the drivers of the disease. Some genetic abnormalities are associated with an excellent response to standard treatment whilst others indicate that the patient will ultimately relapse. This project focusses on a rare subtype of childhood leukaemia which is associated with an extremely poor outcome. We will use state of the art technology to sequence (“read”) the entire genome (“genetic code”) of patients with this poor risk abnormality and compare it to the genome of patients with a related abnormality that has a good prognosis. The results of this study will help us to understand why different patients response differently to the same treatment. It will also help in the development of novel therapies for patients with a poor prognosis.