ESR01 – Aravind Sankar – based in Cambridge - Research Summary
Genome-wide association studies and exome sequencing of melanoma-prone families has identified loci linked to melanoma development, but the genetic determinants contributing to risk for most patients remain unknown. In this project, the student will analyse the whole genome sequence data from over 100 melanoma patients who are highly predisposed to this cancer to identify new genes and pathways involved in this malignancy. In parallel, data from nearly 3500 sporadic melanoma patients will be studied to identify further loci and variants associated with melanoma risk. The project will involve the analysis of next-generation sequencing data and the use of algorithms to predict the consequence of the variants discovered. At the Sanger Institute there is a strong belief that students, even those with an inclination towards computational work, should gain experience in the research lab. Thus, the second phase of the project will involve laboratory experimental work in cells, and potentially in mice, to study the genes and variants identified from the analysis of the sequence data.
ESR02 – Eirini Christodoulou – based in Leiden - Research Summary
The genetic basis of hereditary melanoma is only partially understood. Germline mutations in genes encoding for cell cycle checkpoint regulators and telomere maintenance proteins have been established as causing hereditary melanoma. New high penetrance melanoma susceptibility genes are currently discovered through application of whole exome and genome analysis. This research project aims to discover and establish new melanoma susceptibility genes by whole genome sequencing data analysis and functional genetic analysis of candidate gene variants. The phenotypical characteristics of pathogenic mutation carriers, including the tumour spectrum, will be studied. Additionally screening approaches will be employed to expose tumour cell dependencies and specifically target melanomas in carriers of these mutations.
ESR03 – Marina Juraleviciute – based in Oslo - Research Summary
The identification and characterization of novel genes involved in melanoma development and progression can provide excellent additional therapeutic targets for therapy and prevention. However, identification and validation of these genes is hindered by a lack of good laboratory models. Therefore the focus of the project is on the development of novel melanoma models based on induced pluripotent stem cell technology (iPSC).
Marina will be using iPSCs to produce melanocytes from individuals with melanoma susceptibility genes and investigate their biological functionality and cellular responses after exposure to environmental risk factor, such as ultraviolet light of different wavelengths. We will also investigate the role of the most common sporadic oncogenic mutations found in melanoma. We will perform whole-transcriptome analysis of the cells in order to identify differences in gene networks compared to control cells. We expect to obtain better understanding of how particular genetic variants functionally and molecularly contribute to initiation of melanoma.
ESR05 – Joanna Pozniak – based in Leeds - Research Summary
Immune responses to melanoma have been identified as crucial determinants of survival for this form of skin cancer. Advanced melanoma has been extremely difficult to treat but has recently been shown to respond to treatment with T-cell checkpoint inhibitors. This is a real breakthrough which has changed the scope of melanoma research internationally. Although the discovery has brought about fundamental change in the clinic, there are a number of key difficulties which this project is designed to address. First, the drugs are expensive and toxic: we therefore need to identify biomarkers which will allow clinicians to test the patient samples in order to predict response and then select the optimal treatment (predictive biomarkers). Second, although it is expected that using a combination of drugs, response rates to the drugs will improve, current data suggest that many patients will not respond to the drugs. The aim then is to better understand immune responses to the melanoma in order to identify alternative means of identifying novel treatments, especially in the adjuvant setting.
In Leeds, we have generated gene expression data from in excess of 700 primary tumours. Initial analysis of the first 300 revealed that biological pathways related to the immune system have prognostic value. We have since explored bioinformatics approaches to understanding that immune response. We are also currently using digital pathology to correlate the histopathological correlations with different transcriptomic patterns. In this project Joanna will use bioinformatics approaches to understand the biological basis of those gene expression patterns and in particular to determine if algorithms can be used to dissect those immunological pathways. She will then test the resultant hypotheses in histological and laboratory models as appropriate, with the aim of developing prognostic biomarkers.
ESR06 – Sathya Muraldihar – based in Leeds - Research Summary
Microscopic ulceration of the primary melanoma tumour is a poor prognostic factor which yet appears to predict response to treatment with interferon. Ulceration is therefore of very important but of unclear biological significance. Recent work by the group has established that the transcriptome of ulcerated tumours is indicative of a pro-inflammatory environment and shown that medical conditions associated with systemic inflammation are associated with ulceration. In a new study funded by the MRC, genome-wide gene expression transcriptomic data will be generated from the peripheral blood, primary tumour and draining sentinel lymph node of 300 newly diagnosed melanoma patients. The study will allow us to correlate the transcriptomes of three components of the body in which host/tumour interaction can be studied: tumour, blood and node. The study will also allow us to test the interaction between systemic inflammation and immune responses to the cancer. The objectives of the project are: first, to explore further how persistent inflammation in tumours is related to systemic inflammation (including gene expression in circulating white cells and inflammatory markers) and if both are less evident in the presence of higher vitamin D levels or non-steroidal anti-inflammatory drug usage; second, to establish if tumour inflammation is associated with new blood vessel formation in the tumour and invasion histologically and (using gene expression) characteristic tumour gene expression; and third, crucially, to establish if tumour inflammation is associated with evidence of suppression of adaptive immunity in draining lymph nodes, again using gene expression. The second and third aims will form the core of this ESR project.
Sathya will initially sample tumour and nodal tissues, extract RNA and co-ordinate generation of the transcriptomic data by a commercial provider. Sathya will use statistical and bioinformatic methodology to analyse the resulting data and develop experimental approaches to validate significant results by examining RNA stored from other studies. Sathya will then integrate transcriptomic data from the different tissue types with histopathological and clinical data to look for: evidence of a role for systemic inflammation in melanoma survival, which might be addressed therapeutically; evidence for or against vitamin D as a modifier of systemic inflammation; evidence for an effect of tumour or systemic inflammation on suppression of adaptive immunity to melanoma; transcriptomic patterns in the tumour or blood associated with nodal evidence of reduction of adaptive immunity, which might be explored as a prognostic and predictive biomarker.
ESR07 – Sofia Chen – based in Cambridge - Research Summary
Recent advances in cancer therapy have seen a profound improvement in the outlook for melanoma patients with advanced disease, yet there is still much to learn about how melanoma develops and how host factors such as the immune system and tumour stroma influence this disease. In this project Sofia will initially work with patient samples to analyse expression signatures from the tumour, stroma and immune infiltrate to identify genes associated with response to therapy. She will then work in a range of in vitro and in vivo models to understand how these host factors may influence tumour growth and development.
ESR08 – Catarina Salgado – based in Leiden - Research Summary
Melanoma progression is driven in part by epigenetic alterations and formation of aggressive subpopulations of cells. Recently we have identified promoter CpG island hypermethylation events associated with metastatic behaviour of melanoma. In addition, loss of DNA hydroxymethylation occurs during melanoma progression. DNA methylation and hydroxymethylation patterns will be analysed in metastatic melanoma samples. Microfluidic technologies will be applied to biopsy specimens to generate genomic data at the single cell level and explore clonal diversity and heterogeneity of melanoma cell subpopulations. Correlation of genomic and epigenomic data will provide insight into the biology of melanoma aggressive behaviour. Detection of epigenetic alterations in highly metastatic melanomas will potentially reveal biomarkers for metastatic capacity.
ESR09 – Shamik Mitra – based in Lund - Research Summary
The Lund group has reported the presence of 4 subgroups of melanomas defined by gene expression patterns, which impact on survival. The first objective for this project is to further explore the clinical and immunological associations of these subgroups. The second is to test the associations of the phenotypes with tumour characteristics and survival in other cohorts within MELGEN. If validated robust clinical biomarker development will be explored.
The proposed PhD project includes integration of molecular and clinical data of a large cohort of melanoma tumours with the aim of establishing the clinical significance of integrating molecular data.
ESR10 – Adriana Sanna – based in Lund - Research Summary
The Lund group previously described gene expression subtypes of melanoma tumours of which one is characterized by down-regulation of a transcriptional programme involving melanocyte specific genes including the master regulator of melanocyte development, MITF. We have further demonstrated that down-regulation of MITF can partially be explained by epigenetic mechanisms. In the proposed PhD project we will further examine the MITF transcriptional programme and its role in survival and progression of melanoma. Adriana will estimate MITF protein expression and methylation status in metastatic melanoma and investigate the prognostic importance of the fraction of MITF positive cells in individual tumours a feature that is closely associated to the gene expression phenotype classification. She will then correlate results to other molecular data such as gene expression, CNV and somatic mutation patterns. Furthermore, recent reports demonstrate that melanomas with such transcriptional profiles are less sensitive to targeted therapies and we will further investigate this to find novel treatment options.
ESR11 – Rita Cabrita – based in Lund - Research Summary
The Lund group have previously described transcriptional modules in melanoma corresponding to immune response, stromal cell infiltration, cell cycle and MITF targets. In the proposed PhD project Rita will search for therapeutic modalities affecting the different transcriptional modules and determine the functional effect of such treatment strategies. She will correlate these transcriptional modules with clinical and molecular features in melanoma patients and determine the clinical relevance of the modules in the 800 Lund primary melanoma cases. The findings will be validated in the Leeds Melanoma Cohort samples. The rationale is, as above, the potential that these modules may be nodal or “points of coalescence” genomic changes potentially targetable therapeutically.
ESR12 – Rohit Thakur – based in Leeds - Research Summary
One of the key aims of the MELGEN programme is the identification of prognostic and predictive biomarkers for melanoma outcome and response to treatment. MELGEN groups have already identified signatures associated with survival based on transcriptomic profiling of the tumour. The current programme will build on this work, both to better understand the underlying biology and to develop maximally informative biomarkers of outcome. The first aim, then of this PhD project is to improve statistical and bioinformatics analyses of genomic/transcriptomic data for biomarker development. The second is to use transcriptomics to understand the biology of interaction between the host and the tumour, especially with respect to immunological responses.
Rohit will develop supervised classification methods, including Random Forests and penalised regression, using whole genome transcriptomic data and other tumour characteristics to predict outcome. Methods for incorporating knowledge on relevant biological pathways will also be explored. The performance of these methods, and of previous analyses based on unsupervised clustering, will be evaluated in terms of stability, sensitivity and specificity; different approaches to model validation (including cross-validation) will be explored.
A further aim is to derive the simplest possible biomarker (based on fewer genes) without loss of performance. This could have both practical utility and aid in understanding the underlying biological processes. These methods will be examined using a variety of approaches which may include data simulation, use of other publically available data sets or theoretical considerations. In conjunction with other researchers in MELGEN, methods will be applied to the melanoma data available from Leeds including relapse, melanoma-specific death and response to treatment.
The ultimate aim is to develop a robust methodology for classifying tumours according to expected outcome and the creation of one or more predictive biomarkers for melanoma tumour behaviour.
ESR13 – Joey Mark Diaz – based in Essen - Research Summary
This project aims at identifying molecular genetic profiles with prognostic and predictive significance in melanoma to meet the need for reliable biomarkers and efficient therapy. Mutation data will be generated by next generation sequencing (amplicon panels and capture based enrichment). These data will then be analysed with respect to their clinical relevance for survival (prognostic relevance) and therapy response (predictive relevance), as long term follow up times are available for our patients. Identified profiles of driver mutations and rare mutations with prognostic or predictive value will be validated in the MELGEN data sets. Genome and transcriptome sequencing data from metastatic tumour samples of stage IV melanoma patients will be generated and analysed to complement data on primary tumours (generated in Lunds). The project will additionally integrate datasets from publicly available databases (e.g. TCGA, NCBI GEO, ICGC) to test and validate established hypotheses on mutations, gene expression, prognosis and therapy response.
ESR14 – Sonia Leonardelli – based in Essen - Research Summary
The project will focus on understanding and analysing the mechanisms involved in tumour resistance to targeted and immunotherapeutic approaches in terms of cells entering senescent-like states. Senescence usually refers to an oncogene induced irreversible state where cells express selected markers and stop proliferating. Similar but reversible effects are observed when tumour cells are exposed to either targeted or immune therapies. The goal of this project is to further analyse these cell states, both on a functional level by analysing the phenotype and biological behaviour involved, as well as from a genetic perspective, focusing on gene expression and epigenetic changes occurring. This data, initially generated in in vitro experimental settings will be validated in available patient samples. The goal is to identify biomarkers predicting resistance as well as novel therapeutic approaches targeting resistance mechanisms.
ESR15 – Renata Varaljai – based in Essen - Research Summary
The research focus will be on the establishment of prognostic and predictive cfDNA (circulating cell-free DNA) profiles of patients with metastatic melanoma. The aim is to develop a new monitoring tool that indicates the time point of cancer drug resistance based on the occurrence of mutant driver gene DNA in serum blood samples. Using mutation data from a pre-defined amplicon panel of 150 melanoma genes for routinely treated patients candidate mutations will be identified for biomarker development. QPCR assays will be performed, and deep sequencing will be evaluated, as monitoring tools to track these mutations as potential biomarkers in serum samples from melanoma patients under treatment. DNA from pre- and post-therapeutic tumour samples will serve as validation sets for the cfDNA serum profiles (in conjunction with Joey Mark Diaz, Sonia Leonardelli and other partners within the ETN). Resulting cfDNA data will be correlated with clinical follow-up information. This project is part of our long-term effort of developing individual, fine-tuned sequential therapies to improve patients’ survival.
SWISS1 – Sabrina Hogan – based in Zurich - Research Summary
Cancer immunotherapy is arguably one of the most exciting areas of translational cancer research today. The recent success of both targeted and immuno-modulatory therapies has introduced a new era of precision medicine, with unprecedented clinical benefit to metastatic melanoma patients. Despite these advances, therapeutic resistance frequently occurs in most treatment modalities, and not all patients respond to even these more precise therapies. Thus it is critical to identify early molecular prognostic markers that will help stratify patients for more appropriate first-line therapies, and to plan the most promising second-line treatments. One factor for successful therapeutic response, especially in the case of immunotherapy, is likely to be the ability of the immune system to recognize and clear tumour cells from the patient during therapy. However, the exact types of immune response to targeted or immune therapies is still not completely clear, especially when comparing patients with progressive disease to those with partial or complete responses to therapy.
Sabrina will quantitate T- and B-cell diversity in the blood in melanoma patients undergoing both targeted chemotherapy and immunotherapies, and relate that diversity to histological measures of tumour infiltrating lymphocytes and survival. She will then validate the findings in other data sets within UZH and in MELGEN and work with ImmunID to develop blood derived prognostic and predictive biomarkers.
SWISS2 – Ishani Banik – based in Zurich - Research Summary
The advent of next-generation sequencing technologies and their application to patient-derived tumour material is a tremendous advance in understanding basic disease mechanisms. However, this flood of -omic data is highly descriptive and requires the coupling of in vivo assays to prioritize which genomic observations may have functional significance. Recent publications have demonstrated the effectiveness of zebrafish as a genetic model for in vivo screening of melanoma oncogenes to initiate tumorigenesis when expressed in adult melanocytes in a P53-mutant background. We have analysed the TCGA dataset, which consist of exome data from 238 melanoma tumours, and prioritized a list of frequently mutated genes in tumours with no known oncogenic driver mutation. When normalized for the size of the open reading frame and predicted functional consequence of those mutations, we can reduce the list of potentially interesting mutations to 25. We propose to overexpress each of these mutant variants in zebrafish melanocytes (as has been already done for BRAF, NRAS, and RAC1) and to measure the time to tumour progression in order to prioritize genes that may have a functional consequence in human melanomas. Genes with a significant effect on the time to melanoma formation in zebra fish adults will be further validated in human melanoma samples and cell lines. Future work may also involve the generation of a conditional, tissue-specific mouse model from the most compelling candidate.