Goedgekeurde dataverzoeken

The observed incidence of cancer biomarkers can influence resource investment into key areas such as drug discovery and companion diagnostic technology. The list of clinically relevant cancer biomarkers is growing, however a systematic description of the incidence of these biomarkers is currently lacking. Understanding the prevalence of these biomarkers in late-stage cancer can be especially relevant since many companion diagnostics are utilized in the treatment of stage III and IV cancers.

By using multiple publicly available knowledge management resources that describe clinically relevant alterations, we aim to characterize the frequency of these biomarkers and contextualize their associations within the Hartwig dataset. 

Eivind Hovig,  University of Oslo, Norge

Immune checkpoint inhibitors (ICIs) have been approved for treatment of metastatic urothelial cancer patients. Unfortunately, treatment response rates are low and many patients are being exposed to ineffective treatment with the risk of developing (severe) side effects. Currently, there is a lack of reliable biomarkers to identify patients that will benefit from ICIs. Therefore, the aim of this study is to identify potential genomic and transcriptomic predictive markers using an unbiased approach with machine learning for identification of response to ICIs.

Jeroen de Ridder, Universtair Medisch Centrum Utrecht (UMCU), the Netherlands

Cancer develops by the accumulation of mutations in the DNA of individual cells. In many patients, DNA repair mechanisms are often perturbed early in cancer development, leading to an increased mutation rate and faster development of cancer. However, repair deficiencies can also make the cancer cells more susceptible to certain types of treatments. Better understanding and detection of DNA repair deficiencies may thus help improve the treatment of cancer patients.

Using the cancer genomics data from Hartwig Medical Foundation, we aim to characterize the pattern of DNA deficiencies across different cancer types. We further aim to improve the detection of DNA repair deficiencies through their effect on mutation rates and patterns across the genomes of cancer cells. Finally, we aim further clarify the relationship between DNA repair deficiencies and cancer treatment efficacy.

Jakob Skou Pedersen, Aarhus University, Denmark

Recent advances in melanoma treatments have greatly improved patient survival. However, many patients still do not benefit, and we do not fully understand why some patients respond to treatment, while others do not. We have a cohort of melanoma patients with DNA and clinical data, that may help us understand this better. We aim to combine our study with other melanoma datasets from Hartwiig and other sources, to boost the strength and numbers of our study so that we will be better able to answer these questions.

Samra Turajlic, The Francis Crick Institute, United Kingdom

Tumors from children with underlying DNA damage response or DNA repair defects often carry somatic mutational footprints that can aid their identification. Furthermore, relapses or second tumors in these patients may present with specific signatures that arise due to treatment related damage, particularly in DNA-repair deficient tumors. For the detection and analysis of these signatures in childhood tumors, we request the data from the pan-cancer study on metastatic solid tumors (PMID 31748536) that contain repair and therapy associated signatures that will be used to increase power in signature extraction and detailed analysis of mutational signature extraction. 

Roland Kuiper, Prinses Maxima Centrum, the Netherlands

Thousands of cancer-driving mutations have been detected in the cancer genome. Some of them are used as cancer biomarkers or targets of cancer therapies. However, an overwhelming proportion of currently recognized cancer-driver mutations is located in protein-coding sequences, which encompass barely 2% of the genome. Therefore in this study, we are going to investigate cancer-associated mutations in the non-coding part of the genome. We will focus particularly on the characterization of gene-associated non-coding regions, i.e., 5’UTRs and 3’UTRs, and microRNA genes, to identify potential non-coding cancer drivers.

Piotr Kozłowski, Institute of Bioorganic Chemistry Polish Academy of Sciences, Poland

Cancer of unknown primary (CUP) has a worldwide incidence of 3-5% of all cancers diagnoses. The clinical problem of this group of patients involves long diagnostic period, limited treatment options (anti-cancer treatment is largely based on primary origin) and poor survival. Whole genome sequencing (WGS) helps these patients by both predicting the primary origin and finding all DNA alterations linked with targeted therapy. The Hartwig prediction algorithm of the tissue site of primary origin (CUPPA) is based on a reference dataset of tumors in the Hartwig database. We aim to improve the prediction algorithm by refining the classification of tumors within the reference dataset towards a more clinically relevant setup.

Petur Snaebjornsson, Netherlands Cancer Institute, the Netherlands

In recent years the treatment of patients with lung cancer has improved tremendously with the introduction of immunotherapy. This therapy helps the patient’s own immune system to attack the tumor cells. Unfortunately, not all patients benefit from this treatment. We want to better understand the mechanisms used by the tumor cells to escape the effect of immunotherapy. By improving our knowledge of these mechanisms, we hope to contribute to new cancer therapies in the future.

Willemijn Theelen, Netherlands Cancer Institute, The Netherlands

This project will focus on treatment resistance mechanisms in uterine cancers to develop novel targeted treatments as adjuvants or combination therapy. By investigating pre- and post-treatment uterine biopsies by sequencing, we will detect treatment associated genomic alterations with a specific focus on large genomic re-arrangements. In order to identify actionable resistance drivers, we need to increase the sample size by including the Hartwig uterine samples. In addition, we will investigate the prevalence of detected uterine cancer drivers in non-uterine cancers.

Camilla Krakstad, University of Bergen, Norway

Biliary tract cancer (BTC) is a malignancy of the bile ducts (also called cholangiocarcinoma) and gallbladder. BTC is a rare malignancy with a poor prognosis. Surgical resection is the only treatment with a chance on long-term overall survival but unfortunately only the minority of the patients qualify for resection because of locally advanced or metastatic disease at presentation. For these patients gemcitabine / cisplatin palliative chemotherapy is the standard treatment. Treatments with agents directed to DNA changes (targeted therapy) did not show better survival in different studies because of the heterogeneous nature of BTC. To improve the treatment of advanced BTC, a more comprehensive overview of the genomic changes in advanced BTC is required. lf we can identify subgroups of patients with different prognosis based on genomic changes, new clinical studies could be initiated to study a specific treatment in one or more of these subgroup of patients. 

Heinz-Josef Klümpen, Amsterdam UMC locatie AMC, the Netherlands

Venous thromboembolism (VTE), a venous blood clot, frequently occurs in cancer patients, and substantially affects morbidity, quality of life, and mortality. Yet, it is incompletely understood why cancer patients are at increased risk of this disease. The risk of VTE varies greatly across cancer types, strongly suggesting a role for cancer-specific oncogenic mutations which indirectly cause thrombosis.We aim to investigate this possible association using clinical data and genomic data from tumor biopsies acquired within the CPCT-02 study. Results could help understand the pathogenesis of cancer-associated venous thromboembolism, lead to new therapeutic targets, and identify high-risk patients for thromboprophylaxis.

Nick van Es, Amsterdam UMC locatie AMC, the Netherlands

Summary = Previous efforts in cancer research have been dedicated to identifying genetic alterations on the linear form of DNA, but a circular form of DNA called extrachromosomal DNA (ecDNA), which is only present in cancer cells, is being revealed as an important contributor to cancer progression. Therefore, in this project, we aim to identify genetic alteration patterns caused by ecDNA in metastatic cancer. We will leverage genomics data and their associated clinical data to identify such patterns across different cancer types and metastatic stages. Our project will provide valuable insight for better clinical treatment decisions and prognoses for ecDNA-carrying cancer patients.

Hoon Kim, Sungkyunkwan University, South Korea

Mutations are generated by endogenous (replication errors) and exogenous processes (UV light). Studying the mutational patterns observed across both genes and the entire genome of metastatic patients, we plan to elucidate the interplay of  endogenous and external factors that ultimately shape the mutational processes. Based on the knowledge acquired on mutational processes we will improve the methods to detect signals of positive selection in both coding and non-coding elements, in order to accurately identify those involved in tumorigenesis and study their implications in tumor emergence and evolution.

Nuria Lopez-Bigas, Fundació Institut de Recerca Biomèdica (IRB Barcelona), Spain

Cancer is a disease of the genome. Random factors, environmental exposures and also certain biological mechanisms have the potential to cause mutations in DNA, while the sophisticated DNA repair mechanisms available to human cells prevent such changes in genes. We will use the whole-genome sequences from metastatic tumors to examine how the balance between DNA damage and DNA repair shapes the mutation patterns observed in cancer. This is relevant to patients because we will further investigate how such mutational patterns are associated with risk of cancer progression and with response to various drugs, with implications to prevention and treatment.

Fran Supek, Fundació Institut de Recerca Biomèdica (IRB Barcelona, Spain

We have previously mapped in metastatic prostate cancer how genes are being regulated. Now, we aim to combine these results with mutations and gene activation status in these and other tumor samples, to better understand how gene regulation and mutations are connected in metastatic prostate cancer.The aim is to better understand metastatic disease, and how this deviates from primary prostate cancer and between patients. This knowledge can ultimately guide us to design better treatment strategies and predict which patients are at risk of relapse.

Wilbert Zwart, Netherlands Cancer Institute, the Netherlands

Estrogen Receptor is a hormone-responsive protein,  responsible for breast tumor growth. We found that DNA-binding specificity can predict treatment response, and profiled estrogen receptor/DNA interactions in over 100 breast tumors. Now, we aim to determine whether these bound-regions are mutated in breast cancer, and how this impacts tumor biology. For this, we request access to data from the metastatic breast cancers sequenced at the HMF. In metastatatic breast cancer, estrogen receptor remains critically important. We know that this DNA occupancy differs between patients, whether mutations at these sites alter biology is unknown. This information may explain why some tumors are resistant to therapy, and how we can resolve this.

Wilbert Zwart, Netherlands Cancer Institute, the Netherlands

Endometrium cancer is one of the most common gynecological cancer and accounts for ~6% of all cancers in women. The vast majority of endometrium cancers are dependent on a hormone receptor, Estrogen receptor α (ERα). We aim to identify how changes in the DNA sequence of the tumor would affect the function of ERα, and how this would ultimately impact gene activity and tumor cell growth

Wilbert Zwart, Netherlands Cancer Institute, the Netherlands

DNA carries the instructions to life but is also constantly damaged, causing mutation and disease. Sunlight and cigarette smoke are common damaging agents, but DNA is also attacked by chemicals produced by our own cells. Our work aims to uncover what these chemicals are, how they change DNA and how this contributes to cancer.

Juan Garaycoechea, Hubrecht Institute, The Netherlands

Pancreatic cancer and esophagogastric cancer are among the most deadly forms of cancer: despite intensive treatment with chemotherapy and / or surgery, the majority of the patients die from the disease. In PEGASUS we investigate whether molecular techniques are of value in predicting which patients will benefit from such intensive treatment, both with regard to the change of survival as well as quality of life. We are also investigating whether there are molecular targets that could improve treatment even further with patient tailored drugs. We will weigh the costs of using molecular techniques against the benefits and risks for the patient. Hanneke van Laarhoven, Amsterdam UMC locatie AMC, the Netherlands

The DNA in human cells are continuously exposed to internal and external challenges. The purpose of this project is to computationally derive specific footprints (signatures) in metastatic cancers that can give us clues regarding why they have become cancerous. We will use acquired and inherited mutation calls to explore and understand these characteristic patterns. The knowledge gained is of importance for personalisation of cancer management strategies.

Serena Nik-Zainal, University of Cambridge Clinical School, UK

The purpose of our project is to identify genomic signatures related to DNA damage repair (DDR) defects in advanced prostate cancer and to assess their association with hormonal therapies. The data we plan to use are the 459 whole genomes from HMF patients with metastatic prostate cancer, aiming at determining the association of DDR defects with specific treatments to finally predict which patients could benefit from DDR targeting agent therapies.

Joaquin Mateo, Vall d'Hebron Institute of Oncology (VHIO), Spain

Microsatellite instability (MSI) is a genomic phenomenon characterized by a high number of mutations in tumor DNA. This is due to an error in the DNA repair mechanism, either germline or somatic. MSI-high has been extensively researched in colorectal cancer. MSI-H CRC is enriched for oncogenic fusions, of which some are responsive to targeted therapy e.g., NTRK fusions. In this study, we want to describe the occurrence of oncogenic fusions in MSI-H solid tumors. The second research question is to describe the (co)-occurrence of oncogenic driver mutations in in MSI-H solid tumors. If we find that MSI-H tumors are enriched for oncogenic fusions and/or driver mutations, this could potentially improve treatment possibilities for these patients. Some of these patients might have already been treated with targeted therapy, so we would also be interested to investigate treatment responses.

Barend Sikkema, Erasmus MC, The Netherlands

Genomic structural alterations are prevalent across many tumor types and can affect genes that drive cancer progression. It is important to classify the alterations patterns in tumor genomes in order to understand how they dysregulate genes and lead to treatment resistance in cancer. We propose to analyze these alteration patterns in advanced cancer patients with metastatic tumors. We will develop new computational and analytical methodologies to identify novel alteration signatures associated with response and resistance to therapies. Then, we will compare and contrast between different types metastatic cancers. This work will provide insights into developing biomarkers from genomic alteration signatures for clinical applications.

Gavin Ha, FRED HUTCHINSON CANCER RESEARCH CENTER, United States of America

While localized prostate cancer can be cured, metastastatic disease is associated with high morbidity and mortality. Prostate cancer typically metastasizes to lymph nodes and the bone, while metastases to organs like liver, lung or the peritoneum are less common. This project is aimed to describe what the differences are in the genomic (DNA) make up of prostate cancer metastases in the different sites. Knowledge hereof might spark further research into ‘site-preference’ and ‘site-specific alterations’ of the prostate cancer metastases genome, which ultimately might result in means to better treat or even prevent metastases.

Andre Bergman, Netherlands Cancer Institute, the Netherlands

Only 2% of the human genome contains protein-coding genes. Their expression depends on the regulatory elements in the surrounding “non-coding” genome. High-level amplifications frequently are extrachromosomal amplicons (ecDNA) and their presence in a cancer is associated with worse outcomes for the patient. They include oncogenes that promote cancer growth and regulatory elements, all stitched together into a small circle. The HMF whole genome sequencing data will allow us to reveal exactly which loci are included in ecDNA. In this project, we propose to investigate which specific regulatory elements are required by ecDNAs to drive aggressive tumor growth.  

Anton Henssen, Charité - Universitätsmedizin Berlin, Germany

Tumor metastasis involves dissemination of cancer cells of primary tumors to distant organ sites. Though current cancer research has delineated important mutation features in primary cancer tumors, the identification and understanding of the genetic and genomic profiles is less complete in metastatic cancer. Therefore, understanding the genomic differences between primary untreated and late-stage treated metastatic tumors is essential to understand the cancer evolution and to address potential therapeutic vulnerabilities. In this study, we will perform the largest harmonised analysis (i.e. processed through the same analytical pipeline) of primary tumors (N=2500) from PCWAG and metastatic samples (N=5000) from the Hartwig dataset. The results of this study will provide a better understanding of the processes that are operational at different stages of carcinogenesis. Such information may be used for design of preventive strategies (for tumorgenesis or metastatic seeding) but can also be used for stratifying patients towards targeted treatments based on such characterics (e.g. homologous recombination deficiency as we have already demonstrated by development of the CHORD algorithm, DR-018).

Arne van Hoeck, Universtair Medisch Centrum Utrecht (UMCU), the Netherlands

This study aims to quantify the aggregated effect of mutations in metastatic tumors and compare it with those in primary tumors. We will compute the functional impact of metastatic variants on a pan-cancer level to accomplish this goal. Furthermore, we will compare the functional impact of metastatic variants with those in primary tumors from our previous PCAWG study.

Sushant Kumar, University Health Network, Canada

As part of the EUCANCan project, a federated network for harmonized genomic and phenotypic data sharing, we aim to foster cancer research and its application to precision medicine in a clinical environment which requires an accessible, accurate, and standardized methodology to evaluate the procedures involved in the identification of tumour variants in cancer patients. The lack of easy-to-access and exhaustive benchmarking datasets and procedures together with the intrinsic difficulty of the variant identification process make the implementation of a personalized medicine program in oncology practically impossible for centers without access to bioinformatic expertise and computational power. This project aims to utilize validated tumor patients data from the Hartwig Medical Foundation cohort and other cancer initiatives to provide a complete benchmarking system to homogenize and make variant identification methodologies compatible.

Philippe Hupe, Institut Curie, France

In this project we aim at combining mathematical modelling of tumour evolution with cancer genomic data to measure fundamental evolutionary characteristics of metastatic tumours. These measurements will allow better understanding of how tumours spread to distant parts of the body. Importantly, we anticipate that the patient-specific ‘evolutionary rules’ we will measure from the data may be informative in terms of stratifying patients between good versus poor prognosis. Moreover, some of these measurements may also be predictive of the effectiveness of certain treatments.

Andrea Sottoriva, Human Technopole, Italy

Triple-negative breast cancer (TNBC) is the most challenging breast cancer subtype to treat due to its aggressive characteristics and low response to the existing clinical therapies. Moreover, TNBC is a heterogenous subgroup within breast cancer, but the critical determinants why some TNBC patients demonstrate metastasis and others not, are not well understood. We uncovered that the protein optineurin (OPTN) plays an unexpected role in TNBC during metastasis. OPTN suppresses TNBC metastasis in experimental models. Data mining of publicly available data bases revealed that the OPTN levels in TNBC patients positively correlates with relapse free and distance free survival. We want to validate and extend these findings in independent TNBC patient cohorts. In particular we are interested to find out if OPTN expression levels (high expression or lack thereof) marks a specific subset of TNBC patients.

Peter ten Dijke, Leiden University Medical Center, The Netherlands

Structural variations (SVs) are large scale changes of DNA, which include deletions, duplications, inversion, translocations and other more complex forms. We aim to study the causes of somatic SVs in cancer and elucidate how they contribute to tumor initiation, progression and metastasis. We will develop novel methods to analyze the genomic and transcriptomic sequencing data. We expect to discover new genetic factors and genes contribute to the disease and improve disease prevention and treatment.

Lixing Yang, University of Chicago, USA

The overall objective of our study is to assess if melanoma with a specific DNA mutation (in the CDKN2A gene) may be more sensitive to certain treatments. In order to identify therapeutic vulnerabilities it is essential to obtain a comprehensive picture of the DNA alterations and RNA expression levels in the tumors. The combined DNA and RNA data of patients melanoma tumours, in conjunction with results from our analyses performed on melanoma cells, can be used to generate a computational model of melanoma with CDKN2A mutation, possibly allowing us to predict how treatment of melanoma with this mutation can be improved.

Remco van Doorn, Leiden University Medical Center, The Netherlands

Cells may grow more quickly when a specific subset of genes called oncogenes are activated. Normally, two DNA copies. each gene are present. Oncogenes may become activated when their DNA copies are duplicated. In some instances, hundreds of oncogene DNA copies have been detected in cancer cells. This process is called amplification. In our project, we want to explore which genes are commonly amplified. We also want to evaluate different mechanisms through which amplification can be achieved, which may be linear amplification, complex amplification, or extrachromosomal amplification. Finally, we would like to study how oncogene amplification relates to patient response to therapy.

Roel Verhaak, The Jackson Laboratory, USA

HER2-positive breast cancer is an aggressive subtype driven by amplification of the Her2/Erbb2 gene. This oncogene is amplified and implicated in chromothripsis, a catastrophic event that causes massive genomic rearrangements. Characterizing chromothripsis in HER2-positive breast cancer and other aggressive subtypes of disease is an important step towards improved understanding of the etiology of disease. We will use whole-genome sequencing data to identify structural rearrangements at single nucleotide resolution and infer mechanisms of chromosome shattering and repair. We hope to understand whether chromothripsis associated with poor outcome in HER2-positive breast cancer and other aggressive subtypes of breast cancer, as well as to evaluate its prevalence in metastatic vs. non-metastatic tumors, and the impact of treatment on genomic aberrations.

Anna Poetsch, Technische Universität Dresden, Germany

HER2-positive breast cancer is an aggressive subtype driven by amplification of the Her2/Erbb2 gene. This oncogene is amplified and implicated in chromothripsis, a catastrophic event that causes massive genomic rearrangements. Characterizing chromothripsis in HER2-positive breast cancer and other aggressive subtypes of disease is an important step towards improved understanding of the etiology of disease. We will use whole-genome sequencing data to identify structural rearrangements at single nucleotide resolution and infer mechanisms of chromosome shattering and repair. We hope to understand whether chromothripsis associated with poor outcome in HER2-positive breast cancer and other aggressive subtypes of breast cancer, as well as to evaluate its prevalence in metastatic vs. non-metastatic tumors, and the impact of treatment on genomic aberrations.

Christina Curtis, Stanford University School of Medicine, USA

Cancer-related signaling pathways frequently involve protein kinase C (PKC), a prominent enzyme that phosphorylates alpha-tubulin, a building block of microtubules.  Phosphorylated alpha-tubulin causes human breast cells to acquire metastatic potential, whereas with non-phosphorylated alpha-tubulin, cells are hyperproliferative, as shown in an animal model.  Genomic sequences of primary tumors in publicly available databases showed that alpha-tubulin acquires mutations at the PKC phosphorylation site that are primarily blocking mutations, and were therefore consistent with the genomic samples were from primary tumors.  With the HMF dataset, we will examine data from breast lesions that are authentically metastatic, enabling validation of alpha-tubulin as a bio-marker for predicting metastatic potential and thereby aiding in cancer diagnostics. More generally, we will be alert to novel recurring mutations in other genes that have not yet been studied.

Susan Rotenberg, Queens College of The City University of New York, United States of America

Breast cancer is the second leading cause of cancer-related death in women worldwide. Hormone therapy has been proven effective to increase survival after surgical removal in 80% of patients (Estrogen Receptor alpha-, ERa-positive breast cancers). Nevertheless, in many cases the residual tumor evolves, develops resistance and continues to spread. As compared to the initial tumor, we currently know very little about the resistant disease. This is partly due to scarce use of re-biopsy at time of first relapse in the clinical practice. This dramatically curbs our ability to properly treat relapsing patients, as targeted-therapies remain dictated by the information collected at time of diagnosis. Additional factors restrict our understanding on how genetic alterations drive cancer onset and progression. We often limit our search to the protein-coding part of the human genome; on the other hand, resolution at a genomic scale is poor. As cancer cells tend to accumulate a large number of aberrations that carry no functional significance, one critical problem is how to recognize those propelling the disease, particularly among those residing in regions that control the activity of genes rather than in genes themselves. We are currently developing a novel strategy to overcome these limitations and identify genetic and epigenetic alterations driving resistance. We designed an innovative approach to investigate thousands of potentially critical regions at very high resolution. We are currently collecting this information longitudinally, at diagnosis and at first relapse, for hundreds of patients that underwent hormone therapies. We plan to integrate this information into novel analytical tools to pinpoint driver alterations recurring across multiple patients. This is expected to identify new actionable targets in metastatic breast cancer and – by applying these tools to larger cohorts of breast and other cancers – to reveal new critical biological principles driving tumor evolution. Luca Magnani, Imperial College London, United Kingdom

Cancer is driven by mutations in the genome. Several drugs, including some of the most widely used compound types, induce mutations and genomic instability, and tumor cells that survive drug treatment often involve substantial drug-induced mutations and rearrangements in the genome. Using the HMF data, we aim to study the regional context and genomic location of somatic alterations across different tumor types stratified by drug type, to identify therapy-related mutagenesis at the level of somatic rearrangements.

Joachim Weischenfeldt, Rigshospitalet, Denmark

Head and neck cancer can be cured in approximately half of patients suitable for radical treatment with radiotherapy. Treatment failures often manifest with local relapse, and those patients who relapse often have limited treatment options and poor clinical outcomes – a significant unmet clinical need. The biology underpinning resistance to therapy in head and neck cancer is unknown. We intend to use the Hartwig cohort to compare with patients whose cancers are yet to be treated, and with a cohort who experienced relapse despite radiotherapy in a large clinical trial, to see if similar genomic characteristics can be found. This could help identify future new biomarkers for radiotherapy.

Nick Turner, The Institute of Cancer Research, United Kingdom

We want to achieve better care for patients with CRC and melanoma, by allowing a multidisciplinary team of researchers to being able to query, analyse and integrate DNA and RNA sequencing data with matching clinical, pathology and radiology images and annotations, using different approaches: (1) artificial intelligence (AI), in order to determine whether predictive disease outcome correlations/algorithms using these data types can be identified, and (2) exploration of available data (in different formats than required for AI) on a human readable platform, allowing researchers to determine whether it is possible to explain AI findings by human interpretation of the data.

Mariska Bierkens, Netherlands Cancer Institute, the Netherlands

The two strongest factors predicting a human cancer's clinical behaviour are the primary tumour's anatomic organ of origin and its histopathology. However, roughly 3% of the time a cancer presents with metastatic disease and no primary can be determined even after a thorough radiological survey. A related dilemma arises when a radiologically defined mass is sampled by cytology yielding cancerous cells, but the cytologist cannot distinguish between a primary tumour and a metastasis from elsewhere. Here we use whole genome sequencing (WGS) data from the ICGC/TCGA PanCancer Analysis of Whole Genomes (PCAWG) and the Hartwig Medical Foundation to develop a machine learning classifier able to accurately distinguish among ~30 major cancer types using information derived from somatic mutations alone. This demonstrates the feasibility of automated cancer type discrimination based on next-generation sequencing of clinical samples. In addition, this work opens the possibility of determining the origin of tumours detected by the emerging technology of deep sequencing of circulating cell-free DNA in blood plasma.

Lincoln Stein, Ontario Institute for Cancer Research, Canada

Cancer of Unknown Primary (CUP) accounts for approximately 3% of all metastatic cancers. Despite modern imaging techniques and extensive pathology work-up in CUPs, tumor origin remains unknown in the majority of these patients. Large-scale genome sequencing studies have revealed that tumor types can be classified based on distinct patterns of somatic variants and other genomic characteristics. We aim to investigate the clinical value of genome-driven tumor type prediction using whole genome sequencing (WGS) in the diagnostic workup of patients with CUPs and other tumors that are challenging to diagnose.

Kim Monkhorst, Netherlands Cancer Institute, the Netherlands

EGFR-targeting therapies are used as anti-cancer agents. Not all patients respond well and side-effects can be significant. Therefore, biomarkers that identify those likely to respond have emerged and are used in daily practice. But even among patients positive for these biomarkers, response rates remain variable. We would like to use the HMF data of EGFR-treated patients to identify DNA alterations that are predictive of a poor therapy-response. As such, we aim to ‘fine-tune’ the molecular pre-screening of patients eligible for this therapy, thereby improving response rates and providing the best care for cancer patients.

Jeanine Roodhart Universtair Medisch Centrum Utrecht (UMCU) the Netherlands

Immunotherapy, like nivolumab and pembrolizumab, has drastically improved the survival of many cancer patients. However, some patients do not clinically benefit from treatment with immunotherapy. Many predictive biomarkers have been proposed to anticipate if a patient will be a so called responder or non-responder. However, none seem to be completely predictive. As immunotherapy can cause major side effects and is very costly, unnecessary treatment has to be avoided as much as possible. Therefore, the Erasmus MC and HMF are collaborating in a biomarker study (ARCTIC) to stratify patients who will benefit from ICI. The ARCTIC project combines existing clinical predictors with immune profiles and the genomic features, generated by the HMF in the same individuals, to develop a prognostic/stratification signature for selecting which individuals will benefit from ICI treatment.

Andrew Stubbs Erasmus MC the Netherlands

PurposeAfter treatment with a combination of chemotherapy and radiotherapy for esophageal cancer, the current standard treatment is to continue with resection. However, in circa 20% of patients, no vital tumor cells are found in the resection specimen. On the other side, circa 10% of patients develop metastases shortly after esophageal resection. These patients could have possibly been spared esophagus resection. A nation-wide study, initiated by the department of surgery of the Erasmus MC, is currently recruiting patients (the SANO study, PI prof. dr. van Lanschot) and aims to clarify the role of esophagus resection in patients with no radiological or pathological (on biopsy) evidence of disease after pre-operative chemoradiation. A rapidly developing minimally invasive method to improve detection of minimal amounts of tumor cells or fragments could help to identity those patients in whom true cure has been accomplished. Therefor, we are currently biobanking blood of the patients in the SANO trial to serially measure tumor DNA in the bloodWhich dataTo come to a comprehensive panel of genetic variants to be measured in the blood, we wish to incorporate the genetic variants found in the metastases of esophageal cancer patients, as these variants are expected to be the drivers of resistance to treatment and disease recurrence. Also, we want to explore the genetic characteristics of metastases of esophageal cancer.Relevance for the patientDiscovering new genetic variants in metasatses that have not been identified in primary tumors could lead to new therapeutic approaches. Optimizing the blood assay could lead to a new and minimally inavsive means to monitor tumor response.

Bianca Mostert Erasmus MC the Netherlands

Cancer is a deadly disease affecting millions of people worldwide. Tumor metastasis is the most frequent cause of death. The accumulation of mutations in the DNA explain the onset and progression of this disease. The main attention to understand metastasis has been focused on mutations in the cell nuclear DNA. However, the mitochondrial genome may play an important role in the development and progression of metastasis. The WGS data provided by the Hartwig Medical Foundation will help us to unravel a comprehensive map of mtDNA alterations in tumor metastasis and explore its role in cancer prognosis and treatment response.

Cristina Rodriguez-Antona, Spanish National Cancer Research Center (CNIO), Spain

In this study we will investigate the microbiome in metastatic cancer. We will use whole genome sequencing data from solid tumour cancers. We hypothesise that microbes travels together with the primary cancer cells to metastatic sites in the body and promotes tumoor growth and proliferation.

Robin Mjelle, Norwegian university of science and technology, Norway

Cancer is caused by mutations in the DNA, which can be analyzed by sequencing the tumor’s DNA. We can observe the effect mutations have on various levels, including changes in the amount of protein each gene produces. As proteins perform all the cell’s activities, altered levels could affect the function of our cells. To get a more complete insight into processes causing mutations and their consequences, we will look at the DNA and protein level. This will contribute to a more personalized treatment and help, for example, predict the primary location for those metastatic tumors for which this is unknown.

Ivo Gut Centro Nacional de Análisis Genómico (CNAG-CRG) Spain

Checkpoint inhibitors have failed to demonstrate a survival benefit in unselected castration-resistant prostate cancer (CRPC) patients. Nevertheless, biochemical and radiographic responses have been observed in a subset of trial participants. Here, we will study the association between immune signatures, genomic aberrations and clinical outcome to identify more immunogenic subgroups. A better understanding of factors that influence the immune infiltrate can help to select subgroups for checkpoint inhibitors and might serve as starting point for other immune-based therapies in prostate cancer.

Niven Mehra Radboudumc the Netherlands

We will use whole genome sequencing data derived from blood or normal tissue of colorectal cancer (CRC) patients in a search for new (inherited) predisposition genes, especially rare gene variants with .moderate effects on the risk of CRC.

Ian Tomlinson Cancer Research Centre, University of Edinburgh United Kingdom

The purpose of the project is to search for repeating mutations in genes related to translation in cancer. Such genes lead to impaired translation of proteins. These proteins have a greater tendency to be degraded by the proteosome, and be presented on MHC of cancer cells, to the immune system. For example, T cells can respond and kill the cancer cell. Our aim is to associate these recurring mutations in sequencing data of patients, to patient response to treatment.

Yardena Samuels Weizmann Institute of Science Israel

We observed specific chromosomal copy number changes in germ cell tumors that are proven to be resistant to cisplatin-based chemotherapy. We will use the HMF data to verify that these changes indeed occur frequently in metastasized germ cell tumors and determine whether they also occur in other tumor types that are treated with platin-based chemotherapy. This will help verify the role of these chromosomal changes in general platin resistance and hopefully aid in identifying patients with initial or gained resistant tumors that might benefit from other treatment options.

Leendert Looijenga Prinses Maxima Centrum the Netherlands

Currently, only 20% of pancreatic cancer (PC) patients are qualified for surgery, constituting the only potentially curative treatment available. Despite promising clinical results for targeted therapies, the molecular subtypes of PC is not yet useful to guide clinical decision-making. Therefore, the main purpose of this project is to reveal genomic features, underlying the anti-cancer treatment response. We will perform whole-genome sequencing (WGS) analyses on PC patient’s genomes obtained from HMF and identify the genomic differences between sensitive and resistant tumors. This data mining will improve PC patient stratification toward targeted therapies, contributing to the development of personalized medicine.

Pawel Zawadzki Adam Mickiewicz University in Poznań Poland

Lung cancer (LC) is the leading cause of cancer mortality. Despite a well-known therapeutic targets and thus a wide spectrum of targeted therapies, the mechanisms associated with drug resistance are limited to identification of a point mutations in a particular genes. Therefore, the main purpose of the research project is to reveal all genomic features that differentiate responders and non-responders to a given treatment type. We will perform whole-genome sequencing data analyses on LC patients genomes from Hartwig Medical Foundation and develop a clinical predictor that will improve patient stratification toward anti-cancer therapies and provide long-term benefit from treatment. Pawel Zawadzki Adam Mickiewicz University in Poznań Poland

 

Androgen receptor inhibitors (ARi) are frequently used to treat metastatic prostate carcinomas. In this project, we focus on identifying genetic causes for resistance to these new ARi to facilitate the development of new treatment options.

David Quigley UCSF USA

Mutations are changes in the DNA sequence which can cause cancer. We found that in melanoma and liver cancer different portions of the genome have different mutation patterns. By combining this information with information about the order in which DNA is copied, we could identify subtypes of liver cancer with different survival rates. Expanding our approach to the Hartwig Medical foundation whole genome sequenced metastatic tumors, will allow the definition of tumor subtypes, which may display different clinical behavior in terms of survival, and response to various treatments, helping us to personalize cancer treatment.

Itamar Simon Hebrew University Israel

In this project we aim to use whole genome sequencing data to improve the way we treat patients with prostate cancer. Providing a better understanding of the subgroups with different clinical and genomic characteristics may improve the treatment outcomes. To this end we need to know which alterations are present in the tumor genome of prostate cancer samples.

Martijn Lolkema Erasmus MC the Netherlands

We have created a local copy of the netflow version of th eHMF WGS pipeline and aim to validate the pipeline with samples from the LUMC that were previously sequenced at HMF for DRUP/CPCT studies.

Tom van Wezel Leiden University Medical Center the Netherlands

Many genomic alterations affecting known driver genes are still classified as variants of unknown significance (VUS). Especially VUS other than missense mutations, such as structural variants, have remained understudied. This is of particular concern for VUS affecting genes that are targetable by clinically approved therapeutics. Thus, there is a great need to perform systematic and rapid classification of VUS to allocate carriers to the appropriate personalized treatment. Here, we will use the WGS data from HMF to identify somatic genomic alterations affecting clinically actionable kinases and classify these into known driver events and VUS. We will then functionally validate VUS of interest in vitro and in vivo using cell lines and mouse modelling.

Daniel Zingg Netherlands Cancer Institute the Netherlands

Endocrine therapies remain a core component of systemic therapy for patients with estrogen receptor positive breast cancer, but all patients with advanced disease will eventually relapse. For many of them little is known of the details regarding how or why this happens. We are working on samples taken in a number of different clinical trials from patients who have have breast cancer that has become resistant to endocrine therapy. By comparing the findings from these studies with patients in the Hartwig dataset we hope to understand more about the mechanisms through which resistance develops.

Nick Turner The Institute of Cancer Research United Kingdom

In this study we will investigate the microbiome in metastatic cancer. We will use whole genome sequencing data from solid tumour cancers. We hypothesise that microbes travels together with the primary cancer cells to metastatic sites in the body and promotes tumoor growth and proliferation.

Robin Mjelle Norwegian university of science and technology Norway

Over the last decade we have made significant progress in understanding which mutations are acquired by healthy cells before becoming cancerous. These mutations are called somatic, because they are acquired after the individual was conceived. In parallele, we also know that each person has inherited a unique combination of DNA variants from their parents. These are called germline variants and can predispose individuals towards different cancer types. What we still do not understand is how germline and somatic genetic variants interact with each other. Understanding this could have important implications for prevention and early detection of cancer and the identification of new therapies.

Eduard Porta Pardo Josep Carreras Leukaemia Research Institute Spain

As part of the EUCANCan project, a federated network for harmonized genomic and phenotypic data sharing, we aim to foster cancer research and its application to precision medicine in a clinical environment which requires an accessible, accurate, and standardized methodology to evaluate the procedures involved in the identification of tumour variants in cancer patients. The lack of easy-to-access and exhaustive benchmarking datasets and procedures together with the intrinsic difficulty of the variant identification process make the implementation of a personalized medicine program in oncology practically impossible for centers without access to bioinformatic expertise and computational power. This project aims to utilize validated tumor patients data from the Hartwig Medical Foundation cohort and other cancer initiatives to provide a complete benchmarking system to homogenize and make variant identification methodologies compatible.

David Torrents Barcelona Supercomputing Center Spain

Several large initiatives have profiled cancer cell lines - lab grown models of disease - for their response to panels of drugs. These studies aim to identify genomic features that differentiate responders from non-responders. However, translating predictions back to patients has been challenging due to the differences between cell lines and patients, as well as experimental artifacts. We build upon previous work in integrating across studies from different labs to evaluate whether cross lab analysis results leads to better predictions. The genomic and treatment data collected by the Hartwig Foundation (HMF) will allow us to test predictions learned from cell lines.

Benjamin Haibe-Kains University Health Network Canada

We have shown that many human cancer types can be diagnosed using microbial DNA and RNA from patients’ primary tumors or matched blood (Poore et al., 2020. Nature). Our preliminary data suggests that a metastatic tumor’s tissue-of-origin can also be determined by examining its microbial constituents, as they may be located intracellularly (Nejman et al., 2020. Science). Using the HMF dataset, we hope to comprehensively evaluate if metastatic-associated microbes indeed mirror their primary tumor counterparts and how this may affect local immune responses and survival. The findings, if successful, could support a new type of cancer diagnostic and prognostic.

Rob Knight University of California San Diego USA

The goal of this project is to determine how germline mutations influence somatic mutation signatures, drug response, and survival. We will be analyzing germline, somatic mutation data, and survival data to validate novel findings of new germline mutations that shape the cancer genome and response to chemotherapeutics.

Manish Gala Massachusetts General Hospital USA

In terms of the research on genetic predisposition to cancers, recent advances in DNA sequencing technologies are facilitating the discovery of more and more genetic ‘variants’ that predispose to cancer. However, thwarting these efforts is that fact that the majority of these inherited ‘germline variants’ are located in the ‘non-coding’ regions of the human genome. This project is focused finding new germline (inherited) and somatic DNA mutations affecting the extended gene loci of known cancer predisposition genes, e.g. BRCA1, BRCA2 and ATM and also genes encoding members of the PI3K/AKT signalling and DNA damaging response pathways. We want to determine if these genetic variants predict which people are predisposed to cancer and which patients will respond to a group of drugs that target either the PI3K/AKT signalling or the DNA damage response pathways. A key step towards achieving this goal will be to access large datasets of cancer patient tumour DNA sequencing. We will identify from these datasets the non-coding or ‘regulatory’ mutations that exist in in the gene loci of cancer susceptibility and PI3K/AKT signalling or DNA damage response pathway genes. Once this is done, we will functionally characterise the most promising variants/mutations and use this knowledge to best predict those patients most likely to be predisposed to cancer and also those who we will predict to most likely respond to the targeted therapies.

Alexander Eustace Dublin City University Ireland

This request is an extenstion of DR-129 to include RNA data. The Personalised Breast Cancer Program (PBCP) is an initiative based at the Cancer Research UK Cambridge Institute and the Addenbrooke’s Hospital Breast Cancer Unit to sequence the DNA and RNA of breast cancer patients’ tumours. For every patient, their DNA was checked for mutations which could have an effect on which treatment would work best for them. Crucially, the results of the analysis are returned to the patient and their clinician within 12 weeks. However, most of the data in PBCP comes from primary breast tumours. We aim to understand how our findings from PBCP can be translated to metastatic tumours.

Carlos Caldas Cancer Research UK Cambridge Institute United Kingdom

Numerous environmental triggers have been found to leave a mutational footprint on the DNA of human cancers. In our lab, we want to identify how the treatments of pediatric cancer contribute to the development of second cancers. To robustly link a treatment to a mutational footprint found in pediatric second cancers, we want to investigate the mutations in as many tumors and tumor types that have undergone similar treatment and search for the same signature. Hereby, we identify the treatments with the highest contribution to the development of second cancers, which can be taken into account in new treatment regimens.

Ruben van Boxtel Prinses Maxima Centrum Netherlands

The effect on health related quality of life (HRQoL) of patients treated with or without personalized treatment based on WGS is yet unknown. HRQoL information is necessary in cost-effectiveness analysis, which informs decision makers regarding reimbursement of a new technology, e.g. WGS. In current project, we will compare HRQoL and utilities of patients who received targeted or no targeted treatment, based on WGS and standard diagnostic result. HRQoL data was obtained from patients participating in the CPCT-02 biopsy study. The current request is to match the WGS and clinical data of identical patients from whom we have HRQoL data. Valesca Retel Netherlands Cancer Institute the Netherlands

 

Cancer cells can have very high mutation rates, an effect known as genetic instability. This creates high levels of structural variation, where DNA is gained, lost, and rearranged over time, causing genes to be modified. This is important because versions of cells arise that can be resistant to cancer drugs. Structural variation and whole genome events occur in many solid tumours and understanding them is key to better diagnosis and treatment. In a similar way to how the weather is modelled we will use mathematical models and large amounts of data to try to understand the processes underlying structural variation.

Chris Barnes University College London United Kingdom

The prognosis of patients with metastatic colorectal cancer is poor. Some of these patients are eligible for receiving potentially life-saving immunotherapy, but only if their tumor is classified as having microsatellite instability (MSI). Thus identifying these patients is important. The ability of the tests, that are used in routine medical practice, to correctly classify metastatic tumors as MSI is unclear. Comparing routine test results on MSI status, derived from the Dutch PALGA database, with MSI status based on whole genome sequencing, i.e. gold standard, using data from the Hartwig Medical Foundation, offers a unique opportunity to clarify this important issue.

Petur Snaebjornsson Netherlands Cancer Institute the Netherlands

Cancer is a genetic disease, caused by changes in the cellular hereditary material. When a cancer tumor is detected, a common clinical practice foresees treatment with chemotherapeutic drugs. If the cancer relapses after this treatment, the resulting metastatic tumors will carry information about which of their genetic features made them resistant to the treatment. We aim to analyze the data from metastatic tumors contained in the Hartwig Medical Foundation cohort in order to investigate connections between the tumor genetic makeup and its response to therapy.

Donate Weghorn Centre for Genomic Regulation (CRG) Spain

Cancers develop by the accumulation of "driver" genetic alterations providing tumor cells a growth advantage. Whole genome sequencing allows to identify these driver events, but also the mechanisms at the origin of genome insults. Our project has two major aims. Firstly, we will use deep learning approaches to discover driver alterations in the non-coding part of the genome that remains poorly understood. Secondly, we will analyze mutational signatures to discover genomic instability phenotypes related to treatment response. Leveraging the tremendous resource of HMF genomic and clinical data, we hope to identify new oncogenic mechanisms and therapeutic vulnerabilities. Eric Letouze Cordeliers Research Center France

 

We are investigating drug resistance mechanisms in two different types of breast cancer, 1) invasive lobular carcinoma (ILC) to PI3K/mTOR inhibitors and 2) BRCA-deficient breast cancer to PARP inhibitor. To this end, we used in-vivo mouse models mimicking human ILC and BRCA-deficient breast cancer and obtained drug-resistant tumors through prolonged treatment of mTOR inhibitor and PARP inhibitor. Genomic and proteomic analyses of the treatment-naive/resistant tumors revealed several resistance factors, which were further experimentally validated by in-vitro experiment. To strengthen and validate our findings in clinical relevance, we believe that the HMF WGS, RNA-seq and treatment response data provide valuable resources.

Daniel Zingg Netherlands Cancer Institute the Netherlands

Modern machine learning techniques will be applied to the data in an exploratory study in order to identify possible new insights into chemotherapy response. In particular genomic information will be combined with CT and MRI images to identify common morphological characteristics of different pathways.

Sean Benson Netherlands Cancer Institute the Netherlands

Immune checkpoint inhibitors have been approved for first- and second-line treatment of metastatic urothelial cancer (mUC) patients, and lead to durable clinical responses in a small subset of patients. As a consequence many patients are being exposed to ineffective treatment with the risk of developing (severe) side effects. The aim of this study is to identify potential predictive markers for clinical benefit of immunotherapy. We will use the genomics data to identify predictors based on the sequencing but we will also combine it with other modalities such as staining of tumor biopsies and liquid biopsies. Furthermore, mechanisms underlying primary and acquired resistance to immunotherapy will be studied, potentially facilitating the development of improved (combinatorial) treatment strategies for mUC patients.

Martijn Lolkema Erasmus MC the Netherlands

We aim to understand the effects of the FHIT gene in cancers. FHIT is involved in purine metabolism, and contains a fragile site (FRA3B), leading to translocations which have been associated to many types of cancers. Given its function, FHIT is a good candidate gene for cancer. However, previous genetic studies on this gene were small and not conclusive. Here, we will perform a large and comprehensive genetic study shedding light on FHIT role in cancer.

Claudio Toma Centro de Biología Molecular Severo Ochoa (CBMSO) Spain

The prognosis of patients with metastatic colorectal cancer is poor. Some of these patients are eligible for receiving potentially life-saving immunotherapy, but only if their tumor is classified as having microsatellite instability (MSI). Thus identifying these patients is important. The ability of the tests, that are used in routine medical practice, to correctly classify metastatic tumors as MSI is unclear. Comparing routine test results on MSI status with MSI status based on whole genome sequencing, i.e. gold standard, offers a unique opportunity to clarify this important issue. Therefere, coupling to external parties (PALGA and NKR) is part of this project.

Petur Snaebjornsson Netherlands Cancer Institute the Netherlands

RET fusion gene testing will soon become very important for selection of lung cancer patients for treatment with the novel tyrosine kinase inhibitor, selpercatinib. In our diagnostic setting we observed that RET in situ hybridization (ISH) testing, a molecular test that is used worldwide for screening lung cancer for genomic breaks at the RET locus often yields false-positive results. As a result patients might be(come) treated wrongly. To explain the many false-positive RET ISH results, we would like to investigate WGS data for the presence of structural variations at the RET locus (versus ALK and ROS1 loci) in lung cancers.

Erik Jan Dubbink Erasmus MC the Netherlands

Purpose: To develop an algorithm for robust estimation of the tumor mutational burden (TMB) using targeted sequencing. Data: Somatic analyses data DNA (SNVs, INDELs, SVs, Copy-numbers, MSI+/- etc) from the ~4000 metastatic cases available in the Hartwig Medical Database including the 2,520 cases recently published in Nature (P. Priestley et al. Nature, 2019). Clinical and outcomes data from the patients in the Hartwig Medical Database that have been treated with immunotherapy.Relevance for patient: Immunotherapy, that is, reactivation of the patient’s own immune system to kill the cancer cells, has revolutionized treatment of advanced cancer of various origins. However, robust estimation of TMB by targeted sequencing is crucial to accurately identify patients with an elevated TMB that may benefit from immunotherapy.

Johan Lindberg Karolinska Instutet Sweden

Breast cancer (BC) is the most common malignancy among women, constituting the prominent cause of cancer-associated mortality. Despite considerable knowledge of the BC molecular subtypes, there is no strategy for accurate stratification of BC patients. The main purpose of this project is to identify genomic alterations that differentiate patients, who respond and who don’t respond to a particular treatment type. For this purpose, we will perform whole-genome sequencing data analyses on BC patients genomes downloaded from Hartwig Medical Foundation and build clinical predictor. This tool will help in treatment decision-making, indicating benefits and risks connected with a particular treatment type.

Pawel Zawadzki Adam Mickiewicz University in Poznań Poland

Glioblastoma multiforme (GBM) is the most lethal brain tumour. Mutations in certain genes have been implicated in GBM. However, the DNA sequence encoding for genes (coding regions) only represent 1% of the tumour DNA. We aim to use the whole genome sequencing (WGS) and clinical data from the Hartwig’s recurrent GBM cohort, together with other publicly available datasets, to characterise the mutations in the non-coding regions in GBM patients, and to determine if certain mutations in these regions can predict survival or response to therapy. We are particularly interested in structural variants (SVs) and require BAM/CRAM files to generate comparable new SV calls across all datasets (ie between Hartwig dataset and other datasets generated elsewhere), as we have recently for a combination of previously published and new ovarian tumour BAM files (see https://www.biorxiv.org/content/10.1101/2020.05.11.088278v1).

Colin Semple The University Court of the University of Edinburgh United Kingdom

We aim to apply computational tools to Hartwig Medical Foundation cancer genomes to establish the order in which harmful mutations were acquired, and approximately when they were acquired during a patient’s lifetime. Furthermore, we aim to infer what caused the mutations (for example, smoking, UV light, etc.), and estimate when the cell was exposed to these damaging processes. Across many cancer samples, this approach lets us build a picture of how different cancers evolve over time. By comparing these results to earlier analyses we performed on primary tumours (observed at initial cancer diagnosis), we seek to understand the development of cancer as it spreads from the initial site to another site (known as metastasis). This will show whether tumours that metastasise evolve in the same way as non-metastatic cancers, and may give insight that helps to predict the metastatic step in cancer evolution.

Peter van Loo The Francis Crick Institute United Kingdom

There are different types of breast cancer, one is referred to as Invasive Lobular Carcinoma (ILC). ILC accounts for around 10% of breast cancers, and while considered a distinct tumor subclass, these patients are treated identical as the predominant subclass Invasive Ductal Carcinoma (IDC). ILC cancers have a very specific feature: almost all of these tumors have a mutation in a gene called Ecadherin. Using this mutation to identify ILC, we aim to compare ILC and IDC breast cancer. With a better understanding of ILC, future therapeutic options could be better finetuned for this distinct breast cancer subclass.

Wilbert Zwart Netherlands Cancer Institute the Netherlands

Many tumors show mutations of kinase proteins. When these kinase proteins are mutated, they can provide a catastrophic effect because this favors cell division over cell death of tumor cells. Since every patient can have multiple kinase mutations, this provides options to inhibit these kinases simultaneously. Since kinase inhibitors commonly inhibit multiple kinases at once, their multiple targets could be optimally matched to individual patients. By using laboratory experiments, we have found that simultaneous inhibition of multiple mutated kinases can lead to more effective therapies. Here we aim to provide a proof of concept that optimally matching drug-targets to patients with multiple kinase mutations can lead to more effective therapies.

Bart Westerman VU Medical Center, Amsterdam UMC the Netherlands

See also HMF-DR-031 as this study will be part of DR-031.This project aims to assess the porportion of patients with mUC who have pathognomic germline mutations.

Joost Boormans Erasmus MC the Netherlands

In this project we aim at combining mathematical modelling of tumour evolution with cancer genomic data to measure fundamental evolutionary characteristics of metastatic tumours. These measurements will allow better understanding of how tumours spread to distant parts of the body. Importantly, we anticipate that the patient-specific ‘evolutionary rules’ we will measure from the data may be informative in terms of stratifying patients between good versus poor prognosis. Moreover, some of these measurements may also be predictive of the effectiveness of certain treatments.

Andrea Sottoriva The Institute of Cancer Research United Kingdom

For prostate cancer, identifying patients at risk for developing aggressive disease is clinically challenging [1]. This project aims to discover biomarkers prognostic for disease progression and predictive of treatment response.CPCT-2 DNA and RNA sequencing data.Clinical implementation of prognostic and predictive biomarkers will help personalize therapy, reducing the burden of uneffective treatments and improving patient survival.

Niven Mehra Radboudumc the Netherlands

Recent studies emphasize the importance of combined effects of germline and somatic alterations in cancer. The aim of the project is to study how germline and somatic alterations contribute together to gene expression and drive glioma development. We will use next-generation sequencing data derived from matched normal and tumor samples of glioma patients. The data will be integrated with tissue- and tumor-specific data of regulatory elements to identify cancer-driving gene regulatory elements. The study is expected to produce novel findings, which have potential to be used in diagnostics and therapeutic applications to improve patient care.

Matti Nykter Tampere University Finland

Cancers change over time. Understanding how cancers change is important for determining a patient’s prognosis and to improve the effectiveness of treatment. It is not possible to directly watch a cancer change, and so instead we have to infer the changes using the patterns of mutations across the cancer genome. These mutational patterns are analogous to the rings inside a tree trunk that are a ’secret diary’ of how the tree grew. We have developed mathematical tools to read the patterns and work out how cancers grow and change, and we apply these tools to study the rich data within the Hartwig dataset.

Trevor Graham Queen Mary, University of London United Kingdom

While mutations affecting protein-coding regions have been examined in prostate cancer, structural variants at the genome-wide level are still poorly defined. Here, we try to develop a new structural variants (SVs) detection tool in whole-genome sequence (WGS) data to identify complex structural variants in prostate cancer and characterize the role of SVs in metastatic tumor and gene expression aberrations. In addition, we will also investigate the impact of SVs in patient clinical phenotypes.

Wei Li University of California, Irvine USA

Cancer genomes harbor a substantial number of mutations, but only some contribute to tumor development. These so-called driver mutations serve as targets for new cancer therapies. Over the past years, the search for these driver mutations has been mostly focused on coding regions. The goal of this project is to develop a statistical framework to systematically identify driver mutations in non-coding regions. We will apply this approach to all whole-genome sequencing data in the HMF database, and characterize which mutations in non-coding regions are relevant to carcinogenesis. We will then follow up on these findings using clinical and expression data.

Eliezer Van Allen Dana-Farber Cancer Institute USA

The discovery of driver mutations and mutational processes in melanoma is complicated by high background mutation rates. With expanded melanoma cohorts, it is possible to expand discovery through examination of noncoding mutations and larger somatic events that result in oncogenesis specific to clinically and genomically defined melanoma subtypes. The overarching goal of this project is to examine cancer drivers and mutational processes (“signatures”) in melanoma using computational methodologies and compare these findings within melanoma subtypes. We will apply these frameworks to whole-genome sequencing data in the HMF database and examine signatures for associations with orthogonal somatic molecular and clinical data.

Eliezer Van Allen Dana-Farber Cancer Institute USA

In urothelial cancer, prognosis and responsiveness to treatments, such as chemotherapy and checkpoint inhibitors, differs substantially between patients. Although urothelial cancer is known to be an immunogenic tumour, little is know about the prognostic and predictive value of the immune infiltrate in metastatic urothelial cancer. Here, we will use RNA sequencing data to infer immune cell composition and study its relation with clinical outcome, micronenvironmental factors and long non-coding RNAs. Identification of prognostic or predictive biomarkers is needed to tailor treatment to the individual patient. Moreover, a better understanding of factors that influence the immune infiltrate might provide a rationale for new treatment strategies.

Niven Mehra Radboudumc the Netherlands

In the past two decades, the exploration of molecular mechanisms and genetic information in cancer led to the new approaches for treatment. Despite its undesirable consequences, conventional chemotherapy is still the first-line treatment in metastastic cancer. Nowadays, targeted therapies (monoclonal antibodies and small molecules) are approved for some tumour types. Resistance to these agents is a challenging problem in the clinic and could be a result of activation of alternative molecular pathways involved in tumourigenesis. This finding highlights the need for further research to optimize treatment with targeted therapies by combining them in patients with more than one actionability.

Neeltje Steeghs Netherlands Cancer Institute the Netherlands

The main reason that cancer is lethal lies in its ability to spread in the body, a process called metastasis. The immune system is important in determining tumour development and treatment response. We will conduct in-depth analyses of metastatic cancer by using the transcriptomic data to understand how immune and stromal cell gene expression may be correlated with changes in tumour size and patient outcomes. We will also determine the key genes that the alteration of which may improve disease outcome, which may provide potential therapeutic targets.

Binzhi Qian The University of Edinburgh, QMRI United Kingdom

Despite the advancements in lung cancer treatment in the last decades, most patients diagnosed with non-small cell lung cancer (NSCLC), still suffer from a severe disease characterized by rapid progression and low survival rates. All cancers are caused by genetic alterations in healthy cells. Lung cancer is characterized by a high number of these genetic alterations. We would like to explore the genetic abnormalities that can be found in lung tumors. These abnormalities can help us predict which patients are likely to respond to different therapies. Furthermore, the analysis can probably lead to the discovery of new targets for patient specific therapy.

Joachim Aerts Erasmus MC The Netherlands

Cancer is caused by mutations that occur over time via genetic and environmental exposures. Some mutations called drivers unlock oncogenic cellular properties, while most mutations are neutral passengers. A complete driver catalogue is important for developing therapies, while passengers reflect tumor evolution and exposures. Most known drivers affect protein-coding genes while the non-coding genome is less characterized. Here we use our machine-learning tools to study the non-coding genomes of metastatic tumors. We will compare drivers and passengers to understand the emergence of metastasis and investigate regional mutation processes to decipher tumor evolution and its exposures at functional genomic elements.

Jüri Reimand Ontario Institute for Cancer Research Canada

In this project we will investigate to what extent the cooperation of inherited polymorphisms within regulatory regions and somatic aberration patterns affects evolution of cancer by altering oncogenic signalling pathways and promoting disease progression. Whole Genome Sequencing (WGS) data of metastatic cancer patients (tissue and matched control) will be analysed and compared/integrated to data from other sources (PCAWG, GTeX and ENCODE). This project will lead to the identification of germline biomarkers capable to announce the emergence of specific cancer subtypes and driver cancer aberration patterns, hence help stratifying patients with potential aggressive progression.

Alessandro Romanel University of Trento Italy

The Personalized OncoGenomics (POG) Program sequences cancers from patients with advanced and metastatic tumours to determine which therapies are most likely to help combat their disease. Analysis of the complete tumour DNA sequence and gene activity has revealed how these tumours are affected by the therapies that patient have received and can potentially inform future clinical decision making. In this proposal, we aim to use the sequence data from the Hartwig Medical Foundation to aid in interpretation of patient findings, and also to validate novel observations in advanced cancers.

Steven Jones BC Cancer, Part of the Provincial Health Services Authority Canada

Using mouse models, we found certain variants of the fibroblast growth factor receptor 2 gene (Fgfr2) to be strong tumor drivers. These oncogenic Fgfr2 variants all lack the last exon resulting in shortened FGFR2 proteins. Importantly, cells expressing shortened FGFR2 are very sensitive to inhibitors blocking FGFR2 signaling. Moreover, preliminary data suggest that human cancers might also recurrently harbor genetic alterations of the FGFR2 locus presumably producing shortened FGFR2. To strengthen and validate these findings, the Hartwig Medical Foundation whole-genome sequencing and RNA sequencing datasets represent valuable resources. Ultimately, these findings might refine the inclusion criteria for FGFR2-targeting precision therapies.

Daniel Zingg Netherlands Cancer Institute the Netherlands

HecoPerMed seeks to improve health economic models for personalized medicine, both payment models and evaluation (HTA) models. We seek to develop a model to assess the cost-effectiveness of new tumor agnostic treatments, and the data of the Hartwig medical foundation could help establish the current survival of patients, so we can investigate how good the new treatment is relative to current care.

Irene Santi Erasmus University Rotterdam The Netherlands

We plan to investigate the evolution of lung tumours at different stages of cancer development. The molecular data used to track cancer evolution gives us information about the key genes and processes involved in the tumour's ability to overcome the patient’s immune system and external treatments. With the data from the HMF we additionally plan to investigate such processes at later stages of cancer development. Specifically, we will explore the role of genes and processes relevant at early stages of cancer evolution and at later stages, after treatment. Ultimately, these data will inform the identification of new targets for a more personalised therapy and prognosis.

Nicholas McGranahan University College London United Kingdom

Targeted cancer therapies can be remarkably effective but a considerable number of patients develop resistance as treatment proceeds. Resistance can be caused by specific mutations in cancer cells. Using computational analysis we have shown that a significant fraction of these resistance mutations are predicted to have immunogenic potential across a large proportion of the healthy human population. We intend to use the Hartwig Medical Database to verify if this holds true also in a patients’ population. If immunogenic in patients, recurrent resistance mutations could be extremely interesting in the context of off-the-shelf precision immunotherapies such as therapeutic cancer vaccines.

Stefano Lise The Institute of Cancer Research United Kingdom

Cancer is a complex disease, characterized by a wide spectrum of genome abnormalities. We aim to describe the order and magnitude of carcinogenic events (mutations and copy number changes). This will allow us to construct evolutionary trajectories for individual cancer types. Using the Hartwig Medical Foundation’s data on metastatic cancer samples, we wish to explore the last step in the transformation of the normal cell to lethal disseminated metastatic cancer. By deciphering cancer evolutionary trajectories we may be able to develop tools that can predict how cancer evolves in the presence of therapy, and determine cancer phenotypes associated with response. This may allow for improved patient treatment through stratification.

Nicolai Juul Birkbak Aarhus University Denmark

"After treatment with a combination of chemotherapy and radiotherapy for esophageal cancer, the current standard treatment is to continue with resection. However, in circa 20% of patients, no vital tumor cells are found in the resection specimen. On the other side, circa 10% of patients develop metastases shortly after esophageal resection. These patients could have possibly been spared esophagus resection. A nation-wide study, initiated by the department of surgery of the Erasmus MC, is currently recruiting patients (the SANO study, PI prof. dr. van Lanschot) and aims to clarify the role of esophagus resection in patients with no radiological or pathological (on biopsy) evidence of disease after pre-operative chemoradiation. A rapidly developing minimally invasive method to improve detection of minimal amounts of tumor cells or fragments could help to identity those patients in whom true cure has been accomplished. Therefor, we are currently biobanking blood of the patients in the SANO trial to serially measure tumor DNA in the blood To come to a comprehensive panel of genetic variants to be measured in the blood, we wish to incorporate the genetic variants found in the metastases of esophageal cancer patients, as these variants are expected to be the drivers of resistance to treatment and disease recurrence. Also, we want to explore the genetic characteristics of metastases of esophageal cancer. Discovering new genetic variants in metasatses that have not been identified in primary tumors could lead to new therapeutic approaches. Optimizing the blood assay could lead to a new and minimally inavsive means to monitor tumor response."

Bianca Mostert Erasmus MC The Netherlands

"Since almost 10 years, the outcome of patients with metastatic melanoma has improved significantly and long term durable responses have been realized. However, it remains difficult to predict which patient will benefit from therapy and besides that, severe adverse events occur in about 30-50% of the patients. The aim of this project is to describe the genomic landscape of melanoma patients and to incorporate this into immunological and clinical analyses which we are currently performing. More insight into immunologic and genomic alterations in these patients could contribute to improved and more individualized therapeutic decision making."

Astrid van der Veldt Erasmus MC The Netherlands

The Personalised Breast Cancer Program (PBCP) is an initiative based at the Cancer Research UK Cambridge Institute and the Addenbrooke’s Hospital Breast Cancer Unit to sequence the DNA and RNA of breast cancer patients’ tumours. For every patient, their DNA was checked for mutations which could have an effect on which treatment would work best for them. Crucially, the results of the analysis are returned to the patient and their clinician within 12 weeks. However, most of the data in PBCP comes from primary breast tumours. We aim to understand how our findings from PBCP can be translated to metastatic tumours.

Carlos Caldas University of Cambridge United Kingdom

"Androgen receptor inhibitors (ARi) are frequently used to treat metastatic prostate carcinomas. In this project, we focus on identifying genetic causes for resistance to these new ARi to facilitate the development of new treatment options. DR-127 is an extension of DR-013"

Michiel van der Heijden NKI-AvL the Netherlands

We aim to understand the molecular mechanisms leading to the high degree of mutability observed in certain types of tumors and to clarify how these same mechanisms can influence tumor evolution and therapeutic response, with the ultimate goal to effectively tailor treatment to each individual tumor patient. We will use the Hartwig Medical Foundation (HMF) WGS cancer dataset to compare the type, number and genomic distribution of somatic mutations across different tumor subtypes, ultimately gaining insights into the precise factors that lead to tumor initiation, growth and evolution.

Edison Liu The Jackson Laboratory USA

We recently finalized an extensive genomic analysis of breast and prostate cancer, which provided important novel insights into the biology of these diseases and identified clinically relevant subgroups of patients who might benefit from specific therapies. With this project we aim to identify how transcriptome changes deepen our understanding of disease progression and improve predictive subgrouping. This will be achieved through pan-cancer analysis of different RNA types (mRNAs, lncRNAs, circRNAs, fusions, etc.) and integration with the genomic changes and clinical data. The RNA-Seq analyses will identify novel cancer-associated transcripts and new predictive and prognostic genes, strengthening classification on genome-level. Our ultimate goal is to improve personalized cancer treatment for patients with metastatic cancer. John Martens Erasmus MC The Netherlands

 

Priestley et al. (2019), the study introducing the Hartwig Medical Foundation data collection, used several state-of-the-art analysis tools for calling different types of somatic variation. As a result, they provided a number of new candidates of germline and somatic variation with a potential functional role in metastatic cancers. We have recently developed a new variant caller that provides a unified probabilistic framework for calling different types of variation from short substitutions and indels up to large-scale copy-number events. With it, we would like to further refine variation identification to aid in prioritization for functional testing.

Johannes Koster University Hospital Essen Germany

HER2-positive breast cancer is an aggressive subtype driven by amplification of the Her2/Erbb2 gene. This oncogene is thought to be amplified via chromothripsis, a catastrophic event that causes massive genomic rearrangements. Characterizing chromothripsis in HER2-positive breast cancer is an important step towards improved therapeutic targeting. We will use whole-genome sequencing data to identify structural rearrangements at single nucleotide resolution and infer mechanisms of chromosome shattering and repair. We hope to understand whether chromothripsis associated with poor outcome in HER2-positive breast cancer, the prevalence in metastatic vs. non-metastatic tumors, and how treatment influences chromothripsis.

Christina Curtis Stanford University School of Medicine USA

Oesophageal adenocarcinoma (EAC) is an aggressive cancer with a poor overall prognosis. Our data from shallow sequencing of 95 patients with EAC show a worse prognosis in patients who present a homozygous germline deletion of ADAM3A gene, treated with chemoradiotherapy according to CROSS schema. Approximately 20% of the European population shows deletion in this region, however the frequency and consequences of this deletion are not yet well characterised in patients with cancer, in general. The aim of this data request is to investigate the association of ADAM3A germline and somatic deletions in patients with metastatic cancer and to elucidate the role of this deletion in prediction of clinical outcome in patients treated with different chemo(radiation) regimens.

Liudmila Kodach NKI-AvL the Netherlands

Systematic analysis of cancer genomes has revealed a number of characteristic mutational signatures. Some of these have been linked to endogenous processes induced by inherited or somatic pathogenic variants, or by exogenous challenges, like exposure to UV-light, chemotherapeutic drugs or other types of mutagenic agents. In this research, we will explore, characterize and validate mutational patterns, signatures and processes in primary and metastatic cancer patients. The results of this study will provide a better understanding of the processes that are operational at different stages of carcinogenesis. Such information may be used for design of preventive strategies (for tumorgenesis or metastatic seeding) but can also be used for stratifying patients towards targeted treatments based on such characterics (e.g. homologous recombination deficiency as we have already demonstrated by development of the CHORD algorithm, DR-018).

Edwin Cuppen University Medical Center Utrecht the Netherlands

Pancreatic cancer and esophagogastric cancer are among the most deadly forms of cancer: despite intensive treatment with chemotherapy and / or surgery, the majority of patients die from the disease. In PEGASUS we investigate whether molecular techniques are of value in predicting which patients will benefit from such intensive treatment, both with regard to the chance of survival as well as quality of life. We are also investigating whether there are molecular targets that could improve treatment even further with patient tailored drugs. We will weigh the costs of using molecular techniques against the benefits and risks for the patient.

Hanneke van Laarhoven Amsterdam UMC the Netherlands

If we are able to better predict who will respond to treatment, this could have major implications for clinical practice. By better identification of patients who will benefit from treatment, we would prevent patients from undergoing aggressive treatment in the end stage of their lifes from which they will not benefit, with as an added benefit reduced health care costs.

Harry Groen UMCG the Netherlands

Immunotherapy can achieve great results for NSCLC patients, yet this costly therapy is only efficient in a small number of patients. Our goal is to see whether RNA data, being a level closer to the proteins produced than DNA, can be used to identify which patients will respond. To do so, we plan to combine RNA expression data with clinical data. RNA data to identify which genes are relevant, and clinical data to minimize confounding by other factors and ofcourse to identify responders and non responders.

Harry Groen UMCG the Netherlands

"Glioblastoma multiforme (GBM) is the most common and aggressive primary brain tumor in adults. Since 2005, standard treatment consists of surgery followed by radiation therapy with concurrent and adjuvant temozolomide chemotherapy, taking ± 8.5 months in total. Despite optimal treatment, GBM outcome is almost inevitably fatal, usually within 16 months after diagnosis. There is an urgent need to improve treatment of patients with GBM, both by selection of patients most likely to respond to standard therapy, as well as through the identification of novel targets for [combination] treatment. DR-093 is an extension of DR-060 and DR-076" Filip de Vos University Medical Center Utrecht the Netherlands  

 

This project concerns a feasibility study, in which tumor biopsies collected in different hospitals in the CPCT-02 trial will be used to perform extensive protein analysis. The results of the DNA analysis and protein analysis will be compared to see how changes in the tumor DNA translate to the protein profile in the tumor, for better understanding of tumor biology. In the future, we hope to develop an integrated analysis of tumor DNA and protein profile, in order to better predict which patients will benefit from targeted therapy. Mariette Labots Amsterdam UMC the Netherlands

 

"Colorectal cancer (CRC) is the second most common cause of cancer death worldwide. Cancer is caused by DNA alterations. Detailed information about DNA alterations in tumors from clinically well-characterized CRC patients offers excellent opportunities for translational research. The main purpose of this project is to facilitate querying and viewing of the HMF CRC data by members of the HMF focus group, allowing individual researchers to easily judge whether the HMF CRC dataset is (not) suited to address their specific research question(s). Data: • DNA whole genome sequencing data from CRC samples, indicative for the landscape of genomic aberrations in CRC patients. • RNA sequencing data, indicative for biological consequences of DNA alterations. • Clinical and pathological data, to relate DNA mutations and RNA expression data to clinical characteristics. Relevance for patient: Easy querying and viewing of HMF CRC data in cBioPortal will stimulate the use of HMF CRC data for translational research purposes. Facilitating initiation of discovery and validation studies will ultimately lead to better disease management of CRC patients."

Remond Fijneman NKI-AvL the Netherlands

"Purpose: Colorectal cancer (CRC) is the second most common cause of cancer death worldwide. Cancer is caused by DNA alterations. Little is known about the prevalence and biological and clinical relevance of genomic regions that are affected by chromosomal breakpoints. The aim of this project is to learn more about the prevalence of chromosomal breakpoints in colorectal cancer, and their biological and clinical consequences. Data: DNA whole genome sequencing data from CRC samples, to determine the landscape of genomic regions affected by chromosomal breakpoints. RNA sequencing data, to determine biological consequences of DNA breakpoints; and clinical data to relate DNA mutations to clinical characteristics. We will use lung cancer and breast cancer data in addition to CRC data, to determine organ-specificity of the chromosomal breakpoints. Relevance for patient: Better knowledge about DNA mutations will lead to biomarkers for better diagnosis, prognosis, prediction of therapy response, and treatment monitoring of CRC patients."

Remond Fijneman NKI-AvL the Netherlands

Biliary tract cancer (BTC) is malignancy of bile ducts (also called cholangiocarcinoma) and gallbladder. BTC is a rare malignancy in Western countries with a poor prognosis. Surgical resection in the only available treatment that potentially can cure patients with BTC, but most patients present with advanced BTC (including metastatic disease) that can not be completely removed by an operation. Chemotherapy is the standard treatment for those patients with advanced BTC. Treatments with agents directed to DNA changes (targeted therapy) did not show better survival in different studies because of the heterogeneous nature of BTC. To improve the treatment of advanced BTC, a more comprehensive overview of the genomic changes in advanced BTC is required. If we can identify subgroups of patients with different prognosis based on genomic changes, new clinical studies could be initiated to study a specific treatment in one or more of these subgroup of patients.

Heinz-Josef Klümpen Amsterdam UMC the Netherlands

Estrogen Receptor is a hormone-responsive protein, responsible for breast tumor growth. We found that DNA-binding specificity can predict treatment response, and profiled estrogen receptor/DNA interactions in over 100 breast tumors. Now, we aim to determine whether these bound-regions are mutated in breast cancer, and how this impacts tumor biology. For this, we request access to data from the metastatic breast cancers sequenced at the HMF. In metastatatic breast cancer, estrogen receptor remains critically important. We know that this DNA occupancy differs between patients, whether mutations at these sites alter biology is unknown. This information may explain why some tumors are resistant to therapy, and how we can resolve this.

Wilbert Zwart NKI/AvL the Netherlands

Mutations are generated by endogenous (replication errors) and exogenous processes (UV light). Studying the mutational patterns observed across both genes and the entire genome of metastatic patients, we plan to elucidate the interplay of endogenous and external factors that ultimately shape the mutational processes. Based on the knowledge acquired on mutational processes we will improve the methods to detect signals of positive selection in both coding and non-coding elements, in order to accurately identify those involved in tumorigenesis.

Nuria Lopez Bigas IRB Barcelona Spain

Our group produces WGS data for 100 Non-Melanoma Skin Cancers (NMSC) in order to characterise mutational landscape and driver mutations in relevant tumor types. Produced data will be compaired with the mutational landscapes of publically available skin cancers genomes. Understanding of the mutational process in skin cancers which are characterised by high mutation rates as well as characterisation of the mutated cancer pathways may be useful for chosing optimal treatment strategy.

Sergey Nikolaev Gustave Roussy Cancer Campus France

We have mapped in metastatic prostate cancer how genes are being regulated. Now, we aim to combine these results with mutations and gene activation status in these and other tumor samples, aimed to better understand how prostate cancer gene regulation and mutations are connected. Metastatic prostate cancer samples.

Wilbert Zwart NKI/AvL the Netherlands

Mutational signatures are patterns of mutations that inform us on the mutational process underlying tumorigenesis. We will investigate the mutational signatures caused by chemotherapeutics, to better understand how chemotherapy affects cancer cells. We will use all the sequencing data from tumors that were treated with chemotherapeutics prior to surgery to discover the mutational signatures. Mutations induced by chemotherapy can play a role in how cancers develop resistance to the chemotherapy, therefore understanding where those mutations are likely to form will help us understand how resistance could develop. This knowledge can guide us in avoiding/overcoming resistance to chemotherapy.

Steve Rozen Duke-NUS Medical School Singapore

An aggressive subtype of colorectal cancer (CRC) harbours activating mutations in the BRAF oncogene and inactivating mutations in the RNF43 tumor suppressor gene. The purpose of this study is to assess BRAF and RNF43 copy number variation (CNV) and mutant allele frequency (MAF) in these tumors. Using whole genome sequencing data from BRAF/RNF43 mutant tumors we aim to assess whether selective amplification of the mutant BRAF allele is a frequent event in mCRC, and how this relates to loss of RNF43. Such genetic events may have an impact on the tumor response to drugs targeting the MEK/ERK and WNT pathways.

Onno Kranenburg University Medical Center Utrecht the Netherlands

The treatment of bladder cancer patients with immunotherapy is a very promising and effective treatment. However, only a minority of patients responds to the treatment with only immune therapy (up to 30%). These patients benefit from this treatment and do not need other treatments, whereas other patients do need some other form of treatment. We have performed a study: the RESPONDER study to collect multiple different tissues from patients including a CPCT-02 biopsy. In this proposal we aim to combine the data on immune monitoring/ circulating tumor derived DNA (ctDNA) analysis and clinical outcome with the genomic data obtained by the HMF to improve the identification of those patients that are at risk of failing treatment and those patients that could be spared from potentially toxic combination treatments.

Martijn Lolkema Erasmus MC the Netherlands

The purpose of our project is to derive genomic signatures based on presence of structural rearrangements and copy number from high resolution WGS data of advanced prostate cancer biopsies. The data we plan to use are the 212 high resolution WGS from the Lisanne F. van Dessel, et. al. study (The genomic landscape of metastatic castration-resistant prostate cancers using whole genome sequencing reveals multiple distinct genotypes with potential clinical). We aim to characterize the presence of specific signatures associated with the presence of mutations in DNA repair genes; with the final goal of using them as biomarker for predicting which patients could benefit from specific treatments.

Joaquin Mateo Vall d’Hebron Institute of Oncology Spain

The main purpose of the research project is to develop new diagnostic biomarkers that will be used for personalised therapy selection in glioma patients. We will download the genomes of Central Nervous System (CNS) tumors from the Hartwig Medical Foundation and perform Whole-Genome Sequencing (WGS) data analyses with particular emphasis on the assessment of development of new diagnostic tool predicting the response to temozolomide. This data mining will also demonstrate the trajectories of cancer development, for example genome-wide response to inflammation and molecular mechanisms responsible for the origin of the mutational signatures.

Paweł Zawadzki Adam Mickiewicz University Poland

The main purpose of the research project is to understand the mechanism of interplay between DNA repair pathways associated with acquired chemotherapy resistance. We will download the genomes of ovary cancer patients from the Hartwig Medical Foundation and perform whole-genome sequencing data analyses with particular emphasis on the assessment of DNA repair pathways and its correlation with clinical outcomes. This data mining will reveal the genetic differences between sensitive and resistant tumors and a dominant mechanism that provides the resistance in clinics.

Paweł Zawadzki Adam Mickiewicz University Poland

The nature and impact of genomic changes induced by cisplatin treatment of germ cell tumors is unknown. This is relevant to identify mechanisms of treatment sensitivity and resistance. Genomic data on metastatic germ cell tumors will be compared with primary (non-metastasized) cases, supported by cell line-based observations. Recent data demonstrate that heterogeneity exist in the genomic constitution of primary human germ cell tumors before chemotherapy treatment. Selection of a defined subclone is likely related to treatment resistance, which might help to identify the mechanism involved, being the basis for development of alternatve treatment strategies.

Laurens van der Vlier Princes Maxima Centrum the Netherlands

While localized prostate cancer can be cured, metastastatic disease is associated with high morbidity and mortality. Prostate cancer typically metastasizes to lymph nodes and the bone, while metastases to organs like liver, lung or the peritoneum are less common. This project is aimed to describe what the differences are in the genomic (DNA) make up of prostate cancer metastases in the different sites. Knowledge hereof might spark further research into ‘site-preference’ and ‘site-specific alterations’ of the prostate cancer metastases genome, which ultimately might result in means to better treat or even prevent metastases.

Andre Bergman NKI/AvL the Netherlands

"Venous thromboembolism (VTE), a venous blood clot, frequently occurs in cancer patients, and substantially affects morbidity, quality of life, and mortality. Yet, it is incompletely understood why cancer patients are at increased risk of this disease. The risk of VTE varies greatly across cancer types, strongly suggesting a role for cancer-specific oncogenic mutations which indirectly cause thrombosis. We aim to investigate this possible association by using clinical data and genomic data from tumor biopsies acquired within the CPCT-02 study. Results could help understand the pathogenesis of cancer-associated venous thromboembolism, lead to new therapeutic targets, and identify high-risk patients for thromboprophylaxis." Harry Büller Amsterdam UMC the Netherlands

 

The introduction of new therapeutic options, i.e. antiangiogenic drugs and immunotherapy, has significantly changed the treatment and perspectives of patients with advanced and metastatic renal cell cancer (RCC). Currently, the choice for first-line treatment is based on a risk calculation model, using only clinical parameters such as performance status. The aim of the current project is to describe the genomic landscape of RCC patients focusing on the immunogenic and angiogenic mutations/genes. More insight into immunogenic and angiogenic driver mutations together with the genomic landscape of RCC may contribute to improved and more individualized therapeutic decision making.

Astrid van der Veldt Erasmus MC the Netherlands

We aim to stratify cancer types by the degree of intratumour heterogeneity demonstrated both between metastases (when the cancer spreads) and primary disease (observed at the initial cancer diagnosis) as well as within the metastases or primary tumors themselves. To do this we will use the same techniques to analyse the metastatic data from the Hartwig Medical Foundation as we have previously used in the TRACERx (Tracking cancer evolution through therapy) prospective clinical trial (ID: NCT01888601) to contrast and compare the mutation and chromosomal alteration heterogeneity present in the data held by Hartwig Medical Foundation with data from primary tumors in our studies. By comparing and contrasting these patterns of heterogeneity across cancer types we will determine whether cancers that metastasise have similar or different patterns of intratumor heterogeneity to those that do not. These patterns may give us insight that helps identify which patients are likely to suffer from metastasis. Charles Swanton The Francis Crick Institute United Kingdom

 

The treatment of metastatic colorectal cancer (mCRC) is still based on traditional chemotherapies. mCRC is a very heterogeneous disease, and a tumor from a single patient might respond to one therapy, but not to the other. Unfortunately, personalized cancer treatment for chemotherapy is currently lacking, since we are not able to correctly predict chemotherapy response upfront. Consequently, patients do not always get the optimal (order of) treatment. By looking at the cellular composition of a tumor (i.e. all different cell types), we aim to optimize the chemotherapy choice and select the most effective therapy for every individual patient. Emile Voest NKI/AvL the Netherlands

 

Breast cancer is the most common cancer among women Worldwide. Inherited variants in breast cancer susceptibility genes or other regions in the genome that increase breast cancer risk play an important role in the development of breast cancer. These germline genetic variants facilitate the establishment of risk prediction models for women. Furthermore, understanding of the mechanisms by which these genetic variants cause the development of breast cancer is important to devise novel prevention and treatment strategies for women who carry these variants. Here we aim to discover novel germline genetic variants and the features associated with currently known variants.

John Martens, NKI/AvL,Marjanka Schmidt, Erasmus MC NKI/AvL / Erasmus MC the Netherlands

Numerous environmental triggers have been found to leave a footprint on the DNA of human cancers. By treating human cells with bacteria which frequently occur in colorectal cancer, we have identified the first mutational signature caused by a bacterium. Now we aim to determine how many mutations of cancer patients can be attributed to this signature. Furthermore, we want to search for bacterial genetic information in patient sequencing data and evaluate if more bacterial DNA correlates with this mutational signature. Our results may help identify subgroups of individuals at risk, paving the way for better screening and prevention measures. Hans Clevers Hubrecht Institute Utrecht the Netherlands

 

"Glioblastoma multiforme (GBM) is the most common and aggressive primary brain tumor in adults. Since 2005, standard treatment consists of surgery followed by radiation therapy with concurrent and adjuvant temozolomide chemotherapy, taking ± 8.5 months in total. Despite optimal treatment, GBM outcome is almost inevitably fatal, usually within 16 months after diagnosis. There is an urgent need to improve treatment of patients with GBM, both by selection of patients most likely to respond to standard therapy, as well as through the identification of novel targets for [combination] treatment. DR-093 is an extension of DR-060 and DR-076"

Filip de Vos University Medical Center Utrecht the Netherlands

While localized prostate cancer can be cured, metastastatic disease is associated with high morbidity and mortality. Prostate cancer typically metastasizes to lymph nodes and the bone, while metastases to organs like liver, lung or the peritoneum are less common. This project is aimed to describe what the differences are in the genomic (DNA) make up of prostate cancer metastases in the different sites. Knowledge hereof might spark further research into ‘site-preference’ and ‘site-specific alterations’ of the prostate cancer metastases genome, which ultimately might result in means to better treat or even prevent metastases.

Andre Bergman NKI/AvL the Netherlands

This project concerns a feasibility study, in which tumor biopsies collected in different hospitals in the CPCT-02 trial will be used to perform extensive protein analysis. The results of the DNA analysis and protein analysis will be compared to see how changes in the tumor DNA translate to the protein profile in the tumor, for better understanding of tumor biology. In the future, we hope to develop an integrated analysis of tumor DNA and protein profile, in order to better predict which patients will benefit from targeted therapy.

Mariette Labots Amsterdam UMC the Netherlands

  "Venous thromboembolism (VTE), a venous blood clot, frequently occurs in cancer patients, and substantially affects morbidity, quality of life, and mortality. Yet, it is incompletely understood why cancer patients are at increased risk of this disease. The risk of VTE varies greatly across cancer types, strongly suggesting a role for cancer-specific oncogenic mutations which indirectly cause thrombosis. We aim to investigate this possible association by using clinical data and genomic data from tumor biopsies acquired within the CPCT-02 study. Results could help understand the pathogenesis of cancer-associated venous thromboembolism, lead to new therapeutic targets, and identify high-risk patients for thromboprophylaxis."

Harry Büller Amsterdam UMC the Netherlands

The introduction of new therapeutic options, i.e. antiangiogenic drugs and immunotherapy, has significantly changed the treatment and perspectives of patients with advanced and metastatic renal cell cancer (RCC). Currently, the choice for first-line treatment is based on a risk calculation model, using only clinical parameters such as performance status. The aim of the current project is to describe the genomic landscape of RCC patients focusing on the immunogenic and angiogenic mutations/genes. More insight into immunogenic and angiogenic driver mutations together with the genomic landscape of RCC may contribute to improved and more individualized therapeutic decision making.

Astrid van der Veldt Erasmus MC the Netherlands

We aim to stratify cancer types by the degree of intratumour heterogeneity demonstrated both between metastases (when the cancer spreads) and primary disease (observed at the initial cancer diagnosis) as well as within the metastases or primary tumors themselves. To do this we will use the same techniques to analyse the metastatic data from the Hartwig Medical Foundation as we have previously used in the TRACERx (Tracking cancer evolution through therapy) prospective clinical trial (ID: NCT01888601) to contrast and compare the mutation and chromosomal alteration heterogeneity present in the data held by Hartwig Medical Foundation with data from primary tumors in our studies. By comparing and contrasting these patterns of heterogeneity across cancer types we will determine whether cancers that metastasise have similar or different patterns of intratumor heterogeneity to those that do not. These patterns may give us insight that helps identify which patients are likely to suffer from metastasis.

Charles Swanton The Francis Crick Institute United Kingdom

The treatment of metastatic colorectal cancer (mCRC) is still based on traditional chemotherapies. mCRC is a very heterogeneous disease, and a tumor from a single patient might respond to one therapy, but not to the other. Unfortunately, personalized cancer treatment for chemotherapy is currently lacking, since we are not able to correctly predict chemotherapy response upfront. Consequently, patients do not always get the optimal (order of) treatment. By looking at the cellular composition of a tumor (i.e. all different cell types), we aim to optimize the chemotherapy choice and select the most effective therapy for every individual patient.

Emile Voest NKI/AvL the Netherlands

Breast cancer is the most common cancer among women Worldwide. Inherited variants in breast cancer susceptibility genes or other regions in the genome that increase breast cancer risk play an important role in the development of breast cancer. These germline genetic variants facilitate the establishment of risk prediction models for women. Furthermore, understanding of the mechanisms by which these genetic variants cause the development of breast cancer is important to devise novel prevention and treatment strategies for women who carry these variants. Here we aim to discover novel germline genetic variants and the features associated with currently known variants.

John Martens, NKI/AvL,Marjanka Schmidt, Erasmus MC NKI/AvL / Erasmus MC the Netherlands

Numerous environmental triggers have been found to leave a footprint on the DNA of human cancers. By treating human cells with bacteria which frequently occur in colorectal cancer, we have identified the first mutational signature caused by a bacterium. Now we aim to determine how many mutations of cancer patients can be attributed to this signature. Furthermore, we want to search for bacterial genetic information in patient sequencing data and evaluate if more bacterial DNA correlates with this mutational signature. Our results may help identify subgroups of individuals at risk, paving the way for better screening and prevention measures.

Hans Clevers Hubrecht Institute Utrecht the Netherlands

Personalized medicine driven treatments in major diseases like non-small cell lung cancer (NSCLC) and melanoma offer important health benefits to specific genetically determined subgroups, but can be expensive and may induce severe side effects. Whole Genome Sequencing (WGS) simultaneously tests for all relevant genetic aberrations in tumor tissue from individual cancer patients thereby allowing immediate selection of optimal therapy. This approach is likely to improve patient survival, avoid adverse effects, and to assist in controlling health care costs by offering treatment to only those identified to benefit. The TANGO project aims to detect the most effective way of using WGS in clinical practice by comparing with standard of care diagnostics, in order to let all NSCLC and melanoma patients benefit from personalized medicine driven treatments. Edwin Cuppen University Medical Center Utrecht the Netherlands

 

"In this project we aim to use whole genome sequencing data to improve the way we treat patients with prostate cancer. Providing a better understanding of the subgroups with different clinical and genomic characteristics may improve the treatment outcomes. To this end we need to know which alterations are present in the tumor genome of prostate cancer samples. DR-071 is an extension of DR-011 "

Martijn Lolkema Erasmus MC the Netherlands

"Small intestinal neuroendocrine tumors (SI-NETs) represent a heterogenous group of rare tumors. At time of diagnosis approximately 50% of patients present with metastasized disease (Frilling 2014)1. Patients with metastasized disease are (usually) ineligible for curative surgery and have few treatment options with limited survival benefit. Unraveling the molecular events underlying SI-NET tumorigenesis could facilitate the identification of therapeutic targets and improve prognosis. In order to expand the limited existing knowledge on molecular factors in SI-NETs and move towards personalized treatment, we would like to study somatic and germline mutations of SI-NETs by using DNA next generation sequencing. 1 Frilling A, Modlin I, Kidd M, Russel C, Breitenstein S, Salem R et al (2014). Recommendations for management of patients with neuroendocrine liver metastases. The Lancet Oncology15(1), pp. e8-e21 "

Margot Tesselaar NKI/AvL the Netherlands

Cancer is a disease of the genome. Random factors, environmental exposures and also certain biological mechanisms have the potential to cause mutations in DNA, while the sophisticated DNA repair mechanisms available to human cells prevent such changes in genes. We will use the whole-genome sequences from metastatic tumors to examine how the balance between DNA damage and DNA repair shapes the mutation patterns observed in cancer. This is relevant to patients because we will further investigate how such mutational patterns are associated with risk of cancer progression and with response to various drugs, with implications to prevention and treatment.

Fran Supek IRB Barcelona Spain

Breast cancer is the most common malignancy among women wordwide. To reveal the genomic landscape of metastatic breast cancer, we recently performed an extensive genomic analysis which provided important novel insights into the biology of metastatic breast cancer and identified clinically relevant subgroups of patients who might benefit in the future from specific therapies. With this project we aim to correlate the findings on the DNA level with data generated from RNA. This will lead to increased knowledge of metastatic breast cancer and probably provides new predictive and prognostic information by integrating these two data sources. Finally, our goal is to improve personalized cancer treatment for patients with metastatic breast cancer.

John Martens Erasmus MC the Netherlands

"Many viruses that commonly infect humans have been found to fuse cells. Evidence suggests that the products of cell fusion can retain some characteristics of the original cells, but that they can also develop malignant characteristics, accelerate cancer progression, and facilitate spread to secondary sites (metastasis). Using whole genome sequencing data we aim to identify tumours in which virus-induced cell-cell fusion may have played a role, and potentially contributed to metastasis. This will lead to novel insights into the origins of metastasis, and may be used to identify patients who are at risk of virus-induced cell-cell fusion mediated metastasis."

Maria Jakobsdottir (before David Wedge) University of Oxford United Kingdom

Cancer affects millions of people worldwide. Current DNA sequencing technologies allow routine screening of patients for cancer-associated mutations that can be targeted in personalized cancer treatments. Accurate computational methods for the identification of mutations, in particular structural variants (SVs), are lacking, together with methods that can prioritize mutations for further intervention. To solve this problem, we are developing a new set of machine learning approaches that can learn to detect SVs directly from ‘raw’ sequencing data and prioritize them for clinical intervention based on multiple sources of genomic information. Our work will open up a new avenue of machine-learned SV detection and prioritization, and accelerate the introduction of SV-based screenings for new treatments in clinical settings.

Jeroen de Ridder UMC Utrecht the Netherlands

We will analyse the Hartwig Medical Foundation metastatic cohort with advanced network analysis techniques to identify genes and key pathways that drive the onset of metastatic cancer and in particular metastatic prostate cancer. This fundamental insight in the genetic origin of metastatic prostate cancer will help identifying biomarkers that can optimally stratify patients for metastasis directed therapies.

Kathleen Marchal Ghent University (UGhent) Belgium

Despite the fact that completion of human genome project that was announced over 15 years ago, our knowledge of human genomes is still incomplete. The most challenging pieces missing from our genome “blueprint” are large insertions that are not present in genome of every individual or located in genomic regions that are difficult to assemble. The aim of this project is to identify such segments, finemap them to chromosomes and characterize their molecular function. The completeness of genome reference will greatly simplify, speedup and improve analysis of genomic information from wide range of samples: from healthy subjects to cancer patients.

Victor Guryev UMC Groningen the Netherlands

Malignant pleural mesothelioma (MPM) is a rare and aggressive malignancy that arises from the mesothelial surfaces of the pleural cavity. Currently no curative treatment options are available. Recently, promising new treatment options have emerged, including immune checkpoint inhibitors and cell based vaccines. In order to give the best therapy to each patient, a better understanding of the tumor cells and their interaction with their microenvionment is necessary. We would like to investigate how abberations in the genetic material of the tumor can lead to the development, maintenance and progression of the tumor and how different therapies can influence these processes.

Joachim Aerts Erasmus MC the Netherlands

We aim to apply computational tools to Hartwig Medical Foundation cancer genomes to establish the order in which harmful mutations were acquired, and approximately when they were acquired during a patient’s lifetime. Furthermore, we can infer what caused the mutations (for example, smoking, UV light, etc.), and estimate when the cell was exposed to these damaging processes. Across many cancer samples, this approach lets us build a picture of how different cancers evolve over time. By comparing these results to earlier analyses we performed on primary tumours (observed at initial cancer diagnosis), we seek to understand the development of cancer as it spreads from the initial site to another site (known as metastasis). This will show whether tumours that metastasise evolve in the same way as non-metastatic cancers, and may give insight that helps to predict the metastatic step in cancer evolution.

Peter van Loo The Francis Crick Institute United Kingdom

"Glioblastoma multiforme (GBM) is the most common and aggressive primary brain tumor in adults. Since 2005, standard treatment consists of surgery followed by radiation therapy with concurrent and adjuvant temozolomide chemotherapy, taking ± 8.5 months in total. Despite optimal treatment, GBM outcome is almost inevitably fatal, usually within 16 months after diagnosis. There is an urgent need to improve treatment of patients with GBM, both by selection of patients most likely to respond to standard therapy, as well as through the identification of novel targets for [combination] treatment. Whole-genome sequencing is a personalized medicine approach that can be applied to characterize the complete DNA sequence of individual patient’s tumors. Thus far, tumor specimen of 80 patients with GBM have been collected in the Dutch Center for Personalized Cancer Treatment [CPCT] biopsy study, together with information on the response to temozolomide-based chemoradiation. In this research proposal, a multidisciplinary team of Dutch neurologists, oncologists and neuro-surgeons propose to investigate the relation between the genetic properties of these tumors and response to temozolomide-based treatment."

Filip De Vos UMC Utrecht the Netherlands

In order to enable the next phase of genomic discovery and build a mutational atlas for future Precision Oncology purposes in pancreatic cancer, we wish to create a harmonised genomic dataset spanning all stages, from pre-cancerous lesions, primary resectable tumours to metastatic disease. This will provide an accurate map of cancer driving mutations and the relative timing of their accrual during disease progression, hence allowing early detection of disease, therapy optimisation and to design new drugs.

Sean Grimmond University of Melbourne Australia

The treatment of inoperable metastatic colorectal cancer consists of systemic treatment through chemotherapy with or without additional targeted treatment. Until now, no good model exists to provide personalized treatment and to predict which treatment could benefit individual patients the most. With this project we aim to reveal the genetic landscape of metastatic colorectal cancer. This will not only lead to an increased knowledge of the metastatic disease, but it will provide new information on the possible predictive potential of this information. With this combined knowledge we want to improve personalized medicine options for patients with metastasized colorectal cancer.

Saskia M. Wilting Erasmus MC the Netherlands

We aim to identify and characterize the molecular and cellular factors that distinguish cancer cells that metastasize from cancer cells that do not migrate. Discovering and better understanding these factors will enable the development of treatment options that specifically target the metastatic process.

Roel Verhaak The Jackson Laboratory USA

The use of molecular pathology (for example DNA-testing) has increasingly allowed the selection of the right treatment for the right patient (personalized medicine). Nonetheless, the majority of anticancer treatments remains not patient-specific. As a result, patients still receive ineffective treatments. A possible explanation for this might be found in the tumoral heterogeneity (ie. the presence of different types of cells). Recently, an computional algorithm (Tumor Cell Deconvolution (TCD)) has been developed in breast cancer patients to assess the cellular heterogeneity of a tumor, for every patient specific. With the TCD, we can identify specific cell types, and can explore whether these specific cell types determine the response to therapy.

Emile Voest NKI/AvL the Netherlands

The changes to the DNA (mutations) found in a tumor are expected to be largely unique to each tumor as there are 3 billion places in the DNA that can be changed. However, in a cancer study with 2,583 participants we observed over a million non-unique mutations. We used characteristics of these mutations and all mutations detected, to classify the tumors into 16 groups that have clinical relevance. We want to refine these classes to gain further insights into the biological processes inducing mutations (‘mutagenesis’), inform on treatment, help in diagnosis, and predict the origin of a metastatic tumor.

Ivo Gut Centro Nacional de Análisis Genómico Spain

We use whole cancer genome sequences to understand the order of mutations in cancer development so that we can infer which mutations arise early versus late during the process. The ability to determine the order and importance of mutations is critical for early detection and treatment of disease. We can also understand the environmental exposires that cause cancer based on the pattern of mutations.

Paul Spellman Oregon Health & Science University USA

In this project we aim at combining mathematical modelling of tumor evolution with cancer genomic data to measure fundamental evolutionary characteristics of metastatic tumors. These measurements will allow better understanding of how tumors spread to distant parts of the body. Importantly, we anticipate that the patient-specific ‘evolutionary rules’ we will measure from the data may be informative in terms of stratifying patients between good versus poor prognosis. Moreover, some of these measurements may also be predictive of the effectiveness of certain treatments.

Andrea Sottoriva The Institute of Cancer Research United Kingdom

Cancer genomes typically harbor a substantial number of somatic mutations. However, only few of these mutations actually contribute to tumor development. To identify these so-called driver mutations, large-scale aggregated sequencing datasets are needed. Over the past decade, the search for driver mutations has been mostly focussed on coding regions. Substantially less is known whether and to which degree mutations in non-coding regions are functionally involved in tumor formation. With the drop in sequencing costs over the past years, it became possible to sequence the entire genomes of large cohorts of cancer patients (e.g. the dataset provided by Hartwig Medical Foundation). However, the avaialalbity of these datasets has not been met with the of algorithms that are particularly designed to the underlying biology of these data yet. The goal of this project is to develop a statistical framework which systematically identifies driver mutations in non-coding regions in aggregated whole-genome sequencing data. For this purpose, we will particularly consider the biological differences between coding an non-coding regions, in order to specifically taylor our algorithm to non-coding regions. Applying this refined approach to ~2500 Whole Genome Sequencing data, we will systematically characterize which mutations in non-coding regions are relevant to oncogenesis.

Eliezer van Allen Dana-Farber Cancer Institute USA

"Cancer develops by accumulation of mutations in the DNA of individual cells. In many patients, DNA repair mechanisms are often perturbed early in cancer development, leading to an increased mutation rate and faster development of the cancer. However, repair deficiencies can also make the cancer cells more susceptible to certain types of treatments. Better understanding and detection of DNA repair deficiencies may thus help improve treatment of cancer patients. Using the cancer genomics data from Hartwig Medical Foundation, we aim to characterize the pattern of DNA deficiencies across different cancer types. We further aim to improve the detection of DNA repair deficiencies through their effect on mutation rates and patterns across the genomes of cancer cells. Finally, we aim further clarify the relationship between DNA repair deficiencies and cancer treatment efficacy."

Jakob Skou Pedersen Aarhus University Denmark

Pancreatic cancer is the fourth leading cause of cancer-realated deaths in the world. Current therapeutic options including chemo-, radio- and combinational therapy are not giving satisfactory results. Therefore, there is an emerging need to develop a novel, effective targeted therapy. Immunotherapy engaging the patient’s own immune system is among the newest approaches to target and destroy tumor cells. The immune system can be trained to recognize cancer specific antigens that are encoded by tumor-specific mutated genes (neoantigens). By analysing the seqeuncing data provided by Hartwig Medical Foundation, a list of the most common neoantigens in pancreatic cancer patients will be generated and utilized to synthesize specific peptides in the form of an off-the-shelf vaccine ready to be used for a broad audience. This project is part of the Oncolytic Virus Immune Therapy (OVIT) consortium aiming to employ viro-immunotherapy to treat patients with aggressive cancer types.

Casper van Eijck Erasmus MC the Netherlands

Tumor-specific defect in the DNA repair offer great potential for DNA damage based chemo/radio therapy. Present data implicate that this therapeutic intervention strategy has not been explored sufficiently. To provide optimal therapy with less side effects, there is an unmet need tot take further advantage of cancer-specific defect within the molecular network that controls DNA repair. This project will reveal the novel therapeutic opportunities for patients with tumor-specific DNA repair defects. The mission, optimize tumor intervention strategies to achieve higher cue rates with less side effects by developing optimised pesonalised medicine.

Heinz Jacobs NKI/AvL the Netherlands

Anticancer drugs can (in)directly interfere with DNA and thereby potentially generate a characteristic pattern of somatic mutations (mutational signature). Identifying treatment-specific mutational signatures in a cancer genome will not only improve the mechanistic understanding of the mutational processes of the drug itself, but will also delineate the proportion of the mutational side effects. This research seeks to characterize the mutational patterns that are generated by anticancer drugs and tries to advance our understanding of the mutational signatures with unknown etiology.

Edwin Cuppen UMC Utrecht the Netherlands

Resistance to therapy is one of the main challenges in treatment of metastatic breast cancer. We know that treatment resistance is often driven by changes in the genetic material of the tumor. Tumor DNA is, to some extend, detectable in the blood of cancer patients as circulating tumor DNA (ctDNA). Analysis of ctDNA might enable monitoring of response to anti cancer therapy over time and at low impact. The aim of this study is therefore to identify aberrations in ctDNA related to treatment outcome and to see whether those aberrations change during treatment. Furthermore, we aim to explore to what extend ctDNA represents the genomic landscape of metastastic breast cancer. Therefore, our findings in ctDNA and in metastatic tumor tissue will be compared in a subset of patients.

John Martens Erasmus MC the Netherlands

Immunotherapies stimulate the patients own immune system to kill tumor cells. These relatively new therapies have shown impressive treatment results in many different tumor types. However, these impressive responses are restricted to only a subset of patients; many patients do not have benefit from these treatments, as the tumors keep growing and the patients suffer from treatment side-effects. We set up an ongoing project in which we try to understand why particular patients are not responding based on their DNA. If we understand this, we could (1) prevent treatment of patients with tumors resistant to immunotherapy and (2) develop strategies to overcome such resistance.

Emile Voest NKI/AvL the Netherlands

Some cancers are the result of viral infections. A novel type of cancer therapy, called immunotherapy, has been shown to work exceptionally well against such virus-driven cancers. These therapies stimulate the patients own immune system to kill the virus-infected tumor cells. Many viruses have, like humans, a genome of DNA. In this project we are searching for viral DNA within the tumors of patients, to identify the patients with virus-driven tumors. These patients are potential candidates for immutherapy treatment.

Emile Voest NKI/AvL the Netherlands

Cancer is driven by accumulation of single DNA mutations, but also by catastrophic events involving multiple simultaneous changes. One such catastrophic event is called kataegis (thunder), which concerns multiple DNA mutations occurring in a small region of the DNA. Using Hartwig Medical Foundation whole-genome repository, we can improve our sophisticated computational methods to better detect such somatic events in cancer patients. By identifying and characterizing these events in cancer we gain insight in the molecular mechanisms that give rise to kataegis. These molecular insights will improve our understanding of the biology of (metastatic) cancer.

Harmen van der Werken Erasmus MC the Netherlands

Recent whole genome sequencing efforts have identified a broad spectrum of gains and losses of DNA in cancer genomes. This has led to the identification of novel mechanisms for genome rearrangement like chromothripsis, where chromosomes are shattered and stitched back together in a different order. However, little is known about the prevalence of these types of events throughout different tumor types and their relevance for patient outcome. Furthermore, the data provided by Hartwig Medical Foundation provides unprecedented detail and resolution, enabling us to detect small gains/losses of DNA that would have previously remained unnoticed.

Bauke Ylstra Amsterdam UMC the Netherlands

Endometrium cancer is one of the most common gynecological cancer and accounts for approx. 6% of all cancers in women. The vast majority of enometrium cancers are dependent on a hormone receptor, Estrogen Receptor a (ERa). We aim to identify how changes in the DNA sequence of the tumor would affect the function of ERa, and how this would ultimately impact gene activity and tumor cell growth.

Wilbert Zwart NKI/AvL the Netherlands

"Neuroendocrine tumors are rare cancers most often arising in the lungs, pancreas or gastro-intestinal tract. Its behavior can vary greatly among patients, from very slowly growing tumors that can be kept under control with a relatively low-impact treatment, to very aggressive tumors that quickly progress despite intensive (chemotherapeutic) treatment. Currently, we cannot optimally select the right treatment for the right patient, possibly because the characteristics we are using to select this treatment are not based on the genetic make-up of the tumor. A minimally-invasive method to determine which patient will benefit from a certain treatment, is by taking so-called liquid biopsies. In liquid biopsies we look at tumor-specific genetic variations in the blood of our patients. Thus far, the implementation of liquid biopsies in neuroendocrine tumors have been hampered by the lack of a known genetic variation which occurs in the majority of our patients and is thus a suitable liquid biopsy marker. In this project, we want to explore the genetic charactertistics of metastases of neuroendocrine tumors and use these data to develop a liquid biopsy test by which we can better determine who will benefit from a certain treatment, and who should be spared unnecessary toxicity."

Bianca Mostert Erasmus MC the Netherlands

Histological, immunohistochemical and gene expression markers are routinely used to stratify tumours into clinically distinct subtypes, some of which have targeted therapies that improve prognosis. Recent work has shown that unsupervised clustering using genomic profiling can reveal novel subtypes with distinct clinical outcomes. We will explore if stratifying patients directly from whole genome mutation profiles can identify novel subtypes and driver pathways that could lead to improved personalised therapy opportunities.

David Wedge University of Oxford United Kingdom

To date, translational cancer genomics has focused on somatic genetics, driven by the desire to identify mutated targets that indicate specific therapies. Therapeutic information is also present within the germline genome. However, this germline information is currently ‘subtracted’ in clinical analytic pipelines, discarding valuable information relevant to therapy. Within this project we will identify and study the value of this inherited germline information.

David Thomas Garvan Institute Medical Research Australia

"Patients with head and neck cancer too advanced for surgical resection have a very poor prognosis, despite the currently standard, very intensive treatment which consists of a combination of chemotherapy and radiotherapy. Until now we have not been successful in identifying patients with a very poor prognosis. To spare these patients unnecessary treatment and its side-effects, better upfront patient selection is urgently warranted. A minimally-invasive method to determine which patient will benefit from a certain treatment, is by taking so-called liquid biopsies. In liquid biopsies, we look at tumor-specific genetic variations in the blood of our patients. We are planning to do a study in which we correlate the level of a specific mutation which is often found in head and neck cancer, TP53, with the chance of disease recurrence. We also want to look at other genetic variants in liquid biopsies that are associated with poor prognosis. So far, the known genetic variants have been identified in the original primary tumors. However, genes that are mutated specifically in metastases, which we know can differ from the original primary tumor, are likely largely responsible for its aggressive behavior. Therefore, we will identify recurrent genetic variations in the metastases of patients with head and neck cancer. These variations will then be included in an elaborate liquid biopsy test, which we will correlate with prognosis in our observational study."

Bianca Mostert Erasmus MC the Netherlands

This project aims to unravel molecular mechanisms that drive the high metastatic potential of urothelial cancer.a In addition predictive markers that correspond with response to chemo- or targeted therapy are urgently needed both in the neoadjuvant and metastatic setting to improve patient stratification.

Joost Boormans Erasmus MC the Netherlands

Irinotecan is a widely used cytostatic treatment for a broad range of tumor types. However, response is highly variable making the identifiation of biomarkers that can predict treatment outcome highly desirable. In this project Whole Genome Sequencing characteristics will be explored for all patients that received irinotec treatment and compared with clinical response to the drug.

Martijn Lolkema Erasmus MC the Netherlands

This project aims to characterize the genetic landscape of sarcoma tumors and relate this to treatment outcome data. Improving patient care for this relatively rare cancer indication requires the identification of biomarkers that enable stratification of patients towards the most effective therapies. Understanding underlying mechansisms of carcinogenesis, notably driver genes and pathways, is imperative to reach these goals.

Eric Wiemer Erasmus MC the Netherlands

The effect of anti-cancer drugs depends on the amount of drug that is present in the tumor cells. If these tumor cells have decreased uptake or increased excretion of a drug, the tumor has less exposure to the drug and therefore the drug might be less effective. Uptake and excretion of drugs is regulated by drug transporters, which we will investigate at DNA level. If (subtypes of) tumors have reduced exposure via this mechanism, it might be beneficial to give higher doses of that drug or to switch to an alternative drug.

Stefan Sleijfer Erasmus MC the Netherlands

This project aims to create a systematic overview of the mutation landscape of metastatic breast cancer patients and identify similarities and differences with comparable primary breast cancer data set obtained from elsewhere. This resource may identify patient subcategories that may benefit from different treatment strategies.

John Martens Erasmus MC the Netherlands

"Most studies have focused on mutations in the few percent of the genome that encode proteins. In this study, large amounts of tumor genomes will be used to identify loci in the non-coding regulatory part of the genome that contribute to tumor development when mutated. DR-024 is an extension of DR-017"

Nuria López-Bigas IRB Barcelona Spain

The response to checkpoint inhibitors (immunotherapy) is highly variable between tumor types. The researchers will search for novel biomarkers based on Whole Genome Sequencing and RNA sequencing data accross and within all tumor types to better stratify patients for this treatment type and to understand the underlying biology driving treatment sensitivity or response.

Emile Voest NKI/AvL the Netherlands

The DNA in normal – and tumor cells are continuously exposed to internal and external challenges. The objective of this project is to computationally derive specific footprints (signatures) that are left in the tumor genomes. Previous work has shown that environmental challenges like UV radiation and smoking leave very characteristic patterns, as well as intrinsic process like DNA repair deficiencies. Such knowledge is of importance for cancer prevention and treatment strategies.

Serena Nik-Zainal University of Cambridge United Kingdom

The goal is to investigate the presence of actionable biomarkers in whole genome sequencing (WGS) data of glioblastoma cases and compare these with results obtained with standard diagnostic targeted gene panels. This research should provide insight in what the best diagnostic strategy would be for this category of patients.

Paul van Diest UMC Utrecht the Netherlands

The goal of this study is to systematically compare the mutation landscape and underlying mutagenic mechanisms that are active in primary and advanced metastatic tumors. This study is expected to provide insight in the molecular mechanisms that drive different stages of tumor development and in the mutagenic effects of specific treatments.

Edwin Cuppen UMC Utrecht the Netherlands

Deficiency in DNA repair pathways is a key characteristic in many tumors. The goal of this project is to develop tools based on genome-wide mutation characterists that better detect defects in the homologous recombination DNA repair pathway. This is relevant as such patients may better respond to platinum drug-based therapies or PARPinhibitor targeted treatment.

Edwin Cuppen UMC Utrecht the Netherlands

Most studies have focused on mutations in the few percent of the genome that encode proteins. In this study, large amounts of tumor genomes will be used to identify loci in the non-coding regulatory part of the genome that contribute to tumor development when mutated.

Nuria López-Bigas IRB Barcelona Spain

The researchers will study the involvement of structural variation in the tumor genomes of colorectal cancer patients and investigate their role in tumor development and treatment response.

Remond Fijneman NKI/AvL the Netherlands

"In this study genetic factors will be explored that may explain response to standard chemotherapy (FOLFOX/CAPOX/Irinotecan/capecitabine) in coloroctal cancer patients. These results will be compared and integrated with results obtained with in vitro treated tumor organoids."

Louisa Hoes NKI/AvL the Netherlands

This project is focused on biomarker discovery for specific targeted treatments in breast cancer patients. Specific focus will be on patients treated with aromatase inhibitors.

John Martens Erasmus MC the Netherlands

The research question is focused on the improvement and development of novel treatment approaches in prostate cancer patients that have become resistant to androgen deprivation therapy. The goal is to identify mechansisms that explain resistance and which could be developed in a circulating tumor DNA test for easy monitoring of treatment response.

Michiel van der Heijden NKI/AvL the Netherlands

The goal of this project is to correlate systematic DNA and protein measurement data from the same patients. This pilot may provide indications if multi-level measurements are feasible and provide additive information and form jointly stronger biomarkers. Furthermore, this work may provide insight in the molecular mechanisms that are perturbed by cancer mutations.

Henk Verheul Amsterdam UMC locatie Vumc the Netherlands

This project aims to create a systematic overview of the mutation landscape of metastatic prostate cancer patients with a special focus on structural variation. This resource may identify patient subcategories that may benefit from different treatment strategies.

Martijn Lolkema Erasmus MC the Netherlands

The goal of this project is to identify inherited mutations in genes that result in higher chances to develop cancer (cancer pre-disposition genes).

Edwin Cuppen UMC Utrecht the Netherlands

The response to check-point inhibitors (immunotherapy) is highly variable between tumor types. The researchers will search for novel biomarkers based on Whole Genome Sequencing and RNA sequencing data accross and within all tumor types to better stratify patients for this treatment type and to understand the underlying biology driving treatment sensitivity or response.

Emile Voest NKI/AvL the Netherlands

"This project aims to identify biomarkers that are predictive for response to immunotherapy (check-point inhibitors) in lung cancer patients. As only ~20 to 30% of patients respond to this very expensive treatment, ability to identify these patients before start of treatment would have an enormous impact on quality of life and care budgets, which will also be studied as part of this project. "

Joachim Aerts Erasmus MC the Netherlands

In this project, pairs of tumor biopsies are requested to study evolution of tumors over time. This project may provide insights in potential mutagenic effects of specific tumor treatments and resulting from selection pressure and shed light on treatment resistance mechanisms.

Emile Voest NKI/AvL the Netherlands

The researchers will explore the prevalance of mutations in the KRAS gene in GIST patients and study the response to specific treatments. This work may result in better stratifiation of patients towards targeted treatments.

Neeltje Steeghs NKI/AvL the Netherlands

The goal of this request is to enable Hartwig Medical Foundation Australia team to optimize bio informatic tools and pipelines for the identification of all types of somatic variants to obtain the most complete genetic picture of every individual tumor genome. Data of Hartwig Medical Foundation allows the researchers to test and validate their tools on real-world data.

Hans van Snellenberg Hartwig Medical Foundation the Netherlands

The goal of this project is to identify genetic factors that are involved in chemotherapy resistence in ovarium cancer and to identify novel targets for treatment. To this end, the spectrum of genetic aberrations in chemoresistant ovarium tumors will be compared with chemo sensitive tumors.

Edwin Cuppen (previous Wigard Kloosterman) UMC Utrecht the Netherlands

The objective of this request is to connect genetic information obtained in the CPCT-02 study and in vitro tumor organoid data from the TUMEROID and SENSOR studies form the same patient. This connection allows for the identification and exploration of the potential genetic basis for (in vitro) treatment sensitvity for a broad range of drugs.

Louisa Hoes (previous Fleur Weeber) NKI/AvL the Netherlands

In this project, data from CRC patients treated with anti-EGFR therapy will be used to identify biomarkers that predict treatement response. A specific hypothesis on the involvement of mutations in the MAPK will be tested which provides novel treatment strategies involving existing MEK inhibitors.

René Bernards NKI/AvL the Netherlands

The objective of this project is to determine if the microenvironment of metastases influences mutation patterns. The investigators will test if colorectal cancer tumor lesions that develop as metastases in different tissues show differences in the type of mutations that are accumulated.

Edwin Cuppen (previous Ruben van Boxtel) UMC Utrecht the Netherlands