Project Funding Details

The worldwide costs of cancer care are unsustainable; treatment costs are rising exponentially because of the multitude of promising but expensive treatment strategies entering the market (e.g. Ipilimumab). Therefore, research into biomarkers that can distinguish which patient would benefit from what treatment is crucial to be able to provide the best cancer care to patients now and in the future. To enable the identification of novel biomarkers and discovery of new therapies cutting edge cancer research is required, whereby 1) multiple large (genomics) datasets are used and 2) researchers of different institutions collaborate. Both are currently hampered, because firstly genomics data is typically generated in a large number of different centers with various data collection and analysis protocols. Moreover, to protect patient rights and privacy, data is typically stored as coded or pseudoanonimised records, substantially decreasing the accessibility and interoperability of datasets. Together, this substantially impacts on data quality and integration, and limits the effective collaborative use of (genomics) data for cancer research. Additionally, researchers often have to invest much time and effort to have different types (clinical, genetics, survival, histology etc) of data available. The Hartwig Medical Foundation (HMF), in collaboration with the Center for Personalized Cancer Treatment, aims to overcome the limitations for usage of genomics data by developing a national uniform FAIR data warehouse of whole genome sequencing (WGS) data and clinical/treatment data of cancer patients in the Netherlands. The data warehouse is available – under strict conditions to warrant patient privacy – for scientific research. HMF creates linkages between patient data and genetic information about the cancer. This forms a unique source for innovative scientific and health-economic research, which was not available to researchers before. The HMF data is highly complementary to existing national data resources such as those collected by the Netherlands Cancer Registry NCR (patient’s journey comprising patient- and tumor characteristics, and treatment and outcome data), PALGA (molecular and pathology data), BBMRI-NL (biobanking & cohort data, images & samples) and HUB (cancer organoids and drug sensitivity). The possibility to combine these datasets with HMF’s WGS and clinical data will enable breakthrough discoveries such as those where gene expression profiles coupled to patient survival can classify cancer types to unravel new cancer subtypes. GENONCO aims to create these linkages and provides cancer researchers (~ 3,000 in the Netherlands) with access to all available data. To do so, the first goal of GENONCO is to implement universal keys for database entries through trusted third party (TTP) pseudoanonimization. This increases the interoperability of the different datasets and facilitates research. This objective will build on important work in this area by BBMRI/TraIT. We will demonstrate utility by working out several use cases with key research partners. The second goal of GENONCO is related to the extreme size of WGS data. In February 2017 this was ~0.5 Petabytes of raw data for ~1,000 patients, growing with about 150 patients per month. Both practical and privacy reasons dictate solutions that prevent the researchers having to download these data. In GENONCO we will explore two (available) solutions; 1) a flexible computational infrastructure around the current data storage in a private cloud and 2) the usage of commercial cloud platforms (i.e. Google Cloud Platform and Microsoft Azure). KWF’s support for GENONCO stops after two years. Therefore, self-sustainability is of key importance for GENONCO. We have made a calculation of costs that are expected after that (~110k/year) and will consider different business models to cover those costs that are not covered by the users. Importantly, additional budget for data management and stewardship will become available in Dutch (ZonMW and NWO) and H2020 grants. Researchers (GENONCO’s end-users) can apply this money to fund their usage of GENONCO. GENONCO will align with current infrastructure efforts in the Netherlands to prevent overlap. Finally, we will organize roadshows and (online) workshops and webinars for researchers to demonstrate different research and data access possibilities using HMF data (and in connection with the other 'linked' resources). By creating linkages between relevant data resources and making data more accessible to cancers researchers, GENONCO aims to support; • Breakthrough discoveries: new biomarkers and (targeted) therapies; • More collaboration between cancer researchers; • Higher value (more data) for research money. This will expedite the development of improved diagnostics and novel more effective treatment strategies.