A national research and innovation platform for personalized cancer medicine

Our main objective was to identify all gene mutations in nine different cancer types, to identify those involved in cancer development and impacting treatment responses or providing potential for novel therapeutics. This multidisciplinary project included scientists from the field of oncology, cancer pathology, genomics, bioinformatics and biostatistics, as well as molecular, cell and tumor biology, ethics, health economy and social dialogue. All genes in more than 2000 samples from cancer patients have been analyzed by so-called next generation sequencing (NGS) to identify tumor-specific mutations. The structure of all 22,000 genes is determined in a blood sample and at least one tumor sample from each patient, resulting in about 55 million building blocks being read 100 times over in the blood samples and approximately 300 times in tumours. By now the sequencing analyses have generated approximately 500 terabytes of data. The bioinformatic analysis of these data is extremely complex, both due to the large amounts - and particularly because of - the multiple and complex ways the genetic code could be damaged in the different tumor cells. These injuries can result in removal of parts, or extra copies of genetic code, or DNA parts moved to other sites in the genome. An additional challenge is that the data are sensitive, and so we have established an special protected IT system, separated completely from the internet and where use is strictly controlled and connections are encrypted. In the final stage, when our genomic facility replaced their sequencing machine with the newest version, we encountered unexpected technical challenges due to changed technology details. Novel sequencing errors were discovered, and we had to halt the genomic analysis while numerous time-consuming tests and analyses had to be performed to understand and correct the problem. The consequent delays made it impossible to fulfil our contract within 2017, and the Research Council granted us a six-month project extension, by which the NGS is now completed. The following nine cancer types are studied: breast, ovarian, colon, prostate, malignant melanoma, sarcoma and the haematological cancers leukaemia, lymphoma and myeloma. From each cancer type samples from around 100 patients are analysed, and even though parts of the detailed analyses are still on-going, our high data quality, both with regard to NGS and clinical information, has given and should lead to more high-impact publications. Several of the mechanisms we observe in the patient samples have been examined further in the laboratory for possible subsequent use in new cancer treatments, and the promising ones also in animal models. Based on these experiments several early-phase clinical trials have been proposed. The PhD student on the project has completed his degree. By collaborating with the International Cancer Genomics Consortium (icgc.org) we have secured that our bioinformatic analysis has benchmarked international quality. This is a critical issue due to the complexity of cancer NGS data, which can give dramatically different results depending on analytical details. We have developed an internet-based reporting tool, generating structured reports that present the complex data ranked by clinical relevance. These reports are intended for hospital staff to facilitate the use of genome information for treatment decisions. Despite the uniqueness of the genetic variants within each individual, the frequency of variants within a patient group can be anonymized and thereafter be made available for a wider community. Our national variant database, currently containing all germ-line variants in our patient cohorts (1000genomes.no), is being developed further as part of the national initiative for personalized medicine of the Ministry of Health. A main future challenge is how to maintain our valuable database after the end of the project, and to make it available for other researchers in a secure way. We have an on-going dialogue with the Research Council regarding these concerns, and we have stressed the necessity of a common national infrastructure. In our ELSA-package we focus both on principles for utilization of genomic data in treatment of patients and in research, and on regulation of such use. Additionally, we have examined opportunities and challenges related to the personalization of medical treatment. The ELSA PhD-student has completed her degree, and two PhD-students funded by other grants are investigating the legal aspects of these challenges and aspects of health economics. We have further contributed to societal dialogue through open meetings on genomics-related topics, both for the lay public and health workers, presentation in media, and through local and national newspapers. Our network is continuing, see more at cancergenomics.no.

The Norwegian Cancer Genomics Consortium consists of experts in clinical oncology, cancer pathology, epidemiology and translational biology and genomics, with the objective to investigate the potential for genome-based strategies for personalized cancer t herapy. We propose to extend and open up our research platform to Norwegian and international industry collaborators, as the best way to develop cancer therapies and companion diagnostic tools in a public-private partnership. Ethical and societal aspects are addressed through research, public dialogue and participation of patient groups, as well as knowledge-based cost/benefit analyses in collaboration with similar projects internationally. The research projects will determine the detailed complete seque nces of many or all genes in specific sets of several thousand matched tumour and normal samples treated in a systematic way and with well-defined disease outcome or tumour characteristics. Tumour mutations that may guide treatment choices or predict dise ase characteristics will be searched for. A particular interest is the search for disease-associated targets in minor, or 'orphan', cancers that can be targeted by existing drugs approved for other cancer types, but also a general search in cancers for wh ich there are few therapeutic options. Also mutations providing therapeutic resistance will be a focus, as these may be exploited in improved strategies. At later stages, when sufficient number of patients have been analysed, also normal gene variation th at may predispose to cancer development or adverse or late side effects may be investigated. The project provides infrastructure and analytic capabilities to an extended network of academic, industry and health researchers. All Norwegian Health Regions an d medical faculties are involved, as well as the Norwegian Knowledge Centre for the Health Services, the Norwegian Cancer Registry, the Biotechnology Advisory Board and the Cancer Society, and Oslo Cancer Cluster.