Scientists at deCODE Genetics and academic collaborators from Iceland, the USA, The Netherlands and Spain today report the discovery of variants in the human genome that associate with levels of thyroid stimulating hormone and risk of thyroid cancer. The paper ‘Discovery of common variants associated with low TSH levels and thyroid cancer risk‘ is published today in the online edition of Nature Genetics.

Using data obtained by applying both Illumina whole-genome sequencing technology and Illumina SNP chip technology, deCODE’s scientists performed a genome wide association study on levels of thyroid stimulating hormone (TSH) in 27,758 Icelanders. 22 SNPs with genomewide significance were discovered, of which one, rs965513 had previously been shown to associate with thyroid cancer. The remaining 21 SNPs were genotyped in 561 Icelandic thyroid cancer cases and 40,013 controls. Variants suggestively associated with thyroid cancer were then genotyped in an additional 595 non-Icelandic cases and 2,603 controls.

After combining the results, three separate variants on chromosomes 2q35, 8p12 and 14q13.3 were shown to associate with risk of thyroid cancer, conferring an added risk of 30 – 100%, compared to the general population. These variants were also found to associate with low levels of TSH, a key regulator in the biology and endocrinology of the thyroid gland.

“This study underscores the important role that the genetics of diversity in normal physiologic function can play in understanding the risk of disease. To date, the at-risk alleles of all the variants that confer risk of thyroid cancer associate with decreased serum levels of TSH, suggesting that the primary disorder in non-medullary thyroid cancer is an endocrine one, characterized by decreased concentration of TSH,” said Kari Stefansson, deCODE’s CEO and senior author of the study.

Thyroid Cancer is a malignant thyroid neoplasm, which can be treated with radioactive iodine or surgical resection of the thyroid gland. The contribution of genetics to the risk of thyroid cancer is greater than to any other cancer. Thyroid cancer is classified into four main histology groups: papillary (PTC), follicular (FTC), medullary (MTC), and undifferentiated or anaplastic thyroid carcinomas. The great majority of malignant thyroid tumours are nonmedullary, either PTC (80–85%) or FTC (10–15%).

deCODE’s discovery capabilities combine its extensive population and genetic resources, including DNA samples and medical data, complete genealogical information, next generation sequencing technology, and deCODE’s proprietary bioinformatics and statistical capabilities. Over the next 18 months, deCODE and Pfizer will work together to analyse the genomes of patients to search for sequence variants that would be useful for understanding drug targets and discovering novel drug targets, that may ultimately lead to tools for patient stratification and companion diagnostics.

“This agreement is a part of deCODE’s ongoing strategy to unleash the value of human genetics,” said Dr. Kari Stefansson, founder and CEO of deCODE, “our research platform allows us to understand the genetic basis of disease and modifiers of clinical phenotypes in actual patient populations; by doing so, we can rapidly move from targets to patient stratification and from there to companion diagnostics.”

The research collaboration will utilize the expertise and capabilities of both deCODE and Pfizer: deCODE’s comprehensive population genetics resources and analytical expertise and Pfizer’s dedication to the application of genomic analysis to the discovery and development of drugs.

Scientists at deCODE Genetics and academic collaborators from Iceland, The Netherlands, Spain, Denmark, Germany, Sweden, the USA, the UK and Romania today report the discovery of a variant in the sequence of the human genome associated with risk of developing basal cell carcinoma of the skin (BCC), as well as prostate cancer and glioma, the most serious form of brain cancer.  The study was done in collaboration with Illumina, Inc., and is published today in the online edition of Nature Genetics.

Using Illumina sequencing technology, deCODE scientists determined the sequences of the entire genomes of 457 Icelanders, and identified 16 million single nucleotide polymorphisms (SNPs). Through a combination of SNP genotyping and computational techniques utilizing the extensive Icelandic genealogy, they were able to propagate those 16 million variants into over 40,000 Icelanders for use in this study.

The researchers discovered a single letter variant located in TP53, a gene known to play a central role in tumor biology and for accumulating so called somatic mutations, during the development of cancer in patients.  Until now, however, individuals who are born with defective copies of the gene (germline variants) have been found extremely rarely, only in families with cancer predisposition syndromes, Li Fraumeni syndrome (LFS) and Li-Fraumeni-like syndrome (LFL). The variant found in the present study is an unusual type of mutation that appears to affect the way the gene’s messenger RNA is processed; the messenger RNA in patients with the mutant TP53 gene appears to lack proper termination and polyadenylation.

This is the first evidence of a germline variant in TP53 associated with cancer predisposition beyond LFS and LFL. While the mutations causing LFS and LFL syndromes are very rare (occuring 1:5,000 to 1:20,000 births), the variant described in this paper occurs in ~ 1 in 25 individuals in Iceland, and at comparable frequencies in US and UK populations.

“This mutation is one of a growing number of deCODE discoveries of relatively low frequency sequence variants with large effect,” said Kari Stefansson, deCODE’s CEO and senior author of the study.  “The discovery of such variants is made possible through the breadth and quality of the data that the Icelandic population provides.”

Dr. Stefansson emphasized, “We will, together with our collaborators, including Illumina, extend ourselves to turn this discovery into benefit for patients and those at risk of cancer.”

BCC is the most common cancer in people of European ancestry. Sun exposure is the primary risk factor for BCC, but genetic predisposition also plays a substantial role.   Until now, no mechanistic causal connection between cancers as diverse as BCC, prostate cancer, glioma, and colorectal adenoma was known.

The paper, “A Germline Variant in the TP53 Polyadenylation Signal Confers Cancer Susceptibility” is published online in Nature Genetics at www.nature.com/ng and will appear in an upcoming print edition of the journal.

Iceland was mainly settled by Scandinavian men and women from Irealand and Scotland

In a paper published today scientists at deCODE genetics present the results of the largest study of ancient DNA from a single population ever undertaken. Analyzing mitochondrial DNA, which is passed from mother to offspring, from 68 skeletal remains from approximately 1000 years ago, the study provides the most detailed look to date at how a contemporary population differs from that of its ancestors. The results confirm previous deCODE work that used genetics to test the history of Iceland as recorded in the sagas.

Audio link:  Dr. Kari Stefansson interviewed on BBC WORLD. BBC

These studies demonstrated that the country seems indeed to have been settled by men from Scandinavia – the vikings – but that the majority of the original female inhabitants were from the coastal regions of Scotland and Ireland, areas that regularly suffered raids by vikings in the years around the settlement of Iceland 1100 years ago.

Perhaps the most remarkable finding of the study published today is that the gene pool of contemporary Icelanders appears to have evolved rapidly over the intervening thousand years. As a result, the original female settlers are genetically more closely related to the present day populations of Scotland, Ireland and Scandinavia, as well as those of northwestern Europe and even southwestern Europe, than they are to present day Icelanders. This is an important demonstration of a phenomenon known as ‘genetic drift.’ In essence, in any population certain individuals will have more offspring and, by chance and in this case over the course of 35 generations, many more descendants than others. And as a result, particularly in a small population, the genetic variety of the original population can decrease and change over time. In this study only mitochondrial DNA was studied, but the same phenomenon applies to the Y chromosome, which is passed from fathers to sons, and to any other part of the genome. The paper, ‘Sequences from first settlers reveal rapid evolution in Icelandic mtDNA pool,’ is published today in the open-access journal PLOS Genetics.

“This study is a major contribution to the use of ancient DNA studies in tracing the history not just of single populations, but of our species and how we spread from Africa to every corner of the globe. It is the first such study to be large enough to permit meaningful statistical methods to be applied to ancient DNA. We very much hope this will aid and encourage others to follow with large studies in other parts of the world. In this field, as in the genetics of common diseases, we are pleased and proud to be able to put the knowledge we gain in Iceland to work for the benefit of people everywhere,” said Kari Stefansson, CEO of deCODE.

deCODE scientists have discovered new genetic variants influencing BMI, weight and risk of obesity

As we all know to well, for decades the scales have been tipping in favor of obesity. The epidemic of obesity in many industrialized countries has been driven by many factors, including easy access to fast food, an increasingly sedentary lifestyle, insufficient daily physical activity. All of this while our genomes have evolved on a background of scarcity, often putting a premium on the ability of the body to turn food into fat and store energy for leaner times. A paper published today by deCODE scientists and academic colleagues from the US and Europe provide a significant advance in our knowledge of the underlying genetics and biology of obesity, providing new information for understanding and addressing obesity and perhaps nudging the scales the other way.
In a major study published today, the deCODE-led team reports the discovery of a large number of single-letter variations in the sequence of the human genome (SNPs) influencing body weight, body mass index (BMI) and risk of obesity. (BMI>30kg/m2). The discoveries were made be scanning over 300,000 SNPs in more than  30,000 individuals from Iceland, The Netherlands, and the United States, and then confirming the findings in individuals from Denmark and the multinational GIANT consortium, totaling close to 40,000 individuals. Interestingly, many of the variants discovered are located near genes related to energy sensing or food intake regulation in the brain, suggesting its importance in the development of obesity.   Although these variants only explain a small fraction of the variation in BMI, they provide new insght into the basic mechanisms underlying obesity and a first step towards identifying drug targets that can be used to address the global public health challenge of obesity.

Crown Prince Frederik of Denmark and Crown Princess Marie of Denmark along with deCODE scientist Unnur Thorsteinsdottir during an official visit to deCODE laboratories earlier this year.

Reykjavik, ICELAND, December 8, 2008 – deCODE genetics (Nasdaq:DCGN) today announced the discovery by an international consortium of scientists from deCODE and major European and US academic institutions of a single letter variation in the human genome (SNP) that is associated with increased fasting glucose levels and risk of type 2 diabetes (T2D). deCODE will employ its CLIA-registered genotyping laboratory and existing testing platform to swiftly integrate the finding into its deCODEme™ personal genome scan, and to assess the addition of this new variant to the company’s deCODE T2™ reference laboratory test for assessing individual risk of type 2 diabetes.

The multinational study analyzed a number of SNPs that had been suggestively linked with fasting glucose levels in several major studies involving some 36,000 individuals from Europe and the United States.The analysis identified a version of single SNP within the gene encoding melatonin receptor IB (MTNR1B) that was associated with notable increase in fasting glucose levels. The deCODE team then demonstrated in its Icelandic cohort that this SNP also associated with an increased risk of T2D, a finding that was then replicated in a meta-analysis of data from more than 80,000 cases and controls from Europe and the US. Approximately 10% of the participants in this study carry two copies of the at-risk version of this SNP, putting them at more than 15 percent greater risk of type 2 diabetes than individuals who carry no copies. The paper, entitled “Variants in MTNR1B influence fasting glucose levels,” is published today in the online edition of Nature Genetics, and will appear in an upcoming print edition of the journal.

“This finding is another step towards rounding out our understanding of the genetic factors that underpin glucose regulation and risk of type 2 diabetes. This variant does not confer sufficient risk to be of clinical utility on its own. But when measured in addition to our TCF7L2 variant that is the anchor of the deCODE T2™ test, it may, like other common variants conferring modest risk, enable the test to capture an even larger proportion of inherited risk. We are currently evaluating its integration into deCODE T2™, because understanding genetic risk of T2D enables individuals and their physicians to focus, personalize and improve prevention. In the meantime, we will be enabling our deCODEme subscribers to check their profiles for this new variant, keeping them at the cutting edge of human genetics” said Kari Stefansson, CEO of deCODE.

Type 2 diabetes: A major public health problem
T2D is a chronic condition that develops when the body either becomes resistant to or doesn’t secrete enough insulin. Diabetes affects nearly 200 million people worldwide and, according to the American Diabetes Association, some 21 million in the United States. The vast majority of these have T2D, and as many as one third of Americans with diabetes may not even be aware that they have the disease. More than 50 million Americans have pre-diabetes, a condition characterized by elevated blood glucose levels and which puts these individuals at high risk for developing T2D. T2D can be managed and – most importantly – prevented. If losing weight, eating better and getting adequate exercise aren’t enough, there are also medications that can help to manage blood sugar levels and insulin response to reduce the likelihood of developing diabetes. For more information on T2D and how to prevent it, you can go to the American Diabetes Association’s website.

Here are a few thoughts on Nic Fleming’s piece on personal genome scans, of which one was our own, deCODEme:

Our genomes are all remarkably similar. And so it is the differences that are most interesting and important, and that make us who we are.

The same can be said of genetic testing services. We at deCODE were not at all surprised that Mr. Fleming found that he got some varying results from the three genome scans that he tried. Indeed we would be surprised (and more than a little dismayed) if he hadn’t. Analyzing the genome – accurately detecting which genetic markers individuals have at specific points in the genome, and correlating these variations with risk of a range of common diseases – has been our bread and butter for well over a decade. With the analysis of hundreds of thousands of genomes under our belt, we can say with some authority that it is not a trivial business. We would never ourselves rely on consultants to tell us what variants to look for, what they mean, or to oversee the genotypic analysis itself. And we certainly would not offer such treatment to doctors or members of the public.

That said, it would be a strange logic that therefore suggests that the whole field, or the very well validated science that now exists linking specific markers to risk of common diseases, should be lumped in together as though all are nothing but dot-com storefronts selling DNA analysis today where they might have been selling sofas last year. deCODEme or our diagnostic tests, for example, only detect risk variants that meet exacting criteria: they must have been published in peer-reviewed scientific journals and replicated in large cohorts from several populations. Many genetic risk factors for common diseases have passed this high bar, and have thus been as well validated, as most non-genetic risk factors were when they were first brought into clinical use. As Mr. Fleming notes, there are physicians who are incorporating genetic risk factors into their clinical practice and with some important successes that individuals have been willing to share with the world. (We are posting some of these stories on this blog and on the personal stories page of deCODEme).

So, as suggested by Mr. Fleming’s piece and by Lord Taverne and others he interviewed, it is of pressing importance to establish high scientific and technical standards and regulations for such tests. We therefore hope that the Human Genetics Commission, and other oversight bodies in Europe, the US, and elsewhere will continue to scrutinize how best to provide an effective sheriff for this new territory. Doing so will enable individuals and the healthcare system to take full benefit from the potential of this new technology, while protecting the public from unscrupulous cowboys. At the least – since some people like cavorting with fun-loving bandit types – everyone would know who was who.

Edward Farmer
Chief Communications Officer
deCODE genetics

Link: Read abstract in Nature Genetics

deCODE and SGENE Consortium Discover Deletions in the Human Genome Linked to Risk of Schizophrenia

Findings may provide the foundation for a test to complement standard clinical diagnosis, potentially enabling earlier intervention and treatment

A team of scientists led by deCODE genetics has discovered evidence of three rare deletions in the human genome that confer a greater risk of schizophrenia. This discovery shows that individuals who have one of these deletions may be up to 15 times more likely to develop schizophrenia than the population at large. See “Large recurrent microdeletions associated with schizophrenia” which appeared this afternoon in Nature (www.nature.com)

“Schizophrenia is a disorder affecting thoughts and emotions. It is therefore a quintessentially human disease, but one that is little understood biologically and which is difficult to diagnose. These findings are important because they shed light on its causes and provide a first component to a molecular test to aid in clinical diagnosis and intervention. These discoveries also demonstrate one way in which we can use SNP-chips to find rarer genetic factors conferring risk of disease. In many disease areas we have had great success of late in identifying what these chips are best suited to find: common variants conferring relatively modest increases in risk. But we know that individuals with certain mental disorders such as schizophrenia tend to have few children, and thus that we may have to identify a larger number of rare but high risk variants to understand the genetic contribution to susceptibility. It is encouraging that our efforts to use SNP chips to detect rarer variations such as spontaneous deletions and duplications is now bearing fruit,” said Kari Stefansson, CEO of deCODE.

In the recent wave of discoveries of risk variants for common diseases, those associated with mental disorders such as schizophrenia, autism and others have been conspicuously absent. This phenomenon, and the fact that people with these disorders tend to have few children, suggest that rarer and perhaps spontaneously generated variants may account for a greater proportion of the disease burden in these conditions than in others. SNP-chips are not well suited to finding rare SNPs but can, with sufficiently large sample sizes, be used to identify deletions and duplications – known as copy number variations, or CNVs – which can also be carried by healthy individuals in one generation and contribute to risk of disease in the next.

A look at deCODE, the biotech company in Iceland that is researching genes for common conditions like heart disease, stroke and cancer. Sue Herera interviews NBC’s Robert Bazell on his coverage of deCODE and discusses the scientific, pharmacological and commercial implications of genetic testing.

Originally broadcast July 22, 2008.

Link: Watch the interview.

With this latest discovery, the genetic factors underpinning a very significant proportion of inherited risk of ER+ breast cancer have now been elucidated. Common variants previously discovered by deCODE are together involved in an estimated 25% of ER+ breast cancers.

deCODE made today’s discovery through the analysis of genotypic data from a total of nearly 40,000 patients and control subjects from five countries. deCODE gratefully acknowledges the participation of the patients and researchers who took part in this study.

Link: Read the paper, ‘Common variants on chromosome 5p12 confer susceptibility to estrogen receptor-positive breast cancer’, in Nature Genetics