shape-zoom2 shape-zoom2 chevron-down edit share2 twitter youtube

The New Genetics

SCROLL
TO BEGIN

Tom Ivany

patient with familial hypercholesterolemia

When I was 16, my mom had a massive heart attack at 37. She has a rare disease, HEFH, which is a disease where your cholesterol levels - the LDL, which is your bad cholesterol - are extremely high, which puts you at extremely high risk for cardiac events later on in life.

Given that this is a genetic disorder, my mom's physician thought it would be best to have all her children tested to find out if we had the same disease that she has. When I was tested, I was extremely shocked to find out that my cholesterol levels were upwards of 550, which is extremely high for anyone, let alone a 16 year old kid.

I was put on lipid-controlling drugs to help keep my cholesterol levels down. My doses were raised over the years and I actually ended up being maxed out at a certain age. By the time I turned 25 I actually underwent a massive heart attack myself. From there I ended up having two stents put in.

After my heart attack I realized there weren't many options for me other than my statin drugs that I was on… There are a lot of people out there suffering with the same disease I am. It couldn't be better to find more information out there or other options to help me basically get my cholesterol levels at a normal range.

READ TRANSCRIPT

For better or worse, our genes help determine which illnesses we're likely to get or avoid. Some of us are fortunate to have a genome that seems tailor-made for longevity. Other people inherit genes that may cause disease from birth or early in life.

The sheer size and variability of the genome makes it difficult to untangle this link between our DNA and specific diseases. In book form, our full genetic code would fill more than a million pages.

Variations in a single letter, or nucleotide, might harm us or protect us. Picking out the variants that matter from mountains of background data has been an extremely daunting and slow-moving project. That's beginning to change now, thanks to new gene sequencing technologies and computational tools that help scientists make sense of our genetic complexity.

The cost to sequence a full human genome — all 3 billion pairs of nucleotides — has plunged. The first fully deciphered genome cost an estimated $2.7 billion.1

As recently as 2008, the per-genome price tag still exceeded $1 million.2 But with that cost now at $1,0003 and falling, it's practical to fully sequence the genomes of many people and use this in-depth genetic data to search for disease genes.

Understanding how genes influence disease risk is a core element of Amgen's research and development strategy. This strategy aims to benefit millions of patients with serious illness by using newly-accessible genetic information to set the path for novel, breakaway therapies.

NEXT CHAPTER
A Faster Route
to Clinical Success

A Faster Route to Clinical Success

SCROLL
TO CONTINUE

Genetic insights can give scientists evidence that a particular target protein plays a pivotal role in disease. Better targets, in turn, could reverse a trend that has vexed the biopharmaceutical industry.


In other tech-based industries, the general trend is that products keep getting better and cheaper. Unique to the biopharmaceutical industry, the cost to bring a new medicine to the marketplace has been rising steadily for more than five decades. Only half 4 of all Phase 3 programs lead to regulatory approvals.

Sean E. Harper, M.D.

executive vice president, Research and Development

We're constantly striving to improve our probability of success, reduce our time cycles, and make the whole process of developing a new medicine for patients more efficient. Right now in the industry, we tend to rely very heavily on animal models of disease, usually in species like mice. And unfortunately they can be fairly poor predictors of what actually happens in humans and whether a target, a drug target, is relevant in human disease. But you don't find that out until you've spent ten years and a billion dollars developing a medicine and it doesn't work in clinical trials in humans. So we're taking advantage now of the revolution that's going on in human genetics, driven by the revolution in DNA sequencing, that is allowing us to, in many cases, validate targets in humans, before we engage in that journey. And we validate by looking at the natural human variation that can occur in genes and correlating that with the risk of disease.

READ TRANSCRIPT

Amgen's late-stage pipeline already includes several programs rooted in human genetics. People who lack a key protein called sclerostin have abnormally thick bones.5  A sclerostin-inhibiting therapy is now being studied as a potential bone-building drug.


Different types of mutations in PCSK9, 6, 7  a protein that degrades LDL receptors, can either raise or lower levels of LDL cholesterol. This insight was used to develop an investigational PCSK9 inhibitor for dyslipidemia.

There is evidence that drug candidates backed by genetic evidence have a higher than average 8  success rate in Phase 3 Amgen's experience also shows that genetic validation can be used to accelerate drug development timelines. And by lowering late-stage attrition, genetics can also potentially help lower the total cost of R&D.


While human genetics may make drug development faster, more successful, and more cost effective, achieving these benefits has taken longer than first anticipated. That's largely because the most therapeutically useful genetic variants are also the hardest to find because in many cases they are rare.

In a paper published in Nature Biotechnology, senior scientists from Amgen and deCODE Genetics made the case that human genetics “provides perhaps the single best opportunity to innovate and improve clinical success rates in drug development.” To support this assertion, the authors used industry data to examine hundreds of Phase 3 trials for non-cancer drugs initiated between 2000 and 2008. These studies involved 200 different drug targets, 28 of which were determined to have strong genetic validation.


Of these 28 targets, 21 formed the basis for successful Phase 3 programs (75 percent), and the seven Phase 3 failures observed in this subgroup were due to factors other than the target's relevance (e.g., problems with the drug's pharmacology or lack of an appropriate comparator drug). The authors concluded that “all targets with clear genetic evidence and good pharmacological agents in this set produce the clinical effects predicted by human genetics.”


NEXT CHAPTER
In Search of
Rare Variants

In Search of Rare Variants

SCROLL
TO CONTINUE

People have been awaiting a gene-driven surge in novel therapies since the first draft sequence of a human genome was published in 2001. 9 Predictions that this impact would come quickly were off base for several reasons.


For starters, even successful discovery projects are just the first step in a drug development journey that takes a decade or more to complete. More importantly, most of the studies that followed the Genome Project were GWAS—genome-wide association studies—and the first round of GWAS were designed to uncover common genetic variants, not rare variants.

Sasha Kamb, Ph.D.

senior vice president, Discovery Research

Genome-wide association studies have been very useful in understanding the effect of common genetic variants in human populations. I would say they've been much less impactful in pointing the way towards important disease biology and novel mechanisms of disease and therapy.

READ TRANSCRIPT

Common variants help to determine our place on the normal curve of physiological function. If you're at one end of the curve for a given gene, you may have a somewhat smaller risk of disease; if you're at the other end, you may have a slightly higher risk.10  The confluence of several common variants may elevate risk more substantially, but in general, this type of variant doesn't provide a "smoking gun" in terms of drug targets.


By contrast, rare mutations in genes like sclerostin and PCSK9 can disrupt normal physiological functions, and in doing so, illuminate the underlying biology of disease. Unfortunately, discoveries of rare variants have been rare events themselves—until recently.

Advances in DNA sequencing and computational methods have made it possible to collect and analyze mountains of in-depth human genetic data. The "New Genetics" represents a fundamental upgrade in our ability to find the genetic roots of human diversity and disease.


One of the leading organizations driving this revolution is deCODE Genetics. Founded by Kári Stefánsson and based in Reykjavik, Iceland, deCODE is the world's most accomplished gene discovery company.

Kári Stefánsson, M.D.

Dr. Med., vice president, Research, and president, deCODE Genetics

We have over the past 15 years published about 350 papers on discoveries we have made. We have made discoveries of a very large number of common variants that affect the risk of common diseases. We are now picking up one rare variant after another. We have probably discovered rare variants in 20 to 30 diseases now... They are variants that are in the middle of genes. They show us the gene that is being affected; the gene that is being affected shows us the biochemical pathway that is being affected...

And what you can do is that you can take the mutations, you can take an individual who has a mutation, and you can study that individual. You can use the biochemical perturbation conferred on the individual by the mutation as a means of demonstrating how it is relevant to a disease. So you can actually use very rare variants that you find in one or two or three individuals to bring about completely new insights into the way in which you can manipulate a disease.

READ TRANSCRIPT


deCODE has been part of Amgen since December 2012. As an Amgen subsidiary, deCODE has maintained its long-standing mission of gene discovery, and it continues to share its discoveries through publications in top scientific journals. Amgen, in turn, has early access to deCODE's unpublished research as a source of leads for its drug discovery efforts.


deCODE scientists have also helped to assess the genetic evidence for drug targets represented in Amgen's pipeline. In fact, within weeks of the deCODE acquisition, scientists in Reykjavik helped Amgen to make a major decision—and avoid a large investment that proved questionable.

Sean E. Harper, M.D.

executive vice president, Research and Development

A great example of the way that the kind of information that deCODE can provide to us is valuable is a case where we had a pathway and a molecule and a target that we were working on that was going to require a very expensive development program, with cardiovascular outcomes trials before we could even get that drug to the marketplace to help patients. And the hypothesis was that perturbating that particular gene product would result in changes in lipids in blood, and that it would reduce cardiovascular risk. We were able to identify patients who had natural variation in that gene, and indeed they did have the changes in the lipids in the blood, but they did not have differences in their cardiovascular risk of things like heart attack, stroke, sudden death. And so because of that, we began to view the program as extremely high risk, and to look with skepticism about some of the animal model data that had suggested the benefits of this approach. And we instead invested resources in other, better-validated programs and decided to let others pursue that particular target.

READ TRANSCRIPT

Overall, this analysis strengthened Amgen's R&D pipeline by accelerating 19 programs based on targets backed by positive genetic data and eliminating six targets with negative data.


deCODE's unique capabilities in population genetics complement Amgen's expertise in gene expression research, a discipline that looks for differences in the genes that are active in diseased and healthy tissue. That research is centered in Amgen's Genome Analysis Unit (GAU), a group of 40 scientists who function like a small biotechnology firm embedded in the larger company. The GAU's research has been particularly useful in finding targets that are highly expressed in cancer cells but not in healthy tissue. The group also specializes in probing the biology of newly discovered genes from deCODE and other sources.


Taken together, these capabilities have placed Amgen at the forefront of efforts to use the New Genetics to reengineer how drugs are discovered.

Using Genetic Data to Strengthen the Pipeline


As the quantity and quality of human genetic data keeps improving, these data can be used to re-evaluate the programs in a drug pipeline. After Amgen acquired deCODE Genetics, scientists at deCODE took a closer look at the non-oncology targets Amgen was working on.


"In three-quarters of the cases, the genetics were silent because human genetic variation hasn't been fully characterized," said Sasha Kamb, senior vice president, Discovery Research. "But in cases where we could get a read, 80 percent of the data was positive—the genetic evidence supported the target. In the other 20 percent, the genetic data didn't support the target. That input allowed us to redouble our efforts on 19 programs and terminate six programs. We also resuscitated a program we had killed because a preclinical rodent study provided no evidence of efficacy. But the human genetics provided a good reason to believe the target would be effective."

Sasha Kamb, Ph.D.

senior vice president, Discovery Research

Amgen and deCODE are in some ways ideal drug discovery partners. Amgen has a long history of drug discovery and development. deCODE is arguably the world's leader in the last five to ten years in human genetics discoveries, and so together, deCODE provides starting points to allow Amgen's drug discovery machine to then deliver medicines based on those genetic discoveries.

READ TRANSCRIPT

NEXT CHAPTER
A Trove of
Targets

A Trove of Targets

SCROLL
TO CONTINUE

In Amgen today, a remarkable experiment is unfolding to test the enormous promise of gene-driven drug discovery. The experiment is focused on a trove of new disease genes discovered by deCODE, each one a potential starting point for a novel, breakaway therapy. The genes behind these targets have an unambiguous impact on disease risk, and they drive disease through pathways that are either undiscovered or unsuspected.

Kári Stefánsson, M.D.

Dr. Med., vice president, Research, and president, deCODE Genetics

We found a mutation in a cell surface receptor that causes a decrease in LDL cholesterol and confers about 60 percent protection against myocardial infarction. It's a cell surface receptor that the field has not been focusing on, is fairly unexpected when it comes to its impact both on the blood lipids and the risk of myocardial infarction. And actually, the fact of this mutation both on LDL cholesterol and risk of heart attack and actually on life span is greater than the effect on mutations in the targets that the field is currently working on... So it's an example of a new target in an indication where Amgen is already investing a lot. So it has the infrastructure, it has the knowhow, it has the experience to take such a target forward fairly rapidly.

READ TRANSCRIPT

Amgen's discovery pipeline now includes more than a dozen programs based on genes discovered by deCODE that substantially raise or reduce the risk of disease. But to make the most of this unique opportunity, Amgen's discovery scientists will need to overcome a pair of challenges.


The first challenge is the need to explore and illuminate uncharted regions of human biology. Most drug discovery programs begin on pathways mapped by years of basic research, but many of these newly discovered disease genes lie in unexplored terrain.

Stefan McDonough, Ph.D.

executive director, Research

We don't know where exactly within the gene a variant will come out to be important to be associated with disease. So there is a lot of upfront a biology that really needs to be done to sort of understand how exactly the genetic variation leads to a disease, and that enables us to understand how to treat the disease. For a lot of these, because these are novel insights and novel discoveries, many of them, you really don't have an off-the-shelf toolkit to address what's going to happen. You have to sort of develop the story as you go and develop the tools.

It can be frustrating sometimes but for a biologist it's really a lot of fun. I mean to have the opportunity to actually work with novel genetic associations with disease, try to figure out what they do and then know at the end this is going to be at the basis of something new that really helps people, it's very inspiring.

READ TRANSCRIPT

The second challenge involves the nature of some of the promising genes that deCODE has found. A fair number point to drug targets that may be difficult to engage. They include novel transporters or novel types of ion channel systems that may be difficult to engage effectively.


With targets like these, small molecules or traditional antibodies may not work. Drug discovery teams may need to employ less conventional approaches, such as a bispecific molecule, an antibody fragment, or some other type of novel hybrid molecule.

Amgen is one of the few companies with the skills and know-how to engineer these more sophisticated molecules. The company's product portfolio and pipeline includes nine different drug modalities, or structural templates, each of which has its own distinct way of engaging and modulating human biology.

Les Miranda, Ph.D.

director, Research

At Amgen, we have a rich history in small molecule and large molecule science. We have the ability to create novel hybrid molecules and we do so to meet these increasingly challenging targets. It takes a lot of different people from a lot of different places. It takes biologists. It takes chemists. It takes molecular designers. It takes engineers and other people who can contribute to the process. At Amgen, we have created an environment that inspires us to be creative and innovative and we do that through a collaborative process.

READ TRANSCRIPT

Philip Tagari, M.A.

vice president, Therapeutic Discovery

Protein engineering is a core scientific strength of Amgen. We are very good at being able to express and purify poorly-behaved proteins that we can use as tools to investigate new biology in this, the Century of Biology. We've taken the scientists who perform this activity and combined them into a single organization that we believe is unique in the industry, and we've given them a major increase in resources and technology. We believe that providing our drug discovery teams with the best possible tools is where discovery research will live or die in the next ten years.

READ TRANSCRIPT

Many companies would be reluctant to tackle hard-to-hit targets rooted in novel biology, but Amgen has the people, tools and the extensive experience to proceed with confidence. In fact, seven Amgen therapies approved for use in patients can be traced back to genes that were first cloned in Amgen's own labs.

NEXT CHAPTER
Squaring
the Circle

Squaring The Circle

SCROLL
TO CONTINUE

It typically takes several years for a drug discovery program to generate an experimental therapy that's ready for clinical trials. When you're dealing with novel biology, this preclinical work can be longer and more demanding.


Amgen believes that the extra effort upfront is more than justified by the chance to achieve two goals that are often in conflict.


If you try to deliver first-in-class drugs that are based on unproven targets, you tend to have lower success rates in the clinic. If you try to improve your success by pursuing targets backed by strong evidence, you have lots of competition, which makes it hard to differentiate your therapy.

Sasha Kamb, Ph.D.

senior vice president, Discovery Research

The genetics-based drug discovery approach that we're utilizing at Amgen gives us a chance to square the circle. And by square the circle, I mean the chance to not only discover new biology, new mechanisms to treat disease, ultimately new medicines, but also to discover medicines that have a higher probability of success in the clinic. We are confident in our strategy because we think there are perhaps no other strategies that can plausibly deliver differentiated, novel medicines that also have a higher probability of success in the clinic. And of course there are challenges to the strategy but we're confident in the coming years we will execute and deliver.

READ TRANSCRIPT

This information is current as of December 8, 2014. Amgen's product pipeline will change over time as molecules move through the drug development process, including progressing to market or failing in clinical trials, due to the nature of the development process. This description contains forward-looking statements that involve significant risks and uncertainties, including those discussed in Amgen's most recent Form 10-K and in Amgen's periodic reports on Form 10-Q and Form 8-K, and actual results may vary materially. Amgen is providing this information as of the date above and does not undertake any obligation to update any forward-looking statements contained in this as a result of new information, future events or otherwise.

REFERENCES

References

  1. The Human Genome Project Completion: Frequently Asked Questions. Genome.gov. 2010 Oct 30. Available from: http://www.genome.gov/11006943
  2. Wetterstrand KA. DNA Sequencing Costs: Data from the NHGRI Genome Sequencing Program (GSP) Available at: http://www.genome.gov/sequencingcosts
  3. Illumina System Specification Sheet: Seqeuncing. HiSeqTM Ten. $1,000 human genome and extreme throughput for population-scale sequencing. Accessed at: http://systems.illumina.com/content/dam/illumina-marketing/documents/products/datasheets/datasheet-hiseq-x-ten.pdf 10/30/14.
  4. Hay M, Thomas D, Craighead J, Economides C, Rosenthal J. Clinical development success rates for investigational drugs. Nature Biotechnology. 2014 Jan 9;32:40-51. 5. Brunkow ME, et al. Bone Dysplasia Sclerosteosis Results from Loss of the SOST Gene Product, a Novel Cystine Knot-Containing Protein. The American Journal of Human Genetics. 2001;68.3:577-589.
  5. Brunkow ME, et al. Bone Dysplasia Sclerosteosis Results from Loss of the SOST Gene Product, a Novel Cystine Knot-Containing Protein. The American Journal of Human Genetics. 2001;68.3:577-589.
  6. Abifadel M, et al. Mutations in PCSK9 cause autosomal dominant hypercholesterolemia. Nat Genet. 2003 Jun;34(2):154-6.
  7. Cohen J, et al. Low LDL cholesterol in individuals of African descent resulting from frequent nonsense mutations in PCSK9. Nature genetics 2005;37.2: 161-16.
  8. Kamb A, Harper S, Stefánsson K. Human genetics as a foundation for innovative drug development. Nature Biotechnology. 2013 Nov 8;31:975-978.
  9. Lander ES, Linton LM, Birren B, Nusbaum C, Zody MC, Baldwin J, et al. Initial sequencing and analysis of the human genome. Nature. 2001 Feb 15;409:860-921.
  10. Lohmueller KE, et al. Meta-analysis of genetic association studies supports a contribution of common variants to susceptibility to common disease. Nature genetics. 2003;33.2:177-182.