The Next Small Thing: An Update on the Biomedical Revolution
Chairman - Eli Lilly and Company December 12, 2008
September 8, 2006
Remarks to The Economic Club of Florida - Tallahassee, Florida
Thank you, Dean Lewis. I’m delighted to be here, and I regard this visit as an important opportunity to share ideas.
Florida is a key state in America’s health care map, largely because it’s such a desirable retirement destination. In a sense, it’s a laboratory for the future of health care. It’s a place where the rest of the country can get a glimpse of what it takes to meet the challenges of an aging population. So everyone involved in the health care industry really needs to stay in touch with what’s happening here.
And, turning that around, I have to believe that Floridians have a keen interest in what sort of new answers -- new solutions -- might be forthcoming from companies like mine.
In any case, you honor me with your invitation. For my part, I want to give you a progress report on the advances in the life sciences now unfolding at Lilly ... and at labs all over the world.
This topic hasn’t received much media attention for quite some time and I can’t really explain why. Ten years ago, “the biomedical revolution” was a common phrase in news headlines.
The Human Genome project caught the popular imagination, and served as a catalyst for many upbeat stories. The coverage often projected marvelous advances in understanding human biology … and, just around the corner, related advances in medicine -- especially in pharmaceuticals.
In recent years, however, reports of biomedical advances have all but disappeared from the popular press. Instead, there’s been an unending string of negative stories about my industry, including one thread arguing that the pharmaceutical industry has failed to deliver on its promise and that, in fact, the whole glorious “revolution” has fizzled out.
I would argue that neither attitude is fair or accurate -- neither the earlier hype nor the current scorn. The biomedical revolution has not unfolded in quite the way scientists predicted 10 years ago. But it has unfolded, “exploded” would be more accurate, into a whole set of new ideas and new learning. And, in its applications to medicine, it is already delivering on its promise in some therapeutic areas with the greatest advances yet to come. One observer summarized it this way in a recent article in the journal Scientific American: “Medicine used to be hit or miss. We would find something through drug discovery that performed a useful function such as lowering blood pressure. But lacking effective models of how these interventions worked, many of these drugs turned out to be crude tools with unanticipated side effects. We are now beginning to understand biology as a set of information processes, and we’re developing realistic models and simulations of how disease and aging progress. Moreover, we are developing the tools to reprogram them.”
I believe that this represents the most exciting and important story in any technology field in this new century. Today, I’ll try to give you a brief overview of this story -- and share my view of where we stand now, where the best minds in my business believe we are headed, and what it may mean for all of us.
One disclaimer, though: I’m an MBA, not a scientist. What I’ve gleaned from our research leaders is pretty much the layman’s version. So, if you want to ask me about any very technical issues in our Q and A session, my answer in advance is, “I’ll have to get back to you on that.”
That said, let me share what I’ve seen.
A decade ago, some of our research scientists described their journey as something like the voyage of Columbus and predicted the discovery of a new world. It turned out to be an apt comparison, because, like Columbus, the explorers in the life sciences have indeed found a new world, but not exactly the one they expected. Biologists have known for a long time that the human cell hides an extraordinary world of activity and complexity that governs the normal functions of our bodies as well as the processes that result in disease. But at the same time, many believed that this complexity could eventually be greatly simplified, once they could decipher the “master code” written in our genes.
This expectation was a big part of the motivation behind the Human Genome Project. With powerful computer-driven “sequencing” tools, scientists at the National Institutes of Health set out to scan the entire length of the human DNA molecule -- a chain with 3 billion links -- and pin down the location and chemical composition of every gene that defines a human being.
Originally, there was a broad range of estimates of how many genes would be found. But the consensus was in the range of about 100,000 genes.
Well, it turned out that the actual number of genes was closer to 25,000. This was a startling discovery with powerful consequences for the new biology.
The key implication was not, as some reports had it, that “we’re not as complex as we thought.” On the contrary, the inference the experts drew was that we are even more complex than we supposed. And this complexity cannot be fully explained -- or explored -- through our genes alone.
Ironically, one of the key results of the Genome Project was to refute the hypothesis that drove it -- suggesting that there may be no single “master code.” Rather, there appear to be multiple codes within and beyond the genome that all work together to orchestrate the process of life.
This new understanding validates an emerging concept of “systems biology” grounded in the notion that all the biochemical components of our bodies interact together like the different musicians playing in an orchestra. We can’t fully understand this system by trying to study one part at a time. What really matters is how the interaction of the parts creates the properties of the system.
It also underscores one of the main principles of systems biology which proposes that biological processes are essentially information processes.
This is not a metaphor to these scientists. They cite overwhelming evidence that, at the molecular level, biological processes are information processes sending and receiving digital messages written in chemical form. So at this stage, scientists in both basic and applied research, while still trying to digest the massive output of the Genome Project, are also grappling with a whole new set of fundamental questions that cascade down from all this new information.
What is the purpose or function of each gene?
Genes work by making proteins. But what, exactly does each protein do?
What tells each gene when and where to go into action? Biologists now think the answers may be found in the parts of DNA that don’t contain genes -- parts they used to call “junk DNA.”
Also, if biological activity is really driven by information, what does the circuit diagram look like?
And finally, what can go wrong among all these components to cause disease?
So this is where the life sciences stand today. A bit like Juan Ponce de Leon, having landed on a Florida beach, understanding that this time, what lies before him is not an island, but a vast continent full of unimaginable treasures. It may well take decades for biologists to map all of this territory and fully solve all of these new puzzles. But they have every confidence that the task can be done. One pioneer in the field sees biology as unique among the sciences, in that, I quote, “the digital information of the genome is ultimately completely knowable.”
The good news is patients will not have to wait decades to see the medical benefits of this massive effort. Several key technologies are emerging to help scientists in both basic and applied research.
I’ll just mention three that seem to be finding multiple applications:
One is the use of microarrays sometimes called “biochips.” They can be used to profile the entire genome of an individual or to focus in on specific genes of interest. Biochips are also being adapted to screen for proteins. Protein arrays have the potential to give doctors much more powerful diagnostic tools, capable of detecting disorders like cancer or heart disease long before they produce physical symptoms. They can help us solve the basic research puzzle of what specific proteins do and, along the way, give us a tool for greatly accelerating drug discovery.
Another key technology is a new way of discovering and manipulating gene function called “RNA interference.” I won’t drag you through the details, but basically this allows scientists to figure out what genes do by switching them on and off. In animal testing, they “silence” a given gene and see what changes in the system. Such an experiment can tell researchers whether the gene they’re studying is worth pursuing as a target for drugs. RNA interference also has potential as a new type of therapy itself.
And a third is the group of technologies clustered under the heading of “bioinformatics.” Almost every kind of research in the life sciences is “computer-assisted.” But when people talk about “informatics,” they’re talking about technologies where the computer itself is the primary research tool.
There are numerous applications, but one that really has caught my attention is a program called “Archimedes,” which has shown truly amazing power in its ability to model the results of clinical trials. Again and again, Archimedes has been validated by feeding in all the specifications of a drug candidate, getting the program to predict an outcome. In a hundred such tests, Archimedes has predicted the outcome with 99 percent accuracy. Broadly speaking, every company in my industry is working to harness these and other advances. But at Lilly, we have a very specific strategy for bringing them all together -- a strategy focused on the innovation of tailored therapeutics.
So let me explain what this is and show you what we’re doing to create the next big thing in medicine. Or maybe I should say, since this is a microscopic world, the next small thing. As background, you need to understand that, up to now, our science has been limited to producing drugs that meet a standard of “one size fits all.” Successful drugs have a good probability of benefit and a low risk of serious side-effects.
But in reality, patients have a range of responses to any drug. Some get great benefit; some get little or none; some have negative effects. The only way to know for sure if a drug will work for any individual patient is to give it to them and see what happens.
The question we asked was “what if we could know in advance?”
This is one of the real possibilities we see in the convergence of these new streams in research. We think we can learn so much about the biochemical makeup of diseases of the patients who have them and of the drugs that could treat them that we can identify with a much higher degree of certainty that, YES, these patients will be helped by our medicine, and NO, this other group should not take it.
This is the concept behind a tailored therapeutics model. The ultimate vision would be to predictably deliver to patients “the right dose of the right drug at the right time.”
For the tailored model to work, we need to develop a comprehensive and reliable catalogue of what we call “biomarkers.” A biomarker is any type of chemical or physiological “telltale” that shows a certain kind of biological process is taking place and that other processes or outcomes will likely follow.
It’s not really a new concept -- think of blood sugar level as a biomarker for diabetes, or cholesterol levels as a marker for heart disease. But for the vast majority of diseases, we’ve never had such indicators. Now, we’re able to find them much more rapidly and methodically, thanks to advanced technologies like microarrays and bioinformatics. New biomarkers can help us pinpoint the right patient and right dose for products already on the market. But for the most part, we expect to develop biomarkers and new therapies in tandem.
For instance, we have a very exciting project under way in Alzheimer’s research.
As you may know, Alzheimer’s is typically diagnosed by it behavioral symptoms -- especially memory loss. But these symptoms only begin to appear after the underlying pathology -- the formation of plaques in the brain -- is fairly advanced. To really combat Alzheimer’s, we have to be able to detect it much earlier.
Lilly is working with a large group of partners from both industry and academia to find biomarkers through neuroimaging -- or brain scans. This would have been impossible just a few years ago, but now scanning technologies have advanced to the point that we can actually see an image of plaques forming in the brain, and at quite an early stage at that.
So far, the required high-tech scanners only exist in a couple of labs in the nation. But we are also working to adapt this quality of imaging to machines that any hospital might have.
The value of such a tool would be incredible. Obviously, it would make a huge difference in early detection. But it also would be of enormous value in our search for more effective drugs against the disease.
We have two under development right now. They are very different compounds, but both work by blocking a substance called “amyloid beta” that we suspect is the key to the plaques. Both are now in the middle phase -- Phase II -- of clinical trials
There’s a long road to travel from concept to treatment. But these are some of the most hopeful advances in decades.
Biomarkers are necessary to make our tailoring model work but are far from sufficient. We also have to find new and better therapies for unmet medical needs. And again, the new biology promises to transform pharmaceutical innovation.
The first wave is already here. It’s a relatively new class of pharmaceuticals based on antibodies. These are essentially biotech versions of naturally-occurring substances that our immune system makes to fight invaders -- like viruses and bacteria.
Antibodies are highly specific -- they attack only one target and nothing else. It’s this selectivity that makes them so valuable as drugs and so ideal for our tailored therapeutic approach.
Over the last decade, biotech antibodies have provided some of the biggest breakthroughs in medicine. The list includes Avastin, Herceptin, Rituxan and Erbitux for cancer and Enbrel, Humira, and Remicade for immune and inflammatory disorders. (None of these are Lilly products, by the way.)
Now, a new generation of antibodies is being transformed by another incredible technology -- protein engineering. Proteins are very complex molecules, and that complexity can create problems when we try to use them as drugs. With man-made chemicals -- the kind that make up most pharmaceuticals -- we have the ability to tinker -- to shape and reshape our creations in hundreds of variations until we get them into optimal form. What if we could do that with nature’s proteins? What if we could smooth and shape that complexity and remove their problems without destroying their healing properties?
That’s the idea behind protein engineering, and it’s now a reality.
A couple of years ago, Lilly acquired a San Diego company called Applied Molecular Evolution that we believe sits on the cutting edge of custom-made proteins -- antibodies in particular. The feats these specialists and their colleagues in Indianapolis are accomplishing today would have seemed like sheer science fiction 10 years ago. I’ll share two examples that I find rather mind-boggling.
One of the new-generation antibodies for cancer has oncologists both excited and frustrated. It can produce dramatic benefits against a common type of lymphoma – but only about 50 percent of patients who try this as a standalone treatment see such results.
Our protein engineering group figured out what part of the antibody might be limiting its effectiveness. And then they applied their technology to engineer a new molecule toward the same target in the cancer, but with changes designed to make it effective in the vast majority of patients.
We’ll find out if they’ve succeeded within the next couple of years.
Another example is a molecule currently labeled FGF21.
This began as another one of the interesting discoveries that came to us from fishing in the sea of genomic data -- studying over 400 new proteins. Our researchers got interested in FGF21 when they saw where the body produces this substance -- it had a pattern of distribution that resembled hormones or growth factors from the endocrine system.
One of our young scientists, who happened to train in diabetes, set up some tests to see if these proteins might have any activity related to diabetes. Lo and behold, this insight proved to be right. But as research continued, it soon became clear that the native protein could never be used as a drug for a variety of reasons. This is the common fate of all too many early-stage discoveries. And, a decade ago, that’s where the story would have ended.
But because we had invested in protein engineering, another one of our young scientists was able to create a variation of the molecule that is stable, and appears to be remarkably safe and active in animal models. Whether it will work or not is still to be proved. But in animal testing it seems to not only provide blood sugar control in diabetes, but also lowers triglycerides, lowers LDL (the bad cholesterol), raises HDL (the good cholesterol), and causes obese animals to lose weight.
Needless to say, we’re very excited to see whether it can do the same thing for humans.
Now, I suspect that many of you, as you’ve listened to all of this, have been wondering about the cost of it all.
Customized, personalized, doesn’t this have to be even more expensive than what we have today?
It’s an important question -- and my response is “no,” for a couple of reasons.
First, I believe this new model will greatly improve the cost/benefit ratio for pharmaceuticals. I noted earlier that most drugs aren’t effective for all patients. In fact, the average across all drugs is about 50 percent efficacy. And for the 50 percent of the patients who essentially get little or no benefit, whatever they spend is wasted money.
Tailoring has the potential to eliminate that waste and make every dollar deliver benefit.
Second, we think the technologies behind tailoring -- especially biomarkers -- will enable us to identify early those molecules that are likely to fail and thereby reduce the cost of producing a new drug. Based on the technological revolution I’ve tried to sketch today -- and believe me, I have only sketched a part of what’s happening -- I am certain that the medicines of tomorrow will far outperform even the best of what we have today. I believe we’ll see, in common use, treatments that would have been regarded as miraculous 10 years ago.
Such things as:
- Effective therapies against cancer -- turning it into a chronic disease rather than a death sentence. Survival time for victims will be measured in decades rather than months. Beyond this, we will certainly find ways to detect cancers in their earliest stages. And perhaps sooner than anyone supposes, we will learn how to prevent them.
- Cardiovascular repair and reversal of heart disease. We are already far down the road with treatments to prevent heart attacks. The day will come when we can reverse arterial plaques, and repair damaged tissue in the heart -- without surgery, with nothing more invasive than a drug cocktail.
- A breakthrough in Alzheimer’s. We may be right on the edge of it.
- A much better quality of life for the elderly. When you add up the progress already made in the illnesses that cause much of the disability for seniors and project it forward, you can see major improvement on so many fronts. In addition, our scientists foresee new treatments to preserve cognition and mobility and to fight frailty and pain. “Old age” will take on a new meaning.
And even these things will be overshadowed by the fruits of the new biology. The systems biology pioneers acknowledge only one ultimate boundary: Human mortality will always equal 100 percent. But they are confident that they can move medicine from its current state of treating symptoms to prediction and prevention of disease.
All of this is possible, in terms of the science and technology. But none of it is guaranteed.
Among the many obstacles are policy choices that all of you will have some role in shaping.
It’s a speech for another time, but we all know that our health care system will undergo massive pressure and major change in the near future -- as 76 million baby boomers try to join you all in Florida! As that demographic wave begins to break, we as a society will have to make some tough choices about what to fund, and what is worth what.
You will hear some people argue that we already have all the medical innovation we need.
You will hear others say that whether we need it or not, the biomedical companies have run dry, and so there’s no reason to create space for them to continue their work.
You will hear people say that even if new medicines can be invented, they will not be worth the cost.
My hope is that, when you hear those claims, you will think about what I’ve said today and make your own choice about the value of preserving medical innovation.
Back to Top