Robert S. Langer, "Biomaterials and How They Will Change Our Lives” - Presidential Fellow Lecture

[MUSIC PLAYING]

SPEAKER: It's a real pleasure to introduce the speaker today. This is the second in this year's series of presidential fellow lectures hosted by the Presidential Fellow Society at MIT. The idea is to have a series of lectures that bring together graduate students and faculty colleagues from across the institution to share in some of the exciting research and issues of scholarship that are being undertaken on the campus.

And it's a real pleasure today to have Bob Langer from the Department of Chemical Engineering as our speaker. Now, Bob graduated from MIT with a PhD in chemical engineering in 1974. He did a postdoc at the Children's Hospital with Judith Foxman thereafter, and then he came back and joined the faculty here.

When you look at Bob, you will not think that just looking at him and thinking about 1974 is in my time not very long ago what he has accomplished in the years at MIT. If you read Time Magazine or other popular press, you'll see words like "father of modern drug delivery" associated with his name. He has invented methods for delivering large molecules through membranes into the body, and he'll talk about many of these techniques today. He'll also talk about the applications and impact of the work which is as dramatic as the science and engineering.

He has won an enormous number of honors for his research on innumerable honors from professional societies. I think one of the most outstanding tributes to his work is he is currently an active member of all three National Academies-- The National Academy of Science, The National Academy of Engineering, and The Institute of Medicine. I may be wrong, but I believe Bob is the only active member of all three academies. He's also won the Lemelson Prize for innovation based on the impact both in a societal way and a commercial way of his many research accomplishments.

He holds over 400 patents as well. Most recently, you probably saw the press that Bob was awarded the Draper Prize-- Draper Prize for Engineering awarded by the National Academy of Engineering. Engineers would like to think of the Draper Prize as our version of the Nobel. And if you look at the people who've come before Bob and you look at their inventions, you would, I think, resonate with that idea-- the inventors of the jet engine, the developers of Fortran-- probably the first modern computer language. Now some of you may curse these people, but the developers of Fortran, the developers of modern catalysts, and petrochemical engineering, so Bob is an incredible company there.

Beside all of these things, Bob is an extraordinary human being. He has one property that has gotten in trouble once again. Bob never says no, ever. And he's here today to deliver the lecture "Biomaterials and How They Will Change Our Lives." Bob?

[APPLAUSE]

BOB LANGER: Thank you very much. It's really a pleasure to be able to speak with you today. And what I'd like to do is tell you about some of our research. Whenever I talk about our research, people would always tell me that I'm talking about the future because it's so far off.

But what I'd like to do is tell you about how we've tried to use materials to deliver drugs to our bodies-- sometimes to very specific sites in our bodies, which might lead to new kinds of medical treatments. And towards the end, I'd like to actually tell you how you can actually apply some of these approaches with materials to actually create new tissues in the human body.

So the first thing I want to do is tell you a little bit about this field of drug delivery. A couple of years ago, I got asked actually to give a lecture to graduate students at the University of California at Berkeley, and I wasn't really thinking that much. And the way I started to talk, I said probably everyone here has probably taken drugs at some point or other.

That's what they all did. They all laughed. But what I meant was that they'd probably take an aspirin or things like that.

And the problem is that whenever you do take drugs of any type, the drug level starts out very low. It rises over time, reaches some type of peak, and then goes down. So several hours or a day later, you repeat the same thing.

But the problem with that type of delivery, which happens for all drugs pretty much, is that those peaks can be unsafe, and those valleys, the drugs are not effective. And that the consequence of that's profound. Side effects from taking drugs-- even the way you're supposed to-- can cause many, many deaths every year. The journal, American Medical Association, estimates that over 100,000 deaths each year occur because of these kinds of adverse drug effects often caused by this type of way.

So if you were able to deliver drugs at a steady controlled rate to where you want to, the hope would be that could have a profound impact on human health. And actually, even though it's a very young field, I believe it's already doing that. If one looks at this from a purely economic standpoint, in 1980, the sales of advanced drug delivery systems, which I'll be talking to you about, would be about zero dollars. But if you look at last year, 2001, in the United States alone, the sales of these types of delivery systems exceeded $20 billion. So it's actually moving quite quickly.

And so what I'd like to do today is tell you a little bit about this field and maybe some of the things you might expect to see sometime in the future. Let me see if I do this right. I've never used this computer presentation before, so do I just press this thing? Where?

AUDIENCE: [INAUDIBLE]

BOB LANGER: No, I do. Oh, it's up-- jeez, okay. Yeah, usually, I do this, and they're all I got a pleasant surprise today that I was going to have a much more efficient presentation for you.

So anyhow, this. Is an example of a controlled drug delivery system. It's a nitroglycerin patch. It delivers nitroglycerin over 24 hours. And so this is kind of a one-day system-- it was one of the earliest ones to come out in 1982. But over half a billion of these were used last year.

And one of the great things about chemical engineering is you can employ principles like slow diffusion, even through very tiny systems. This is a graph of a system known as the Norplant. And what you can see is these systems are the sizes of matchsticks. Yet, by slow diffusion to silicon rubber, you can get released for over five years from a single implant, and then you remove it. Well I'd like to actually now go back in time and tell you a little bit about how I got interested in this field.

And as Bob mentioned, after I graduated from MIT, I went to work with Judah Folkman at Boston Children's Hospital. And really, when I started this work that I'm going to talk about. It was pretty basic kind of stuff.

I was studying this process of how blood vessels grow in the body, and I was trying to develop assays, bioassays to study that. But these processes would take a long time, like 30 days, and the problem is the molecules we were studying were often large molecules. And they were destroyed rather quickly in the body. So what I became interested in doing to develop these bioassays was seeing if I could have a way to slowly deliver these macromolecules.

Now, in 1974, we started this work. I don't think really other than that field anybody cared. But what happened is just about four or five years later, we'd see the whole advent of biotechnology and genetic engineering.

So for the first time, it became possible to produce large molecular weight drugs, peptides, proteins and so forth. And these drugs pose very serious drug delivery challenges. You couldn't swallow them because they were so big, very hard to take to the skin because they're so big. And if you injected them, what you'd often see are very short lifetimes in the body. But if you wanted to use one of these types of drugs on a chronic basis, it would be critical to have a way to deliver it in an unaltered form and yet protect it from harm.

Now when we started this work, the conventional wisdom in this young field was that you couldn't really get large molecules through biocompatible polymers. In fact, in the 1970s, when you look at quotes-- and I'll just show you this one from 25 years ago-- when people were describing this young field, which was then largely in the research stage, this is what they would say. And I'll just read this to you. It's a direct quote.

It says, "the agent to be released is a small molecule with a molecular weight no larger than a few hundred. One would not expect that macromolecules, for example, proteins, could be released because of their extremely small permeation rates through polymers." Actually, this quote was very helpful to us in getting some of those patents that Bob mentioned.

I spent several years in the laboratory, and I was able to find that you could take certain very hydrophobic polymers. Polymers that absorb very little water. And by dissolving them in the right kind of solvent, adding powdered drug, and doing this at a very low temperature, we could make little microspheres. Here's one cut in half. And by using these approaches, what we were able to show is that you could get released of a variety of different molecules, peptides, and proteins ranging in molecular weight from 14,000 to a quarter of a million for over 100 days.

Now in these early studies, you can see that the release rate weren't constant. So we used some chemical engineering modeling approaches to figure out why this was, and we were actually able to construct microspheres that could get us absolutely constant release. And this is an example of that where we're releasing albumin for over 50 days.

When I first started doing my work, I mean, now I think this is about the seventh or eighth talk I've given this week. But the thing is when I first started my career, I was not a good speaker at all, and I was very scared, actually, of giving talks.

And the very first time I got asked to talk at an international meeting was in 1976 when I was a postdoc, and it was in Midland, Michigan. That's where Dow and Dow Corning is. And I was actually quite nervous about it. So I stopped, actually, work about 2 and 1/2 weeks before the talk, and I kept practicing my 20-minute talk over and over again into a tape recorder.

So finally, the day came and that was, like Bob said, 26 years ago. So I was quite a bit younger than I am now. And I was giving this talk to this pretty distinguished audience of more elder polymer scientists and engineers. So I got up, and I gave the talk.

And actually, at the end of it, I actually felt pretty good about it. I didn't forget too much of what I wanted to say. I didn't stammer too much.

And so I thought when I was done with this talk that this more older scientists being nice people would want to encourage me, this young guy. So when I got done though, a whole bunch of them came up to me after the talk, and they said, well, we don't believe anything you just said. It was kind of my introduction to how scientists sometimes interact with other scientists.

At any rate, it wasn't until another three years later, 1979, that a variety of groups like the National Institutes of Health in other places repeated what we'd done. And then the question shifted to how could this possibly happen? So Rajan Bawa, one of my graduate students, started taking techniques that were normally used in pathology, like using a microtone. And he would cut thin sections through these polymers, and we would visualize them.

So here's a five-micron thin section of one of the polymers we used, ethylene vinyl acetate. And if a molecule was 300 molecular weight or greater, it could not diffuse from one side of this to the other. But if you put a protein in, like here's a reddish protein myoglobin, if you put that in and you cut a slice before any release, what you see is what we call a phase separation where the polymer's here, and the protein's here.

Now if you take one of these systems-- millimeter-thick system-- and release it, say, for a year and then cut a thin section, what you see is left behind-- where the red protein was-- are these pores. And clearly, it's the way we've made these that enable these pores to occur because you didn't see it two slides ago. But it turns out that these pores you can do a lot of microscopy on it. These pores have a lot of tight constrictions in them, and they're also incredibly winding and torturous. So it takes a very long time to get through.

One way I explain this to people around the country is it's kind of like driving a car through Boston. Boston has a very high torch philosophy. And so if anybody's ever been there, they understand.

Now, it turns out, we've also done a lot of work kind of modeling these systems and understanding how to regulate the kinetics of these systems. And it turns out that by manipulating the pore structures, you can take very, very tiny systems and get your release almost anywhere from a day to several years. In fact, the biggest problem that I have found as a professor in getting long-term release actually turns out not to be the design of these systems but rather my ability to find a graduate student who will do the experiment that long.

And actually, it turned out in the 1980s, I had a very good graduate student named Marc Saltzman, and he actually did this experiment for over three years for his thesis. And then one day, he came up to me. He said, Bob, I've been thinking. He said what will I do if this experiment fails? He said do I have to repeat it?

So we stopped at three years and two months and five days, but I've always thought with the right student, we could probably go six or seven years. That's not to scare anybody. Mark, actually, had gotten endowed chair at Cornell. And now, Yale just recruited him to head up there a bioengineering program.

Now one of my goals also has been to not just do the research, but we wanted to really try to see the work have an impact, ultimately, in people. And so some of my students, like Larry Brown, who actually was diabetic himself, would actually develop systems, in this case, for delivering insulin. So here we have a tiny system no bigger than an aspirin pill, and you can take this and implant it in a rat and lower the blood sugar from 400, which is diabetic, to 100 which is normal, and keep it there for over 100 days. I'm going to come back to this point a little later to show you how you could regulate it. And this would begin to establish the in vivo feasibility of these kinds of things.

Now how could we get people interested, companies interested? Well, when we do these things ourselves as professors, we always kind of think that what we're doing is important. And I would try to get people interested. And I could tell you for the first 9 or 10 years I was doing this, absolutely nobody cared.

And finally at about 1983, '84, a Japanese company, actually Takeda, they called up, and they said, well, we'd like to give you some grant money. Any professor hears that, that's magic. But they said what we'd like in return is we want to send a scientist to your lab each year. I said that's great.

And so actually, they did that. But what they did is actually then went back, and they made actually the very first controlled release system for a peptide. That's shown here. They actually developed a drug called leuprolide acetate. It's about 1,200 molecular weight, and it's very powerful drug.

But if they gave it by any conventional method, it was destroyed right away, so they put it in these little microspheres. And originally, they would last for a month. They were injectable.

Now they actually have made them to last for four months. And today, this is actually the most conventional way of treating advanced prostate cancer and endometriosis or precocious puberty. It's over a billion dollar a year product.

Now to get people interested in protein delivery, that was even harder. And in fact, we couldn't get any company to be interested in that. So Alex Klibanov, who is a professor of chemistry, and myself actually decided to start a company, which would ultimately merge into Alkermes. And Alkermes is working together with Genentech using this approach basically developed the very first controlled release system for protein.

And this is human growth hormone, which normally you have to take once a day. Now you can take it once a month. And there are many, many others in various stages of clinical trials. There's one again by Alkermes that's at the FDA. It's gone through all the trials for schizophrenia, many others at various other stages.

So basically, we've been able to get some of the fundamentals done to release these kinds of molecules at either decreasing or steady rates, but you always like to do better. And one of the things that I wanted to do was someday to be able to make these systems smart. Like say for that diabetic situation, could you ever release more drug when you want to? So a couple of years ago, I was watching a TV show, and usually, I was watching a PBS show. And they were talking about how microchips were made in the computer industry.

And so I started thinking, gee, this might be an interesting way to develop chemical delivery systems. So I called up Michael Sema, who is a professor in material science, and I asked him if he could help us. And what we came up with was this kind of design as a prototype. And instead of like a regular chip, this one contains real chemicals that you can store and release to the outside world.

So here's the idea that basically you have an impermeable coat down here, like an epoxy. You build little wells in, and you cover them with gold. This is a cutaway. You could, by the way, use of various materials. You don't have to use these, but turns out, these are pretty good in the body.

At any rate, this the drugs will stay in there for pretty much as long as you want. But if you apply just one vault selectively to any of these little micro pattern wires, they will actually, as I'll show you, dissolve the gold, and then whatever's underneath it can come out. It will come out right away if you put no resistance underneath it. But if I want to make it take longer to come out, I could put a polymer or gel or something like That and essentially, hopefully, get almost any delivery pattern.

So I had a very good graduate student, John Santini, and he actually designed some of these. This one-- what you're looking at is the United States dime, and these are the two sides of the chip. There's 34 wells on this. He actually has a company to do this-- over 100 wells-- and I think pretty soon, it will be over 1,000.

Now sometimes people ask could you put much drug in the answer to that really depends on the clinical application but a way of thinking about this is let's say you could put a 1 cubic centimeter chip as an implant or a pill. Well, the way to think about that is 50% of the volume of the chip could actually be the drug. So if you make a conservative assumption that the density of the drug is the density of water, one gram per cc, that means you could put 500 milligrams of drug in such a system. What does that mean? 5,000.1 milligram dosages or 501-milligram dosages or 50 10 milligram dosages-- that's a way that you might think about it.

So we made these chips and just to look at how they work-- here's an example, looking at again a micrograph of just a single well. And it'll stay like this for years. Hermetically sealed, the drug in place. But if you just apply one volt to this, in 10 seconds, here's what happens.

The gold comes off. And as soon as it does, the drug can come right out. So what we've been able to do is you could take one drug and deliver it in multiple doses. Here's an example where we've taken a single drug and arbitrarily released it at one and a half days-- all different doses in the well at one and a half days, 2 and 1/2 days, 4 and 1/2 days, and 5 and 1/2 days. Each different well is a different amount.

And again, these are purely arbitrary. We could have selected essentially any pattern we want. Someday our thinking is that you could almost have a pharmacy on a chip.

Let's say you wanted to deliver multiple drugs at multiple times. And as a prototype of that, John released multiple drugs here. And here what you're seeing is the drug released at 2 and 1/2 days, another drug from a different well at 3 and 1/2 days, this same drug from a different well at five days, and this same drug from a different well at six days. Again, it's arbitrary.

Now since this time, we've also taken this in vivo in rats and done pretty extensive biocompatibility studies on these materials, and they seem quite safe in animals. Our long range goals for this, which we're currently working on, are to see about ultimate applications in a variety of diseases. But some of the things that we've been trying to do or could, for example, you deliver this by remote control, like by telemetry? So probably you could open these wells up but by telemetry in much the same way you open up a garage door. You could open up any individual well whenever you'd want.

Also, what we're trying to do is incorporate biosensors into these chips, along with a power source and a microprocessor so that you could essentially have a direct feedback control system. Also, what we hope to do since it's a chip, you could actually download the information. Whenever you deliver a drug, you could actually have the information be transmitted to your computer in your home, your doctor's office, or the hospital. So all these are going to happen in the future someday, we believe, and hopefully, do some very useful things in medicine.

Now so far what I've told you about is how we can kind of take materials and engineer them to do things that they couldn't do before. But another concern I've had is the materials themselves. And when I was working in the hospital many years ago, I was very curious to see how materials would find their way into medicine. And being a chemical engineer, I guess I always thought that the driving force for bringing materials into medicine would be other chemical engineers that were before us or chemists or material scientists.

But as I examined this, I found that was rarely the case. Almost always the driving force for bringing materials into medicine were like companies and very often clinicians. And what the clinicians would do is they actually wanted to solve a problem relatively quickly, and the way they would do it is they would go to their house and kind of find a household object that resembled the organ or tissue they wanted to fix.

Let me explain that. And this is all true. I mean, people may think it's funny, but they're all true.

So let me tell you about the artificial heart, which you read about a lot in the newspapers. today, but you probably don't read about the origins of the material so much. Well, the origins of all this started in 1967 at the National Institutes of Health, and there some of the clinicians were trying to figure out how to make an artificial heart.

And what they thought about from a material standpoint is something with a good flex life, and they thought what object has a good flex life? They said a lady's girdle. They said, what's it made out of? Polyetherurethane.

So that's what they made the artificial heart out of in 1967. Problem is that's what they make it out of in 2002 as well. And you can look at various journals, like the Time or Newsweek or scientific journals. It doesn't really matter.

One of the biggest problems in the artificial hearts-- it turns out the mechanical engineers have done a pretty good job. It beats long enough. But when blood hits the surface of the artificial heart, the ladies' girdle material, it can form a clot, which could go to the brain. And the patient could get a stroke and die. And yet, if you think about it, something that was designed to be a lady's girdle might not be the optimal material to put in contact with blood.

And this problem pervades all of medicine. Dialysis tubing was originally sausage casing. Vascular grafts-- that's an artificial blood vessel-- that was the surgeon in Texas going to a clothes store to find what he could sew well with.

And by the way, I don't want to minimize these approaches because they really enabled the first wave of progress to occur in the 20th century, but they also create problems as, obviously, people know. You can't, for example, make vascular grafts under six millimeters in diameter because, again, blood sees an unnatural surface. It will clot.

The last one-- I hope there are no lawyers here. But if you look at breast implants, one was a lubricant, and the other actually a mattress stuffing. I see people are thinking of the logic, but it's true. That's often how these things happen.

So what we decided to do was take a very different approach in the 1970s and that is rather than take the material off the shelf, rather to ask the question, what do you really want in a biomaterial from an engineering standpoint, from a chemistry standpoint, from a biology standpoint, and then could you synthesize it? So we picked an example and that example had to do with biodegradable polymer implants. And when we started this work, there was only one synthetic polymer that had been approved by the FDA for implantation. Those were the polyester suture materials, and they dissolved like this-- a process we call bulk erosion.

So you uniformly distribute the drug in the system, but over time, it would get spongy and could fall apart. And that's fine for some drugs, but it could lead to bursts of drug coming out, which, for example, if you had a really toxic drug like, say, insulin or an anti-cancer drug, this might not be so good. So we said, from an engineering standpoint, what you'd like to do is surface erosion. That's analogous to the way a bar soap dissolves or the way an onion peels.

And if you could do that, the thinking was you shouldn't get this dose dumping. Not only that, the design should be very easy-- if you want to double the lifetime of the device, you just double the thickness, for example. The challenge was how to do it.

And we went through a pretty detailed engineering design analysis to figure out the right bonds, the right monomers, the right chemical structures. I won't go through all of that with you but other than to show you the final answer we came up with is shown by this chemical structure. It's the only structure I'll show.

But it turns out to be a polyanhydride-- that's the bond. And we put two monomers, two units-- one which is very hydrophobic to keep water out and the other slightly less hydrophobic-- and we based this particular structure on not only chemistry but also toxicology. Michael Marletta, who was a professor at MIT at the time helped us figure out what monomers would be safe in the human body. And so this was a hypothesis, but some of our students like Harry Rosen, synthesized these polymers. And what we found this was work by Janet [? Tomada ?] was that, basically, by looking at percent degraded or released, this is 0%.

It's a basic acid, one of the units. This is 15%, 55%. The whole point is by simply adjusting the ratio of these, this millimeter thick slab will take four years to dissolve. But as you increase it, it will dissolve faster and faster with 79%. You're all gone in two weeks.

So you could simply dial in the ratio of these monomers, and it would last for whatever length of time you want. So that would give you the opportunity to design systems that could last anywhere from days to literally years. So I was going to tell you we've actually, over the years, been looking at a number of applications. I thought I'd show one of them for you.

This is a brain cancer, glioblastoma multiforme. It's uniformly fatal, and regardless of how you treat it, the mean lifespan is less than a year. So it's a terrible disease. And the drugs that you give for this disease are really bad drugs, and they cause a lot of side effects.

The major drug that's given is a drug called BCNU shown here. And normally, you give it intravenously, so it travels throughout the whole body, causing a lot of serious side effects on the liver, kidney, and spleen. So Henry Remini-- Henry was a young neurosurgeon at the time. He's now chief actually of neurosurgery at Johns Hopkins. But he, in 1985, came to see me to see whether we could try to apply some of these ideas to treating brain cancer and actually more broadly, other localized diseases, which I'll mention shortly.

But basically, the idea that we started talking about is could you develop what we'll call local therapy rather than systemic therapy? And the idea was that you could operate-- now, in this case, for neurosurgery-- operate on the patient and take the tumor out because you're going to do that anyhow. That's what the neurosurgeon would do. But before we close up the patient, he'd line the surgical cavity with a polymer containing BCNU-- that's the drug. They wanted the polymer to be degradable, so it wouldn't accumulate in the brain, to have the surface degrading properties, so it couldn't dump this toxic drug.

And from animal studies they'd done, they wanted to last for a month, and you could do that with these polymers by simply adjusting the ratios of those units. At any rate, so the polymer protects the BCNU new from degradation. So if the polymer lasts for a month, the drug will last for a month. But most importantly is this concept of local therapy-- high concentrations in the brain where you want it and lower almost nonexistent concentrations in the rest of the body where it could cause harm.

Now whenever you get an idea and you're a professor, I mean, what you have to do is raise money for it. So we would write grants to the federal government-- generally, the National Institutes of Health-- and my peers would review it. And over the years, I actually saved a lot of their grant reviews, which were not particularly good. And I thought I would show you a couple of them today. So I entitle this slide "This Approach Will Not Work Because."

In 1981, when I first wrote about this-- this is purely from a polymer standpoint-- the chemists would look at it, and they said, well, these polymers can't be synthesized. So Harry Rosen and Bob Linhardt in our lab worked out ways to synthesize them. So then two years later, the reviews came back. They said, well, okay, they can be synthesized, but these polymers will react with any drug you put in there.

So Kam Leong, another one of our postdocs at the time, worked out a way so that wouldn't happen by making salts and other things. So a couple of years later, we wrote again, and they said well, but these polymers are low molecular weight. They're fragile. They'll break up in the body-- can't be useful. So [? Javi ?] [? Dom, ?] another one of our postdocs, worked out ways figuring out the right catalysts and time and temperature conditions in the synthesis to make them very high molecular weight and strong.

Actually, all these, by the way, made for great doctoral theses and postdoctoral projects, and all these guys that did it are like full professors today-- different places. Anyhow, the next issue was well, these are brand new materials. They're certainly not going to be safe in the body, but as it turns out Cato Laurencin, who was one of my students, did a lot of studies on animals showing that, in fact, they work.

So finally, we're now in 1988, and this was not actually a review. I was running one of these conferences called the Gordon Conference, which are these conferences where you actually like to encourage a lot of discussion. So being a chemical engineer, I thought it would be fun to invite a lot of other chemical engineers, but I also thought it would be nice to invite Henry Bram who a clinician because I want to show people how people were starting to think about this young field from a medical standpoint.

So I asked Henry to give a talk, and he went over his talk where he was showing how he was treating this disease in animals. And at the end of his talk, a number of the chemical engineers raised their hand. And they said, well, Dr. [? Brown, ?] that was a very nice talk, but we've developed these mathematical models that show that the drug will not diffuse far enough to kill the remaining tumor-- you have to pardon our spelling.

But at any rate, Henry then said but I just did show that it killed the tumor. And the chemical engineer said, no, no, that's not important. They said the models show it doesn't diffuse far enough, and they kind of went back and forth about what was important. In fact, it wasn't until about five years later that Marc Saltzman developed some better models to predict this. The next issue was that even if it did diffuse far enough, it's not a very good drug that. By the way, is true.

And one of the hopes we have is that with this kind of approach, you could put much better drugs in and the Hopkins group and others are doing that now with newer drugs like Taxol and cytokines and so forth. And the next issue was manufacturing-- that this would be very difficult to manufacture. And Guilford pharmaceuticals, which licensed it from MIT, worked those things out. And finally, in 1997, the FDA actually approved this procedure. It was the first time in over 20 years a new treatment had been approved for treating brain cancer.

Now, I'd like to actually show you what this operation looks like, but if anybody's squeamish and doesn't like the sight of blood-- I'm really serious about this-- you really shouldn't look at the next two slides. But what you do is you operate-- this is going to be human patients. And here you put one little disk in the brain here. And usually, you put seven or eight and then close the tumor site.

It's very hard to get good advice whenever you give a talk. But a couple of years ago, actually, I was lecturing to a group of chemical engineers, and my wife, Laura, came to the lecture. And I asked her at the end of the talk, I said, what did you think of the talk? And she said, well, she said the talk was okay. That's actually very high praise. But she said, Bob, you had those two bloody slides on for 12 minutes, and you explained every detail of the operation.

I was kind of excited about it. It was like our polymer is going to people. And she said, you explained every detail of the operation. I don't know if you were looking, but all those poor chemical engineers were turning green and looking at the floor. So ever after that, I've done just what I did today, but I do want to tell you the sequel to this.

About two years ago, Mike Moskowitz, a friend of mine, he's a neurologist. He was running a conference called Princeton Conference. And he actually asked if I would give the dinner speech. But you see, this conference is attended pretty much by neurosurgeons and a few neurologosts. Now there's all clinicians.

So I actually gave them the same talk. I told them the same story. And at the end, a number of the neurosurgeons came up to me. And they said, you know those two bloody slides you showed? I said yes. They said you could have left those on as long as you wanted. They said that was fine. But they said, those chemical formulas--

[LAUGHTER]

That's also true, both of them. And so now when I give the talk to neurosurgeons, it's very different about what I put on.

At any rate, to show you a little bit of the clinical data, I mean, this is by no means a cure, but there is approved safety and efficacy. And what you see is if you look at the treated group at the end of a year, you see about 63% survival in the treated group, 19% in the controls. And at the end of two years, 31% in the treated group and 6% in the controls, and it does relieve suffering as well. And as I mentioned, the FDA has approved it for different uses.

Now whenever you do something like this, there's different milestones that maybe you hit-- one, we were excited when we synthesized it, another when it first went into people, finally, when the FDA approved it. But about a year and three months ago, I found something out that I guess I hadn't expected, which was that it actually got onto national television, and I'll tell you a story. This is for those of you that aren't familiar with this.

So there's this TV show I don't usually watch it, but it's called ER. Some people probably do watch it, and it turns out that the star of the show is a guy named Anthony Edwards. And he plays this clinician named Mark Green.

And he has a brain tumor. By the way, for those of you that don't watch the show, If you ever watched this movie Top Gun that stars Tom Cruise, and he's Tom Cruise's partner in that. He actually didn't do too well there either.

But so at any rate, so what happens in the show is he had a very bad brain tumor. He was going to be off the show in about two weeks. He'd go from doctor to doctor, and they all would tell him he's off the show in a couple of weeks.

And finally, he goes to one last doctor who actually turns out to be Susan Sarandon's ex-husband in real life. But anyhow, this guy when you see him-- and I'll show you this video clip in a second-- you sort of get the impression he may have gone to Harvard Medical School. You'll see what I mean. But at any rate, so I'm going to show you this video clip from ER see if this works. Here he is.

[VIDEO PLAYBACK]

- What did he say?

- I mean, his trial--

- I thought you were going to be able to sell.

- So did I. Rachel called. She wants to come out a day early for Christmas.

- Marjorie, Dr. Burke, nice to meet you.

- You too. You too.

- And you are?

- Dr. Elizabeth Courdain, Dr. Green's fiancee.

- Oh, congratulations.

- Thank you.

- Sorry to keep you waiting. I met some resistance in tumor board.

- Is that good or bad?

- The oncologists and radiation therapists are having trouble embracing the future.

- Meaning?

- Come over here. When they see a lesion like yours, they jump to the conclusion it's inoperable.

- But it's not?

- Not in my hands. There's your tumor. Now as you were doing verbal tasks, there was increased blood flow to language centers. Those light up in orange. As you can see one of the tumors adjacent to Broca's-- it hasn't invaded yet.

- That's not what the neurosurgeon in Chicago said.

- That's why you're here.

- So you'll operate?

- Yeah, we'll do an awake craniotomy. You'll be conscious and talking while I map out Broca's and resect the tumor, but the real bonus is my ability to insert high dosage chemotherapy wafers into the tumor cavity and use the malignant cells themselves to create a cancer vaccine.

- Thank god for second opinions.

- Yeah, those idiots from the tumor board are satisfied with 12 months survival, where most of the patients on my protocol have disease-free for 24 months and counting.

- So when can you fit him into the schedule?

- How about December 31? You can ring in the new year with a load off your mind-- brain surgeon humor.

- All right.

- You okay?

- Yeah. It's good news.

- It's very good news. You came to the right place, Mark

[END PLAYBACK]

BOB LANGER: So anyhow, the bad news about this is that it looks like now he wants to get off the show. It's about a year and a half later. So his tumor may be coming back.

But you see, that would actually probably mimic what may happen in real life. These treatments extend life. They relieve suffering. They probably double lifespan, but they're not a cure. And so the hope is by finding better drugs and better materials, we'll be able to do better still.

Also, one of the exciting things about this is now many groups are extending it into other areas. So there's actually already another localized spinal tumor that's been approved, another one for ovarian cancer is in clinical trials. And it's being extended to other areas as well like one of the big areas which has been very exciting is up in the whole area of heart disease. One of the most common ways of treating heart disease is to use balloon angioplasty to open up a blood vessel, but the blood vessels close off over about six months.

So people put in stents. These are little like Chinese finger puzzle things. The problem even there about 25% of the times, the blood vessels close off, and the patients die. So what they're now doing-- and Johnson & Johnson is probably a leader but about five companies have done this-- is develop, again, this whole idea of localized chemotherapy right to the blood vessel, keeping the blood vessels open with really actually 100% open after the six-month time period and greatly increased survival. And I think, again, it's only the imagination of clinicians that as people move forward to think about other applications one might use.

The next example I wanted to use is, actually, it's an interesting personal story. About seven years ago, I had a postdoc come to me, named David Edwards. Actually, Bob recommended him to see me.

And David actually had already done one postdoc and actually had never done experiments. He was a chemical engineer, extremely bright, but always had done mathematical modeling. And in fact, the very first time he came to our lab, the first paper he wrote had 300 equations in it. But then we got him interested in thinking about other delivery problems, and one of them was to the lung. And I'll tell you a little bit about that.

Right now, when people take aerosols-- and probably different people know people who've taken a metered dose inhalers or nebulizer-- it's unbelievably inefficient. You're lucky if you get 5% of the drug from the device into the lung. The reason for that is you've got these tiny aerosols, usually liquids, and they've got enormous surface area, so they aggregate. It's kind of like wet sand.

In fact, more recently, people have even come up with dry powder inhalers. But because aerosolization is so inefficient, I'd say over the last 10 years, we've seen about 40 different companies develop through mechanical engineering better aerosolizers, better devices. And the reason they do that is because if you could make them better maybe you can go from like 5% to 8% or 9%-- that's pretty much what's been done. And that would mean you'd save a lot on the drug, but also you could make the device a lot smaller.

People want to use these for insulin and other things, and you don't want to carry around a giant inhaler. But David came up with a very different way of looking at that. He said everybody else has looked at the device. Nobody's looked at the aerosol. And you might think well, how could you redesign water or other aerosols? But see, what everybody had been doing was designing these tiny aerosols but what David thought is what if we make a really big aerosols?

And first of all, why might bigger be better? Well, the idea bigger could be better because they wouldn't aggregate as much. There's a lot less surface area. So if you think about it, what basketballs aren't going to aggregate like wet sand?