Making Drugs Smarter

The world of drug delivery—at least at its cutting edges—has begun to converge with diagnostics, tissue design, and materials science in ways that promise to transform some areas of medicine. One of the most interesting developments involves "intelligent therapeutics" that detect specific substances in the patient's bloodstream and respond to them by releasing an appropriate dose of drug. One of the key figures in the world of intelligent therapeutics is Nicholas Peppas of the University of Texas at Austin. Peppas got on the phone with Pharm Exec recently to talk about emerging work in the field. What follows is an edited transcript.

Nicholas Peppas

Pharm Exec: What has changed about the work being done in drug delivery compared to 10 or 15 years ago?

Peppas: In the 1970s and 1980s, we were talking about a once-a-day tablet that could be taken in the morning and be available in a patient's body for the next 16 hours. But as we solved that problem, we realized that there were diseases where a drug didn't have to be released continuously, but only as necessary.

The classic case is type 1 diabetes, in which insulin has to be delivered only when glucose levels are high. People started talking about coming up with systems that would respond to the patient and release the drug only when the patient needed it.

What form did that take?

In diabetes, for instance, some of the early—and unsuccessful—approaches involved devices or systems that could be left in the body for a period of time. They contained glucose oxidase, the enzyme that breaks down glucose.When glucose levels were high, glucose oxidase would break it down and the system would release insulin. As I said, this hasn't worked. Other systems have been more successful.

How does one of these systems recognize an undesirable compound?

Different investigators are using different techniques. In my laboratory, we use patented methods of what we call molecular recognition, with which we are able to recognize a specific undesirable compound and only that compound.

For example, we have nanoparticles that are able to recognize cholesterol. Different nanoparticles are able to recognize glucose.

Other investigators use other systems. Some use a chemical reaction.Others use an electrochemical reaction to recognize a chemical compound and start the release process.

How does the intelligent material go from recognizing the substance to reacting to it?

Again, there are different techniques. The mechanism might be a chemical reaction that breaks down a polymer and allows the drug to come out. Or you might have a coating of polymer that contains enough of the drug for the first treatment of the patient. And later, a second recognition process will lead to a second coating being released, and so on.

It might be something simpler—for example, a swelling process that allows one layer to be released and then a second layer and then a third. In some of our new developments, we have what we call multi-depot structures. Basically, you have a big matrix, inside of which you have minute spheres, each one encapsulated to release at specific times.

It sounds like you're talking about techniques that have been used for pills to determine when they're released in the gut, and you're now taking it all the way into the bloodstream.

It is true that systems using oral delivery have had an impact in our way of thinking. But the response in the blood is significantly different than what it is in an oral delivery system. And so one has to be careful when one generalizes. And many of the nanoparticulate systems we're discussing here, although they seem to work from an engineering point of view, don't work in medical applications where there is a strong immune response to them.

Are there other approaches to intelligent delivery besides nanoparticles?

We are also developing microchips that hopefully will be miniaturized to the nanochip level. The chips will be implanted in the body and will have the ability to recognize and release at specific intervals. Some companies are working on this approach right now—MicroCHIPS and Immed, for example.

A lot of earlier work on drug design aimed to find ways to attach the drug to a particular target. Your work seems to be much more about time.

There are diseases in which there is a certain chronobiological response that needs to be taken into consideration. So as we're developing intelligent systems that respond to internal conditions, we are also looking at whether particular patients might need a drug to be released only at specific times of the day.

For example, it is well documented that heart attacks happen primarily early in the morning, and so we are interested in developing systems that can be taken by the patient the previous evening and that will be triggered and release the drug at about six o'clock in the morning.

But our work is much more general than that. Our work really has the major vision that you can trigger a process of release if you identify the undesirable compound that is responsible for a disease. For example, we're working a lot now with high blood pressure and we know that angiotensin II is present in high blood pressure situations. We're trying to see how we can use that information to detect very early changes in the blood pressure and immediately respond by providing an appropriate drug.

Some of these systems seem to blur the line between drugs and diagnostics.

True. As you know, it is impossible for a doctor to be checking an individual patient all the time, and if we had a system that could automatically check for particular functions and report to the doctor's office, that would be a much improved system.

In some conditions, the progress of the disease takes place over six months, a year, two years. Yet because of the way our insurance system is run, we don't have the ability to test the patient continuously. Multiple sclerosis, for example, is an autoimmune disease that can develop over the period of several years. One of its major exhibitions is the formation of lesions characterized by demyelinization of the nerves. Sometimes the patient will have symptoms—for example, numbness or neuropathy. But often it is not obvious to the patient that the number of lesions has been increasing.

If we could come up with nanoparticles that detect minute amounts of a chemical compound involved in the demyelinization, and if, let's say, we could read that information through a wristwatch that the patient would wear to the doctor's office, the doctor would know if there is deterioration in the patient in real time and not have to wait for an annual MRI.

When we talk about intelligent therapeutics and smart delivery, we're talking at the same time of some form of telemedicine. And that's why organizations like NASA and NIH are very interested supporting this type of work.

Are there applications of intelligent therapeutics in oral drugs?

Yes. For example, we are working on systems to recognize undesirable activities in the stomach or upper small intestine. We're interested for example in how celiac disease [a digestive disease in which the body has an autoimmune reaction to a protein found in wheat and other grains] could be treated.

Chaitan Khosla at Stanford University has identified a 33–amino acid compound that seems to be the main compound responsible for starting the autoimmune response. We are looking at ways to use that as a triggering mechanism, so as soon as the stomach of the celiac patient detects this compound, immediately something will come in and break it down.

These nice ideas, but at the same time I want to remind you, when you play with the immune system, there have to be a lot of studies before anything will be approved. We're talking about systems that probably our kids will be working with 20 years from now.

Nicholas Peppas is the Fletcher Stuckey Pratt Chair in engineering at the University of Texas at Austin. He is the author of 900 publications and 25 books and holds 20 US and international patents. Products that he has developed, patented, or commercialized include intraocular lenses for cataract patients; improved materials for cartilage replacement; biogels for epidermal release of growth factors to improve wound healing; new materials for artificial heart linings; materials for vocal cord replacement or reconstruction; and oral delivery systems of insulin to type I diabetic patients. He received the 2005 Founders Award of the Society for Biomaterials for "seminal and pioneering contributions to the field of biomaterials," and in June, he was inducted into the French Academy of Pharmacy.