Thoughtleader: Martin Mattingly, Ambrx

November 1, 2006

Pharmaceutical Executive

Volume 0, Issue 0

The ability to customize small molecules-to make them better, safer, and easier to use-has long been a staple of pharmaceutical development. But until recently, scientists had few options for enhancing biologics. San Diego-based Ambrx wants to change that.

The ability to customize small molecules—to make them better, safer, and easier to use—has long been a staple of pharmaceutical development. But until recently, scientists had few options for enhancing biologics. San Diego-based Ambrx wants to change that.

Martin Mattingly has been president and CEO of Ambrx since September 2005. He joined the company from CancerVax, where he was executive vice president and chief operating officer. He has also held leadership roles at Agouron Pharmaceuticals and Pfizer, and oncology marketing positions at Eli Lilly. He holds a doctorate in pharmacy from the University of Kentucky.

The company's ReCODE (reconstituting chemically orthogonal-directed engineering) technology allows researchers to modify virtually any amino acid, even those without their own chemical activity. In the past, it was only possible to make modifications to a protein's 20 natural amino acids.

So, in essence, what Ambrx has done is create the "glue" that allows researchers to attach activity-enhancing molecules to amino acids where they couldn't before. These attachment points have the potential to increase the length of time a protein is active in the body, improve its safety profile, or make it more convenient for patients to use.

At just three years old, Ambrx has already forged its first partnership with Big Pharma: It teamed up with Roche to develop the pharmaceutical giant's next-generation proteins and peptides. (It also has several other partnerships in the works.) Ambrx's own lead protein—a weekly version of human growth hormone—is ready to enter the clinic.

Here, CEO Martin Mattingly discusses ReCODE in detail, explaining its potential to improve not only forthcoming biologics but also those that already exist. He also answers a burning question: How did Ambrx develop this technology so quickly?

PHARM EXEC: With small molecules, you have structure-based design, which allows you to customize. Is this what you're doing with biologics?

MATTINGLY: It's similar. We've used the term "protein medicinal chemistry" to describe what we do. In the past, the prevailing belief was that natural substances have a set of amino acids that can't be modified. And what we now know is that they can be, and we can improve on them. Roche has done it with a major product called Pegasys (pegylated interferon-alpha-2a). Amgen has done it with several of their proteins. Eli Lilly has done it with insulin. We think of it as applying the traditional mindset of medicinal chemistry, of making minor modifications to a small molecule to improve upon its characteristics.

But what Ambrx brings to the table for the first time is the ability to make a very precise and specific modification to a protein, because what we have found is that the site of modification matters. So if you have a protein that has 200 amino acids, we can pick any one of those amino acids and change it, and then confer new properties on the protein as a result of that change. So, in a way, we do use a structure-based design approach because there are certain places you don't want to modify the protein because you would inhibit it or you'd prevent it from being active. We know the general areas that we can modify without affecting the activity of the protein.

The beauty of our technology is that once we go to a particular region of the protein, we can make changes anywhere within that region; literally, we can make any amino acid change. We enable a level of control that's never been seen before.

Are you adding something to the amino acid through this technology?

Yes. We use an Ambrx amino acid that is specially designed to have certain properties. In the case of our first two products, it is designed to make a chemical attachment point. We've referred to it variously as a Velcro-strip concept or glue. We're creating a chemical attachment point—glue, if you will—that you can then use to put other things on the protein.

With our first two products, we're attaching PEG (polyethylene glycol) molecules, which are standard in the industry to lengthen the time-activity of proteins. But our technology allows us to attach the PEG molecule exactly where we want it—whereas other companies are using technology that limits them to the natural amino acids that are already chemically reactive, because that's the only place that the PEG molecule will go.

We can insert that amino acid wherever we like—usually in multiple places—and then we test the molecules. We can play with it, attach it in multiple places, and see which one's the best in animals before we put it into humans.

How will this technology change the way biologics are developed?

It enables companies to develop protein drugs the way they've always developed small molecules, and that is simply a capability they do not have today. And with this technology, they can make 50 different versions of a protein that are different only in a single site of substitution. And then they can test those molecules with lab assays and in animals and then pick the best molecule to go into humans. That's what everybody does with small molecules and nobody can do with proteins.

Partner of Choice

We are in active discussions with a number of the major pharmaceutical companies that are committed to proteins, and I think it's safe to say there's a tremendous amount of interest and enthusiasm in the technology.

How might ReCODE improve existing products?

Take a really exciting area right now: antibody fragments. The antibody market has really become huge, with a lot of successful cancer antibodies and arthritis antibodies that are on the market.

Researchers are now looking at using antibody fragments, which are a part of the full antibody but smaller, less expensive to manufacture, and might penetrate tissues more easily, for instance, in treating diseases like cancer.

The problem with them is they don't last very long in the body. They get chewed up very quickly, before they can really have an effect.

That's an area that we're working on. We have two programs right now with pegylated antibody fragments. Using our Ambrx glue, we can attach a molecule to the fragment that will allow it to last longer and protect it so that it has enough time to be active in the body.

Separately, we also have a lead project that is a weekly version of human growth hormone. All of the current products are daily, so we're applying our technology to growth hormone so it only has to be given on a weekly basis. We'll start clinical trials within the next six-to-nine months with that program.

At what stage of a drug's clinical life will you begin to look for partners?

Generally speaking, if you have the ability and the finances, it's better to carry a clinical asset as far as you can before you partner it. The value keeps going up. If you partner your own products very early, you simply do not get the same amount of value for it. So if you believe in your product, which we do, it's better to take on some risk and hopefully reap greater rewards.

Now, other types of partnerships happen much, much earlier. If a company is working on its own existing proteins, we can then work with them early in the development process to take their products and optimize them with our technology. And those are some deals that we're looking at right now.

How many partnerships have you done so far?

So far, Roche is our only partnership. We're a very young company. We're less than three years old at this point. We anticipate in the next six to nine months signing several additional partnerships. We're in very active discussions right now with multiple companies.

It's been a very rapid rise. One of the things that we are most proud of is that two years after we started lab work here at Ambrx, we were manufacturing our first modified protein, which is the growth hormone, at a scale that could be commercialized. And we will put our first modified protein in the clinic almost three years after we started.

In a short period of time, we did something that is very difficult to do in the pharmaceutical industry, and that is take a core technology platform that has never been used before and apply it rapidly to products.

Nobody wants to wait 10 to 15 years for you to prove that your exciting technology can actually make a better product. And so that's why we started from the get-go and focused on products and, again, we'll put one in a clinic very quickly.