co-develop a genetically engineered drug based on FGF21 for diabetes. That alliance has ended, and the molecule is now being co-developed by Ambrx and Bristol-Myers Squibb, Allen says.
While biotech scientists have dreamed for three decades of making “smart bombs” that destroy diseased cells and spare healthy tissue, it’s been easier said than done. But there have been major strides in this work the past few years, as Seattle Genetics won FDA approval in August for one such “empowered antibody” for a couple of rare lymphomas, and Genentech showed this month that its souped-up version of trastuzumab (Herceptin) was better at slowing tumor growth, and likely better at extending life, than the original. Those successes have helped energize a number of startups around the world working to link antibodies to toxins, which I listed here at Xconomy in May.
So what does Ambrx have that’s different and special? The key, Allen says, is in the company’s ability to precisely control the way it binds antibodies to the toxins. The company’s technology enables it to use an extra-long stretch of DNA, which allows it to insert a non-natural amino acid at the same spot every time in a large, Y-shaped antibody molecule, Allen says. By creating such a specific binding site on the antibody, Ambrx scientists believe they can control exactly how many toxins bind to the antibody during manufacturing, and exactly where they belong, to give the drug ideal pharmaceutical properties, Allen says.
Today’s leading technologies for linking antibodies and toxins don’t provide that same degree of specificity, Allen says. Bothell, WA-based Seattle Genetics (NASDAQ: [[ticker:SGEN]) and Waltham, MA-based ImmunoGen (NASDAQ: [[ticker:IMGN]]), the leading developers of such linker technology, use manufacturing processes that aren’t that consistent, which means that there can be variability in the number of toxins that bind to each antibody, and the place they bind, Allen says. If not enough toxins get bound to the antibody, the antibody-drug combo might not be effective. If too many get bound, or too many get bound to the wrong part of the antibody, it could cause the immune system to react negatively and neutralize the drug before it can have the desired effect on the biological target, he says.
More consistent antibody-drug binding properties should make it possible to develop drugs that are still potent against diseased cells, but also have the kind of mild side effect profiles that are essential for drugs against chronic conditions like autoimmune diseases, Allen says.
“We think there’s a lot of room for improvement in conjugation,” Allen says.
All this work is still at an early stage, so it’s way too early to start forecasting if or when any of the drug candidates will get into clinical trials. But Ambrx, an R&D operation with 57 employees, has been cash-flow positive for more than a year, thanks to a series of partnerships with Merck, Pfizer, Bristol-Myers Squibb, and Merck KGaA of Germany.
“We’ve been cash-flow positive for a number of years, which is unique for a biotech at our stage of development. That will enable us to drive our own internal pipeline of candidates that use Ambrx technology into the clinic,” Allen says.
