If you have an ACS member number, please enter it here so we can link this account to your membership. (optional)

ACS values your privacy. By submitting your information, you are gaining access to C&EN and subscribing to our weekly newsletter. We use the information you provide to make your reading experience better, and we will never sell your data to third party members.



Bispecific Antibodies Deliver Two For The Price Of One

With new biotech drugs, scientists try to pack more punch into a single molecule

by Lisa M. Jarvis
December 17, 2012 | A version of this story appeared in Volume 90, Issue 51

Credit: Merrimack
Merrimack scientists have developed a pipeline of bispecific antibodies.
Scientists in Merrimack’s labs are working on bispecific antibodies.
Credit: Merrimack
Merrimack scientists have developed a pipeline of bispecific antibodies.

Short of a course of antibiotics for a simple infection, few diseases these days are treated with a single therapy. Whether doctors are trying to tackle something as serious as cancer or as common as allergies, they are likely to prescribe a cocktail of drugs to effectively quiet a patient’s disease.

But cocktails don’t make sense for some kinds of drugs, including monoclonal anti­bodies. These protein-based therapies, typically manufactured in mammalian cells and administered by injection, have become critical tools in doctors’ medicine bags, making up six of the top 10 selling drugs this year, according to the health care consultancy EvaluatePharma. They are known to hit specific therapeutic targets, particularly ones related to cancer, with high precision.

Yet monoclonal antibodies are expensive on their own and even more so when a combination is called for. Moreover, different types of antibodies do not always work well in concert. Given these drawbacks, drug researchers have turned to new strategies for amping up anti­bodies’ therapeutic impact. One approach gaining traction in big and small drug companies alike is the bispecific antibody: one molecule that binds to two targets.

“The monospecific antibody therapeutic arena has matured over these past few decades, and people have now realized in the clinic that even though these targeted-molecule drugs are quite effective, there is room for improvement,” says Germaine Fuh, a scientist with Genentech. “There are diseases where just one target may not be sufficient.”

Consequently, companies boasting technology for making bispecific antibodies are proliferating. In recent years, dozens of players have emerged, each offering its own twist on how to engineer two functions into one molecule. Meanwhile, the most advanced bispecifics are starting to trickle through the pipeline, providing the first clues about the safety and effectiveness of the approach.

Interest in bispecifics is underscored by the sizable investments many big drug companies are making in the technology. Among the most notable deals is Amgen’s $1.2 billion purchase early this year of Micromet, a company with a bispecific drug, blinatumomab, in Phase II studies.

Eli Lilly & Co. has launched an aggressive program to develop internal capabilities for multispecific therapies. Since 2011 it has hired some 40 scientists dedicated to those efforts across its labs in San Diego and Indianapolis, says Tom Bumol, Lilly’s vice president of biotechnology discovery research.


Drug companies are angling for access to bispecific antibody technology.

June 2011 Eli Lilly & Co. unveils a multi-million-dollar investment, including hiring 40 scientists, to develop multispecific therapeutics.

August 2011 Merck & Co. inks bispecific antibody deal with Zymeworks, which could receive up to $187 million in research support and milestones.

January 2012 Amgen pays $1.2 billion for Micromet, gaining a bispecific antibody for leukemia in Phase II studies.

April 2012 Adimab expands relationship with an undisclosed partner to include a project in bispecific antibodies.

June 2012 Novartis taps Genmab to develop bispecific antibodies against various target combinations.

July 2012 Janssen and Genmab team up in a bispecific antibody discovery pact. Genmab scores $3.5 million to generate panels of antibodies across several target combinations.

September 2012 Servier pays MacroGenics $20 million for an option to license three bispecific antibodies in preclinical studies.

October 2012 Mitsubishi Tanabe Pharma pays $5 million for access to Covagen’s bispecific antibody technology. Mitsubishi agrees to cover research expenses and could pay another $140 million in milestone payments.

November 2012 Innovent Biologics, a Chinese bispecific antibody developer, raises $25 million from the Asian arms of Fidelity and Lilly’s venture capital group.

SOURCE: Companies

Companies point to several reasons for wanting to combine two functions in a single molecule. At the most basic level, the approach could sidestep the clinical challenges and high price of giving patients antibody-drug combinations.

“Even with generics, it would be very expensive to prescribe two antibodies,” notes Julian Bertschinger, chief executive officer of Zurich-based Covagen. Figuring out the correct dose of each drug would require a complex study with multiple arms—an expensive and timely proposition for the drug developer. And at the end of the day, patients would prefer to have one injection instead of two.

Bispecifics can also, in principle, offer a therapeutic advantage over combinations of individual drugs. By coaxing a molecule to hit two targets, “you get something more active than a combination,” says Genentech’s Fuh, who invented the company’s “two in one” platform for bispecific antibodies.

Genentech’s dual-action antibody against HER3 and EGFR, two receptors in a signaling pathway often implicated in cancer, has multiple ways to hang on to the cell surface, Fuh says, translating into a tight bond with its target receptors. Studies in cancer cells suggest that the drug candidate is more potent than a cocktail of different antibodies that inhibit HER3 and EGFR.

In some cases, a bispecific antibody enables a therapeutic response that other­wise couldn’t be achieved at all. “Sometimes, you cannot get the preferred biology by hitting two targets separately,” notes Tillman U. Gerngross, CEO of Adimab, a Lebanon, N.H.-based biotech firm with bispecific antibody technology.

Amgen’s lead bispecific, blinatumomab, for example, engages two receptors: CD3, which activates disease-fighting T cells, and CD19, a target found only on the surface of B cell-derived cancers such as leukemia. The rationale is that by recruiting T cells near the tumor site, the immune system can be better put to work to shut down the disease.

Merrimack Pharmaceuticals is also pairing functions in one molecule for a therapeutic effect that wouldn’t be possible with separate antibodies. Merrimack’s most advanced bispecific antibody, MM-111, homes in on HER2, the target of the breast cancer drug Herceptin, and the receptor ErbB3. Although Herceptin works well against breast cancers driven by an amplification of HER2, cancer cells will often find a way to survive.

Ulrik B. Nielsen, Merrimack’s chief scientific officer, explains that overexpression of HER2 causes the protein to dimerize with ErbB3, turning on a signaling pathway that allows tumors to proliferate. But blocking ErbB3 on its own is ineffective. “An ErbB3 monoclonal antibody is not very potent in HER2-overexpressing cells,” Nielsen notes. “You need a bispecific.” MM-111 first latches on to HER2 then forms a trimeric complex with ErbB3, in theory shutting down the resistance mechanism.

Another reason for the bispecific approach is to generate a new mode of action by binding to two parts of the same receptor on a cell. One example, from the early-stage pipeline of the Swiss biotech firm Covagen, latches on to two different sites on HER2.

Covagen develops its drugs by fusing pairs of small binding proteins it calls Fynomers to multiple sites on an antibody and then screening the resulting bispecific architectures in search of the most active ones. With this approach, the firm arrived at a molecule with a mode of action different from that of existing HER2 inhibitors such as Herceptin. “We found that only a very specific orientation—one specific location on the antibody where we’ve fused our Fynomer—achieves that effect,” notes Dragan Grabulovski, Covagen’s chief scientific officer.

The therapeutic rationale for bispecifics is clear, designing an effective one presents a series of challenges. The first hurdle, not surprisingly, is choosing the two functions to put in the molecule. Often companies are tackling targets that have already been broached with limited effectiveness with single monoclonal antibodies, Genentech’s Fuh notes.

The next challenge is designing a molecule with the right architecture. In order to maintain the antibody’s overall activity, each functional component needs to be in the right position to interact strongly with a targeted cell’s binding site, or epitope. Controlling the placement of the components is also critical to keeping antibodies from glomming on to one another. “We think the geometry is going to require optimization for every epitope,” says Ho Cho, chief technology officer at Ambrx. “There isn’t a whole lot of literature here, and it’s going to be quite empirical.”

Although companies are making headway in overcoming the architectural challenges to designing bispecific antibodies, biotech executives say some of the most advanced approaches might not be flexible enough to work in many therapeutic settings. “The design element for bispecifics is really tough,” says Trevor Hallam, chief scientific officer of Sutro Biopharma. “Most of the formats out there are not efficient and they’re not optimized, but they’re good enough.”

Once scientists have struck upon the right spatial configuration, scaling up the manufacture of a properly folded molecule is also difficult. “It’s easy to draw two molecules on a piece of paper and expect them to assemble properly,” Adimab’s Gerngross notes. “But they tend to aggregate, to precipitate, not express well, and fall apart for various reasons. I’d argue that the ability to manufacture has been one of the central issues that has bogged this field down.”

Stability is a particularly tough issue, as many of the early technologies for assembling bispecifics rely on fusion proteins, molecules made by connecting two genes to generate a single protein chain.

Amgen’s blinatumomab highlights the limitations of the fusion protein approach. Industry executives note that yields of blinatumomab are low compared with traditional monoclonal antibodies, and its half-life in the body is short. Despite its short half-life, the drug’s potency has kept it alive. Still, extreme measures must be taken to elicit a therapeutic effect: Patients must endure a continuous infusion of the drug over the course of several weeks. Amgen declined to comment.

As biotech companies work their way through the developmental hurdles, the handful of bispecifics that have made it into clinical trials are being closely watched. Data are starting to roll out for the most advanced drug candidates, including Amgen’s blinatumomab, Genentech’s dual HER3/EGFR inhibitor, and Merrimack’s MM-111. They provide the first clues about the therapeutic potential of the approach.

Last week,Amgen reported impressive results from a Phase II study of blinatumomab to treat an aggressive blood cancer called acute lymphoblastic leukemia. Of the patients given the bispecific antibody, 69% saw their disease go into complete remission.

Leerink Swann stock analyst Howard Liang surveyed three leading hematologists who said they are encouraged by the high response rates seen with blinatumomab, but they pointed out that it is less practical than competing drugs in development because it is administered by continuous infusion.

Earlier this year, Genentech moved its HER3/EGFR blocker, also known as RG7597, into Phase II studies as a treatment for head and neck and colorectal cancer. Merrimack, meanwhile, just began recruiting patients for a Phase II study of MM-111 to treat gastric cancer. The company also has launched a Phase I study of its second bispecific antibody to enter the clinic, MM-141, in tests to treat solid tumors.

Other companies, including Lilly, hope to start their first Phase I studies in 2013. Ambrx is on the brink of nominating candidates for human studies and expects by mid-2013 to begin clinical trials of its multispecific antibodies, either on its own or with a partner. Covagen expects to push its lead bispecific antibody into Phase I studies in psoriasis in early 2014.

With the bispecific antibody pipeline starting to fill, industry watchers expect to see a corporate shakeout. The field will need several approaches to assembling bispecific molecules, and dozens of formats have emerged. The expectation is that some technologies will fade away as results from preclinical studies are announced. “Some of the constructs that some very clever engineering scientists have generated have, in our view, no chance of being commercial,” Lilly’s Bumol says.

After years of work in the lab, scientists are now eager to see data from clinical trials. “Research in humans is so critical to understanding the performance of these new ideas,” Bumol says. “In 2013 and beyond, we’re going to learn a lot about how good our strategies were.”


This article has been sent to the following recipient:

Chemistry matters. Join us to get the news you need.