When Amgen Inc.'s Blincyto (blinatumomab) gained accelerated approval from the FDA in December to treat patients with relapsed or refractory Philadelphia chromosome-negative precursor B-cell acute lymphoblastic leukemia, or B-cell ALL, the compound turned a new chapter in drug development, heralding the arrival of bispecific antibodies. (See BioWorld Today, Dec. 4, 2014.)

Technically, Blincyto wasn't the first approved bispecific. That title belongs to Removab (catumaxomab), developed by Trion Pharma GmbH and Fresenius Biotech GmbH. The molecule gained European approval in 2009 to treat malignant ascites (abdominal fluid) in cancer patients by targeting the T-cell antigen CD3 and epithelial cell adhesion molecule (EpCAM; CD326).

Removab made little headway in the market, but uptake could improve with Blincyto, the first asset to emerge from the bispecific CD19-directed CD3 T-cell engager (BiTE) antibody platform developed by Rockville, Md.-based Micromet Inc. In 2011, Amgen, of Thousand Oaks, Calif., inked a license agreement with Micromet valued at up to $976 million, including $14 million up front, to develop BiTE antibodies to three solid tumor targets. Less than a year later, the big biotech purchased Micromet outright for $1.16 billion. (See BioWorld Today, July 12, 2011, and Jan. 27, 2012.)

At the time, blinatumomab was in phase II development. The drug acts as a connector between the CD19 protein, found on the surface of most B-cell lymphoblasts, and CD3, a protein on T-cell lymphocytes, and works by engaging T cells to destroy leukemia cells. In a multicenter, single-arm, open-label phase II study of 185 adults who were treated with Blincyto for at least four weeks via infusion, 32 percent showed complete remission for approximately 6.7 months.

Blincyto is not expected to reach blockbuster status, with 2019 sales projected at just $354 million, according to Cortellis Competitive Intelligence (CI). But it's a foot in the door, and biopharmas are salivating over the chance not just to follow in the footsteps of Amgen and Micromet but to surpass the clinical and commercial success of Blincyto.

"Obviously, everyone's excited about the approval," said Sean McCarthy, CEO of Cytomx Therapeutics Inc., of South San Francisco, whose "masked" antibodies, known as Probodies, have drawn deals with New York-based pharmas Pfizer Inc. and Bristol-Myers Squibb Co. "It's energized the field." (See BioWorld Today, June 7, 2013, and May 28, 2014.)

LEVERAGING BIOLOGY 'TO MAKE ONE PLUS ONE EQUAL FIVE'

The field is focused on taking the learnings from that approved drug and focusing on next-generation approaches, according to McCarthy. "I think they will extend to full-length bispecific formats in the full IgG context, where one would expect improvements in pharmacokinetics," he said.

Cytomx's Probodies are fully recombinant antibody prodrugs. When administered, they are distributed to normal and tumor tissue but become activated only in the tumor through tumor-associated proteases. The company already is building the next generation of its technology, which it calls T-cell engaging multivalent Probodies, or TMVPs.

"We believe the migration of T-cell bispecifics to the solid tumor environment is going to be quite difficult in creating a therapeutic index for conventional T-cell engaging bispecifics," McCarthy told BioWorld Insight. "The molecules are so potent that they will be, we think, quite unforgiving of very low levels of antigen expression on normal tissue. Our program is designed to mask the antigen and CD3-binding moieties in the bispecific, so the bispecific will be unable to bind and recruit T cells into normal tissue. We think that will open up the solid tumor landscape."

At Xencor Inc., of Monrovia, Calif., the focus of its Xmab technology is to engineer small changes in the Fc domain to improve binding and create a truly modular platform, according to Bassil Dahiyat, the company's co-founder, president and CEO.

"If I'm going to make an Fc domain that is more cytotoxic – and we've got a bunch of these in the clinic with partners – I want it to be the same Fc domain I can plug and play in any antibody," he explained. "I can basically replace the natural Fc domain with our domain. It's just a couple of amino acid mutations; it's not a big change. Nature cooperates, because the Fc domain is constant, so it should be usable as a scaffold."

Other biotechs are using similar techniques with their bispecifics, he said, citing Regeneron Pharmaceuticals Inc., Genmab A/S and Zymeworks Inc. as examples.

Genmab, of Copenhagen, Denmark, has a combo EGFR tyrosine kinase receptor inhibitor/hepatocyte growth factor receptor modulator, known as EM1-mAB, in discovery with partner Janssen Research & Development LLC as part of their Duobody deal, inked in 2012. (See BioWorld Today, July 16, 2012.)

Zymeworks has ZW-34, a B-lymphocyte antigen CD19 modulator, in discovery, as part of its 2011 agreement with Merck & Co. Inc., of Whitehouse Station, N.J. In recent months, Zymeworks attracted two more partners to its Azymetric platform for creating bispecific antibodies. In October 2014, Eli Lilly and Co., of Indianapolis, broadened an earlier deal by adding new milestone payments and an equity stake of about $27 million in Zymeworks. Last month, Celgene Corp., of Summit, N.J., took an option to advance an undisclosed number of bispecific candidates through clinical development and commercialization in return for an up-front fee, $164 million in milestones per candidate and an equity stake of roughly $8.2 million. (See BioWorld Today, Aug. 30, 2011, Oct. 24, 2014, and Jan. 22, 2015.)

Ali Tehrani, president and CEO of Zymeworks, recalled the early days of antibody development, when prescient researchers at companies like Amgen and Genentech Inc. were already thinking about how to use antibodies to target two epitopes rather than on a one-to-one ratio.

"They believed if you could take advantage of the synergies of biologies and combine biologies, you could make one plus one equal five," Tehrani told BioWorld Insight. "That's where the field started."

'A LOT OF STUFF WORKS AT THE BENCH LEVEL'

Bispecifics weren't exactly overnight sensations. Medarex Inc. – which spun out Genmab before becoming part of New York-based Bristol-Myers Squibb Co. – was working on a bispecific antibody that recognized the Fc receptor and triggered an attack on the AIDS virus in 1991. Two years later, the company started its first trial with a so-called "second-generation cancer therapeutic" designed to kill cancer cells by anchoring them to white blood cells such as monocytes, macrophages and neutrophils through the immune cells' Fc receptors. (See BioWorld Today, Nov. 1, 1991, and May 24, 1993.)

Forerunners in the field "showed that it was possible" to build bispecifics "and to take advantage of the biologies of two different targets to go above and beyond the basics in defeating a disease," Tehrani said.

The bigger question was whether such research could move outside the lab, which was dependent on the ability to manufacture early candidates at the right purity, the right stabilities and the right yields required for the clinic.

"A lot of stuff works at the bench level, but when you're talking about going through FDA approvals and GMP, you really have to nail it perfectly," Tehrani said. "This is where the field stopped, until Micromet entered and tried to look at the problem a different way."

Those early efforts may seem primitive by today's standards, but the field would not have advanced without them, Dahiyat agreed. Bispecifics emerged from "years and years of lots of tinkering that, frankly, didn't really get anywhere for a long time," he observed, crediting the "heroic effort" behind blinatumomab, which he said targets "an extremely promising biology." Alongside the Amgen/Micromet alliance, others were attempting to create bispecifics that involved alternative scaffolds, different variable domains and different kinds of binding molecules.

But each of those platforms, which were reengineered as antibody derivatives rather than built from the ground up, faced the same problem.

"Antibodies are really good at their job because nature engineered them to have long half-life and durability and to be produced in massive quantities in the body," Dahiyat pointed out. Although the biotech industry quickly learned how to replicate those attributes in monoclonal antibodies, achieving the same feat in bispecifics was not so easy.

"It's only been in the last two or three years that people have been able to bring antibody-like features to bispecific antibodies," he said, citing the need to incorporate the natural symmetry of monospecific antibodies while retaining the potency of the T-cell attack. Xencor has focused its platform on engineering a series of Fc domain variants that spontaneously form stable, bispecific antibodies that can be manufactured and purified with standard antibody production methods and used to generate a range of drug candidates.

"Now, we're looking at how to go beyond bispecific into multispecific antibodies to access new biology," Dahiyat said.

'THERE'S A GORGEOUS PROLIFERATION OF TECHNOLOGIES'

The industry is rife with other approaches. Abbvie Inc., of North Chicago, has three bispecifics in development, including ABT-122, a dual anti-tumor necrosis factor (TNF)/IL-17 ligand inhibitor that has one phase II under way and a second planned, according to Cortellis Clinical Trials Intelligence. A second TNF ligand inhibitor and a dual IL-12/IL-18 ligand inhibitor remain in discovery.

Ablynx NV, of Ghent, Belgium, is advancing two bispecific candidates from its nanobody platform: the von Willebrand factor ligand inhibitor caplacizumab (subcutaneous) to treat thrombocytopenic purpura, and ozoralizumab, a TNF alpha ligand inhibitor, targeting inflammatory disease. Both have advanced to phase II. (See BioWorld Today, June 18, 2014.)

Medimmune, of Gaithersburg, Md., has an anti-Psl/PcrV multi-specific monoclonal antibody in discovery to treat Pseudomonas aeruginosa infection. The global biologics research and development arm of Astrazeneca plc also has MEDI-3902, a PcrV protein type III modulator, in phase I development in the same indication.

Paris-based Sanofi SA has moved SAR156597, a bispecific antibody targeting interleukin-4 and interleukin-13, into phase II development in idiopathic pulmonary fibrosis.

And Roche AG unit Genentech has a bispecific molecule in discovery to treat wet age-related macular degeneration and duligotuzumab, an EGF antagonist and Erbb3 tyrosine kinase receptor inhibitor, in phase I development to treat solid tumors.

Enthusiasm about the scientific underpinnings and potential of bispecifics – or multivalent antibodies, as some prefer to call them – has fueled dealmaking in the space. And nearly everyone expects to see more.

"There's a gorgeous proliferation of technologies now, fueled by the general health of our industry," said Laura Vitez, principal business analyst for Thomson Reuters Recap. The flow of capital "allows increased investment in innovation and increased alternatives to move forward the results of that innovation," whether through M&A deals, initial public offerings, "interesting and mutually fulfilling" licensing deals or continued venture funding, she added.

'PHARMA WILL CONTINUE TO MAKE DEALS IN THIS SPACE'

An exhaustive look at the bispecific space is nearly impossible, but a sampling of deals suggests the vigor of the technology.

In June 2013, Immunocore Ltd., of Oxford, UK, inked a multitarget deal with Genentech to generate bispecific constructs called Immtacs, or Immune mobilizing mTCR against cancer, designed to exploit the power of T-cells to recognize peptide antigens that are produced by tumors. The deal was valued at a potential $320 million or more per program, plus royalties, with Genentech handing over two targets at the outset. (See BioWorld Today, June 27, 2013.)

Days later, Immunocore inked a potential $525 million discovery agreement with Glaxosmithkline plc, of London, to discover bispecific Immtacs against multiple targets that could not be addressed with antibody-based technologies. (See BioWorld Today, July 9, 2013.)

In December 2013, Oncomed Pharmaceuticals Inc., of Redwood City, Calif., signed a potential $3 billion deal with Celgene that included $177.25 million up front, $22.25 million as an equity investment and options on up to six oncology candidates. The lead candidate in that deal was demcizumab, which targets Delta-like ligand 4 and was undergoing phase Ib testing in non-small-cell lung cancer and pancreatic cancer. But payments for an undisclosed asset in the lineup, an anti-DLL4/VEGF bispecific antibody, could amount to as much $505 million. (See BioWorld Today, Dec. 4, 2013.)

Hedging its bets, Celgene inked two additional bispecific deals with Sutro Biopharma Inc., of South San Francisco. A potential $500 million pact in 2012 led to a broader collaboration last year – and one with a buyout option. The specifics of those antibodies remain under wraps, but both are targeting cancer. (See BioWorld Today, Dec. 19, 2012, and Oct. 24, 2014.)

Prowess in bispecific antibody development drove Johnson & Johnson's $218 million acquisition of Covagen AG, of Schlieren, Switzerland. The small biotech, which is developing engineered binding proteins called Fynomers derived from the Src homology 3 domain of Fyn tyrosine kinase, already had a lead asset in a phase II program with three others in discovery, according to Cortellis CI. (See BioWorld Today, Aug. 27, 2014.)

Late last year, Janssen Biotech Inc. inked a potential $675 million deal for MGD-011, from Macrogenics Inc., which targets CD19 and CD3 as a potential treatment for B-cell malignancies. (See BioWorld Today, Dec. 23, 2014.)

McCarthy also pointed to Roche's potential $488.75 million acquisition of Dutalys GmbH, of Vienna, which specializes in the discovery and development of bispecific antibodies based on its Dutamab technology. "Roche obviously has incredibly deep experience in biologics, including experience in bispecific of their own, but they still saw value in going after the market and acquiring a novel platform with its own twist on how to create and optimize bispecific formats," he said. "That's one of the very strong signals that pharma will continue to make deals in this space."

As Recap's Vitez observed, even small companies with bispecific assets are attracting funding to advance their internal pipelines. In February 2014, Geneva-based Novimmune SA closed a series B financing round for CHF60 million (US$66 million) to advance multiple candidates, including two assets from its kappa/lambda body bispecific drug platform, NI-1701 and NI-1801, that may selectively inhibit CD47 on cancer cells.

Other names in the space include Adimab LLC – which inked yet another bispecific deal, with Sanofi, just last week – F-star Biotechnology Ltd., Innovent Biologics Inc., Mabvax Therapeutics Inc., Merus BV, Sorrento Therapeutics Inc., Swedish Orphan Biovitrum AB, Tesaro Inc. and Triphase Accelerator Corp.

In short, the roster of bispecific antibody developers reads like a Who's Who of the biopharma industry. All told, Cortellis CI identified 83 bispecific antibodies in development – three-fourths still in discovery and most targeting oncology indications – with almost as many companies involved as sponsors or partners and using a variety of target-based actions. The chart below shows the 20 most common mechanisms.

'WE'RE GOING TO SEE A FLOOD OF BISPECIFIC MOLECULES'

Moving forward, it's clear that "the drug development world wants bispecific antibodies that can be readily manufactured and, very importantly, have the same manufacturing protocols as their monovalent antibodies because that affects their cost of goods," Zymeworks' Tehrani observed. "They don't want to change processes. They don't want to change their purification protocols. They don't want to change their cell line development protocols. They don't want new vectors."

A number of companies now have the technology to address those challenges while still providing the flexibility that pharma also seeks, he added.

"This is the future," Tehrani said. "Whether it's our bispecific or someone else's bispecific, the one sentiment we hear is that the industry needs to combine biologies – not in a cocktail format but in a single, targeted drug format. It is possible now, and I believe drug design in the next five to 10 years is going to be dominated by bispecifics."

The big question for the field now is which technologies, and companies, will survive the inevitable winnowing as competing assets seek to prove their mettle in the clinic, which could be a daunting task.

In light of Blincyto's approval, "it's a little easier to see – maybe a lot easier to see – the path to market for bispecifics" compared to related technologies such as CAR-T cells, McCarthy observed. "But the technical challenges we've been grappling with remain. What is the right relative affinity of the two binding arms of a bispecific? What is the right format, in terms of the presentation of those two antigen-binding arms? How do those formats relate to manufacturability and overall CMC considerations? And what are the right pairings or targets to really drive efficacy?"

The field has learned a great deal, and the tipping point was Blincyto's approval, he added, but that won't remove all of the twists and turns in the road ahead.

"It's noteworthy that most if not all of the late-stage clinical work so far is in hematologic malignancy," McCarthy said. "How do we extend the technology to solid tumors? That's where the field is wrestling with the challenge of target selection and target distribution. The doses of a T-cell engaging bispecific are so much lower than that of a conventional therapeutic antibody and the potency of these molecules is so much higher – given the cell-based efficacy and the amplification that one obtains by recruiting professional killers to the tumor – that to have a therapeutic index, targets will have to be carefully addressed."

But the field is moving rapidly to overcome those challenges, Xencor's Dahiyat maintained.

"We're going to see a flood of bispecific molecules enter and advance through the clinic in the next two to three years," he predicted. "We're really at the early days of the platforms that are going to change the landscape."