Patricia F. Fitzpatrick Dimond Ph.D. Technical Editor of Clinical OMICs President of BioInsight Communications

Genentech’s Knobs-into-Holes production method proves fruitful.

Monoclonal antibodies are now in wide use and recognized as therapeutically and commercially successful molecules for the treatment of cancer, inflammatory and infectious diseases, and other disorders.

But some scientists argue that despite the clinical success of antibodies such as rituximab, trastuzumab, and cetuximab, all directed at single targets, room for improvement in functionality remains. Clinical studies have shown that many patients don’t adequately respond to monospecific therapy and tumor resistance often develops. Limitations of monoclonals have spurred antibody engineers to improve and prolong the efficacy of these molecules and enhance their function by developing antibodies with two or more targets.

By simultaneous recognition of two different targets, bispecific antibodies (bsAbs) and/or “dual specificity” antibodies can serve as mediators to redirect immune effector cells, such as Natural Killer cells and T cells to tumor cells to enhance tumor destruction. In addition, by targeting two different receptors in combination on the same cell, these antibodies can induce modifications of cell signaling, including the inactivation of proliferation or inflammatory pathways.

The therapeutic potential of these antibodies has galvanized major biotech company research efforts, with more than 35 formats of such molecules in early and late clinical development. The cumulative value of six bispecific antibody deals over last two years reached about $3.5 billion.

Scientists at Genentech have developed an approach to production challenges posed by making an antibody with two or more specificities that are non-immunogenic and that retain Fc region functionality. Christoph Spiess, Ph.D., and colleagues devised a production strategy that relies on the co-culture of two bacterial strains, each expressing a half-antibody and each with its natural light chain. The half antibodies are then assembled into functional bispecific antibodies. The researchers demonstrated that these half-antibodies could self-assemble after co-fermentation and co-purification, further streamlining production. 


The novel method is based on knobs-into-holes’ technology, originally developed at Genentech, that formed the basis of the efficient heterodimerization of the antibody heavy chains. But due to the way these antibodies assemble during production, addition of a second antibody had been limited to pairing antibodies that use a so called common light chain.

“The idea of bispecifics is not a new idea and was thought of around the same time as regular monoclonals for cancer treatment, basically to recruit natural killer or cytotoxic T cells to tumor cells,” reports Dr. Spiess.” But back in the 80s no good technologies for making bispecifics existed. These early bispecific antibodies were made by co-transfection of two different light and heavy chains, resulting in nine mismatched pairs when the antibodies were expressed. This required that the correct antibody with the desired heavy and light chains, and the correct heterodimizeration, needed to be purified out, a process that was not too productive.”

The half-antibody bacterial co-culture approach was developed by Dr. Spiess and his team. He says it offers many advantages for producing bispecific antibodies in that it is basically a “simple” technique and can be used with virtually any two existing antibodies. It also eliminates the need for a common light chain and requires no reengineering of antibody functional domains. The resulting bispecific antibodies maintain the native architecture of a typical antibody and, therefore, probably retain the favorable pharmacokinetic properties and specificity of conventional antibodies.

Dr. Spiess and colleagues have produced 28 unique bispecific antibodies using this method, including an antibody against the receptor tyrosine kinases MET and EGFR that binds both targets monovalently, inhibits their signaling, and suppresses MET- and EGFR-driven cell and tumor growth.

Amgen’s Efforts

Amgen is preparing to present new clinical data on its bispecific T-cell engager (BiTE) antibody, blinatumomab, directed against both the CD19 antigen on tumor cells and the T-cell CD3 antigen on T cells. The company will report its findings at the upcoming American Society of Hematology (ASH) meeting in San Francisco. Blinatumomab became Amgen’s property after its 2012 acquisition of Micromet.

Blinatumomab, the most clinically advanced of BiTE antibody constructs, is a single-chain protein that consists of the antigen-binding domains of two different antibodies joined by a nonimmunogenic linker. The linker allows for the rotational flexibility to bind two different antigen epitopes on separate cells in close proximity. Blinatumomab contains binding regions for the B cell lineage-specific antigen CD19 as well as the invariant CD3ε subunit of the T-cell receptor present on all T lymphocytes.

According to David Reese, vp of translational research at Amgen, a unique linker protein between the variable domains of two monoclonal antibodies plays a key role in the construction of BiTe antibodies.

“We have found that the size and shape of the linker protein between the variable domains critically determine the close contact of the T cell and the antibody,” he explains. “How these are brought together seems to affect how the T cells are activated.”

Clinical trial updates at ASH on blinatumomab will include two analyses from a pivotal Phase II trial in patients with acute lymphocytic leukemia (ALL); results from a confirmatory single-arm, Phase II study (study 211) in patients with minimal residual disease positive ALL; and long-term follow-up data from an exploratory Phase II study in patients with relapsed/refractory B-precursor ALL.

According to Reese, the company will be providing Phase II clinical trial data, including pivotal Phase II data, which have been filed with the FDA. In one study, results of which were reported at the 2014 ASCO meeting, 43% of 189 patients (82) with relapsed or refractory Philadelphia chromosome negative ALL achieved a complete response or a complete response with partial hematological recovery.”

“We were also able to show that among complete remission patients 40% went on to receive a stem cell transplant, the only curative option for these patients, continues Reese.” One of the key goals of this BiTE study is to serve as a bridge to allogeneic transplants.”

Additionally, he says, in the 211 study, 73 patients were evaluable for minimal residual disease. By using molecular detection, 82% had no residual disease detected in bone marrow. MRD has been, he noted, correlated with an improved prognosis in patients with relapsed or refractory ALL.

In October, Amgen announced that the FDA has accepted for review its Biologics License Application (BLA) for the blinatumomab antibody construct. The BLA is for the treatment of adults with Philadelphia-negative (Ph-) relapsed/refractory B-precursor ALL. As part of its acceptance, the FDA granted blinatumomab priority review with a PDUFA date of May 19, 2015.

Meanwhile, Roche says it has designed a new format for bispecific antibodies called CrossMAbs. In contrast to other technologies CrossMAb allows production and correct chain assembly using a standard process applied for the manufacture of therapeutic antibodies. Christian Klein, Ph.D., one of the CrossMAb inventors at Roche pRED, notes that “The correct pairing of heavy chains from different antibodies was already achieved by a technology invented at Genentech back in 1996, with the `knob and hole’ built into the two heavy chains of different antibodies promoting correct pairing.”

But, he continued, while advances in achieving correct heavy chain association have been made over the recent years, progress in enforcing correct light chain association has remained challenging.

“The bio-engineers solved it like LEGO® players. By exchanging the molecular bricks between heavy and light chain, only specific interaction can take place in the short arm of the Y-shaped antibody,” adds Dr. Klein.

In early November OncoMed Pharmaceuticals, a clinical-stage company developing therapeutics that target cancer stem cells (CSCs) or tumor-initiating cells, announced it had been granted a U.S. patent (No. 8,858,941) for OncoMed’s anti-DLL4/anti-VEGF bispecific antibody (OMP-305B83) and use of the bispecific antibody in the treatment of cancer.

“OncoMed’s anti-DLL4/ant-VEGF bispecific antibody is one of many exciting programs in our preclinical development portfolio,” reports Paul J. Hastings, company chairman and CEO. “The activity of anti-DLL4, with its robust anti-cancer stem cell and dysangiogenic properties, is expected to complement the anti-angiogenic activity of anti-VEGF to create a potent antitumor combination.”

The company says it plans to initiate clinical trials of the bispecific antibody in late December or early 2015.

A key driver, analysts say, for interest in the potential success of bispecifics was the European Union’s 2009 approval of catumaxomab (Removab), a rat/murine antibody for the treatment of malignant ascites in patients with EpCAM positive carcinomas in cases where standard therapy is either unavailable or no longer feasible. Catumaxomab has been designed to attach to two antigens, EpCAM, found in high levels on some types of cancer cells, and CD3 on T cells. This brings the cells close together so that the T cells can kill the cancer cells. Catumaxomab also attaches to a third target, the Fc-gamma receptor, which helps the body’s immune system to target the cancer cells.

In a June 2014 report Marketwatch estimated that by 2023, the market for bispecifics will approach $4.4 billion. And, the report further notes, ongoing “evolution of technology platforms and increased venture capital interest will act as strong enablers to support the growth of this market.” In particular, report authors said that owing to the complexity of the structure of specific antibodies, “there are abundant opportunities for contract manufacturers in this area.”

Patricia Fitzpatrick Dimond, Ph.D. ([email protected]), is technical editor at Genetic Engineering & Biotechnology News.

Previous articleWatson’s Nobel Prize Medal to Be Auctioned
Next articleIllumina, Sequenom Settle NIPT Patent Infringement Dispute