Understanding the Bulk Production of Broadly Neutralizing Antibodies

How will researchers manufacture broadly neutralizing antibodies for HIV prevention and treatment studies?

By Michael Dumiak

The isolation of a new crop of antibodies that can inactivate a wide variety of HIV strains remains one of the most important recent discoveries in HIV vaccine research. Antibodies are proteins made by the immune system in response to pathogens such as viruses. So-called broadly neutralizing antibodies (bNAbs), dozens of which have now been isolated, can inactivate a much broader range of HIV strains than any of the HIV-specific antibodies isolated before. Ideally, vaccine researchers will identify an immunogen or set of immunogens—the vaccine components that induce an immune response—which can spark production of these bNAbs in uninfected people.

Meanwhile researchers are using these antibodies in other ways, including what is referred to as passive administration or passive immunization studies (see VAX May 2010 Primer on Understanding if Broadly Neutralizing Antibodies are the Answer). In passive immunization studies, researchers directly inject one or more of the recently isolated bNAbs into uninfected people to see if they are able to protect against natural exposure to HIV. Preliminary safety studies of the passive immunization approach are already underway with one of the recently identified antibodies, known as VRC01, which was discovered by researchers at the National Institute of Allergy and Infectious Diseases (NIAID). Two trials with VRC01 are ongoing in groups of 15 to 25 people—one is a group of healthy volunteers and the other is a group of HIV-infected volunteers. Results from the safety trials are expected in as little as three months, and scientists are already discussing what the next steps could be.

Based on recent non-human primate studies, there is also the potential for using these bNAbs for HIV treatment or even research toward an HIV cure (see VAX March 2014 Primer on Understanding the Expanding Role for Broadly Neutralizing Antibodies).

Taken together, all of these approaches prompt a more basic question: how will researchers produce these antibodies in bulk?

Machinery for manufacture

Isolating bNAbs is a painstaking procedure, involving finding an HIV-infected volunteer, extracting their immune cells, isolating the genetic material from those cells, screening the genetic material to isolate the code for the antibody, cloning that code in cells, and then finally screening and testing those cells until the antibody can be isolated. This is complex and difficult work that is done on a small scale. While it is sufficient for use in lab research to try to understand the antibody, each procedure generally only produces a maximum of 30 micrograms of protein.

By comparison, the ongoing clinical trial testing VRC01 in humans involves doses ranging from one milligram of antibody per kilo to 40 milligrams per kilo. If an average person weighs 80 kilos, the necessary dose of antibody could be as much as 80 milligrams, or 80,000 micrograms. To manufacture this quantity of antibody for research studies in humans, a more industrial process is required. Making this process more efficient and cost effective could become a pressing issue should these bNAbs be used in the future as part of a treatment or cure strategy.

A dozen or so companies can manage protein production on an industrial scale in what are called bioreactors. Contemporary bioreactors are specialized stainless steel tanks used to cultivate cell banks. These are groups of cloned cells into which scientists have inserted the genetic material of the bNAb in order to multiply. This cultivation is a grueling process which can take months: workers need to find the right temperature and medium in which to grow the cell bank, and then the antibody needs to be purified from the solution in the bioreactor and tested rigorously. If it is to be made for clinical use, this needs to be done under very strict lab conditions, all of which must be thoroughly documented.

While biotech companies have made advances in purification and bioreactor design in recent years, the basic procedures remain fairly constant. NIAID has its own pilot plant with a bioreactor capable of producing enough VRC01 antibody doses for a Phase I trial. A typical pilot plant bioreactor size runs from 100 to 500 liter capacities, which can deliver five or six kilos of protein per batch. A manufacturing company is likely to use bioreactors that can handle 1,000 to 20,000 liter capacities.

But even given all the technological or feasibility issues, the greatest challenge in scaling up antibody manufacturing is likely to be the cost. Market rates for bulk bNAb protein production could run from US$30-$50 per gram, by some estimates. For use on a broad scale, either a developer would need to take on that cost or drive it down to $3-$4 per gram, which will be difficult to achieve. The vast quantities of purification solution and the filtering needed to effectively purify the antibody in the bioreactor keep costs high. Manufacturing antibodies is a very sophisticated and challenging operation.

To cut costs researchers are searching for cheaper media in which to cultivate cell lines and purify and extract the antibody solution, and are also considering using smaller single-use or even disposable bioreactors with faster production times. Researchers are also looking for new modes of production. Some scientists think using corn or tobacco plant cells is another way to quickly and inexpensively grow antibodies in bulk. But for now, the cost challenge for production remains daunting.

Michael Dumiak reports on global science, technology, and public health and is based in Berlin.