Prepping for the Immunome

Could a group of experts lead the way to ambitious  new vaccine trials that aim to map the human  immune response and speed vaccine development?

By Michael Dumiak

In New York City this summer, Marie-Paule Kieny, assistant director-general for health systems and innovation at the World Health Organization, and Ted Bianco, innovations director at the Wellcome Trust, convened a small group of researchers, public health experts, and medical industry insiders pursuing an audacious idea: mapping the human immune system.

What is starting with small steps could develop into big science. The July gathering is the second in a set of workshops underwritten by the Robert Wood Johnson Foundation (RWJF) aimed at developing a scientific and, equally vital, business plan for what the International AIDS Vaccine Initiative’s chief science officer Wayne Koff, University of Melbourne microbiologist Ian Gust, leading University of Pennsylvania vaccine authority Stanley Plotkin, and others are dubbing the Human Vaccines Project. Analogous in scope, if not method, to the Human Genome Project of the 1990s, the group outlined the initiative’s objective last May in the journal Science and this June in Nature Immunology as a comprehensive assessment of human immune responses to licensed and experimental vaccines in rapid, focused, and iterative clinical research trials.

The Brains Behind the Human Vaccines Project  



It would be a vast undertaking: a small number of volunteers, a huge number of experiments, and testing untold hundreds of thousands of antigens and the immune responses they induced at a level of detail and scale that is only now becoming possible due to technological and methodological advances in antigen discovery, genomics, and immunological monitoring. Koff describes it as an effort focused on the major problems impeding vaccine development. “One is how to elicit a specific, broad, potent, and durable immune response in people,” he says. “Another is how can we optimize the efficacy of licensed vaccines and improve the potential for the next generation of vaccines for specific populations: like in the developing world, in newborns, or in the elderly.”

Overall, the goal is to speed development of new and improved vaccines for major global diseases such as HIV, influenza, cancers, dengue, and other infectious illnesses. The Human Vaccines Project aims to do this by creating a reproducible platform for screening related vaccine immunogens in humans to determine which are able to elicit the broadest, most potent, and durable immune responses. The immune responses to these immunogens could be tracked and analyzed using the new molecular tools and technologies coming online over the last decade. By cataloguing the comprehensive responses from these small human trials, Koff and the others behind the Human Vaccines Project hope to create a map of the immunome—much like the map of the human genome—which details all of the genes and proteins associated with the human immune response to vaccine antigens and as much as can be learned about their modes of action.

“You don’t want to just immunize one or two people here and there and make a few anecdotal findings. You want to do this on a big scale and in a systematic way so it’s absolutely clear what the conclusions are,” says Dennis Burton, an immunologist and HIV vaccine researcher at the Scripps Research Institute in La Jolla, California. “The strategy is to look at the response of different individuals and find those magic antibodies, the ones that are extraordinarily effective, and garner information allowing you to design a more effective vaccine.” One example would be using this information to develop a universal influenza vaccine. Now, vaccination against flu requires an annual shot that is manufactured based on what strains of the virus are predicted to be in circulation in a given year. But if researchers could identify an immunogen that could induce antibodies that would protect against many different types of flu, it might be possible to develop a vaccine that would provide life-long protection against several strains of the virus.

Inner workings

In some ways, the easy vaccines have been made. Current vaccines have been so successful, saving tens of millions of lives, but most existing vaccines were developed without a deep understanding of how they work, Burton says. Over time there’s been a progression: in the 1950s, the strategy for developing vaccines was to isolate a pathogen, kill it or weaken it, and inject this modified form into a person. This strategy worked well for polio and smallpox. But this strategy isn’t practical for viruses like HIV: it’s not safe. Many of the pathogens researchers are now trying to create vaccines against are also more challenging foes. “A lot of the diseases we and others are looking at now are either complex pathogens, persistent pathogens, immunovasive pathogens, or hypervariable pathogens,” Koff says. “Or cancers.”

Koff, Gust, and Plotkin outline a landscape of problems facing contemporary vaccine research in their Nature Immunology article. Genetic variation impedes development of vaccines against HIV, blood-stage malaria, and influenza; population-specific characteristics, such as the immaturity of the immune system in newborns and its deterioration in the elderly, limit the efficacy of some already-licensed vaccines and pose hurdles for creating new ones; and identifying potential antigens that can induce the necessary protection is still difficult. Animal models for vaccine research are another limiting factor—as Burton says, ultimately researchers want to gain information in humans because that’s who the vaccines are for.

Several vaccines tested in the last decade targeting these more complex pathogens have either completely failed in efficacy trials or only provided a modest level of protection: vaccines to prevent HIV, malaria, herpes simplex, staphylococcus aureus, melanoma, and pancreatic cancers all make this grim list. “When you look at all the time and expense, you’re looking at decades of work and billions of dollars,” Koff says. Meanwhile, the Human Genome Project cost about $3.5 billion, including public and private efforts. A Human Vaccines Project might be equally thrifty—comparatively speaking—if it could eliminate some of the hurdles in vaccine development and get new products to market more quickly. But this doesn’t mean funding a project like the one Koff and colleagues are proposing is easy.

“I think one of the key points is that any study in humans costs a lot of money. We need a lot of funding for this,” says Burton. “You’d need stakeholders with very deep pockets. But the rewards could be massive,” he adds, referring to the ability to design vaccines with much greater certainty and reliability. Venture capitalists and industry representatives came to the table for the last workshop and will no doubt play a vital part, given limited prospects for government funding of big initiatives.

The group behind the Human Vaccines Project is putting polishing touches on its business and scientific draft plans and expects to publish an update in the first part of next year, in advance of a third workshop funded by the RWJF grant. At that point they hope to reach agreement on the infrastructure and resources—the “enabling environment”—necessary to create something that resembles the Human Genome Project. The group’s already met with senior leadership of the US Food and Drug Administration to discuss a key element of their plans: how to ensure that products for a large number of clinical trials involving a relatively small number of individuals can be delivered safely.

The Human Vaccine Project will also need a home. The question is should this effort be independent, part of a research institute or university, a for-profit or a non-profit endeavor? “In that sense it is similar to the Human Genome Project. It would require collaboration among many different groups with different expertise in order to develop the answers we want,” Burton says. There are also other questions regarding intellectual property and leadership that need to be addressed.

Technology leads the way

Efforts to better characterize the human immune response are possible now because new technologies are creating unprecedented potential for analysis and cataloguing. Researchers can now use genetic sequencing to analyze tens or even hundreds of thousands of different antibodies induced by a vaccine. “If we’re talking about a large number of clinical research studies with comprehensive immunologic assessments, looking at the whole antibody repertoire, this is a huge amount of data,” says Koff. You want to be able to manipulate that data and get as much information out of the study as you possibly can. We now have the data management and informatics tools to be able to do that.”

Pharmaceutical companies could also benefit from collecting the types of data Koff is referring to. In recent weeks Sanofi claimed success in a second large clinical trial with its experimental dengue vaccine, into which it has invested $1.7 billion already. But against dengue serotype 2, protection was only 42%. Knowing whether this was because of a population or immunology effect would make it easier to make adjustments to the vaccine to improve the response against this serotype.

Technology is also aiding antigen design. Vanderbilt Vaccine Center director and pediatric infectious disease specialist James Crowe points to the widening collection of three-dimensional antigen structures made with sequences and molecular modeling systems like the one developed at the University of California, San Francisco, known as Chimera. These along with pathogen genome sequencing and the identification, expression, and screening of protein antigens are leading to the increasing viability of optimized structure-based vaccine candidate design or reverse vaccinology. Earlier this year Crowe was part of a large group publishing a proof of principle for epitope-based vaccine design (Nature 507, 201, 2014).

Advances in synthetic biology—namely the ability to synthesize genes at a large scale, to the point of the whole organism genome—are also allowing researchers to study the effect of genetic diversity upon antigens and the immune repertoires. In terms of immunologic monitoring, new techniques for measuring gene expression, such as RNA-Seq, the use of metabolomics, and immune repertoire sequencing bring extremely detailed views of the human immune response into view. As Koff points out, the identification of broadly neutralizing monoclonal antibodies has recently re-energized efforts toward the development of vaccines against HIV and influenza. In part this is thanks to advances in computational and structural biology.

Despite these advances, researchers are still groping somewhat in the dark when it comes to eliciting these antibodies through vaccination. “We don’t know how to elicit neutralizing antibodies because we don’t really know the rules of immunogenicity around affinity maturation,” Koff says, referring to the way in which immune cells mature and mutate in lymph nodes in response to pathogen exposure and are able to produce higher affinity antibodies. For those developing HIV vaccine candidates, knowing how different antigens guide the immune system to make more highly affinity-matured antibodies is a huge challenge. Researchers need a map. Enter the immunome.

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