IN early February 2020, we presented on our first Controlled Human Infection for Vaccination Against Strep A (CHIVAS) study at the Third Human Challenge Trials in Vaccine Development meeting in Oxford. The meeting “brought together a broad range of international stakeholders, including academia, regulators, funders and industry, with a considerable delegation from Low- and Middle-Income Countries”, yet outside of that room few of our colleagues in clinical medicine and research, and fewer among the general public, had ever heard of a human challenge trial.
Just months later, tens of thousands of adults had signalled their interest in volunteering for human challenge trials to study a novel pandemic coronavirus, and clinicians, scientists and commentators were debating the merits of modern experimental human infection research in leading journals and on social media (here, here, here, and here).
Human challenge trials for COVID-19 were alternately cast as an express route to miraculous deliverance from the perils of the pandemic, or a dangerous and unethical fool’s errand. For human challenge researchers, neither framing sat comfortably.
In the plainest terms, human challenge trials involve giving people a carefully monitored infection, with health care support, to learn how diseases work and how to stop them. By 2018, more than 22 000 volunteers had participated in modern human challenge trials studying at least 22 different pathogens, including bacteria, viruses, and parasites.
Human challenge trials may help to accelerate product development and understanding of infectious diseases in a number of ways: establishing pathogenicity of an organism; understanding pathogenesis of a disease; identifying risk factors for infection and disease; estimating the infective dose; characterising the human immune response; assessing infection-derived immunity (challenge after natural infection or repeat challenge); evaluating vaccine (or drug) efficacy; screening for new diagnostics; and helping to identify or validate correlates of protection.
For the uninitiated, mere mention of experimental human infection may conjure images of the worst historical examples of unethical human research, from the sexually transmitted disease studies in Guatemala to Auschwitz-Birkenau. The common theme in those criminal endeavours was dehumanisation, with the participants deprived of information and agency, and with morbidity or even mortality as intended outcomes. In the starkest possible contradistinction, modern human challenge research is grounded in continuous ethical self-reflection and quality improvement, mitigating participant risks and maximising autonomy, by design. The safety record of modern human challenge trials compares well with conventional early-phase clinical trials, with very few serious adverse events and (unlike standard Phase 1 drug trials) no deaths or permanent disability related to participation.
For any given human challenge trial proposal, the critical ethical questions are usually: (i) is this trial useful enough?; and (ii) considering all measures taken to mitigate risks, is this trial safe enough? Really, these are generic questions for all clinical research involving humans. There are always unanswered questions related to the experimental intervention – that is the point of doing the research. In most cases, we can actually make a more informed risk assessment of how an infectious challenge might affect human volunteers than is possible in early-phase trials of novel therapeutics.
As recently highlighted by our Western Australian colleagues in the MJA, Streptococcus pyogenes (Strep A) remains a formidable human pathogen with an immense global burden of disease – there is a clear unmet need for development of a vaccine (here, and here). A successful Strep A vaccine would be like hepatitis B and human papillomavirus vaccines in preventing a spectrum of communicable and non-communicable disease. The death of a previously well 7-year-old girl in Perth this April was a tragic reminder that Strep A is a leading cause of community-acquired sepsis. As the microbiological cause of rheumatic heart disease, uncontrolled Strep A disease undermines the campaign to close the health and life-expectancy gap between Aboriginal and Torres Strait Islander peoples and non-Indigenous Australians. It is fitting that Australian leadership has reinvigorated global Strep A vaccine development efforts.
To overcome longstanding limitations of laboratory assays and animal models of Strep A disease, the 2018 World Health Organization roadmap for Strep A vaccine development called for development of new human models. We led a multinational team in establishing the world’s only Strep A controlled human infection model as a platform for human challenge trials to accelerate development of new Strep A vaccines. The opportunity for early evaluation of vaccines in the challenge model begins to address a chicken-and-egg strategic dilemma impeding Strep A vaccine development – that doubt surrounding the efficacy of candidate vaccines has prevented their progression to large and expensive field trials to establish their efficacy (here, here, and here).
Drawing on the expertise of our diverse investigator group and encouraged by the success of historical human infection studies with Strep A, we selected a strain that would cause pharyngitis (“strep throat”) but was very unlikely to cause invasive infections or post-infectious complications such as acute rheumatic fever and glomerulonephritis. We took the same approach to designing the study protocol, aiming to produce a clinically relevant strep throat syndrome while protecting participants from complications of streptococcal infection. First, we needed to prove that we could safely cause strep throat in most of participants. We carefully screened healthy adult volunteers, as is routinely done for clinical trials, and did echocardiograms to rule out subclinical rheumatic heart disease. For the challenge, volunteers were admitted to a dedicated clinical research ward supported by a tertiary adult hospital.
Starting at one-tenth of the dose used in historical studies, we used a swab to paint our special Strep A strain on the back of each participant’s throat. Some investigators worried that the intensive efforts to design a safe trial might mean that nothing happened, so there was great relief when, from the very first participant at the very low starting dose, most participants developed a convincing strep throat. Of the first 20 participants, 17 (85%) had a red and sore throat, some had enlarged tonsils with exudates, and a few had a fever and chills. Pharyngitis or no pharyngitis, every participant was treated with antibiotics. We tried an even lower dose but only one of five participants had a strep throat. The participants all recovered quickly, and antibiotic treatment eradicated the bacteria from the throat of every participant. There were no other infections besides strep throat and no post-infectious complications. We followed every participant closely for 6 months as outpatients. We collected blood, saliva and throat swabs, and scientists at Murdoch Children’s Research Institute (MCRI) and around the world are now doing careful scientific research, studying these samples alongside samples from natural human Strep A infections, to describe how humans respond to Strep A infection in greater detail than has ever been possible before.
Next, we are going to test new vaccines developed by researchers in Australia and elsewhere, in challenge trials funded by the Heart Foundation and the Australian Strep A Vaccine Initiative. In these randomised controlled double-blinded studies, about 50 healthy adult participants will either get a Strep A vaccine or a placebo. Then we will paint our challenge strain onto their throats and compare the proportion who get a strep throat in each group. In another trial, we are working with researchers at Telethon Kids Institute to use the model to help develop better and longer-acting penicillin formulations to improve secondary prevention of rheumatic heart disease.
The limitations of human challenge trials are mostly intrinsic to model-based research and relate largely to their scientific generalisability and their relative strategic usefulness compared with more conventional product development pathways. Human challenge trials generally recruit only healthy younger adults and generally study self-limited, asymptomatic, or relatively mild symptomatic infections. Can our healthy adult model of pharyngitis in Melbourne using a single Strep A strain inform development of a vaccine eventually aiming to prevent severe syndromes around the world which disproportionately affect young children, older people, pregnant women and Indigenous communities?
Development of vaccines and other medicines, by whatever pathway through Phase 1, 2 and 3 and more, is always the result of motivation, inspiration, perspiration, good luck and educated steps into the unknown. Human challenge trials will never answer every question. Phase 3 trials do not do that either. Even in a pandemic, there are narrow opportunities for phase 3 trials – “vaccine development requires someone to be infected somewhere” (here, here, and here). And even the biggest field trials struggle with generalisability and are usually underpowered to detect rare adverse events, such as thrombosis with thrombocytopenia syndrome after adenovirus-vector vaccines, and myocarditis after mRNA vaccines (here, and here).
Notwithstanding Nobel laureate Sydney Brenner’s 2008 comment that “We don’t have to look for a model organism anymore. Because we are the model organisms”, it is unusual for vaccine efficacy in a human challenge trial to be decisive in licensure decisions without a subsequent phase 3 field trial. Human challenge trial data will not suffice for licensure of a first generation of modern Strep A vaccines but it may also be true that the necessary late-phase field trials will not happen without human challenge trials. Over a century of in vitro and animal research has not delivered a human Strep A vaccine or addressed critical knowledge gaps regarding human immunity against a pathogen that only naturally infects humans. Instead, there is a growing pile of pre-clinical vaccines proven to protect laboratory mice from experimental Strep A challenge and just one Strep A vaccine has proceeded to an adult human phase 2 safety and immunogenicity study in the past 40 years.
Development of a modern controlled human infection model of Strep A pharyngitis is a statement of intent to finally cross the species barrier and develop an effective vaccine against this highly adapted, co-evolved and human-restricted pathogen.
Strep A human challenge trials using our model and done according to stringent modern scientific and ethical standards can meaningfully contribute to shifting strategic and scientific barriers still standing in the way of vaccine development. We do not and cannot know how far experimental pharyngitis in healthy adults caused by a single strain can propel our understanding of the full spectrum of natural Strep A diseases, but pursuit of perfection would have been a trap. We did not aim to build a better mouse. As the statistical aphorism goes: “All models are wrong, but some are useful”. We have developed a model that will serve its intended purpose of helping to accelerate towards the goal of an effective vaccine to finally eliminate the deep and enduring global burden of Strep A diseases.
Dr Joshua Osowicki is a consultant paediatrician and paediatric infectious diseases physician at the Royal Children’s Hospital Melbourne and completed his PhD in the Tropical Diseases research group at the Murdoch Children’s Research Institute.
Professor Andrew Steer is a consultant paediatrician and paediatric infectious diseases physician at the Royal Children’s Hospital Melbourne. He is Director of the Infection and Immunity theme, and Group Lleader of the Tropical Diseases research group at the Murdoch Children’s Research Institute.
The statements or opinions expressed in this article reflect the views of the authors and do not represent the official policy of the AMA, the MJA or InSight+ unless so stated.