A Brief History of Human Challenge Studies (1900-2021) Emphasising the Virology, Regulatory and Ethical Requirements, Raison D'etre, Ethnography, Selection of Volunteers and Unit Design.

Venetian quarantine 400 years ago was an important public health measure. Since 1900 this has been refined to include "challenge" or deliberate infection with pathogens be they viruses, bacteria, or parasites. Our focus is virology and ranges from the early experiments in Cuba with Yellow Fever Virus to the most widespread pathogen of our current times, COVID-19. The latter has so far caused over four million deaths worldwide and 190 million cases of the disease. Quarantine and challenge were also used to investigate the Spanish Influenza of 1918 which caused over 100 million deaths. We consider here the merits of the approach, that is the speeding up of knowledge in a practical sense leading to the more rapid licensing of vaccines and antimicrobials. At the core of quarantine and challenge initiatives is the design of the unit to allow safe confinement of the pathogen and protection of the staff. Most important though is the safety of volunteers. We can see now, as in 1900, that members of our society are prepared and willing to engage in these experiments for the public good. Our ethnology study, where the investigator observed the experiment from within the quarantine, gave us the first indication of changing attitudes amongst volunteers whilst in quarantine. These quarantine experiments, referred to as challenge studies, human infection studies, or "controlled human infection models" involve thousands of clinical samples taken over two to three weeks and can provide a wealth of immunological and molecular data on the infection itself and could allow the discovery of new targets for vaccines and therapeutics. The Yellow Fever studies from 121 years ago gave the impetus for development of a successful vaccine still used today whilst also uncovering the nature of the Yellow Fever agent, namely that it was a virus. We outline how carefully these experiments are approached and the necessity to have high quality units with self-contained air-flow along with extensive personal protective equipment for nursing and medical staff. Most important is the employment of highly trained scientific, medical and nursing staff. We face a future of emerging pathogens driven by the increasing global population, deforestation, climate change, antibiotic resistance and increased global travel. These emerging pathogens may be pathogens we currently are not aware of or have not caused outbreaks historically but could also be mutated forms of known pathogens including viruses such as influenza (H7N9, H5N1 etc.) and coronaviruses. This calls for challenge studies to be part of future pandemic preparedness as an additional tool to assist with the rapid development of broad-spectrum antimicrobials, immunomodulators and new vaccines.

Influenza is now a preventable disease.

The world is waiting with apprehension for the predicted pandemic of H5N1 (avian) influenza as an increasing number of countries in Asia, Europe and Africa report cases of influenza in migrating birds. All is not 'despondency', however. Targeted and controlled administration of antiviral drugs, alone or in combination, to contacts and cases, together with well tried public health measures, should slow down the spread of the infection and allow time for vaccines to be developed, thus preventing a worldwide pandemic of the type that occurred in 1918.

World War I may have allowed the emergence of "Spanish" influenza.

The 1918 influenza pandemic caused 40 million deaths, and so dwarfed in mortality and morbidity the preceding pandemic of 1889 and the 1957 and 1968 pandemics. In retrospect, much can be learnt about the source, the possible subterranean spread of virus, and the genetic basis of virulence. The World Health Organization has urged every nation to prepare a pandemic plan for the first global outbreak of the 21st century. We present an appraisal of epidemiological and mortality evidence of early outbreaks of respiratory disease in France and the UK in the years 1915 to 1917. Certain of these earlier focal outbreaks--called epidemic bronchitis rather than influenza--occurred during the winter months when influenza was known to be in circulation, and presented with a particular heliotrope cyanosis that was so prominent in the clinical diagnosis in the world pandemic outbreak of 1918-1919 (the Great Pandemic). The outbreaks in army camps at Etaples in France and Aldershot in the UK in 1916-1917 caused very high mortality in 25-35 year olds. Increased deaths from bronchopneumonia and influenza were also recorded in England. We deduce that early focal outbreaks of influenza-like disease occurred in Europe and on the balance of probability the Great Pandemic was not initiated in Spain in 1918 but in another European country in the winter of 1916 or 1917. We suggest that the pandemic had its origins on the Western Front, and that World War I was a contributor.

Antivirus antibodies in myasthenia gravis.

Serum antibodies to influenza A, measles, rubella, cytomegalovirus, varicella zoster, herpes simplex type 1, and mumps have been assayed in 104 patients with myasthenia gravis, grouped according to clinical features plus thymus pathology, and compared with matched controls. No significant differences in incidence or antibody titer were detected. In 37 patients with recent onset of symptoms, the incidence of antibody to coxsackieviruses B1-B6 was less than in controls. Juvenile-onset cases also demonstrated antibody to Epstein-Barr virus at the expected frequency. These results weaken the case for any of these common viruses, or the response to them, contributing to the pathogenesis of myasthenia gravis.

Sequence analysis of the equine H7 influenza virus haemagglutinin gene.

The nucleotide sequences of ten haemagglutinin genes of representative H7N7 equine influenza viruses isolated between 1956 and 1977 have been determined by primer extension sequencing. Their nucleotide and deduced amino acid sequences demonstrate a high degree of homology. These equine viruses can be divided into two distinct subgroups, the prototype-like, and a group comprising the early American isolates and the remaining equine viruses. The equine H7 haemagglutinins form a quite distinct group compared to H7 haemagglutinins isolated from other species. Each of these equine H7 haemagglutinins possess a tetrabasic amino acid cleavage site separating the HA1 and HA2 domains but, in addition, all ten contain a nine amino acid insertion prior to the tetrabasic sequence. The haemagglutinin glycoproteins of all ten viruses are capable of cleavage activation in virus infected primary chicken embryo fibroblast cells.

Biochemical and serological studies of influenza B viruses: comparisons of historical and recent isolates.

The genetic characteristics of 24 representative influenza B viruses isolated in widely different geographical areas of the world between 1940 and 1980 were analysed using either RNA:RNA hybridisation or oligonucleotide mapping. Additional biochemical characterisation included electrophoretic analysis of virus-induced polypeptides and virion RNAs. A panel of monoclonal antibodies to virus HA was used to investigate serological relationships between the viruses. The influenza B viruses examined constituted a genetically and serologically related group but mutational changes were detected in all eight genes of the viruses isolated in different eras and also in genes of viruses isolated in the same epidemic year. Regardless of the overall and dominating similarities, at a higher level of discrimination it was clear that certain genetic and serological relationships were more complex than expected and, for example, some recently circulating field viruses were apparently more closely related antigenically and genetically to viruses isolated five to twelve years previously than to other viruses isolated concurrently. No evidence of recombination with hitherto undescribed influenza B viruses and with genes coding for internal proteins was detected.

HIV type 1 envelope glycoprotein 120 carboxy-terminal peptide-induced human T cell lines selectively suppress heterogeneous proliferative T cell responses to soluble antigens.

It has been proposed that the highly conserved human immunodeficiency virus type 1 (HIV-1) envelope gp120 carboxy-terminal sequence, TKAKRRVVEREKR (CT120), may represent a functional mimic of the human leukocyte antigen (HLA) class II DR beta-chain third hypervariable region (HVR3) sequence motif located at position 69-81. Presentation of this potentially pathogenic fragment by HLA class I and/or II molecules, in a manner analogous to the indirect pathway of allorecognition, may induce both widespread cellular activation and also break self-tolerance, resulting in the selective and progressive anti-self HLA class II-directed immune suppression, which is a central feature of HIV-1 infection and the associated acquired immune deficiency syndrome (AIDS). To investigate the functional role of the HIV-1 gp120 C-terminal fragment T cell lines (TCLs) were raised from three healthy HIV-1-seronegative subjects at low risk of HIV-1 exposure, by repeated stimulation with a short synthetic 13-mer CT120 peptide in vitro. Graded concentrations (10[3] to 5 x 10[4]) of CT120 TCLs suppressed the primary 6-day proliferation of autologous PBMCs in response to the soluble antigens tetanus toxoid (TT) and purified protein derivative (PPD). In contrast, CT120 TCLs demonstrated no suppressive effect on 3-day phytohemagglutinin (PHA), concanavalin A (ConA), and pokeweed mitogen (PWM) mitogenic responses. Fractionation of CT120 TCLs into highly purified CD4+ and CD8+ T cell subsets demonstrated that the CD8+ T cell fraction mediated the suppressor effector function. HLA restriction analysis revealed a complex pattern as both anti-HLA class II DR and anti-HLA class I (A, B, C) MAbs inhibited proliferation of oligoclonal CD8+ CT120 TCLs. Strategies aimed at specifically inhibiting such putative immunopathogenic HIV-1-encoded T cell epitopes may be an important consideration for development of future HIV-1 immunotherapy.

The H274Y mutation in the influenza A/H1N1 neuraminidase active site following oseltamivir phosphate treatment leave virus severely compromised both in vitro and in vivo.

Oseltamivir carboxylate is a potent and specific inhibitor of influenza A and B neuraminidase (NA). Oseltamivir phosphate, the ethyl ester prodrug of oseltamivir carboxylate, is the first orally active NA inhibitor available for the prophylaxis and treatment of influenza A and B. It offers an improvement over amantadine and rimantadine which are active only against influenza A and rapidly generate resistant virus. The emergence of virus resistant to oseltamivir carboxylate in the treatment of naturally acquired influenza infection is low (about 1%). The types of NA mutation to arise are sub-type specific and largely predicted from in vitro drug selection studies. A substitution of the conserved histidine at position 274 for tyrosine in the NA active site has been selected via site directed mutagenesis, serial passage in culture under drug pressure in H1N1 and during the treatment of experimental H1N1 infection in man. Virus carrying H274Y NA enzyme selected in vivo has reduced sensitivity to oseltamivir carboxylate. The replicative ability in cell culture was reduced up to 3 logs, as was infectivity in animal models of influenza virus infection. Additionally, pathogenicity of the mutant virus is significantly compromised in ferret, compared to the corresponding wild type virus. Virus carrying a H274Y mutation is unlikely to be of clinical consequence in man.

Lack of detection of influenza genes in archived formalin-fixed, paraffin wax-embedded brain samples of encephalitis lethargica patients from 1916 to 1920.

A method was developed for detection of influenza genes in formalin-fixed brains of mice that had been experimentally infected with influenza A/NWS/33 (H1N1) virus. Using this technique, messenger ribonucleic acid (mRNA) of the beta-actin gene was detected in eight clinical brain samples from the 1916-1920 outbreak of encephalitis lethargica, showing preservation of particular mRNAs. However, we did not detect influenza nucleotide sequences of M, NP, and NS genes from these same samples. We conclude either that influenza was not the causative agent of encephalitis lethargica or, possibly, that the virus had a hit-and-run mechanism and was no longer present in the brain at the time of death of the patients.

Direct sequence determination of the influenza B HA-1 gene after PCR amplification of clinical specimens from an infected volunteer.

When clinical isolates of influenza A and B viruses are propagated in embryonated hens' eggs or tissue culture cells, different selective pressures in vitro result in specific amino acid substitutions in the haemagglutinin (HA) gene. A proportion of such viruses which lose a potential glycosylation site near the receptor binding region of the HA at amino acid positions 196-198 appear to have reduced virulence. Direct polymerase chain reaction (PCR) amplifications of cDNA and subsequent nucleotide sequence analysis of part of the HA-1 gene of the original infecting influenza B strain and the nasal wash material from an infected volunteer were performed. The nucleotide sequences of the viral HA-1 from the nasopharynx of the infected volunteer were the same as that of the original infecting strain. Antigenic analysis of both the original infecting virus and the viruses isolated from sequential samples collected from the volunteer, all of which were cultivated on Madin-Darby Canine Kidney (MDCK) cells and in embryonated hens' eggs, revealed variation in the HA of viruses only after egg adaptation. In particular, we describe the use of direct nucleotide sequencing techniques without the use of cloning strategies in order to determine the sequence of the HA-1 gene after direct PCR amplification of clinical nasal wash material.

Evidence of diffusion artefacts in diaminobenzidine immunocytochemistry revealed during immune electron microscope studies of the early interactions between influenza virus and cells.

Anti-haemagglutinin-labelled antibodies have been used to search for influenza entry into cells by fusion of viral and plasma membranes. The plasma membranes of infected cells were stained by immunoperoxidase but not by immunoferritin reagents. It is suggested that the staining obtained with the peroxidase conjugate was due to diffusion of the diaminobenzidine reaction product away from the enzymic site. Immunoferritin labelling provided no evidence for entry of influenza by fusion of viral and plasma membranes under conditions of physiological pH.

Treatment of epidemic and pandemic influenza with neuraminidase and M2 proton channel inhibitors.

A small armentarium of anti-influenza drugs now exists, and includes the M2 blockers (amantadine and rimantadine) and the neuraminidase inhibitors (Relenza and Tamiflu). The neuraminidase inhibitors have certain advantages, including a broader spectrum of antiviral activity, including influenza A and B viruses. On the other hand, there is now much clinical experience with the M2 blockers, and these drugs are inexpensive. It is clear that influenza in different community groups needs to be managed in specific and targeted ways. For example, in the over-65-years and at-risk groups, vaccination will remain a mainstay of disease prevention. However, up to 40% of those in these groups may fail to receive vaccine, and therefore the antivirals can be used therapeutically, or, in defined circumstances, as prophylactics. At present, influenza is hardly managed in the community. The infrequent global outbreaks, pandemics, present further problems. The more extensive use of the two classes of antivirals, and also vaccines, in the important interpandemic years will provide a very significant investment in health benefits in the face of a new pandemic virus in an otherwise completely vulnerable population.

A designer drug against influenza: the NA inhibitor oseltamivir (Tamiflu).

The description of the first two designer antiviral drugs to fight influenza was a ground breaking advance. Targeted against the influenza neuraminidase enzyme these inhibitors have been shown to reduce both the severity and duration of influenza illness. Importantly, it is expected that these neuraminidase inhibitors would be effective against influenza pandemic strain and could therefore be vital at reducing the potentially devastating consequences of such an outbreak. Despite the demonstrated efficacy of these drugs, they are not commonly used, particularly in the UK, and there is substantial concern that in the event of a pandemic or even a severe epidemic there could be substantial morbidity and mortality. SARS has shown that the public and media response to a serious epidemic is not always rational and this could easily become panic if it became apparent that treatment was possible, but not available.

Attenuation of virulence in influenza B viral infection of volunteers.

A study was performed with volunteers in order to determine whether the virulence of an influenza B viral infection could be attenuated. Of a total of 62 persons, 26 received intranasal inoculations of an unpassaged influenza B virus isolate [virus U], while 26 received the same virus isolated passaged in cell culture and then in special pathogen-free embryonated hens' eggs [virus P]. The remaining 10 persons received uninoculated cell culture medium. Daily nasal wash samples were collected post-infection and scores for illness in the volunteers were evaluated. Viruses were isolated in cell culture and in eggs. Isolates were analysed by means of monoclonal antibodies (MAbs) raised against the influenza B viral haemagglutinin (HA) glycoprotein. One-step and nested polymerase chain reactions as well as direct nucleotide sequence analysis of part of the HA gene of the unpassaged specimens, together with the influenza B viruses isolated from those specimens, were performed. Nine of 26 volunteers who received the unpassaged virus became ill (illness scores from 13-84/100) whereas 7/26 of the volunteers who received the passaged virus had very mild illness (illness scores from 3-7/100). All the other volunteers remained well. Results of analysing the clinical specimens collected from the two groups of volunteers were compared. A difference in MAb reactivity, together with an aspartate for asparagine amino acid substitution at position 196 in a 432 base pair region of the viral HA gene, was observed. The loss of a potential glycosylation site at amino acid position 196-198 in the viral HA was associated with attenuation of virulence. This finding may have implications for the formulation of influenza vaccines.

Natural and 'in vitro' selected antigenic variants of influenza A virus (H2N2).

We provide data on the prevalence of SRH antibody to influenza A/Singapore/1/57 (H2N2). Approximately 10.3% of sera had antibody to the influenza A (H2N2) subtype virus in comparison to the 36.9% of positive sera to a representative influenza A (H3N2) and 31.5% to influenza A (H1N1) viruses. The percentage of subjects with antibody constantly decreased from the older to the younger age groups. Persons born after 1968 were essentially seronegative, whereas subjects born before 1900, and in the decade 1950-1959, showed the highest antibody levels to influenza A (H2N2) viruses. These age groups also appeared to have 'protective' levels of anti-HA antibody to influenza A (H2N2) virus. An antigenic variant of A/Singapore/1/57 virus was selected in the laboratory using a monoclonal antibody to HA. Serological comparison of the new in vitro variant with the parental virus and two naturally occurring viruses, namely A/England/12/64 and Tokyo/3/67, showed that certain human sera were able to distinguish the variant, indicating a restricted antibody repertoire in these adult and children's sera, providing an explanation of how such variants could actually arise in nature.

Zanamivir (Glaxo Wellcome).

Zanamivir, a neuraminidase inhibitor under license from Biota, has been developed and launched by Glaxo Wellcome for the treatment of influenza virus infection [277127]. While registered as drugs, neuraminidase inhibitors are technically vaccines as they are believed to work by provoking an immune response, thus preventing the spread of infection from one cell to another within the respiratory tract [314356]. It is administered, using an inhaler, directly to where the virus replicates, minimizing the potential for side effects. Zanamivir has been launched, under the name Relenza, for the treatment of influenza in Australia [325148], New Zealand [335591] and the UK [341871]. It is registered in Mexico and Armenia [346474], and approved in Japan [351580]. In October 1999, it was launched in the US [352392] and Glaxo confirmed in January 2000, that zanamivir has been launched all over Europe, with the exception of Greece and Spain, where launch is imminent [352513]. In October 1999, the British Pharma Group (BPG) wrote to the UK Prime Minister to seek a discussion on the implications for the British-based pharmaceutical industry of the recommendation from NICE's Raised Appraisal Committee that Relenza should be banned from the NHS [342597]. An FDA submission was made in October 1998 [302732,304779]; the FDA advisory committee met on February 24 1999 to discuss the application [313583]. The Antivirals Drugs Advisory Committee voted against recommending zanamivir for FDA approval [316196]. Members suggested that additional studies using a different endpoint or targeting specific high-risk populations would be more likely to garner support [317454]. Following a discussion with Glaxo Wellcome in April 1999, the FDA indicated that it would continue to review the NDA for zanamivir [320565]. In July 1999, the company received final FDA approval for the drug for the treatment of both influenza A and B [334127], and aimed to initiate a phase IV trial program for the detection of consumer problems with the product, following launch of the product in time for the 1999-2000 flu season [352392,335048]. In September 1998, Glaxo applied to the Canadian Health Protection Branch (HPB) for approval to market zanamivir for the treatment and prevention of influenza. A separate application for Fast Track approval was also made with the HPB [299375,299472]. In November 1999, zanamivir received approval in Canada [346474]. Glaxo expects to make MAA and NDA filings for the use of zanamivir as a prophylactic in 1999 [333494]. In July 1998, Merrill Lynch predicted that as zanamivir offered only modest benefits gained from a narrow treatment window, potential would be limited and forecast annual sales of US dollar 30 million [302048]. In January 1999, Paribas predicted sales of pounds 75 million in 1999, rising to pounds 280 million in 2003 [317650]. In April 1999, ABN Amro predicted sales of pounds 23 million in 1999, rising to pounds 326 million in 2003 [328676]. In August 1999, Lehman Brothers estimated that sales of zanamivir will reach US dollar 400 million by 2003, and that it is an obvious candidate for an OTC switch, although this is not likely until 5 years after launch [336356].