Skip to main content

Oral Vaccination against Clostridium difficile Infection

Final Report Summary - CDVAX (Oral Vaccination against Clostridium difficile Infection)

Executive Summary:
Clostridium difficile, a spore forming Gram-positive bacterium is now the leading cause of nosocomial infection in the developed world. C. difficile infection (CDI) primarily affects elderly hospitalised patients who have received antibiotic treatment, the use of which results in a disturbance of the bacterial flora in the gastro-intestinal (GI) tract. This allows spores of C. difficile to germinate and bacteria to proliferate, with the concurrent production of toxins that lead to severe diarrhoea and acute colitis. CDI is associated with a significant mortality rate, particularly in the elderly. It is extremely difficult to contain a C. difficile outbreak within the hospital setting, and large outbreaks of multidrug resistant epidemic strains have been documented throughout Europe. Finally, after antibiotic therapy and remission, a proportion of patients (~30%) enter a state of relapse or recurrence where the disease re-appears. C. difficile is a particularly important pathogen since most mammals are known carriers and for livestock such as pigs, spores of C. difficile are shed into the environment where they make contact with farm workers and ultimately the general population. Similarly, they contaminate raw meat and enable entry into the food chain. C. difficile is a potent carrier of antimicrobial resistance genes and some strains are multidrug resistant where this resistance can easily transfer to other strains and species. C. difficile also is adept at evolving into more virulent strains that are noticeably more infectious. Taken together C. difficile presents a major problem for public health and there is currently no prophylactic treatment. Treatment of CDI is by use of antibiotics which is a particular problem for a pathogen that is extremely capable of acquiring and developing resistance to antimicrobials. Good hygiene practiced in hospitals has been shown to halt the increase in CDI but what is needed is a vaccine. It should be noted that vaccines under development are all injectable and do not prevent colonisation of the host by C. difficile but rather, only neutralise toxins and alleviate symptoms of disease.

The CDVAX consortium has adopted a novel oral vaccine consisting of recombinant spores of the bacterium Bacillus subtilis that have been engineered to display an antigen which is part of C. difficile toxin A. Oral immunisation with spores displaying this antigen have been shown to protect hamsters from oral challenge with virulent C. difficile and provide the impetus for a phase 1/2a clinical trial. The CDVAX vaccine is unique and unlike other injectable vaccines since it prevents colonisation of the host by the pathogen as well as neutralising toxins and has obvious long-term utility as a best-in-class vaccine for CDI.

The CDVAX consortium comprises five partners. First, is Royal Holloway University of London (UK) who are the inventors of the oral vaccine. The remaining four partners are SMEs each of whom have played a key role in the phase 1/2a clinical trial. Leads-2-Development (France) have provided the regulatory support for the clinical trial and have interfaced with the Paul Ehrlich Institute for approval of the clinical studies. CRIS (France) have provided all the bioanlaytical tests required for the development of the test product while Amatsi-QBiologicals (Belgium) have produced the cGMP test product. Lastly FGK (Germany) are supervising the human studies. The CDVAX project has, over 42 months, involved evaluation of the optimal dosing regimens in animals of in animal models of CDI as well as defining the mechanism of protection. Concurrently, Amatsi-QBIO have developed methods to produce a cGMP test product consisting of CDVAX spores. CRIS have developed and validated numerous analytical test methods to evaluate the clinical trial data as well as conducting GLP-accredited toxicology analysis. FGK are now supervising the human studies involving oral administration of cGMP CDVAX in healthy volunteers.

The CDVAX study represents the first-in-man study of a Bacillus spore vaccine using oral delivery. The key attributes of the CDVAX vaccine other than oral delivery are that it 1) generates immune responses that neutralise C. difficile toxins, and 2) these immune responses prevent colonisation of the host by C. difficile.

Project Context and Objectives:
Clostridium difficile, a spore forming Gram-positive bacterium is now the leading cause of nosocomial infection in the developed world. C. difficile infection (CDI) primarily affects elderly hospitalised patients who have received antibiotic treatment, the use of which results in a disturbance of the bacterial flora in the gastro-intestinal (GI) tract. This allows spores of C. difficile to germinate and bacteria to proliferate, with the concurrent production of toxins that lead to severe diarrhoea and acute colitis. CDI is associated with a significant mortality rate, particularly in the elderly. It is extremely difficult to contain a C. difficile outbreak within the hospital setting, and large outbreaks of multidrug resistant epidemic strains have been documented throughout Europe. Finally, after antibiotic therapy and remission, a proportion of patients (~30%) enter a state of relapse or recurrence where the disease re-appears. C. difficile is a particularly important pathogen since most mammals are known carriers and for livestock such as pigs, spores of C. difficile are shed into the environment where they make contact with farm workers and ultimately the general population. Similarly, they contaminate raw meat and enable entry into the food chain. C. difficile is a potent carrier of antimicrobial resistance genes and some strains are multidrug resistant where this resistance can easily transfer to other strains and species. C. difficile also is adept at evolving into more virulent strains that are noticeably more infectious. Taken together C. difficile presents a major problem for public health and there is currently no prophylactic treatment. Treatment of CDI is by use of antibiotics which is a particular problem for a pathogen that is extremely capable of acquiring and developing resistance to antimicrobials. Good hygiene practiced in hospitals has been shown to halt the increase in CDI but what is needed is a vaccine. It should be noted that vaccines under development are all injectable and do not prevent colonisation of the host by C. difficile but rather, only neutralise toxins and alleviate symptoms of disease.

The CDVAX consortium has adopted a novel oral vaccine consisting of recombinant spores of the bacterium Bacillus subtilis that have been engineered to display an antigen (TcdA26-39) which is part of C. difficile toxin A. Oral immunisation with spores displaying this antigen have been shown to protect hamsters from oral challenge with virulent C. difficile and provide the impetus for a phase 1/2a clinical trial. The CDVAX vaccine is unique and unlike other injectable vaccines since it prevents colonisation of the host by the pathogen as well as neutralising toxins and has obvious long-term utility as a best-in-class vaccine for CDI.

The consortium comprised 5 partners. Royal Holloway University of London the project coordinator and trial sponsor. Leads 2 Development a French company who were responsible for the regulatory steps, CRIS Pharma who conducted the bioanalytical work, Q-Biologicals (now Amatsi Q-Biologicals, Belgium) who performed the cGMP production of the vaccine and FGK (Germany) who were charged with the clinical studies.

The project carries 4 RTD workpackages. WP1 lead by RHUL focused on the optimisation of the B. subtilis spore vaccine, termed PP108. Almost all aspects of the vaccine were considered including how best to inactivate the recombinant using formalin and the optimal dose to use in a vaccination program. Mice and hamsters were used extensively for analysis of protection and immune responses. Ultimately, our work showed that the PP108 vaccine has reduced immunogenicity when inactivated with formalin and the live vaccine was reproducibly immunogenic and conferred levels of protection of about 75%. Immunity correlated with the production of mucosal antibodies (sIgA)which prevented colonisation. This is the first evidence that mucosal antibodies could prevent colonisation. Mucosal antibodies were shown to block adhesion of C. difficile spores to mucus and the findings published in Infection & Immunity (Hong et al, 2017: in press). The identity of the protein targets that are recognised by anti-TcdA26-39 antibodies formed the basis of a patent filed by RHUL in the UK in April 2016. This patent provides IPR relating to how to protect against CDI as well identifying new vaccine candidates.

WP2 lead by CRIS Pharma was focused on the development of bioanalytical techniques for the clinical study and has completed all of its assigned tasks. WP3 lead by Q-Biologicals was dedicated to the production of cGMP PP108 spores. This was achieved and methods for process development and scale up were validated extensively. Master cell banks were made and the final CDVAX vaccine prepared for the clinical studies.

The clinical study comprising WP4 is lead by FGK. The project had entered difficulties that made it unlikely that the project could complete by June 31st 2016. The two key reasons were:
(i) problems encountered with the GMP production and a delay of a disposable fermentor.
(ii) A required change in the classification of the CDVAX vaccine to a live GMO.
The project requested and was granted a 6-month extension with completion Nov 31st 2016. Subsequently we have experienced further difficulties with the ethical committee (EC) review and then the Paul Ehrlich Institute (PEI). Initially we received approval from the EC and then conditional approval from the PEI. However, we subsequently received further questions from the EC which then required referral back to the PEI. Since the PEI had 27 days to respond this introduced yet further delays and we did not receive final PEI approval till January 2017. The study is listed on clinicaltrials.gov with the identifier NCT02991417.

However, post project the vaccine was taken to clinic and administered to a human volunteer.

Project Results:
WP1: OPTIMISATION AND CHARACTERISATION OF CDVAX SPORES (RHUL)

Objectives:

The objective of this WP is to develop and optimise the CD-VAX vaccine and to undertake in vitro and in vivo pharmacological evaluation of B. subtilis spores expressing C. difficile TcdA26-39 termed here as CD-VAX.

SO1.1: To develop the methodologies for formalin inactivation of spores.
SO1.2: To characterise the in vivo efficacy of CD-VAX to provide protection against primary CDI in rodent models.
SO1.3: To determine the optimal oral dosing regimen in rodents that provides protective immunity.
SO1.4: To determine the utility of the spore vaccine in protecting against a range of highly virulent multidrug resistant strains of C. difficile (including hypervirulent strains).
SO1.5: To characterise the in vivo efficacy of CDVAX to provide protection against CDI relapse
SO1.6: To characterise the humoral and cellular immune responses induced by CDVAX.
SO1.7: To determine the memory of mucosal immune responses following oral dosing of vaccine.

Description of work completed:

The CDVAX vaccine is a recombinant spore of Bacillus subtilis that expresses the TcdA26-39 antigen on the spore surface. TcdA26-39 is itself a C-terminal segment of C. difficile toxin A. As a recombinant our first objective was to determine whether we could inactivate the spore such that it did not provide any risk to the environment upon release. Vaccines are normally inactivated using formalin (formaldehyde) so we used this approach first. By empirical experimentation we identified a strategy for successfully killing all spores of CDVAX. However, although spores were killed the spores had reduced immunogenicity and in a hamster model of infection failed to provide levels of protection (~17%) equivalent to the live CDVAX spore (75%). Therefore, the consortium decided to progress with a live recombinant vaccine for clinical evaluation. It must be emphasised that vaccine companies will often spend years developing optimal inactivation strategies and we do not consider this initial failure with formalin to be conclusive. The fact that we obtained some level of protection implies that the method for inactivation might be further refined. Alternative methods for inactivation such as the use of γ-irradiation were considered but we are unable to find a suitable of willing collaborator able to perform a small-scale inactivation.

Oral Immunization with PP108 Spores Prevents Colonization of C. difficile
Spores of CDVAX were produced to cGMP standards and used to dose hamsters either in the live form or following formaldehyde-inactivation. The dosing strategy involved a combined sub-lingual-oral (s.o.) administration of four cycles after which animals were treated with clindamycin to purge the intestinal microbiota before challenge (i.g.) with 100 spores of C. difficile strain 630. Control groups included animals dosed s.o. with live B. subtilis PY79 spores (identical to PP108 but not presenting the TcdA26-39 antigen), intramuscular (i.m.) with rTcdA26-39 protein and naïve. All naïve animals and those parenterally immunized with rTcdA26-39 succumbed to CDI and died within 50h (Fig. 1.1A). For animals dosed with live PP108 spores, 67% were protected. For animals dosed with formaldehyde-inactivated PP108 spores, 17% were protected and for those dosed with live PY79 spores, 33% survived. Using s.o. delivery we have obtained similar levels of protection using non-cGMP PP108 spores in two independent trial experiments (data not shown).

In surviving animals, we showed complete protection to recurrence (Fig. 1.3A) in agreement with our previous study (Permpoonpattana et al., 2011). Anti-TcdA26-39 IgG responses in serum were measured (Fig. 1.1B) showing that although high IgG titres were induced in animals dosed orally with PP108 and parenterally with rTcdA26-39 protection was restricted to animals immunized with PP108. The formaldehyde-inactivated PP108 spores gave low IgG titres indicating that the inactivation method impaired the immunogenicity of the expressed antigen correlating with the low level of protection observed. As would be expected, PY79-dosed animals and naives produced no IgG response. Counts of C. difficile spores (CFU) in feces 24h post-challenge showed that in all protected animals no viable C. difficile was detectable indicating complete pathogen clearance. In contrast, high counts were present in all non-protected animals (Fig. 1.1C). The levels of toxin A and toxin B in feces of protected (no toxins) and non-protected (toxins present) animals were determined (Fig. 1.1D-E and Fig. 1.3) and confirmed using cytotoxicity assays (data not shown). This analysis together with bacteriological assessments (Fig. 1.1C) demonstrated the absence of infection (toxins and CFU), in protected animals. Similarly, for protected animals in which recurrence was induced, no toxins were present during days 3-15 following reinfection (Fig. 1.3B-C).

To measure SIgA in hamster feces, we used a goat anti-mouse secondary antibody to detect the heavy chain of IgA in both mouse and hamster fecal samples. Curiously, using this assay we did not detect SIgA in feces taken seven days after the final immunization (data not shown). In an earlier unpublished study however we were successful in detecting fecal IgA 14 days after the final immunization of PP108 spores. These data are shown in Fig. 1.1F and illustrate that in hamsters, protection against CDI correlates with high levels of SIgA and that sampling at day 14 may be preferable. To show that mucosal IgA correlates with protection, we also immunized CD-1 mice with cGMP compliant PP108 spores (s.o.: live and inactivated) together with rTcdA26-39 (parenteral). We then showed that vaccination with live PP108 spores induced neutralizing SIgA in feces as well as neutralizing serum IgG specific to TcdA26-39 in contrast to injection of rTcdA26-39 that induced neutralizing serum IgG only (Fig. 1.4).

This study demonstrates firstly, that protection to CDI correlates with a failure to colonize the host together with the production of mucosal antibodies rather than serum IgG. Secondly, spores of B. subtilis that do not express a C. difficile antigen can provide a low level of protection to CDI.

Parenteral Vaccination with Toxoids Does Not Prevent Colonization
Hamsters were given three i.m. injections of toxoids A and B (plus adjuvant) and, together with a naïve group, then challenged (i.g.) with 100 spores of C. difficile strain 630. Compared to naïve animals, symptoms of CDI were delayed but all animals ultimately succumbed to infection (Fig. 1.5A). In toxoid-immunised animals serum IgG levels specific to toxins A and B were high (Fig. 1.5B) but counts of C. difficile spores in feces 24h post-challenge demonstrated that animals were fully colonized with C. difficile (Fig. 1.5C) at levels indistinguishable from those in naïve animals. Consistent with these observations, toxins A and B were readily detectable in cecum samples of sacrificed animals (Fig. 1.5D-E).

Defining the Mechanism of Prevention of Colonisation
In the course of this project we have discovered that antibodies recognising TcdA26-39 also recognised a number of other C. difficile proteins (Figure 1.6). These were LdhA (lactate dehydrogenase), AdhE1 (alcohol aldehyde dehydrogenase) and CdeC. CdeC is a spore coat protein preent in the exosporium while LdhA and AdhE1 are proteins found in the vegetative cell, probably surface exposed and for LdhA also found in the exosporium. In the case of AdhE1 this protein is also present in the spore exosporium. We have shown that anti-TcdA26-39 antibodies reduce adhesion of spores and veg cells to mucus and we believe mucosal antibodies raised to TcdA26-39 could prevent adhesion and therefore colonisation of C. difficile in the GI-tract. This finding is important since it identifies additional antigens as potential vaccine candidates but also defines a mechanism. The identity of these antigens have been protected through a UK patent application as of April 2016.

Protection to Hypervirulent Strains
We have observed in mice good levels of protection to hypervirulent strain R20291 Irobotype) as well as VPI 10463 (ribotype 087). This is good and shows that the vaccine has the potential to provide protection to more than one variety of C. difficile.

Dosing Regimens
A dosing regimen has been defined empirically in mice and then hamsters. This is described in Deliverable 1.3. In brief multiple doses (12 total over 4 cycles) was shown to be optimal for vaccination. This is consistent with other oral vaccines. We also have identified a combined sub-lingual/oral dosing regimen that works well in hamsters and will be used in humans.

WP1 has produced:

Deliverables 1.1-1.5
One publication in press with Infection & Immunity
One publication under review by PLOS Pathogens
One patent (filed UK April 17th 2016)


WP2: DEVELOPMENT OF ANALYTICAL AND PHARMACOLOGICAL TOOLS (CRIS)

Objectives:

The objective of this WP was to establish and validate the analytical and bioanalytical methods required to characterise the CDVAX product (including its stability, biodistribution, efficacy and immunogenicity) that were used during both the preclinical and clinical phases of this research programme. The specific objectives were established on five levels:

SO2.1: Validation of methods for characterisation of the recombinant B. subtilis bacterial strain
SO2.2: Development and validation of analytical methods
SO2.3: Development and validation of bioanalytical and immune-monitoring methodologies
SO2.4: Development of a quantitative CDVAX potency test
SO2.5: Validation of immune-monitoring methodologies for the clinical studies

Description of work completed:

Task 2.1: Characterisation of the recombinant B. subtilis bacterial strain

The B. subtilis PP108 strain expressing TcdA26-39 on the spores was transferred from Partner 1 to Partner 4. During the generation of B. subtilis PP108, it cannot be excluded that growth media containing components derived from animal origin were used. In order to minimize/exclude the transmission thereof, a clone dilution experiment was performed before banking.

The strain was plated out as single colonies on LB agar containing chloramphenicol and erythromycin at 37°C. One colony was picked up, dissolved in LB medium with antibiotics and plated out on LB agar plates supplemented with antibiotics. This was repeated twice.

A single colony was then picked up, grown in LB medium supplemented with antibiotics at 37°C to prepare the Research Cell Bank (RCB). After 8 hours incubation, glycerol was added to the culture to a final concentration of 15%. One ml of culture was aliquoted in each cryovial. The RCB was composed of 50 individual cryovials stored at -70°C.

From the characterized RCB, a GMP compliant Master Cell Bank (MCB, 310 vials) was produced by Partner 4.

In the process, one component from animal origin was used (BactoTM tryptone). Corresponding TSE/BSE free certificates of analysis and of origin are available, as well as for BactoTM yeast extract.

Analytical methods were developed and validated by Partner 4 that were used to characterise the MCB:

- Visual inspection by Gram staining and microscopic evaluation
- Culture selective plating
- Presence of TcdA26-39 by western blot
- B. subtilis identity by partial sequencing
- Viability by agar plate counting before freezing
- Viability by agar plate counting after freezing
- Turbidity (European pharmacopoeia (EP) method 2.2.25)
- % vegetative cells before freezing by agar plate counting before and after heat treatment
- Microbial purity: culture in TSA medium, culture in SDA medium, bile tolerant Gram negative organisms, E. coli, S. aureus and P. aeruginosa
- Inserts identity: sequencing of cotB flanking region and TcdA26-39 antigen, sequencing of cotC flanking region and TcdA26-39 antigen

Task 2.2: Development and validation of analytical methods

Several analytical procedures were developed and validated by Partner 4 that were used to test and release the CDVAX drug substance and the Investigational Medicine Product (IMP):

- For the drug substance
o pH by EP method 2.2.3
o B. subtilis identity by partial sequencing
o Appearance by microscopic inspection of spores
o Spore identity by culture selective plating by replica plating on LB agar
o Presence of TcdA26-39 by western immunoblotting
o Quantity of spores by microscopic counting

- For the drug product
o Appearance by EP method 2.2.1
o pH by EP method 2.2.3
o Homogeneity by visual inspection
o B. subtilis identity by partial sequencing
o Appearance of spores by microscopic inspection
o Spore identity by culture selective plating by replica plating on LB agar
o Presence of TcdA26-39 by western immunoblotting
o Quantity of spores by microscopic counting
o Quantity of live spores by agar plate count method
o Uniformity of dosage units by EP method 2.9.40
o Microbial purity: culture in TSA medium, culture in SDA medium, bile tolerant Gram negative organisms, E. coli, S. aureus and P. aeruginosa
o Innocuity by mouse weight gain assay
o Potency test by presence of TcdA26-39 on spores (ELISA)
o
The appearance of the spores was controlled by microscopic inspection. A dilution series of the CDVAX drug substance was prepared in 0.9% NaCl, and a few µl of the diluted solution were placed into a Neubauer improved counting chamber (Petroff-Hausser). The spore appearance was controlled under a phase contrast microscope using a 1000-fold magnification. The quantity of spores was determined by microscopic counting at the same time and on the same sample.

The homogeneity was assessed by visual inspection of the suspension in clear tubes. No aggregation should be visible after re-suspension.

The identity of spores was verified by culture selective plating using replica plating on LB agar, with or without chloramphenicol and erythromycin. A hundred single colonies of a master plate derived from the agar plate counting method were individually picked and transferred to a LB agar plate without antibiotics and then to a LB agar plate with antibiotics (chloramphenicol and erythromycin) in a defined square. Two negative controls were included on each plate. The plates were incubated at 37°C for 18 to 48 hours. Colony-forming units (CFU) were counted and the percentage of antibiotic resistant colonies was calculated based on the number of colonies on the plate with antibiotics compared to the number of colonies on the plate without antibiotics.

The presence of TcdA26-39 antigens was controlled by western immunoblotting. Spore suspensions (2.0x108 spores) were centrifuged for 5 minutes at 21,000 g. Supernatant was removed and pellets were re-suspended in sample buffer with reducing agent. The samples were then heated for 10 minutes at 95°C and centrifuged. The supernatants (20 µl, 1.0x108 spores) were loaded on a 4-12% Bis/Tris gel, with MES SDS running buffer. Four different quantities of recombinant TcdA26-39-his6 were loaded as a reference. Following electrophoresis, the gel was transferred to a nitrocellulose membrane and immunostained with a polyclonal rabbit anti-TcdA antibody and a polyclonal swine anti rabbit IgG secondary tagged with alkaline phosphatase. The antigen-antibody immune complexes were revealed with BCIP-NBT. For the reference material TcdA26-39, one major band was visible at 37kDa. For the CDVAX samples, two protein bands were detected at 50 kDa (CotC-TcdA) and 66 kDa (CotB-TcdA).

The quantity of the live spores in the CDVAX drug product was determined by plating on LB agar. Three independent dilution series per sample were prepared, three dilutions were plated in triplicate (100 µl/plate) and incubated for 18 to 48 hours at 37°C. CFU were counted after incubation and a visual inspection if the culture was performed.
The uniformity of dosage unites of the drug product was determined according to the EP method 2.9.40.

Task 2.3: Development and validation of bioanalytical and immune-monitoring methodologies

Immunological assays were developed and validated by Partner 3 to characterize both the mucosal and systemic humoral response in mice.

A commercial kit was available to detect anti-toxin A from Clostridium difficile mice IgG but the sensitivity was not sufficient enough compared to the in house method developed to detect anti-TcdA mice IgG. For both methods, no commercial standards exist, as mouse IgA or mouse IgG against the TcdA recombinant protein; no standard curve could be therefore realized for an absolute quantitation of positive response.

Three ELISA methods were developed and validated to allow the detection of anti-TcdA26-39 IgA and IgG antibodies in serum samples and anti-TcdA IgA antibody detection in faecal samples in mice. Validation was performed using positive control serum and positive control faecal extract. These positive samples are a pool of samples from several mice immunized with recombinant TcdA protein and cholera toxin. Serum pool gives positive response for IgA and IgG directed against TcdA recombinant protein and faecal extract sample gives positive response for mouse specific IgA. Negative control pool was also constituted with samples from naïve mice.

ELISA methods are based on the following protocol: plates were coated with TcdA and detection of bound antigen-specific IgA or IgG from samples (tested at several dilutions) were done using HRP-conjugated anti-mouse IgA or IgG antibodies. Absorbencies measured are linked to the specific antibodies amount in serum or faecal samples. Methods were validated to demonstrate their reliability for the determination of specific immunoglobulins in serum or faeces samples by determining the following parameters: specificity, precision and linearity.

Positive and negative controls were used to test the specificity of methods, their ability to measure specific immunoglobulins in the presence of other components present in matrix.
Several serum dilutions of serum samples from 1/1000 to 1/128000 were used to define the linearity of the response of IgG level in serum (ability to obtain results which are directly proportional to the amount in the sample). The coefficient of determination was 0.985 until dilution 1/64000 and is within specifications.

The precision of methods was assessed using 4 dilutions of positive samples freshly prepared tested in 3 runs conducted over several days. For each run, 5 replicates per dilution were analysed. The precision around the mean value was expressed by the coefficient of variation (CV) calculated for each dilution. The intra-assay CV evaluates the variability of 5 replicates of samples whereas the inter-assay CV evaluates the variability of QC samples over 3 experiments. CV values should not exceed 20 % on a same run. The precision of the analytical method to detect specific IgA in faeces was successfully validated. Intra- and inter-assay precisions were within specifications for samples dilution ranged from 1/5 to 1/40. The precision of the analytical method to detect specific IgA in serum was successfully validated. Intra- and inter-assay precisions were within specifications for samples dilution ranged from 1/250 to 1/2000. The precision of the analytical method to detect specific IgG in serum was successfully validated. Intra- and inter-assay precisions were within specifications for samples dilution ranged from 1/5000 to 1/32000.

For the toxicology study, positive and negative controls used in this validation were used as controls in each run performed. Mucosal and systemic humoral response could be characterized by relative quantitation based on dilution series of samples (comparison between samples based on dilutions series).

In pharmacology and toxicology studies, CDVAX did not exhibit any toxicity after oral administration. The biodistribution study consisted of a dissemination study of CDVAX from the mice. It was assessed by measuring CDVAX elimination in excreta (faeces and buccal secretions) of CD-1 mice after a single oral administration.

The presence and the quantification of CDVAX spores were determined in faeces and in buccal secretions over a 4-week period. The method was validated by spiking both matrices (faeces and buccal secretions) with two amounts of CDVAX spores. Experiments were repeated 5 times to assess the precision of method and the intra- and inter-assay precisions were determined.

The stability of spores in human stools was also evaluated for the clinical study. Spiked human faeces with CDVAX spores were stored at 4°C up to seven days and were plated on DSM agar plates with antibiotics (100 µl per plate) and incubated for 48 to 72 hours at 37°C. The mean of CFU counts for each incubation condition will be compared to the mean of CFU counts for the time T0 and will allow determine the stability of CDVAX spores in human faeces. CDVAX spores were stable in human faeces up to seven days when stored at 4°C.

Task 2.4: Development of a quantitative CDVAX potency test

A potency test was developed by Partner 3 to be used as a drug substance and drug product specification for batch release and stability studies. A potency test should consist of either an in vitro or in vivo test designed to quantify specific biological activity. As the CDVAX product consists of an antigenic domain displayed at the surface of the delivery vehicle (spore), its biological properties is to induce a humoral immune response targeted against the TcdA26-39 peptidic sequence. However, at least two vaccinations cycles are required to induce a measurable immune response. Thus, the most appropriate potency test to measure the active ingredient of the product is a quantitative TcdA26-39 ELISA method, which determines the quantity of TcdA26-39 per unit quantity of spores.

Quantification of TcdA is performed by means of a standard curve constituted with 8 concentrations of TcdA recombinant protein. Plates were coated with TcdA recombinant protein and detection of bound antigen-specific anti-toxin A antibody (from goats) were done using HRP-conjugated anti-goat antibodies. Absorbencies measured are linked to the quantities of TcdA coated. Method was validated in terms of accuracy, precision, linearity, limit of quantification and range of quantification.

Validation steps were assessed using calibration curves and 3 QC samples freshly prepared and performed in 6 runs conducted over several days.

The range of quantification was determined for each ELISA Assay and The Lower Limit Of Quantification (LLOQ, lowest amount of TcdA recombinant protein coated which can be quantitatively determined with suitable precision and accuracy) is 10 ng/ml and the Upper Limit Of Quantification (ULOQ, the highest amount of TcdA recombinant protein coated which can be quantitatively determined with suitable precision and accuracy) is 1280 ng/ml.

The accuracy was reported as the percentage of total error (bias) obtained for each QC. The mean relative bias of each QC was comprised between -30 and 30% and is within specifications.

The precision intra-assay around the mean value was expressed by the coefficient of variation (CV) calculated for each QC and was within specifications (comprised between 8.54 and 11.90%, which is below the accepted cut off of 20%). The inter-assay CV evaluates the variability of QC samples over 6 experiments and was within specifications (comprised between 15.01 and 19.58%, which is below the accepted cut off of 20%).

An ELISA method has been developed and validated to measure the quantity of TcdA per unit quantity of spores to test and release the CDAVX drug substance and drug product.

Task 2.5: Validation of immune-monitoring methodologies for the clinical study

An immune-monitoring method for human use was developed by Partner 5 to analyse the healthy volunteers samples in the clinical study for vaccination-induced TcdA-specific antibodies (specifically antigen-specific IgA in faeces and antigen specific IgG and IgA in serum).

For this analysis, sensitivity and other performance parameters of the method was compared between a commercially available ELISA Kit (tgcbiomics assays no.: TGC-E101) and a Ligand Binding Assay on the MesoScale platform. The commercial ELISA was performed according to the manufacturer’s protocol using the materials and reagents supplied with the kit and the in house MSD assay based on the following strategy: MSD plates were coated with TcdA and detection of bound antigen-specific IgA or IgG were done using sulfo-tagged anti-human IgA and anti-human IgG antibodies respectively. For both assays positive controls used to test the sensitivity consist of a pool of human serum from healthy volunteer blood donations and a supernatant of faeces extract spiked separately with recombinant IgA and IgG specific for C.difficile toxin A. Assay sensitivity was judged by using serial dilutions and comparing detection limits and signal strength for each assay.

The results showed that the assay of tgcBiomics is suitable for human use. A good titration row was seen with the control antibodies (IgG and IgA anti-TcdA). Spiked samples in matrix (IgG in serum and IgA in stool samples) result in signals which are considered to be sufficient for the detection of antibodies after vaccination. Only a semi-quantification is possible as no material is included in the kit for a standard curve. Relative comparison between samples based on the dilution series will be used during sample analysis.
During screening of subjects and as requested in the clinical protocol, detection of anti-TcdA antibodies (IgG and IgA) in human serum and detection of anti-TcdA antibodies (IgA) in stool samples were performed to determine evidence of C.difficile infection and Anti-C.difficile (Toxin A) immunoreactivitiy, suggesting previous C.difficile exposure.

The commercial kit from tgc biomics was used for this screening and in addition 6-10 blood and stool samples from healthy volunteer were included on the assay for the determination of an assay cut point.

Conclusions:
A MCB of B. subtilis PP108 strain free from animal derived material was generated and analytical methods were developed to characterise and release the MCB as well as the CDVAX drug substance and drug product. A potency test consisting of an ELISA method was also developed and validated to measure the quantity of TcdA per unit quantity of spores to test and release the CDVAX drug substance and drug product. Before preclinical studies, immunological assays were validated to characterise both the mucosal and systemic humoral response in mice. For the dissemination study, the measurement of CDVAX in mouse excreta (faeces and buccal secretions) was validated. These methods had to be adapted for clinical trial. Immune-monitoring methodologies were adapted for human samples and the stability of CDVAX spores in human stool was assessed.


WP3: PRECLINICAL DEVELOPMENT (L2D)

Objectives:

The primary objective of WP3 was to undertake and complete preclinical development of the CDVAX vaccine; including GMP production of the clinical trial material, a formal GLP-compliant toxicology study, preparation of a regulatory dossier and submission of the phase I clinical trial application. The specific objectives were established on five levels:

SO3.1: Manufacturing of technical and clinical (GMP compliant) batches of CDVAX
SO3.2: Formulation of the CDVAX product and formulation of the administered dosage form
SO3.3: Complete efficacy, toxicology and in vivo pharmacology studies
SO3.4: Undertake CDVAX stability testing
SO3.5: Obtain scientific advice for this novel vaccination approach from both the national (PEI) and international (EMA) competent authorities
SO3.6: Complete regulatory activities together with submission of the clinical trial application


Description of work completed:

Task 3.1: Manufacturing

A research cell bank (RCB, 50 vials) was generated by Partner 4 from a single colony of the recombinant strain PP108 supplied by Partner 1. Partner 4 characterized the expression of the two antigen expressing constructs integrated in strain PP108. Partner 4 used the RCB to undertake initial scale up process development activities (see below). From the characterized RCB a cGMP-compliant Master Cell Bank (MCB, 310 vials) was created. The quality of the cell bank is of paramount importance as this MCB will be used throughout the entire life of the CDVAX product, both prior to and post-market authorization. The GMP compliant MCB was characterized and released according to the specifications.

Within the research setting, sporulation of bacteria had been induced on solid agar plates, a methodology that is not suitable for scale up. Therefore it was necessary to optimize biomass production and sporulation. This was evaluated for different growth parameters, using growth on solid and liquid medium where the latter proved more robust and scalable using a bioreactor. Firstly, the effects of nutrients, heat and pH on growth conditions were evaluated and optimal growth conditions (biomass production) defined. Secondly, sporulation conditions were optimized to increase spore yield and the robustness of spore production. One of the major problems encountered during sporulation was the effect of bacterial specific proteases on the stability of the CDVAX antigens. This was unexpected and has taken considerable time to resolve. It is now considered by the consortium that a robust working method for CDVAX spore production has been achieved. It is also important to note that the problems encountered and solved as part of manufacturing process optimization are proprietary, patentable and represent a significant advancement in the state of the art.

The optimized protocols were transferred to the cGMP production unit and two batches of drug substance were produced at the current 45 L processing scale. A first, non-GMP batch of drug substance was produced and characterized (batch n°100374). This batch was used for analytical work, stability studies (including drug product) and a preliminary toxicology study. A second, GMP batch of drug substance (batch n°100817) was produced for stability studies (including drug product) and the GLP toxicology study, and used to manufacture the IMP. This batch was characterized and released according to the specifications. The analytical results obtained show that, globally, the drug substance used to manufacture the clinical trial material is of appropriate quality.

Three batches of drug product (non-GMP and GMP) were produced and characterized. The two non-GMP batches n°100905 and n° 100903 were produced for the main purpose of generating initial long-term stability data. The third batch (batch n°101062), manufactured in compliance with GMP, was the clinical trial material used in the first-in-human clinical study. This batch was QP-released according to the specifications.

Task 3.2: Formulation

Formulation for the phase 1 clinical study is a liquid formulation similar to that already adopted for in vivo experiments. The CDVAX IMP batch used in the clinical trial (batch n° 101062) consisted of 480 cryotubes filled with 1 ml of a 4x109 spores/ml suspension of the CDVAX spores in 0.9% NaCl. A total of 900 cryotubes were filled, of which 50 were used for weight checks during the process, and 370 were reserved for quality control, stability studies and as retain samples.

The finished product was packaged and stored in sterile, polypropylene cryotubes closed with sterile, polypropylene screw caps. The cryotubes and caps were CE marked according to EU Directive on medical devices. The IMP was labelled according to the Good Clinical Practice, GMP and in compliance with the legal requirements of the country. As the clinical trial was conducted in Germany, the label texts were in accordance with German regulatory requirements.

Task 3.3: In vivo toxicology and pharmacology studies

A regulatory compliant study was undertaken by Partner 3 to characterize the safety of CDVAX in an appropriate animal model prior to the launch of the clinical study. In this GLP toxicology study, the safety and immunogenicity of CDVAX after oral administration of the clinical dose and by the clinical route (oral route) was assessed in the mouse. No in vivo safety pharmacology study was required.

Five oral immunisation cycles of CDVAX at 2x1010 spores/dose (the clinical dose) was well tolerated in the CD-1 mouse. CDVAX treatment did not induce any changes in the observed parameters. CDVAX was detected in the faecal samples of mice collected after the first, third and fifth administrations, confirming that treatment was correctly administered. No difference in anti-TcdA26-39 systemic IgG, systemic IgA or mucosal IgA between the CDVAX-treated and vehicle groups were detected in this study, most likely as a result of a technical issue. An increase of the white blood cell count was reported at D60 for males and at D100 for both genders in the CDVAX-treated group compared to the control group, but these values remained within the normal range. No gender effect was observed.

Task 3.4: Stability testing

Real-time and stressed stability testing of CDVAX was initiated by Partner 4 to define the shelf-life (expiry date) of the clinical trial material.

Long-term drug substance stability studies were initiated using the non-GMP CDVAX drug substance batch n°100374 and the GMP drug substance batch n°100817. The objective of these studies was to obtain information on the stability of CDVAX in actual current storage conditions. For these studies, the drug substance was filled in sterile polypropylene tubes equipped with high-density polyethylene screw caps, of the same type as the tubes used for storage of the bulk drug substance. The stability samples were stored in accordance with ICH guidelines, and tested versus the current specifications. Upon storage between -18°C and -25°C, results of both batches have so far remained within specifications. The product is stored between -18°C and -25°C, with a shelf life that is currently set to 12 months.

Stress testing studies were also performed with the three batches of CDVAX drug product produced so far using the same manufacturing process. Stress testing studies were performed at 5 ± 3°C for 16-24 hours and 160-168 hours with the non-GMP drug product batches n°100903 and n°100905, and the GMP batch used in the clinical study (batch n°101062). The stress testing samples were stored in adequate, monitored cold rooms complying with ICH guidelines, and tested according to the stability plan described in the IMPD. Overall the results confirmed the stability of the CDVAX drug product at least up to 160-168 hours at 5 ± 3°C.

Stability studies were initiated with the three batches of CDVAX drug product produced so far using the same manufacturing process. A first long-term stability study was undertaken with the non-GMP drug product batches n°100903 and n°100905 in the current storage conditions (-18°C to -25°C). The stability samples were stored in adequate, monitored cold rooms complying with ICH guidelines, and tested according to the stability plan described in the IMPD. Stability studies were also initiated in the same conditions with the IMP batch used in the clinical study (GMP batch n°101062). Upon storage at -18°C to -25°C, results have so far remained within specifications. The CDVAX drug product is considered to be stable for at least 6 months at -18°C to -25°C. Based on these results the shelf life of the CDVAX drug product, stored at -18°C to -25°C, is currently set at 6 months.

Task 3.5: Obtain scientific advice from competent authorities

A written scientific advice request was submitted to the EMA by Partner 2 to define the classification of the CDVAX product. This was required as the B. subtilis bacterial strain used to obtain the CDVAX spores is genetically modified to express a C. difficile antigenic sequence (toxin A, TcdA). Two antibiotic resistance markers (chloramphenicol resistance: CmR and erythromycin resistance: ErmR) are also present within the strain in order to select for the recombinant events. As such the starting material is a GMO for which specific guidance exists. However, the initial CDVAX product aimed to be an inactivated (killed) spore expressing the C. difficile CDTA antigen on its surface. This product would have had no replicative capacity and the fixation process used to kill the spores would also have resulted in DNA damage, further preventing DNA replication or DNA transfer between organisms. The EMA reviewed the classification request and confirmed that the initial CDVAX product was not an Advanced Therapy Medicinal Product (ATMP) as it should have contained no live cells and because the product will be used as a prophylactic vaccine against an infectious disease causing bacteria.

Additional scientific advice was requested by Partner 2 from the German National Competent Authority, the Paul Ehrlich institute (PEI). This was required to review and validate the regulatory aspects of CDVAX production, toxicology and the first-in-human clinical study that took place in Germany. This process was also designed to introduce the PEI to the programme in order to ensure that all regulatory aspects were considered in good time and to prevent any delays in obtaining clinical trial authorisation. To this end, a briefing document was generated, including the phase I clinical trial synopsis, a monograph for the product defining specifications of the CDVAX product together with quality and release criteria, and a toxicology/immunogenicity plan. This document used as the basis of a face-to-face meeting with the German regulators to review, discuss and define the regulatory requirements for the CDVAX product. A follow up teleconference meeting was conducted afterwards.

Task 3.6: Regulatory activities together with generation and submission of the clinical trial dossier

The objective of this programme was to develop a vaccine for human use. As such the quality of both clinical trial material and the non-clinical data had to be sufficiently high to ensure the safety of the healthy volunteers in the phase I clinical trial, together with patients who will receive the CDVAX product in later phase II and III studies. Strict regulatory guidance (from both the EMA and FDA) had to be followed. The regulatory activities were therefore associated with a number of other tasks including scientific advice and manufacturing, and aimed to ensure robust and reproducible production methods and a well-characterized CDVAX product batch at sufficient quality for clinical evaluation.

The clinical trial application dossier was generated by Partner 2. This dossier contained a list of required documents and was submitted to the competent authorities:

- A cover letter
- An EudraCT number and application form
- The clinical protocol, the informed patient consent and insurance
- The Investigator’s Brochure
- The Investigational Medicinal Product dossier
- The certificate of analysis of the clinical IMP batch and the manufacturing authorisation of Partner 4
- The texts for labelling of the IMP
- A declaration on data protection
- The scientific advice briefing documents and meeting minutes
- Annex IIIa, environmental risk assessment and SNIF application form
- A GLP study director statement from Partner 3

Conclusions:
All deliverables and milestones of WP3 were achieved and all tasks were completed. The preclinical development of the CDVAX vaccine (including GMP production of the clinical trial material, formal GLP-compliant toxicology study, preparation of a regulatory dossier and submission of the phase I clinical trial application) was completed.


WP4: PHASE 1 CLINICAL EVALUATION OF CDVAX (L2D)

Objectives:

Key Objective
The primary objective of WP4 is to perform early clinical (“first-in-man‟, phase I) evaluation of the CD-VAX vaccine in healthy volunteers in order to characterise the vaccine’s safety and immunogenicity. RHUL will be the sponsor of this study.

Specific objectives
SO4.1: Undertake first-in-man studies to evaluate the safety and tolerance of CD-VAX in healthy volunteers.
SO4.2: Characterise the clinical mucosal immune response induced by CD-VAX.
SO4.3: Characterise the clinical systemic immunity induced by CD-VAX (both cellular and humoral).


Description of the work completed:
The clinical trial material has been generated in WP3 and the methods to perform the immuno-monitoring of the clinical study have been developed and validated in WP2. The clinical trial was to be conducted by FGK in connection with the clinical site of the Phase I Unit of Nuvisan, Neu-Ulm.

Task 4.1: Generation and validation of the clinical trial protocol
A synopsis of the planned clinical study was written by a medical writer of FGK for the Scientific Advise Meeting at PEI in autumn 2015. This meeting was also attended by the Project-Coordinator/Medical Monitor of FGK. After the meeting the synopsis was adapted to the wishes of PEI.

The clinical protocol was developed based on the preclinical characterisation of the CD-VAX product (WP1, 2 & 3). Therefore, a draft clinical trial protocol was written during December 2015 and January 2016 by FGK’s medical writer. The review and finalization of the clinical protocol was performed subsequently until May 2016. Based on the final protocol the patient information sheet and informed consent form was prepared by FGK’s subcontracted Phase I unit at Nuvisan located in Neu-Ulm, Germany. The application dossier for getting the favourable opinion from the responsible ethics committee was prepared and submitted in May 2016 by the Phase I Unit Nuvisan. Since the application to the ethics committee and also the application to the competent authority (Paul Ehrlich Institute, PEI) was not accepted without questions and conditions, FGK participated at several teleconferences from July until December 2016 to discuss how to modify the clinical trial documents to meet the requirements of the ethics committee and the regulatory authority. FGK’s project-coordinator, medical writer and project manager - together with the Phase I unit at Nuvisan – modified the protocol and the patient information sheet and informed consent form numerous times during the multi-step discussion process with the ethics committee and the competent authority. The final approval of the ethics committee was obtained early January 2017. The approval by the PEI is still pending (January 17th, 2017).

PROTOCOL SYNOPSIS

TITLE Safety and immunogenicity study of a Clostridium difficile vaccine in healthy adult volunteers

SPONSOR Royal Holloway, University of London. Egham Hill, Egham, Surrey, UK. TW20 0EX

CLINICAL PHASE Phase I

INDICATION This study will be performed in healthy adult volunteers.

OBJECTIVES Primary objective: Safety evaluation of CDVAX including the nature, frequency and severity of treatment-specific adverse events post-vaccination.
Secondary objectives: Monitoring of both mucosal and systemic immune response. To characterise different immunisation schedules in order to determine the optimal schedule

STUDY DESIGN Single-centre, safety and immunogenicity study.
Subjects will be allocated to one cohort consisting of 24 subjects receiving CDVAX.
The first subject will receive CDVAX. If seven days after the second administration of CDVAX no adverse effects are seen, then a further two subjects will receive CDVAX. If seven days after the second administration of CDVAX no adverse effects are seen in these three subjects, then a further nine subjects will receive CDVAX.
The second immunisation schedule will be initiated in the same manner as the first but will not be initiated until the third subject has been treated in the first schedule.
In case adverse events occur, further treatment is at the discretion of the combined decision of the Data Safety Monitoring Board, the Principle Investigator, the Medical Monitor and a Sponsor representative.
Drop-out subjects will be replaced to achieve 12 evaluable subjects.

NUMBER OF SUBJECTS 24

STUDY POPULATION

Inclusion criteria:

1. Written informed consent
2. Male
3. Age: 18-50 years
4. Body mass index within 18.5 and 29.9 kg/m²
5. Ability to read and comprehend study information
6. Non-smokers or light smokers (<4 cigarettes per day)
7. In good physical and mental health as determined by the following:
a. Complete medical history
b. Complete physical and neurological examination
c. Vital signs including blood pressure, heart rate, respiratory rate, and temperature
d. Standard 12-lead ECG
e. Clinical laboratory (biochemistry, haematology and urinalysis) tests. (Note: Blood and urine samples may be drawn up to 3 weeks prior to the baseline visit of the study provided all data are available and evaluated prior to administration of study drug.) Values of laboratory results outside normal reference ranges may be acceptable if the investigator considers that they do not compromise the safety of the subjects or the conduct of the study
f. Vital signs, clinical laboratory measurements and ECG measurements may be repeated at the discretion of the investigator

Exclusion criteria:
1. Evidence of C. difficile infection
2. Anti-C. difficile (Toxin A) immunoreactivitiy, suggesting previous C. difficile exposure.
3. Diarrhea, active or inactive inflammatory bowel disease, irritable colon syndrome, chronic abdominal pain or other chronic diarrhoea
4. History of malignancy within 5 years
5. History of anaphylaxis, asthma or severe vaccine or severe allergic drug reaction
6. Known or suspected history of immunodeficiency, active or inactive immune-mediated or inflammatory disease
7. Receipt of antibiotic therapy within the previous 30 days
8. Blood or organ donation within the previous 60 days
9. Evidence of clinically significant psychiatric, gastrointestinal, neurologic, neuromuscular, hepatic, pulmonary, cardiovascular, or renal disease (as judged by the investigator)
10. Use of prescription medication or regular use of over-the-counter medicines or herbal or dietary supplements. Acetaminophen/paracetamol may be used intermittently as needed for pain
11. History or current evidence of abuse of any drug substance, licit or illicit, including alcohol; a positive urine screen for drugs of abuse
12. Positive hepatitis C antibody (HCV), hepatitis B surface antigen (HBsAg) or positive human immunodeficiency virus (HIV)-1/2 antibodies
13. Participation in any other investigational drug or device study within 60 days prior to the first study drug administration
14. Relatives of, or staff, directly reporting to the Principle Investigator
15. Vulnerable subjects

SUBJECT WITHDRAWAL
Subjects must be withdrawn from the study prematurely for one of the following reasons:

1. Subject withdrew consent
2. Adverse event(s) which in the opinion of the Principle Investigator may jeopardise subject’s health or may compromise study objectives
3. Abnormal laboratory value(s) which in the opinion of the Principle Investigator may jeopardise subject’s health or may compromise study objectives
4. Any elevated liver transaminases (aspartate aminotransferase [AST] and/or alanine aminotransferase [ALT]
5. Resting blood pressure:
a. systolic blood pressure (BP) ≥160 mmHg or ≤90 mmHg
b. diastolic BP ≥100 mmHg or ≤40 mmHg
6. Resting pulse: <40/min or >100/min
7. Protocol violation (e.g. administration of concomitant/prohibited medication) which in the opinion of the Principle Investigator may jeopardise subject’s health or may compromise study objectives

INVESTIGATIONAL MEDICINAL PRODUCT
The investigational medicinal product (IMP) comprises genetically modified spores of Bacillus subtilis expressing two recombinant fusion proteins on their surface. The two recombinant proteins comprise the non-toxic C-terminus of toxin A (TcdA26-39) from Clostridium difficile fused to one of two Bacillus subtilis spore coat proteins (CotB or CotC) enabling the display of the TcdA26-39 antigen on the outer surface of the spores. The IMP will be tested as a mucosal vaccine following oral administration.

DOSE
MODE OF ADMINISTRATION
TREATMENT SCHEDULE One dose level: 2x1010 CDVAX/administration.

Oral
First immunisation schedule: Four administrations on days 0, 14, 28 and 42
Second immunisation schedule: Four administrations on days 0, 7, 14 and 21

REFERENCE PRODUCT None

STUDY ASSESSMENTS

SAFETY

IMMUNOGENICITY

Treatment-emergent adverse reactions, clinical laboratory, ECG, vital signs

Evaluation of both mucosal and systemic immunity (secretory IgA and serum IgG), pre-treatment and post-treatment

STATISTICAL ANALYSES All analyses will be done descriptively.
STUDY PERIOD 65 days, and follow-up periods to be defined

Task 4.2: Initiate recruitment of healthy volunteers

The validation and analysis of genetically modified Bacillus spores in faeces of the study participants by C.RIS Pharma was organized by FGK’s project manager. The draft CRF was developed plus finalised, site setup performed and the first contact with possibly available healthy volunteers initiated at the Phase I Unit at Nuvisan. The draft site documents (e.g. Investigator Site File, Lab Manual etc.) were prepared by FGK’s project manager.

Since the PEI approval is still pending, no site initiation visit or recruitment of healthy volunteers could be performed so far.

Potential Impact:
The Impact of C. difficile on Public Health

Clostridium difficile, a spore forming, anaerobic, Gram-positive bacterium is now the leading cause of nosocomial infection in the developed world. C. difficile infection (CDI) primarily affects elderly hospitalised patients who have received antibiotic treatment, the use of which results in a disturbance of the bacterial flora in the gastro-intestinal (GI) tract. This allows spores of C. difficile to germinate and for the bacteria to proliferate, with the concurrent production of toxin A and toxin B that leads to severe diarrhoea and acute colitis. CDI is associated with a significant mortality rate, particularly in the elderly (Bauer et al., 2011; Rupnik et al., 2009). It is extremely difficult to contain a C. difficile outbreak within the hospital setting, and large outbreaks of multidrug resistant epidemic strains have been documented throughout Europe (Birgand et al., 2010). Finally, after antibiotic therapy and remission, a proportion of patients (~30%) enter a state of relapse or recurrence where the disease reappears (Williams and Spencer, 2009).

• Mortality rates: In the UK during 2009, four-times as many patients (~4,000) died from CDI than from Methicillin-resistant Staphylococcus aureus (MRSA) (HPA, 2009). In a Europe-wide study (34 countries) incidence rates were 4.1 per 10,000 patient days. In both of these studies CDI related mortality was around 10% (Bauer et al., 2011).
• Epidemics have arisen recently with the emergence of ‘hypervirulent’ strains (Merrigan et al., 2010).
• Almost all C. difficile strains carry antibiotic resistance (e.g. to erythromycin, clindamycin, moxifloxacin, tetracycline) and over half of clinical strains carry multidrug resistance (e.g. to rifampicin and fluoroquinolones) (Spigaglia et al., 2011).
• C. difficile has been shown to transfer from animals to humans with the emergence and transfer of at least one ribotype (type 078 strains) that are hypervirulent (Goorhuis et al., 2008). Since antibiotics are used extensively in the livestock industry this is an alarming situation with the potential threat of the emergence of new antibiotic resistant (AbR) strains (Smits et al., 2016)of C. difficile to humans.
• C. difficile can conjugate and transfer genetic information with relative ease. In the gut environment this situation is of particular concern.

The issue of C. difficile multidrug resistance has been highlighted in recent position papers from the EU (e.g. IMI COM (2011) 748 ). The existence of epidemic and hypervirulent strains shows that this organism can change rapidly; therefore, the continued evolution of drug resistance is a realistic scenario especially under current treatment regimens. Management and treatment of nosocomial CDI is therefore set to become an increasing challenge and will be associated with escalating healthcare and social costs.

Recent reviews on CDI have revealed that in the USA the disease shows no signs of decline whereas in the UK improvements in hospital hygiene and antibiotic usage do appear to show signs of reducing the incidence of disease (Martin et al., 2016).


Current Therapeutic and Prophylactic Measures

Vaccines

A multitude of vaccines are under pre-clinical development but the most significant are those in clinical trial. (Table 1)

The Sanofi Pasteur vaccine is currently mid-way through a phase III trial. In reality this vaccine is likely to reach the market first.


Antibiotic Therapies (Table 2)


Antibiotic therapies are attractive and some, such as Ridinazole look promising but will require funding for phase 3 evaluation. It should be noted that some of these drugs are narrow-spectrum and of course resistance could develop.


Other approaches (Table 3)


The non-toxigenic approach (Table 3) utilising spores of a strain of CD that does not produce toxins is attractive (developed by Dale Gerding, Chicago) and was taken forward by ViroPharma. The use of non-toxigenic spores appeared to prevent colonisation of the host by C. difficile. However, this product has now been dropped following the acquisition of ViroPharma by Shire in 2014.
The monoclonal approach (Table 3) of passive immunisation taken forward by Merck (as Bezlotoxumab) has resulted in the registration in late 2016 of a clinical product. It should be noted that this is a therapeutic approach and by injection. The costs are high and believed to be ~$30,000/dose.
The FMT approach (Table 3) developed by SERES Therapeutics looked promising with a cocktail of spore-forming bacteria administered as a medicine (SER-109). SER-109 appeared to be able to prevent CDI as well as colonisation in pre-clinical studies. However, completion in mid 2016 of a phase 2 trial showed no efficacy at all and resulted in a 79% fall in share value for this public company and raised questions over the validity of this approach.

The Need for a Vaccine that Prevents Colonisation

The non-toxigenic approach as well as the SER-109 fecal microbiota transplantation cocktail both are able to prevent colonisation of the host. For CDI there are two aspects that are of critical importance for control of this disease:

1) neutralisation of toxins
2) prevention of colonization

To neutralise toxins it is clear that this will prevent symptoms of disease arising from the cytotoxicity of the toxins and resulting inflammation. However, one aspect that has not been considered fully is the impact of colonisation. It is apparent that spores of C. difficile have the ability to germinate, proliferate and then resporulate in the GI-tract. As spores and as cells they can reside in the GI-tract in complex biofilms where they remain silent and hidden from both antibiotics and from the host immune response. The ability to remain silent in the GI-tract makes it possible for the host to relapse where weeks or months later the disease recurs presenting a significant problem for the patient and hospital. There are additional factors that go beyond the immediate need to treat the patient. Spores are shed in the feces where they can survive indefinitely in the environment in a dormant state. It stands to reason that spores originating from infected patients may have acquired AMR and may also be of more virulent strain varieties. The ease at which C. difficile can acquire AMR is alarming as well as its continued ability to evolve into more virulent forms, especially originating from hospital outbreaks.

A final factor is that animals are carriers of C. difficile. In livestock, notably pigs, C. difficile can cause neonatal gastroenteritis although in Europe at least CDI has not been properly documented or recognised (Songer and Anderson, 2006; Songer and Uzal, 2005). Livestock will continually be excreting spores in their feces where it can contaminate farm workers and potentially the general public. Likewise ~10% of raw meat is thought to be contaminated with C. difficile spores of animal origin (Songer et al., 2009). The overuse of antibiotics in animals as growth promoters makes the danger of transfer of new varieties to humans particularly real. Indeed, evidence shows that at least C. difficile variety (ribotype 078) has transferred to humans (Rupnik et al., 2008).

The ultimate purpose of vaccination is often forgotten but Smallpox and Polio are two examples worth considering and remind us that vaccination should ultimately lead to the eradication of the pathogen. In the case of C. difficile a dangerous situation is now apparent where the pathogen (potentially AMR) is able to transfer to humans and a continued recycling of C. difficile occurs between animals-humans and the environment coupled with the acquisition of AMR genes and development of hypervirulence. This is further compounded by an inability to diagnose or detect the spore form of C. difficile.

Therefore, in the case of CDI it is clear that a vaccine must provide 2 roles, to prevent symptoms of disease but also to prevent colonisation of the host that ultimately would lead to the eradication of this emerging disease.


The Potential Impact of CDVAX and its Exploitation

The CDVAX vaccine is comprised of spores of Bacillus subtilis that express a fragment of C. difficile toxin A (aka TcdA26-39). Expression is such that TcdA26-39 is displayed on the spore surface. This expression is also stable. In previous work we have shown that oral immunisation of hamsters with spores expressing TcdA26-39 provides ~75% protection to CDI. Moreover, antibodies raised against TcdA26-39 cross-react and neutralise Toxin B as well as toxin A (Permpoonpattana et al., 2011).

In the CDVAX project we have made an important discovery that B. subtilis spores expressing TcdA26-39 completely prevent colonisation of C. difficile in the host as well as neutralisation of toxins. Further, we have deciphered how this occurs by showing that anti-TcdA26-39 antibodies also bind to a number of proteins displayed on the surface of vegetative cells of C. difficile. By binding to these proteins colonisation is prevented. The identity of these proteins has been patented and this has value to support the exploitation of the CDVAX vaccine.

In sum, the CDVAX vaccine (B. subtilis expressing TcdA26-39) neutralises toxins A and B but also prevents colonisation. The critical aspect here is the production of mucosal antibodies that can only be induced following mucosal delivery and thus are critical to preventing colonisation.

So, CDVAX is the only vaccine that is likely to prevent colonisation. It is apparent also that typically, mucosal (sIGA) antibodies are short-lived but we believe this not important for C. difficile since the primary target group is the elderly who are planning on entering hospital. Thus, the vaccine would be considered for those aged >65 and potentially for those about to enter hospital. The final product being an oral tablet/capsule (or possibly a liquid formulation)would be simple to administer requiring no clinical expertise.

The outcome of the phase 3 clinical trial by Sanofi will most likely deliver an injectable vaccine but for the reasons already outlined above we feel it unlikely that this vaccine will fulfil all the requirements of a vaccine that fully protects against CDI. In our own hands we have been unable to demonstrate any prevention of colonisation using injection of toxoids so we feel there is room for a best in class vaccine.

It must be emphasised that the CDVAX vaccine might also be considered as an adjunct vaccine to an injectable vaccine and one route for exploitation may be to consider an injectable vaccine followed by oral administration of CDVAX. Although prime boost vaccines have not yet reached the market this approach may well work enabling a combined injection-oral vaccine to prevent both symptoms of CDI as well as colonisation.

The CDVAX consortium has discussed and agreed on a basic exploitation strategy which is outlined in Deliverable D6.5. In brief the RHUL patent covering the mechanism for how prevention of colonisation occurs will be assigned to the consortium. We, as a consortium have not yet defined how we will assign but most probably this will be via the RHUL start up SporeGen Ltd which has a technology transfer agreement with RHUL. The CDVAX partners who wish to participate will be given rights to the exploitation of the CDVAX product.

Funds to enter further clinical trials will be sourced most probably using the NIH as the first step since the USA has the greatest problem with CDI and will fund foreign ventures. All 5 CDVAX partners are fully committed to taking this vaccine forward and our belief is that the value of this vaccine is that it prevents colonisation.


References

Bauer, M.P. D.W. Notermans, B.H. van Benthem, J.S. Brazier, M.H. Wilcox, M. Rupnik, D.L. Monnet, J.T. van Dissel, and E.J. Kuijper. 2011. Clostridium difficile infection in Europe: a hospital-based survey. Lancet 377:63-73.
Birgand, G., K. Blanckaert, A. Carbonne, B. Coignard, F. Barbut, C. Eckert, B. Grandbastien, Z. Kadi, and P. Astagneau. 2010. Investigation of a large outbreak of Clostridium difficile PCR-ribotype 027 infections in northern France, 2006-2007 and associated clusters in 2008-2009. Euro Surveill 15:
Goorhuis, A., D. Bakker, J. Corver, S.B. Debast, C. Harmanus, D.W. Notermans, A.A. Bergwerff, F.W. Dekker, and E.J. Kuijper. 2008. Emergence of Clostridium difficile infection due to a new hypervirulent strain, polymerase chain reaction ribotype 078. Clinical infectious diseases : an official publication of the Infectious Diseases Society of America 47:1162-1170.
HPA. 2009. Clostridium difficile infection: How to deal with the problem. In Health Protection Agency, 130.
Martin, J.S. T.M. Monaghan, and M.H. Wilcox. 2016. Clostridium difficile infection: epidemiology, diagnosis and understanding transmission. Nat Rev Gastroenterol Hepatol 13:206-216.
Merrigan, M., A. Venugopal, M. Mallozzi, B. Roxas, V.K. Viswanathan, S. Johnson, D.N. Gerding, and G. Vedantam. 2010. Human hypervirulent Clostridium difficile strains exhibit increased sporulation as well as robust toxin production. J Bacteriol 192:4904-4911.
Permpoonpattana, P., H.A. Hong, J. Phetcharaburanin, J.M. Huang, J. Cook, N.F. Fairweather, and S.M. Cutting. 2011. Immunization with Bacillus spores expressing toxin A peptide repeats protects against infection with Clostridium difficile strains producing toxins A and B. Infection and immunity 79:2295-2302.
Rupnik, M., A. Widmer, O. Zimmermann, C. Eckert, and F. Barbut. 2008. Clostridium difficile toxinotype V, ribotype 078, in animals and humans. J Clin Microbiol 46:2146.
Rupnik, M., M.H. Wilcox, and D.N. Gerding. 2009. Clostridium difficile infection: new developments in epidemiology and pathogenesis. Nat Rev Microbiol 7:526-536.
Smits, W.K. D. Lyras, D.B. Lacy, M.H. Wilcox, and E.J. Kuijper. 2016. Clostridium difficile infection. Nat Rev Dis Primers 2:16020.
Songer, G., H.T. Trinh, G.E. Killgore, A.D. Thompson, L.C. McDonald, and B.M. Limbago. 2009. Clostridium difficile in retail meat products, USA, 2007. Emerging Infectious Diseases 15:819-821.
Songer, J.G. and M.A. Anderson. 2006. Clostridium difficile: an important pathogen of food animals. Anaerobe 12:1-4.
Songer, J.G. and F.A. Uzal. 2005. Clostridial enteric infections in pigs. J Vet Diagn Invest 17:528-536.
Spigaglia, P., F. Barbanti, P. Mastrantonio, and Esgcd. 2011. Multidrug resistance in European Clostridium difficile clinical isolates. Journal of Antimicrobial Chemotherapy 66:2227-2234.
Williams, O.M. and R.C. Spencer. 2009. The management of Clostridium difficile infection. British medical bulletin 91:87-110.


List of Websites:
www.cdvax.org