A Study to Quantify Bacille Calmette Guerin (BCG) and Characterise the Immune Response After BCG Vaccination

Over the last 3 years, the McShane group in Oxford have established a clinical trial programme to evaluate the safety and immunogenicity of this BCG prime-MVA85A boost vaccination strategy in a series of Phase I studies in the UK.

To date, in Oxford, the safety and immunogenicity of MVA85A (dose 5x10^7 pfu) has been assessed in 14 mycobacterially and BCG naïve individuals, 27 individuals previously vaccinated with BCG and 12 individuals latently infected with M.tb. In all 53 individuals there have been no serious or severe adverse events, and a single vaccination with MVA85A has been highly immunogenic in BCG naïve volunteers, and significantly moreso in BCG primed volunteers. (23) Following the success of the UK studies, 21 healthy individuals (11 BCG naïve, 10 BCG primed) in the Gambia and 24 healthy adults in Cape Town, South Africa (ongoing age de-escalation phase II trial), have now been vaccinated with MVA85A, with no serious or severe adverse events, and a safety and immunogenicity profile comparable to the UK. A study in infants in the Gambia is now underway and the adolescent arm of the age de-escalation studies in Cape Town is now also underway.

Interferon-gamma as a protective correlate and the need to identify other immunological correlates of protection:

In all of the studies to date, (both preclinical and clinical), the main immunological readout has been the ex-vivo interferon-gamma (IFN- ) Elispot assay, used to assess specific T cell responses to tuberculin PPD, purified antigen 85 complex and pools of overlapping 15mer peptides spanning the length of antigen 85A. The secretion of IFN-y from antigen-specific T cells is the best available immunological correlate of protective immunity. The rationale and evidence supporting the use of assays for determining IFN- production as a correlate of protection are substantial, but alone, it may not be sufficient. There is increasing evidence that CD4(+) T cells possess IFN-gamma-independent mechanisms that can limit the growth of an intracellular pathogen and are dominant in secondary responses to M. tuberculosis.

It is clear that new immunological correlates of protection need to be identified, and given the considerable time and cost involved in any Phase III efficacy trial of a new TB vaccine, there is therefore an urgent need to identify protective correlates which will help distinguish between candidates demonstrated to be safe and immunogenic in phase I trials.

Whilst we cannot infect human volunteers with TB to address this issue, we can use another replicating mycobacteria, BCG, as a surrogate. A good vaccine for TB should also reduce BCG replication and increase BCG clearance. The objective of this proposal is to perform the preliminary work necessary to develop such a BCG 'challenge' model in which bacterial (BCG) load in the skin is assessed following vaccination with MVA85A.

Such a challenge model is based on evidence that animals given subunit vaccination against TB have attenuated T cell responses to replicating mycobacteria administered as challenge agents. The hypothesis for this model is that replication of BCG and hence the induction of a BCG induced immune response will be abrogated by pre-existing immunity induced by a subunit vaccine, MVA85A. In addition, replication of intradermal BCG administered as a challenge agent can be assessed by biopsies of volunteers' skin. So, these methods would allow correlation of immune responses elicited by MVA85A vaccination with direct or indirect measures of diminished intradermal mycobacterial (BCG) replication in humans.

There are three parts to this model:

1. Preliminary mouse pilot study

   This involves current ongoing work to identify the time at which the mycobacterial load at the BCG vaccination site (ear) may be at its highest (and therefore most detectable by bacterial quantification methods), in a murine model:

   Mice are being vaccinated with intradermal BCG, into the ear, and then sacrificed at different timepoints, the ears and lymphoid tissues being harvested for BCG quantitation.

   Real -time PCR (RTPCR), histological analysis using staining with Ziehl-Nielson and mycobacterial culture techniques are being performed on the tissues to quantify the BCG present. Numbers of BCG present at all timepoints are being compared. Preliminary work suggests that BCG replication at the vaccination site decreases with time, but this is being investigated further. The results from this have advised on the timings of biopsy and blister in the human pilot study, such that the chances of recovering significant numbers of BCG will be maximised.

2. Human pilot study based on results of 1) This proposal refers to this part of the model. 24 BCG naïve human volunteers will be vaccinated with BCG (SSI strain) intradermally into the deltoid muscle of each arm. The total dose given will be the standard dose used in BCG vaccination programs worldwide: 0.1ml of 2-8x10^6 pfu/ml, i.e., 2-8x10^5 pfu. In this study half this dose (1-4x10^5 pfu) will be administered into each arm. All volunteers will then have a punch biopsy at one vaccination site and a suction blister formed at the other, (so that each volunteer will be their own comparison), at 3 different timepoints post-vaccination.

   Preliminary BCG quantitation methods on the biopsy tissue from the first 6 volunteers (who had a biopsy at 2 weeks post-vaccination) show that between 5x10^2 and 4x10^3 BCG cfu are present at the vaccination site, 2 weeks post-vaccination. We do not yet know if the replication of BCG in the skin peaks before, at, or after this timepoint.

   Results from the preliminary murine experiments suggest that BCG replication in the skin decreases over time, which would seem to necessitate another early timepoint in the human trial, to "catch" this replication. However, it is also known (from murine experiments) that maximal BCG replication in the local draining lymphoid tissue occurs at 3-4 weeks post-vaccination, so by not quantitating BCG in the skin at 4 weeks we may miss the peak of BCG replication. In addition we do not know if BCG replication intradermally in the mouse ear will necessarily resemble that in human skin.

   For these reasons, and in order to fully characterise the kinetics of BCG in human skin post-vaccination, we shall vaccinate 3 groups of volunteers as outlined in section 4, "Trial Design", incorporating both early and later timepoints for biopsy and blisters (1, 2 and 4 weeks).

   The suction blister is created using a standard technique used by a number of dermatology research groups; a negative pressure of 25-40 kPa (200-300mmHg) below atmospheric pressure is applied via a suction chamber, for 2 to 4 hours using a clinical suction pump (VP25 portable suction unit, Eschmann, Sussex, UK) until a unilocular blister measuring 10-15mm in diameter is formed. This procedure is painless.

   Mycobacterial quantification techniques will be applied to the biopsy and fluid specimens (RT-PCR and culture), and the timepoint yielding the highest numbers of BCG will then be chosen for the randomised human trial. In addition, flow cytometry assays will be applied to the blister fluid specimens to characterise the local cellular reaction to BCG.

3. Future Prospective randomised human trial (based on results of 1 & 2):

2 groups of BCG naïve subjects will be randomised to receive either a single immunization with MVA85A or saline. The numbers per group will be decided on the basis of the results from the murine work and BCG pilot study, accounting for the level of variability in BCG replication, such that the number per group should be able to reliably detect a significant difference in BCG replication between the 2 groups (should it exist). The vaccine-induced immune response will be fully characterised. At 4 weeks, all volunteers will then be "challenged" (intradermally) with BCG. BCG-induced immune responses will be monitored for a further 6 months. In addition, a punch biopsy (+/- suction blister) will be taken from the BCG vaccination site and the amount of bacteria present quantified. The correlation between any reduction in BCG replication and MVA85A-induced immune response will then be assessed, in order to identify a potential correlate of protection.

Future candidate vaccines can then be assessed against their capacity to induce such an immune response.

Potential risks to human subjects:

The general risks to participants in this human pilot study are associated with phlebotomy, vaccination, punch biopsy and suction blister techniques. Mild tenderness, bruising, or fainting may result from venepuncture. The volume of blood drawn over the study period (504.5ml +/- 20ml) should not compromise these otherwise healthy subjects.

BCG is a licensed vaccine which has been in use for the last 100 years, given to 4 billion people, and has an excellent safety profile.

Potential vaccination risks include the following:

(Local reactions: An inflammatory reaction as manifested by redness and swelling, that can form a small ulcer, may occur at the site of vaccine injection. This usually heals and leaves a small flat scar. It is also possible to develop some swelling of glands in the armpit, but less than one centimetre across. These are natural reactions to the vaccine.

Uncommonly (less than 1 in 100 people) swelling of glands in the armpit to more than 1cm across or an ulcer that discharges fluid at the injection site may occur.

Rare side effects (less than 1 in 1000 people) include inflammation of glands, sometimes with abscesses and discharge of fluid from the swellings.

Systemic Reactions: Systemic reactions to immunization with BCG that could potentially occur are rare (less than 1 in 1000 people) and include low-grade fever and headache, and allergic reactions. Disseminated complications of BCG such as osteitis or osteomyelitis have also been reported but are extremely rare and have usually been reported in immunocompromised, not immunocompetent individuals.

Punch biopsy and suction blister risks:

There is a small risk that the punch biopsy site or blister site may become infected. If this did occur, treatment with antibiotics may be required. The punch biopsy site will heal to form a small scar. A small scar might develop at the blister site but often this is barely visible after several months.

Allergic Reactions and Anaphylaxis: As with any vaccine, allergic reactions are possible. Very rarely allergic reactions to local anaesthetic (administered prior to the punch biopsy) may occur.

Known Potential Benefits:

Volunteers will benefit from partial protection against TB, by vaccination with BCG. Volunteers will gain some information about their general health status at the pre-study screening assessment. It is also hoped that the information gained from this study will contribute to the development of a safe and effective TB vaccine for HIV negative and positive individuals.