Clinical Testing of Pre-erythrocytic Malaria Vaccines
Hotel De Wolfsberg, Groesbeek, The Netherlands, 3 - 4 November 2000

Moderator and Rapporteur: R.W. Sauerwein

June 2001

List of Participants

Sirima, Bienvenue, Burkina Faso
s.sirima.cnlp@fasonet.bf

Jepsen, Søren, Denmark
sje@ssi.dk

Bouzidi, Ahmed, France
ahmed.bouzidi@wanadoo.fr

Corradin, Giampietro, Switzerland
giampietro.corradin@ib.unil.ch

Engers, Howard, Switzerland
Engersh@who.ch

Karbwang, Laothavorn, Switzerland
karbwangj@who.ch

Kilima, Wen, Tanzania
wkilama@AfricaOnline.co.Tz

Bojang, Kalifa, The Gambia
kbojang@mrc.gm

Hollingdale, Mike, UK
bgymrh@lucs-mac.leeds.ac.uk

Hill, Adrian, UK
adrian.hill@molecular-medicine.oxford.ac.uk

Reece, Wil, UK
mailto:adrian.hill@molecular-medicine.oxford.ac.uk

Sinden, Bob, UK
r.sinden@ic.ac.uk

Epstein, Judy, USA
epsteinj@nmrc.navy.mil

Heppner, Gray
USA
donald.heppner@na.amedd.army.mill

Leroy, Odile, France
odile.leroy@wanadoo.fr

Eling, Wijnand, The Netherlands
w.eling@mmb.azn.nl

Telgt, Denise, The Netherlands
d.telgt@aig.azn.nl

Sauerwein, Robert, The Netherlands
r.sauerwein@mmb.azn.nl

II. Objective

An increasing number of malaria vaccine preparations are in the pipeline or actually tested in clinical trials. Although product definitions and regulations for admittance to clinical trials are well circumscribed by (inter)-national regulatory agencies, the protocols for clinical testing and evaluation often follow more institutional rules and policies. In addition, the enormous task to clinically and immunologically evaluate a malaria vaccine, includes a number of questions, which are specific for malaria. Stage specific malaria vaccines i.e pre-erythrocytic, asexual blood stage and sexual/ sporogonic stage vaccine, all have their own clinical applications and evaluation.

The objective of the workshop was to specifically discuss issues related to clinical testing and evaluation of pre-erythrocytic (pre-E) vaccines and to identify areas for further exploration.

III. Programme

Friday 3rd November 2000, 14.00 - 18.00 hrs

Opening
R. Sauerwein

Ongoing and planned human pre-E vaccine trials
J. Epstein, A. Hill, G. Heppner, G. Corradin

Aspects of clinical trials
Production/Safety
J. Karbwang, A. Bouzidi

Saturday 4th November 2000, 900 - 12.00 hrs

Aspects of clinical trials
Animal models
A. Hill

Laboratory evaluation of pre- clinical trials
W. Reece, G. Corradin, R. Sauerwein, R. Sinden, G. Heppner

Clinical trials in Africa
AMVNT - W. Kilama

Burkina Faso - B. Sirima

The Gambia - K. Bojang / A. Hill

Ghana - J. Epstein

Closing remarks
R. Sauerwein

IV. Report

Ongoing and planned human pre-E vaccine trials

As presented by Epstein, the multistage DNA based vaccine operation (MuSTDO) of the US Navy programme, has the objective to reduce all manifestations of P falciparum malaria in 95% of the cases for 6 month and to reduce mortality by at least 50% for at least one year after vaccination. The vaccine strategy which is based on duplication of whole organism immunity, involves a prime-boost strategy with 15 DNA-based genes, 8 genes in recombinant pox-virus and 4 recombinant proteins and is carried out in collaboration with Aventis/ Pasteur and SKB. Antigens are both from pre-E and asexual blood stages. Four trials were presented:

I. PfCSP DNA (20-2500 ug) Phase I trial induced CTL responses in humans which were HLA restricted but redundant. No side effects were reported (Wang et al).

II. In the second PfCSP DNA vaccine trial the route of administration was investigated (Needle IM vs BiojectorIM vs Biojector IM(70%) plus ID (30%). Of the three routes the IM biojector appeared the most immunogenic as assessed by frequency and magnitude of induced T cell responses. There was no antibody response by any route.

III. PfCSP DNA prime, RTS,S recombinant protein boost using the biojector, started in August 2000.

IV. MuSTDO 5.1 (5 pre-E antigens) is tested in combination with GM-CSF based on experiments in mice showing enhanced immunogenicity and protection (Weiss et al, 1998). Dosages are used up to 500 ug plasmid with 3 doses at 4 wks interval.

In the near future at least two MuStDO 8 trials are planned for. This vaccine consists of MuStDO 5 with addition of three blood stage antigens: MSP-1, AMA-1, EBA 175. The first MuStDo trial will start in 2001 all genes will be synthetic for optimal codon usage, the gene gun will be evaluated and there will be an experimental challenge. The second planned MuStDO 8 trial will be a prime boost trial. The prime will be the plasmids from the first MuSTDO-8 trial, the boost will consist of rec poxvirus proteins, or recombinant proteins or a combination of both.

Hill: Results RTS/S trials in non-endemic countries and the Gambia show that high levels of interferon-secreting cells are induced but responses are of short duration and directed against polymorphic peptides (J Inf Dis 1999). IFN-g response is most CD8-cell derived. Modified Vaccinia Virus Ankara (MVA) is used for heterologous prime-boost vaccination. MVA has an excellent safety record in humans.

Upon priming with sporozoietes in murine models, the CTL response is boosted by MVA expressing a multi-CTL epitope protein of 6 pre-E antigens. A trial will be carried out in the naturally exposed population of the Gambia with modified MVA. Phase I studies with DNA and/or MVA are planned and/or in progress in the Gambia.

ME-Trap DNA has been used to prime in several vaccination schedules to investigate the magnitude of the immune response (DDD, MM, DDD-MM etc). A phase I and phase IIa has been carried out in the UK with ME-Trap DNA and MVA. The phase IIa showed no protection. For vaccination the best results are obtained with a Powderject gene gun that injects gold beads with DNA at the junction of epidermis and dermis.

The efficacy of a vaccine is determined by host factors (i.e HLA type, pre-existing immune responses against the vaccine components, etc), vaccine composition and presentation (route, regimen) and protection. In the future studies will be carried out adressing questions including whether RTS,S will be improved by MVA-CS boosting? Is Fowlpox-CS with MVA-CS better?

Heppner: 43 malaria vaccine trials with 22 constructs studies have been carried out between '86-'00. Over 300 volunteers have been challenged with 97% success rate; prepatent period varies between day 9-13.

Objectives: i) 95% protection ii) protection against severe and fatal disease in the 5% failure, iii) protection for at least 6 mnth.

RTS/S vaccine trial in the Gambia: safe and well tolerated but soreness in the arm is common. 2/3 shows reduction of parasitemia for 2 mnth.

Future strategic plans include i) RTS,S/ SBAS2 studies to optimise the vaccination schedule, phase I and IIa studies. ii) MSP-1(3D7)/ SBAS2 vaccine trials for children and travelers phase I and IIa. Next, field studies (Phase Ib) with stepdown approach in age (adults, 6-11years, 0.5-5 years). iii) Combination of RTS/S SBAS2 with MSP1.

Further antigens in the programme: AMA1 and LSA3

Corradin: PfCS 102 (residues 282-383) long synthetic peptide induces 83% protection in mice mediated by CD8 effector cells. Studies were carried out with freunds, alum and Montanide 720 in mice. Phase I clinical trials with alum or Montanide showed no major adverse reactions and a strong immunogenicity.

Vaccine production/ Safety

Karbwang opened this session by giving a overview of procedures and requirements for trials. Design of trials, IRB’s and SOP’s were separately discussed. It was stressed that IRB’s should look at safety and immungenicity data from pre-clinical studies. For dose ranging studies it is needed to have evidence for the rationale of the proposed dosages. One obvious key question is to what extent animal data can be transferred to the human situation. If there is evidence based on animal studies that the proposed dose is acceptable, then there is no need to increase the dose levels to potentially toxic ranges. Another important aspect is a proper risk/benefit assessment for the population to be studied. The question was raised s to what extent the existing ethical guidelines (Helsinki declaration 2000) are applicable for Africa?

An overview was presented by Bouzidi on the production and quality control of MSP 3.181-276 which is one of the three long synthetic peptide vaccines in production. Product specifications require information on purity ( acceptable is 80% in HPLC), physio-chemical stability; toxicology including sterility, LAL-assay, pyrogen tests in rabbits, toxicology tests in small (rats,mice) and large animals (rabbits, possibly monkeys). Montanide was tested alone and in combination with peptide. Potency tests can be carried out along the clinical trial. There are several protocols for these potency tests in mice.

Animal Models

Primate models do have their shortcomings to study protection as pointed out by Hill; Aotus is unreliable with 50% infection rates and lack of immunological tools. Chimpanzees need high challenge dosages and do not show natural bloodstage infections. Rhesus macaques with P cynomolgi or P knowlesi are probably the best primates for challenge studies, in particular for P vivax. P knowlesi transfectants with P falciparum may be an attractive alternative.

Alternatively, in the absence of appropriate challenge models, one may alternatively optimise immune responses to candidate vaccines. Although the magnitude of immune responses in mice show a poor correlation with human reactivity, a negative or very poor response can be considered as a "no-go" criteria. In case of a satisfactory response, maximum antibody, CD4 and CD8 responses can be pursued in rhesus monkeys before going into human clinical trials (Hill, Corradin, Epstein). Immunisations and sporozoite challenge studies were presented with PkCSP/ PkSSP1 and PkMSP1-41 were presented by Epstein. From their mouse studies it was concluded that addition of GM-CSF as co-stimulant induced higher protection rates and studies in humans are planned based on these results.

Corradin compared different adjuvants in mice and humans and got discordant Elispot results weakening the case for mice as model for human vaccine testing. Levels of IFN-g in men may correlate with mice (Hill). There is a need to acquire more data on the level of IFN-g spots after natural infections.

Laboratory assays for pre-E trials

The Elispot assay for interferon-gamma is the prime read-out for immunogenicity and probably best correlate of protection. This assay can be carried out ex-vivo and with cultured lymphocytes. 17 ex-vivo and 31 cultured IFN-g test were carried out by Reece with PBMC's from vaccinees with a panel of peptides from the multi-epitope and TRAP. IFN-g producing cell numbers are magnitudes higher in cultured cells. Objective for the near future is to study inter and intra-assay variation using mathematical modelling and to determine the best analytical method for linkage to protection. Corradin mentioned that Elispot IFN-g test with frozen cells from Burkina Faso showed no positive reaction in ex-vivo but spots were obtained with purified CD8 cells after in vitro culture. Read-out can be performed by individuals or by machine with dramatically different results.

Sauerwein discussed experimental challenges in human volunteers with mosquitoes infected with NF54. It was shown that 100% infection rate was obtained in 2 challenge experiments with bites of 5 infected mosquitoes; the challenges were carried out safely and tolerated with moderate discomfort . A real time quantitative PCR was developed which detects 20 parasites in a ml of blood. Strikingly similar patterns of blood parasitemia starting day 6.3 post-infection was observed. This system can be used in phase IIa trials for pre-erythrocytic vaccine and has potential for evaluation of asexual vaccines; the latter needs further discussion with EMVI. Further studies will include extension of the period of asexual parasitemia before treatment, the use of other strains and less parasites to determine the minimum 100% infective dose. Sinden has produced infected mosquitoes for human challenges in collaboration with Hill. The need for standardisation of reagents and protocols between centres was emphasised. He also stated that minimum infective dose and determination of sporozoite load in mosquitoes should be done.

CLINICAL TRIALS IN AFRICA

Kilama: Introduction on the history of the African Malaria Vaccine Trial Network ( AMVTN) was presented. AMVTN was founded in Arusha, Tanzania in 1995 as a network to strengthen the capacity of African institutions to carry out clinical malaria vaccine trials. An overview of its activities was presented including a series of workshops and a directory of potential trial sites. The future plans are focussed on continuation of the workshops, strengthening of human resources in specific sites and establishment of network for immuno-assays in Africa i.e. standardisation and harmonisation of protocols.

Sirima, Bojang/Hill, Heppner and Eptein subsequently presented recent achievements and future trials in respectively Burkina-Faso, the Gambia and Ghana.

In planning and actual performing clinical trials in Burkina-Faso, a number of points need to be taken into consideration. The EIR can vary between 1-10 in the urban areas whereas EIR can go up to values of 500-1000 in the rainy season in rural areas; the P falciparum prevalence in children < 5 years can be 95% in the rainy season and decline to 40-45% in the dry season. In addition, there are many mixed infections with P ovale and susceptibility to malaria in different ethnic groups shows a significant variation. The Fulani produce more antibodies to infection and show less disease (Modiano, PNAS ’93).

In the Gambia a phase IIb trial (n=306) is carried out with 3 doses of RTS-S vs rabies over a 5 month period following clearance of parasitemia with Fansidar. The objective is to see whether MVA can boost natural infections. Data in mice show that indeed MVA boosts responses after priming with sporozoites. The choice of a control vaccine was actively discussed in the context of the rabies vaccine as a control to a HepB-CS vaccine. Another point of discussion from this presentation was the policy to treat malaria cases as detected by active or passive case detection; this is particularly relevant in relation to the age of the children in the study and the chances for severe disease. On has to weigh the toxicity data to the chances to acquire cerebral malaria. With the planned transfer of DNA vaccine and different regimens and combinations of prime/boost (MVA or fowl pox FP9) protocols, ethical and safety issues are of prime relevance and need careful consideration. Group seize planned for Phase I trials will be 20.

In Kisumu, Kenya trials are planned with MSP1-42 and SBAS2 by the US Army.

In Ghana two sites have been identified for clinical trials in collaboration with the NMIR: Navrongo in the north and Mengpang which is near the capital Accra. Prime vaccine candidate will be MuStDO 8 with optimised human codon usage possibly in combination with GM-CSF and rabies vaccine as a control. It is intended to go down from trials in adults directly to 1-2 year old children without an evaluation in the 6-11 age group as it was conceived that information obtained would have no specific value.

Summing-up:

Rationale, policy and/or standardization of a number of products and topics are to be considered for pre-E vaccine design as came up from the lively and constructive workshop discussions as presented by Sauerwein.

Delivery systems of antigens: The ones currently under include MVA, DNA, FP9, RTS, long synthetic peptides and recombinant proteins. Which is the best or best combination?

Adjuvantia: The ones currently under include Alum, montanide 720, GM-CSF, SBAS-2, CpG, QS21. Which is the best or best combination?

Dosages and schedules as well as routing of immunisation 10.0pt; font-family:Arial">: ID, IM, IE, one arm, two arms? The use of gene guns (Powderjector/ Biojector) vs needle. Justification for the choices is important and preferably harmonisation and development of a standard protocols need to be pursued.

Animal models: Current models are primarily using mice and monkeys. Strength and weaknesses for prediction of outcome in humans need further evaluation. So far the Rhesus monkey seems to be the most realistic and best alternative for predicting immunogenicity in humans.

Immunogenicity: Obstacles for the IFN-gamma Elispot are low number of spots with fresh cells as compared to cultured, choice of cells for evaluation ( PBMC, CD3 or subsets),variation in read-out depending on the methodology used (manually or machine). Additional topics to cover are the results with titrated cells and the way of expression of the results so that results can be compared between labs. There is a need to acquire more data on the level of IFN-g spots after natural infections.

As this assay is the generally accepted most important immunological assay for pre-E vaccines, harmonisation and standardisation is of prime importance in order to compare different trial results. A separate workshop on Elispot may be considered.

There was consensus on the value of Phase IIa trials in the continuation of testing process.

The value for the critical path of measurement of lymphocyte proliferation assays, CTL assays, specific antibody levels needs consideration.

African trials: A variety of relevant topics for trials in endemic countries were discussed in greater and lesser extent including i) the necessity to include safety trials in chidren 6-11 years when the target group is small chidren ii) the choice of the control vaccine iii) endpoints for phase I and II trials iv) the possibilities and advantages to carry out Phase IIa trials in Africa. v) standardisation of assays between centres (e.g. afro-immunoassay). vi) The use of the declaration of Helsinki for the African situation.

2001-06-22

R Sauerwein
With support from O. Leroy and D. Telgt