This single blind controlled phase I study was conducted at the Aster-Cephac S.A. facility in Paris. Ethical Review Board approval was obtained and each patient provided voluntary informed consent. Eligible subjects were healthy adult HBV vaccine naïve volunteers, aged 18–55, with no serologic evidence of previously resolved or current HBV infection. However, three of these were later found to be seroconverted (but not seroprotected) at baseline in the post study HBsAg antibodies determination (subjects #019, 035 and 044). Other exclusion criteria included liver enzyme levels outside the normal range, chronic HIV or HCV infection, or evidence of any other clinically significant acute or chronic disease. Subjects receiving immune suppressive medication, and those diagnosed with an immune or autoimmune dysfunction were not considered for this study. Female subjects had to have gone through the menopause for a least one year, as evidenced by lack of menstruation for the last 12 months and hormones (FSH, estradiol) blood level measurement at screening confirming menopausal status.

For the production of a clinical batch of IMP321, CHO DHFR^- cells were transfected with a plasmid coding for the D1-D4 extra-cellular domains of human LAG-3 fused to the Fc tail of a human IgG1 11. A production clone was selected after amplification in methothrexate. The final container clinical batch used in the present study has a concentration of 1.1 mg/ml IMP321 (a 200 kDa dimeric protein) and 0.09 EU/mg endotoxin, 0.4 ng/ml DNA and 6 ng/ml host cell protein contents. Experimental vaccines contained 10 μg yeast-derived recombinant HBsAg (provided by Rhein Biotech GmbH, Düsseldorf) alone or with 3, 10, 30, 100 μg IMP321 (hLAG-3Ig). All vaccines were prepared by an unblinded pharmacist at the trial site and were administered within 1 h of mixing using a 200 μl injection volume. Each subject received three sub-cutaneous (s.c.) doses at 0, 4 and 8 weeks. The first and the third injections were done in the deltoid area of the dominant arm. The second injection was done in the deltoid area of the non dominant arm. Subjects in another comparative arm received an adult dose (1 mL) of Engerix-B^® (GlaxoSmithKline, Rixensart, BE) that contains 20 μg of alum-absorbed yeast-derived recombinant HBsAg, which was administered intramuscularly.

Subjects were enrolled sequentially into four cohorts according to dose level of IMP321. Within cohorts, subjects were randomised to receive an experimental vaccine or control HBsAg alone in a 4:1 ratio. A total of 48 subjects were immunized according to the planned three administration schedules, 8 receiving control vaccines, 8 receiving Engerix-B^® and 32 receiving experimental vaccines with IMP321 (n = 8 in each group). Two subjects were prematurely discontinued from the study after the first injection and were replaced.

All subjects who received a dose of the study drug were included in the safety evaluation (n = 50). Adverse effects were identified by clinical examination at baseline and at the following times post administration: first dose at 4 h, 48 h, one week, and 4 weeks (just prior to second dose); second dose at 4h, 48 h, one week and 4 weeks (just prior to third vaccine dose); third dose at 4 h, 48 h, one week and 4 weeks. In addition, vital signs (blood pressure and pulse rate) and oral body temperature were recorded at pre-dose, 0.5 h, 1 h, 1.5 h, 2 h and 4 h post-dosing as well as 48 h and 1 week after each injection. Laboratory tests included a complete blood count, serum chemistry, liver and renal function, and coagulation measures. Rheumatoid factors, anti-nuclear antibody titres (ANA) and anti-IMP321 antibodies were measured at baseline and weeks 12.

Immunogenicity results were analysed using the population which completed the study (n = 48). To assess anti-HBsAg responses, blood samples obtained at baseline and 8 and 12 weeks after the initial vaccine dosing, were allowed to clot at room temperature for 15 minutes. Samples were centrifuged at 1,500 g at about 4°C for 10 minutes and the serum was aliquoted and stored in airtight stoppered polypropylene tubes at -20°C. Sera were tested by the Abbott AUSAB-MEIA (Abbott, Abbott Park, IL, USA) and anti-HBs titres were expressed in mIU/mL based on comparison with standards defined by the World Health Organization (WHO). A protective titre was defined as ≥ 10 mIU/mL. The commercially available hepatitis B vaccine Engerix^®-B (20 μg HBsAg adsorbed on alum) was used to ensure that our 3-months protocol schedule was able to induce antibodies in most subjects. Geometric mean of titres (GMT) was calculated using the formula 10^{mean [log (Ab titers)]} for each group at each time point. Seronegative subjects have been given the arbitrary value of 1 mIU/mL for GMT calculation.

The analyses for safety and tolerability parameters were performed on all randomised subjects who received at least one dose of study medication and who had post-dose safety information (n = 50). Immunogenicity results were analysed on the population which completed the study (i.e. subjects who received 3 injections and had their post-study visit) (n = 48). Anti-HBsAg titres measured in mIU/mL were expressed as geometric mean titres (GMT) for each group. The differences between GMTs achieved at a given time point for each of the HBsAg plus IMP321 groups or the Engerix-B^® group were compared with the HBsAg alone group by Student's two-sided t-test. The proportions of subjects achieving seroconversion (anti-HBsAg ≥ 1 mIU/mL) and seroprotection (anti-HBsAg ≥ 10 mIU/mL) were compared in the combined IMP321 groups and in the Engerix-B^® group versus the control HBsAg alone group.

Before evaluating HBsAg-specific T cell responses to follow the efficacy of the immunization protocol, validation experiments were performed on four PBMCs samples collected from volunteers who had been previously immunized with commercial hepatitis B vaccine. PBMCs were thawed and stimulated using a set of 22 20-mers peptides (overlapping by 11 aa) that span the entire HBsAg protein sequence (1 μM of each peptide) or cultured with the vehicle (DMSO), in the presence of FastImmune CD28/CD49d costimulation cocktail (BD Biosciences) for 18 h and in the presence of brefeldin A (BD Biosciences) for the last 16 h. In another series of experiments, PBMC samples from three other donors were stimulated with a cytomegalovirus (CMV) pp65 peptides pool (1.75 μg/ml, BD Biosciences) or Staphylococcus Enterotoxin B (SEB, 1 μg/ml, Sigma Aldrich) in the same conditions. PBMCs unstimulated or stimulated with peptides or SEB were fixed, permeabilised using CytoFix/CytoPerm, stained with fluorochrome-conjugated CD3-PerCP-Cy5.5, CD4-PE-Cy7, CD8-APC-Cy7, IFN-γ-FITC, TNF-α-APC and IL-2-PE antibodies and extensively washed with PermWash buffer (all from BD Biosciences). Cells were then analysed using a 6-colour FACSCanto flow cytometer (BD Biosciences) to determine the percentage of CD3⁺CD4⁺ and CD3⁺CD8⁺ cells expressing IFN-γ, TNF-α and/or IL-2. The percentage of cells expressing cytokines in unstimulated conditions was subtracted from the percentage of cells obtained after peptide stimulation. Following completion of the protocol, a series of samples grouping the whole kinetics for each individual included in the clinical trial were thawed and analysed after 18 h of ex-vivo restimulation using the same set of HBsAg peptides. Cells were fixed, permeabilised and stained as above. A very large number of PBMCs were analysed (as an average 0.9 × 10⁶ cells) by flow cytometry to secure the validity of small percentages and/or differences. Results following FACS analysis were defined as the difference in response to HBsAg-peptides at D29, D36, D57 or D85 versus D1. The confidence interval depended on the numbers of relevant events (CD3⁺CD4⁺ or CD3⁺CD8⁺ events) collected in each sample, the amount of background stimulation at D1 and difference between D1 and D29, D36, D57 or D85 time points. This difference was significant with a power of 90 % (p < 0.05) if the number of CD4⁺or CD8⁺ cells collected was larger than calculated CD4⁺ or CD8⁺ events using the formula:

2 × ( ( D ÷ 100 + D 1 ÷ 100 ) ÷ 2 ) × ( 1 − ( ( D ÷ 100 + D 1 ÷ 100 ) ÷ 2 ) × 8.6 ) ( D ÷ 100 − D 1 ÷ 100 ) 2 MathType@MTEF@5@5@+=feaafiart1ev1aaatCvAUfKttLearuWrP9MDH5MBPbIqV92AaeXatLxBI9gBaebbnrfifHhDYfgasaacH8akY=wiFfYdH8Gipec8Eeeu0xXdbba9frFj0=OqFfea0dXdd9vqai=hGuQ8kuc9pgc9s8qqaq=dirpe0xb9q8qiLsFr0=vr0=vr0dc8meaabaqaciaacaGaaeqabaqabeGadaaakeaadaWcaaqaaiabikdaYiabgEna0kabcIcaOiabcIcaOiabbseaejabgEpa4kabigdaXiabicdaWiabicdaWiabgUcaRiabbseaejabigdaXiabgEpa4kabigdaXiabicdaWiabicdaWiabcMcaPiabgEpa4kabikdaYiabcMcaPiabgEna0kabcIcaOiabigdaXiabgkHiTiabcIcaOiabcIcaOiabbseaejabgEpa4kabigdaXiabicdaWiabicdaWiabgUcaRiabbseaejabigdaXiabgEpa4kabigdaXiabicdaWiabicdaWiabcMcaPiabgEpa4kabikdaYiabcMcaPiabgEna0kabiIda4iabc6caUiabiAda2iabcMcaPaqaaiabcIcaOiabbseaejabgEpa4kabigdaXiabicdaWiabicdaWiabgkHiTiabbseaejabigdaXiabgEpa4kabigdaXiabicdaWiabicdaWiabcMcaPmaaCaaaleqabaGaeGOmaidaaaaaaaa@7456@

where D is the percentage of CD3⁺ CD4⁺ or CD3⁺CD8⁺ cells expressing at least one cytokine on D29, D36, D57 or D85 upon stimulation and D1, the percentage of CD3⁺ CD4⁺ or CD3⁺CD8⁺ cells expressing at least one cytokine on D1.

After completion of the protocol, PBMC harvested from a HLA-A2⁺ donor on D1, D57 and D85 were thawed and cultured with two HLA-A2-restricted HBsAg peptides (GLSPTVWLSV and WLSLLVPFV, 1 μM each) in the presence of IL-2 (20 IU/ml) for 10 days. Fresh autologuous PBMC were loaded with the two peptides and added to the culture for additional 10 days. Fresh IL-2 was added every two days during the two rounds of stimulation. Cells were then incubated with the two HBsAg peptides/HLA-A2 pentamers (HLA-A*0201) conjugated to PE, washed, stained with CD3-PerCP-Cy5.5, CD4-APC-Cy7, CD8-FITC, CD14-APC antibodies and analysed by flow cytometry. After exclusion of CD14⁺ monocytes, the binding of pentamers on CD3⁺CD8⁺ cells was determined.

This study was conducted between May 2005 and December 2005. A total of 113 subjects were screened, of which 50 were enrolled and received at least one dose of vaccine. Baseline characteristics and demographics were evenly distributed among the six cohorts, with the exception of age in the HBsAg plus 10 μg IMP321 group (Table 1). All but two subjects completed the study. One subject in the HBsAg + 3 μg IMP321 and one subject in the HBsAg + 10 μg IMP321 withdrew from the study after the first immunization for personal reasons. They were replaced by 2 other subjects.

Overall, IMP321 plus HBsAg was characterised by a good tolerability profile at the four doses tested. A lower incidence of subjects experiencing AEs was reported after injection of IMP321 plus HBsAg (38 %) or HBsAg alone (25 %) than after injection of Engerix^®-B (62.5 %). The most common observed non-serious adverse events included local injection site pain (4/35) and erythema (2/35), as well as systemic symptoms such as nausea (2/35) and headache (5/35) (see Table 2). Injection site pains and erythema were considered certainly related to the study drugs, whereas nausea and headache were considered possibly related. Most of these AEs were of mild to moderate intensity and resolved without any corrective treatment. Following vaccine injection, oral temperature, blood pressure, and pulse rate remained stable from baseline to hour 4, as well as on day 3 and day 8 post-dosing (data not shown). One subject from the HBsAg plus 100 μg IMP321 group developed a pruritus and a papular rash 2 hours after the first injection, which could be indicative of an allergic reaction; the symptoms were transient, not reproduced after the following injections and no medical or pharmacological intervention was required.

There were no consistent or dose-related changes in biochemical haematological or rheumatological measures (data not shown). Moreover, antibodies to IMP321 were not detected in sera collected from subjects on D29, D36, D57 and D85 (not shown). Altogether, these data show that the injections of IMP321 were well tolerated with few reported non serious AEs and no sign of induced autoimmunity.

Before evaluating HBsAg-specific T cells response by intracellular staining to detect cytokines in T cells by flow cytometry analysis after short term ex-vivo stimulation with a pool of 22 HBsAg overlapping peptides (20 aa overlapping by 11), standard operation procedures were established and fixed. First, blood samples from 4 different donors previously immunized by a commercial hepatitis B vaccine were collected and PBMCs independently purified and frozen by two different operators. PBMCs were then stimulated by the HBsAg peptide pool for 18 hours in the presence of brefeldin A and stained wit