CPG 7909 adjuvant improves hepatitis B virus vaccine seroprotection in antiretroviral-treated HIV-infected adults

Background: HIV patients are vaccine hyporesponsive.

Methods: We evaluated CPG 7909, a synthetic oligodeoxynucleotide containing immunostimulatory CpG motifs, as an adjuvant to Engerix-B. A randomized, double-blind controlled trial was conducted to determine safety and hepatitis B virus (HBV) immunogenicity in adult HIV subjects on effective antiretroviral therapy. HBV-susceptible subjects, half of whom had failed previous vaccination, were vaccinated at 0, 1 and 2 months with a double dose of Engerix-B with/without (±) 1 mg CPG 7909. HBV immune subjects (anti-HBsAg titres ≥ 10 mIU/l) received either CPG 7909 alone or saline. Safety, anti-HBs titres and lymphocyte proliferation response (LPR) to HBsAg were assessed over 12 months.

Results: Vaccinations with Engerix B ± CPG 7909 were well tolerated locally and systemically. HIV suppression and CD4 cell counts were maintained. Anti-HBs titers were significantly higher in vaccinees receiving CPG 7909, for all time points after the second dose. Seroprotective titres (≥ 10 mIU/ml) by 6 and 8 weeks, and 12 months were found in 89, 89, and 100% of subjects receiving CPG 7909 compared to 53, 42, and 63% of controls respectively (P = 0.029, 0.005, and 0.008). HBsAg LPR was increased at all time-points up to 12 months after vaccination with addition of CPG 7909 (P < 0.05).

Conclusions: Addition of CPG 7909 achieves rapid, higher, and sustained HBV seroprotection and increases HBV-specific T helper cell response to HBV vaccine in HIV subjects. These results confirm a potential adjuvant role for CPG 7909 in vaccine hyporesponsive populations including those living with HIV.

Introduction

HIV-infected individuals are at increased risk of hepatitis B virus (HBV) infection due to shared modes of transmission. Furthermore, when infected with HBV, HIV-infected individuals are more likely to become chronic carriers ^[1] and develop complications such as cirrhosis and hepatocellular carcinoma ^[2,3]. Thus HIV-infected patients have a special need for effective HBV vaccination. By virtue of immune deficiency, HIV-infected individuals are hyporesponsive to vaccinations, including against HBV ^[2-4].

Adjuvants augment antigen-specific immune responses by physical localization and improved presentation of antigen, and by provocation of inflammatory or innate immune responses. Synthetic oligodeoxynucleotides (ODN) containing CpG motifs are potent stimulants of innate immune responses ^[5] that act as agonists for the innate immune receptor Toll-like receptor 9 (TLR9). The mechanism by which CpG promotes Th-1 specific immune responses is summarized elsewhere ^[6-9]. CpG ODN induce potent, T-helper type I-directed response when administered with a wide range of antigens ^[10], including HBsAg in mice ^[11-14] and in non-human primates ^[12,15,16].

Pre-clinical efficacy of CpG ODN in an immune-compromised situation has been demonstrated by improved humoral immune response to Engerix B in simian immunodeficiency virus-infected rhesus macaques ^[17]. Clinical trials of CPG 7909 have demonstrated this adjuvant to be safe and well tolerated in healthy, adult volunteers when administered with influenza ^[18] and HBV vaccines ^[19]. In the latter case, the addition of CPG 7909 to Engerix B stimulated earlier and higher titers against HBsAg (anti-HBs) than control subjects receiving Engerix-B alone ^[19]. We report the safety and efficacy of CPG 7909 as an adjuvant to Engerix-B in HIV-infected adults.

Methods

Study design

This phase Ib/IIa, randomized, controlled, double-blind vaccine study was conducted at The Ottawa Hospital Clinical Investigation Unit, Ottawa, Canada. The study was evaluated and approved by The Ottawa Hospital Research Ethics Board. Otherwise healthy, consenting HIV-seropositive volunteers, aged 18 to 55 years were eligible for enrolment if on highly active antiretroviral therapy (HAART; defined as at least three antiretrovirals including either a protease inhibitor or efavirenz) for a minimum of 6 months, with CD4 T-lymphocyte counts ≥ 200 × 10⁶cells/l and HIV RNA < 50 copies/ml for a minimum of 3 months. Once enrolled, all were tested for 12 weeks after first study injection with clinical follow-up for 12 months for repeated immunologic measurements.

Experimental groups

Anti-HBc, HBsAg and HBV DNA-negative subjects were enrolled in one of two arms of the study depending on their anti-HBs levels. Subjects with anti-HBs titres < 10 mIU/mlL were assigned to the 'susceptible' arm. Subjects were randomized to receive Engerix-B (control vaccine group) or Engerix-B plus CPG 7909 (experimental vaccine group). In the susceptible arm (n = 38), randomization was stratified between 'naïve' subjects (n = 19) with no detectable antibodies or history of previous vaccination, and 'hyporesponders' (n = 19) who had received at least three previous doses of a commercial HBV vaccine. Twenty subjects with anti-HBs titres ≥ 10 mIU/ml, by virtue of previous successful vaccination or resolved natural infection, were assigned to the 'immune' arm, and randomized to receive CPG 7909 (experimental non-vaccine) or saline (control non-vaccine).

Vaccines and control injections

All subjects were dosed at 0, 1 and 2 months. Susceptible subjects received two intramuscular injections, one into each deltoid, of an adult dose of Engerix-B (Glaxo SmithKline, Rixensart Belgium), thus a total of 40 μg HBsAg adsorbed to alum. Experimental vaccine subjects also received 0.5 mg CPG 7909 in 100 μl mixed with each injection of vaccine for a total dose of 1 mg CPG 7909. Control vaccine subjects received 100 μl of saline added to each of their bilateral vaccine injections. In all cases, the volume injected into each arm was 1.1 ml. Subjects in the immune arm received bilateral injections of either 0.5 mg CPG 7909 (total dose 1 mg) or saline, with each injection being given in a volume of 1.1 ml. Investigators and participants were aware of whether an individual was within the susceptible or immune arms, but not whether experimental or control injection was administered. The volume and appearance of each injection product were identical.

CPG 7909, a B-Class CpG ODN of sequence 5′-TCGTCGTTTTGTCGTTTTGTCGTT-3′ was synthesized with a wholly phosphorothioate backbone (Coley Pharmaceutical Group, Wellesley, Massachusetts, USA). This ODN contains three copies of the CpG motif GTCGTT that was optimized for stimulation of the human TLR9 receptor ^[15].

Safety evaluation

Safety and tolerance were assessed for 12 weeks. Intensity of adverse events was graded according to the common toxicity criteria (CTC version 2.0). Laboratory tests including a complete blood count, serum chemistry, urinalysis, and coagulation measures were performed at baseline and weeks 2, 4, 6, 8, 10, and 12, as well as 24 h after each dose at weeks 2, 4 and 8. Plasma HIV RNA (Chiron 3.0 system; Chiron, Emeryville, California, USA), CD4 and CD8 cell counts were assessed at these time points as well. Rheumatologic measures including CH50 (hemolysis), anti-dsDNA antibody, rheumatoid factor, and anti-nuclear antibody titre (ANA) were measured at baseline and at weeks 2, 4, 6, 8, 10, and 12.

Immunological evaluations

To assess serological responses to the experimental and control vaccines, blood samples were obtained at screening, weeks 2, 4, 6, 8, 10, 12 on study and at weeks 24 and 48 in clinical follow-up. Serological responses were tested in triplicate by the Abbott AUSAB-EIA (Abbott, Abbott Park, Illinois, USA). Anti-HBs titers were expressed in mIU/ml by World Health Organization-defined standards. Seroconversion and seroprotection were defined as anti-HBs titers of ≥ 1 and ≥ 10 mIU/ml, respectively.

To assess HBV- or HIV-specific cell-mediated immune responses, whole blood was collected from all subjects at baseline and at weeks 4, 8, 12 on study and at weeks 24 and 48 weeks in clinical follow-up. Peripheral blood mononuclear cells (PBMC) were isolate by centrifugation over a Ficoll gradient. Lymphocyte proliferation, as measured by ³H-thymidine incorporation after 6 days of incubation of PBMC with HBsAg (2.5 μg/ml), HIV p24 antigen (10 μg/ml) or media, was performed ^[20].

Statistical analysis

The number of patients in this study was chosen to identify important differences in measures of safety between those patients receiving CPG ODN 7909 (n = 29) and those receiving placebo (n = 29). Measures of safety were compared between experimental and control groups by chi-squared test. HIV RNA and CD4 T-lymphocyte counts were described by mean and 95% confidence interval (CI) for each study group. The change from baseline in HIV RNA and CD4 T-lymphocyte count was compared between experimental and control groups by Student's t-test.

The HBV-specific immunologic measures, namely anti-HBs titers and lymphoproliferative responses were also summarized using group-wise means and 95% CI. Geometric mean anti-HBs titres were compared by the Mann-Whitney U-test. The proportion of subjects achieving seroconversion, seroprotection, and high titre seroprotection (anti-HBs titre ≥ 100 mIU/ml) was evaluated by chi-squared test. SPSS 11.0 (SPSS Inc., Chicago, Illinois, USA) was used for these calculations. Primarily, all available data was included in analysis (intention-to-treat), without imputing missing values. Serologic data after a missed vaccine dose was not included in secondary analysis (per protocol). Change from baseline in proliferation counts after re-stimulation with HBsAg or HIV p24 were used as dependent variables in a repeated measures analysis of variance. The SAS mixed procedure (SAS Institute, Cary, North Carolina, USA) was used with an autoregressive covariance structure. Since this method requires equal time intervals between measurements, two analyses were done for each dependent measure, one using weeks 4, 8 and 12 and a second using weeks 24 and 48. Three effects were modeled: treatment group, time and the treatment group by time interaction.

Results

Population characteristics

This study was conducted between January 2001 and August 2002. Eighty-five adult HIV seropositive volunteers were screened. Twenty-seven subjects were excluded due to: withdrawal of consent (n = 8), anti-HBc positivity (n = 12), history of malignancy (n = 4), elevated HIV RNA level (n = 2), and elevated liver enzymes (n = 1). Thirty-eight HBV-susceptible subjects (19 HBV seronegative vaccine naive, 19 HBV vaccine failures) and 20 HBV-immune subjects received at least one study injection. All but one subject (see below) completed the study (Table 1).

Table 1

Safety and tolerance

During the course of the study, most subjects reported at least one adverse event, most of which were mild to moderate in intensity. The most common adverse events included local injection site reactions (pain, erythema, bruising) and influenza-like symptoms (fatigue, pyrexia, chills, arthralgia, headache) (Table 2). Injection site pain occurred more frequently in those receiving CPG 7909 alone than saline placebo (P = 0.02).

Table 2

Serious adverse events in three individuals included one case of unstable angina 2 months after the last vaccination with Engerix-B and CPG 7909 with a myocardial infarction 4 months after that; one with bacteremia with portal vein thrombosis 7 months after receiving Engerix-B alone;, and one with laparoscopy for chronic pelvic inflammatory disease 16 days following vaccination with Engerix-B and CPG 7909.

Two subjects did not complete three doses. One male subject developed generalized urticaria 1 day following his second Engerix-B plus CPG 7909 injection. This resolved in conjunction with a short course of oral prednisone. This subject was followed for the entire study duration without further injection. The third injection was withheld in a male receiving CPG 7909 alone with non-adherence to HAART and rebound HIV plasma viremia.

Laboratory

HIV RNA suppression was maintained over the 12-week course of this study in all but one subject (see above). Fluctuations in HIV RNA, ranging from 51 to 268 copies/ml, occurred in fourteen individuals and were without patterns in time, treatment groups or between susceptible versus immune subjects. HIV RNA was < 50 copies/ml on the next follow-up evaluation in eleven of these patients and in all fourteen by the measure thereafter.

A transient decline in total mean lymphocytes (data not shown) and mean CD4 T-cell counts (-199, -244, and -287 × 106 cells/l the day following the first, second and third Engerix-B plus CPG 7909 vaccinations, respectively) was noted in recipients of CPG 7909 (Fig. 1). These clinically inconsequential declines returned to baseline at the next measure.

Fig. 1

Transient elevations above normal range for anti-dsDNA were noted in two subjects receiving Engerix-B plus CPG 7909 and in two subjects receiving CPG 7909 alone. In no case was there a concurrent positive ANA or related clinical findings of autoimmunity.

Vaccine immunogenicity

Geometric mean anti-HBs titers (GMT) were higher for all measured time points after the second injection for susceptible subjects receiving Engerix-B vaccine with CPG 7909 (both vaccine-naive and prior vaccine failure combined) than those receiving control vaccine (Fig. 2). This was significant at weeks 6, 8 and 10 (P < 0.01) and weeks 12, 24, and 48 (P < 0.05) (ITT). GMT for hepatitis B vaccine-naïve, susceptible subjects were higher at weeks 6, 8, 10 and 12 weeks (P < 0.05) in recipients of Engerix-B vaccine plus CPG 7909. GMT for prior hepatitis B vaccine failure subjects control were higher over the entire course of this study in Engerix-B plus CPG 7909 recipients reaching significance at week 10 (P < 0.05).

Fig. 2

All subjects seroconverted by week 10. Detectable antibody persisted to 48 weeks in all vaccine plus CPG 7909 recipients (19/19) and 89% (17/19) of controls. Seroprotective anti-HBs titers were achieved more rapidly and in more subjects receiving vaccine plus CPG 7909 than control vaccine (Fig. 3). All CPG 7909 recipients (n = 19) and 17/19 (89%) control subjects achieved seroprotection by 2 weeks after the third and final injection (10 weeks). Seroprotective titres remained in all CPG 7909 recipients at 48 weeks versus 12 of 19 (63%) for controls (P = 0.008). Similar trends were seen for CPG 7909 versus controls when compared within the subgroups of vaccine-naive and hyporesponder subjects. Addition of CPG 7909 to Engerix B resulted in more subjects achieving high anti-HBs titre (≥ 100 mIU/l) as early as week 8 (13/19 CPG versus 5/19 control; P = 0.022), and maintaining them as long as 48 weeks (14/19 CPG versus 7/19 control, P = 0.049). In HBV-immune subjects, who did not receive vaccine, anti-HBs titres remained unchanged over the course of this investigation, whether or not they received CPG 7909 or saline (data not shown).

Fig. 3

Lymphocyte proliferative responses to ex vivo restimulation with HBs showed higher mean counts per minute in Engerix-B plus CPG 7909 recipients compared to controls at all time points measured (weeks 4, 8, 12, 24, 48). By repeated measures analysis of variance of the week 4, 8 and 12 data, only CPG 7909 group was identified as a significant predictor of proliferative response (P = 0.0039, n = 38). In week 24 and 48 data, the interaction between group and time was significant (P = 0.0271, n = 38) indicating that the increase in counts from baseline between weeks 24 and 48 was greater for CPG 7909 recipients. The same individuals were evaluated in both of the above analyses. Secondary analyses confined to individuals for whom measures were available for all visits confirmed significance (P = 0.0024 for weeks 4, 8, and 12, n = 32; P = 0.034 for weeks 24 and 48, n = 33). Cellular immune response measures to HIV p24 were assessed as well. No significant difference in proliferation to HIV p24 was identified between Engerix-B plus CPG 7909 recipients and controls in this population of subjects with maximal HIV RNA suppression.

Discussion

This study demonstrates again that CPG 7909 is safe and well-tolerated as a vaccine adjuvant, in this study being administered to HIV seropositive adults on effective anti-HIV therapy. We have published similar findings in healthy, adult volunteers receiving Engerix-B plus CPG 7909 adjuvant ^[19]. The frequency of local injection site symptoms likely reflects the proinflammatory effect of TLR9 stimulation by CpG ODN ^[5]. Comparable local reaction is seen with monophosphoryl lipid A (MPL) added to the same Engerix-B HBV vaccine ^[21,22] or a CpG ODN of a different sequence (1018 ISS) mixed with HBsAg but without alum ^[23].

CPG 7909 significantly enhanced the immunogenicity of Engerix-B in HIV-infected subjects. This appears consistent in both vaccine-naive and previous vaccine failures. It is well accepted that antibodies alone are sufficient for protection against HBV, and in the present study, the humoral response was faster, higher, and lasted longer. The durability of detectible anti-HBs may address the issue of waning seroprotection in vaccine hyporesponsive populations including those with HIV.

Several recent studies have shown that stimulation of TLR9 by CpG ODN can transcriptionally activate the HIV LTR, resulting in increased HIV replication in vitro, in transgenic mice, and apparently in vivo in humans ^[24-26]. These reports suggest potential danger from in vivo stimulation of TLR9 in humans. However, we found no increase in viral load in the subjects in our trial, who were maintained on HAART throughout the trial. We conclude that in vivo stimulation of TLR9 with the low doses of CPG 7909 used in this vaccine adjuvant study does not induce detectably increased plasma viremia in the presence of HAART.

Other novel adjuvants including monophosphoryl lipid A (MPL) ^[27], thymopentin ^[28], AM3 (Immunoferon) ^[29], and levamisole ^[30] have been tested with HBV vaccines in hyporesponder populations. In comparison to these adjuvants, CPG 7909 generally achieved superior results in time to seroconversion and seroprotection, absolute anti-HBs titres achieved, and durability of response. The increased immunogencity of CpG 7909 as an adjuvant points to the utility of TLR9 as a vaccine adjuvant target. These results in subjects with compromised T-helper function further suggest that in vivo TLR9 stimulation in humans can substitute for the normal requirement for T-cell help in generating humoral immunity.

The addition of CPG 7909 to Engerix-B resulted in increased HBsAg-specific cell-mediated immune (CMI) responses as measured by lymphocyte proliferation. Most other studies with novel adjuvants and HBV vaccines have not evaluated CMI ^[23,28,29]. Enhancement of CMI has been reported by addition of MPL to Engerix-B in healthy, adult volunteers ^[21] and in healthy, adult vaccine non-responders ^[31]. The increased proliferation observed in this study, which presumably reflects T-cell help, may have contributed to the augmented humoral response seen with the inclusion of CpG. Improved CMI is also recognized as an important factor in clearance of acute infection ^[32] and in immunologic control of chronic HBV disease ^[33] which suggest a potential role for CPG 7909 in therapeutic vaccines for HBV.

This study is limited by its relatively small size and short duration of evaluation. Long-term follow-up to determine duration of anti-HBs titers and to evaluate protection from natural infection continue. Safety of CPG 7909 in large numbers of patients is yet to be established. These concerns not withstanding, this randomized controlled trial demonstrates that improved seroprotection and higher anti-HBs titers can be achieved rapidly with the addition of CPG 7909 to an accelerated double-dose Engerix-B vaccination program in treated HIV seropositive vaccinees, and that protection is sustained for at least 48 weeks. The enhanced humoral and cell-mediated immunity produced with CPG 7909 justify evaluation of this adjuvant in the realm of therapeutic HBV vaccination and other vaccine development, including HIV.

Acknowledgements

The authors wish to thank Dr D. P. Kilby for his assistant in the recruitment of patients for this study, and Dr J. Clinch for statistical consultation.

Sponsorship: This study was supported by the AIDS Program Committee of the Ontario Ministry of Health. DWC is a Career Scientist of the Ontario HIV Treatment Network. This trial was approved and supported by the Canadian HIV Trials Network (CTN-158). C.L.C. received supported as a Canadian HIV Trials Network / Canadian Institutes of Health Research Fellow and as an Ontario HIV Treatment Network Scholarship recipient.

Conflict of interest: H.L.D., M.L.M., S.M.E. and A.M.K. are employees of Coley Pharmaceutical Group, which provided CPG 7909, monitoring of adverse event reportage, data management and statistical analysis.