Expert Review of Vaccines Downloaded from informahealthcare.com by Nyu Medical Center on 07/22/15 For personal use only.
Review
Improving immunogenicity and efficacy of vaccines for genital herpes containing herpes simplex virus glycoprotein D Expert Rev. Vaccines 13(12), 1475–1488 (2014)
Sita Awasthi1, Carolyn Shaw2 and Harvey Friedman*3 1 522F Johnson Pavilion, Infectious Disease Division, University of Pennsylvania, Philadelphia, PA 19104-6073, USA 2 506 Johnson Pavilion, Infectious Disease Division, University of Pennsylvania, Philadelphia, PA 19104-6073, USA 3 522E Johnson Pavilion, Infectious Disease Division, University of Pennsylvania, Philadelphia, PA 19104-6073, USA *Author for correspondence: Tel.: +1 215 573 8432 Fax: +1 215 349 5111 [email protected]
No vaccines are approved for prevention or treatment of genital herpes. The focus of genital herpes vaccine trials has been on prevention using herpes simplex virus type 2 (HSV-2) glycoprotein D (gD2) alone or combined with glycoprotein B. These prevention trials did not achieve their primary end points. However, subset analyses reported some positive outcomes in each study. The most recent trial was the Herpevac Trial for Women that used gD2 with monophosphoryl lipid A and alum as adjuvants in herpes simplex virus type 1 (HSV-1) and HSV-2 seronegative women. Unexpectedly, the vaccine prevented genital disease by HSV-1 but not HSV-2. Currently, HSV-1 causes more first episodes of genital herpes than HSV-2, highlighting the importance of protecting against HSV-1. The scientific community is conflicted between abandoning vaccine efforts that include gD2 and building upon the partial successes of previous trials. We favor building upon success and present approaches to improve outcomes of gD2-based subunit antigen vaccines. KEYWORDS: acyclovir • complement • genital herpes • glycoprotein C • glycoprotein D • glycoprotein E • HSV • IgG Fc receptor • immune evasion • vaccine
A half billion people worldwide are infected with herpes simplex virus type 2 (HSV-2) [1]. The infection cycle for genital HSV includes replication at the site of entry in the genital mucosa or skin, transport of viral components, including DNA to sensory neurons in the dorsal root ganglia (DRG), establishment of latent infection in the DRG with frequent reactivation that results in viral components traveling back down the nerve fibers to produce recurrent genital infections. Some individuals have multiple recurrences of genital lesions; however, the majority of subjects have subclinical (asymptomatic) infection that is manifested by frequent short episodes of genital shedding of HSV-2 that poses a risk for transmitting infection to their partners or to newborns delivered through an infected birth canal [2–4]. Genital herpes infection is emotionally distressing for infected subjects and their partners, is
informahealthcare.com
10.1586/14760584.2014.951336
potentially life threatening for exposed newborns, and importantly, increases the risk of acquiring and transmitting HIV infection by three- to fourfold [5,6]. No vaccine is currently available for prevention or treatment of genital herpes. Developing these vaccines is a major public health priority. HSV glycoprotein D (gD) was the first well-characterized HSV entry molecule, which explains why many vaccine trials performed to date have included gD2 subunit antigen. The goal is to induce neutralizing antibodies that block infection. As discussed below, the gD2 subunit antigen vaccine trials failed to meet their primary end points. This review focuses on approaches to improve gD2-containig subunit antigen vaccines. Other approaches that use attenuated live viruses, HSV DNA or viral vectors to express HSV antigens are not considered in this review.
2014 Informa UK Ltd
ISSN 1476-0584
1475
Review
Awasthi, Shaw & Friedman
Human trials involving HSV-2 gD2 subunit antigen
Expert Review of Vaccines Downloaded from informahealthcare.com by Nyu Medical Center on 07/22/15 For personal use only.
Prophylactic vaccine trials using gD2 antigen
One of the earliest double-blind, placebo-controlled trials of HSV vaccines for prevention of genital herpes used glycoproteins purified from HSV-2-infected chick embryo fibroblast cells [7]. Extremely low ELISA titers to HSV-2 gB2 and gD2 were obtained after three immunizations. The vaccine did not induce neutralizing antibodies or provide protection to high-risk subjects who had partners with genital herpes. The primary end point of preventing HSV infection was not achieved. Subsequently, large placebo-controlled, double-blinded HSV vaccine trials have either used HSV-2 gD2 as a monovalent subunit antigen vaccine (GlaxoSmilthKline [GSK] Biologicals, Rixensart, Belgium) or gD2 with gB2 as a bivalent vaccine (Chiron Corp. Emeryville, CA) [8–10]. Both gB2 and gD2 serve essential roles in virus entry and are expressed on the virus envelope, making them accessible to neutralizing antibodies [11]. The Chiron studies enrolled HSV-2 seronegative patients into multicentered trials that evaluated recombinant gB2 and gD2 antigen each used at 30 mg. The antigens were administered three-times over a 6-month period with MF59, an oil and water adjuvant comprising squalene, polysorbate 80 and sorbitan trioleate [10]. The results of two studies were reported in a single paper by the Chiron investigators [10]. One study enrolled HSV-2 seronegative partners of HSV-2-infected subjects, while the other enrolled HSV-2 seronegative persons attending sexually transmitted disease clinics. The primary end point of both studies was the time to acquisition of herpes infection, with or without genital lesions, as defined by seroconversion to herpes antigens not contained in the vaccine or by virus isolation in culture. An analysis of both studies together indicated that the vaccine group had 50% lower acquisition rates than placebo recipients over the first 5 months of the trial; however, over the entire course of the study, the vaccine was not efficacious despite inducing high titers of neutralizing antibodies. Multicentered, double-blind, randomized trials were performed by GSK and academic investigators in 2002 and 2012 using recombinant gD2 antigen formulated with AS04 adjuvant, which contains alum (aluminum hydroxide) and 3-Odeacylated-monophosphoryl lipid A (MPL) [8,9]. The 2002 publication reported two studies involving immunization of subjects whose regular sex partner had genital herpes [8]. In the first study, the subjects were seronegative for HSV-1 and HSV-2, while in the second study subjects were of any HSV serologic status. The primary end point was preventing genital herpes lesions in all subjects in the first study and in female subjects who were HSV-2 seronegative in the second study. A secondary end point in the second study was prevention of genital disease in female subjects who were seronegative at enrollment for HSV-1 and HSV-2. The gD2 vaccine was administered three-times over 6 months using 20 mg per dose. Overall, the vaccine did not provide significant protection against genital herpes lesions; however, the analysis showed approximately 74% efficacy in HSV-1 and 1476
HSV-2 doubly seronegative females, who met the criteria established as a secondary end point. The vaccine was not efficacious in seropositive women or in men, regardless of serologic status. These studies helped shape the experimental design of a subsequent vaccine trial reported in 2012, the Herpevac Trial for Women, and raised the possibility that protection may differ in males and females [9]. The Herpevac Trial subjects were HSV-1 and HSV-2 doubly seronegative women who were 18–30 years old. This population differs from the prior study that enrolled HSV-2 discordant couples, which makes direct comparisons between the two studies difficult. The recombinant gD2 antigen was administered at 20 mg with MPL/alum as adjuvants at times 0, 1 and 6 months. The primary end point was genital disease caused by HSV-1 or HSV-2 between months 2 and 20 after the first immunization. The overall vaccine efficacy was 20% and the primary end point was not achieved. The vaccine elicited ELISA and neutralizing antibody titers, although neutralizing titers were low and the median level declined to undetectable levels by month 16. No significant protection was observed against HSV-2; however, a subset analysis revealed 35% protection against HSV-1 infection (genital and/or oral) and 58% protection against HSV-1 genital disease compared with the control group (both statistically significant). This finding is important, since HSV-1 is now more common than HSV-2 as the cause of first-time genital herpes disease [12]. Using sera taken 1 month after the third immunization from subjects enrolled in the Herpevac Trial, neutralizing antibody titers were noted to be significantly higher (3.5-fold higher) to HSV-1 than HSV-2, which potentially explains the results of the Herpevac Trial [13]. Neighboring glycoproteins on HSV-2 appear to shield gD2 from neutralizing antibodies more effectively than neighboring glycoproteins on HSV-1 [13]. Based on low titers of neutralizing antibodies generated in subjects immunized in the Herpevac Trial, questions have emerged whether 20 mg of gD2 antigen was too little; however, in a separate study, no significant differences were detected in gD2 ELISA titers, neutralizing antibody titers and cellular immune responses comparing 20, 40 and 80 mg concentrations of gD2 [14]. Taken together, the results suggest that gD2 concentrations >80 mg may be required for protection against HSV-2, that gD2 amino acids 1–281 (of 393 amino acids) that were included in the vaccine formulation may be too limited in scope to be protective, that MPL/alum may not be an optimum adjuvant or that different antigens and adjuvants are needed to prevent HSV-2 genital infection. Therapeutic vaccine trials using recombinant gD2 antigen
The term therapeutic vaccine refers to immunization as a form of treatment to reduce the frequency and/or severity of recurrent genital herpes disease or recurrent genital shedding of HSV DNA. A double-blind, placebo-controlled trial of gD2 subunit antigen was performed to assess immunization as therapy for genital herpes in subjects reporting 4–14 genital herpes outbreaks annually [15]. Recombinant gD2 subunit Expert Rev. Vaccines 13(12), (2014)
Expert Review of Vaccines Downloaded from informahealthcare.com by Nyu Medical Center on 07/22/15 For personal use only.
Improving immunogenicity & efficacy of vaccines for genital herpes containing HSV gD
antigen at 100 mg per dose was administered with alum twice at 2 months interval or alum alone was used as a control. The gD2 vaccine recipients boosted their neutralizing antibody and gD2 ELISA titers, and had fewer virologically confirmed recurrences over a year of observation. These encouraging results were followed by a study using recombinant gD2 combined with recombinant gB2 subunit antigen as a bivalent vaccine each used at 10 mg with MF59 as adjuvant [16]. Subjects received four immunizations over 14 months with the bivalent vaccine or MF59 alone as a control. The bivalent vaccine induced neutralizing antibodies and decreased the duration and severity of the first outbreak; however, the overall recurrence rate was not significantly reduced. It is important to note that in one study gD2 was used at a concentration of 100 mg, while in the follow-up study gD2 and gB2 each were given at 10 mg raising the possibility that suboptimal concentrations of antigen were used in the second study. Immune correlates of protection against genital herpes
Immune correlates of protection refer to immune responses that statistically correlate with vaccine efficacy; however, correlation does not necessarily imply a mechanism [17]. For example, neutralizing antibody titers may be a mechanistic correlate of protection, while ELISA titers are likely non-mechanistic correlates that serve as surrogates for an antibody response that mediates protection, such as neutralizing antibody or antibody-dependent cellular cytotoxicity (ADCC). In general, the antigens selected for immunization will influence the correlates of protection that emerge. If a successful vaccine includes antigens involved in virus entry, then correlates of protection are likely to be neutralizing antibodies, while if the candidate vaccine contains immunogens aimed at inducing potent T-cell responses, then CD4+ and/or CD8+ T cells are likely to correlate with protection. Humoral immunity may be of greater significance for a prophylactic HSV-2 vaccine, while T-cell immunity may be essential for a successful therapeutic vaccine. For pathogens that infect at mucosal surfaces, correlates of protection may involve locally produced IgA antibodies or IgG that reaches the mucosal surface by transcytosis from serum, as demonstrated by protection against influenza virus [18]. For a complex pathogen, such as HSV-2, many aspects of immunity are likely to contribute to protection; therefore, more than one correlate may emerge. Varicella vaccine provides an example of a human herpes virus where correlates of protection have been assessed. Antibody responses and cell-mediated responses both correlate with protection, suggesting the importance of multiple components of immunity [19–23]. In support for a role for antibodies in protection against varicella and HSV, passive administration of varicella-zoster immune globulin provides protection against varicella infection and transplacental transfer of maternal antibodies protects infants exposed to HSV-2 at birth [24,25]. The Herpevac Trial for women had the unexpected result of protecting HSV-1 and HSV-2 doubly seronegative women against genital disease caused by HSV-1 but not HSV-2 [9]. informahealthcare.com
Review
ELISA antibody responses, rather than CD4+ or CD8+ T-cell responses, correlated with protection [9,26]. Among gD2 vaccine recipients, HSV-1 infection rates were 2.5% in subjects with low ELISA titers and
Review
Improving immunogenicity and efficacy of vaccines for genital herpes containing herpes simplex virus glycoprotein D Expert Rev. Vaccines 13(12), 1475–1488 (2014)
Sita Awasthi1, Carolyn Shaw2 and Harvey Friedman*3 1 522F Johnson Pavilion, Infectious Disease Division, University of Pennsylvania, Philadelphia, PA 19104-6073, USA 2 506 Johnson Pavilion, Infectious Disease Division, University of Pennsylvania, Philadelphia, PA 19104-6073, USA 3 522E Johnson Pavilion, Infectious Disease Division, University of Pennsylvania, Philadelphia, PA 19104-6073, USA *Author for correspondence: Tel.: +1 215 573 8432 Fax: +1 215 349 5111 [email protected]
No vaccines are approved for prevention or treatment of genital herpes. The focus of genital herpes vaccine trials has been on prevention using herpes simplex virus type 2 (HSV-2) glycoprotein D (gD2) alone or combined with glycoprotein B. These prevention trials did not achieve their primary end points. However, subset analyses reported some positive outcomes in each study. The most recent trial was the Herpevac Trial for Women that used gD2 with monophosphoryl lipid A and alum as adjuvants in herpes simplex virus type 1 (HSV-1) and HSV-2 seronegative women. Unexpectedly, the vaccine prevented genital disease by HSV-1 but not HSV-2. Currently, HSV-1 causes more first episodes of genital herpes than HSV-2, highlighting the importance of protecting against HSV-1. The scientific community is conflicted between abandoning vaccine efforts that include gD2 and building upon the partial successes of previous trials. We favor building upon success and present approaches to improve outcomes of gD2-based subunit antigen vaccines. KEYWORDS: acyclovir • complement • genital herpes • glycoprotein C • glycoprotein D • glycoprotein E • HSV • IgG Fc receptor • immune evasion • vaccine
A half billion people worldwide are infected with herpes simplex virus type 2 (HSV-2) [1]. The infection cycle for genital HSV includes replication at the site of entry in the genital mucosa or skin, transport of viral components, including DNA to sensory neurons in the dorsal root ganglia (DRG), establishment of latent infection in the DRG with frequent reactivation that results in viral components traveling back down the nerve fibers to produce recurrent genital infections. Some individuals have multiple recurrences of genital lesions; however, the majority of subjects have subclinical (asymptomatic) infection that is manifested by frequent short episodes of genital shedding of HSV-2 that poses a risk for transmitting infection to their partners or to newborns delivered through an infected birth canal [2–4]. Genital herpes infection is emotionally distressing for infected subjects and their partners, is
informahealthcare.com
10.1586/14760584.2014.951336
potentially life threatening for exposed newborns, and importantly, increases the risk of acquiring and transmitting HIV infection by three- to fourfold [5,6]. No vaccine is currently available for prevention or treatment of genital herpes. Developing these vaccines is a major public health priority. HSV glycoprotein D (gD) was the first well-characterized HSV entry molecule, which explains why many vaccine trials performed to date have included gD2 subunit antigen. The goal is to induce neutralizing antibodies that block infection. As discussed below, the gD2 subunit antigen vaccine trials failed to meet their primary end points. This review focuses on approaches to improve gD2-containig subunit antigen vaccines. Other approaches that use attenuated live viruses, HSV DNA or viral vectors to express HSV antigens are not considered in this review.
2014 Informa UK Ltd
ISSN 1476-0584
1475
Review
Awasthi, Shaw & Friedman
Human trials involving HSV-2 gD2 subunit antigen
Expert Review of Vaccines Downloaded from informahealthcare.com by Nyu Medical Center on 07/22/15 For personal use only.
Prophylactic vaccine trials using gD2 antigen
One of the earliest double-blind, placebo-controlled trials of HSV vaccines for prevention of genital herpes used glycoproteins purified from HSV-2-infected chick embryo fibroblast cells [7]. Extremely low ELISA titers to HSV-2 gB2 and gD2 were obtained after three immunizations. The vaccine did not induce neutralizing antibodies or provide protection to high-risk subjects who had partners with genital herpes. The primary end point of preventing HSV infection was not achieved. Subsequently, large placebo-controlled, double-blinded HSV vaccine trials have either used HSV-2 gD2 as a monovalent subunit antigen vaccine (GlaxoSmilthKline [GSK] Biologicals, Rixensart, Belgium) or gD2 with gB2 as a bivalent vaccine (Chiron Corp. Emeryville, CA) [8–10]. Both gB2 and gD2 serve essential roles in virus entry and are expressed on the virus envelope, making them accessible to neutralizing antibodies [11]. The Chiron studies enrolled HSV-2 seronegative patients into multicentered trials that evaluated recombinant gB2 and gD2 antigen each used at 30 mg. The antigens were administered three-times over a 6-month period with MF59, an oil and water adjuvant comprising squalene, polysorbate 80 and sorbitan trioleate [10]. The results of two studies were reported in a single paper by the Chiron investigators [10]. One study enrolled HSV-2 seronegative partners of HSV-2-infected subjects, while the other enrolled HSV-2 seronegative persons attending sexually transmitted disease clinics. The primary end point of both studies was the time to acquisition of herpes infection, with or without genital lesions, as defined by seroconversion to herpes antigens not contained in the vaccine or by virus isolation in culture. An analysis of both studies together indicated that the vaccine group had 50% lower acquisition rates than placebo recipients over the first 5 months of the trial; however, over the entire course of the study, the vaccine was not efficacious despite inducing high titers of neutralizing antibodies. Multicentered, double-blind, randomized trials were performed by GSK and academic investigators in 2002 and 2012 using recombinant gD2 antigen formulated with AS04 adjuvant, which contains alum (aluminum hydroxide) and 3-Odeacylated-monophosphoryl lipid A (MPL) [8,9]. The 2002 publication reported two studies involving immunization of subjects whose regular sex partner had genital herpes [8]. In the first study, the subjects were seronegative for HSV-1 and HSV-2, while in the second study subjects were of any HSV serologic status. The primary end point was preventing genital herpes lesions in all subjects in the first study and in female subjects who were HSV-2 seronegative in the second study. A secondary end point in the second study was prevention of genital disease in female subjects who were seronegative at enrollment for HSV-1 and HSV-2. The gD2 vaccine was administered three-times over 6 months using 20 mg per dose. Overall, the vaccine did not provide significant protection against genital herpes lesions; however, the analysis showed approximately 74% efficacy in HSV-1 and 1476
HSV-2 doubly seronegative females, who met the criteria established as a secondary end point. The vaccine was not efficacious in seropositive women or in men, regardless of serologic status. These studies helped shape the experimental design of a subsequent vaccine trial reported in 2012, the Herpevac Trial for Women, and raised the possibility that protection may differ in males and females [9]. The Herpevac Trial subjects were HSV-1 and HSV-2 doubly seronegative women who were 18–30 years old. This population differs from the prior study that enrolled HSV-2 discordant couples, which makes direct comparisons between the two studies difficult. The recombinant gD2 antigen was administered at 20 mg with MPL/alum as adjuvants at times 0, 1 and 6 months. The primary end point was genital disease caused by HSV-1 or HSV-2 between months 2 and 20 after the first immunization. The overall vaccine efficacy was 20% and the primary end point was not achieved. The vaccine elicited ELISA and neutralizing antibody titers, although neutralizing titers were low and the median level declined to undetectable levels by month 16. No significant protection was observed against HSV-2; however, a subset analysis revealed 35% protection against HSV-1 infection (genital and/or oral) and 58% protection against HSV-1 genital disease compared with the control group (both statistically significant). This finding is important, since HSV-1 is now more common than HSV-2 as the cause of first-time genital herpes disease [12]. Using sera taken 1 month after the third immunization from subjects enrolled in the Herpevac Trial, neutralizing antibody titers were noted to be significantly higher (3.5-fold higher) to HSV-1 than HSV-2, which potentially explains the results of the Herpevac Trial [13]. Neighboring glycoproteins on HSV-2 appear to shield gD2 from neutralizing antibodies more effectively than neighboring glycoproteins on HSV-1 [13]. Based on low titers of neutralizing antibodies generated in subjects immunized in the Herpevac Trial, questions have emerged whether 20 mg of gD2 antigen was too little; however, in a separate study, no significant differences were detected in gD2 ELISA titers, neutralizing antibody titers and cellular immune responses comparing 20, 40 and 80 mg concentrations of gD2 [14]. Taken together, the results suggest that gD2 concentrations >80 mg may be required for protection against HSV-2, that gD2 amino acids 1–281 (of 393 amino acids) that were included in the vaccine formulation may be too limited in scope to be protective, that MPL/alum may not be an optimum adjuvant or that different antigens and adjuvants are needed to prevent HSV-2 genital infection. Therapeutic vaccine trials using recombinant gD2 antigen
The term therapeutic vaccine refers to immunization as a form of treatment to reduce the frequency and/or severity of recurrent genital herpes disease or recurrent genital shedding of HSV DNA. A double-blind, placebo-controlled trial of gD2 subunit antigen was performed to assess immunization as therapy for genital herpes in subjects reporting 4–14 genital herpes outbreaks annually [15]. Recombinant gD2 subunit Expert Rev. Vaccines 13(12), (2014)
Expert Review of Vaccines Downloaded from informahealthcare.com by Nyu Medical Center on 07/22/15 For personal use only.
Improving immunogenicity & efficacy of vaccines for genital herpes containing HSV gD
antigen at 100 mg per dose was administered with alum twice at 2 months interval or alum alone was used as a control. The gD2 vaccine recipients boosted their neutralizing antibody and gD2 ELISA titers, and had fewer virologically confirmed recurrences over a year of observation. These encouraging results were followed by a study using recombinant gD2 combined with recombinant gB2 subunit antigen as a bivalent vaccine each used at 10 mg with MF59 as adjuvant [16]. Subjects received four immunizations over 14 months with the bivalent vaccine or MF59 alone as a control. The bivalent vaccine induced neutralizing antibodies and decreased the duration and severity of the first outbreak; however, the overall recurrence rate was not significantly reduced. It is important to note that in one study gD2 was used at a concentration of 100 mg, while in the follow-up study gD2 and gB2 each were given at 10 mg raising the possibility that suboptimal concentrations of antigen were used in the second study. Immune correlates of protection against genital herpes
Immune correlates of protection refer to immune responses that statistically correlate with vaccine efficacy; however, correlation does not necessarily imply a mechanism [17]. For example, neutralizing antibody titers may be a mechanistic correlate of protection, while ELISA titers are likely non-mechanistic correlates that serve as surrogates for an antibody response that mediates protection, such as neutralizing antibody or antibody-dependent cellular cytotoxicity (ADCC). In general, the antigens selected for immunization will influence the correlates of protection that emerge. If a successful vaccine includes antigens involved in virus entry, then correlates of protection are likely to be neutralizing antibodies, while if the candidate vaccine contains immunogens aimed at inducing potent T-cell responses, then CD4+ and/or CD8+ T cells are likely to correlate with protection. Humoral immunity may be of greater significance for a prophylactic HSV-2 vaccine, while T-cell immunity may be essential for a successful therapeutic vaccine. For pathogens that infect at mucosal surfaces, correlates of protection may involve locally produced IgA antibodies or IgG that reaches the mucosal surface by transcytosis from serum, as demonstrated by protection against influenza virus [18]. For a complex pathogen, such as HSV-2, many aspects of immunity are likely to contribute to protection; therefore, more than one correlate may emerge. Varicella vaccine provides an example of a human herpes virus where correlates of protection have been assessed. Antibody responses and cell-mediated responses both correlate with protection, suggesting the importance of multiple components of immunity [19–23]. In support for a role for antibodies in protection against varicella and HSV, passive administration of varicella-zoster immune globulin provides protection against varicella infection and transplacental transfer of maternal antibodies protects infants exposed to HSV-2 at birth [24,25]. The Herpevac Trial for women had the unexpected result of protecting HSV-1 and HSV-2 doubly seronegative women against genital disease caused by HSV-1 but not HSV-2 [9]. informahealthcare.com
Review
ELISA antibody responses, rather than CD4+ or CD8+ T-cell responses, correlated with protection [9,26]. Among gD2 vaccine recipients, HSV-1 infection rates were 2.5% in subjects with low ELISA titers and
