Prospects for Control of Herpes Simplex Virus Disease through Immunization
- Lawrence R. Stanberry1,
- Anthony L. Cunningham4,
- Adrian Mindel5,
- Laurie L. Scott2,
- Spotswood L. Spruance3,
- Fred Y. Aoki6, and
- Charles J. Lacey7
- 1Division of Infectious Diseases, Children's Hospital Medical Center, and Department of Pediatrics, University of Cincinnati College of Medicine, Cincinnati, Ohio
- 2Fort Lauderdale Perinatal Associates, Plantation, Florida
- 3Department of Medicine, Health Sciences AIDS Center, School of Medicine, University of Utah, Salt Lake City
- 4Westmead Millennium Institute, Westmead Hospital, and University of Sydney, Westmead
- 5Academic Unit of Sexual Health Medicine, University of New South Wales, Sydney Hospital, New South Wales, Australia
- 6Department of Medical Microbiology, University of Manitoba, Winnipeg, Manitoba, Canada
- 7Department of Genito-Urinary Medicine, Imperial College School of Medicine, Jefferiss Research Trust Laboratories, London, United Kingdom
- Reprints or correspondence: Dr. L. R. Stanberry, Division of Infectious Diseases, Children's Hospital Medical Center, 3333 Burnet Ave., Cincinnati, OH 45229-3039 (l.stanberry{at}chmcc.org).
Abstract
Herpes simplex viruses (HSVs) can cause a variety of infections, including genital herpes. Despite effective antiviral therapy, HSV infections remain a significant worldwide public health problem. Vaccines offer the best hope for controlling spread and limiting HSV disease. This article discusses the pathogenesis and immunobiology of mucocutaneous HSV infections, summarizes the spectrum of diseases caused by HSV, and provides a review of the field of HSV vaccine research. This article also discusses what might be realistically expected of a vaccine intended for control of genital herpes and explores the question of whether a vaccine that is effective in controlling genital HSV disease might also be effective in controlling nongenital HSV disease. The efficacy of such vaccines for the full spectrum of HSV disease will eventually determine the timing and targeting of immunization, ranging from selective immunization in preadolescence to universal childhood immunization as part of the routine childhood regimen.
Herpes simplex virus types 1 (HSV-1) and 2 (HSV-2) cause a variety of illnesses, depending on the portal of entry, the immune state of the host, and whether the infection is initial or recurrent. Although HSV-1 and HSV-2 differ in their genetic and antigenic makeup, they are similar with regard to pathogenic properties and cause comparable illnesses. HSV commonly causes oral-facial infections (e.g., pharyngitis herpetica, herpes labialis, and herpetic gingivostomatitis), genital herpes, herpes keratitis, and cutaneous infections (e.g., herpetic whitlow and herpes gladiatorum). Less common but potentially deadly illnesses include herpes encephalitis, erythema multiforme, eczema herpeticum, neonatal herpes, and disseminated herpes. Inasmuch as the same etiologic agents cause these illnesses, the question arises whether a single vaccine could prove effective in preventing them. Three types of prophylactic vaccine are in clinical trials: adjuvant subunits, a replication-incompetent viral mutant, and a DNA vaccine. Other strategies, including genetically attenuated mutants and vectors, are in preclinical development. If a prophylactic vaccine is shown to be effective in controlling genital herpes, it will be important to consider vaccination for prevention of other HSV diseases. Although not discussed in this article, therapeutic vaccines may also be useful in the management of genital and nongenital HSV infections [1–3].
A key issue regarding HSV vaccines is whether they will prevent symptomatic disease but not asymptomatic infection. In the case of genital herpes, asymptomatic infection might permit latency to be established with the subsequent development of symptomatic or asymptomatic recurrences. Whether any vaccine can prevent asymptomatic infection (i.e., HSV replication at a mucosal portal of entry) is an important question that has obvious public health implications. In extending the use of a genital herpes vaccine to prevent other HSV diseases, it will be important to consider the impact of asymptomatic infection on the natural history of these illnesses.
This article reviews contemporary knowledge of HSV disease, epidemiology, pathogenesis, immunobiology, and management, and examines the possible benefit of a prophylactic genital herpes vaccine in controlling other HSV diseases. With regard to genital herpes, it explores the issue of prevention of infection versus prevention of disease, assesses the likely effect of immunization on the natural history of genital infection, and reviews efforts to develop HSV vaccines. The purpose of this article is to stimulate thought and discussion regarding the goals for prophylactic HSV vaccines and their potential use in controlling a variety of HSV diseases, including neonatal herpes, oral-facial herpes, and herpes keratitis.
Pathogenesis and Immunobiology
Pathogenesis of Oral-Facial and Genital HSV Infections
HSV-1 and HSV-2 both have a 152-kb double-stranded DNA genome encoding at least 80 gene products. These proteins are either structural (arrayed in the inner capsid enclosing the DNA core, in the surrounding tegument, or as glycoproteins in the envelope) or nonstructural (often enzymes present only in infected cells). Neutralizing antibodies are directed at the major envelope glycoproteins B, D, and H/L. T lymphocytes may be directed at any viral protein. The proteins and their antigenic epitopes are generally well conserved between HSV-1 and HSV-2. However, type-specific B cell epitopes have been defined on glycoproteins B, D, C, and, especially, gG, and type-specific CD4+ T cell epitopes have been defined on these glycoproteins, VP16, and others [4, 5].
The pathogenesis of HSV infection [6] has important implications in the development of realistic vaccine goals. The patho-genesis of mucocutaneous HSV infection is illustrated in figures 1 and 2. For oral-facial and genital herpes, infection typically begins when a mucosal epithelial cell in the oral cavity or genital tract of a susceptible person is exposed to virus present in the oral or genital lesions or secretions of an infected person. Virus can also infect apparently intact keratinized skin, although microabrasions are probably required for the virus to gain access to the epithelial cells under the keratinized layer. The virus attaches and binds to the epithelial cell, entering by fusing with the cell membrane. Once inside, replication results in the production of more virions that cause cell death. At the same time, virus enters sensory nerve endings that innervate the oral cavity or genital tract. The virions are transported to neuronal cell nuclei located in sensory ganglia: the trigeminal ganglia in oral-facial infections and the sacral dorsal root ganglia in genital herpes.
Figure 1
Pathogenesis of mucocutaneous herpes simplex virus infection: primary infection. 1, Transmitted virus infects an epithelial cell. 2, Virus enters sensory nerve endings. 3, Virus spreads by retrograde axoplasmic transport to the nucleus of neurons located in sensory ganglia. 4, Virus either (a) undergoes further replication in the neuron or (b) enters into a nonreplicating latent state. 5, Newly produced virions spread from the infected neurons by anterograde axoplasmic transport to mucocutaneous sites. 6, Virus is released from the nerve ending at the dermal-epidermal junction. 7, Released virus infects epithelial cells and replicates further. 8, Local viral replication and immunologic responses result in formation of a herpetic vesicle or lesion. 9, Virus is shed into the environment.
Figure 2
Pathogenesis of mucocutaneous herpes simplex virus infection: recurrent infection. 1, Reactivation of latent virus harbored in neurons in sensory ganglia. 2, Virus replicates to produce infectious virus. 3, Newly produced virions spread from the infected neurons by anterograde axoplasmic transport to mucocutaneous sites. 4, Virus is released from the nerve ending at the dermal-epidermal junction. 5, Released virus infects epithelial cells and replicates further. 6, Local viral replication and immunologic responses result in formation of a herpetic vesicle or lesion. 7, Virus is shed into the environment.
Virus in the infected neuron may replicate to produce progeny, or the virus may enter a nonreplicating latent state. Productively infected neurons send progeny virus back to cutaneous sites where they are released and infect epithelial cells adjacent to the nerve endings, thus further contributing to viral spread and injury. Animal studies suggest that the clinical disease resulting from genital infection is due largely to the virus that descends from the acutely infected neurons [7, 8]. Prevention or reduction of the acute ganglionic infection prevents the development of clinically apparent disease. Eventually, immune responses limit viral replication, allowing for recovery from the acute infection. However, these responses do not eliminate the latent infection, which persists in the ganglia for the life of the host. Latency, per se, causes no disease; however, latent infection can reactivate producing virions that, when released from nerve endings, can infect adjacent epithelial cells, resulting in recurrent skin lesions or asymptomatic shedding. The severity of disease resulting from recurrent infection is generally less than that resulting from initial infection, possibly due to a modifying influence of HSV-specific immunity (which limits viral replication and, hence, virus-induced injury).
An effective HSV vaccine might act by interfering with one or more steps in the pathogenesis of infection. Ideally, vaccine-induced immunity would block infection of epithelial and neuronal cells, thus preventing both disease and the establishment of the latent infection. If such complete protection is not possible, an alternative goal might be for vaccine-induced responses to reduce viral infection of epithelial and neuronal cells to a point where overt clinical disease does not occur. In this setting, asymptomatic infection and latency would occur but possibly with a diminished magnitude of the latent infection, thereby preventing or reducing subsequent recurrences.
Immunobiology of Mucocutaneous and Ocular HSV Infections
Host resistance to HSV infections includes nonspecific mechanisms such as IFNs, neutrophils, complement, macrophages, and natural killer cells, as well as specific mechanisms including humoral (antibody) immunity, T cell-mediated immunity (such as cytotoxic T cells [CTLs] and T helper activity), and cytokine release. The relative importance of these various mechanisms is different in initial or recurrent HSV disease. Animal studies suggest that activated macrophages, IFNs, and, to a lesser extent, natural killer cells are important in limiting initial HSV infection, whereas humoral immunity and cell-mediated immunity (CMI) are important in controlling both initial and recurrent infections. Adoptive transfer studies suggest that either virus-specific antibody or lymphocytes can protect animals against initial HSV infection, but several lines of evidence suggest that CMI responses play the central role in controlling recurrent HSV infections [9–11]. Mucocutaneous herpes is more severe in patients with impaired or defective CMI [9, 12, 13], but not in patients with agammaglobulinemia. Better understanding of the responses important in the control of recurrent herpes might help predict the types of responses that should be induced by an effective vaccine.
Studies comparing infected persons who do and do not have recurrences suggest that alterations in cytokine production may correlate with the development of recurrent infection [10, 14–18]. IFN-γ may be critical in resistance to HSV infection. HSV-specific IFN-γ; production by cultured peripheral blood mononuclear cells (PBMC) is lacking in some patients who have frequent episodes of herpes labialis [16–19]. For patients with recurrent infections who produced IFN-γ, there was a positive correlation between the concentration of IFN-γ in lesion vesicle fluid or in the culture supernatants of unstimulated or viral antigen-stimulated PBMC collected during a recurrence and the time to the next recurrence [10, 20]. A recent study [21] compared titers of antibody to HSV and cytokine production by cultured PBMC for seropositive patients with or without a history of herpes labialis. Those patients with a history of frequent recurrences were found to have higher median titers of serum antibody to HSV and trends for lower levels of HSV-specific IFN-γ and IL-2 production by PBMC. The findings suggest that a Th1-like cytokine response (IFN-γ and IL-2 production) may be associated with resistance to recurrences of herpes labialis. This conclusion is supported by the observation that vesicle fluid from recurrent herpetic lesions contains relatively high levels of IL-12 but a lack of IL-4 and by the demonstration that HSV infection of human keratinocytes stimulates IL-12 production 4- to 5-fold [22].
CD4+ lymphocytes that predominate in the early stages of recurrent herpes lesions [4, 10] are probably the main source of IFN-γ [23]. IFN-γ stimulates major histocompatibility complex (MHC) class II expression in keratinocytes, allowing for CD4+ lymphocyte cytotoxicity. It also partially reverses the downregulation of MHC class I expression induced by HSV infection of keratinocytes, allowing CD8+ CTLs to lyse infected cells [11, 24, 25]. Hence, CD4+ and CD8+ T cells may act sequentially in recurrent lesions before and after the induction and effects of IFN-γ [4, 24, 26, 27]. The major targets for restimulated (memory) blood CD4+ CTLs are the glycoproteins gD, gB, and gC [28]. CD4+ and CD8+ lymphocytes isolated from lesions also recognize other HSV targets [4, 29].
Experiments have examined whether it is possible to induce immunity that protects the genital tract from infection. Animals inoculated intravaginally with HSV develop evidence of genital herpes, and virus can be found in genital secretions and sensory ganglia. Those animals that survive the infection develop both local and systemic HSV-specific immune responses [30–34]. When the animals are reinoculated with a different HSV strain, immunity induced by the initial infection does not prevent reinfection of the vaginal mucosa, but does significantly reduce the magnitude and duration of viral replication occurring in the vagina after reinoculation, and protects the ganglia from reinfection [35, 36]. Mice depleted of CD4+ T cells after recovery from the initial infection have a pattern of viral replication in the vagina similar to that seen in immunologically naive mice. This effect could be partially duplicated by neutralization of IFN-γ, implying that the protective effect of the CD4+ cells is mediated in part by IFN-γ. These results suggest that HSV-specific CD4+ T cells play a more important role than CD8+ T cells or antibody in protecting the vaginal mucosa from HSV infection. It is interesting that depletion of either CD4+ or CD8+ lymphocytes from immune animals does not abolish the protection of the ganglia afforded by prior vaginal infection. Administration of HSV-specific antibody also protects the sacral ganglia from HSV infection resulting from vaginal challenge [37], suggesting a role for antibody in protecting the neuron early in primary HSV infection, possibly by blocking infection of the sensory nerve ending or alternatively by acting at the level of the ganglia.
Regarding the immunology of ocular herpes, reinfection with a second homologous strain (e.g., HSV-1a or HSV-1b) at the same location occurs rarely in humans and is difficult to accomplish experimentally. Prior HSV infection of rabbit cornea leads to milder infection when the same cornea is reinfected with the same virus strain [38, 39]. Prior infection at a nonocular site may [40] or may not protect against subsequent ocular HSV infection [38]. Animal studies suggest that the CMI responses correlate better with protection due to prior ocular infection than do antibody responses [41]. Experiments showed that distant and contralateral prior infection conferred substantial immunity but that a still greater protective effect resulted from prior ipsilateral corneal infection [42]. Prior oral-facial herpes may afford protection against ocular herpes, since infection of the eye resulting from exposure to the infectious secretions of another person probably takes place when the recipient is HSV-1-seronegative. Likewise, ocular autoinoculation probably occurs during primary herpetic gingivostomatitis, when the patient is effectively still seronegative.
In general, clinical observations and experimental studies suggest that if an HSV vaccine is to be effective it will need to induce both virus-specific CMI and humoral immunity. It is likely that local HSV-specific immunity will also be important for protection against genital herpes.
HSV Immunopathology
Host immune responses to HSV infections may themselves induce pathology [43]. HSV infections with clinically significant immunopathologic components include herpes retinitis, keratitis, and encephalitis. Erythema multiforme, an immunopathologic illness, can be triggered by mucocutaneous HSV infections. Studies with murine models of ocular infection have shown that immunopathologic reactions against HSV involve T lymphocytes [21, 43–46]. It is theoretically possible that administration of HSV vaccines to subjects previously infected with HSV could trigger immunopathologic reactions. Therefore, individuals with HSV retinitis or keratitis should not be vaccinated. It is possible, however, that if universal vaccination decreased HSV transmission, there would be a decrease in the number of individuals with immunopathologic reactions.
HSV Diseases
Genital Herpes
HSV infection of the genital tract may be either symptomatic or asymptomatic. Symptomatic infections can range from mild to severe. Mild infections may lack classical signs and symptoms and may not be recognized as genital herpes [47]. In addition to classical vesiculoulcerative skin lesions, patients with severe infections may have complications such as urinary retention and meningitis [48, 49]. Patients diagnosed with genital herpes may have substantial psychological morbidity, sometimes out of proportion to the severity of their infection [50]. Genital herpes is also a significant cofactor in HIV transmission and/or acquisition in both heterosexuals and homosexuals [51].
HSV-2 is usually sexually transmitted and is clinically manifested in the genital or adjacent area as primary or recurrent infection. HSV-1 is becoming an increasingly common cause of primary genital infection but is less commonly a cause of recurrent infection [52]. Of individuals with genital HSV infection, ∼20% have symptomatic disease that they recognize as genital herpes, another 60% have symptoms that they attribute to genital herpes only after being taught about the manifestations of the infection, and the remaining 20% lack any sign or symptom of genital herpes. In the absence of symptoms, HSV can be isolated from the genital area in men and women. This asymptomatic shedding may occur from the vulva, cervix, and urethral and perianal skin in women and from the urethral skin and perianal area in men [53–56]. Although the risk of transmission is greatest during initial symptomatic infection, most spread of HSV-2 occurs during recurrent infections as a consequence of asymptomatic shedding. The annual rate of transmission varies from 4% to 17%, depending on whether the at-risk subject is male or female and whether the subject has had a previous nongenital HSV infection [57].
The presence of HSV type-specific antibody indicates latent infection with the corresponding HSV type. Seroprevalence studies, therefore, indirectly reflect the burden of HSV infection in the population. HSV-2 seroprevalence varies considerably between different countries and subgroups. In the United States, population-based studies [58, 59] covering the periods of 1976–1980 and 1988–1994 found a 30% increase in HSV-2 seroprevalence from 16% to 22% among adolescents and adults. This increase was surprising in view of widespread antiviral (acyclovir) use and safe sex campaigns during an era of increasing HIV awareness. The factors that influence the acquisition of HSV-2 include age, ethnic origin, socioeconomic group or educational level, sex, geographic variables, HSV-1 sero-status, and sexual behaviors, including age at first incident of sexual intercourse, number of lifetime sexual partners, and, of course, selection of a partner with genital herpes [60–64].
Recent studies suggest that cross-reacting immunity resulting from an oral HSV-1 infection affords limited protection from genital HSV-2 infection [57–65]. People with preexisting immunity to HSV-1 appear to be at reduced risk of acquiring genital HSV-2 infection, and if they do become infected, the resulting disease is often milder than that in an HSV-1-seronegative individual.
Oral-Facial Herpes
First episodes of oral-facial HSV infection are nearly always caused by HSV-1 but also can be caused by HSV-2 [52, 66, 67]. The first episode usually occurs in the oral cavity as a primary infection and, when symptomatic, results in herpetic gingivo-stomatitis or pharyngitis. First episodes of nonprimary oral-facial HSV-1 infection also occur in patients with cross-reacting antibodies to HSV from prior genital HSV-2 herpes infection [68]. The severity of HSV-1 infection varies greatly; the existence of more or less virulent strains of HSV-1 has been proposed [69] but, as yet, remains unproved.
Seroprevalence studies indicate that the incidence of primary HSV-1 infection is greatest during childhood. Symptomatic primary oral HSV-1 infection has accounted for 10% of cases of sore throat in college students [70]. From 20% to 40% of the population have had episodes of herpes labialis [71–74]. The frequency of recurrent episodes is extremely variable, ranging from rare outbreaks every 5–10 years to >12 episodes yearly. Evidence from a 12-year longitudinal study suggests that the frequency and severity of recurrent HSV-1 disease decrease with time [75]. Virus transmission from a person with oral-facial HSV-1 infection to a susceptible individual can occur by direct contact with primary or recurrent oral-facial lesions or virus shed asymptomatically in saliva. There are few data on the frequency of oropharyngeal virus shedding among HSV-1– seropositive persons who do not have a history of herpes labialis, although it does occur [76, 77].
Primary HSV gingivostomatitis can be a painful, protracted illness that may require hospitalization. Recurrent herpes labialis is most often mild, although it can be uncomfortable, interfere with eating and sleeping, and adversely affect self-esteem. The cosmetic disfigurement associated with herpes labialis can be significant in a number of personal and professional situations. Severe herpes labialis can result from exposure to sunlight and occur with operations on the trigeminal ganglion and with abrasive, laser, and chemical facial cosmetic procedures. In immunosuppressed persons, recurrences are of longer duration and may lead to major morbidity, including esophagitis and pneumonia [78].
HSV gingivostomatitis may be treated with oral acyclovir. Severely ill individuals may need to be hospitalized to maintain hydration, and iv acyclovir may be appropriate treatment. Episodic therapy for herpes labialis may be given as topical penciclovir cream or oral acyclovir, valacyclovir, or famciclovir. Topical acyclovir cream is approved in many countries for treatment of herpes labialis, but there is sparse evidence for its efficacy. Individuals with very frequent episodes of HSV-associated erythema multiforme may be managed with suppressive antiviral therapy [79, 80].
As described in later sections, oral-facial HSV-1 infection may also cause or be the source of virus for other HSV infections or associated conditions, including eczema herpeticum, erythema multiforme, ocular herpes, herpes encephalitis, and primary genital HSV-1 infection [81–84].
Neonatal Herpes
Neonatal HSV infection occurs throughout the world. The highest rates of neonatal herpes are in the United States, with an incidence of 1 case per 2500–8800 deliveries [85, 86]. The incidence in other countries is lower: 1 case per ∼65,000 births in the United Kingdom, 1 case per 15,000 deliveries in Sweden, 1 case per 35,000 births in the Netherlands, and 1 case per ∼17,000 deliveries in Japan [87]. In the United States, ∼75% of cases are due to HSV-2, and 25% are caused by HSV-1, whereas HSV-1 causes about 50% of cases in the United Kingdom [88] and >60% of cases in Japan [87]. Most cases result from intrapartum exposure to the mother's HSV-infected secretions or genital lesions. Intrauterine infection can also occur if virus present in the mother's lower genital tract ascends to the uterus or if maternal viremia develops with subsequent transplacental spread [89]. Intrauterine infection may account for 6%–14% of the cases [90–92]. Postpartum acquisition of the virus, usually resulting from exposure to nongenital herpetic lesions [85, 91, 93–95], may account for up to 12% of cases [96].
In the United States, 5% of reproductive age women have a history of genital herpes [97], but it is estimated that up to 1 million pregnancies occur in HSV-2-seropositive American women annually [59, 98]. About one-half of all the cases of neonatal herpes occur in women with first episodes of genital infection. Two-thirds of these infections are asymptomatic, occurring around the time of delivery. An additional 30% of cases of neonatal herpes result from recurrent maternal genital herpes, and two-thirds of these cases result from asymptomatic genital HSV shedding [99]. The highest risk of transmission (up to 50%) occurs when a woman with symptomatic primary genital infection gives birth vaginally [100]. The risk is lowest (<4%) for the infant exposed to HSV shed asymptomatically during a recurrent infection [101]. It is believed that exposure to higher viral loads increases the risk of transmission [6], whereas the presence of passively acquired maternal-derived antibody decreases the risk [7, 91, 102].
Three categories of neonatal HSV disease have been identified: disseminated, CNS, and localized [94]. Disseminated and CNS infections frequently are fatal, and survivors may have significant neurological sequelae [103–107]. HSV infection localized to the skin, eyes, and mucous membranes is rarely fatal [103, 104]; however, morbidity occurs in 5%–10% of these infants, typically in those who have multiple cutaneous recurrences in the first 6 months of life [107]. Prompt treatment of infected infants with iv acyclovir or vidarabine can improve outcome, although a high proportion of infants with CNS or disseminated infection still die or are left with neurological sequelae [103, 104]. Without appropriate antiviral treatment, localized infection will generally progress to disseminated in-fection and/or meningoencephalitis and, ultimately, neonatal death [104, 106, 107].
At this time, the major strategy for preventing neonatal herpes focuses on pregnancies in women known to have genital herpes. If, at parturition, an identifiable lesion is present, a cesarean section is undertaken to avoid intrapartum exposure [108]. This approach has many drawbacks. It does not prevent the cases of neonatal herpes that result from asymptomatic or unrecognized herpetic infections at delivery [88, 109, 110]. Some women with identifiable lesions inadvertently deliver vaginally, and their infants miss the limited protection afforded by cesarean delivery. Finally, some neonates acquire HSV infection after birth, and for these infants, mode of delivery has no bearing on their risk of becoming infected. Cesarean delivery is not completely effective, since 20%–30% of infants who develop neonatal herpes have been delivered by cesarean section [88, 91]. Furthermore, delivery by cesarean section is associated with substantial risk of maternal morbidity and mortality. For every 7 herpes-related neonatal deaths or severe disabilities averted, a policy of cesarean section for mothers with recurrent genital herpes results in 4 maternal deaths and has a price tag of $2.5 million to prevent each neonatal death due to recurrent HSV infection [111].
Another approach is to use acyclovir for virus suppression in mothers with genital herpes, to prevent recurrent genital HSV infection at delivery. Although this approach has been demonstrated to be effective in reducing recurrent infections near the time of delivery [112, 113], it has not been proven to reduce risk of transmission to the offspring. Furthermore, it is also limited to those women who are known to have genital herpes. Finally, questions remain as to the safety of this therapy for the fetus or neonate.
Herpetic Keratitis
Either virus type may cause keratitis, although HSV-1 is the cause of 78%–98% of cases [114, 115]. Ocular HSV-2 infection can occur from autoinoculation from a genital lesion [112] or in cases of neonatal herpes. The frequency of ocular HSV reactivation appears to be type-specific, with frequent HSV-1 reactivations and infrequent HSV-2 reactivations [116–118]. Conflicting data from animal studies exist with regard to the propensity of HSV types and strains to infect the eye and cause injury [119–121]. For humans, there are little data to support the concept of oculotropic HSV strains. There is no evidence that cases of keratitis cluster and the testing of HSV isolates from patients in an animal model found no correlation between disease pattern in the animals and that in the patients, suggesting that host not viral factors are important in determining the severity of ocular HSV disease [122].
The exact prevalence of ocular HSV disease is unknown [123]. Extrapolation of results of a Mayo Clinic study performed in the 1980s on the US population yields an incidence of 22,000 new cases of ocular HSV disease, a total of 60,000 episodes of ocular herpes per year, and 400,000 persons with a history of ocular herpes [124]. The incidence of initial ocular infection appears to be highest in the first decade of life, although 8%–27% of new cases occur in individuals aged ≥55 years [125–127]. In 1 study [128], patients often had recurrences: 53% had ≥1 recurrence, and 40% had >1 recurrence. Factors associated with risk of recurrence include male sex, history of an ulcer, and history of recurrences close together [128, 129].
Keratitis is probably contracted by exposure to persons with oral-facial herpes. Contributing to transmission are the high prevalence of herpes labialis, the high titer of virus in lesions, asymptomatic shedding in the oral cavity between lesion episodes [130–132], and the frequent presentation of the lips, oral-facial lesions, and saliva of 1 person to the lips and eyes of another, either directly or indirectly by contamination of the hands. There is a strong association between oral disease and ocular disease: 56%–68% of patients with ocular herpes have a history of oral herpes, 1% of patients with herpes labialis have a history of ocular herpes, 22% of patients with primary ocular herpes have concurrent herpetic gingivostomatitis, and 12%–37% of patients with recurrent herpetic keratitis have a concurrent attack of herpes labialis (S. L. Spruance et al. [unpublished data], [125, 128, 133–135]). Besides spread via contaminated secretions, HSV-1 may also reach the eye via intraneuronal virus spread from the trigeminal nerve during a primary oral-facial infection [136]. Most persons with first episodes of ocular disease are HSV-seropositive at their first presentation and do not have an increasing titer [126], indicating that the primary infection took place earlier. The clinical consequences of ocular HSV infection range from first episodes of keratoconjunctivitis to recurrent bouts of lytic corneal infection that cause dendritic ulcers and recurrent disciform keratitis and/or stromal disease that may lead to blindness.
Herpetic Encephalitis
There are only limited data on the incidence of HSV encephalitis. In the United States, the estimated incidence is 1 case per 200,000 population or ∼1250 cases per year [137]. The incidence in northern Europe appears to be slightly lower, with an estimated yearly incidence of 1 case per 400,000 population in Sweden and the United Kingdom. Data on incidence in other countries appear to be equally low [138].
Of cases of HSV encephalitis, ∼30% occur in patients aged <20 years, 20% in those aged 20–50 years, and 50% in those aged >50 years. Most cases are probably due to reactivation of latent HSV-1, since ∼70% of subjects with HSV encephalitis are seropositive at presentation [139, 140]. Encephalitis due to HSV-2 is very rare and essentially only seen in neonates and HIV-infected individuals. In adults, HSV localizes to the temporal lobe and limbic system and presents as focal encephalitis [141, 142]. However, only 50% of cases of focal encephalitis are due to HSV. The diagnosis of HSV encephalitis has been markedly improved by the use of PCR-based assays to detect HSV DNA in CSF. This test is highly specific and very sensitive within the first 7–10 days after onset of symptoms [143].
The mortality rate among untreated individuals is >70%, and only 2.5% of survivors will be neurologically normal. Mortality rates are considerably reduced with acyclovir treatment (19%) or vidarabine therapy (44%–54%) [137–139]. In addition, the proportion of individuals who return to normal function is greater with acyclovir treatment (38%) than with vidarabine treatment (14%).
Neurological Complications
Because HSVs are neurotropic, HSV infections can cause neurological complications. Recognized complications include aseptic meningitis, autonomic nervous system dysfunction, transverse myelitis, sacral radiculopathy, and Bell's palsy [68, 144, 145]. HSV vaccines protect against neurological complications associated with genital infection in animal studies [146] and might be expected to also prevent neurological complications in humans.
Dermatological Complications
Erythema multiforme is an acute inflammatory disease of the skin and mucous membranes [147]. Two US population-based studies estimated that the annual incidence of erythema multi-forme that results in hospitalization ranged from 7.4 to 46.5 cases per 1 million population [148, 149]. HSV is one of several known precipitants of erythema multiforme [150]. “Typical” erythema multiforme, where <10% of the total body surface area is involved [147], is frequently associated with a history of recent HSV infection. The etiological connection with HSV infection has been strengthened by the demonstration of HSV within such lesions by culture, immunofluorescence, and, more recently, PCR-based assays; HSV DNA was detected in ∼70% of samples by PCR-based assays in 4 studies [151–154].
The pathogenesis of erythema multiforme and its relationship to HSV remain incompletely defined. An association with DQB1*0301 has been described in some series [155] but not in others. The concept of HSV being involved in the etiology of most cases of erythema multiforme is strengthened by studies of response to treatment. In an observational study of continuous oral acyclovir twice daily for 6 months [153], this treatment completely suppressed recurrent erythema multiforme in 15 of 37 patients, whereas partial suppression was seen in 8 of 37 patients. A randomized, double-blind study over 6 months, in which the same regimen was used, showed a reduction in the number of episodes of erythema multiforme (0.5 episode in the acyclovir group vs. 3.7 episodes in the placebo group) and of HSV infection (0.6 episode in the acyclovir group vs. 2.1 episodes in the placebo group) [80]. However, a few resistant cases only respond to immunomodulatory drugs such as dapsone or azathioprine [156].
Eczema herpeticum, also referred to as Kaposi's varicelliform eruption, is due to HSV infection in persons with atopic eczema or other preexisting skin disease [150]. Most cases represent primary HSV infections, although recurrent cases of eczema herpeticum have been described [150, 157]. If the disease remains undiagnosed or not treated with systemic acyclovir, it can be fatal. More aggressive forms of HSV infection, involving multiple cutaneous and mucosal sites but not >20% of the total body surface area, are also more commonly seen in atopic eczema. Atopic eczema appears to be characterized by exaggerated Th2 responses to common allergens, and such responses may be involved in the pathogenesis of eczema herpeticum. This is a rare condition (annual incidence, ∼1.5 cases per 1 million population) [157].
Economic Burden of HSV Infections
Despite the prevalence of HSV infections, there are very limited data regarding their economic impact. A recent study estimated that the total annual direct medical care costs for genital herpes in the United States was $980 million, with indirect costs of $210 million [158]. A recent cost-benefit analysis calculated that neonatal herpes added $255,000 lifetime costs per affected infant, resulting in $1.68 million in lifetime expenses incurred each year [159]. The interventions to prevent neonatal herpes are estimated to cost $80 million each year. Together, the estimated total cost of genital herpes and its sequelae is >$1 billion annually in the United States alone.
HSV Vaccines: What Might They Do?
HSV Vaccines
HSV vaccines may be divided broadly into 2 groups: live or inactive. Live vaccines contain organisms capable of at least limited replication in vivo, whereas inactive vaccines are incapable of replicating (table 1). The vaccines may be further subdivided into 6 categories: attenuated HSV vaccines, replication-limited (or replication-incompetent) HSV vaccines, vaccines consisting of live nonpathogenic replicating vectors engineered to express HSV gene products, inactivated HSV vaccines, HSV component or subunit vaccines, and nucleic acid (plasmid) vaccines [160]. The advantages and disadvantages of each vaccine type are briefly discussed below, and those vaccines in commercial development are presented in table 2. For more detail, the reader is referred to several recent reviews on HSV vaccines [3, 5, 170–172].
Table 1
Categories of herpes simplex virus (HSV) vaccines.
Table 2
Status of genital herpes simplex virus (HSV) vaccines in commercial development.
Live virus vaccines generally induce broader and more durable immune responses than do inactivated virus vaccines. Traditionally, live vaccines are produced by serial passage of a virulent virus in cell culture that results in an attenuated virus that is immunogenic but not pathogenic. This strategy has not worked for HSV, because the attenuation does not result in a genetically stable virus. There are also concerns regarding the ability of a live virus to establish latency, to reactivate, and to recombine with virulent wild type virus, as well as theoretical concerns regarding the oncogenic potential of some HSV genes. An alternative strategy that addresses the stability issue is to use molecular genetic methods to engineer stable attenuated viruses [173]. Initial testing of a genetically attenuated HSV vaccine by Pasteur Merieux Connaught (Lyon, France) showed that the approach was feasible but that the resulting vaccine was overly attenuated and minimally immunogenic [5]. Aviron (Mountain View, CA) is further developing this approach [161].
A novel variation on genetically attenuated viruses is the development of replication-incompetent or replication-limited mutants [174, 175]. The approach involves deletion of a gene(s) that is essential for viral replication. The defective virus is then cultivated on a genetically engineered cell line that expresses the missing viral gene product(s). The resulting virus is capable of infecting normal cells, but it cannot make the missing gene product(s) required to produce new infectious virions and, hence, is limited to a single infectious cycle without spread of infection to other cells. These replication-incompetent mutants do not cause disease in animals but engender HSV-specific humoral immunity and CMI responses that protect the animals from experimental infection [176, 177]. AVANT Immunotherapy (Needham, MA) has a candidate replication-incompetent HSV vaccine in preclinical development. Cantab Pharmaceuticals (Cambridge, UK) has developed an HSV-2 mutant lacking the gene encoding the essential glycoprotein gH. This vaccine has been shown to be immunogenic and well tolerated in phase 1 trials in the United Kingdom and the United States [162, 163]. This vaccine has been recently licensed to Glaxo-Wellcome (Research Triangle Park, NC), and further clinical development is planned.
Although vaccines containing genetically attenuated HSV mutants or replication-incompetent mutants probably induce broad and durable immunity, questions remain regarding their safety, especially concerning their potential to recombine with wild type virus, to establish latency, and to reactivate, as well as their potential oncogenicity. Another strategy that retains the advantages of a live virus vaccine while avoiding safety concerns is the use of replicating vectors. In this approach, HSV gene(s) encoding immunogenic protein(s) is inserted into a replication-competent viral or bacterial vector. When immunized with the vector, the host has humoral immunity and CMI responses to the proteins encoded by the vector, including the HSV protein(s). A number of vectors have been proposed, including vaccinia virus, adenovirus, poliovirus, rhinoviruses, canarypox virus, and Salmonella. Studies have shown that live vectors encoding HSV genes can induce HSV-specific immunity and protect animals against disease [178–180]; however, these vaccines have not yet entered clinical trials.
Vaccines derived from killed-virus preparations offer a potential safety advantage over live virus vaccines since the killed virus cannot replicate. However, the theoretical concern regarding oncogenicity remains, unless HSV DNA is completely removed from the preparation. Killed-virus vaccines have the disadvantage that they may induce less broad and less durable immune responses than do live-virus vaccines. This disadvantage may be overcome by including adjuvants in the formulation. Killed-HSV vaccines have a long and unsuccessful history [160, 181]. Vaccines made in the 1930s consisted of formalin-treated HSV-infected guinea pig footpad or rabbit brain emulsion. In the 1950s, vaccines were prepared by use of ultraviolet radiation to inactivate virus grown in embryonated eggs. By the 1960s, virus was being grown in cultured cells and was inactivated by a variety of means, including ultraviolet irradiation, formalin treatment, or heat. These early vaccines were tested largely in open clinical trials; hence, their efficacy was never satisfactorily proven [181].
Since the 1970s, several killed-virus vaccines have been developed for therapeutic and/or prophylactic use [160, 181]. These vaccines include Lupidon H and Lupidon G (Hermal-Chemie, Germany), Dundarov vaccine (Bulgaria), Skinner vaccine (Porton International, UK), the Kutinova vaccine (Czechoslovakia), Cappel vaccine (Belgium), and, more recently, a glycoprotein vaccine made from infected cells (Merck, West Point, PA). None of these vaccines have been proven to be efficacious in well-designed carefully controlled clinical trials.
Subunit vaccines became possible because of developments in recombinant DNA technology that permitted production of large quantities of immunogenic proteins without the need to work with virus-infected cells. This approach has the safety advantage of ensuring that the preparation is free of infectious virus or viral DNA. However, subunit vaccines have some shortcomings. Because they contain only a fraction of the antigens present in a whole-virus preparation, the subunit vaccine can induce only immune responses to a limited number of epitopes. In addition, like killed -virus vaccines, subunit vaccines probably induce a less durable and less broad immune response. Finally, purified proteins alone tend not to induce a full array of CMI responses, especially MHC class I-restricted CTL responses. To address this problem, purified protein(s) can be formulated with an adjuvant. This strategy, while enhancing vaccine immunogenicity, also increases its reactogenicity [170].
The development of subunit HSV vaccines has largely focused on 2 envelope glycoproteins, gB and gD. These are major targets for neutralizing antibody, and although it is uncertain whether they are the predominant viral targets for cell-mediated responses [170], they have been shown to be major targets for restimulated CD4+ CTLs in human keratinocytes [28]. The recombinant glycoproteins gB and gD are immunogenic and protective in animal challenge studies [182]. Four recombinant subunit vaccine preparations have been evaluated in clinical studies. A vaccine developed by Chiron (Emeryville, CA) that contains recombinant truncated HSV-2 glycoprotein gD (gD2) adsorbed to alum was tested as a therapeutic vaccine for the control of frequent recurrent genital herpes. The vaccine was found to be both immunogenic and modestly effective [183]. A second Chiron vaccine containing gD2 with a muramyl tripeptide adjuvant was shown to be immunogenic in humans but had an unacceptable reactogenicity profile, especially in HSV-seropositive subjects.
This problem led to the development of a third vaccine containing recombinant gD2 and gB2 with MF59, an adjuvant consisting of squalene, polysorbate 80, and sorbitan trioleate. This formulation was immunogenic and only mildly reactogenic in humans [184] but was ineffective in the treatment of recurrent genital herpes [185]; the results of 2 large prophylactic studies reportedly showed that the vaccine did not protect recipients from acquiring genital HSV-2 infection [164]. The 2 studies, which included 2393 volunteers, found a lower acquisition rate during the first 5 months of the trial, but the effect was lost by the end of the 1-year follow-up, with an overall efficacy rate of only 9%. It is interesting that the efficacy was affected by sex, with an efficacy rate of −4% for males but an efficacy rate of >26% for females. For these studies, the primary outcome measure was prevention of infection as assessed by seroconversion to HSV-2 antigens not contained in the vaccine. Animal studies suggested that the subunit vaccine could not prevent infection but might be capable of preventing or ameliorating symptomatic genital herpes [182]. In these studies, immunization with the Chiron vaccine did not appear to reduce the likelihood that the infection would be asymptomatic, nor did it reduce the severity of the symptomatic disease.
SmithKline Beecham Biologicals (Rixensart, Belgium) has also tested a subunit vaccine in humans. It contains recombinant truncated gD2 and alum combined with the potent adjuvant 3-de-0-acylated monophosphoryl lipid A. The vaccine appears to be well tolerated and induces humoral and cellular immune responses superior to those produced by the glycoprotein gD and alum alone [165]. Depending on the success of these trials, future development of subunit vaccines may need to focus on identifying new viral antigens that are important targets of CMI, as well to expand the range of potent immunopotentiating agents that can be used in the vaccine formulations.
The development of nucleic acid-based vaccines has provided a new strategy for controlling HSV infections. Vaccines consisting of plasmid expression vectors containing the genes encoding HSV-1 glycoprotein gB or gD or gD2 have been shown to induce humoral immunity and CMI responses and to protect mice and guinea pigs against HSV challenge [166, 186–189]. This is a burgeoning field with both large and small companies exploring its potential. In the United States, Vical (San Diego, CA) and Merck are involved in the preclinical development of DNA-based vaccines for the prevention of HSV infections. Apollon (Malvern, PA) has initiated a phase 1 trial of their gD2 DNA vaccine.
Objective of Vaccines Against Genital Herpes: Prevention of Infection vs. Prevention of Disease
What might be expected of a genital herpes vaccine? Ideally, it would induce a “sterilizing immunity” that would prevent replication of virus introduced onto the genital mucosa; thus, the immunized host would neither develop symptomatic genital herpes nor become latently infected. From the perspective of clinical trial design, vaccine failure could be assessed by determining whether the subject had an antibody response to HSV antigens not present in the vaccine (i.e., nonvaccinal antigens). Because it would be expected that the virus would be prevented from replicating, the immunized host would not be expected to mount immune responses to nonvaccinal antigens. Although it is possible that vaccine-induced immune responses would increase the amount of virus required to cause genital infection (i.e., the threshold of infection), our current understanding of the immunobiology of HSV and of vaccinology in general suggests that induction of a broadly effective sterilizing immunity is probably not achievable. As discussed earlier, animal studies and clinical reports indicate that even the responses induced by initial genital HSV infection do not protect the host against genital reinfection, and it is uncertain that a vaccine will be able to engender immune responses that are more protective than those induced by “natural” infection. One caveat regarding the potential effectiveness of vaccines vis-à-vis immunity induced by natural infection is that HSV encodes viral gene products known to facilitate immune evasion. Hence, it might be possible to engineer live virus vaccines devoid of immune evasion genes that would be capable of inducing protection superior to that resulting from natural infection.
If induction of a sterilizing immunity does not appear achievable, a second more realistic goal might be the development of a vaccine that, while failing to prevent infection of the genital tract, does prevent symptomatic genital herpes and protects the sensory ganglia from latent infection. This situation might be somewhat analogous to the inactivated poliovirus vaccine that prevents poliovirus infection in the nervous system but does not block viral replication in epithelial cells of the gastrointestinal tract. Animal studies have shown that although immune responses resulting from an initial genital HSV infection do not protect the genital tract from reinfection, they nevertheless protect the sensory ganglia from reinfection [36, 37]. Theoretically, it should be possible for a vaccine to provide the same protection to the ganglia that natural infection does. A subclinical primary infection of the genital tract would be of little clinical or public health significance if a vaccine prevented both symptomatic genital herpes and the establishment of latent infection in sensory ganglia.
A third possibility would be the development of a vaccine that provides partial protection against symptomatic genital herpes (i.e., a reduction in the severity of the disease caused by the initial infection). It is probable that the amount or inoculum of virus required to cause mucosal and/or ganglionic infection would be greater in immunized individuals than in nonimmunized individuals. In this setting, immunization would be expected to completely prevent symptomatic genital herpes in some individuals and reduce the severity of the primary infection in others. This situation might be analogous to the milder course of influenza seen in some people who receive the influenza virus vaccine but still become symptomatically infected [190]. On the basis of animal studies [36, 37], vaccines would also partially protect the ganglia, resulting in a reduction in the magnitude of the latent infection and a significant decrease in the frequency and severity of recurrent infections. Vaccine recipients who become subclinically infected would be intermittently at some risk for transmitting the infection to susceptible partners. However, it is likely that subclinically infected vaccine recipients may shed smaller amounts of virus and shed less frequently than do nonimmunized infected persons, thus making them less contagious than those who become asymptomatically infected without the benefit of the vaccine.
Although subclinically infected immunized people might inadvertently spread infection to others with resultant disease, this problem might be effectively managed by universal immunization, so that immunized persons exposed to someone who is shedding virus in the absence of symptoms might become infected but would not have clinical disease. If the vaccine altered the natural history of genital herpes so that those persons who became infected were less contagious and also made those who were uninfected less susceptible, the net effect would be to reduce the spread of genital HSV infection. When anticipating the effect of a vaccine on the spread of genital herpes, it must be considered whether a partially effective vaccine might facilitate the spread of infection by encouraging some people to engage in riskier sexual behaviors. Therefore, the issue of whether a partially protective vaccine would have a positive or negative impact on the epidemic of genital HSV infection is an important public health question that demands thorough examination. A partially protective vaccine, however, would be expected to reduce the likelihood of developing symptomatic disease, thereby preventing the pain and discomfort associated with genital herpes. Two other important issues regarding the use of a vaccine that prevents genital HSV disease, but not infection, are whether the vaccine will impact the spread of virus from infected mother to fetus or infant and whether subclinical genital infection in immunized subjects places them at any increased risk of acquiring HIV infection through sexual exposure.
Vaccines and Control of Nongenital HSV Infections
The HSV vaccines currently in development are intended for the prevention of genital herpes. The choice of genital herpes rather than another HSV disease to assess vaccine efficacy and safety is based on public health need, perceived market size, and the feasibility of conducting vaccine trials. If a “genital herpes” vaccine can be successfully developed, it will be important to consider its use for controlling other HSV diseases. On the basis of our understanding of the pathogenesis and immunobiology of the different HSV diseases, it is likely that a vaccine that does not prevent infection will be useful in controlling some illnesses but not others. The possible effects of a genital herpes vaccine on nongenital HSV infections are summarized in table 3. Although it may be possible to directly assess vaccine efficacy for relatively common illnesses such as herpetic gingivostomatitis, for uncommon conditions direct assessment will be problematic. The potential for success and the special issues related to vaccine evaluation in nongenital herpetic illnesses are discussed next.
Table 3
Possible effects of prophylactic immunization on herpes simplex virus (HSV) diseases.
Neonatal herpes. There is an obvious need for more effective strategies for protecting the newborn from HSV infection. Because of the natural history of neonatal herpes, immunization of the newborn with a safe vaccine would, at best, have limited effectiveness. An alternative strategy would be to immunize women before or possibly during pregnancy. If the vaccine prevented maternal genital infection caused by HSV-1 and HSV-2, it would be expected to reduce the incidence of neonatal herpes by ∼80%–90%. Epidemiological data suggest that immunization of women might afford some protection to the fetus or neonate, even if the vaccine does not completely prevent maternal genital HSV infection. This protection would result from protective, maternally derived, transplacentally passaged antibody and/or a reduction in the amount of virus present in the genital tract around the time of delivery [89, 110]. Depending on the effectiveness of the passively acquired antibody, maternal immunization might also afford the infant some protection against infection acquired in the postpartum period.
Because of the low incidence of neonatal herpes, it would be impractical to conduct clinical trials specifically designed to assess the effectiveness of maternal immunization in preventing neonatal infection. Effectiveness might be assessed indirectly by population-based surveillance programs designed to determine the impact of immunization on the incidence of neonatal herpes.
Oral-facial herpes. An effective vaccine for genital herpes could have a significant effect on oral-facial herpes and its complications. This effect might come about in 3 ways: (1) reduction in the severity of primary and recurrent oral-facial diseases; (2) reduction of virus excretion associated with primary and recurrent oral-facial diseases, with reduced transmission of the virus; and (3) increased resistance to acquisition of the infection. Initial assessment of the efficacy of a vaccine for oral-facial infections might be conducted in seronegative college students. The impact of vaccination on controlling oral-facial infection would depend on the immunization plan (universal immunization in childhood vs. immunization of targeted groups) and the nature of the response to the vaccine. A program of universal childhood immunization might over time gradually lower the prevalence of HSV-1 and HSV-2 infections with a concomitant reduction in associated morbidity. Given the widespread nature of oral-facial HSV infections, mass childhood immunization would appear to be the preferred strategy.
Herpetic keratitis. An HSV vaccine could impact the frequency of herpetic keratitis directly or indirectly. Seronegative vaccine recipients may have increased resistance to primary infection and/or a reduced severity of infection, such that extension of the disease from the oral cavity to the eye, or from neurons in the trigeminal ganglion, is less likely to occur. A vaccine may indirectly reduce the risk of ocular disease by lowering the frequency and severity of oral-facial disease and the likelihood of virus transmission.
Stromal keratitis may be an immunopathologic disease determined by the immune characteristics of the host. Accordingly, alteration of the immune status of someone with ocular herpes with a vaccine should be approached with caution, and persons with preexisting ocular disease probably should be excluded from an immunization program. However, preliminary data indicate that an HSV glycoprotein vaccine injected subcutaneously did not exacerbate preexisting or postimmunization experimental ocular HSV infection in mice [198]. Limited epidemiological data suggest that the incidence of ocular herpes is too low to permit direct evaluation of the efficacy of an HSV vaccine in controlling ocular disease. Effectiveness might be assessed indirectly by developing population-based surveillance programs to determine the impact of immunization on preventing ocular herpes after vaccine licensure.
Herpetic encephalitis. The effects of a genital herpes vaccine on HSV encephalitis after the neonatal period are difficult to estimate, unless the vaccine is completely protective against HSV infection. The effect of enhancing immunity to HSV without protection against infection is difficult to predict since HSV encephalitis is not increased in incidence or severity in immunocompromised patients (e.g., those with AIDS). However, HSV encephalitis associated with primary HSV infection may be prevented or reduced by immunization. Clinical trials of HSV vaccines to determine their effects on HSV encephalitis would be impractical because of the large number of subjects required for appropriate statistical power. The effect of a vaccine on herpetic encephalitis could be assessed only by population-based surveillance studies.
Dermatological complications. A prophylactic vaccine effective against genital herpes caused by HSV-1 or HSV-2 would probably prevent most cases of eczema herpeticum, if given universally at a population level in early childhood. Because of our incomplete understanding of the pathogenesis of erythema multiforme, the effect of an HSV vaccine is difficult to predict. It will not be possible to directly assess the effect of HSV vaccines on dermatological complications; rather, the impact will be determined after universal childhood or adolescent prophylactic vaccination through postmarketing surveillance programs.
Conclusions
The need for a vaccine to control genital herpes has been acknowledged for decades. Recent recognition of the extent of asymptomatic and underrecognized genital herpes and the frequency of HSV shedding in the genital tract emphasize that a vaccine is the only practical measure to control disease and the spread of infection. From earlier discussions, it is clear that the morbidity, mortality, and likely economic impact of nongenital HSV infections are also significant. Therefore, broadening the target population for an HSV vaccine beyond those persons at risk for genital herpes should be considered.
It will also be important to reach a consensus regarding the goals for an HSV vaccine. Should we insist that a vaccine provide complete protection against infection or is this an unrealistic goal? The recent failure of a recombinant subunit vaccine to prevent infection suggests that it may be so. Prevention or amelioration of disease with or without partial protection against infection may be achievable, but it is difficult to know whether a partially protective vaccine will favorably impact the ongoing epidemic of genital herpes. In considering the potential benefit of an HSV vaccine that prevents disease but not infection, it is worth noting that other vaccines including pertussis vaccine [191–193] and influenza virus vaccine [194–197] also prevent disease without preventing infection.
As discussed earlier, there are some theoretical benefits and potential problems that could be associated with use of an HSV vaccine that prevents disease but not infection. The extent of the benefits and the problems will be influenced by how widely and at what age the vaccine is used (i.e., universal administration vs. immunization of targeted groups and immunization of infants vs. immunization of adolescents or adults). Immunization of targeted groups may benefit the vaccine recipient but may have little impact on public health. Alternatively, universal immunization in infancy with administration of a booster dose in adolescence might protect the vaccine recipient against symptomatic oral-facial, ocular, and genital herpes and also reduce the risk of spread to people who are not immunized. Although the latter strategy is the most expensive, it could be incorporated into existing pediatric vaccine regimens, and prevention of all HSV diseases would have significant medical and economic benefits. With promising new HSV vaccines in clinical trials, now is the time for public discussion regarding how these vaccines should be used and their potential cost-effectiveness, benefits, and limitations.
Footnotes
-
Financial support: This work was supported, in part, by the National Institutes of Health (AI22667 and AI45252) and SmithKline Beecham Biologicals.
-
The authors are members of the SmithKline Beecham Biologicals Herpes Vaccine Advisory Board. The authors developed the concept for this paper at a meeting of the advisory board. The contents of this paper reflect the independent views of the authors.
- Received December 17, 1998.
- Revision received October 26, 1999.
- © 2000 by the Infectious Diseases Society of America
References
- ↵
-
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
-
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
- ↵
This Article
- AbstractFree
- » Full Text (HTML)Free
- Full Text (PDF)Free
Classifications
Services
Citing Articles
Google Scholar
PubMed
Related Content
Share
- Email this article
-
CiteULike
Connotea
Delicious
Facebook
Google+
Mendeley
Twitter
Published on behalf of
IDSA Members: For your free access to this journal, log in via the IDSA members area.
The Journal
Impact Factor: 8.186
5-Yr impact factor: 7.898
For The Media
For Authors
- Services for authors
- Self-archiving Policy
- IDSA Conflict of Interest policy
- Instructions for Authors
- ICMJE Form
- Submit Now!
Open access options for authors visit Oxford Open
This journal enables compliance with the NIH Public Access Policy
Corporate Services
Alerting Services
Related IDSA Publications
