Immunization to reduce the frequency and severity of herpes zoster and its complications

The past three decades have seen herpes zoster (HZ) research reach a significant threshold. Investigators are now in a position to determine whether immunization can reduce the incidence and/or severity of HZ and its complications. Such a determination will, at last, provide a test of Edgar Hope-Simpson's prescient hypothesis ^[1] that the occurrence of HZ is determined by host immunity to varicella-zoster virus (VZV). Summarized below are some of the important observations that provide a rationale for this approach to the prevention of HZ and its complications.

HZ is a localized disease characterized by unilateral radicular pain and a vesicular cutaneous eruption that is generally limited to the dermatome innervated by a single spinal or cranial sensory ganglion. It results from reactivation of an endogenous VZV infection that has persisted in a latent form within the sensory ganglia following an earlier attack of varicella. ^[1,2] Both the incidence and the severity of HZ increase with advancing age, and complications, which are rare in normal children and younger adults, occur in almost one-half of older individuals. ^[1-4] The most frequent complication of HZ in older individuals is postherpetic neuralgia (PHN), a syndrome of pain, allodynia, dysesthesia, and hyperesthesia that persists or develops after the dermatomal rash has healed. PHN can be prolonged and disabling, and its frequency and severity also increase with advancing age.

Pathogenesis of HZ: the role of host immunity.

Hope-Simpson proposed that host immunity controls reactivation of latent VZV and limits the spread of reactivated virus within the ganglion and from the ganglion to the skin. He hypothesized that HZ occurs when VZV-specific immunity falls below some critical threshold Figure 1. ^[1] He suggested that this immunity declines as a consequence of the aging process itself but may be boosted periodically by exogenous exposure to children with varicella, as well as by endogenous exposure to VZV antigens produced during episodes of ganglionic reactivation that are contained by host immune responses ("contained reversions'') and thus do not result in HZ. He also hypothesized that HZ itself induces a marked increase in immunity to VZV and that this explains the rarity of second episodes of HZ.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Pathogenesis of herpes zoster. (Modified, with permission, from Figure 1 in reference 2, which was based on the original by Hope-Simpson. ^[1])

Most of the elements of Hope-Simpson's remarkable hypothesis have since been validated. Endogenous reactivation in the absence of rash has been documented and has been shown to boost cell-mediated immunity (CMI) to VZV, and HZ itself has been demonstrated to markedly increase VZV-specific CMI. ^[5-9] Exogenous reexposure resulting from contact with children with varicella has also been shown to boost humoral immunity and CMI to VZV. ^[10,11] Finally, a marked decline in host immunity with advancing age has been documented by a number of investigators ^[9,11-16] Figure 2. In fact, the only aspect of our current understanding not encompassed by Hope-Simpson's model is the recognition that it is VZV-specific CMI rather than humoral immunity that is critical in limiting the occurrence of HZ and its complications. ^[9,11-16]

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. Age-related decline in VZV-specific cell-mediated immunity. PHA equals phytohemagglutinin. (Modified, with permission, from Burke et al, Arch Intern Med 1982;142:291-293. ^[14] Copyright 1982, American Medical Association.)

Ganglionic latency: VZV versus herpes simplex virus.

Virus-specific CMI also plays an important role in the pathogenesis of recurrent herpes simplex virus (HSV) infections, but differences in the cellular sites and mechanisms of HSV and VZV latency appear to dictate important differences in the pathophysiology of reactivation and the characteristics of the resulting diseases Table 1. These differences have important implications for the potential efficacy of immunologic interventions.

Aided by the availability of excellent animal models, a number of investigators have identified the sensory neuron as the ganglionic site of HSV latency (reviewed by Fraser et al ^[17]). Furthermore, during HSV latency, transcription is limited to only one small region of the viral genome, resulting in a set of latency-associated transcripts that overlap the 3 prime end of one of the immediate early HSV genes (ICPO) in an antisense manner. Latency-associated transcripts are not polyadenylated and remain confined to the cell nucleus, and there does not appear to be any HSV-encoded protein synthesized in latently infected neurons. Recent studies by Straus and his colleagues ^[18] indicate that the site of VZV latency is not the sensory neuron but the satellite cells that surround it. Moreover, in contrast to HSV, polyadenylated messenger RNAs corresponding to several of the immediate early and early genes of VZV are transcribed in the latently infected satellite cells. These anatomic and molecular biologic differences provide a possible explanation for many of the differences between the natural history of HZ and that of recurrent HSV Table 1.

Biologic implications of the site and nature of viral latency.

The potential role of virus-specific CMI in controlling VZV latency and reactivation.

Two features of VZV latency and reactivation, in particular, suggest an important role for host immunity in limiting the frequency and severity of HZ and its complications. Several immediate early and early VZV genes appear to be expressed during latency, and this may permit latently infected cells to be "recognized'' by the host's immune system. Consequently, VZV-specific immune responses may play a direct role in limiting reactivation itself. When reactivation does occur, the requirement for virus replication and spread within the ganglion provides ample opportunity for host immune responses to halt the process before the skin is involved, and this is almost certainly the explanation for zoster sine herpete, as well as for episodes of asymptomatic reactivation detectable only by the resulting boost in VZV-specific immune responses. ^[5-8,22-24] These considerations support the supposition that the marked increase in the frequency and severity of HZ and its complications observed in elderly persons ^[1-4] Figure 3 is caused by the age-related decline in CMI to VZV Figure 2. ^[12-15]

• Download figure
• Open in new tab
• Download powerpoint

Figure 3. Incidence of herpes zoster versus age. (Reprinted, with permission, from Ragozzino et al, Medicine 1982;61:310-316. ^[3])

Therapeutic implications.

Given these observations, what can be done to reduce the frequency and severity of HZ and PHN in older persons? One approach is to inhibit VZV replications as soon as possible after the rash of HZ has appeared by the prompt administration of antiviral agents. However, while antiviral therapy can speed the resolution of the rash and may also affect the duration of PHN, such treatment is unlikely to eliminate this complication because many of the pathologic changes responsible for PHN will already have occurred before the rash appears. The best hope of reducing the incidence of HZ and its complications, particularly PHN, lies in preventing reactivation itself, or at least blocking the subsequent replication and spread of reactivated virus within the ganglion, events that precede the development of the cutaneous eruption. The recently licensed live attenuated Oka/Merck varicella vaccine offers the possibility of accomplishing this goal by boosting immunity to VZV in elderly persons.

Immunization: live attenuated varicella vaccine.

In the early 1970s, Takahashi et al ^[25] isolated a strain of VZV from a Japanese child with varicella and attenuated it by serial passage in human and guinea pig cells. This vaccine and derivatives of it (eg, the Oka/Merck strain) induce VZV-specific humoral immunity and CMI in normal and immuno-compromised children and in normal adults who are susceptible to VZV. ^{[9,11,16,25-34]} The vaccine is clearly attenuated ^[32] and has not been associated with significant side effects in normal vaccinees. ^[16,25-34] In addition, these vaccines have been shown to boost VZV-specific CMI when administered to seropositive elderly adults. ^[9,16,33-38] Levin et al ^[16,35] have immunized more than 200 seropositive elderly adults (mean age equals approximately 67 years) with live attenuated Oka/Merck varicella vaccine. The vaccine was well tolerated and the majority of recipients exhibited a significant boost in VZV-specific cellular immunity, as indicated by an increase in VZV-specific responder cell frequency from a prevaccination mean below 1/70,000 (1.4 responder cells per 10⁵ peripheral blood mononuclear cells [PBMC]) to a mean of approximately 1/40,000 (2.5 responder cells per 10⁵ PBMC) 6 months after vaccination Figure 4, a response comparable to that induced by an episode of HZ. ^[9,16] The level of VZV-specific CMI achieved by vaccination is typical of that observed in 40-year-old individuals, in whom the incidence and severity of HZ is markedly reduced.

• Download figure
• Open in new tab
• Download powerpoint

Figure 4. VZV-specific cell-mediated immune response of elderly subjects to live attenuated Oka/Merck varicella vaccine. Vaccine administered on day 0 (based upon data in reference 35). PBMC equals peripheral blood mononuclear cells.

The boost in VZV-specific cellular immunity induced by varicella vaccine was long-lasting, having a half-life of 54 months. ^[35] During more than 4 years of observation representing more than 800 patient-years, eight episodes of dermatomal pain, skin lesions, or both were reported in vaccine recipients. Only one was virologically confirmed and two were immunologically confirmed as HZ, and none were associated with prolonged pain, extensive rash, or PHN. ^[35]

Immunization to reduce the frequency and severity of HZ.

The finding of an age-related decrease in VZV-specific CMI Figure 2 and the observation that the incidence and severity of HZ and its complications are increased in elderly and immunocompromised persons ^[1,3,4] Figure 3, taken together, raise the obvious question of whether this association is causal or merely coincidental. The availability of a safe and effective live attenuated varicella vaccine capable of boosting VZV-specific CMI in elderly seropositive individuals offers an opportunity to answer this question. However, because of the relatively low annual incidence of HZ and PHN, and because of the impracticality of a postimmunization surveillance period much longer than 3 years in elderly subjects, the answer can be obtained only by means of a large, prospective, placebo-controlled clinical trial.

The protocol for such a trial, involving a total of 27,000 subjects 60 years of age or older randomized to receive either vaccine or placebo and then actively followed for a minimum of 3 years, has been developed in collaboration with Merck and Company and approved by the Veterans Administration Cooperative Studies Program. With a total of 160 to 240 cases of HZ and 30 to 50 cases of PHN expected to occur in the placebo group, this study should provide a definitive answer to the question of whether immunization can reduce the frequency and/or severity of HZ and PHN. It should also provide the first real test of Hope-Simpson's hypothesis that host immunity is the primary determinant of the incidence and severity of HZ and its complications.