8 Malhotra, R. et al. (1990) 1. Exp. Med. 172,955-959 9 Malhotra, R. et al. (1993) Biochem. J. 293,15-19 10 van den Dobbelsteen,M.E.A. et al. (1993) J. Immunol. 151,4315-4324 11 Malhotra, R. (1993) Behring Inst. Mitt. 93,254-261 12 Daha, M.R. et al. (1989) Eur. 1. Immunol. l&783-787 13 Geertsma, M.F. et al. (1994) Am. J. Physiol. 267, L578-L584 14 Burnet, F.M. and McCrea. 1.F. (1946) Aust. i Exp. Biol. Med. Sczy24,277:282 15 Gottschalk, A., Belyavin,G. and Biddle, F. (1972) in The Glycoproteins.
Composition, Structure and Function (Gottschalk, A., ed.), pp. 1082-1096, Elsevier 16 Hartley, C.A., Jackson, D.C. and Anders, E.M. (1992) J. Virol. 66, 4358-4363 17 Wakamiya, N. et al. (1992) Biochem. Biophys. Res. Commlm. 187,1270-1278 18 Hartshorn, K.L. et al. (1993) 1. Clin. Invest. 91,1414-1420 19 Hartshorn, K.L. et al. (1993) J. lmmunol. 151,6265-6273 20 van Iwaarden, I.F.C. et al. (1993) Am. Rev. Respir. I%. (Abstr.) 148, 146a 21 Malhotra, R. et al. (1994) Biochem. 1. 304,455-461
22 Hartshorn, K.L. et al. (1994) 1. Clin. Invest. 94,311-319 23 Jones, D.H. et al. (1994) Vaccine 12, 250-258 24 Anders, E.M. et al. (1994) 1. Gen. Virol. 75,6lS-622 25 Ezekowitz, R.A.B. et al. (1989) J. Exp. Med. 169,185-196 26 Haurum, J.S. et al. (1993) AIDS 7, 1307-1313 27 Andersen, 0. et al. (1993) Stand. 1. Immunol. 33,81-88 28 Ushijima, H. et al. (1992) \Dn1. Cancer Res.‘Gank 83,458-464 “. ” 29 Ushijima, H. et al. (1992) Res. Viral. 143,97-99
Herpes simplex virus vaccines as immunotherapeutic agents Lawrence R. Stanberry or most viral infections, the host makes a variety of immune responses that restrict virus replication and eliminate the agent from the host. Some viruses, however, have evolved unique strategies to evade the immune system and establish persistent infection. These infections may result in episodic disease [for example, recurrent infection with herpes simplex virus (HSV)], chronic progressive illness (for example, infection with hepatitis B virus or HIV) or may be clinically silent except in the immunocompromised subject (for example, infection with cytomegalovirus) (Table 1). HSV is a useful model for studying the pathogenesis and control of persistent virus infections. In the normal host, initial infection with HSV is generally self-limited. Early in infection, virus is transported within sensory nerves from the site of entry to regional sensory ganglia, the site of persistent infection. The immune system responds to infection by producing humoral, cellular and cytokine responses that suppress the replicating virus. Despite host responses, some virus escapes immune-mediated destruction by establishing a latent infection in
As persistent viruses can escape immune surveillance, chronic or recurrent disease can be a major problem. Only after nearly 60 years of work have recent reproducible data, using herpes simplex virus infection as a model for persistent viral disease, established that vaccine immunotherapy is effective in the treatment of such viral infections.
L. R. Stanberry is in the Divn of Infectious
Diseases, Children’s Hospital Research Foundation, University of Cincinnati College of Medicine, 3333 Burnet Avenue, Cincinnati, OH 45229-3039, USA. tel: +I 513 559 4578, fax: +1 513 559 7655
sensory ganglion neurons. During latent infection, the virus resides in a nonreplicating state, apparently undetected by the immune system. Unfortunately, for some patients, the latent virus periodically reactivates into a replication-competent form and, after transport to the periphery, may cause recurrent mucocutaneous HSV infections (recurrent genital herpes or recurrent herpes labialis). Because not all HSV-infected subjects experience recurrent infections, it is likely that host factors are important in controlling recurrent disease. Recurrent infecn
tions would be expected in the individual whose immune system failed to prevent reactivation or to eliminate reactivated virus at the periphery efficiently. In such individuals, induction or augmentation of specific anti-HSV immune responses might reduce the frequency and/or severity of recurrent disease. One tactic for enhancing host responses is treatment with an immunogenic vaccine (other immune-based tlierapies that might be useful in the control of persistent viral infections are listed in Box 1). Background As early as 1936, Brain suggested that vaccines could be used as therapeutic agents to treat viral diseases]. During the first half of the 20th century, two approaches to the control of recurrent HSV infections were put forward. One strategy involved treatment with live attenuated virus vaccines, such as vaccinia. This type of treatment was believed to cause a nonspecific upregulation of the immune system, which, in turn, resulted in a reduction in the frequency of recurrent HSV disease’. The alternative approach was to treat the patient with either live or inactivated HSV to stimulate undefined
1995 Elsevier Science Ltd
virus-specific immune responses important in the control of recurrent HSV infections. Not surprisingly, treatment with live virus resulted in some problems. There were some cases of disseminated vaccinia, and inoculation of unattenuated HSV into anatomical locations that were distinct from the site of recurrent disease often resulted in the development of recurrent disease at the treatment site3*4. Perhaps the greatest problem associated with vaccine immunotherapy has been the lack of rigorous scientific data to support the concept. Since 1938, there have been numerous clinical trials that have claimed that vaccine immunotherapy is effective for the treatment of recurrent HSV infections5,6. Unfortunately, most of these trials were poorly designed, and often controls were lacking. After nearly six decades of research, it is only recently that studies have provided solid experimental support for the concept of vaccine immunotherapy. Experimental studies The development of a guinea-pig model of recurrent genital HSV infections allowed the first rigorous tests of the concept of vaccine immunotherapy’. Genital infection in the guinea-pig is remarkably similar to genital herpes in humans, with self-limited, primary, vesiculoulcerative skin disease developing after intravaginal or intraurethral HSV inoculation. Virus replicates locally and is transported to sensory ganglia where, despite a full range of host responses, the virus establishes a latent state in sensory neurons. Periodically, the latent virus reactivates, resulting in clinically apparent or inapparent (viral shedding in the absence of detectable lesions) recurrent infections. To examine the concept of vaccine immunotherapy, latently infected guinea-pigs that had recovered from primary genital HSV-2 infection were immunized with a vaccine containing genetically engineered HSV glycoproteins B and D (gB and gD) and complete Freund’s adjuvant. Treatment produced a 4-244-fold rise in virus-specific antibody titer and significantly reduced the frequency and severity of recurrent
Table 1. Persistent viral infections in humans VirUS
Clinical consequences of persistent lnfectlon
Herpes simplex virus types 1 and 2 Varicella-zoster virus Cytomegalovirus Epstein-Barr virus HIV Enteroviruses Papillomaviruses Hepatitis B virus Hepatitis C virus Parvovirus B19 Measles virus Human T cell leukemia viruses
Recurrent mucocutaneous infections, encephalitis Zoster (shingles) Pneumonia, retinitis, colitis, encephalitis Lymphoproliferative disorder AIDS Chronic/recurrent meningoencephalitis Warts, papillomatosis, cervical carcinomas (?) Chronic hepatitis, hepatocellular carcinoma Chronic hepatitis Agranulocytosis, granulocytic aplasia Subacute sclerosing panencephalitis Leukemia
genital HSV infections*. This was the first controlled experimental study demonstrating that treatment of a persistently infected host with an immunologically potent vaccine could alter the natural history of the infection favorably. Subsequent experiments in guinea-pigs extended the original observations and explored factors that influence efficacy. Dose, route of administration and timing relative to primary infection are all important9. The type of adjuvant used in the vaccine formulation is critical for therapeutic efficacy. Vaccines containing HSV gD, together with a potent adjuvant, such as complete Freund’s or synthetic muramyl tripeptide (MTP), were effective in reducing recurrent disease, while gD combined with the relatively weak adjuvant aluminium hydroxide (alum) was ineffective9.i0. The effectiveness of the gD-based vaccines was enhanced further by incorporating the immunogen with MTP in a liposome carrier”. The gD-MTP liposomal vaccine enhances antigenspecific lymphocyte stimulation and reduces the frequency and severity of recurrent genital infections to 25 % of those in untreated animals’ I, A novel approach ro increasing the immunogenicity of therapeutic vaccines is the construction of fusion proteins consisting of a viral antigen and human interleukin 2 (IL-2)“. Studies showed that a fusion-protein-based vaccine, consisting of truncated HSV-1 gD and IL-2, is highly immunogenic and reduced recurrences to 35% of those in
untreated animals13. Replicationincompetent viruses are also being explored as possible therapeutic vaccines. These viruses are constructed to lack an essential gene and, therefore, must be propagated using special cell lines that have been engineered to express the missing gene product. As a consequence, they can infect normal cells, but they undergo at most one replication cycle. Normal cells infected with these viruses synthesize immunogenic viral proteins, but fail to produce infectious virions. Prophylactic immunization with viruses lacking gH or infectedcell protein 8, a DNA-binding protein, protects animals from HSVinduced disease14-16. Currently, the gH-deleted HSV-2 vaccine is being evaluated in the guinea-pig model as a therapeutic vaccine (S. Inglis, pers. commun.). Both guinea-pigs and humans can shed HSV from the genital tract in the absence of clinically recognized lesions. Because asymptomatically shed HSV is an important source of transmissible virus, it was important to determine whether immunotherapy simply converts symptomatic
Box 1. Immune-based therapies potentially useful in the control of persistent viral infections *Transfer of immunologically lymphocytes lImmunoglobulin therapy *Vaccine immunotherapy lcytokine therapy lImmunomodulator therapy
recurrences to asymptomatic virus shedding. Using the guinea-pig model, immunotherapy of latently infected animals with HSV-1 gB and gD together with complete Freund’s adjuvant was shown to reduce the frequency of both clinical recurrences and cervicovaginal shedding”. The mechanism by which vaccine immunotherapy reduces recurrent HSV infections has not yet been established. Treatment of latently infected animals with most HSV vaccines causes a rise in anti-HSV antibody titers. However, the ability of a vaccine formulation to induce high-titer antibody does not predict its effectiveness at reducing recurrent infections. Recent studies have shown that therapeutically effective vaccines enhance antigen-specific lymphoproliferative, cytolytic and cytokine responses”J*. This suggests that vaccine immunotherapy may act by enhancing cell-mediated immune responses. Clinical studies
One of the first trials of a vaccine to treat recurrent HSV infections was reported by Frank in 193819. The vaccine consisted of a suspension of formalin-inactivated, HSVinfected rabbit brain. A multidose course of treatment was given to 14 volunteers; 13 subjects reported a subjective increase in the interval between herpetic recurrences. The results were surprising because the treatment did not increase antibody titers, suggesting that the vaccine was not immunogenic. The study also lacked a placebo control group and relied on subjective end points, rather than an objective assessment of the frequency of recurrent disease. Between 1946 and 1982, at least 27 similar studies were conducted to test the effectiveness of inactivated or subunit HSV vaccines at controlling recurrent HSV infections. Depending on the trial, O-100% of patients reported improvement. Like the original study by Frank, these studies relied on a subjective assessment of efficacy, and most lacked placebo control groups. Not all studies lacked appropriate controls. In 1964, Kern and Schiff conducted a randomized trial in which the subjects did not know
whether they received vaccine or placebo20. In their study, volunteers treated with an inactivated wholevirus vaccine reported a 70% reduction in recurrent infections, while 76% of the placebo recipients reported subjective improvement. This trial failed to establish that vaccines are useful in treating recurrent HSV disease, but it illustrates the significance of the placebo effect and the importance of including appropriate controls when evaluating HSV vaccine immunotherapy. Design flaws continue to complicate the interpretation of morerecent clinical studies. In 1983, Woodman reported a study in which volunteers recovering from the initial episode of genital herpes were either immunized with an HSV-lderived subunit vaccine or were unimmunized and acted as contro1s2i. Genital recurrences were reported by 7 of 22 (3 1% ) immunized subjects, compared with 17 of 20 (85%) unimmunized volunteers (mean follow-up periods were 12 and 8 months, respectively). These data suggest that vaccine-induced enhancement of host immunity improves the control of recurrent disease. Unfortunately, the study was not placebo controlled, recurrent infections were not confirmed by medical personnel and the virus type causing primary infection was not established for 16 of the 42 volunteers. Virus type is an important variable that must be controlled because primary genital infection due to HSV-2 is twice as likely to cause recurrent genital infections and recurs 8-lo-fold more frequently than does genital HSV-1 infection22. In 1985, Cappel and colleagues published their results using an HSV-2 envelope-glycoprotein vaccine. In their study, the 59 volunteers treated with vaccine reported fewer and less-frequent recurrences than did the 33 unimmunized patients23. When recurrences did occur, they were described as ‘less severe’ and ‘of shorter duration’. Unfortunately, the two comparison groups knew whether or not they had been immunized; hence, there may have been a placebo effect in the immunized group. Furthermore, Cappel et al. used a mixed study population including subjects with
recurrent genital herpes or recurrent herpes labialis. Because these illnesses have different natural histories, it is not appropriate to combine the two populations to analyze vaccine efficacy. A study published in 1988 using 42 volunteers examined the efficacy of an HSV-1 glycoprotein vaccine to treat frequent, recurrent genital or oral/labial herpes24. Again, the study design was complicated by using a mixed population of patients with respect to the type of recurrent herpes infections that they had. While comparable selfreported improvement in both the vaccine (57%) and placebo (45%) recipients was seen, the vaccine was only minimally immunogenic, producing little or no rise in antibody titers, and so it was not surprising that the study failed to demonstrate a therapeutic effect. In a recent, double-blind placebocontrolled trial, Straus and colleagues evaluated the efficacy of an immunogenic, HSV-2 gD (gD2) vaccine for treating recurrent genital herpes in 98 subjects with documented, symptomatic HSV-2 infection2s. Patients with frequent recurrences randomly received either two doses of gD2 in alum (100 pg at the start and after 2 months) or alum alone. The subjects were evaluated for 12months. Recurrent infections were confirmed by the isolation of virus from lesions and/or the observation of typical herpetic lesions by an investigator who did not know which treatment group the patients were in. The vaccine recipients experienced fewer recurrences per month than did the placebo recipients (mean 0.422 0.05, compared with 0.55kO.05; P=O.OSS), had fewer virologically confirmed recurrences per month (0.18+0.03, compared with 0.282 0.03; P=O.O19) and had a lower median number of recurrences in the study year [4 (range O-17), compared with 6 (range O-15); P=O.O39]. The vaccine was immunogenic and caused a fourfold increase in HSV-2-neutralizing antibody titer and a sevenfold increase in gD2-specific antibody levels. This study was the first controlled trial documenting the usefulness of vaccines in treating patients with
Questions Why do some individuals infected with herpes simplex virus develop frequent severe recurrences, while others never experience recurrent infections? l What are the elements of an ideal clinical study designed to examine vaccine immunotherapy? What have been the problems associated with most studies of vaccine immunotherapy? l Does vaccine immunotherapy act by blocking the reactivation of latent infection or by preventing the reactivated virus from replicating at cutaneous sites? l What is the mechanism(s) by which vaccine immunotherapy reduces herpetit recurrences (for example, enhancement of cytotoxic T cell responses, enhancement of antibody-dependent cell-mediated cytotoxicity or enhance ment of cytokine responses)?
recurrent HSV infections. Currently, several other trials are under way that are designed to explore the immunotherapeutic efficacy of new vaccines in the treatment of persistent viral infections. Conclusions The concept of using vaccines to treat persistent viral infections has a long but dubious history. While many trials were said to prove the effectiveness of vaccines in treating viral diseases, most of these studies were poorly designed or used vaccines that were not highly immunogenic. Moreover, many physicians and scientists held the opinion that if the immune responses of the host to infection could not prevent or limit chronic or recurrent viral disease, then it was unlikely that treatment with a vaccine could provide better control. With the development of a unique guinea-pig model of genital herpes, investigators were finally able to prove that highly immunogenic vaccines were effective in treating recurrent HSV infections. Not surprisingly, even these rigorously controlled studies were met with much skepticism. Recently, however, a carefully conducted clinical study has confirmed the animal research, and shown that an HSV vaccine is effective in treating patients with recurrent genital herpes25. What is now required is further research to define the immunological responses that are important in the control of recurrent HSV infections. When these are known, it should be possible to develop vaccine formulations that optimally enhance these specific responses. This research will not only provide
new treatments for patients with recurrent herpes infections, but may also yield information about why some patients never experience recurrent disease after recovering from primary HSV infection, while others suffer frequent recurrences. It is obvious from the studies of HSV vaccine immunotherapy that it is possible to stimulate the immune system of a persistently infected host in a way that results in better control of recurrent disease. Hence, it is possible to produce responses that are superior to those that result from infection. Depending on the mechanism by which a persistent virus escapes immune surveillance, it may also be possible to develop vaccines that are effective in treating other chronic and recurrent viral diseases.
Virus: Pathogenesis, lmmunobiology and Control (Rouse, B.T., ed.), pp. 15-30, Springer-Verlag 8 Stanberry, L.R. et al. (1988)J Infect. Dis. 156,156-163 9 Stanberry, L.R. et al. (1989) Antiviral Res. l&203-214 10 Berman, P.W. et al. (1988)J. Infect. Dis. 157,897-902 11 Ho, R.J.Y. etal. (1989)J. Virol. 63, 2951-2958 12 Hinura, S. et al. (1991) FEBS Lett. 288, 138-142 13 Nakao, M. et al. (1994) J. Infect. Dts. 169,787-791 14 Knipe, D.M. et al. in Vaccines 1995: Modern Approaches to New Vaccines b&ding Prevention of AlDS (Channock, R. et al., eds), Cold Spring Harbor Laboratory Press (in press) 1.5 Morrison, L.A. and Knipe, D.M. (1994) I. Virol. 68,689-696 16 Farrell, H.E. et al. (1994)J. Viral. 68, 927-932 17 Myers, M.G. et al. (1988) Antiviral Res. 10,83-88 18 Bernstein, D.I. et al. (1991) J. lmmunol. 146,3571-3577 19 Frank, S.B. (1938) I. Invest. Demratol. 1, 267-282 20 Kern, A.B. and Schiff, B.L. (1964) Arch. Dematol. 89,844-845 21 Woodman, C.B.J. et al. (1983) BY.J. Vener. Dis. 59,311-313 22 Webb, D.H. and Fife, K.H. (1987) Infect. Dis. Clin. N. Am. 1, 97-122 23 Cappel, R. et al. (1985) J, Med. Viral. 16,137-145 24 Kutinova, I. et al. (1988) Vaccine 6, 223-228 25 Straus, S.E. et al. (1994) Lancet 343, 1460-1463
Acknowledgements I thank my friends and colleagues who have played important roles in developing and testing the concept of vaccine immunotherapy; they include Drs Martin Myers, David Bernstein, Nigel Bourne, Rae Lyn Burke and Moncef Slaoui. This work was supported by grants from the National Institutes of Health (AI 22667, AI 29687, AI 15101, AI 37940 and AI 45252). References 1 Brain,R.T.(1936) BY.].Dematol. Syph. 48,21-26 2 Schiff, B.L. and Kern, A.B. (1954) Postgrad. Med. l&32-36 3 Lyon, E. (1961) 1s~.Med. 20,103-108 4 Jawetz, E. et al. (1955) Am. 1. Med. 299, 477-485 5 Stanberry, L.R. (1990) in Herpesviruses, the Immune System, and AlDS (Aurelian, L., ed.), pp. 309-341, Kluwer Publishing 6 Shanley, J.D. and Stanberry, L.R. (1994) Rev. Med. Viral. 4, 105-118 7 Stanberry, L.R. (1992) in Herpes Simplex