How Does Herpes Simplex Virus Type 2 Influence Human Immunodeficiency Virus Infection and Pathogenesis?

In this issue of The Journal of Infectious Diseases Reynolds et al. [1] present results from a prospective cohort study of patients at sexually transmitted infection (STI) clinics in Pune, India, that show that antecedent herpes simplex virus type 2 (HSV-2) infection markedly increases the rate of human immunodeficiency virus type 1 (HIV-1) infection. An interesting and novel finding in their study was a 2-fold higher risk of HIV-1 acquisition among those with recent incident HSV-2 infection (0–6 months after acquisition of HSV-2), compared with those with preexisting prevalent HSV-2 infection The relationship between genital ulcers and HIV-1 has been noted by HIV epidemiologists for >20 years, but it is only recently that HSV infection has been recognized as the major factor behind this association. Why now? Perhaps first and foremost has been the recognition in the last 10 years that the association between HSV and HIV is driven not only by clinically apparent genital ulcers but also by the invariable subclinical reactivation and attendant microscopic mucosal ulcerations of genital HSV-2 infection. Clinically recognized genital herpes is truly the tip of the HSV-2 iceberg, regardless of locale. It is noteworthy that Reynolds et al. [1] found a baseline HSV-2 prevalence of 43%, and as high as 89% among female commercial sex workers. Nevertheless, genital ulcers were infrequently reported. These high rates of HSV-2 seroprevalence in the face of infrequently recognized disease have been consistently reported from all regions of the world and in diverse patient populations Such seroepidemiologic studies are sobering when placed in the context of recent biological studies which show that >95% of HSV-2-seropositive persons, regardless of whether they report clinically apparent genital ulcers, have viral reactivation in the genital region [2, 3]. Mucosal HSV ulcerations, whether symptomatic or asymptomatic, are associated with the influx of activated CD4+ T cells, which are easily infectable by HIV-1 [4, 5]. Episodes of subclinical HSV-2 reactivation are frequent, even in the healthy human host. HSV shedding is found on a median of 10%–15% of days among immunocompetent men and 20%–25% among immunocompetent women with HSV-2 infection [3, 6]. Interestingly, rates of reactivation are 2–3 times higher in the first 6 months after acquisition of genital HSV-2 than at later time points [7–9], which provides a biological explanation for the higher incidence of HIV-1 among persons with recent incident HSV-2 infection, compared with those with prevalent HSV-2 infection, reported in Reynolds et al. [1] To us, the question is why the link between HSV-2 and HIV-1, which was reported in 1988 in the United States as increasing susceptibility to HIV-1 among men who have sex with men [10], as well as in Europe as part and parcel of the heterosexual transmission of HIV-1 [11], has not been seriously considered as a major factor in the heterosexual transmission of HIV-1 in the Southern hemisphere. As true believers in this link, we may not be the best source of speculation [12] The commercial development of type-specific serologic tests for HSV-2 infection has been critical in amassing the large body of seroprevalence data showing the high prevalence rates of unrecognized HSV-2 infection worldwide [13–15]. Similarly, it is clear that studies using polymerase chain reaction technology have resulted in the recognition of HSV as the cause of genital ulcers at rates 3–5 times higher than those found using culture methods [16–18]. Studies now show that, in all regions of the world, HSV is the most common cause of genital ulcers [19–21]. These new investigative tools have brought recognition of the largely silent genital herpes epidemic in resource-poor regions of the world [22–24]. Moreover, clinical appreciation for genital herpes in the developing world did not occur until the HIV-1 epidemic. In developing countries, persons with HIV-1 and HSV-2, who have more extensive and more persistent genital herpes, are more likely to present for care than are persons with the mild and short-lasting lesions typical of immunocompetent persons with genital herpes To date, the focus of STI control has traditionally been on bacterial STIs—those that can be diagnosed clinically or with simple laboratory tests and those that can be cured, preferably with a simple regimen of oral therapy. Unfortunately, the viral STIs, especially HSV-2 infection, do not fit into this paradigm. The need to include genital herpes in STI control programs is perhaps best illustrated by retrospective analyses of a study in the Rakai district of Uganda, a study in which no impact of mass treatment of bacterial STIs was found on HIV-1 incidence [25]. HSV was the most common inflammatory STI in this population [26]; the prevalence of HSV-2 infection was 31% among men and 61% among women aged 15–29 years. Importantly, prevalent HSV-2 infection was associated with a higher per contact rate of HIV acquisition The critical issue for all of us in investigative medicine is to learn from the past and move forward. As discussed by Reynolds et al. [1], all observational and cross-sectional studies of this subject are subject to criticism about the effect of confounding by sexual behavior on the association between genital herpes and HIV-1 acquisition. Acquisition of HSV-2 could be a marker for exposure to HIV-1-seropositive persons, who are likely to be shedding HSV more frequently and at higher amounts than persons without HIV-1 [27–29]. This may, in fact, explain why HIV-1-seropositive persons appear to have a higher rate of HSV-2 acquisition than do HIV-1-seronegative persons: the cause may be not increased susceptibility among HIV-1-seropositive individuals, but contact with sex partners who are infected with both viruses and are likely to be more infectious for both. The inability to disentangle, even in prospective studies, the extent of confounding by sexual behavior has led some people to doubt that the proposed causal pathway from HSV-2 to HIV-1 exists, despite strong evidence for biologic plausibility. Overcoming this dilemma requires an interventional approach. An intervention study with anti-HSV therapy is, in our opinion, the best method of ascertaining and quantifying the risk of HIV-1 that can be attributed to HSV-2 infection and to define the true biology behind the issues raised in the study by Reynolds et al. [1] Generic acyclovir (400-mg tablet twice daily) is inexpensive—it can be obtained for 20 cents per day, or $73.00 per year, and it has a safety profile that requires no clinical or laboratory monitoring, even with daily use for many years. Acyclovir is highly effective in reducing both clinical and subclinical HSV-2 reactivation, and its effectiveness is maintained after many years of continued use [30–32]. Biologic resistance is low because of the unique role the viral enzymatic target of acyclovir, the viral thymidine kinase, plays in neurovirulence, transmission, and reactivation [33]. Thymidine kinase-deficient strains are less fit epidemiologically. Worldwide use of acyclovir-like drugs has increased 6-fold in the last decade, with little alteration in the frequency of resistance [34]. Thus, although monitoring and evaluation of resistance must be undertaken, there is optimism that the use of generic acyclovir for treatment of HSV infection in the developing world is a policy whose time has come The issues of amplification of HIV-1 acquisition (increasing susceptibility) and of HIV-1 transmission (increasing infectiousness) by HSV-2 infection need to be tackled separately. As the study by Reynolds et al. [1] demonstrates, prevalent HSV-2 infection increases the risk of HIV-1 acquisition. Because the increased risk of HIV acquisition in HSV-2-seropositive persons has been noted in every population studied and on most continents, the hypothesis should be tested in a diverse population to ensure that the results are generalizable. The conceptual basis for such a study would be to randomly assign persons who are HSV-2 seropositive to receive acyclovir or placebo and determine whether daily acyclovir reduces HIV-1 acquisition by reducing the frequency of symptomatic and subclinical HSV-2 reactivation. Bacterial STI control would also be a necessity in such a trial. Perhaps the greatest challenge of such an approach would be to motivate healthy at-risk persons to adhere to a regimen of a daily medication. Sustaining alterations in behavior for preventative reasons has proven to be difficult in most contexts There are also considerable published data that indicate that both subclinical and clinical HSV-2 reactivation increase the frequency and amount of mucosal HIV-1 shedding [35–37]. Whether reducing HSV-2 reactivation will reduce HIV infectivity in persons with both HIV and HSV-2 is, perhaps, more difficult to study, but exploration of this question may bring greater benefits. There are relatively few prospective studies of HIV transmission, because it requires a couples design: the exposure and the outcome are measured in different persons. This complicates follow up, and the success of the study partly depends on the vagaries of the relationship. In addition, transmission events in prospective studies are relatively infrequent. The causes are likely multiple and include counseling of both partners, as well as the frailty effect—those most susceptible are likely to have acquired infection already [38–40]. Despite these caveats, this approach may have a more powerful public health impact and may be relevant in both the developed and the developing world. With the exception of condom use, there is no available biomedical intervention for prevention of HIV transmission. Antiretroviral therapy should work, and such a study is planned. But antiretroviral therapy, even in the United States, is not generally initiated until CD4+ T cell counts are <350 cells/mm3. Most HIV infections worldwide are transmitted by persons who do not need antiretroviral therapy for their own health. It is in this context that anti-HSV therapy may prove to be beneficial, if it decreases the risk of HIV transmission from persons with both infections. Availability of a biomedical intervention to decrease the risk of HIV transmission to sex partners may encourage individuals to be tested for HIV and medical providers to encourage voluntary HIV testing In summary, the superb work by Reynolds et al. [1] helps to provide an understanding of the interactions between 2 divergent pathogens. Evidence for an epidemiologic and clinical synergism between HSV-2 and HIV-1 continues to accumulate. The task for investigative medicine is to see whether these data and design strategies can be used to interrupt these 2 interacting epidemics