Preventing neonatal herpes infections through maternal immunization

Pathogens that infect the fetus and newborn are known collectively by the acronym TORCH: Toxoplasma gondii, other, rubella virus, cytomegalovirus and herpes simplex virus (HSV). ‘Other’ pathogens include syphilis, varicella-zoster virus, parvovirus B19, HIV and the most recent addition, Zika virus. These infections are a significant cause of mortality and morbidity, manifesting with neurological sequelae in the survivors. Sensorineural deafness, seizures, microcephaly and intellectual disability are some of the neurological outcomes common to many of these congenital infections [1]. For HSV, neurological infection is present in about a third of neonatal HSV infections. Even with antiviral therapy, there is still a high likelihood of permanent neurological abnormalities [2]. Here we present evidence that maternal immunization to induce virus-specific antibodies would be an effective strategy to prevent neonatal HSV.

The global incidence of neonatal HSV is estimated to be 1.03 per 10,000 live births annually [3]. However, estimates of neonatal HSV have been reported to be as high as 1 in 3200 live births in the USA [4]. Globally, HSV-2 causes approximately 70% of neonatal HSV, but HSV-1 is thought to be the major cause in the Americas, Europe and Western Pacific [3]. Acquisition occurs perinatally when the neonate is exposed to vaginal HSV during parturition. When active maternal lesions prompt a cesarean delivery, HSV transmission is prevented, but HSV shedding is often asymptomatic and prophylaxis is often unexploited [4]. Therefore, new prophylactic avenues are being explored to prevent the global burden of this disease.

The idea that maternal antibodies can protect neonates from HSV originated from clinical observations [4,2]. Fifty percent of babies delivered to women with first-episode primary infection develop neonatal HSV disease while only 1–2% of babies delivered to women with recurrent HSV develop the disease [4,5]. The most likely explanation for this observation is that antiviral maternal antibodies induced during previous infection transfer to the neonate and protect them from HSV disease [6]. This idea was further supported by animal studies in the 1980s demonstrating that prior to maternal infection and maternal neutralizing, antibody titers correlate with protection in pups [7,8]. Despite these clinical and animal studies, there has been surprisingly little effort to translate these findings to the clinic. While the development of acyclovir in the 1980s reduced the urgency for development of immune-based approaches, treatment with intravenous acyclovir for neonatal herpes has limitations. A high-degree of clinical suspicion and empirical therapy are critical for successful acyclovir treatment. This is challenging since neonatal HSV symptoms are nonspecific and accurate, and timely diagnosis is often hampered due to initial suspicion of bacterial sepsis [9]. Moreover, even when high-dose acyclovir is given, the risk of neurological sequelae remains high [10]. We believe that these shortcomings may be overcome by strategies to stimulate maternal antibody production during pregnancy. In a recent study by our group, we found that antiviral maternal antibodies transfer efficiently to the naive neural tissue of newborn mice prior to infection and effectively prevent neurological HSV infection [11]. We also uncovered evidence that these maternal antibodies persist in the neural tissues of neonates in humans. This work therefore documents a neuroprotective role of maternal antibodies and underscores the potential of maternal immunization to prevent congenital neurological infections.

Despite the multiple failures in the HSV vaccine field, we propose that maternal vaccination aimed at preventing neonatal herpes is feasible. To date, all vaccine clinical trials for genital HSV have been discontinued due to low efficacy in horizontal (adult-to-adult) transmission. The bar for success of these vaccines is high: they are required to induce a combined cell-mediated and humoral immune response that prevents HSV shedding for the entire sexually active phase of the recipient. While the development of vaccines to prevent the morbidity of adult genital herpes is important, the predominant situation in which genital herpes leads to mortality is when HSV is transmitted to neonates. Therefore, specific immunization strategies to prevent neonatal HSV warrant independent consideration. Compared with the adult genital herpes vaccines, we hypothesize that a successful maternal vaccine would only need to clear a relatively low immunological bar. It would only need to induce a robust humoral response during the perinatal period when HSV transmission occurs. Given that primary maternal infection leads to the greatest risk of transmission, the target population of this vaccine would be the most at-risk women who are seronegative for HSV-1 and HSV-2. This maternal vaccine would need to: induce a robust anti-HSV IgG response against HSV-1 and HSV-2; cross the placenta to the fetus; persist in the neonate during the perinatal period; and ultimately, protect the neonate from HSV disease and sequelae. Since naturally occurring maternal IgG readily crosses the placenta and persists in the newborn for up to 6 months [12], the induction of maternal antibodies is an ideal strategy for preventing such perinatal infections.

There are several existing genital HSV vaccines that induce a robust humoral response. The Chiron subunit vaccine induced neutralizing antibodies against gD and gB that were greater than or equivalent to naturally acquired infections [13,14]. Newer vaccines in current pipelines, including replication-defective vectors [15,16] and trivalent subunit vaccines [17], can all potently induce protective antibodies in animal models. Additionally, most adult humans are infected with HSV-1 and HSV-2, so purifying hyperimmune antibodies from pooled human sera for passive immunization would also be a feasible alternative. At last, these IgGs could be engineered for specific epitope recognition, specific FcR binding, or greater half-life stability through in vitro methods. It remains to be tested whether antibodies from these strategies could transfer, persist and protect the newborn from HSV infection.

In addition to testing the efficacy of these maternal vaccines, there are still many outstanding questions that remain to be addressed. Which antibodies, in terms of subclass and antigen specificity, are most protective against neonatal HSV? How do these antibodies exert antiviral function? Do antibodies in milk also play a role in preventing neonatal HSV disease? And at last, how do these antibodies gain access to and persist in naive neural tissue? Recent studies show that maternal antibodies can transfer into the developing peripheral and central nervous system during gestation [11,18]. These findings coupled with reports that maternal antibodies are effective against congenital Zika [19] and cytomegalovirus [20] infections suggest a broader applicability of maternal immunizations to other TORCH pathogens.

In summary, the efficacy of maternal vaccines against neonatal HSV has been predicted by studies as early as the 1980s. Compared with the long-term protection required of vaccines against genital HSV, maternal vaccines would only need to induce an antibody response during the perinatal period to be effective in the newborn. Whether a vaccination strategy would supersede or synergize with acyclovir treatment remains an open question, but the newly-discovered ability of maternal antibodies to protect the newborn nervous system provides compelling impetus for this approach. Development of maternal immunizations against neonatal HSV is highly warranted for preventing this global disease.