RSV vaccine: Beating the virus at its own game

THE NATURE OF THE BEAST

The genesis of severe RSV disease is multifaceted and has several simultaneous causes: The contributions of the virus itself, the underlying genetic predisposition of the host, and the components of the inflammatory response form a complex picture that is far from fully understood. RSV replicates initially in the nasopharyngeal epithelium and later spreads to the lower respiratory tract. Viral replication may lead to plugging of the small airways, atelectasis, and airway narrowing and obstruction. This lower respiratory tract disease is the combined result of both cellular damage mediated by the virus and injury caused by the immune response of the host.

This immune-mediated injury has been a major roadblock to the development of an RSV vaccine. In the 1960s, a formalin-inactivated RSV vaccine (FI-RSV) was tested in infants and children. RSV-naïve children developed worse RSV disease upon subsequent natural exposure to virus (3), and the intense inflammatory infiltrate in the lungs of the vaccinated children suggested an immunopathological cause of enhanced disease.

The devastating clinical experience of the 1960s and the need to understand the mechanism of disease enhancement spurred fundamental research over the past two decades (4) that has highlighted several characteristics that a successful vaccine for RSV-naïve infants would require. A successful vaccine would need to induce both adequate levels of neutralizing antibody and CD8⁺ virus-specific cytotoxic T lymphocytes (CTLs), as well as a CD4⁺ T cell response to correspond to the response to natural infection with RSV. Secretory and serum antibodies have been shown to protect against infection of the upper and lower respiratory tract, respectively, whereas cell-mediated responses directed against internal viral proteins are important for clearing infection.

Recently, there has been a renewal of interest in RSV vaccine development founded on the past few decades’ research, including the use of replicating and nonreplicating approaches to generate RSV vaccine candidates. Many RSV candidate vaccines are currently being explored, of which 12 are in clinical development as of this writing. Nonreplicating candidate vaccines include RSV subunit vaccines composed of purified surface glycoprotein, most commonly the fusion protein (F), which elicits cross-reactive neutralizing antibodies. Alternately, vectored vaccines with replication-competent or deficient viral vectors are being used to deliver one or more RSV glycoprotein genes (5, 6).

For protection of RSV-naïve infants and young children, live attenuated RSV vaccines have several advantages (Fig. 1). Intranasal immunization with a live virus should induce local immunity in the respiratory tract as well as systemic immunity, and live virus contains the full complement of viral proteins, including all of the humoral and cellular response epitopes. Moreover, live attenuated virus vaccines have been successful for other paramyxoviruses—measles and mumps. The immune response to attenuated RSV should most closely resemble the response to natural infection and thus would be less likely to induce enhanced disease upon exposure to RSV in the field; indeed, live attenuated RSV vaccine candidates have never been associated with enhanced disease (7).

• Download high-res image
• Open in new tab
• Download Powerpoint

In spite of these advantages, the development of live RSV vaccines has been hampered by difficulty in achieving the right balance between viral attenuation (dampening of viral replication) and induction of immunity (8). Recently, deletion (Δ) of nonessential genes (SH, NS1, NS2, or M2-2) in combination with known attenuating cp and ts mutations yielded two candidates, designated rA2cp248/404ΔSH and rA2cp248/404/1030ΔSH, that were evaluated in young children. Although these candidates were well tolerated, they induced only modest levels of antibody, highlighting the difficulty of striking the right balance between viral attenuation and the immune response, which would presumably correlate with efficacy.

A TRICKY PLAY

Now, Karron et al. report a live attenuated RSV vaccine (RSV MEDI ΔM2-2), developed through rational design and expanded knowledge of RSV gene function, that appears to offer a game-changing solution to the problem of attenuation. The new candidate is an engineered virus with a deletion in the open reading frame of the RNA synthesis factor M2-2. The M2-2 protein mediates the regulatory switch from transcription to RNA replication (9). Deletion of this gene leads to a virus that is shifted toward transcription at the expense of replication. Infected cells accumulate viral mRNA and serve as viral protein factories, but relatively few full-length genomes are produced and, therefore, little progeny virus.

Preclinical evaluation of this intriguing vaccine candidate developed by Collins and Bermingham showed promising attenuation and immunogenicity (10). The current study evaluates the virus sequentially in a phase I clinical trial in adults, RSV-seropositive children, and, most importantly, RSV-seronegative children (presumably RSV-naïve), who would represent the target population for immunization. The virus is indeed highly restricted in replication yet more immunogenic than rA2 cp248/404/1030/ΔSH, the previous lead live attenuated RSV vaccine candidate. There was no enhancement of disease in vaccinees, as expected. The ΔM2-2 virus also primed for a potent anamnestic response after natural exposure to wild-type RSV.

The M2-2 defect is highly stable during replication in vitro and appears to be so during growth in humans as well. The increase in synthesis of viral proteins includes the viral neutralization and protective antigens, F and G (the receptor binding protein of RSV), likely to induce a neutralizing antibody response. In RSV-seronegative children, the vaccine was more restricted in replication and induced higher levels of antibody than that of the previous lead candidate.

Six children followed during surveillance demonstrated extraordinarily high titers of neutralizing antibodies, with RSV illness only documented in one child; this illness was caused not by natural infection with the RSV A strain vaccine contained in the vaccine but by RSV B. Placebo recipients who experienced their first infection had substantially lower neutralizing titers. Although the numbers of patients is certainly very small, it is tempting to interpret these data to mean that vaccine primed for an anamnestic response—as expected for a live vaccine—and may have afforded some protection against illness induced by RSV A in children harboring the virus. Larger studies will sort out this important question, as well as address whether the increased rhinorrhea in vaccinated children is real and, if so, clinically important. When weighed against the risk of RSV lower respiratory tract disease, mild rhinorrhea may be an acceptable side effect.

One limitation of this study is that cellular immune responses have not been assessed yet in the young subjects of this study. The authors speculate that the increased expression of the entire set of viral antigens may also lead to productive RSV-specific T cell responses in the vaccinees. Considering the importance of these cellular responses for clearance of infection, the presence and nature of cellular responses, whether beneficial or harmful, will be key questions to be addressed for this vaccine candidate and its derivatives over the coming years.

PERSPECTIVE

Live attenuated vaccines will not be suitable for all people at risk for RSV. A panel of vaccine strategies, along with a complementary set of antiviral drugs for use in RSV prevention and treatment, will ultimately be needed. Maternal immunization may use nonreplicating vaccines to protect the very youngest infants, whereas live attenuated vaccines will protect the slightly older infants and young children. The link between viral replication and induction of an immune response by RSV vaccines has long stymied efforts at vaccination. By hijacking the virus’s transcriptional machinery and forcing the virus to delay its own replication while pumping out its proteins, this candidate vaccine reported by Karron et al. “delinks” viral replication from the desired immune response for the first time. It is tantalizing when a mechanism identified through beautiful basic virology research appears to be borne out in humans. Further trials in larger cohorts will be required to know whether this vaccine candidate can deliver on its promise.