Passive immunization with polyclonal SHIVIG given to RMs that were subsequently exposed repeatedly to low doses of the pathogenic R5 SHIV-2873Nip yielded partial protection at 400 mg/kg as evidenced by statistically significant lower peak viral RNA loads compared with control RMs. The SGA data suggest the possibility of increased viral acquisition at the low SHIVIG dose of 25 mg/kg. When tested ex vivo in the presence of fresh complement, low SHIVIG concentrations showed significant C’-ADE when tested with viruses carrying envelopes related to the challenge virus or other HIV-Cs. These results suggest the possibility that the polyclonal SHIVIG contained Abs with the ability to either partially protect or to facilitate virus acquisition.
These surprising data were generated in a primate model that sought to replicate as many aspects of HIV-1 transmission among humans as possible. Specifically, we chose to perform an upfront heterologous challenge, in contrast to earlier studies that matched the IgG preparation with the challenge virus Env [9, 10, 19, 20]. We also elected to perform multiple low-dose challenges instead of using a single high-dose of virus. The low-dose challenge regimen resulted in a low number of transmitted viral variants in our control animals, a situation similar to that observed in humans. Lastly, the exclusive R5 tropism of our challenge virus reflects that of acutely transmitted founder viruses isolated from humans.
Ab-mediated enhancement of HIV-1 acquisition has been implicated in a recently published subgroup analysis of the Vax004 AIDS vaccine efficacy trial that involved gp120 immunogens . This study linked Fcγ receptor IIIa (FcγRIIIa) genotype with a significantly increased HIV-1 acquisition rate for vaccinees with low behavioral risk of infection and homozygosity for the FcγRIIIa V allele. Other evidence for in vivo increased viral acquisition following immunization stems from experiments involving SIV , feline immunodeficiency virus [23–25], and equine infectious anemia virus . Enhancement of viral acquisition and/or higher viral loads or earlier viremia by pre-existing Abs are well-known phenomena for dengue virus [27, 28], Murray Valley encephalitis virus , respiratory syncytial virus , Ebola virus  and measles virus .
Another recent report described a passive immunization study conducted in HIV-infected pregnant women and their infants in Uganda with the aim of testing anti-HIV-1 Ig (termed HIVIGLOB) for its ability to lower the risk of mother-to-infant virus transmission . Two groups of pregnant women were enrolled; all mothers and their infants were treated with single-dose nevirapine (NVP) according to standard local protocol. Half of the women and their children also received HIVIGLOB (maternal dose, 200 mg/kg at week 36–38 of gestation). The infants received HIVIGLOB at 400 mg/kg within 18 h of delivery. At birth, 9.1% of infants born to HIVIGLOB-treated mothers were HIV positive compared with 4.1% of control infants; the difference was statistically significant. The higher HIV-1 infection rate of HIVIGLOB-treated infants persisted throughout six months of follow-up, although differences at later time points were not statistically significant. In essence, passive immunization with HIVIGLOB did not prevent HIV-1 acquisition in any infants born to infected mothers, and may have enhanced in utero HIV-1 transmission.
Ab-mediated increased viral acquisition was also suggested in a recent study of Burton et al. , who performed passive immunization with human monoclonal Abs (mAbs), including the anti-CD4 binding site mAb b6, which is weakly neutralizing. The latter provided no protection from virus acquisition when tested against intravaginal challenge with SHIVSF162P4 (tier 1) or SHIVSF162P3 (tier 2). SGA revealed a significantly higher number of newly transmitted quasispecies among the b6-treated monkeys compared with the control groups, which is compatible with increased virus acquisition.
Ab-mediated enhancement of lentiviral infection can occur through different mechanisms. The VAX004 data implied that increased risks of HIV-1 acquisition occurred through FcγR-mediated enhancement. Such a mechanism was first described by Takeda and Ennis . Their studies involving cultured monocytes/macrophages demonstrated IgG-linked enhancement of infection only in the presence of surface-expressed Fc-receptors (FcRs) and the IgG constant region. This enhancement still required virus entry through CD4, implying that FcR-bearing cells may enhance infectivity in trans. The second major mechanism was first described by Robinson et al., who demonstrated the critical role of complement in C’-ADE . This activity was found to be highly prevalent in individuals with acute HIV-1 infection who had developed binding Abs but no autologous nAbs yet; strikingly, enhancement of infection reached levels up to 350-fold and was not only seen with autologous virus, but also with different virus isolates .
Prompted by our observation of an increased number of quasispecies seen in RMs given the low SHIVIG dose, we examined this preparation for C’-ADE activity in vitro. Significant enhancement of infection was seen at low SHIVIG concentrations, whereas higher concentrations of SHIVIG showed some virus-inhibitory activity, thus providing a potential mechanism for the increased number of quasispecies seen in the RMs treated with the low-dose SHIVIG. The complex interactions of HIV-1 with complement, including enhancement of infection, have been reviewed . In this context, it is worth mentioning the recent suboptimal outcome of the phase IIb HVTN 505 trial . The multicomponent immunogens contained envelope and possibly may have induced low levels of anti-Env Abs.
One of our initial goals for the passive immunization using the tier 2 SHIV-2873Nip had been to compare in vivo protection with in vitro neutralization titers of serum samples collected at the time of virus exposure. We now realize that the role of Abs directed against HIV-1/SHIV is more complex in vivo and that the currently used neutralization assays have a narrow focus on prevention of virus entry. A more complex interaction of Abs and virus with primary cells is captured to a certain degree in PBMC-based assays but only if the Ab-virus mixture is left with the cells for several days, a protocol we have followed when assessing SHIVIG in vitro. In contrast, most neutralization assays include a washing step, in which the virus-Ab mixture is removed, thus not allowing ADCC activity to impact the final readout. Currently, routine neutralization assays do not query the influence of complement, and their lack of predictive value may be due to their oversimplified readout and lack of probing mechanisms that may influence the outcome in vivo.
To summarize, passive immunization with SHIVIG yielded partial protection at higher doses and may have increased viral acquisition at the low dose – a perplexing finding. Future studies are required to address the following questions: Can the Ab-mediated infection-enhancing activity be separated from protective functions, such as neutralization, ADCVI and ADCC? More important still, can immunogens be designed that will elicit protective but not infection-enhancing Abs? Will it be possible to induce durable nAb responses at sufficiently high levels to counteract any potential Ab-mediated enhancement of infection?