Interest has increasingly been focused on the potential of virus-neutralizing monoclonal antibodies (mAbs) to treat COVID-19 by passive immunization. These antibodies generally target the viral spike protein to prevent infection by blocking ACE2 receptor binding to its receptor binding domain (RBD). There are several issues, however, that need to be addressed to determine when and how they might be used. First, the levels of neutralizing antibodies present in COVID-19 patients do not always correlate to the severity of the disease, and the mean time for seroconversion in SARS patients is known to be about 2 weeks after disease onset. It is possible that timing is critical; i.e., perhaps lack of such antibodies early in infection allow vigorous viral replication, and high levels of antibody production after the development of symptom can be too late to do much good. It may be that the T cell response also plays an important role in the course of infection. Second, although immune serum from recovered patients has been touted as a potential therapy with some successful cases, evidence of its effectiveness is at best contradictory (1).
Another consideration is the nature of the antibodies themselves. This is challenging because protective immunity against SARS-CoV-2 remains unknown. Specifically, the levels and types of antibodies required for the protection need to be defined. Neutralizing antibodies against SARS-CoV-2 appear to have two sets of targets. Some bind the RBD, while others bind the spike protein outside the RBD. Some antibodies bind to the spike protein but do not neutralize the virus (non-neutralizing antibodies). Yet another consideration is that some neutralizing antibodies can lead to antibody-dependent enhancement (we have covered this in an earlier GVN Perspectives) when they are at sub-optimal concentrations. It is obvious that great care must be taken in selecting mAbs for therapeutic development. It also seems intuitive that a mixture of mAbs targeting different epitopes would provide an extra measure of protection. In clinical use, successful treatment will depend critically upon the time of mAbs administration, doses, levels of concentrations, and duration of treatment.
The general approach to identify effective antibodies has been to identify infected patients with high titers of neutralizing antibodies, sort and recover their memory B cells, sequence the heavy and light chain of mRNAs from single sorted cells to characterize the antibody produced by each cell, synthesize their mRNAs with codon optimization for high levels of expression, clone them into expression vectors and express them in transfected cells, and then screen the resultant antibody library. This has become a standardized approach to generate recombinant neutralizing antibodies (2-8). In general, they fall into two categories. The primary group of antibodies, as might be expected, targets the receptor binding domain (RBD) of the spike protein, presumably blocking its binding to the ACE2 receptor. Others bind the N-terminal domain (NTB) of the spike protein. It is not clear how these NTB antibodies neutralize SARS-CoV-2. Presumably, this is by causing allosteric changes in the tertiary or quarternary structure of the spike trimer. Currently identified neutralizing antibodies against SARS-CoV-2 are closely related to germline sequences and do not show signs of hypermutation. This indicates that neutralizing antibodies have been derived with relatively few changes from germline sequences and is a favorable sign for successful vaccine development. It would likely be best for mAb therapeutic use to include antibodies targeting both the RBD and the NTD domains of the spike protein. It should be pointed out that most neutralization assays were done with Vero (monkey kidney) cells, which are not infected by the same pathway as human airway cells(9), although ACE2 binding is required for both pathways.
Importantly, administration of neutralizing mAbs have proven to be protective in several animal models of COVID-19, including rhesus macaques, hamsters, and mice expressing human ACE2(3, 6, 8, 10-12). It is important to note that the mAbs were administered prior to or shortly after challenge. Thus, the results apply to prophylactic or early therapeutic use, but not necessarily to general therapeutic use.
As with other antivirals, it will likely be critical to administer mAbs early in disease, or as a prophylactic. Many patients with serious disease already have high levels of neutralizing antibodies, but these do not appear to ameliorate disease severity, presumably because dysregulated immune responses are driving pathogenesis in these settings. It will also be important to determine what constitutes an effective dose as well as the biologic half-life of any protective effect.
We should mention some recently described “dark horses” for prophylactic passive immunization. These are engineered single chain single domain antibody-like proteins, called nanobodies, that are derived from camelid species (llamas, alpacas). Camelids make normal antibodies, but they also make antibodies comprised of a single heavy chain containing constant and hinge domains linked to a variable region capable of binding antigens. Interestingly, the variable domain alone is also capable of binding antigens. The variable domain is about 1/10th the size of normal antibodies; hence the name nanobody. Because of this, they are far more stable than normal antibodies, allowing them to be lyophilized, heated, and aerosolized. Their small size makes them able to penetrate tissue more readily than do normal antibodies. And they are easy to produce, which makes them scalable.
Nanobodies that are able to bind the SARS-CoV-2 spike protein (both within the RBD and outside the RBD) were identified from a yeast library of synthetic nanobody sequences. Their tiny size allowed them to be trimerized with gly-ser linkers, thus yielding a multivalent molecule with greatly increased potency. They were then subjected to saturation mutagenesis and affinity maturation, yielding at least one nanobody with picomolar neutralization potency and sub-picomolar affinity. Similar nanobodies have been isolated from a llama immunized with stabilized SARS-CoV-2 spike protein(13). Potency was increased by creating a bivalent molecule with two linked nanobodies.
Because of their size and stability, nanobodies are easily produced and stable to aerosolization. This should make it feasible to package them in inhalers. Introduction to airway tissue would be direct to potential sites of infection. They might, thus, serve as an effective prophylactic, since it would be simple to self-administer them daily. If so, they could be a game changer, but as with other potential therapeutic possibilities, the proof will be in the results.
More than 70 antibody therapies are being developed for the treatment of COVID-19. There has been anticipation that monoclonal antibodies may provide short-term protection from SARS-CoV-2 and could serve as important components of the COVID-19 pandemic response until vaccines become available. Two Phase 3 clinical trials are currently underway in the US. In general, mass-produced antibodies are complex to manufacture and are expensive. Therefore, their availability in low- and middle-income countries can be very limited. It will be also important to have a breakthrough to produce cost-effective, large quantities of antibodies.
- L. Li et al., Effect of Convalescent Plasma Therapy on Time to Clinical Improvement in Patients With Severe and Life-threatening COVID-19: A Randomized Clinical Trial. JAMA, (2020).
- P. J. M. Brouwer et al., Potent neutralizing antibodies from COVID-19 patients define multiple targets of vulnerability. Science 369, 643-650 (2020).
- Y. Cao et al., Potent Neutralizing Antibodies against SARS-CoV-2 Identified by High-Throughput Single-Cell Sequencing of Convalescent Patients’ B Cells. Cell 182, 73-84 e16 (2020).
- X. Chi et al., A neutralizing human antibody binds to the N-terminal domain of the Spike protein of SARS-CoV-2. Science 369, 650-655 (2020).
- B. Ju et al., Human neutralizing antibodies elicited by SARS-CoV-2 infection. Nature 584, 115-119 (2020).
- L. Liu et al., Potent neutralizing antibodies against multiple epitopes on SARS-CoV-2 spike. Nature 584, 450-456 (2020).
- Y. Wu et al., A noncompeting pair of human neutralizing antibodies block COVID-19 virus binding to its receptor ACE2. Science 368, 1274-1278 (2020).
- S. J. Zost et al., Potently neutralizing and protective human antibodies against SARS-CoV-2. Nature 584, 443-449 (2020).
- M. Hoffmann et al., Chloroquine does not inhibit infection of human lung cells with SARS-CoV-2. Nature, (2020).
- A. O. Hassan et al., A SARS-CoV-2 Infection Model in Mice Demonstrates Protection by Neutralizing Antibodies. Cell 182, 744-753 e744 (2020).
- J. Kreye et al., A SARS-CoV-2 neutralizing antibody protects from lung pathology in a COVID-19 hamster model. bioRxiv, (2020).
- R. Shi et al., A human neutralizing antibody targets the receptor-binding site of SARS-CoV-2. Nature 584, 120-124 (2020).
- D. Wrapp et al., Structural Basis for Potent Neutralization of Betacoronaviruses by Single-Domain Camelid Antibodies. Cell 181, 1004-1015 e1015 (2020).