Broadly neutralizing antibodies against COVID-19

Following its emergence in late 2019, COVID-19 rapidly established a pandemic, which has caused a public health crisis and economic recession. According to data from WHO, there have been more than 660 million reported cases and more than 6.7 million deaths, as of January 2023. Vaccines have been widely and effectively used to reduce disease severity and a number of drugs have been approved for clinical use, including the small-molecule drugs Paxlovid and Veklury and several monoclonal antibodies (mAbs): bebtelovimab, bamlanivimab, etesevimab, Xevudy (sotrovimab), REGEN-COV (casirivimab and imdevimab), and Evusheld (cilgavimab and tixagevimab) [1][2][3].

Since the first cases were reported in China, Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has mutated rapidly and multiple variants have appeared. From the Wuhan strain to currently dominant Omicron strains, the virus has gained increased transmissibility and immune escape. Many mAbs, including approved therapeutic ones that neutralize earlier variants, have largely or totally lost their ability to neutralize new variants [3,4]. It is therefore essential to develop broadly neutralizing antibodies for the ongoing Omicron variants and, ideally, new variants that will emerge in the future.

The major antigens of SARS-CoV-2 are the nucleoprotein and the trimeric spike glycoprotein, and numerous spike-binding antibodies have been characterized as potent neutralizers [5,6]. Spike consists of the S1 and S2 subunits linked by a furin cleavage site. S1 mediates binding with the receptor angiotensin-converting enzyme 2 (ACE2) and S2 facilitates membrane fusion. The S1 subunit comprises the N-terminal domain (NTD), subdomain 1 (SD1), and receptor-binding domain (RBD). The RBD harbors the ACE2-binding site and adopts up and down conformations, with only the up conformation engaging ACE2. Previously, a naming convention described RBD epitopes for neutralizing mAbs: left shoulder, neck, right shoulder, left flank, and right flank. Most potent neutralizing mAbs target the RBD and interfere with ACE2 binding. In line with this, many mutations in variants occur on or near the ACE2-binding site, allowing escape without significantly impairing ACE2 binding. In just over three years, variants account for changes at more than 30 of the ~200 RBD residues. Neutralizing antibodies against an NTD supersite are generally narrow due to poor conservation. Less common neutralizing antibodies target SD1, the stem-helix region, and the fusion peptide. Here, we mainly review broadly neutralizing mAbs directed against highly conserved epitopes in portions of the RBD and SD1 on S1 and the stem-helix region and the fusion peptide in S2.

Receptor-binding domain (RBD): Potent neutralizing mAbs typically overlap the ACE2 footprint at the neck and shoulders of the RBD, blocking receptor engagement. Many such mAbs, including commercial therapeutics (e.g., LY-CoV555/bamlanivimab, REGN10987/imdevimab), lost efficacy due to variant substitutions (e.g., E484K in Beta/Gamma; N440K, G446S in Omicron BA.1; Q493R, Y505H in BA.1 impacting Beta-27). BA.1 introduced 15 RBD substitutions, causing widespread escape; BA.2 and subsequent sublineages added convergent mutations near the ACE2 footprint (e.g., R346X, K444X, V445X, N450D, N460X) further reducing mAb potency. A small set of mAbs retain broad activity across Alpha, Beta, Gamma, Delta, and Omicron sublineages: left-shoulder back binders (F61, Omi-42) and right-shoulder binders (LY-CoV1404/bebtelovimab, 2-7, XGv265, XG005, AZD1061, MB.02, SP1-77, 002-S21F2). However, several of these are vulnerable to newer sublineage mutations (e.g., LY-CoV1404 impaired by V445P in BJ.1/XBB; AZD1061 knocked out by R346T in BA.2.75.2). Omi-42, isolated from a BA.1 infection, binds the back of the left shoulder with substantial overlap with ACE2, maintaining potency despite K417N, S477N, Q493R, Y505H; a derivative (AZD3152, combined with cilgavimab as AZD5156) is in clinical trials.

Beyond ACE2 blockers, weaker but broadly neutralizing anti-RBD mAbs target conserved regions: the left flank (e.g., S2H97, COVOX-45) and a cryptic epitope on the back of the RBD below the Omi-42 site (e.g., S304, IY-2A, EY6A). S2H97 shows pan-sarbecovirus binding, neutralizes BA.1, and induces receptor-independent conversion of spike to the postfusion state. S309 (sotrovimab) targets the right flank and was broadly neutralizing against early variants but is impaired against BA.2 (e.g., S371F). Cryptic epitope binders disrupt spike trimer to neutralize; epitope conservation persists across variants, but accessibility can depend on spike conformation: S304 loses activity against Omicron sublineages due to one-RBD-up packing that blocks binding; EY6A potency is reduced against BA.2/BA.5, while IY-2A remains active.

N-terminal domain (NTD): NTD-directed neutralizing mAbs predominantly target a single supersite (loops N1, N3, N5). The NTD is extensively glycan-shielded and variable (including insertions/deletions), with BA.5 bearing multiple changes, many within the supersite. Consequently, NTD-specific mAbs generally lack breadth.

Subdomain 1 (SD1): SD1 is highly conserved (reported mutations T547K, A570D are buried). Neutralizing mAbs (S3H3, P008_60, sd1.040) recognize distinct SD1 epitopes, sometimes bridging to RBD termini, and can neutralize multiple variants. P008_60 and sd1.040 appear to destabilize the spike trimer, likely recognizing transient conformations; S3H3 may block S1 release. These mAbs tend to be weaker than potent RBD ACE2 blockers.

Stem-helix region (S2): The stem-helix (residues 1144-1158) is highly conserved among betacoronaviruses. mAbs including CV3-25, S2P6, B6, CC40.8, 1.6C7, 28D9 inhibit membrane fusion but target different faces/segments of the helix. Structural studies show many bind the hydrophobic face, buried in prefusion spike, implying binding induces conformational changes; CV3-25 binds the hydrophilic face but to a conformation distinct from native prefusion spike. Broad reactivity is observed, but potency is generally weak.

Fusion peptide (S2): The fusion peptide (residues 816-837) is highly conserved (>99.7% identity across 7.9 million sequences) and conserved across coronaviruses. mAbs (COV44-62, COV44-79, VN01H1, C77G12, 76E1) broadly neutralize alpha- and betacoronaviruses and reduce disease/viral load in animal models. Structural data indicate the epitope is inaccessible in prefusion trimer and becomes exposed upon ACE2-induced conformational changes. Potency is weaker than anti-RBD antibodies.

• Most potent anti-RBD therapeutic mAbs developed early in the pandemic have been rendered ineffective by mutations in circulating variants, especially Omicron sublineages; BA.1 introduced 15 RBD substitutions and subsequent BA.2-derived lineages accumulated convergent changes around the ACE2 footprint (e.g., R346X, K444X, V445X, N450D, N460X).
• A minority of RBD-directed mAbs retain broad potency: Omi-42 (back of left shoulder, ACE2-overlapping) remains highly active against all variants to date; right-shoulder mAbs like LY-CoV1404 (bebtelovimab) retained activity against many variants but are impaired by specific mutations (e.g., V445P in BJ.1/XBB); AZD1061 is knocked out by R346T in BA.2.75.2.
• Conserved non-ACE2-overlapping RBD epitopes (left flank and cryptic back epitope) yield broadly reactive but generally weaker neutralizers (e.g., S2H97, EY6A, S304, IY-2A); epitope accessibility can be limited by spike conformational states (S304 loses activity against Omicron due to one-RBD-up packing).
• NTD-directed mAbs mainly target a variable supersite and lack breadth due to extensive variation and glycan shielding.
• SD1-specific mAbs (S3H3, P008_60, sd1.040) target conserved regions and neutralize via spike destabilization or blocking S1 release but are comparatively weak.
• Stem-helix (S2) mAbs (CV3-25, S2P6, B6, CC40.8, 1.6C7, 28D9) and fusion peptide mAbs (COV44-62, COV44-79, VN01H1, C77G12, 76E1) are broadly reactive across coronaviruses and reduce disease in animal models; their epitopes are conserved but typically buried in prefusion spike, resulting in weaker neutralization potency.
• Therapeutic strategies suggested include cocktails of mAbs targeting distinct conserved epitopes and bispecific antibodies to enhance breadth and reduce escape risk; AZD3152 (derived from Omi-42) combined with cilgavimab (AZD5156) is in clinical trials for pre-exposure prophylaxis in immunocompromised patients.

The review addresses how SARS-CoV-2 antigenic evolution undermines most potent anti-RBD mAbs while highlighting conserved spike epitopes that enable broader neutralization. Antibodies overlapping ACE2 at the RBD neck/shoulders achieve strong neutralization but are vulnerable to mutations selected by immune pressure. In contrast, antibodies targeting conserved regions (RBD left flank and cryptic back site, SD1, stem helix, fusion peptide) maintain breadth across variants due to functional and structural constraints on these epitopes; however, these regions are often recessed or conformationally masked, limiting potency. The standout exception, Omi-42, retains potency across all tested variants because it targets a conserved ACE2-overlapping site where escape would compromise receptor binding.

These insights inform antibody and vaccine design: selecting epitopes constrained by viral fitness may maximize breadth; combining antibodies with complementary epitopes can mitigate escape; and engineering bispecifics that engage multiple conserved sites on a single spike can enhance binding avidity and neutralization. Structural understanding of epitope accessibility and spike conformational dynamics is critical to anticipate and counter future variant evolution.

SARS-CoV-2 has accrued extensive mutations, especially around the RBD ACE2-binding site, driving escape from most early therapeutic mAbs. Very few anti-RBD antibodies, notably Omi-42, retain potent neutralization across current variants, likely because their epitopes overlap essential receptor-binding residues. Outside the ACE2-binding site, multiple conserved epitopes exist in the RBD left flank and cryptic back site, SD1, the stem helix, and the fusion peptide, supporting broad but generally weaker neutralization. To enhance clinical effectiveness and minimize escape, cocktails of broadly neutralizing mAbs targeting distinct conserved epitopes or bispecific formats are advocated. Continued surveillance of variant mutations, structural mapping of conserved vulnerable sites, and rational antibody/vaccine design are needed to sustain efficacy against ongoing and future SARS-CoV-2 evolution.

• Broadly neutralizing mAbs outside the ACE2-binding region tend to have weak potency and may be ineffective as monotherapies; their epitopes can be transiently accessible or conformationally masked (e.g., S304 loses activity against Omicron due to tighter one-RBD-up packing).
• Even ACE2-overlapping broad mAbs like Omi-42 may eventually lose potency as the virus continues to evolve, especially via convergent mutations that abrogate families of antibodies without severely compromising receptor binding.
• Evidence for breadth and efficacy of many non-RBD mAbs relies on in vitro assays and animal models; translation to human clinical efficacy remains to be fully established.
• As a narrative review, the work synthesizes published findings and does not present new experimental methodology or systematic meta-analysis, which may limit quantitative comparison across studies.