Bamlanivimab (Page 9 of 12)

15 MICROBIOLOGY/RESISTANCE INFORMATION

Antiviral Activity

The cell culture neutralization activity of bamlanivimab and of etesevimab against SARS-CoV-2 was measured in a dose-response model quantifying plaque reduction using cultured Vero E6 cells. Bamlanivimab, etesevimab and a 1:1 (weight/weight) ratio of bamlanivimab and etesevimab together neutralized the USA/WA/1/2020 isolate of SARS-CoV-2 with estimated EC₅₀ values = 0.14 nM (0.02 μg/mL), 0.97 nM (0.14 μg/mL) and 0.14 nM (0.02 μg/mL), respectively.

Bamlanivimab demonstrated antibody-dependent cell-mediated cytotoxicity on reporter Jurkat cells expressing FcγRIIIa following engagement with target cells expressing spike protein. Bamlanivimab did not elicit complement-dependent cytotoxicity activity in cell-based assays.

Etesevimab did not demonstrate detectable antibody-dependent cell-mediated cytotoxicity on Jurkat reporter cells expressing FcγRIIIa. Etesevimab did not elicit complement-dependent cytotoxicity activity in cell-based assays.

Antibody Dependent Enhancement (ADE) of Infection

The risk that bamlanivimab and etesevimab could mediate viral uptake and replication by immune cells was studied in THP-1 and Raji cell lines and primary human macrophages. In general, experiments with bamlanivimab, with etesevimab, and with bamlanivimab and etesevimab together did not demonstrate productive viral infection in immune cells exposed to SARS-CoV-2 at concentrations of mAb(s) down to at least 100-fold below the respective EC₅₀ value(s).

Antiviral Resistance

There is a potential risk of treatment failure due to the development of viral variants that are resistant to bamlanivimab and/or etesevimab (Table 5). There are other authorized treatments available and healthcare providers should choose an authorized therapeutic option with activity against circulating variants in their state, territory, or US jurisdiction. Variant frequency data for states, territories, and US jurisdictions can be accessed on the following CDC website: https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/variant-proportions.html.

Resistant variants were identified using directed evolution of the spike protein and serial passage in cell culture of SARS-CoV-2 in the presence of bamlanivimab or etesevimab individually. Resistant variants were not identified when bamlanivimab and etesevimab were tested together using the same methodology. Viral variants identified in these studies that had reduced susceptibility to bamlanivimab included spike protein amino acid substitutions E484D/K/Q, F490S, Q493R, and S494P, and variants that had reduced susceptibility to etesevimab included substitutions K417N, D420N, and N460K/S/T/Y. Neutralization assays using SARS-CoV-2 and vesicular stomatitis virus (VSV) virus-like particles (VLP) pseudotyped with variant SARS-CoV-2 spike protein confirmed reductions in susceptibility to the selecting antibody. Retention of susceptibility to the other antibody alone was observed, with the exception of the E484D and Q493R substitution. All variants maintained susceptibility to bamlanivimab and etesevimab together, with the exception of those with E484D, E484K, E484Q, and Q493R substitutions, which had reduced susceptibility of 145-fold, 24-fold, 17-fold, and 1,054-fold, respectively in a pseudotyped VLP assay.

Evaluation of susceptibility of variants identified through global surveillance in subjects treated with bamlanivimab and etesevimab is ongoing. Pseudotyped VLP evaluation of amino acid substitutions identified in global surveillance showed that the V483A substitution reduced susceptibility to bamlanivimab 48-fold, but activity was maintained with etesevimab, and with bamlanivimab and etesevimab together. N501Y and N501T substitutions reduced susceptibility to etesevimab approximately 5-fold and 20-fold, respectively. Activity against variants with N501Y or N501T substitutions was maintained with bamlanivimab alone, and with bamlanivimab and etesevimab together.

Bamlanivimab and etesevimab together retained activity against a SARS-CoV-2 B.1.1.7 lineage (Alpha; UK origin) virus and related pseudotyped VLPs expressing the spike protein found in the B.1.1.7 variant (Tables 5 and 6). SARS-CoV-2 B.1.351 lineage (Beta; South Africa origin) virus and related pseudotyped VLPs expressing spike proteins from B.1.351 lineage or substitutions K417N + E484K + N501Y found in this lineage had reduced susceptibility to bamlanivimab and etesevimab together of >324, 431-fold or >45-fold, respectively. Pseudotyped VLPs expressing spike protein from the P.1 lineage (Gamma; Brazil origin) or K417T + E484K + N501Y found in the P.1 lineage had reduced susceptibility to bamlanivimab and etesevimab together of 252-fold or >3,351-fold, respectively.

Bamlanivimab and etesevimab together and etesevimab alone retained activity against SARS-CoV-2 B.1.617.2 lineage (Delta; India origin) virus and related pseudotyped VLPs, but bamlanivimab alone had reduced activity (>1,136 and >1,868-fold, respectively). Bamlanivimab and etesevimab are expected to retain activity against B.1.617.2 sublineage AY.3 (India origin). B.1.617.2 sublineages AY.1/AY.2 (India origin) have an additional K417N substitution; pseudotyped VLPs expressing AY.1/AY.2 related spike sequence had a reduced susceptibility to bamlanivimab and etesevimab together of 1,235-fold. SARS-CoV-2 recombinant virus containing the L452R substitution present in B.1.427/B.1.429 lineages (Epsilon; USA [California] origin) and pseudotyped VLPs expressing the full-length spike protein or the L452R substitution found in this lineage showed reduced susceptibility to bamlanivimab and etesevimab together of 11-fold, 9-fold or 5-fold, respectively. Pseudotyped VLPs expressing spike protein from the B.1.617.1 lineage (Kappa; India origin) showed reduced susceptibility to bamlanivimab and etesevimab together of 6-fold; for this variant, susceptibility to etesevimab alone was maintained, but not to bamlanivimab alone (>1,030-fold reduction). Bamlanivimab and etesevimab together and etesevimab alone retained activity against pseudotyped VLPs expressing the full-length spike protein from the C.37 lineage (Lambda; Peru origin), but bamlanivimab alone had reduced activity (>2,112-fold reduction). Pseudotyped VLPs expressing spike protein from the B.1.621 lineage (Mu; Colombia origin) show reduced susceptibility to bamlanivimab and etesevimab together of 116-fold, due to susceptibility reductions to bamlanivimab (>1,863-fold) and etesevimab (17-fold) alone. Pseudotyped VLPs expressing the spike protein from the B.1.1.529/BA.1 lineage (Omicron; South Africa origin) show reduced susceptibility to bamlanivimab alone (>1,465-fold), etesevimab alone (>616-fold), and bamlanivimab and etesevimab together (>2,938-fold).

Table 5: Pseudotyped Virus-Like Particle Neutralization Data for SARS-CoV-2 Variant Substitutions with Bamlanivimab and Etesevimab Together (1:2 Molar Ratio)

a Key substitutions occurring in the receptor binding domain of spike protein are listed. Pseudoviruses containing the full-length spike protein reflective of the consensus sequence for each of the variant lineages were tested.
b No change: <5-fold reduction in susceptibility.
c Bamlanivimab and etesevimab together are unlikely to be active against variants from this lineage.
d Etesevimab retains activity against this variant.
e Isolates of the B.1.526 lineage harbor several spike protein amino acid substitutions, and not all isolates contain the E484K substitution (as of February 2021).
Lineage with Spike Protein Substitution	Country First Identified	WHO Nomenclature	Key Substitutions Tested a	Fold Reduction in Susceptibility
B.1.1.7	UK	Alpha	N501Y	no changeb
B.1.351	South Africa	Beta	K417N + E484K + N501Y	431c
P.1	Brazil	Gamma	K417T + E484K + N501Y	252c
B.1.617.2/AY.3	India	Delta	L452R + T478K	no changeb
AY.1/AY.2(B.1.617.2 sublineages)	India	Delta [+K417N]	L452R + T478K + K417N	1,235c
B.1.427/B.1.429	USA (California)	Epsilon	L452R	9d
B.1.526e	USA (New York)	Iota	E484K	30
B.1.617.1	India	Kappa	L452R + E484Q	6d
C.37	Peru	Lambda	L452Q + F490S	no changeb
B.1.621	Colombia	Mu	R346K + E484K + N501Y	116c
B.1.1.529/BA.1	South Africa	Omicron	G339D + S371L + S373P + S375F + K417N + N440K + G446S + S477N + T478K + E484A + Q493R + G493S + Q498R + N501Y + Y505H	>2,938c

Table 6: Authentic^a SARS-CoV-2 Neutralization Data for Bamlanivimab and Etesevimab Together (1:2 Molar Ratio)

a The B.1.1.7 variant was assessed using cell culture-expanded virus isolates and tested using an immunofluorescence based microneutralization assay and by plaque reduction assay; B.1.351 and B.1.617.2 variants were assessed using cell culture-expanded virus isolates and tested using a plaque reduction assay; the B.1.526/E484K and B.1.427/B.1.429/L452R substitutions were assessed using recombinant SARS-CoV-2 (USA/WA/1/2020 isolate with E484K or L452R) and tested using a plaque reduction assay.
b Key substitutions occurring in receptor binding domain of spike protein which are associated with each lineage.
c No change: <5-fold reduction in susceptibility.
d Isolates of the B.1.526 lineage harbor several spike protein amino acid substitutions, and not all isolates contain the E484K substitution (as of February 2021). This assay was conducted using recombinant SARS-CoV-2 with the E484K substitution only.
Lineage with Spike Protein Substitution	Country First Identified	WHO Nomenclature	Key Substitutions Tested b	Fold Reduction in Susceptibility
B.1.1.7	UK	Alpha	N501Y	no changec
B.1.351	South Africa	Beta	K417N, E484K, N501Y	>324
B.1.617.2/AY.3	India	Delta	L452R, T478K	no changec
B.1.427/B.1.429	USA (California)	Epsilon	L452R	11
B.1.526d	USA (New York)	Iota	E484K	11

Due to the large reduction of pseudotyped VLP neutralization activity of both bamlanivimab and etesevimab against the substitutions in B.1.351 (Beta; South Africa origin), P.1 (Gamma; Brazil origin), AY.1/AY.2 (Delta [+K417N]; India origin), B.1.621 (Mu; Colombia origin), and B.1.1.529/BA.1 (Omicron; South Africa origin), it is unlikely that bamlanivimab and etesevimab together will be active against these variants.

It is unclear how small reductions in susceptibility to bamlanivimab and etesevimab seen in authentic or recombinant SARS-CoV-2 or pseudotyped VLP assays correlate with clinical outcomes.

In authentic SARS-CoV-2 assays, bamlanivimab and etesevimab together retained activity against variants of B.1.1.7 (Alpha) and B.1.617.2/AY.3 (Delta) lineages (Table 6), although bamlanivimab alone had reduced activity to B.1.617.2/AY.3 (Delta) in this assay (>1,136-fold). SARS-CoV-2 (USA/WA/1/2020 isolate) engineered to express the E484K substitution present in the B.1.526 lineage (Iota; USA [New York] origin) or the L452R substitution present in the B.1.427/B.1.429 lineage (Epsilon; USA [California] origin) showed reduced susceptibility to bamlanivimab and etesevimab together of 11-fold. Susceptibility to etesevimab alone was maintained for both isolates, but not to bamlanivimab alone (>833-fold and >1,460-fold reduction for E484K and L452R viruses, respectively). Available nonclinical and clinical PK data indicate that etesevimab at the authorized dose may retain activity against the B.1.526 variant clinically, although only very limited data are currently available from patients infected with this variant in clinical trials. Preliminary clinical evidence indicates that the administration of bamlanivimab and etesevimab together result in similar viral load reductions in participants infected with the L452R variant (Epsilon; USA [California] origin) as observed in those who were infected with bamlanivimab-sensitive strains. Of the 134 participants infected with the L452R variant at baseline in the Phase 3 portion of BLAZE-1, 3 of the 50 individuals treated with placebo (6%) and 1 of the 84 participants treated with bamlanivimab 700 mg and etesevimab 1,400 mg (1%) were hospitalized (p=0.15).

Genotypic and phenotypic testing are ongoing to monitor for potential bamlanivimab- and etesevimab-resistance associated spike variations in clinical trials. Analysis of baseline samples show that 8.4% (188/2246) of clinical trial patients were infected with viral variants containing single amino acid substitutions at positions associated with reduced susceptibility to either bamlanivimab or etesevimab as predicted by pseudotyped VLP or authentic SARS-CoV-2 neutralization assays. No patients were infected with a variant that was predicted to have reduced susceptibility to both bamlanivimab and etesevimab by these assessments.

Patient samples were also analyzed for treatment-emergent viral variants, defined as variants with single amino acid substitutions at positions that had reduced susceptibility to either bamlanivimab or etesevimab present at an allele fraction of ≥15%.

• In the Phase 3 portion of BLAZE-1, treatment-emergent variants were observed in 9.0% (42/467) of patients treated with bamlanivimab 2,800 mg and etesevimab 2,800 mg together, in 5.3% (21/394) of patients treated with bamlanivimab 700 mg and etesevimab 1,400 mg together, and in 4.0% (27/674) of patients treated with placebo. The majority of these were only detected at one time point in the sequential series with 0.9% (4/467), 1.0% (4/394), and 0.3% (2/674) of patients having multiple instances of detection in the bamlanivimab 2,800 mg and etesevimab 2,800 mg together, bamlanivimab 700 mg and etesevimab 1,400 mg together, and placebo groups, respectively.
• In patients treated with bamlanivimab and etesevimab together, substitutions detected in one or more patients included ones with reduced susceptibility (≥5-fold) to bamlanivimab only: L452R/W, E484K, G485V, F490L, and S494P; and ones with reduced susceptibility to etesevimab only: D405G/Y, K417N, D420N/Y, N460H/I/T, A475S/V, Y489H, and N501I/Y. While these variants had reduced susceptibility to either bamlanivimab OR etesevimab compared to wild-type in a pseudotyped VSV VLP or authentic virus assay they still retained susceptibility to the other antibody in the combination.
• There were also observations of variants with reduced susceptibility (≥5-fold) to both bamlanivimab and etesevimab and to bamlanivimab + etesevimab tested together: E484D (n=1; 145-fold reduction to bamlanivimab + etesevimab tested together at a molar ratio of 1:2), Q493K/R (n=9; 584-fold and 1,054-fold reduction to bamlanivimab + etesevimab tested together at a molar ratio of 1:2 for Q493K and Q493R, respectively) out of a total of 861 patients treated with bamlanivimab and etesevimab together.
• In a subgroup of participants infected with virus harboring L452R substitution found in the B.1.427/B.1.429 (Epsilon) lineage, a S459P treatment-emergent substitution was identified in one subject. Concurrent L452R+S459P substitutions conferred a 1,656-fold reduction in susceptibility to bamlanivimab + etesevimab together (1:2 molar ratio).
• Additional treatment-emergent substitutions in patients treated with bamlanivimab and etesevimab together, with no phenotypic data, include D405del, D420G, C480R, G485D, S494L, and P499L. The impact of these substitutions on susceptibility is not currently known.
• In a subgroup of 53 pediatric subjects who were infected with a B.1.617.2 (Delta)-related variant, which has reduced susceptibility to bamlanivimab (>1,136-fold), the following treatment-emergent substitutions with reduced susceptibility to etesevimab were detected: D420A (n=2), N460T (n=1), N460Y (n=1). Three of these four subjects had high viral load (>5.27 log10) on Day 7.
• Additional treatment-emergent substitutions with no phenotypic data detected in other pediatric subjects who were infected with a B.1.617.2 (Delta)-related variant at an allele fraction of ≥50% included: F347C, V401L, G431S and I434V.

It is possible that bamlanivimab and etesevimab resistance-associated variants could have cross-resistance to other mAbs targeting the receptor binding domain of SARS-CoV-2. The clinical impact is not known.

Immune Response Attenuation

There is a theoretical risk that antibody administration may attenuate the endogenous immune response to SARS-CoV-2 and make patients more susceptible to re-infection.