Science - USA (2022-04-22)

C144 (Fig. 2, B and C, and fig. S5A). In addition,
whereas binding of CT-P59 to B.1.351 and P.1
variants was minimally changed (37 to 100%),
neutralization was decreased 26- to 43-fold (Fig.
2, B and C). The remaining antibodies showed

minimal binding changes and a <3.6-fold differ-

ence in neutralization half-maximal inhibitory

concentration (IC 50 ) (Fig. 2, B and C, and fig.

S5A). An evaluation of the antibodies in our

panel against B.1.617.2 revealed minimal changes

in binding and neutralization for all antibodies

except REGN10987, A19-46.1, and LY-CoV555

(Fig. 2, B and C). As previously reported

( 14 , 39 Ð 42 ), REGN10987 binds B.1.617.2 spike

but has 22-fold less neutralization, and the

Zhouet al.,Science 376 , eabn8897 (2022) 22 April 2022 3 of 12

Pseudotyped virus neutralization of monoclonal antibodies (ng/mL)

Antibody Generic name Class IC 50 IC 80 IC 50 IC 80 IC 50 IC 80 IC 50 IC 80 IC 50 IC 80 IC 50 IC 80

A23-58.1 -- I 1.3 4.5 2.1 4.7 4.7 11.6 1.6 5.7 1.6 3.5 231 1132

B1-182.1 -- I 0.9 2.4 1.7 3.9 2.0 4.3 <0.6 1.51.03.5 281 1301

COV2-2196 tixagevimab * I 2.0 3.2 2.7 5.4 3.5 10.6 1.9 7.51.44.5 269 900

S2E12 -- I 1.4 2.9 6.8 3.3 2.2 4.1 <0.6 2.51.12.438.1 112

CB6 etesevimab I 50.5 109 22.7 141 > 10,000 > 10,000 > 10,000 > 10,000 14.8 492 > 10,000 > 10,000

REGN10933 casirivimab I 6.1 16.0 9.7 29.1 > 10,000 > 10,000 1536 > 10,000 3.0 3.8 > 10,000 > 10,000

CT-P59 regdanvimab I 1.5 4.6 5.5 22.3 65.8 233 39.6 153 14.7 78.3 > 10,000 > 10,000

ADG2 -- I,IV 5.1 14.7 4.7 17.1 15.5 43.9 6.4 21.8 7.6 20.8 2037 8113

A19-46.1 -- II 19.4 40.6 39.2 106 57 157 39.7 69.6 > 10,000 > 10,000 223 376

LY-COV555bamlanivimab II 3.6 10.4 7.7 24.9 > 10,000 > 10,000 > 10,000 > 10,000 > 10,000 > 10,000 > 10,000 > 10,000

C144 -- II 5.1 9.8 5.7 22.4 > 10,000 > 10,000 > 10,000 > 10,000 > 10,000 > 10,000

A19-61.1 -- III 7.7 20.1 31.3 49.4 10.8 19.6 7.3 15.0 18.7 27.5 > 10,000 > 10,000

REGN10987 imdevimab III 20.0 412 13.5 72 24.4 102 6.5 53.6 455 4926 > 10,000 > 10,000

COV2-2130 cilgavimab * III 3.7 10.9 6.3 12.5 5.4 13.3 14.1 18.2 25.0 80.5 5850 > 10,000

C135 -- III 10.8 64.3 13.1 191 34.0 n.d. ‡ 14.7 115 > 10,000 > 10,000

S309 sotrovimab * III 36.1 163 30.6 194 27.7 97.4 41.8 359 45.2 113 281 1336

LY-CoV1404bebtelovimab III 3.0 5.8 30.6 194 4.1 8.7 11.5 17.6 3.7 8.5 5.1 14.4

D614G B.1.1.7 B.1.351 P.1 B.1.617.2 B.1.1.529 (BA.1)

4.9 20.6

24.7 n.d. ‡

B.1.1.7 ()

2 del., 7a.a. substitutions

1 RBD substitution

N501Y

Variant

substitutions

RBD substitutions

B.1.351 ()

1 del., 7 a.a. substitutions

3 RBD substitutions

K417N E484K

N501Y

P. 1 ()

12 a.a. substitutions

3 RBD substitutions

K417T E484K

N501Y

B.1.617.2 ()

7 a.a. substitutions

2 RBD substitutions

L452R T478K

B.1.1.529 ()

3 del., 1 ins., 30 a.a. substitutions

15 RBD substitutions

G339D

S373P

K417N

G446S

T478K

Q493R

Q498R

Y505H

Cell surface binding of monoclonal antibodies to variant spike proteins

S371L

S375F

N440K

S477N

E484A

G496S

Antibody Class N501Y

A23-58.1 I

B1-182.1 I

COV2-2196 I

S2E12 I

CB6 I

REGN10933 I

CT-P59 I

ADG2 I, IV

A19-46.1 II

LY-COV555 II

C144 II

A19-61.1 III

REGN10987 III

COV2-2130 III

C135 III

S309 III

LY-CoV1404 III

100

100

100

100

100

100

100

100

100

100

100

100

100

100

100

100

100

200%

100%

0%

Normalized

to D614G

D614G B.1.1.7 P. 1

204

213

146

160

158

132

204

163

216

139

197

141

142

158

139

137

191

B.1.351

116

118

110

102

0

0

11

104

121

102

100

0

106

98

113

97

104

37

39

40

37

0

0

3

42

38

34

37

0

41

40

40

36

35

107

0

0

89

114

115

85

112

105

90

120

117

122

119

93

B.1.617.2 B.1.1.529

99 13

97 9

0

47

0

0

0

0

3

32

0

0

0

24

15

44

4

C

B

A Location of variant amino acid substitutions on SARS-CoV-2 spike protein

T a.a.

Fig. 2. SARS-CoV-2 monoclonal antibody binding and neutralization.
(A) Models of SARS-CoV-2 WA-1 spike protein (PDB: 6XM3) with the locations
of substitutions present in variants indicated as red dots. Also indicated is the
total number of amino acid substitutions and the number and locations of RBD
substitutions in VOC spike proteins. (B) Full-length spike proteins from the indicated
SARS-CoV-2 variants were expressed on the surface of transiently transfected
293T cells, and binding to indicated monoclonal antibodies was assessed by means
of flow cytometry. Antibody mean fluorescence intensity (MFI) binding signal was
adjusted according to spike protein expression level (fig. S4). Shown is the ratio of
the adjusted antibody MFI binding to the indicated spike expressing cells to the
adjusted MFI of the same antibody bound to D614G spike–expressing cells. The data
are expressed as a percentage. Shown is a representative experiment (n=2
replicates). (C) Lentiviruses pseudotyped with SARS-CoV-2 spike proteins from

D614G, B.1.1.7, B.1.351, P.1, B.1.617.2, or B.1.1.529 (BA.1) were incubated with serial

dilutions of the indicated antibodies, and IC 50 and IC 80 values were determined.

S309 was tested on 293 flpin-TMPRSS2-ACE2 cells, whereas all the other antibodies

were tested on 293T-ACE2 cells. Ranges are indicated with white (>10,000 ng/ml),

light blue (>1000 to≤10,000 ng/ml), yellow (>100 to≤1000 ng/ml), orange (>50 to

≤100 ng/ml), red (>10 to≤50 ng/ml), maroon (>1 to≤10 ng/ml), and purple

(≤1 ng/ml). n.d.‡, not determined because of incomplete neutralization that

plateaued at <80% (fig. S5B). Where available, generic names of antibodies

under therapeutic investigation are shown. Gray shading indicates antibodies that

previously or currently have received Emergency Use Authorization from the US

Food and Drug Administration. Generic names with an asterisk indicate therapeutic

antibody products with the same binding regions as those of the antibodies being

tested but containing amino acid changes in their Fc domains.

RESEARCH | RESEARCH ARTICLE