Optimal Efficacy and Safety of Humanized Anti-Scg3 Antibody to Alleviate Oxygen-Induced Retinopathy

Abstract

The retinopathy of prematurity (ROP), a neovascular retinal disorder presenting in premature infants, is the leading causes of blindness in children. Currently, there is no approved drug therapy for ROP in the U.S., highlighting the urgent unmet clinical need for a novel therapeutic to treat the disease. Secretogranin III (Scg3) was recently identified as a disease-selective angiogenic factor, and Scg3-neutralizing monoclonal antibodies were reported to alleviate pathological retinal neovascularization in mouse models. In this study, we characterized the efficacy and safety of a full-length humanized anti-Scg3 antibody (hAb) to ameliorate retinal pathology in oxygen-induced retinopathy (OIR) mice, a surrogate model of ROP, by implementing histological and functional analyses. Our results demonstrate that the anti-Scg3 hAb outperforms the vascular endothelial growth factor inhibitor aflibercept in terms of efficacy and safety to treat OIR mice. Our findings support the development of anti-Scg3 hAb for clinical application.

Keywords: Scg3; aflibercept; anti-Scg3 therapy; anti-VEGF; anti-angiogenic therapy; humanized antibody; oxygen-induced retinopathy; retinopathy of prematurity; safety; secretogranin III.

Figure 1

Intravitreal administration of anti-Scg3 hAb inhibits pathological RNV and reduces central avascular area in OIR mice. (A) Flat-mount retinas of OIR mice treated with anti-Scg3 hAb at 0.1, 0.5, 2, or 4 µg/0.5 µL/eye. Retinal vessels were stained with AF488-IB4. Yellow box corresponds to the magnified area below, showing changes in vascular pattern. (B) Quantification of RNV in (A). (C) Quantification of the central avascular area in (A). n = 8 eyes (0, 0.1, and 2 μg) and 6 (0.5 and 4 μg) in (B,C). (D) Flat-mount retina of OIR mice treated with different reagents (2 µg/0.5 µL/eye). (E) Quantification of RNV in (D). (F) Quantification of the central avascular area in (D). n = 33 eyes (PBS), 14 (control IgG), 26 (aflibercept) and 21 (anti-Scg3 hAb). Mean ± SEM; **** p < 0.0001; ns, no significance. one-way ANOVA test. Scale bar = 500 μm (whole retina).

Figure 2

Intravitreal anti-Scg3 hAb ameliorates aberrant retinal artery tortuosity and decreases vein width in OIR mice. (A) Representative FA images of OIR mice treated with the indicated agents (2 µg/0.5 µL/eye). Arrowheads indicate tortuous arteries; arrows show dilated veins. (B) Quantification of retinal arterial tortuosity in (A). Tortuosity index = actual vessel length/linear length. (C) Quantification of vein width in (A). Mean ± SEM; n = 10 eyes (non-injection), 18 (PBS), 20 (control hIgG), 20 (aflibercept), and 23 (anti-Scg3 hAb). **** p < 0.0001; ns, not significant; one-way ANOVA test.

Figure 3

Intravitreal anti-Scg3 hAb improves short- and long-term visual function in OIR mice. (A–C) Short-term pfERG of OIR mice at P21 (7 DPI). a-wave amplitude remained unchanged between groups (A). Intravitreal aflibercept suppressed b-wave amplitude (B). n = 20 eyes (PBS), 24 (aflibercept), and 25 (anti-Scg3 hAb). (D–F) Long-term pfERG of OIR mice at P42 (28 DPI). a-wave amplitude was unchanged (D). Anti-Scg3 improved b-wave amplitude, whereas aflibercept decreased the amplitude (E). n = 27 eyes (PBS), 32 (aflibercept), 36 eyes (anti-Scg3 hAb). (G–I) Short-term sfERG of OIR mice at P21 (7 DPI). Anti-Scg3 hAb significantly improved both a- and b-wave amplitudes at P21 in (G,H) with more improvement at higher light intensity. Aflibercept improved only b-wave amplitude at the highest light intensity (H). n = 14 eyes (PBS), 12 (aflibercept), 10 eyes (anti-Scg3 hAb). (J–L) Long-term sfERG of OIR mice at P42 (28 DPI). The results are similar to P21. n = 12 eyes (PBS), 22 (aflibercept), 22 eyes (anti-Scg3 hAb). (C,F,I,L) are representative graphs of ERG in each group. (I,L) show light intensity at 1 cd·s/m². Mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001; ns, not significant; one-way ANOVA test.

Figure 4

Intravitreal anti-Scg3 hAb does not adversely affect visual function in healthy neonatal mice. (A–D) Short-term pfERG of healthy mice at P21 (7 DPI). n = 20 eyes (non-injection), 20 (PBS), 27 eyes (anti-Scg3 hAb), and 30 (aflibercept). (E–H) Long-term pfERG of healthy mice at P42 (28 DPI). n = 12 eyes (non-injection), 12 (PBS), 10 (anti-Scg3 hAb), and 10 (aflibercept). (A,E), a-wave amplitude. (B,F), b-wave amplitude. (C,G), b-wave latency. (D,H), average ERG graphs of all mice in each group. Aflibercept decreased a- and b-wave amplitudes at P21 and P42 (A,B,E,F). Additionally, aflibercept also increased b-wave latency at P21 and P42 (C,G). No adverse effects of anti-Scg3 hAb on ERG were detected at either P21 or P42. Mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001; ns, not significant; one-way ANOVA test.

Figure 5

Systemic treatment with anti-Scg3 hAb does not adversely affect visual function in healthy neonatal mice. (A–E) Short-term pfERG of healthy mice at P21 (i.p. treatment began at P7, followed by continually i.p. injection every other day until P19) detected no change in a-wave amplitude and latency (A,B) as well as b-wave amplitude and latency (C,D) with anti-Scg3 hAb. Aflibercept therapy demonstrated a decrease in all parameters except in the latency of a-wave. Average ERG graphs of all mice in each group were depicted in (E). Mean ± SEM. n = 12 eyes (PBS), 10 eyes (aflibercept) and 12 eyes (anti-Scg3 hAb). * p < 0.05, ** p < 0.01, *** p < 0.001; ns, not significant; one-way ANOVA test.

Figure 6

Systemic treatment with anti-Scg3 hAb was without adverse effects on the development of healthy neonatal mice. (A) Representative images of body size at P21 for healthy mice treated with PBS, anti-Scg3 hAb or aflibercept (i.p. treatment as in Figure 5A). (B) Quantification of the body weight of the mice in A. Aflibercept hinders the body weight gain. n= 9 mice (PBS), 10 mice (aflibercept), and 9 mice (anti-Scg3 hAb) for each group. (C) Representative images of kidney size at P21 for mice treated with PBS, anti-Scg3 hAb, or aflibercept (i.p. treatment as in Figure 5A). (D) Western blot analysis of kidneys in (C) using anti-CD31 antibody (endothelial cell marker). Aflibercept reduced kidney endothelial cells. (E) Quantification of CD31 signal intensity in (D). Data were normalized against β-actin. n= 4 kidneys in 4 mice/group. (F) Representative image of kidney H&E staining. Arrows indicate aberrant glomerular morphology in the aflibercept group. Mean ± SEM. * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001; ns, not significant; one-way ANOVA test.