Table 2.

Summary of in vivo studies utilizing an AAV vector encoding erythropoietin (EPO) as gene therapy.

Reference/Study	Country	Injury Model and Animal	Groups and Sample Size	Parameters for Efficacy	Main Result	Remarks
2. Viral vector encoding erythropoietin (EPO)
a. AAV2-EPO gene therapy Tao Y. et al., 2020 [17]	China	N-methyl-N-nitrosourea (MNU) induced retinal degeneration, mice	10 negative control, 10 damage + treatment, 10 damage only, 10 treatment only	Visual function via ERG responses Behavioural function via vision-guided optokinetic tests Retinal layer thickness Cone photoreceptor count Multi-electrode array (MEA) field potentials RGC spontaneous firing spikes Protective mechanism	Scotopic B-wave amplitudes and photopic ERG responses in the damage + treatment group were significantly higher than the damage only group. Visual acuity and contrast sensitivity was significantly higher in the damage + treatment group than in the damage only group. Retina thickness was significantly larger in the damage + treatment group compared to the damage only group. The number of peanut agglutinin (PNA)+ cone photoreceptors was significantly higher in the damage + treatment group compared to the damage only group. The mean amplitude of field potential was significantly larger in the damage + treatment group compared to the damage only group. The spontaneous firing rate of RGCs in the damage + treatment group was significantly lower compared to the damage only group. The mRNA levels of caspase-3, CHOP, and Bax were significantly downregulated in damage + treatment group vs. damage only. The mRNA level of Bcl-2 was significantly upregulated in damage + treatment group vs. damage only.	The damage + treatment group was not significantly different than the control group. Visual acuity and contrast sensitivity in the damage + treatment group were not significantly different from the control group. The damage + treatment group was not significantly different than the control group. PNA staining was absent in the damage only group. Field responses in the damage + treatment group were not homogenous. Peripheral regions were significantly higher than mid-peripheral and central regions. Mid-peripheral regions were significantly larger than central regions. The damage + treatment group still had significantly higher spontaneous firing rates compared to normal mice.
b. rAAV.EpoR76E gene therapy Hines-Beard J. et al., 2016 [18]	United States	DBA/2J mouse model of glaucoma	15 negative control, 25 GFP vector control, 30 treated mice	Protection against vision loss Microglia number Microglia morphology Pro-inflammatory cytokines/chemokines Antioxidant enzymes	The N1 and P1 amplitudes in treated mice were not significantly different from negative control or GFP vector control mice. The number of microglia were significantly reduced in treated mice vs GFP vector control mice in both central and peripheral retina. The average soma area of treated mice was significantly smaller compared to GFP vector control mice. Treated mice had significantly reduced IL-1α, IL-1β, IL-17, IL-12p40, IL-12p70, CCL4 (MIP-1β), CCL5 (RANTES), and IL-13 mRNA levels compared to GFP vector control mice. There was an increase in the expression of several antioxidant proteins in both GFP vector control and treated mice compared to negative control.	IOP was not significantly different between treated and GFP vector control mice. Microglia numbers increase prior to the onset of IOP elevation in this mouse model. The average soma area of treated mice was significantly different than negative control mice There was no significant difference in TNFα levels between treated and vector control mice. These increases were not significant.