Mesenchymal stem cell surgery, rescue, and regeneration in glaucomatous optic neuropathy

Glaucomatous optic neuropathy (GON) is an anatomofunctional impairment of the optic nerve triggered by glaucoma. Recently, growth factors (GF) have been shown to produce retinal neuroenhancement. The suprachoroidal autograft of mesenchymal stem cells (MSC) by Limoli Retinal Restoration Technique (LRRT) has proven to achieve retinal neuroenhancement by producing GF directly into the choroidal space. This retrospectively registered clinical study investigated the visual function changes in patients with GON treated with LRRT. Methods: Twenty-ve patients (35 eyes) with GON in progressive disease conditions were included in the study. Each patient underwent a comprehensive ocular examination, including analysis of Best Corrected Visual Acuity (BCVA) for far and near visus, sensitivity by Maia microperimetry, and the study of the spectral domain-optical coherence tomography (SD-OCT). The patients were divided into two groups: a control group, consisting of 21 eyes (average age 72.2 years, range 50–83) and an LRRT group, consisting of 14 eyes (average age 67.4, range 50–84). Results: After 6 months the BCVA, close-up visus, and microperimetric sensitivity signicantly improved in the LRRT-treated group (p < 0.05), whereas the mean increases were not statistically signicant in controls (p > 0.5). Conclusions: Patients with GON treated with LRRT showed a signicant increase of visual performance (VP) both in BCVA and sensitivity and an improvement of residual close-up visus, in the comparison between the LRRT results and the control group. Further studies will be needed to establish the actual signicance of the reported ndings.

being considered the main risk factor [3]. Hypotensive therapies, together with neurotrophic supplements, represent the recommended treatment for patients with GON to stop or slow down neurodegeneration.
The recent appearance of cell therapy in regenerative medicine has represented a promising tool in glaucoma therapy [6]. On the one hand, embryonic stem cells (ESC) and induced pluripotent stem cells (iPSC) have been used in preclinical and clinical studies to replace dead or diseased RGCs, but, although expressing RGC markers, they have not shown to effectively restore retinal connections, as they remain close to the injection site [6].
On the other hand, mesenchymal stem cells (MSC) can be exploited for their paracrine secretion of different molecules that have been shown to activate RGC-intrinsic regenerative programs after optic nerve injury, promoting cell survival and axonal regeneration [7]. Speci cally, the effectiveness of MSCs is expressed through several mechanisms, including hemorheological, anti-oxidative, anti-in ammatory, anti-apoptotic, neurotrophic and cytoprotective ones [3,6,7]. These mechanisms clinically could lead to the improvement of visual performance (VP) and the overall prognosis of glaucoma.
This study aimed to evaluate both the e cacy and safety of LRRT suprachoroidal MSC graft in patients with GON.

Study participants
This retrospectively registered study was performed at the Low Vision Center in Milan between January 2015 and September 2019. The study was performed in accordance with the tenets of the 1964 Declaration of Helsinki and was approved by the Institutional Review Board of the Low Vision Academy Written informed consent was obtained from all the patients prior to enrollment.
Individuals who met the following inclusion criteria were recruited for the study: Diagnosis of GON highlighted by microperimetry or, when necessary, coherence optical tomography (OCT).
Excavated optic papilla with cup/disc < 0.6; Good therapeutic balance (IOP ≤ 15 mmHg) obtained with hypotonizing therapy; the pressure must be corrected also after LRRT surgery.
Best-corrected visual acuity (BCVA) between + 1 and 0 logarithm of the minimum angle of resolution (LogMAR); Age between 50 and 84 years old; Stable eye conditions without previous surgery or ocular trauma.
Subjects who met any of the following criteria were excluded from the study: Bad therapeutic balance (IOP > 15 mmHg) obtained with hypotonizing therapy; IOP must be also corrected after LRRT surgery.
Refractive error 6 diopters of myopia, hypermetropia, and astigmatism; Presence of cataract or other media opacity that could interfere with a functional response; Presence of chorioretinal diseases including macular pucker with the altered foveal area, age-related macular diseases (AMD), or eredodistrophy, etc.; Intravitreal injection treatment and/or intraocular surgery; Inability to provide written informed consent; Inability to attend all follow-up visits; Systemic diseases including multiple sclerosis, epilepsy, vasculitis, Parkinson's disease, renal and hepatic diseases, malignant neoplasms, decompensated diabetes mellitus, etc.
All the eyes enrolled in this study were divided into two groups: the LRRT group who underwent autologous suprachoroidal graft of mesenchymal cells, and the control group of GON patients who did not undergo LRRT surgery.
Participants used as control were matched with GON patients according to the sensitivity alteration measured by microperimetry.

Ophthalmologic examination
The diagnosis of GON was established for each patient by the clinical analysis of visual performances (VP). Then, evaluation by slit-lamp biomicroscopy with and without dilatation, applanation tonometry, and retinal mapping with an indirect ophthalmoscope were performed. Also, BCVA, close-up visus, sensitivity measured by microperimetry (MY) with Maia 100809 (CenterVue S.p.A., Padua, Italy), spectral domain-optical coherence tomography (SD-OCT) with Cirrus 5000 (Carl Zeiss Meditec AG, Jena, Germany), and ocular electrophysiology with the Retimax electromedical system (C.S.O. Srl, Scandicci, Italy) were performed. All the ophthalmologic analyses were carried out by the same examiner at baseline (T0) and 6 months (T180) in both groups. Finally, the subjective improvement of VP in the LRRT group at 6 months after surgery was reported. BCVA was always measured according to the standards recommended by the early treatment diabetic retinopathy study charts (ETDRS) at 4 meters and expressed in logMAR. The visual acuity for near distance (close-up visus) was recorded in points (Pts). Microperimetry was performed using a Maia apparatus (Centervue spa, Padua, Italy) with images acquired by scanning laser ophthalmoscopy. Sensitivity was measured from 0 to 25 decibel (dB), and the ≥ color was coded. The eld of the infrared image was 36°x 36°, and perimetry was performed in a eld of 30° x 30° with a luminance of 4 asb. The Full-Threshold 4-L test was used to assess the retina in detail.

LRRT: cell isolation and grafting procedures
The autograft of MCs in the suprachoroidal space, i.e., LRRT consisted of the following triad: ASCs, ADSCs contained in SVF of adipose tissue, and platelets obtained from the PRP. ASCs were collected from the orbital fat during the surgical procedure according to previously published methods [8]. A scleral pocket with deep sclerectomy was created in each patient's eye to expose the surface of the choroidal space [9].
After exposing the choroid, the pedicle of adipose tissue derived from the orbital space was placed on the choroid's surface (Fig. 1). ADSCs contained in the SVF were grafted in the suprachoroidal space. The SVF was isolated from the abdominal fat according to the Lawrence and Coleman technique [14]. Brie y, 10 mL of adipose tissue were manually harvested from the abdominal subcutaneous layer of each patient using a 3 mm blunt cannula connected to a locking syringe. After adding 50 mL of saline solution to the freshly harvest lipoaspirate for 10 minutes to eliminate the blood component, the supernatant was extracted and centrifuged at 1500xg for 5 minutes at 20°C in order to isolate SVF from the mature adipocytes, connective tissues, cellular debris, and oil.
The platelets were obtained from PRP gel according to established methodologies [9]. Eight mL of human peripheral blood was collected with a 22 G needle and put in a Regen-BCT tube (RegenKit; RegenLab, Le Mont-sur-Lausanne, CH) for PRP preparation. The collected blood was centrifuged at 1500 x g for 5 min at 20° C in order to isolate the PRP.
The adipose pedicle was in ltrated with platelets derived from the PRP gel ( Fig. 1).
Finally, a mixture of ADSCs from the SVF and PRP was used to saturate the residual volume of the scleral pocket, where the pedicle of adipose tissue-derived from the orbital space was previously placed (Box 1).

Box 1. Surgical phases of Limoli Retinal Restoration Technique (LRRT).
Anchoring of the sclera with 6 − 0 silk suture, near the inferior-temporal limbus, and globe deviated to the superonasal quadrant.
Opening of the sub-conjunctival and Sub-Tenon's space at 11 mm from the inferior-temporal limbus, using 5.5" Westcott Tenotomy curved scissors.
Insert the Limoli-Basile conjunctival retractor in the space to make a scleral surgical eld.
To pre-cut the ap on the side in the sclera at 8 mm from the limbus using a 5-mm crescent knife angled up with the ap hinge always radial and to the left of the surgeon.
Open a deep scleral ap of about 5 x 5 mm at the inferotemporal quadrant, maintaining the radial hinge. The sclerectomy has to be deep enough to allow viewing the color of the choroid.
Remove a little operculum in the distal part of the ap in order to facilitate blood circulation in the subsequent suprachoroidal autograft.
Extract the orbital fat with forceps from a gap above the inferior oblique muscle. The fat must su ciently be vascularized to allow it to survive after its implantation Place the autologous fat ap on the choroidal bed and suture with choroidal 6/0 polyglactin ber at the proximal edge of the door.
Suture the scleral ap to avoid compression on the fat pedicle or its nutrient vessels.
In ltrate the stroma of the fat pedicle with 1 mL of PRP gel (obtained by centrifugation of the blood material, separation of the component, and platelet degranulation) using a 30 G angled (30°) cannula.
Leave a small exible plastic tube to insert the autologous ADSCs in the space between the ap, the choroid, and the suprachoroidal autograft, before closing.
Fill the remaining space between the autologous fat graft, choroid, and scleral aps with 0.5 cc of ADSCs in SVF and 0.5 of PRP using a 25 G cannula and close the suture.
After surgery, administer three days of antibiotic therapy with 500 mg of azithromycin. Also, provide eye drop therapy with an antibiotic and steroid combination, such as Chloramphenicol and Betamethasone, for about 15-20 days.

Cell identi cation by ow cyto uorimetry
Flow cytometry analyses were performed in order to identify the phenotypic characteristics of the population of cells within the graft, speci cally ADSCs and platelets. PRP and SVF were obtained from patients of the LRRT group who underwent LRRT surgery and were isolated under fresh conditions. SVF was manually isolated from each patient's lipoaspirate in a clean room near the operating room, according to a previously described method [15]. Brie y, the adipose portion of the lipoaspirate was washed with the phosphate-buffered saline (PBS; Biological Industries) and mixed with 2.5 mg/mL of collagenase type II (Worthington Biochemical Corporation, Lakewood, NJ) for enzymatic digestion. The collagenase/adipose mixture was placed in a 37°C water bath for 30 minutes in order to create a singlecell suspension and then ltered through a 100 µ cell strainer and nally a 45 µ mesh. It was centrifuged for 5 minutes at 1200 x g to collect the cellular SVF as a pellet. Once isolated, characterization of the cell composition of freshly-isolated SVF was obtained by multi-color ow cytometry (CytoFLEX Flow Cytometer, Backman Coulter, USA) that allows the in-vitro identi cation of the surface marker expression of the cells. The panel of cell surface antigens was chosen in agreement with the International Federation for Adipose Therapeutics and Science (IFATS) and International Society for Cellular Therapy (ISCT) recommendations [16]. The immunophenotypic analyses were performed to con rm the mesenchymal nature of isolated cells. The following uorochrome-labeled monoclonal antibodies were used for SVF analysis: CD31-PE, CD34-PC, and CD45-APC (Backman Coulter, USA). The markers were used in combination with ViaKrome (Beckman Coulter, USA), which determines cell viability, excluding debris and dead cells induced by the isolation protocol. Cells were incubated with speci c mAbs for 15 minutes. At least 105 cells were acquired from each sample. The software CytEpert Version 2.2.0.97, CytoFLEX (Beckman Coulter, Inc) was used to create dot plots and to calculate the cell composition percentages according to the pro le of the surface marker expression. Immunophenotyping of platelets by ow cytometry was performed on platelets in PRP. The following markers were used for platelet analysis: anti-CD41-FITC and CD61-PE [17].

Statistical analysis
All statistical analyses were performed using software SPSS Statistics (version 20.0, SPSS Inc, Chicago, IL). Data were summarized with the mean ± standard deviation (SD) and minimum and maximum (minmax) values were also reported. Student's t-test was used to compare values between controls and the study group. Paired t-test was run to compare study subjects and controls at baseline and after 6 months. A p value < 0.05 was considered statistically signi cant.

Patient Characteristics
A total of 35 eyes from 25 patients affected by GON (10 females and 15 males; mean age 70.7 ± 9.9 years, range 50-84 years) met the inclusion criteria and were enrolled in the study. Twenty-one eyes of the total composed the control group (8 males and 7 females; mean age 72.2 ± 9.6 years, range 50-83 years), while the remaining 14 eyes constituted the LRRT group of patients (7 males and 3 females; mean age 69.4 ± 9.2 years; range 50-84 years). Two patients from the latter group were excluded from the analysis because of the close-up visus that was not assessable, and the BCVA that was greater than 1 LogMAR. The baseline characteristics are summarized in Table 1. No adverse event associated with the surgery either intra-operatively or post-operatively was observed throughout the period. Mean values of the IOP recorded before and after surgery did not change signi cantly. All completed 6 months of evaluation.
Phenotype of platelets and freshly isolated SVF The positive expression for cell surface antigens CD61 and CD41 identi ed PLTs. We observed that the PRP contained a mean of 79.2 ± 13.7% PLT on a total of 10 5 cells.
The positive expression for cell surface antigens CD34 and negative expression for CD31 and CD45 identi ed ADSCs. The percentage of the phenotypically identi ed ADSC population was 44.9% ± 11% on a total of 10 5 cells and the ADSC/ L was 590.3/ L (127.2-1485.3/ L) (Fig. 2). BCVA After 6 months, the BCVA went from 0.0947 to 0.0937 logMAR in the control group with a mean increase of -0.001 (+ 1.09%; p > 0.05) and from 0.213 to 0.155 logMAR in the LRRT group with a statistically signi cant increase of -0.0582 (+ 27.32%; p = 0.0264) ( Fig. 3; Table 2). The difference between the 6month increase in LRRT treated group compared to evolution in the control group is considered to be statistically signi cant (p = 0.0353).

Close-up Visus
After 6 months, the close-up visus went from 6.57 to 6.9 pts in the control group with a mean reduction of -0.33 pts (-5.02%; p > 0.05) and from 10.21 to 8.29 pts in the LRRT-treated group with a mean increase of 1.93(+ 18.81%; p > 0.05) (Fig. 4, Table 2). However, the latter increase did not reach statistical signi cance. The difference between the 6-month increase in the LRRT-treated group compared to the evolution in the control group is not considered to be statistically signi cant (p = 0.0818).

Microperimetry
After 6 months, sensitivity went from 13.20 to 12.64 dB in the control group with a mean reduction of -0.56 dB (-4.24%; p > 0.05) and from 10.00 to 11.12 in the LRRT-treated group with a statistically signi cant increase of 1.12 dB (+ 11.24%; p = 0.0033) (Figs. 5 and 6, Table 2). The difference between the 6-month increase in the LRRT-treated group compared to the evolution in the control group is considered to be statistically signi cant (p = 0.0014).

Compliance
The subjective experience of all LRRT treated patients was surveyed. At 6 months, the VP increased in 11 eyes out of 14 (79%) and remained unvaried in 3 eyes (21%). Notably, the VP worsened in no eyes (Fig. 7).

Discussion
In this study, patients affected with GON received the LRRT treatment to preserve the residual VP. All of the eyes in the study group showed improvement in BVCA, sensitivity, and residual close-up visus with no ocular and systemic complications; whereas the majority of the eyes in the control group showed a decrease in the same measured parameters. The improvements were consistent through the 6 months follow-up.
GON is currently recognized as a progressive neurodegenerative disease, resulting in due course permanent visual loss [2]. Up to date, there are no curative treatments, however many potential options are being investigated in the clinical setting, including retinal prostheses, gene therapies, and cell-based treatments. Among these different therapeutic alternatives, growing research interest has been developed towards the MSCs, i.e. adult stromal cells, as promising candidates for cell therapy in retinopathies [6][7][8]18] These cells are ubiquitously distributed in the body and play a key role in organogenesis, tissue remodeling and repair [18]. A growing body of evidence points that MSCs can restore VP in different ocular degenerative disorders through various therapeutic pathways involving cell differentiation to replace the lost cells, paracrine activity to trigger cell survival and repair, and modulation of the local immune response [6][7][8]18].
These biological mediators are well known for providing a vital microenvironment by inducing gene expression changes that lead to neuroprotective, regenerative, anti-in ammatory, and anti-apoptotic effects; hence, they can support cell survival and rescue the damaged tissue [6][7][8]18]. This complex interplay has been shown to cause functional neuro-enhancement of the residual retinal cells and to regulate regeneration by reversing cell death or damage in different diseases. Especially due to their paracrine trophic activity, MSCs have emerged as trending regenerative biologic agents for retinopathies [18].
MSCs can be isolated from adult and fetal tissues, including bone marrow, adipose tissue, Wharton's jelly, dental pulp, and placenta [20,25]. ADSCs have been emerging as ideal MSCs among the other cell sources because of their sustainable costs, manageability, easy harvesting, and wide distribution in the adult tissues. Compared with bone marrow, adipose tissue contains a higher number of MSCs and of pericytes, which are the precursors to MSCs, and a lower amount of leukocytes [20,25]. Furthermore, the adipose tissue is one of the most attractive sources for MSCs due to the lack of ethical concerns involved in their application, and the high paracrine trophic and immunomodulatory effects. Most notably, ADSCs have been shown to have no risks of uncontrolled growth and malignant transformation, no rejection or immune reactions, demonstrating their long-term e cacy and compatibility in the transplanted tissue [22,25].
For all the provided reasons, ADSCs are ideal for autologous cell transplants and we chose to use them for our surgical procedure.
The LRRT is a cell therapy consisting of autologous ADSCs within the SVF, ASCs, and PRP [9]; it is administered intra-ocularly with a supra-choroidal delivery method.
Alongside ADSCs, also ASCs have shown regenerative potential as well as autologous PRP that is a source of growth factors [8,25].
Freshly isolated cells were subjected to ow cytometry analyses to con rm the immunophenotyping characterization of ADSCs within SVF and PRP.
Different routes of administration of MSCs have been explored in different clinical studies for the management of degenerative retinal diseases. We used the suprachoroidal method, which is reported to have no serious complications and is considered to be safer compared to the intravitreal or subretinal applications [11].
The suprachoroidal area has been shown as natural drug storage and an immune-protected region [9][10][11][12][13]. The GFs secreted by the ADSCs can effectively pass through that space and reach the retinal target without producing immune reactions, making the suprachoroidal region ideal as the site of MSCs administration.
The effect of ADSCs is thought to be related to the expression of several GFs, including bFGF, BDNF, NGF, CNTF, GDNF, and HGF [3,6,7]. GFs secreted by MSCs in the suprachoroidal space can either trigger the retinal cells in the quiescent phase to re-enter the cell cycle and activate the progenitor cells or act directly on the damaged cells supplying neuroprotection and reducing the retinal oxidative damage. GFs have been shown to inhibit apoptosis in the diseased retina, to mediate a neuro-cytoprotective action, and to suppress retinal chronic in ammation that occurs in glaucoma through an anti-in ammatory and immunomodulating action. Furthermore, several in vivo and in vitro studies have shown MSCs-mediated pleiotropic activity in stimulating angiogenesis in ischemic disease, myelination, dendritic and axonal regeneration through IGF secretion and mTOR pathway activation [20][21][22][23][24][25][26][27][28][29].
In this way, the can promote RGC survival and stimulate both axonal regeneration and myelination in the optic nerve, restoring both dendritic and synaptic connections with bipolar and amacrine cells [23][24][25][26][27].
According to these ndings, MSCs might promote the RGC function and survival through the paracrine release of GFs, exosomes, and microvesicles over time, slowing retinal degeneration.
These biochemical mechanisms could underlie the positive clinical results we observed following the autologous MSC graft performed by LRRT treatment in the suprachoroidal space in patients affected with GON.
The LRRT treatment has been applied in other studies of our group in retinal diseases, such as retinitis pigmentosa, AMD, and optic neuropathies, and it has been shown its safety and effectiveness with improvements of both VP and electroretinographic parameters [9][10][11][12]. In this clinical study, the LRRT demonstrated healing potential in patients with GON.
In accordance with our results, many investigators evaluated the safety and e cacy of the MSCs use for retinal diseases and suggest MSCs-mediated neuroprotection [13,30,31].
Oner et al. [13] showed that the suprachoroidal implantation of ADSCs in patients with optic nerve disease caused functional improvement in VP in terms of visual acuity, visual eld, and mfERG recordings. These outcomes are believed to be related to the paracrine secretion of neurotrophic and angiotrophic GFs from ADMSCs and angiotrophic GFs from PRP, suppressing the in ammation and protecting RGCs from death.
In Brazil, Siqueira et al. [31] conducted a study with intravitreal injection of bone-marrow derived stem cells in patients with RP, showing the safety of the cell therapy and observing an increased quality of life.
Finally, a Californian group [32] obtained similar results with the intravitreous use of BMDSCs in patients affected with retinal vascular occlusion, non-exudative age-related macular degeneration, or retinitis pigmentosa. The investigators assessed the safety and feasibility of cell therapy, showing the important role that MSCs may play in tissue repair.
Alongside ADSCs, also ASCs have shown regenerative potential as well as autologous PRP that is a source of growth factors [8,25].
Freshly isolated cells were subjected to ow cytometry analyses to con rm the immunophenotyping characterization of ADSCs within SVF and PRP.
Different routes of administration of MSCs have been explored in different clinical studies for the management of degenerative retinal diseases. We used the suprachoroidal method, which is reported to have no serious complications and is considered to be safer compared to the intravitreal or subretinal applications [11].
The suprachoroidal area has been shown as natural drug storage and an immune-protected region [9][10][11][12][13]. The GFs secreted by the ADSCs can effectively pass through that space and reach the retinal target without producing immune reactions, making the suprachoroidal region ideal as the site of MSCs administration.
The effect of ADSCs is thought to be related to the expression of several GFs, including bFGF, BDNF, NGF, CNTF, GDNF, and HGF [3,6,7]. GFs secreted by MSCs in the suprachoroidal space can either trigger the retinal cells in the quiescent phase to re-enter the cell cycle and activate the progenitor cells or act directly on the damaged cells supplying neuroprotection and reducing the retinal oxidative damage. GFs have been shown to inhibit apoptosis in the diseased retina, to mediate a neuro-cytoprotective action, and to suppress retinal chronic in ammation that occurs in glaucoma through an anti-in ammatory and immunomodulating action. Furthermore, several in vivo and in vitro studies have shown MSCs-mediated pleiotropic activity in stimulating angiogenesis in ischemic disease, myelination, dendritic and axonal regeneration through IGF secretion and mTOR pathway activation [16,[20][21][22][23][24][25][26][27][28][29].
In this way, the can promote RGC survival and stimulate both axonal regeneration and myelination in the optic nerve, restoring both dendritic and synaptic connections with bipolar and amacrine cells [23][24][25][26][27].
According to these ndings, MSCs might promote the RGC function and survival through the paracrine release of GFs, exosomes, and microvesicles over time, slowing retinal degeneration.
These biochemical mechanisms could underlie the positive clinical results we observed following the autologous MSC graft performed by LRRT treatment in the suprachoroidal space in patients affected with GON.
The LRRT treatment has been applied in other studies of our group in retinal diseases, such as retinitis pigmentosa, AMD, and optic neuropathies, and it has been shown its safety and effectiveness with improvements of both VP and electroretinographic parameters [9][10][11][12]. In this clinical study, the LRRT demonstrated healing potential in patients with GON.
In accordance with our results, many investigators evaluated the safety and e cacy of the MSCs use for retinal diseases and suggest MSCs-mediated neuroprotection [13,[30][31][32].
Oner et al. [13] showed that the suprachoroidal implantation of ADSCs in patients with optic nerve disease caused functional improvement in VP in terms of visual acuity, visual eld, and mfERG recordings. These outcomes are believed to be related to the paracrine secretion of neurotrophic and angiotrophic GFs from ADMSCs and angiotrophic GFs from PRP, suppressing the in ammation and protecting RGCs from death.
In Brazil, Siqueira et al. [31] conducted a study with intravitreal injection of bone-marrow derived stem cells in patients with RP, showing the safety of the cell therapy and observing an increased quality of life.
Finally, a Californian group [33] obtained similar results with the intravitreous use of BMDSCs in patients affected with retinal vascular occlusion, non-exudative age-related macular degeneration, or retinitis pigmentosa. The investigators assessed the safety and feasibility of cell therapy, showing the important role that MSCs may play in tissue repair.
The study has some limitations. First, our sample size was small and the study was not masked. A larger number of patients will be necessary to evaluate the effects of this therapy.
Second, the duration of action of LRRT treatment is unknown. Even though long-term research is necessary to determine the duration of e cacy, PRP booster injections after 12 months have been shown to maintain the outcomes. Another limitation of the study is that we do not measure whether additional treatments such as electrical stimulation may increase MSCs activity. The latter limitation forms the basis for near-future studies.

Conclusions
The LRRT treatment has proven safe and effective in treating patients affected with GON.
In our experience, both visual acuity and retinal sensitivity measurements showed statistically signi cant improvements in 80% of GON patients after LRRT during the follow-up period of 6 months, and no ocular or systemic effects were reported. Therefore, autologous MSC graft combined with PRP into the suprachoroidal space could contribute to restoring optic nerve function, improving the clinical, prognostic, and rehabilitative aspects in patients affected with GON, that currently have no curative treatment options.
Further studies are needed to validate our ndings and to unveil the potential of MSCs as therapeutic agents in regenerative medicine especially for degenerative retinal and optic nerve diseases.