With aging of the world population, neurodegenerative diseases have come to the forefront of public health concerns. In the current study we report anatomic reversal of optic nerve neurodegeneration in OAG, measured objectively by OCT.
Progressive thickening (improvement) of the NFL and GCC have not been previously observed in OAG. 1–8 In the current study, positive NFL and GCC trends were noted predominantly in VPT treated OAG eyes (p < 0.0001 each). Regarding the magnitude of change, while control eyes demonstrated an average NFL loss of -0.57 um/year, VPT treated eyes showed a net + 0.42 um/year improvement, for a + 0.99 um/year difference favoring VPT eyes (p < 0.0001). Likewise, VPT treated eyes demonstrated less GCC loss, for a + 0.24 um/year advantage to VPT compared to control eyes (p = 0.0175). These findings in VPT treated eyes go clearly against the grain of expectation in OAG. 1–16 How might these findings be understood?
The mechanism of action of SDM has been discussed in several prior publications and supported by numerous in vitro and in vivo studies.10–40 SDM preferentially affects the retinal pigment epithelium (RPE) and has no direct effect on the neurosensory retina, transparent to the 810nm wavelength.10–15, 18–27, 35 SDM causes photothermal activation of RPE heat-shock proteins (HSPs) at energy levels sublethal to the RPE, upregulating and accelerating the endoplasmic reticulum (ER) unfolded protein response (UPR) to improve and normalize cell function and inhibit apoptosis. 28–31,34,40 The reaction is catalytic and triggers multiple secondary cascades of restorative processes, including inflammation reduction and therapeutic local and systemic immunomodulation. 14, 17–20 By so doing, SDM activates a biologic “reset” of the RPE, acting as a non-specific trigger of disease-specific repair agnostic to the cause of protein misfolding and subsequent cellular dysfunction. 11, 14, 35 SDM is hormetic, improving cell function by inflicting a significant but sublethal acute physiologic stress on the RPE to which the cell responds as an existential threat, improving RPE, and thence retinal, function. 14, 28–34, 37, 40
The fundamental effect of the SDM-elicited reset mechanism, and HSP activation in general, is thus to normalize the function of dysfunctional RPE cells directly exposure to treatment. The reset effect is, by definition, agnostic to the underlying cause of the dysfunction and has no notable effect on normally functioning cells. In this, it has been described as “pathoselective” and “a non-selective trigger of disease-specific repair”. 11,14,35,36,38 Prior studies have shown that SDM-elicited responses occur in in proportion to the level of baseline dysfunction: the greater the dysfunction, the greater the degree of improvement.10–12 Thus, a response to SDM indicates RPE dysfunction as normally functioning tissue cannot be further normalized. The observations in the eyes reported in the current study are consistent with these observations, revealing the presence of a modifiable and until recently unrecognized dysfunction of the RPE in eyes with OAG, or the “retinopathy of OAG” (ROAG). 10–12, 14 ROAG appears to be characterized by a hyponeurotropism specific to OAG leading to characteristic glaucomatous optic atrophy and associated visual loss.10, 13, 14 (Fig. 1) ROAG is responsive to SDM, indicated by improvements in visual function and electrophysiology and visual function following treatment. 10–14 (Fig. 5)
The improvements in retinal and visual function elicited by SDM have been shown to be highly predictive surrogates for key anatomic and clinically important endpoints, such as slowed progression of geographic atrophy and inhibition of neovascular conversion in dry AMD, and resolution of diabetic macular edema and reversal of retinopathy in diabetes.10,11,14, 20,24, 26, 41–46 Thus, while the trends in retinal layer thickening in OAG eyes managed with VPT we report are novel, there is ample precedent for anatomic improvements following SDM VPT for indications other than OAG, assessed by other means, resulting in significant long-term benefits.
Hou and associates found an average rate of GCC thinning of -1.18um / year in OAG, while Shin et al found average GCC thickness loss was significantly higher in eyes with progressive disease (-1.05 ± 0.98 µm/year for mild glaucoma and − 0.66 ± 0.30 µm/year for moderate to advanced glaucoma) than eyes without progression (-0.47 ± 0.54 µm/year for mild glaucoma and − 0.31 ± 0.50 µm/year for moderate to advanced glaucoma). 7 The average rate of GCC loss in control eyes in the current study is thus intermediate to these measures (-0.554 µm/year), while GCC loss in VPT treated eyes was much lower (-0.148 µm/year), for a net VPT treatment effect of + 0.406 µm/year (p = 0.0175). Miki and associates found the estimated mean rate of NFL loss significantly associated with visual field loss (− 2.02µm/year in eyes with visual field loss vs. −0.82µm/year without, P < 0.001), finding that each 1µm/year increased rate of NFL loss was associated with a 2.05 times higher risk of visual field loss.6 In the current study, control eyes had an average NFL loss of -0.692 µm/year, while NFL trends of VPT treated eyes demonstrated positive trends, with net average thickening of the NFL of + 0.347 µm/year, for a net VPT treatment effect of + 1.039 µm/year (p < 0.0001). Thus, the improvements we report in the current study represent significant departures from the typical course of OAG, indicating reversal of disease progression at anatomical level which should lead to reduced risks of visual loss. 3–7
Thinning of the NFL and GCC (and thus negative trends) in OAG is generally attributed to loss of RGCs and their axons. 1–9 Thus, the simplest explanation for the progressive NFL and GCC layer thickening in the eyes we report following VPT would be neuroregeneration. 1–8, 47,48 Retinal thickening, such as macular edema or from vitreoretinal traction, or retinal thinning such as from myopia, does not typically alter OCT-measured NFL or GCC layer thicknesses.1–8 It is possible that the retina layer thickenings observed in this study could be due to thickening of something else in those layers, such as glial tissue. 48 This is particularly a potential concern for the much thicker GCC. However, the combined improvement of both form (progressive thickening of NFL and GCC) and function (previously reported electrophysiologic and visual function improvements) (Fig. 5) suggest possible SDM-elicited retinal neuroregeneration, as proliferation of adventitia would not be expected to significantly improve retinal and visual function. 10–14
“Neuroregeneration” may describe either axonal or neurocyte (retinal ganglion cell, RGC) regeneration, or both. 9, 47, 54–59 Under normal circumstances regeneration of neurocytes in the central nervous system is not observed. A brief perusal of the internet, however, will show that virtually every major research university and many companies currently devote significant resources to discovering the conditions under which CNS neuroregeneration might be achieved, indicating the widespread belief in the possibility of success. The primacy of NFL layer improvement in VPT treated eyes we report suggests that RGC axonal regeneration may be the major neuroregenerative effect of SDM VPT treatment. Because the NFL is a component of the GCC, axonal regeneration would manifest in both measures, but make smaller contribution to the much thicker GCC layer, possibly accounting for the higher percentage of VPT treated eyes with improved NFL than GCC trends. 1–8 Regeneration of RGC dendrites that make up the inner plexiform layer of the GCC could also contribute to thickening of the GCC, as well as increases in retinal macro- and microglia, noting that Mueller cell activation has been documented following SDM in patients with diabetic macular edema. 7–8, 16, 24, 49 Regeneration of RGCs themselves, i.e., such as from activated HSP-mediated differentiation of pluripotent stem cells, cannot be directly determined, as it is difficult to measure the RGC layer by OCT or visualize human RGCs in vivo. 51–53 Thus, while we present evidence consistent with retinal neuroregeneration, proof will require further study. It has been estimated that loss of 25–35% of RGCs precedes clinical visual field loss. 16 Thus, the improvements in RGC function reflected in improved visual, retinal, and optic nerve function in OAG and other CPRs following SDM VPT, as well as the anatomic improvements reported in the current study in OAG are noteworthy, as they represent reversals of disease progression. These functional improvements could result from neuroenhancement alone, representing restored and/or improved function of pre-existing neural elements. 9,47 However, it is difficult to also explain the highly significant anatomic improvements in NFL and GCC thicknesses reported here absent neuroregeneration. (Fig. 1, 2, and 5)
Neurocytes, such as RGCs, may survive by retrenching and sacrificing axons and dendrites in response to a hostile environment or trauma. 54–64 In the peripheral nervous system, such neurocytes can regrow or “re-perfuse” abandoned axons to reestablish axonal flow and their distal interneural connections to restore function when conditions improve. This is not the case with central nervous system neurons, such as RGCs. 54–63 Despite this difference, peripheral neuroregeneration, which is permitted, may be the best model for central nerve regeneration, which is not normally permitted. 54–60 Because peripheral and central nerves share the same genetic information, the difference in their ability to regenerate, conferred at the time of differentiation, appears to reflect epigenetic control of gene expression. 54–60 The heat shock stress response, triggered in the retina by SDM, is a known modulator of gene expression. 24, 30, 52, 53, 64–66 Thus, it is possible that SDM-induced HSP-mediated epigenetic modifications may release RGCs from their native CNS epigenetic inhibition against axonal regeneration.
HSPs are influential maintaining stem cell viability, activation, differentiation, and maintenance of pluripotency. 11–14, 24 SDM-elicited retinal homing of bone-marrow derived mesenchymal, monocytic, and CD34 + stem cells have been observed in an animal model.17 Recruited to the retina by SDM RPE HSP activation, this same HSP activation may also induce differentiation of these stem cells into retinal neural elements. 61–66 HSP-initiated immunoactivation has been found to have neuroprotective effects in animal models of retinal degenerations measured by electrophysiology. 63 Thus, HSP activated immune cell-mediated paracrine regeneration has been proposed for neuroprotective therapy for all CPRs. 60–66 Because such laboratory findings echo the clinical improvements seen in eyes with various CPRs following SDM VPT RPE HSP activation, it is also possible that SDM-elicited immune cell activation may contribute to the NFL and GCC layer thickening observed in the current study. 10–15, 41–46
Finally, retinal micro- and macroglia might also play a role. 24 Axonal degeneration has been correlated with inflammatory microglial activation in glaucoma and shown to precede RGC death and visual field loss. 52, 55, 57, 64–66 Microglial and Mueller cell (macroglia) activation generally occurs in response to retinal cytokine signals having distinctive pro-degenerative / pro-inflammatory (M1 and A1) and pro-regenerative / anti-inflammatory (M2 and A2) phenotypes. 24, 55, 57, 64–66 Most of these cytokines are of RPE origin and thus subject to SDM-induced therapeutic modulation. 11, 14, 35,39 For example, Midena and associates have identified biomarkers of neuroregenerative (M2) Mueller cell (macroglia) activation in the aqueous humor of patients treated with SDM for diabetic macular edema. 24
This report has notable limitations. It is retrospective, and thus non-randomized. To our knowledge, this is the first report of therapeutic anatomic restoration of the neurosensory retina, and thus the optic nerve. Novelty is a special limitation, in that violates convention and there are no prior corroborating studies. Novelty is how we learn and progress, however. The study groups differ in two respects. First, the retinal layer trend data in the VPT group could go no further back than October 2020. The resulting shorter follow up of a median 708 days for VPT compared to 730 days for control eyes is mitigated by the constant slopes of the retinal layer trends over time. (Fig. 2) Second, the VPT group was significantly older than the control group. If anything, this difference would favor the younger control group as the retinal layers tend to thin with age. 1.,2 However, because consistent thickening of either the GCC or NFL has never been previously observed, let alone in association with any particular datum (age, IOP, VA, glaucoma severity, surgery, etc), no imbalance in group demographics – other than VPT in the treatment group – could account for the positive in NFL and GCC trends in the VPT group. Finally, natural NFL and GCC trend increases are sufficiently rare (considered no more than sampling artifact) that it seems highly unlikely that our findings could be even deliberately generated by selection bias. 1 Thus, the most reasonable explanation for the highly significant improvements in retinal NFL and GCC layer thickness trends in the VPT group is VPT treatment. Although novel, the observations we report are explicable based on known science and consistent with improvements in electrophysiology and visual function in OAG eyes following SDM. These suggest the possibility of clinically therapeutic CNS neuroregeneration. Such a neuroregenerative response might be cell-mediated and/or reflect epigenetic disinhibition of CNS neuroregeneration. If so, it is possible that similar responses might be produced in other organs, including other CNS tissues, in the same way by appropriate means. 14 Confirmed, the results of the current study would have important implications for the future management of OAG and other CPRs, and neurologic injuries and neurodegenerations in general. 67–69