Unformatted Attachment Preview
Tissue-Engineered Solutions for
Cell Replacement Therapy in
Age-Related Macular Degeneration
BIOE689Y Fall 2017
Dr. Martha Wang
Mostafa Lotfi, Niloofar Massahi, Shannon Schreiner (Team 4)
Fischell Department of Bioengineering
“We pledge on our honor that we have not given or received any
unauthorized assistance in this assignment”
Abstract
According to the World Health Organization, adult macular degeneration (AMD) ranks third among global
causes of visual impairment with a blindness prevalence of 8.7%. AMD is the p
rimary cause of visual
deficiency in industrialized countries and affects 2 million Americans.(Nazari et al.) As RPE tissue does
not have regenerative capacity and the root cause of AMD is unknown at this time, research is focused
on treatment strategies which include the use of human embryonic stem cell derived retinal pigmented
epithelium (hESC-RPE). This proposal seeks to deliver a monolayer of hESC-RPE to the ocular
subretinal space via porous, biodegradable, ultra-thin parylene scaffold, impregnated with an anti-VEGF
therapeutic, to repair existing RPE damage while simultaneously preventing future neovascularization.
Implant viability will be evaluated based on cytotoxicity, including any adverse mechanical effects that
may result from its subretinal placement, implant degradation rates, and suppression of
neovascularization of the choroidal layer through two potential anti-VEGF drugs, bevacizumab and
ranibizumab.
Clinical Need
There are two types of adult macular degeneration (AMD): dry (non-neovascular) and wet (neovascular).
Both forms of AMD are characterized by discoloration of the retinal pigment epithelium (RPE), a buildup
of cellular debris known as a drusen formation, followed by a thickening of Bruch’s membrane.(Birch and
Liang)Early stage indicators and risk factors for the two types of AMD are similar, but late stage
development and presentation are distinct.
Dry AMD is characterized by drusen formations in the back of the eye.(Birch and Liang) Drusen
formations in patients with AMD, although present in normal eyes as well, have entered a vicious cycle
and accumulated to form larger, more rigid aggregates.(Sarks, Sarks, and Killingsworth) In these
regions, drusen eventually fade away and are replaced by a patch of dead RPE termed geographic
atrophy (GA).(Nazari et al.) The primary challenge in the treatment of dry AMD is a lack of knowledge of
its root cause and although a variety of risk factors have been identified, currently there is no consensus
regarding the specific cause of AMD.(Nazari et al.)
Wet AMD is characterized by the formation of blood vessels below the retina.(Kaarniranta et al.) The
barrier function of the new vessels is compromised and results in leakage of plasma to form edematous
regions under the retina or into the eye.(Kaarniranta et al.) It is known that inappropriate regulation of
vascular endothelial growth factor (VEGF) is responsible for the growth of these new vessels.(Nazari et
al.)1 Management of VEGF is a strategy for slowing down the progression of wet AMD but does not stop
the inevitable result.
Figure 1. Fundoscopy of eyes with regional hyperpigmentation and hypopigmentation
where GA will occur (Bird et al.)
Current Strategies for Treatment
Dry AMD Treatment
At this time, there are no clinical treatment options for dry AMD as the root cause is unknown (Rickman,
C.B., LaVail, M.M., Anderson, R.E., Grimm, C., Hollyfield, J., Ash) however, there are a variety of cell
1
therapy treatments being investigated. Current research is focused on the use of RPE implants obtained
from either human embryonic or induced pluripotent stem cells by either bolus injection of RPE cell
suspension or via prefabricated RPE stem cell sheets implanted into the subretinal space. (Olmos et al.,
2015) According to Vugler at al, in vitro experiments demonstrated hESC-RPE could form a monolayer
of regular, polarised cells which demonstrated apical and basal regions and expressed markers of both
developing and mature RPE.(Vugler et al.) Additionally, characteristics of RPE cells cultured on collagen
membranes have been shown to develop similar differentiation indicators to RPE cells in vivo.(White
and Olabisi)
Experimental evidence indicates hESC-RPE has a survival period of 12 months in immunocompromised
laboratory models and a polarized monolayer of hESC-RPE had a higher survival rate than suspensions
without a teratoma or ectopic development.(Diniz et al.) This suggests the probability of a positive
association of hESC-RPG in human retinal subjects as well. Due to the possibility of implant rejection,
immune response needs to be considered, and immunosuppressant medications required for these
therapies(Olmos et al.). Experimental evidence has also indicated favorable survival rates, and
functionality of hESC derived RPE in the form of monomer on polymer substrates (rCPCB-RPE1
cells/pluripotent stem cells) in cultured cells in rats.(Thomas et al.; Tian et al.; Thumann et al.)
Wet AMD Treatment
Since it has been shown that VEGF regulation is important in the progression of wet AMD,
pharmaceutical companies have begun producing anti-VEGF treatment options. Examples of these
anti-VEGF drugs are Pegaptanib (ss nucleic acid), Ranibizumab (monoclonal antibody), and Aflibercept
(recombinant fusion protein).(Avery et al.) Stated briefly, all of these treatments work by binding or
blocking VEGF-A or a combination of VEGF isomers.(Nazari et al.) All 3 are FDA approved for use in the
treatment of wet AMD and require monthly or semi-monthly injections of the agent into the eye.(Avery et
al.) The motivation behind these treatments is, if the rate of angiogenesis can be decreased, patients will
have a better prognosis; however, this strategy requires early diagnosis. In this same vein as the above
treatments, a cursory search of ClinicalTrials.gov revealed there are many co-treatment trials where
anti-VEGF therapies have been paired with anti-platelet derived growth factor (anti-PDGF) which is
important in the formation of blood vessels.
Macular translocation and RPE transplantation surgery have also been options for patients suffering
from AMD.(Steven D Schwartz et al.) Both of these surgical interventions demonstrate healthy RPE
tissue can have vision restorative capabilities which have been demonstrated to last at least 5
years.(Oshima et al.) However, these surgical options can be technically challenging and may result in
vitreoretinopathy or torsion. (Nazari et al.)
Current Investigated Tissue Engineering Strategies for Treatment
hESC-RPE derived
Transplantation of human embryonic stem cell - derived retinal pigment epithelium (hESC-RPE) has
been demonstrated as an effective and promising strategy in the treatment of AMD. In a study published
in 2016, Schwartz, et al. produced results that indicated it was possible to implant hESC-RPE in patients
suffering from dry AMD resulting in restoration of function of both photoreceptors and visual function.
(Steven D. Schwartz et al.) These conclusions were based primarily on the tolerance and safety of the
hESC-RPE cells and secondarily on the efficacy of the transplanted cells.
In this study, 18 patients (9 dry AMD, 9 STGD) were injected sub-retinaly with a target dose of
resuspended hESC-RPE. Each group of 9 patients was further subdivided into 3 cohorts, each receiving
a differing number of hESC-RPE cells: 50,000, 100,000, or 150,000. The patients were then given a 6
week regimen of immunosuppressants including tacrolimus and mycophenolate mofetil; after the first 6
weeks had elapsed, the tacrolimus was discontinued and mycophenolate mofetil was continued to be
administered for an additional 6 weeks. As a follow up to the procedure, all patients were serially
observed for overall changes in health as well as changes in visual capabilities. To deem the procedure
as safe, patients were physically examined, had their vitals taken and underwent electrocardiograms,
2
hematological and serological testing as well as cancer screening. Although no patient in the study
suffered an intraocular or systemic event as a result of the hESC cells, a number of patients were
observed to have adverse effects as a result of the immunosuppressants administered.
To examine the visual outcomes, the patients were examined for best-corrected visual acuity (BCVA),
visual fields, slit-lamp biomicroscopy, ophthalmoscopy, OCT, fluorescein angiography, autofluorescence,
fundus photography and electroretinography. Thirteen out of 18 patients exhibited an increase in
subretinal pigmentation and at the 6 month mark, 9 out of the 18 eyes demonstrated stable or better
visual performance on a visual acuity test.
While these results are promising, the use of hESCs is not without risk or controversy. Although hESCs
are not sourced from an abortion, and do not require embryonic loss, their use remains contentious.
Additionally, the use of hESCs can result in the formation of a teratoma, unwanted differentiation, and
immunologic rejection.
iPSC-RPE derived
In 2006, Yamanaka et. al. showed that by overexpressing 4
factors (OCT3/4, Sox2, c-Myc, Klf-4) fibroblasts can be induced
into an embryonic-like pluripotent state.(Takahashi and
Yamanaka) These cells were called induced pluripotent stem
cells (iPSC) and are another renewable source of stem cells that
can be differentiated into a variety of tissues. Also, because
these cells can be generated from readily available fibroblasts,
this method opens up the potential for creating autologous
transplant where a patient’s own cells can be reverted to the iPS
state, differentiated, and then implanted back into the patient.
There have been two primary methods of utilizing iPSC-derived
RPE (iPSC-RPE) in a cure for AMD. Both involve the
transplantation of iPSC- RPE in the eye at locations where later
stage macular degeneration has caused the loss of the RPE
layer. The first method is the transplantation of a suspension of
RPE cells at the back of the eye.(Westenskow et al.) The
second method is the transplantation of a monolayer of iPSC-RPE on a scaffold at the back of the eye.
Both methods have drawbacks and design challenges.
In the injection of iPSC-RPE suspension, the cells are cultured, differentiated, loaded into syringe, and
injected into the eye. There are a few similar culturing methods available in literature, the details of which
are beyond the scope of this paper however, it is necessary to note that in culturing iPSCs the purity of
the final cellular product is an omnipresent hurdle regardless of the cell type being produced. In both
methods, the integration of the cells into the eye is a concern that needs to be addressed. In the
suspension-injection method integration is optimized by placing the cells at a very specific location in the
back of the eye. Unfortunately, this requires expert surgical skills and misplacement is
common.(Westenskow et al.)
In the iPSC-RPE monolayer placement method, a single layer of cultured RPE is grown on a scaffold
and placed in the back of the eye.(Bird et al.) Many of the challenges in the injection method are
applicable here as well. Purity is a concern, placement can be a hassle, and this method has the added
challenge of a scaffold. This method has many advantages over the bolus approach which include: a
rigid scaffold for easier placement of the cells in the desired location and the ability to analyze the cells in
their macro-structural formation (as a monolayer sheet).
Photoreceptor replacement
There are several transplant strategies that meet the requirement for a successful photoreceptor
replacement therapy however, in all cases, donor cells must be able to migrate to the recipient retina and
differentiate into a functional and connected photoreceptors.(Westenskow et al.) Whole retinal sheets
3
from embryonic and neonatal retinae have the ability to differentiate. In a study completed by Radtke et
al, scientists saw an improvement in vision in 7 out of 10 AMD patients after receiving fetal retinal sheet
transplants.(Pearson) Transplantation of brain derived neural stem cells into the neural retina is another
possibility that researchers have investigated. However, these integrated cells do not show intrinsic
characteristics of mature retinal neurons.(Iyer et al.) To overcome this issue stem cells from embryonic
retinas can be used as they are able to differentiate into glial cells and cells expressing specific
photoreceptors.(Pearson) However, they are not able to integrate into the laminar structure of retina.
Successful transplantation also depends on the developmental stage of the donor cell. This was seen in
the study completed by MacLaren et al (MacLaren et al.) where postnatal donor cells were transplanted
into the recipients at the exact developmental stage. The cells migrated and matured into normal
photoreceptors.(Qiu et al.)
One of the challenges faced after transplanting photoreceptors is their ability to improve vision. A study
completed by Pearson et al suggests that there is a direct correlation between the number of integrated
cells and the amount of vision restored.(MacLaren et al.) The researchers also used gene replacement
therapy to show that 120,000 functioning photoreceptors are needed to produce scotopic
electroretinogram response.(Pearson et al.)
The described strategies for photoreceptor replacement are derived from postnatal retina which occurs
in the second trimester in humans and therefore supply is limited. Researchers are investigating
reproducible sources of photoreceptor precursors. Embryonic stem cells derived from the inner cell
mass of a preimplantation blastocyst have also shown great promise in their ability to generate retinal
cells. In a study published by Eiraku et al, researchers were able to develop a 3D culture in which the
ESCs mimicked retina morphology.(Eiraku et al.) The reprogramming of nuclei of somatic cells, to
pluripotent state, can bypass the problems associated with the use of human embryos. These induced
pluripotent stem cells (IPSs) should have similar capability as ESCs; in a study published by Lamba et
al, investigators showed that the iPSCs can generate photoreceptor like cells but can have some
resistant to retinal differentiation.(Lamba et al.)
The realistic prospect of photoreceptor transplantation presents an important landmark in therapies for
patients suffering from AMD as these cells are able to integrate into recipient retina and restore vision.
Project Aims
Age related macular degeneration, which stems from a loss or damage to the Retinal Pigmented
Epithelium (RPE), is the leading cause of visual disability in the western world.(Idelson et al.; Oshima et
al.) RPE tissue is characterized by a monolayer of hexagonal cells, which are filled with pigment, and is
located between the highly vascularized choroidal layer of the eye and the neuroepithelium, which
contains rod and cone photoreceptors.(Maminishkis et al.) These hexagonal cells have well defined
apical and basolateral regions as well as tight junctions between cells which help to form a blood-retinal
barrier.(Peyman, Spitznas, and Straatsma) The RPE’s main functions are: phagocytosis of rod and cone
outer segment fragments, shed from their distal ends; uptake, processing, transport, and release of
vitamin A (retinol); regulation of net ion and water movement in the apical to basal direction; absorption
of stray light, thus improving visual images; and to aid in the attachment of the retina to the choroidal
layer.(Idelson et al.) Derived from the neuroectoderm, RPE, once differentiated, does not self-renew and
must remain functional for the lifetime of the individual. (Klimanskaya et al.; Bok)
In the eyes of patients with “wet” AMD, neovascularization of the choroidal layer, through Bruch’s
membrane and into the subretinal space, leads to leakage of blood and serum, causing a severe
decrease in central vision.(Maminishkis et al.) This neovascularization has been attributed to an
overabundance of Vascular Endothelial Growth Factor present in the region.(Spilsbury et al.) If the RPE
below the fovea suffers sufficient damage, potential medical interventions are few in number and have
limited effectiveness in the treatment of this degenerative disease.(Oshima et al.) Existing therapeutic
treatments of AMD, which include intravitreal injections of anti-vascular growth factor, and triamcinolone,
aim only to slow or halt the progression of neovascularization; they lack the ability to reverse the loss of
best-corrected visual acuity (BCVA).(Idelson et al.) However, recent developments have indicated new
treatment strategies are on the horizon.
4
The use of Human Embryonic Stem Cell RPE (Idelson et al.) (hESC-RPE), to replace dysfunctional
RPE, has been demonstrated to be a viable method to restore visual function in those with AMD.
(Maminishkis et al.; Steven D. Schwartz et al.; Song et al.) Participants of clinical studies have shown
stabilization or improvement of visual acuity, while simultaneously supporting the safety of the use of
hESCs in this application.(Steven D. Schwartz et al.; Song et al.) Therapeutic replacement hESC-RPE
can be delivered to the intraretinal space via cell suspension or as cell layer(s) adhered to an
implantable material. Multiple studies have demonstrated that the use of a monolayer of hESC-RPE has
an outcome superior to that of a cell suspension.(Diniz et al.; Thomas et al.) There is a wide variety of
materials to which the hESC-RPE could be adhered.(Montezuma et al.) In studies investigating the
intra-retinal implantation safety of parylene-C, a poly(p-xylylene) polymer, outcomes indicate this
material demonstrates a low incidence of histological disruption.(Montezuma et al.; Yu et al.; Wang et
al.) 15,16,17
Once in place, the replacement RPE should be protected from damage associated with
neovascularized AMD. Anti-VEGF therapies, such as bevacizumab and ranibizumab, have been shown
as equally effective to one another in the prevention of the formation of new vasculature as a treatment
for wet AMD.(Martin et al.; Berg et al.; Told et al.) In both instances, patients who received anti-VEGF
treatments displayed gains in visual acuity, especially when therapeutics were delivered on a scheduled
versus as-needed basis.(Martin et al.; Berg et al.; Told et al.)
The global hypothesis of this proposed work is the delivery of a monolayer of hESC-RPE to the ocular
subretinal space via a porous, biodegradable, ultra-thin parylene scaffold, impregnated with an
anti-VEGF therapeutic, will repair existing RPE damage while simultaneously preventing future
neovascularization. Specifically, we hypothesize:
1.
2.
3.
Parylene-C film is an amenable scaffold for hESC-RPE while demonstrating biocompatibility
with hESC-RPE, choroidal, and retinal tissues
Scaffold degradation rates can be controlled by the physical dimensions of the implant
Neovascularization will be suppressed by the delivery of an anti-VEGF therapeutic to the local
tissues via the implant
To address these hypotheses, the following specific aims are proposed:
1.
2.
3.
Investigate implant toxicity i.e. genotoxicity, carcinogenicity, blood response, and cytotoxicity.
Reproductive toxicity will not be investigated as AMD patients are past their reproductive
window. Implant will also be investigated for any adverse mechanical effects that may result
from its subretinal placement.
Investigate implant degradation rates by varying the radial dimension, thickness, porosity, and
pore size of the implant.
Investigate bevacizumab and ranibizumab for in vivo suppression of neovascularization of
choroidal layer
Research Design & Aims
In order to treat conditions like AMD, via hESC-RPE, it is essential to verify the efficacy of the different
approaches in animal models prior to initiating human trials. A mouse model is an efficient analog for
testing the viability of hESC-RPE for potential carcinogenicity, blood pathology, genotoxicity, etc.
However, there is a lack of information on the use of these methods in human models and instances of
human trials are absent. Moreover, the evidence collected to date indicates potential changes in stem
cells as well as their microenvironments. There is also little data regarding the cell-intrinsic stem cell
changes, as well as the potential toxicity in human implants.(Piccin and Morshead) Although hESCs are
considered pluripotent, they appear to demonstrate a predisposition for specific cell lineage
differentiation. An additional, significant hurdle in the use of hESCs is their potential to form teratomas or
other neoplasms; preclinical assessment of this tissue is required to mitigate the long-term tumorigenic
potential of hESC-RPE.(Cunningham et al.) The development of proliferative vitreoretinopathy (PVR),
macular pucker or macular hole, as well as crust or scar tissue development can lead to loss of visual
acuity due to an implant.(Alexander et al.) Due to these varied risks, all research needs to be validated
by experimentation, and in depth analysis conducted, to ascertain a Parylene-C film, coated with a
5
monolayer of hESC-RPE and anti-VEGF growth factors, is an amenable, biocompatible, and effective
scaffold, which can be used for the treatment of AMD.
In Silico
Phase I of this research will focus on the theoretical optimization of the dimensions of the implant that
will replace the dysfunctional RPE. This investigation will examine how implant degradation rates are
affected by varying the radial dimension, thickness, porosity, and pore size of the implant. Determination
of the range of the radial dimension is based on documented values obtained for a mouse retinal region,
plus or minus 10%. The central value for the radial dimension will be 0.75mm,(Remtulla and Hallett)
thus resulting in a modeling radial range of 0.675 – 0.825mm. Ideally, the value to be used for the
thickness of the parylene-C implant would be rooted in experimental values obtained for the basement
membrane thickness of 54nm.(Cuthbertson and Mandel) However, as this thickness would prove
extremely challenging to manufacture, coat with cells, and physically implant in a subject’s eye, the
thinnest value of the implant to be modeled will be based on actual material availability of
0.2µm(Parylene Engineering), and previous studies which indicate ranges of values of 0.15-0.30µm
approximate the permeability of Bruch’s membranes.(Lu et al.) The range of values to be modeled for
the thickness parameter will represent a variability of baseline to baseline plus 25% or 0.2 – 0.25µm.
The third parameter to be considered in the in silico model is the size of the pores in the parylene-C film.
The desired range would be too small to support neovascularization but large enough to easily
accommodate the transport of nutrients and waste. The range to be modeled is representative of a
manufacturing reproducible minimum porous membrane array (PMA) which has pore diameters of
110nm(Thakar et al.) to 10x this value, or 1100nm (1.1 µm) which is well below the minimum diameter
value of rodent or other mammal’s capillaries(Iatridis) of 3µm. Finally, the last parameter to be modeled
is the porosity of the scaffold; the optimal value would accommodate the cell density required to have
functional RPE while also being sufficient for waste/nutrient transport. Based on the data presented in a
study completed in 2002 by Del Priore, Kuo, and Tezel(Del Priore, Kuo, and Tezel), RPE cell density
should meet or exceed 4980 ± 90 cells/mm2.(Iatridis) With a given hexagonal diameter of 12µm,(Bhatia
et al.) each RPE cell would occupy 124.77µm2, thus in order to maintain the 4980cells/mm2 threshold,
0.62135mm2 of every square millimeter of scaffold would need to be present. Thus, porosities would be
modeled for values ranging from 0-35%. The vitreous humor turnover rate will be modeled using values
obtained by Laurent and Fraser(Laurent and Fraser) indicating diffusion controlled transport through the
vitreous body; rate constants for components with high molecular weights (>500,000) will be set to
0.024/d, mid-range molecular weights (~18 000) to 0.16/d and low molecular weight (