PREVENTION OF REGRESSION IN REFRACTIVE KERATOPLASTY
BACKGROUND OF THE INVENTION
All methods used to alter the anterior corneal radius of curvature for the correction of refractive error are collectively known as "refractive keratoplasty." Refractive keratoplasty is based upon the hypothesis that the cornea is responsible for about eighty per cent (80%) of the light bending power of the eye. Nominal modification of the front surface radius will, therefore, result in marked changes in the focusing power of the cornea. The correction of nearsightedness (myopia) requires that the front radius be lengthened or flattened, while steepening of the central cornea, or shortening of the radius, is performed for the correction of farsightedness (hypermetropia).
Various methods have been developed to change the radius of curvature of the cornea. For example, refractive keratoplasty can be surgical, laser-assisted, and conductive. Laser- assisted procedures include laser in situ keratomileusis (LASIK), photo-ablative refractive keratectomy (PRK), and laser thermal keratoplasty (LTK). PRK and LASIK have in common the use of the excimer ultraviolet laser for photodisruption and the removal of corneal stromal collagen. Each method results in epithelial damage and apoptosis as well; PRK requires that the epithelium be removed entirely in the area of ablation, while LASIK-induced epithelial damage results from the microkeratome that is used to cut a thin corneal flap beneath which the stroma is ablated. Radiofrequency keratoplasty and laser thermal keratoplasty each cause thermal epithelial trauma and subsequent apoptosis. The stabilizing heat-labile hydrogen bonding of the collagen fibrillar triple helices are also destroyed by the thermal keratoplasty method. All of the techniques lead to stromal remodeling.
The advantage of laser light in the treatment of various types of tissue is that its monochromatic, high energy beam can be focused and manipulated to obtain specific photobiological effects. Irradiation exposure parameters can be matched to specific physical, chemical and biological properties of the target tissues to get a desired result. Tissues can be defined by their (1) optical properties (absorption, scattering and scattering an isotropy), (2) thermal properties (heat capacity and heat diffusivity), (3) mechanical properties (viscoelasticity, tensile strength and rupture points), (4) chemical composition (water and other endogenous and exogenous absorbers), (5) anatomy (physical arrangement of organelles, cells and tissues) and (6) physiology (tissue and organismal metabolic status and function). Depending upon the irradiation conditions and the desired endpoints, some properties will dominate over others as the major determinants of the final effects of the laser-tissue interaction.
For example, lasers emitting in the infrared domain of the electromagnetic spectrum interact with tissue with a photobiologic effect which is substantially photothermal. Photothermal effects result from the transformation of absorbed light energy to heat, leading to contraction, coagulation or destruction of the target tissue. The nature and extent of photothermal effects of laser-tissue interactions are governed by (1) the distribution of light within the tissue, (2) tissue temperature, (3) duration of time the tissue is maintained at temperature and (4) the tissue's thermal properties, diffusivity and heat capacity. These factors are collectively called the "thermal history" of the tissue.
As the tissue temperature approaches the threshold temperature of vaporization of water (100°C), the photothermal effects of the laser-tissue interactions come under the influence of the energy requirements of the phase changes of the water, tissue desiccation, formation of steam vacuoles within the tissue, and the mechanical effects of the rapidly expanding steam vacuoles trapped within the tissue.
Water strongly absorbs light at 2000 nanometers, leading to rapid vaporization of water. Tissue desiccation radically changes the optical characteristics of tissues, especially their absorption characteristics of infrared laser irradiation. In addition to the optical property changes, water loss reduces the thermal conductivity and specific heat of the tissue. Tissue
"thermal history" is a dynamic function and must therefore be constantly monitored in order to attain accurate refractive correction. Each method of corneal recurvature requires some mode of traumatic intervention, although some methods are more invasive than others. As one creates ever-more complex radius modification upon the anterior surface of the cornea, the trauma is compounded. In view of this trauma, the efficacy and predictability of each method becomes an important issue. Although these methods effect a change in the refractive characteristics of the cornea, an ongoing problem with all refractive methods is the subsequent regression that leads to a loss of the desired refractive effect over time. The tendency of the cornea to reverse the desired effects generated by the keratoplasty is referred to as "refractive regression." This regression response has frustrated attempts to design procedures that are both efficacious and predictable, while remaining safe. Efficacy is intrinsically related to the change in tissue structural bio-mechanics induced by the trauma. US Patent No. 6,033,396 describes a method of performing LTK using laser impact shapes that reduce regression, but does not address cellular responses after surgery leading to regression.
Observations of the post-operative corneal radius creep and deformation have elicited an understanding of the changes that affect the stability and predictability of desired effect. Some of the changes seen in regression are paradoxical, while others appear to be directly related to the
type of trauma sustained. The molecular and cellular biological responses manifested by these processes are the result of several separate, integrated events. Some of these events are cellular apoptosis, cellular necrosis, accumulation of stress or heat shock proteins, and destruction of the lamellar-stabilizing molecular cross-links. It is known that corneal surgical procedures including refractive keratoplasty procedures will elicit a wound healing process that can cause scarring of the cornea and impair vision through corneal opacity. Treatments have been developed that are directed to reducing the fibrosis that accompanies wound healing and induces subsequent corneal scarring. Such treatments include antimetabolites, immunosupressants, corticosteroids, and growth factors such as TGF-β. However, it is only now being considered that this comeal wound healing response is also responsible for regression of the desired refractive effect.
A need clearly exists to find a way to prevent the regression that characterizes so many post-refractive procedures, in order to maintain the visual improvements brought about by the refractive intervention.
SUMMARY OF THE INVENTION
The methods and compositions of the present invention provide therapeutic agents designed to accurately intersect with certain stages of the corneal wound-healing process and inhibit the repair mechanisms that contribute substantially to refractive regression. A method of thermal refractive keratoplasty is featured which prevents or inhibits subsequent loss of refractive effect that is caused by thermally induced corneal epithelial apoptosis. The method comprises applying thermal energy to the cornea such that the energy increases sub-epithelial stromal temperature in at least one selected location sufficient to induce hydrothermal collagen fibrillar shrinkage and induce refractive changes, and cooling the corneal epithelium prior to, subsequent to, or concurrent with the application of thermal energy, so that any heat generated by the energy to the cornea that radiates to the epithelium is unable to increase the epithelial temperature in at least some epithelial cells to a level sufficient to induce apoptosis so that fewer epithelial cells die than would without cooling. The thermal energy used is preferably photothermal coherent energy or radiofrequency conductive energy, and the cooling is preferably achieved by spraying a cryogenic solution onto the corneal surface or placing a cooled corneal contact device onto the cornea. The corneal contact device is preferably fabricated of a material having thermal mass sufficient to remain cooler than ambient temperature for the duration of thermal energy application and which transmits about 100% of mid-infrared irradiation, such as sapphire or quartz. The corneal device can be pre-cooled or cooled in place on the cornea by spraying a pharmaceutically acceptable cryogen upon it, such as 1,1,1,2
tetrafluorethane. Preferably, the corneal contact device is fabricated with a base radius of curvature flatter than the apical radius of curvature of the comea upon which it is placed, so that it remains in close contact with the cornea and water does not pool beneath it. An advantage of this method is that by preventing corneal epithelial apoptosis, it thus prevents release of cytokines from the epithelium which permeate into the stroma and induce apoptosis of corneal keratocytes, a cause of stromal remodeling.
Also featured is a method of preventing or inhibiting a loss of refractive effect in a patient who is undergoing or has undergone a refractive keratoplasty procedure, where the loss of effect is caused by stromal remodeling. This method comprises interrupting at least one step in the stromal remodeling response of the cornea elicited by the refractive procedure by providing an effective amount of a composition that modulates the step, where the interrupted step results in inhibition of epithelial apoptosis, inhibition of keratocyte apoptosis, inhibition of keratocyte necrosis, inhibition of keratocyte proliferation and migration into the wounded cornea, inhibition of glycosaminoglycans synthesis, or inhibition of collagenesis. The refractive procedure used is preferably ablative refractive keratoplasty or thermal refractive keratoplasty. The step inhibited can be the cascade associated with Connective Tissue Growth Factor (CTGF) or biologically active fragments thereof, and can be an antibody or fragment thereof, a nucleic acid or other molecule which inhibits expression of the gene or protein, or one which blocks a receptor of CTGF. Alternatively, the composition can be an antibody/fragment, chemical compound, or nucleic acid that inhibits the accumulation of heat shock protein-70 and result in an inhibition of collagen assembly to its properly folded state, or inhibits the effects of C- proteinase in stromal remodeling, or the effects of prolyl hydroxylase. Preferred inhibitors of prolyl hydroxylase include FG-1648 and FG-041 (proprietary compounds of Fibrogen, Inc., San Francisico, CA). An advantage of this method lies in that it disrupts one or more phases of the wound repair cascade pharmacologically without disrupting all phases of fibrosis and wound repair, thus selectively preventing specific causes of loss of refractive effect.
Another featured method of preventing or inhibiting a loss of refractive effect in a patient who is undergoing or has undergone a refractive keratoplasty procedure comprises applying an effective amount of a composition that creates or restores cross-links between stromal lamellae, where the cross-links increase the stability of the corneal stroma and cause it to hold the corneal shape induced by the refractive keratoplasty procedure longer than it would have if the composition had not been applied. Compositions of the method preferably create or restore hydrogen bonds, electrostatic forces, or covalent bonds between lamellar fibers. A preferred composition is glycerol, which requires no heat or light activation to induce cross-links. Other preferred compositions include photoactivatable substances, and when these are used, the method
further comprises applying photoirradiation to the cornea after application of the composition to activate the composition and create the cross-links. A preferred photoactivatable composition is riboflavin, which is activated by light energy in the ultraviolet spectrum. Other preferred compositions are activated by heat to form cross-links. An advantage of this method is that it is effective regardless of whether the destruction of the stabilizing molecular cross-links result from ablative refractive keratoplasty that has mechanically compromised the corneal stroma, or from thermal destruction of the heat-labile stromal cross-links.
Another feature of the invention is a composition for preventing or inhibiting refractive regression of the cornea following a refractive keratoplasty procedure, where the composition comprises an agent which inhibits the deposition of collagen matrix in response to keratocyte apoptosis or necrosis and an excipient capable of combining with the agent and carrying it across the corneal epithelial barrier and into the corneal stroma, and which is physiologically compatible with the eye. Preferably, the composition does not affect keratocyte proliferation or migration.
In preferred embodiments, the composition further comprises a topical ocular insert which serves as a reservoir and delivery device for the agent. Preferred inserts include pleggets, polymer contact lenses soaked in the agent, collagen shields impregnated with the agent, and liposomes, flexible capsules, or flexible wafers containing the agent, or other membrane or reservoir systems. The agent and excipient are preferably efficacious and bioavailable in doses practical to administer to the eye, and are non-toxic and physiologically compatible with the eye. Preferably, the agent is an antibody, a small molecule, or a nucleic acid, more preferably an antibody against CTGF or a fragment thereof, an antibody against prolyl hydroxylase or a fragment thereof, an antibody against hsp-70 or a fragment thereof, or an antibody against C- Proteinase or a fragment thereof.
Another feature of the invention is a composition for preventing or inhibiting refractive regression of the cornea following a refractive keratoplasty procedure which comprises an agent that creates or restores intermolecular cross-links between lamellar fibers in the corneal stroma, and an excipient capable of combining with the agent and carrying it across the corneal epithelial barrier and into the corneal stroma, and that is physiologically compatible with the eye. Preferably, the agent comprises glycerol. In preferred embodiments, the composition further comprises a topical ocular insert which serves as a reservoir and delivery device for the agent. Preferred devices include pleggets, polymer contact lenses soaked in the agent, collagen shields impregnated with the agent, and liposomes, flexible capsules, or flexible wafers containing the agent, or other membrane or reservoir systems. The agent and excipient are preferably efficacious and bioavailable in doses practical to administer to the eye, and are non-toxic and physiologically compatible with the eye. Preferably, the agent is an antibody, a small molecule,
or a nucleic acid
The invention also features a method of measuπng a change in comeal πgidity compπsing applying a first weight to a cornea using an indentation tonometry device to obtain a first tonometπc scale reading, applying a second weight of different value to the cornea using the indentation tonometry device to obtain a second tonometπc scale reading, wherein the difference between the first and the second tonometπc scale readings is indicative of a first particular comeal πgidity then waiting for a selected time or for a selected event to transpire, and re-applying the first weight to the cornea using the indentation tonometry device to obtain a third tonometπc scale reading, and re-applying the second weight to the cornea using the indentation tonometry device to obtain a fourth tonometπc scale reading, where the difference between the third and the fourth tonometπc scale readings is indicative of a second particular comeal πgidity, and where a difference between the first particular comeal πgidity and the second particular comeal rigidity is indicative of a change in comeal rigidity In preferred embodiments, the selected event is a refractive keratoplasty procedure or an application of a lamellar cross-linking therapy
A different embodiment of the comeal ngidity measure involves applying an indentation tonometry device to a cornea to obtain a first reading, on a tonometnc scale, applying an applanation tonometer to the cornea to obtain a second reading, of intraocular pressure in rnrnHg, where the difference between the first and the second readings is indicative of a first particular comeal πgidity After waiting for a selected time or for a selected event to transpire, the indentation tonometry device is re-apphed to the cornea to obtain a third reading, on the tonometπc scale, and the applanation tonometer is re-applied to the cornea to obtain a fourth reading, of intraocular pressure in mmHg, where the difference between the third and the fourth readings is indicative of a second particular comeal rigidity The difference between the first particular comeal rigidity and the second particular comeal πgidity is indicative of a change in comeal πgidity
DESCRIPTION OF THE DRAWINGS
These and other features and advantages of the present invention will be better understood by reference to the following detailed descπption and accompanying drawings wherein
Figure 1 is a graph showing the overlapping phases of wound repair, where (A) is the inflammation stage (early and late), (B) is re-epithehazation and granulation tissue
formation, and (C) is matrix formation and remodeling. Collagen accumulation begins shortly after the onset of granulation tissue formation.
Figure 2 is a schematic of the formation of hydroxylated procollagen from nonhydroxylated procollagen catalyzed by prolyl hydroxylase.
Figure 3 is a schematic of the formation of collagen fibers following the cleavage of the C and N terminals of the procollagen molecule by C-Proteinase and N-Proteinase, respectively.
Figure 4 illustrates the manner in which hydrogen bonds form between α chains in the tropocollagen molecule and provide stability to the triple helical structure. Heat above 60 °C will denature collagen and cause the collapse of the helical configuration and the separation of individual α chains into random coil configuration. The intramolecular cross link is unaffected and the two chains remain linked in the denatured state.
Figure 5 shows the intermolecular cross-linking between collagen molecules. At the top, the cross-link is shown as bonding a lysine in the amino terminal of a torpocollagen molecule with a hydroxylysine near the carboxyl terminal of another tropocollagen molecule. In the lower part of th efigure are the proposed reactions leading to stable intermolecular cross-links.
Figure 6 shows intramolecular cross-linking. At the top is shown amino acid sequences of the amino ends of the three peptide chains of the tropocollagen molecule. A cross-link between lysines in an αl and the α2 chain is shown. Postulated steps in the formation of this cross-link are shown in the lower figure.
Figure 7a-7c show various methods of instilling therapeutics into the eye, such as by irrigation (7a), periodic irrigation via drops (7b), and aerosol spraying (7c).
Figure 8a-8d demonstrates application of dmg reservoirs such as soft contact lenses (8a), pledgets (8b), sponges (8c), flexible capsules, wafers, or other membrane or reservoir systems (8d).
Figure 9 illustrates subconjunctival injection.
Figure 10 details a method for evaluating comeal rigidity and a protocol to indicate the success of remedial cross-linking of the stroma.
Figure 11 schematically depicts the cellular events associated with various refractive modalities.
DETAILED DESCRIPTION OF THE INVENTION
It has been discovered that prevention of epithelial apoptosis and/or modulation of the comeal wound healing cascade at selected points, particularly stromal remodeling, can inhibit or prevent the process of refractive regression following refractive keratoplasty. The wound and wound repair responses affected by the methods and compositions of the invention include the epithelial and keratocyte apoptosis and necrosis, and the disruption of the stromal lamellae of the cornea. These processes lead to loss of the desired refractive effect, and prevention of any of the processes, or particular manifestations, such as the accumulation of heat shock proteins, fibrillogenic response of wound repair, or repair of stromal cross-links, can prevent or inhibit the loss of refractive correction effected by the refractive keratoplasty. In addition, a novel method of thermal quenching during refractive procedures that induce thermal damage is disclosed that reduces the apoptosis associated with thermal injury, thus reducing the subsequent wound healing response. This method, alone or in combination with the wound healing inhibition and remedial cross-linking methods disclosed herein, is designed to reduce the amount of refractive regression experienced by patients who have undergone refractive procedures.
An important component of invention is concerned with regression accompanying laser and radiofrequency conductive refractive keratoplasty. The biological responses to trauma and the wound repair cascade following these thermal refractive procedures are becoming better understood, as is their relationship to refractive regression (Figure 1 1). Apoptosis and direct cellular necrosis induces the wound repair response in each case. Damage to the stabilizing cross-links also occurs with each method. These processes are induced and integrated during the various techniques for refractive keratoplasty and tend to reverse the desired effects generated by the refractive process.
1. Control of Thermal Injury During Refractive Procedures
In the refractive methods commonly used today, thermal effects are an important means of correcting the anterior comeal slope and refractive properties. Because coupled radiative- thermal effects are present during and after pulsed iπadiation by laser or radiofrequency energy, determination of the optical and/or the thermal properties of the test material can be complex. For example, if the irradiation of the tissue is repeated before the tissue completely cools, the
elevation of the tissue temperature is additive When, however, the repetition rates are fast, the pathophysiological effects of pulsed laser irradiation can be essentially like those produced by continuous wave lasers The pathologic changes of thermal injury can be placed mto temperature-time categoπes in increasing order of seventy The tissue response to injury or the wound repair response thus also vanes
A Methods to Regulate Energy Dose for Optimal Refractive Effects and to Minimize Damage (1) Calcluation of Thermal Effects An understanding of the photothermal tissue effects from lasers emitting in the infrared has led to the study of pulsed photothermal radiometry (PPTR) PPTR is a technique for determining tissue reaction with reference to its thermal properties PPTR has been investigated as an indirect modality for determining the appropπate laser treatment for vaπous tissue, such as skin, tendon and cornea
Photothermal effects are produced within the target tissue when, by means of the appropπate laser exposure parameters, the radiant energy exceeds the threshold required for tissue modification The photothermal changes tngger a biological response which culminates in a complex sequence of events within the madiated tissue These changes may be only represented by a phase transition or may proceed to tissue destruction with a wound healing response and new tissue synthesis In any case, the definitive change will be determined by the magnitude of the thermal response, or the "thermal history" of the tissue.
PPTR is a non-contact method that uses a rapid acting infrared detector to measure the temperature changes induced in a test mateπal exposed to pulsed radiation Heat generated as a result of light absorption by subsurface chromophores in the test mateπal diffuse to the surface and results in increased infrared emission levels at the surface By collecting and concentrating the emitted radiation onto an infrared detector, one obtains a PPTR signal that represents the time evolution of temperature near the test mateπal's surface Useful information regarding the test mateπal (e g comeal or skin tissue) may be deuced from analysis of the PPTR signal
Experiments have been conducted at the Beckman Laser Institute (University of California, Irvine) to determine depth profiles of laser light absorption in skin tissue It has been discovered that strong scattering compared with absorption tends to raise the front surface temperature, as some of the scattered light is absorbed while back-scattering through the front surface If the scattering is a significant event, the radiation transport, temperature distπbution and penetration depth are all pπmaπly dominated by the scatteπng and not by the chromophore absorption.
When the desired endpoint is collagen stress compaction or hydrothermal shrinkage, as in photothermal refractive keratoplasty, an appropriate thermal profile within the therapeutic window must be attained. Collagen phase transition from the tightly wound triple helical form to the relaxed contracted form begins as the tissue temperature reaches about 55° C. If a sufficient volume of tissue does not reach that state, comeal front surface radius changes do not occur. If the temperature exceeds 68° C, collagen denaturation proceeds to destruction. The contracted state then regresses, as does the desired refractive comeal recurvature.
PPTR is a novel method, when combined with infrared pulsed laser irradiation, to determine objective patient-specific laser treatment conditions. When combined with front surface cooling or the conduction of heat away from the front surfaces (discussed below), it provides an even more refined ability to determine and exquisitely control the laser generated thermal profile within the comeal tissues. Superficial tissue cooling processes may be either static as in the form of a simple heat sink, or dynamic as in the addition of an evaporative cooling agent. PPTR permits patient-specific determinations with or without additional cooling capabilities.
One design of the thermal monitoring system embodies the use of a rapid acting thermal sensor that is focused upon the delivery site of the laser. The steady state tissue temperature is obtained and integrated with the known or experimentally determined thermal gradient between the surface and the subsurface target tissue after sub-threshold radiant exposure. Computing electronics within the laser system can automatically determine the appropriate dosimetry.
This system can be applied to any coherent or non-coherent light source, as well as to radiofrequency energy sources used to perform any beneficial treatment within the human body that is dominated by thermal or photothermal biological mechanisms.
When integrated with a dynamic cooling system, when the superficial temperature decrease resulting from evaporative cooling is not within programmed range, as in the case of reduced level of cryogen within the coolant reservoir, the laser operation will not be enabled, providing a fail-safe operation. This control allows the surgeon to effect the refractive change in the desired temperature range without risking collagen denaturation caused by accidental overheating.
(ii) Dynamic Cooling Methods
Inasmuch as thermal injury to the comeal epithelium in LTK and Radiofrequency
Conductive Keratoplasty causes secondary keratocyte apoptosis immediately beneath the irradiated sites, a novel dynamic cooling process is presented to protect the epithelium from damage. This procedure is refeπed to as "thermal quenching." A similar procedure has been
used in aesthetic dermatology to protect the epidermis while an infrared laser achieves thermal denaturation of the proteins in the papillary and reticular dermis to reduce rhytides or superficial cutaneous wrinkles. A modification of this procedure is disclosed that allows the cooling method to be used in refractive keratoplasty. It has been discovered that appropriate application of pulsed laser irradiation and cryogen spray cooling may be used to protect the comeal epithelium and selectively confine the spatial distribution of thermal injury to the stroma. The ability to limit preferentially spatial distribution of thermal injury is supported by histological studies in an animal model (Milner et al. JOSA A
12:1479-1488 (1995); Anvari et al, Phys Med. Biol, 40:241-252 (1995)). In this procedure, the epithelium is rapidly cooled prior to pulsed laser irradiation by a short, gentle spurt of cryogen sprayed upon the comeal surface. Subsequent absorption of pulsed laser energy by water in the comeal stroma results in a temperature increase sufficient to cause spatially selective protein denaturation and phase-transition or shrinkage. Because epithelial temperature at superficial depths is decreased by dynamic cooling, thermal injury in the epithelium is reduce or eliminated.
In an alternate embodiment, this technology can be used to pre-cool a sapphire contact lens or lens , which would transmit 100% of the coherent thermal energy to the stroma while the contact lens acts as a heat sink to conduct heat away from the superficial cornea, thus preventing thermal trauma and the subsequent apoptotic keratocyte death. This process is a combination static-dynamic cooling process which can be temporally integrated with the laser irradiation.
B. Prevention of Apoptosis from Thermal Trauma
There appears to be a correlation between keratocyte apoptosis and comeal thermal truama occurring during refractive surgery. Studies elucidating this correlation have suggested the use of various pharmacologic and mechanical methods in the modulation of this biological response. The goal is to prevent, control or eliminate the apoptosis and the wound healing response that is stimulated by the apoptosis that is responsible for these regressive changes.
This keratocyte destruction precipitates a wound repair process culminating in the synthesis of glycosaminoglycans (GAG's) and collagen. It is desirable to inhibit this process before it is precipitated. Conduction of heat away from the comeal epithelial surfaces will prevent the epithelial destruction during thermal refractive keratoplasty. It is the epithelial damage which causes the epifhelial-stromal interaction precipitating the apoptotic keratocyte death seen in laser refractive keratoplasty.
The process of thermal quenching is applicable to refractive keratoplasty, as discussed above. Various methods of thermal quenching are efficacious and include both passive and dynamic cooling of the comeal epithelial surface.
Static cooling may be accomplished by means of a comeal device which transmits the IR radiation used in the thermal keratoplasty, such as 2 micron THC: YAG irradiation, but has enough thermal mass to serve as an efficient heat-sink to the comeal epithelium. Such materials include sapphire and quartz. The cornea is best served if the comeal device has a base curve parallel to or flatter than the flattest radius of the cornea. Pre-cooling of the device can be done.
Dynamic cooling or thermal quenching can be accomplished by spraying the cornea with a cryogen, such as 1,1,1,2 tetrafluoroethane (R 134a according to the National Institute of
Standards and Technology; boiling point approximately -26° C). As it rapidly evaporates, the comeal surface cools. The longer the cryogen is in contact with the epithelium, the deeper the cooling effect, therefore, a short burst of the magnitude of 5 milliseconds or less by a fine mist is efficacious. The finer the mist, the more rapid the evaporation from the surface. The infrared laser irradiation would be automatically enabled following the spray application-
The timing of the application of the comeal device or cryogenic spray precedes the laser irradiation. In this way, the epithelium is rendered hypo hermic, thus protecting it from destruction as the photothermal energy is transmitted through and absorbed by the universal chromophore, water, within the stroma.
2. Modulation of the Wound Healing Response to Inhibit Refractive Regression
It is an object of the invention to interrupt the process of repair undertaken by the cornea so that it does not revert to its original (pre-surgical) shape and condition. When procedures such as LASIK are performed, comeal stroma is lathed away, leaving portions of the cornea thinner or reduced, and a certain amount of necrotic debris. Thermal procedures cause keratocyte death via apoptosis and/or necrosis. The cornea's response is to try and correct the insult and return to its original state by initiating a wound healing cascade of events and stromal remodeling. A. Phases of Wound Repair
Wound repair is an integration of dynamic interactive processes. In the absence of intervention, these wound repair processes follow a specific time sequence and can be temporally categorized into three phases: (A) inflammation, (B) tissue formation, and (C) tissue remodeling. These phases of wound repair are not mutually exclusive, but rather overlap in time (FIG. 1), and are roughly as follows. (A) Inflammation (early and late) 1) blood vessel disruption with concomitant extravasation of blood constituents.
2) blood coagulation and platelet aggregation generate fibrin-rich clot.
3) provides matrix for low impedance cell migration.
4) platelets secrete growth factors that initiate granulation tissue formation
5) neutrophils attack foreign bodies. 6) monocyte infiltration continues and promotes phagocytosis.
7) monocytes change into reparative macrophages which secrete growth factors
(B) Granulation Tissue Formation
1) begins to form in 4 days. 2) consists of new vessels, macrophages, fibroblasts and loose connective tissue.
3) fibroplasia consists of granulation tissue components (fibroblasts and extracellular matrix (ECM)).
(C) Matrix Formation and Remodeling 1) structural molecules of early ECM contribute to tissue formation by providing a scaffold for fibronectin and collagen, low impedance cellular mobility and reservoir for cytokines.
2) cytokines stimulate fibroblasts to migrate and switch their major role to protein (collagen) synthesis 3) fibroblasts participate in ECM remodeling, e.g., by depositing fibronectin as a second-order provisional matrix (the appearance of fibronectin, then collagen in healing wounds is consistent with fibronectin serving as a template for collagen fibril organization).
4) hyaluron is a major component of early granulation tissue, promoting cellular movement.
5) proteoglycans are secreted by the ECM and mature fibroblasts; their function in the ECM is to regulate collagen fibrillogenesis.
6) collagen fibrillar deposition into the wound site peaks around 7 to 14 days and provides the structural support. 7) myofibroblasts differentiate and cause tissue contraction by means of dynamic linkages between actin bundles and ECM.
8) by the third week, wounds gain about 20% of their final strength.
9) collagen remodeling during the transition of granulation tissue to mature scar is dependent upon both continued collagen synthesis and collagen catabolism controlled by a variety of enzymes.
Figure 1 shows these wound repair processes plotted as a logarithmic function of time. The phases of wound repair overlap considerably with one another. Inflammation is divided into early and late phases denoting neutrophil-rich and mononuclear cell-rich infiltrates, respectively. Wound contraction begins after granulation tissue is well established. Collagen accumulation begins sharply after the onset of granulation formation.
These processes are generally mirrored in ocular histopathology, except for the vascular events which are modified in the cornea due to its avascularity. Several general cellular and molecular events are precipitated by traumatic comeal intervention; 1) apoptosis, or programmed cell death, 2) necrosis from direct cellular destruction, 3) lamellar instability, which is caused by damage to the stabilizing inter-molecular and intra-molecular collagen cross-links within the collagen matrix, and 4) accumulation of heat shock proteins. Epithelial hyperplasia may also play a role, but this appears to be precipitated by apoptosis. B. Apoptosis Apoptosis differs from cellular necrosis, in which the cell contents including the proteolytic enzymes are released into the surrounding area and stimulate an inflammatory response and further damage. Apoptosis is a programmed cell death process in which the cells die in a complex and controlled way with minimal collateral tissue damage or inflammation. Apoptosis occurs when the keratoplasty procedure is accompanied by injury to the comeal epithelium, during which various growth factors or cytokines are released from the injured epithelial cells. Among these growth factors are interleukin-1 (IL-1), α and β, and FAS ligands. The cytokines diffuse into the stroma and, thereby, bind to specific receptors located on the cellular surface of the stromal keratocytes and fibroblast like cells.
Interleukin- lα binds to IL-1 receptor type 1 (keratocytes) thus modulating epithelial- stromal communications. This leads to keratocyte disappearance beneath the epithelial injury site. The increased levels of IL-la and IL-lb induce apoptotic death of keratocytes in the stroma.
FAS ligand (FasL) is a membrane-bound protein, which also induces apoptotic cell eath in cells expressing the FAS receptor. All keratocytes to a depth of 50% of the stroma (200 microns) or more may die, depending upon the type of epithelial injury. The morphologic changes in the stromal apoptotic keratocytes include chromatin shrinkage and fragmentation, cell shrinkage and cellular blebbing of apoptotic bodies. Because the highly organized structure of the stroma is essential for the proper refraction of light though the cornea, the elimination of keratocytes induced by comeal epithelial trauma can result in stromal haze such as that observed after various refractive keratoplasty procedures.
It now appears that these cytokine systems are inter-related. Without wishing to be bound by a mechanism, evidence shows that the IL-1 release from the damaged epithelial cells facilitates expression of FasL mRNA and protein at the time that cell death in response to IL-lα is noted. It appears that IL-1 may trigger autocrine suicide of keratocytes by induction of FAS ligand in cells already expressing the FAS receptor. It thus, appears that several systems may mediate keratocyte apoptosis in response to the epithelial wounding.
This interaction then induces fibroblasts to up-regulate hepatocyte growth factor (HGF), keratinocyte growth factor (KGF), and protein levels, all of which aid in the wound remodeling. For example, HGF and KGF both stimulate epithelial cell proliferation. Concuπent stimulation of proliferation and inhibition of differentiation in comeal epithelial cells can produce thickening of the comeal epithelium and hyperplasia. Epithelial hyperplasia has been shown to be another factor that leads to regression of refractive effect after PRK (see, e.g., Spadea et al, J. Refractive Surgery, 16: 133-139 (2000); Salchow et al J. Refractive Surgery, 15:590-593 (1999), both herein incoφorated by reference). The destruction and removal of the apoptotic cells stimulates the migration of new keratocytes from the posterior and peripheral stroma to the site of injury in the anterior stroma. These repopulating keratocytes differ in appearance from the apoptotic cellsin that they have dark nuclei and prominent intracellular organelles, and are indistinguishable from fibroblasts in other parts of the body. These activated keratocytes are associated with the increased stromal collagen synthesis and collagen disorganization accompanying stromal remodeling (correlating with the increased haze as seen in excimer photorefractive keratectomy) and the regression of effect or comeal topographic hysteresis as observed in all refractive surgical procedures.
Keratocyte apoptosis is, therefore, an initiating event in the wound healing response after the epithelial removal in PRK and/or thermal damage in photothermal keratoplasty and radiofrequency keratoplasty. While somewhat diminished in laser-assisted in situ keratomileusis (LASIK), there is also a significant wound healing response within the stroma with the level of regression after LASIK varying from essentially zero to as much as 50% in individual cases. Interference with one or more steps in the apoptosis triggering mechanism, or with the migration of new keratocytes into the recently remodeled comea, can prevent the deleterious keratocyte and/or collagenous matrix proliferation that leads to regression.
C. Cellular Necrosis
The wound repair-related regression in LASIK appears to be more the result of cellular necrosis inducing collagenesis and stromal remodeling than the result of cellular
apoptosis and epithelial hyperplasia (as the epithelial damage in this procedure results only from the laser microkeratome). Cellular necrosis (not apoptosis) results from the ablation of the stroma. The necrotic cell's appearance differs from apoptotic cells and is characterized by the rapid loss of membrane function and abnormal permeability. There is early disruption of organelles and irreversible damage to mitochondria.
A histological study has revealed that the two cellular destructive processes are distinguishable from each other based upon staining and histological criteria in tissue specimens. Clinically, they both stimulate wound repair cascades culminating in collagenesis that in turn causes the regression of effect.
D. Distinguishing Apoptosis from Cellular Necrosis
Standard, art-known H & E staining characteristics can be used for identification of necrosis and TUNEL stain can be used to identify apoptotic nuclei with their characteristic DNA fragmentation. Nuclear DNA fragmentation occurs relatively early in the process of apoptosis, preceding significant disruption of the nuclear membrane, but it is a relatively late event in necrosis, occurring when there is already severe nuclear and/or cellular membrane disruption. The TUNEL stain labels fragmented DNA and is not specific for either apoptosis or necrosis. Therefore, the presence of apoptotic cell death is determined by identifying cells stained by TUNEL that have an intact nuclear membrane. In contrast, necrotic tissue will stain positively in the TUNEL assay and display a dismpted nuclear membrane. The use of TUNEL to identify fragmented DNA, in conjunction with H & E staining of the nuclear membrane, provides that necrosis and apoptosis can be reliably distinguished, allowing for the separate analysis of each.
It now appears that the epithelial-stromal apoptotic responses and the post-necrotic wound repair processes each play important roles in the stromal remodeling responsible for the regression of refractive effect. Both processes occur immediately after comeal injury and may persist for long periods of time following the injury. Thus, they are prime targets for intervention to prevent regression.
E. Heat Shock or Stress Proteins - Relation to Prevention of Regression in Laser Refractive Keratoplasty
Immediately after a sudden increase in temperature, all cells increase production of a certain class of molecules. When first observed some 35 years ago, this phenomenon was called the "heat shock response." It is now commonly referred to as the stress response and the expressed molecules as stress proteins.
These proteins are far more than just defensive molecules. Throughout life of a cell, many of these proteins participate in essential metabolic processes, including the pathways by which other cellular proteins are synthesized and assembled. Some stress proteins appear to orchestrate the activities of molecules that regulate cell growth and differentiation. The stress proteins do play an active role in cellular defense. For example, the stress response plays an important role in the ability of animals to withstand brief exposures to high temperatures. The stress response facilitates the identification and removal of denatured protein from the traumatized cell.
Many of the agents that induce the stress response are protein denaturants and can lead to loss of the protein's biological function. It has been suggested that the accumulation of denatured or abnormally folded proteins in a cell initiates a stress response. This response has been observed in the phase-transition precipitated by laser thermal keratoplasty.
A highly inducible heat shock protein, hsp-70, accumulates inside the nucleolus after heat shock. The nucleolus manufactures ribosomes, the organelles on which proteins are synthesized. After heat shock, cells stop making ribosomes and their nucleolus becomes awash in denatured ribosomal particles, hsp-70 recognizes denatured intracellular proteins and restores them to their correctly folded, biologically active shape.
There is a family of hsp 70-related proteins. All of them share certain properties, including a high affinity for adenosine triphosphate (ATP). All of these related proteins, with one exception, are present in cells growing under normal conditions, yet in cells experiencing metabolic stress, they are synthesized at much higher levels.
It has been observed that one form of hsp-70 is identical to immunoglobulin binding protein (BiP). BiP is involved in the preparation of immunoglobulins, as well other secreted proteins. BiP binds to newly synthesized proteins as they are being folded or assembled into their mature form. If the proteins fail to fold or assemble properly, they remain bound to BiP and are eventually degraded. In addition, under conditions in which abnormally folded proteins are accumulated, the cell synthesizes more BiP.
Taken together, these observations indicate the BiP helps to orchestrate the early events associated with protein secretion. BiP seems to act as a molecular overseer of quality control, allowing properly folded proteins to enter the secretory pathway but holding back those unable to fold correctly.
Biochemical studies have indicated that other proteins known as hsp- 10 and hsp-60 are also essential to protein folding and assembly. The hsp-60 molecule consists of two seven-member rings stacked one atop the other. This large structure appears to serve as a "work-bench" onto which unfolded proteins bind and acquire their final three dimensional
structure. According to cuπent through, the folding process is extremely dynamic and involves a series of binding and release events. Each requires energy, which is provided by the enzymatic splitting of ATP, and the participation of hsp- 10 molecules. Through multiple rounds of binding and release, the protein undergoes conformational changes that take it to a stable, properly folded state.
It is likely that both hsp-60 and the hsp-70 families work together to facilitate protein maturation. As a new polypeptide emerges from a ribosome, it becomes bound to a form of hsp-70 in the cytoplasm. Such an interaction may prevent the growing polypeptide from folding prematurely. Once the synthesis is complete, the new polypeptide, still bound to its hsp-70 escort, is transferred to a form of hsp-60, on which folding of the protein and its assembly with other protein components commences. Hsp-60 and hsp-70 act as "molecular chaperones." Although the molecules do not convey the information for the folding or assembly of proteins, they do ensure that those processes occur quickly and with high fidelity. They expedite self-assembly by reducing the possibility that a maturing protein will head down an inappropriate folding pathway.
Temperatures that are sufficient to activate the stress response may eventually denature some proteins inside cells. Heat-denatured proteins, like newly synthesized and unfolded proteins, would therefore represent targets to which hsp-60 and hsp-70 can bind. Over time, as more thermally denatured proteins become bound to hsp-60 and hsp-70, the levels of available chaperones drop and begin to limit the ability of the cell to produce new proteins. The cell somehow senses this reduction and responds by increasing the synthesis of new stress proteins that serve as molecular chaperones.
A rise in the expression of stress proteins may also be a requirement for the ability of cells to recover from a metabolic insult. If heat or other metabolic insults irreversibly denatures many cellular proteins, the cell will have to replace them. Raising the levels of those stress proteins that act as molecular chaperones will help facilitate the synthesis and assembly of new proteins. In addition, higher levels of stress proteins may prevent the thermal denaturation of other cellular proteins.
The magnitude of the resulting stress response appears to correlate with the relative severity of damage sustained. This has focused the utility of using the changes in stress protein levels as markers for tissue and organ injury. Our pre-clinical studies to develop an animal model for photo-thermal trauma has utilized this principle in determining appropriate markers of laser-induced trauma.
Cells that produce high levels of stress proteins appear better able to survive ischemic damage than cells that do not. Consequently, raising the levels of stress proteins, for example
by pharmacological means, may provide additional protection to injured tissues and organs For example, raising the levels of such proteins pπor to refractive keratoplasty could lead to protection from apoptosis of keratocytes following the procedure A corollary to this pπnciple is applicable to the inhibition of the cascade responsible for comeal stromal remodeling that results from mterventional laser refractive keratoplasty Thus, loweπng the levels of stress proteins produced would limit stromal remodeling by impaiπng the expression or secretion of collagen
F Control of Stromal Remodeling Stromal remodeling is one of the important factors in refractive regression The wound healing process that results in the remodeling is a complex cascade of events which, in most cases, is hematogenous in oπgm An exception to this is the response to injury in the comea, which is an avascular tissue
Collagenesis, the culmination of all tissue regeneration or healing, progresses naturally from the migration of the multi-potential keratocyte into the wound sites
Fibroblasts differentiate and are guided by a chemotactic gradient into the provisional matπx- filled wound space An active proteolytic system also cleaves a way for the fibroblast migration
After the fibroblast has migrated into the wound, it gradually switches its major function to synthesis of great quantities of collagen in response to TGF-β Once an abundant collagen matnx is deposited in the wound, the fibroblasts cease collagen production, despite the continuing expression of TGF-β Signals responsible for down-regulating fibroblast proliferation and matrix synthesis may include factors such as gamma-interferon and the collagen matnx itself (suppressing both fibroblast proliferation and fibrobtastic collagen synthesis) A complex interaction and feedback control between cells-cytokines/growth factors-enzymes-matnx is what controls the production of collagen
Hyaluronic acid (a glycosaminoglycan or GAG) appears to promote cellular migration in early granulation tissue and as it becomes hydrated, it promotes expansion of the interstitial spaces, allowing more cell recruitment and proliferation in these areas Once the wound is filled w ith granulation tissue and covered with neo-epidermis, myofibroblasts differentiate to contract the wound, and epithelial cells differentiate to establish the permeability barπer
The role of TGF-β in scar formation is not well understood, but this is where certain pπor art methods to control the fibrotic response have been directed These methods are very non-specific TGF-β suppresses the proliferation of epithelial cells, which, by suppressing re- epithelia zation of denuded basement membrane, may potentiate scarπng and fibrosis that
appear to be initiated in the absence of efficient re-epithelialization; and this growth factor has quite profound inhibitory activities on cells on the immune system. Hence, by suppressing the accumulation and/or activation of T-cell sub-population, TGF-β may also indirectly potentiate tissue scarring and fibrosis.
(i) Anti-Fibrotic Therapeutics
Fibrosis takes many forms and is the result of diverse causes. However, each of the many fibrotic diseases and disorders involves an excessive deposition of collagen and the accumulation of scar tissue. Scar tissue is composed of dense and inelastic collagen fibers. As a consequence, the formation of scar tissue causes the distortion and loss of normal histology and function. Regardless of the underlying cause, the cascade resulting in fibrotic tissue is the same. Currently available pharmacologically based treatments for fibrotic disorders are not very effective in preventing the pathological progression of fibrosis. However, recent advances in cell and cytokine biology have brought a new understanding of the molecular events underlying tissue fibrosis and reveal promising new therapies.
This invention takes advantage of these new developments to remedy the consequences of biological response to trauma in ocular pathology, not particularly with respect to scar formation, but rather with a view to limiting and/or preventing refractive regression of the comea. Tissue engineering studies have resulted in the development of selective and focused compounds which inhibit collagen repair mechanisms, and human neutralizing antibodies which block the fibrogenic cascade.
Targets for this inhibition of effect include: (1) Heat Shock Protein 70 (hsp-70), (2) fibrogenic cytokines (CTGF), (3) intra-cellular collagen synthesis enzymes, (prolyl hydroxylase) and (4) extra-cellular collagen fibrillar assembly enzymes (C-proteinase). Encouraging preclinical data on these programs has validated the molecular targets.
(a) Inhibition of Heat Shock Proteins
The various laser-assisted modalities that coπect refractive eπor by altering the anterior comeal radius induce traumatic stress, resulting in the accumulation of high levels of stress proteins. The fibrillogenesis of these new proteins culminate in comeal stromal remodeling and a reversal or regression of the desired anterior comeal radius shape change.
Therefore, pharmacological inhibition of these stress proteins facilitates enhancement and predictability of this change by reducing the action of these molecular chaperones.
Regression of the shape modification as a function of time is thus prevented as well.
When these cells experience metabolic stress and produce more stress proteins, they also make a reporter enzyme which can be detected by various assays. Small molecule inhibitors or neutralizing antibodies to the stress proteins and/or their reporter enzyme(s)
prevents their activity in the same way specific inhibitors are used prevent the fibrogenic cascade. Such antibodies and inhibitors provide specific therapeutics designed to block refractive regression. An example of a commercially available antibody is anti-hsp-70 from StressGen, Inc. (800.661.4978, Victoria, British Columbia, Canada). (b) Inhibition of Connective Tissue Growth Factor (CTGF)
As one studies the therapies which block the cascade of fibrogenic cytokines that drive the destruction of the ocular structure and visual function, it becomes obvious that no satisfactory therapies exist for these indications. Particularly interesting therapeutic candidates are antibodies and compounds which block the production of specific growth factors that enhance connective tissue growth. Excess accumulation of extracellular matrix and fibrosis in ocular disease processes has focused attention upon the role of certain fibrogenic cytokines or growth factors.
Growth factors are a class of secreted polypeptides that stimulate target cells to proliferate, differentiate and organize in developing tissues. The action of growth factors is dependent upon their binding to specific receptors which stimulate a signaling event within the cell. Examples of well-studied growth factors include platelet-derived growth factor
(PDGF), insulin-like growth factor (IGF-1), transforming growth factor beta (TGF-β), transforming growth factor alpha (TGF-αJ epidermal growth factor (EGF), and fibroblast growth factor (FGF). PDGF is a cationic, heat-stable protein found in the alpha-granules of circulating platelets and is known to be a mitogen and a chemotactic agent for connective tissue cells such as fibrolasts and smooth muscle cells. Because of the activities of this molecule, PDGF is believed to be a major factor in normal wound healing and may contribute to such diseases as atherosclerosis and fibrotic diseases. PDGF is a dimeric molecule consisting of an A chain and a B chain. The chains form heterodimers or homodimers, and all combinations isolated to date are biologically active.
Studies on the roles of various growth factors in tissue regeneration and repair have led to the discovery of PDGF-like proteins. These proteins share both immunological and biological activities with PDGF and can be blocked with antibodies specific to PDGF. These new growth factors likely play a significant role in the normal development, growth and repair of human tissue. Therapeutic agents derived from these molecules may be useful in augmenting normal or impaired growth processes involving connective tissues in certain clinical states, e.g., wound healing. When these growth factors are involved pathologically in diseases, therapeutic developments from these proteins may be used to Control or ameliorate uncontrolled tissue growth. Various cell types produce and secrete PDGF and PDGF-related molecules. In an attempt to identify the type of PDGF dimers present in the growth media of
cultured endothelial cells, a new growth factor was discovered. This previously unknown factor, termed Connective Tissue Growth Factor (CTGF), is related immunologically and biologically to PDGF, but it is the product of a distinct gene. CTGF (MW=34,000) is one of the CCN family of early response genes. CCN, an acronym for a family of genes related to connective tissue growth factor, derives its name from the CYR61, CTF and NOV growth factors.
CTGF stimulates matrix production and the proliferation of cells. CTGF does not substitute for TGF-β, which appears to prepare fibroblasts for responding to CTGF, but can supplement the action of TGF-β in inducing cell growth. CTGF is a heparin binding growth factor which may be involved in the activation and tissue storage of TGF-β (Kothapalli at al, Cell Growth & Differentiation, 1997). The TGF-β response element in the CTGF promoter confers the CTGF gene's responsiveness to TGF-β (Grotendorst et al, Cell Growth & Differentiation, 1996). Cells exposed to TGF-β produce CTGF, and this response can be used in a high throughput screen for compounds that block this pathway. Alternatively, since CTGF binds to at least two classes of receptors, small molecule agonists or antagonists can be developed using screening methods known in the art, and CTGF's activity modulated at the receptor-binding step in the cascade.
(i) Active Fragments of CTGF Sequence data from other studies have identified an active fragment of CTGF, referred to as the EENI fragment (MW=10,000), a C-terminal protein of CTGF. Current data suggest that this and other CTGF fragments are produced by proteolytic cleavage. Biological activity is associated with the fragments as well as the intact molecule. The data are consistent with distinct regions of the CTGF molecule inducing different responses (proliferation vs. matrix induction) following proteolysis by different cellular receptors. It has been further suggested that enhanced proteolytic cleavage of CTGF is a progression factor enhancing the local fibrotic response. Antibodies have been developed and optimized to neutralize the various activities of CTGF, and these antibodies or other specific antibodies and small molecules can also be used to inhibit refractive regression.
(ii) Specific Anti-CTGF Therapeutic Targets a) Human Monoclonal Antibodies which Neutralize CTGF or its Receptor High affinity fully human antibodies have been developed which neutralize CTGF or its receptor. The antibodies are drawn from certain transgenic mice bearing human immunoglobulin genes to produce human antibodies. Antibodies that neutralize growth
factors or receptor binding have been shown to neutralize effects for peπods from 15 days to 2 months b) Generation of Active Fragments
The C-terminal fragments have the ability to stimulate cell proliferation, while the N- terminal region stimulates matnx production Antibodies specific to biologically active fragments of CTGF are far more specific in their antagonistic activities Because of these specific functions, control of refractive regression can be achieved by carefully controlling either or both proliferation or matnx production In this way, comeal wound healing and healthy cell turnover is promoted without the accompanying stromal remodeling that promotes regression
c) Small Molecules Inhibiting CTGF Expression
CTGF receptors have been isolated and identified and the signal transduction pathways activated by these receptors have been characteπzed Small molecules which block the induction of CTGF have also been identified It appears that these compounds have the ability to block collagen production in cells exposed to fibrogenic cytokines They will thus also have the ability to be used to inhibit refractive regression, since reduced collagen production will allow the refractive changes induced by keratoplasty to persist intact without stromal remodelling
d) Antisense Inhibition of CTGF Pathway
Portions of nucleotide sequence complementary to DNA or RNA sequences encoding CTGF or portions thereof, or CTGF promotor or other regulatory sequences, may be used to prevent induction of CTGF Methods known in the art to produce and implement antisense therapeutics are within the scope of the invention, and are applicable to inhibit CTGF or any other component of the molecular cascade discussed herein Antisense therapy directed to any step that inhibits apoptosis, necrosis, keratocyte migration or proliferation, or collagenesis is included in the invention
(c) Inhibition of Prolyl Hydoxylase
Collagens make up about 30% of the total body proteins. They are the major structural proteins that hold cells together and determine tissue and organ architecture. In fibrotic disorders, collagen molecules are synthesized and secreted in excess, and the resulting fibrotic scar tissue is comprised of a strong cross-linked collagen scaffold. Attention has been focused upon key enzymes required for collagen formation and deposition as a target for inhibition of the formation and accumulation of collagen which leads to fibrotic scar.
Prolyl-4-hydroxylase (P4Hase) is an intracellular enzyme required for the synthesis and formation of all the 20 known types of collagen, therefore representing an important target for anti-fibrotic therapies. P4Hase works inside the cell by modifying polypeptide chains to allow them to form stable triple-he ϋeal structures. Prolyl hydroxylase catalyzes the hydroxylation of prolyl residues to 4-hydroxyproline during post-translational modification of collagen. The formation of hydroxyproline enables the molecule to form the stable triple helical conformation typical of collagen (FIG. 2). Hundreds of thousands of compounds have been screened in high throughput screening (HTS) for assaying diverse organic compounds against this target. A recombinant expression system has been established to produce large quantities of prolyl hydroxylase. Effective inhibition of the activities of this enzyme prevents the assembly of all types of collagen. Novel compounds that inhibit prolyl hydroxylase have been identified, and these have been tested in secondary screens to determine efficacy, toxicity and potency. Additional compounds have been developed from promising leads by focused combinational chemistry.
Selected compounds have been tested for oral and topical administration, and the compounds have been found to be active and effective in both forms. Safety has been demonstrated in 14-week administration. In these pre-clinical tests, functional improvement was also demonstrated.
Collagen synthesis has been blocked using a novel inhibitor identified as FG-041 in an animal model. Scar formation has been prevented and wound healing and tissue regenaration has been improved. The effect is maintained on withdrawal of the dmg after local administration. In other studies, the drug was well tolerated and showed good pharmacodynamics after oral dmg delivery. Another compound, FG- 1648, a water-soluble inhibitor of prolyl hydroxylase, has been shown to be effective by means of topical administration in a pig model for dermal wound scarring. The safety and efficacy of these compounds make them promising therapeutics for topical ophthalmic administration for use in regression therapy. Their short term use would inhibit regression via stromal remodeling,
but would avoid the side effects of long-term suppression of collagen formation that could impair comeal health.
The regression of refractive effect observed in laser refractive keratoplasty is the result of the trauma-induced wound repair cascade. Collagenesis is the culmination of this event. Thus, ophthalmic administration of compounds inhibiting the prolyl hydroxylase required for collagen biosynthesis and assembly would prevent or control the refractive regression.
(d) Inhibition of C-Proteinase
After procollagen is secreted from the fibroblast, C-proteinase is the enzyme that then coverts the procollagen into collagen, a necessary step in the formation of types I, II, and III collagen by removing the carboxy terminal domain of the procollagen (FIG. 3).
US Patent 6,020,193 to Prockop (herein incoφorated by reference) teaches purified C- proteinase and sufficient amino acid sequence to identify, clone and express this protein. C- Proteinase is an important target for inhibition in refractive regression due to its selective effects on formation of types I and III collagens. It converts fibrillar collagens to collagen by removing the carboxyl terminal globular domain, thereby enabling assembly into larger collagen fibrils. By blocking the action of C-proteinase, procollagens can not assemble into collagen fibrils. As a consequence, C-proteinase inhibitors can block the proliferative effects of types I and II collagen that induce refractive regression, without affecting the formation of other types of collagen.
Three alternative forms of C-Proteinase have been identified, including bone moφhogenic protein, or BMP-1, M-tolloid and M-tolloid-like protein (see, e.g., US Patent 5,939,321). Two methods have been created for the blocking of these enzymes, monoclonal antibodies and small molecule compounds. It is understood for this and all other antibody therapeutics discussed herein, that polyclonal antibodies or active fragments (e.g., FAb) may be used.
These-therapeutic antibodies offer a significant advantage in that they can be directed to the key enzymes allowing the development of therapeutics for specific local, topical and systemic uses. Alternatively, compounds have been and can be identified with high throughput screening processes as anti-regression therapeutics.
Inhibitors identified to date have shown good selectivity in blocking the C-Proteinase at 200 to 1000 fold lower levels than matrix metalloproteinases MMP- 1, 9 and 13. The development of an animal model for wound healing following laser refractive keratoplasty has shown that matrix metalloproteinase-9 (MMP-9) is found in damaged corneas undergoing wound healing. It presents an all-or-none marker since it is not found in the normal comea,
but correlates with the magnitude of laser trauma and resolves in 48 hours. C-Proteinase inhibition should correlate well with the disappearance of MMP-9 in the ophthalmic model and thus provides a sensitive method of monitoring intervention during anti-regression therapy.
(e) Alternative Therapeutics
The methods of the invention can also make use of the following drugs as alternatives for the prevention of refractive regression: 1. Interferon-α 2. 3 hydroxypyridine-2-carboxamidoesters
3. tolloid-like gene product (mTll) and its cognate gene
4. hydroxypyridone compounds
5. 3,4 dihydroxybenzoic acid
6. 3,4 dihydroxyphenylacetic acid 7. P 1894B of Okazaki
8. pyridine-3-carboxylic acids and esters containing a substituted or unsubstituted 5- tetrazolyl group in a 6 position of the pyridine ring
9. pantethine
10. 1 ,2 diamine compounds 1 1. 3,3 dihalo-2-propenlyamine compounds and their salts
12. 2,4 or 2,5-disubstituted pyridine N-oxide compounds
13. 2,3-dialkyl-5,6-dimethoxy-p-benzoquinone compound
14. 5-hydroxy- 1 ,4-naphthoquinone compounds
15. oxalylamino acid compounds and their physiologically active salts 16. 2-amino,-6-phenyl-4H-pyran-4-one compounds containing an N-moφholino substituent in the 2 position of the pyranone ring
17. calcium channel blockers chosen from nifedipine, hydropyridine, verapamil, colbalt chloride, and biologically acceptable cobalt salts
18. peptide derivatives 19. pyridine-2,4-and-2, 5- dicarboxylic acid amides
20. phenanthroline derivatives
21. pyridoxal benzoyl hydrazones or analogs such as 3-hydroxyisonicotinaldehyde benzoyl hydrazone and salicylaldehyde benzoyl hydrazone
22. polypeptide including a sequence of at least about 5 amino acids corresponding substantially to an amino acid sequence from within the 33kD fragment of the A chain
of fibronectin, within the G domain of the A chain of laminin, or within the NCI domain of the α2 chain of the type IV collagen.
Properties of compounds useful in this invention and potential targets for inhibiting regression, as well as information regarding their therapeutic use, can be found, for example, in US Patent Nos. 5,939,321; 6,020,193; 6,037,139; 5,716,633; 5,811,446; 5,626,865;
6,013,628; 6,020,350; 5,789,426; 5,807,981 ; 5,916,898; 5,863,530; 6,005,009; 4,499,295;
6,965,586; 5,965,585; 5,620,996; 5,981,717; 5,993,845; 5,998,422; 5,408,040; 5,837,258;
5,770,209and 5,783,187, all herein incoφorated by reference, and other sources known to those of skill in the pharmaceutical arts.
3. Remedial Cross-linking to Restore Lamellar Stability Following Thermal Refractive Keratoplasty A great deal has been written about the regression of the desired refractive effect refractive keratoplasty. These changes observed as a function of time have been attributed to the wound repair cascade and subsequent epithelial hypeφlasia and stromal remodeling. A frequently overlooked adjunctive process involves the destruction of the stabilizing collagen molecular cross-links. The various intra- and intermolecular cross-links impart to collagen its unique structural integrity. While most of the stability of the triple helical conformation comes from heat-labile hydrogen bonding, surgical intervention also results in proteolytic enzymatic destruction of other cross-links, and may be the first step in the catabolism of insoluble collagen in the comea.
A. The Nature of the Structural Integrity of Collagen Primary structure:
The essential features of the primary structure of collagen include the presence of glycine as every third residue and the presence of the amino acids hydroxyproline and hydroxylysine. Proline and hydroxyproline together account for approximately one-third of the amino acids present. There is evidence that the collagen molecule is stabilized by one or more disulfide cross-links in addition to hydrogen bonding.
The individual peptide chains of collagen molecules are twisted into a right-handed helix. This helical twist is the result of the presence of large numbers of amino acid residues, proline, and hydroxyproline. Because one side of the five-sided ring structure of these amino acides constitutes one part of the peptide backbone of the molecule, free rotation of the chain cannot occur at these sites and the backbone of these chains are bent at these points. The net
result of the presence of these residues is that the chain of amino acids is forced into the helical configuration. However, there is a short sequence of amino acids (15 in the case of type I collagen) at the N-terminal portion of the molecule that not have glycine as every third residue. This is know as the "telopeptide" region. This region is not helical, can be attacked by ordinary proteolytic enzymes, and is important in cross-linking.
Secondary structure:
The three peptide chains must be held together in some fashion to make a molecule. When first assembled, the three chains are held together by relatively weak forces. The molecule thus formed, before it has become polymerized into fibrils and fibers, is referred to as "tropocollagen." The term collagen usually refers to the polymeric aggregation of a number of tropocollagen units to form a long polymeric chain. Covalent cross-links can form among the three chains.
Tertiary structure:
The spatial aπangement of three chains in the molecule also is unique to collagen. The α chains are right-handed helices, and the three helices are then twisted into a left-handed "super helix." The entire structure is held together by hydrogen bonds (FIG. 4).
Hydrogen bonds are formed between a hydrogen atom and a strongly electronegative atom such as F. O, or N. In proteins, a hydrogen attached to an amino group or a hydroxyl group may form a bond within adjacent oxygen derived from a carboxyl group. Hydrogen bonds are quite weak and be broken by mild heat, such as that which is generated by laser thermal or conductive radiofrequency keratoplasty. When this happens, the helical structure of the molecule is destroyed and individual chains separate as randomly coiled structures. In the case of collagen, this gentle heating leads to hydrogen bond mpture and denaturation of the collagen. This process is inherent in the thermal keratoplasty methods and results in collagen fibrillar shrinkage or contraction. If the thermal profile is uncontrolled and exceeds the thermal shrinkage temperature of collagen (Ts), collagen is reduced to parent gelatin and loses all structural integrity.
Quaternary structure:
The quaternary structure of collagen, or the way tropocollagen is aggregated into a stable biological unit of great mechanical strength, is important to the mechanical properties of collagen (FIG. 5).
In the course of aggregation and during cross-linking, the solubility of collagen undergoes change. Collagen is no longer soluble in water when so aggregated but can be solubilized in neutral salt solutions. Initially the forces holding chains of tropocollagen together are electrostatic. Salt serves to neutralize the electrostatic forces holding the molecules together and thus solubilizes collagen. As intermolecular cross-links form, collagen losses its salt solubility, but can be solubilized in dilute acid, such as 0.5 M acetic acid. The major intermolecular covalent cross-links that form are aldimines which hydolyze in acid solution. As these aldimine bonds become reduced or substitutions occur across the double bond, collagen fibrils become completely insoluble, even in acid. A model of the collagen fibril, based upon physical chemistry, describes the micro fibril of collagen as held together by covalent cross-links and the bundles of microfibrils are "cemented" together to form fibrils by protein polysaccarides of glycoproteins.
As a fibril grows, each succeeding chain is distorted or twisted a bit more because of increasing surface of the fibril. This introduces a stress into the polymer and it requires force to maintain this stressed position. This force is provided by ionic or charge attraction between bonding sites in the tropocollagen chain.
Collagen Maturation: Maturation of collagen can be defined as the process by which the fragile, soluble fibers of collagen change into strong, insoluble fibers and how they proceed from a disorganized, random, and not very useful aπangement to an oriented, organized structure providing mechanical strength to a tissue.
One of the most stable cross-links arises from a shift of the double bonds to yield a ketone. Thus, these bonds and others such as more complex reaπangements involving other amino acid side chains, including the imidazole group of histadine, are the major force holding fibrils and fiber bundles together. Their presence is the chief contributing factor in the tensile strength of collagen.
It is important that for intermolecular bonds to form, the participating groups must be packed closely together. There is evidence that mechanical tension may play a role in packing the chains so that these cross-linking reactions are facilitated. The formation of an intramolecular bond does not alter the solubility of collagen, bu7t it does make the molecule much more resistant to attack by enzymes. There is also evidence of covalent linkage fibrils to the glycoproteins of the stromal ground substance.
Covalent cross-linking is important in the maturation of collagen and in the tensile strength of wounds. The introduction of cross-links by local treatment of a healing wound with cross-linking agents should hasten the increase in tensile strength. This has, in fact, been the case in controlled studies. The thermal shrinkage temperature of the individual collagen fiber was increased in treated wounds as compared to fibers from comparable control wounds.
Disclosed herein is the administration of cross-linking agents by topical ophthalmic delivery to a thermally damaged comea to enhance the structural recovery of the tissue. We have applied this concept to laser and radio frequency conductive keratoplasty.
B. Causes of Lamellar Structural Instability
One of the great paradoxes in refractive keratoplasty is demonstrated in laser thermal keratoplasty. LTK requires heat to reduce the hydrogen bonds in order to obtain efficacious comea shape change by means of collagen fibrillar shrinkage. Hydrogen bonds are quite weak and can be broken by mild heating. When this happens, the helical stmcture of the molecule is destroyed and individual chains separate as randomly coiled structures. In the case of collagen, mild heating, which leads to hydrogen bond mpture, produces denaturation; parent gelatin is the resulting product. The absence of the triple-helical formation of the molecule leads to extreme fragility of the structural integrity. When collagen is denatured from exposure to elevated temperature and its tertiary stmcture destroyed, it is quite susceptible to attack by trypsin, pepsin and other proteases. This increases its heat-lability and lowers thermal shrinking temperature of the collagen, thus reducing its tensile strength even more. It may also lead to the secondary destruction of other collagen stabilizing cross-links. This, then, forms the basis for lamellar instability and regression of desired comeal shape change following laser thermal keratoplasty and other modalities which rely upon heat to induce comeal recurvature for refractive modification.
The hydrothermal shrinkage of the collagen molecule occurs within a small temperature range, from 60 to 70 C, and is attributed to the cleaving of the internal stabilizing cross-links. Exceeding this temperature range will further damage these cross-links and culminate in comeal lamellar instability.
Early relaxation (leading to refractive regression) of the tension exerted by the hydro- thermal shrinkage of collagen fibers will depend upon the nature of the inherent cross-links in the collagen network Hydrolysis of the heat-labile cross-links has been shown to be one of the causes. Two phenomena will compete during the hydrolysis of these heat-labile bonds: relaxation following the decrease in the number of meshes and eventually the larger diameter
of the remaining ones, and the rise in tension induced by the temperature effects on the resistant network. The parameters influenced by the hydrolytic reaction of the network nodes are time and temperature, as well as "stress" dependence. Other parameters depend upon cross-link density and the presence of the heat-stabile cross-links that are responsible for the residual tension.
The fibroblast, or keratocyte in the case of the cornea, is the cell primarily responsible for the synthesis of collagen. When a wound occurs, the fibroblast is among the first cells to appear. Amino acids are picked up by specific transfer RNAs and incoφorated into growing peptide chains within these cells and become collagen precursors or procollagen. Collagen chains exist as a helical coil aπanged parallel to each other. The alpha chains are right-handed helices, and three helices are then twisted into a left-handed "super- helix." This entire stmcture is held-together by hydrogen bonds. Hydrogen bonds (FIG. 4) are formed between a hydrogen atom and a strongly electro-negative atom. In proteins, a hydrogen attached to an amino group or a hydroxyl group may form a bond with an adjacent oxygen derived from a carboxyl group. Hydrogen bonds are quite weak and can be broken by mild heat. When this happens, the helical stmcture of the molecule unwinds and the collagen molecule contracts.
If covalent cross-linking is important in the maturation of collagen and in the tensile strength of wounds, the introduction of cross-linking by local treatment of a healing wound with cross-linking agents would be expected to hasten the increase in tensile strength. This experiment has been performed with formaladehyde in one instance and l-ethyl-1-3 (3- dimethylaminopropyl) carbodiimide hydrochloride in another. It was suggested that formaldehyde would introduce methylene bridges between adjacent tropocollagen chains. The carbodiimides has been shown to be useful in forming amide bonds during protein synthesis. With both agents there was a sizable increase in tensile strength of the wounds as compared with controls.
The thermal shrinkage temperature of the individual collagen fibers from the wounds was increased in treated wounds as compared to fibers from comparable controls. These facts suggest that cross-linking after fibril formation is an extremely important aspect responsible for the mechanical properties of collagen, particularily tensile strength.
The greatest bulk of comeal stroma consists of collagen fibers of uniform diameter gathered into bundles or comeal lamellae. Each successive lamellar layer has collagen fibrils oriented at right angles to the ones above and below it. Each fiber is imbedded in a characteristic matrix, or ground substance. The possible stabilization of collagen fibrils
through covalent bonding with glycoprotein to collagen also plays a role in the physical properties of the collagen fibril and may regulate the size attained by the fibril.
All of the physical properties of collagen, such as the tensile strength, are affected by cross-linking. The lamellar instability which results from thermal keratoplasty, whether by coherent, radiofrequency radiation or any other method of thermally modifying the comeal collagen matrices, is directly related to mpture of the collagen intermolecular hydrogen bonds which are heat labile. The heat generated by thermally mediated comeal processes may also alter certain intramolecular cross-links (FIGS. 5 and 6).
Regressive modification of the desired refractive change following thermal keratoplasty is directly related to these biochemical alterations. The introduction of cross- linking agents at the time of, or subsequent to, thermal injury would hasten the increase in tensile strength of the comea and would provide stability to the resultant refractive modification.
C. Methods and Devices for Restoration of Lamellar Stability
Incomplete cross-linking of tissues and the structural collagenous network can lead to enhanced biodegradation, antigenicity and loss of mechanical function. Prior art cross- linking techniques have made use of homo- and heterobifunctional reagents, both of which form chemical cross-links by introducing bridges between amino acid chains. Limitations of some cross-linkers have been obviated by the use of rapidly cross-linking photosensitive agents. Upon photoactivation by the use of ultraviolet or visible iπadiation, the photoactivatable site is converted to a species of very high chemical reactivity which forms a covalent linkage with another amino acid side chain. The absoφtion of the UV or visible radiation by the bifunctional reagent can give rise to two general classes of species produced by the cleavage of chemical bonds.
Two methods exist for the photoactivation of heterobifunctional compounds. One is accomplished by means of iπadiation with a short wave UV lamp. The half-time of photolysis is in the order of 10 to 50 seconds depending upon the reagents used. The other method is flash photolysis for an extremely short period on the order of milliseconds. In past investigations, chemical agents, in particular glutaraldehyde, were found to have application for the biosynthesis of intramolecular and intermolecular cross-links. This process has been used commercially in vitro to stabilize pig heart valves used in valvular replacements. Other applications have been in the development of tissue bioadhesives for sutureless closures of wounds and in the development of contact lens material, lens or comeal
implant material, a wet occlusive bandage, patch graft, implant material to replace silicone in cosmetic plastic surgery, artificial joint lining material and as a dmg delivery mechanism.
In each case, all of the applications of this process for cross-linking any amino acid- containing polymer have been in vitro. The introduction, in vivo, of collagen cross-linking agents during or immediately following thermal modification of the comes for refractive keratoplasty is proposed for enhancing the lamellar structural integrity thus weakened by the themal keratoplasty. This will prevent the transient stress-induced regression normally observed.
One such method and apparatus is the utilization of spray or drop instillation. A biodegradable collagen shield saturated with this photosensitive agent can also be utilized followed by photoactivation by short-wave UV or visible photonic radiation. Unique to this method is the fact that while one end of the bifunctional reagent form peptide-like bonds with the collagen amino acid side chains, the other end remains unbound until photoactivation. This end is then converted to a highly reactive compound called a "nitrene" or a "carbene", which in turn bonds with an amino acid side chain of tissue collagen.
The concentration of the cross-linking reagent mixtures used in this application may vary between 5 μM and 250 mM dissolved in a biologically compatible solvent such as, but not limited to, DMSO.
Photoactivation of the reagents can be achieved within a wave length range of 220 nanometers to 310 nanometers. The duration of the photoactivation will vary depending upon the cross-linker used.
An additional cross-linking system is presented which utilizes the vitamin B2 or riboflavin as the photomediator. It can be instilled in several ways to be outlined. Once the comea is so treated, the stained comea is iπadiated or exposed to a 400 nm ultraviolet light source for about 30 minutes. This photoactivation yieldsO:, which creates new cross-links between the collagen fibers. The comeal structure has been shown to stiffened by a factor of approximately 1.6 using this method.
Photoactivation, depending on the wavelength and time required, may cause additional comeal epithelial trauma. It may, therefore, be preferable to utilize cross-linking agents, which do not require photoactivation. Broad classes of cross-linking reagents, in addition to those requiring photoactivation, are included in this invention for the puφose of re-establishing the structural integrity of the comeal lamellae. Among these classes are those reagents that use additional non-toxic chemicals or heat to complete the desired chemical reaction.
A prefeπed cross-linker of the invention is glycerine or glycerol. Glycerol is the simplest trihydric alcohol, with the formula CH 2OHCHOHCH 2OH. The name glycerol is prefeπed for the pure chemical, but the commercial product is usually called glycerin. It is widely distributed in nature in the form of its esters, called glycerides. The glycerides are the principal constituents of the class of natural products known as fats and oils. It is completely soluble in water and alcohol and has a very low toxicity in mammals. It has properties that allow it to stabilize lamellar stmctures without the need to photoactivate it. Because it requires no potentially damaging photoactivation, has very low toxicity, and is a common additive to food items, it provides the advantage of an effective, non-toxic, inexpensive cross- linking reagent that is very suitable for human use.
4. Pharmacokinetics
A. Factors Influencing Drug Penetration Into the Cornea
While there are several factors that differentiate the use of anti-fibrotic and cross- linking therapy in the comea from other tissues in the body, pharmokinetics is of prime importance. Dmg absoφtion into the comea presents unique challenges.
The anti-fibrotic and cross-linking dmgs and reagents can be instilled in several different ways but these agents must penetrate the physical barrier that the comea presents. Dmgs appear to penetrate the comea by diffusion and the rate of diffusion parallels the dmg concentration. The comeal epithelium is the first baπier and dmgs must enter this layer rapidly or be washed away. Cell membrane lipids, which are present within the five layers of the epithelium, limit drug penetration. The epithelium contains hydrophilic constituents, which also retard dmg penetration.
For there to be clinical significance in the restoration of stromal lamellar stability by means of reagents or dmgs, the dmgs must be presented to the comea in a "bioavailable" manner. The bioavailability of a drug product is defined as the percentage of the reagent that is absorbed. In addition to bioavailability, the other important parameter to be considered is the pharmokinetics of the dmg in the patient's tissues. Many factors will alter each parameter and will result in the choice of method or means of administration of the reagent of choice. Among the pertinent factors are the excipients or inactive compounds present in the dmg such as diluants, lubricants, binders and the like, which are important determinants of bioavailability. Variations in dissolution rate are also important. The pH of the dmg solution determines whether it is the ionized or non-ionized form. The non-ionized form, being more lipid soluble, is better able to penetrate the comeal epithelium. Viscosity will affect the drug- comeal contact and, therefore, its bioavailability. Surfactants or detergents may be used to
increase the solubility of dmgs that are hydrophobic. Osmotics may be added to adjust the tonicity of the ophthalmic solutions to that of the tears.
Lacrimal volume will affect the concentration of the dmg, but reflex tearing can be minimized by the prior administration of a local anesthetic and dmg absoφtion can be increased by altering the comeal epitheliun. Eye drop size and/or punctal occlusion will result in greater reagent concentration being available.
B. Modes of Drug Delivery The reagents or dmgs are presented to the comea by one of three modes of administration: 1) Topical administration of solutions, suspensions, ointments, powders, particulates, and the like, by periodic or continuous irrigation (FIGS. 7a and 7b), and aerosol spraying (FIG. 7c), iontophoresis, or the like.
2) Applications of dmg reservoirs, such as pledgets (usually, squares of pressed cotton), or sponges soaked with a drug (FIGS. 8b and 8c), soft contact lenses soaked in a reagent (FIG.8a), polymer dmg delivery systems, collagen shields soaked in the dmg
(FIG. 8a), liposomes, flexible capsules or wafers, or other membrane or reservoir systems (FIG. 8d), and
3) Subconjunctival injection (FIG. 9).
Solutions and aqueous suspensions are the pharmaceutical forms most widely used to administer dmgs that must be active on the eye surface or in the eye after passage through the comea or conjunctiva. To increase bioavailability of dmgs, to extend therapeutic efficacy, and to improve patient compliance, various dosage forms have been developed over the years. These include soluble inserts (undergoing gradual dissolution/or surface erosion), insoluble inserts (e.g., medicated contact lenses such as Ocusert™, etc.), gels (e.g., Gelrite™), liposomal and dmg delivery via nanoparticles (emulsion, suspension, etc.), and ointment (See Edman, BIOPHARMACEUTICS OF OCULAR DRUG DELIVERY, CRC Press, 1993). Suspensions require shaking well before administration to ensure the particles are well dispersed in the suspension solution. The combinations of powders and periodic irrigation, such as by eyedropper 44 in
FIG. 7b, or continuous irrigation, such as by reagent reservoir 40 and irrigation tube 42 in FIG. 7a, may be useful. As shown in FIG. 7b, periodic irrigation may be achieved by the use of eyedropper 44, having reagent reservoir 46, and pressure bulb 48, to deliver reagent to the eye 10.
Alternatively, such periodic irrigation can be achieved by similar periodic drip systems, either manual or automated. As shown in FIG. 7a, continuous irrigation may be achieved by use of reagent reservoir 40 which feeds reagent, either by gravity or controlled feeding, to irrigation tube 42, and thereby to the eye 10. Alternatively, such continuous irrigation can be achieved by similar continuous reagent supply systems, either manual or automated.
Similarly, aerosols may be employed, such as by use of a reagent reservoir 50 and aerosol device 52, including reagent access tube 54, of FIG. 7c. Periodic or continuous irrigation, or aerosol spraying, should be used to iπigate, or to spray the eye, respectively, with reagent for about five minutes or until the reagent has sufficiently penetrated the comeal epithelium. When periodic irrigation systems, continuous irrigation systems, or aerosol devices are used, the reagent should be applied before the blink reflex. Alternatively, means can be employed to inhibit the blink reflex and, thus, facilitate the delivery of reagent by any of these systems. Iontophoresis is largely of historical significance to aid the penetration of ionizable dmgs having limited lipid solubility. This technique works by placing one electrode upon the comea usually imbedded within a scleral contact lens in the presence of a dmg solution. The other electrode can be placed anywhere else on the body. In iontophoresis, the electrode, the cuπent, the topical anesthetic, the hypertonic solution, and the non-physiologic pH-aided dmg penetration may cause some comeal epithelial damage. If the epithelium is breached, the cuπent alters the steady state of the stroma and, therefore, the Ts (thermal shrinking temperature), thus adding another variable that may pertubate the results.
In administration by subconjunctival injection, a syringe 64 of FIG. 7 is used to penetrate the subconjunctiva, the "mucous membrane" of the eye which lines the inner surface of the eyelids and is reflected over the fore part of the sclera and the comea, and to deliver a reagent to the subconjunctival tissue. Syringe 64 includes needle 66, typically a 25 gauge, reagent reservoir 68, and plunger 70 (shown in part) of FIG. 9. In use, the needle 66 is placed adjacent the conjunctiva, as schematically illustrated by the dashed line of FIG. 9, pressed against the conjunctiva to penetrate into the subconjunctival tissue, and plunger 70 is depressed to deliver reagent from the reagent reservoir 68 to the subconjunctival tissue. The injection is usually directed to the perilimbal portion of the subconjunctiva.
Certain aspects of the injection of the reagent, such as patient apprehension, subsequent inflammatory response, pain, inconvenience, and expense should be evaluated against the advantageous aspects of this method.
Pledgets 58 of FIG. 8b or sponges 60 of FIG. 8c, usually smaller than a contact lens 56 of FIG. 8a, are soaked (as indicated by a droplet) with a reagent and then placed inside one or both lids, adjacent the comea 11 of FIG 8a, for about five minutes or until the reagent sufficiently penetrates the comea 11. Pledger 58 may be composed of a fibrous material, such as cotton or cellulose, as illustrated by the curved and overlapping lines in the interior of pledger 58 of FIG. 8b. Sponge 60 may be composed of known sponge materials, such as cellulose, its sponge-like characteristics being illustrated by the void, and non-void spaces in the interior of the sponge 60 of FIG. 8c. While pledger 58 is often a cotton square, when placed adjacent the comea 1 1 in the same manner as described below with respect to the contact lens 56 of FIG. 8a, it substantially conforms to the shape of the comea 1 1. Similarly, sponge 60 substantially conforms to the shape of the comea 1 1 when it is placed adjacent the comea 1 1 in the same manner. For simplicity, pledget 58 and sponge 60 are shown in two dimensions, it being understood that pledger 58 and sponge 60 have substantially the three- dimension configuration of a contact lens 56 of FIG. 8a. While the above described delivery systems are quite effective, dmgs contained within a membrane device 62 of FIG. 8d and placed within the conjunctival sac produce a more even release of dmg than any topical system. Membrane device 62 is left in the conjunctival sac for about five minutes or until the dmg or reagent sufficiently penetrates the comea 1 1. These membrane devices 62 may take the form of flexible capsules, shaped as semi-circular wafers, which are placed in one or both eyelids (usually the lower eyelid), although other forms are contemplated. While the lipophilic or hydrophilic nature of the membrane, the pore size, and the membrane thickness may mitigate the rate of dmg release, such membrane system may have advantages over other modalities. For example, excipients can be avoided, tear pH is not acutely lowered and lacrimal washout is not a factor. Liposomes, which are synthetic phospholipid vesicles, absorb to the comeal epithelium cell membrane and transfer dmg directly. Liposomes are generally described in published international patent application under the Patent Cooperation Treaty, International Publication No. WO 86/03938, of inventors John H. Crowe and Lois M. Crowe (International Filing Date of Jan. 8, 1986), on pages 1 and 2. The entire disclosure of these pages is incoφorated herein by this reference. The primary limitations of liposomes are their limited binding power to the epithelium and their expense.
Hydrophilic soft contact lenses 56 of FIG. 8a have been used as dmg reservoirs with success. Polymer dmg delivery systems, while effective, are generally more rigid than soft contact lenses 56. Particularly prefeπed are cross-linked collagen shields 56, also shown of FIG. 8a, which are shaped like contact lenses 56 and are used to both promote comeal
epithelial healing and to provide drug delivery. Because they conform to the shape of the comea 11, multiple base curves are not required. They are also biodegradable and thus provide protection to the comea during the post-laser recovery period after the dmg is dissipated and do not require removal. As schematically shown in FIG. 8a, a pre-soaked (as indicated by a droplet) contact lens 56 or collagen shield 56 which is shaped to conform substantially to the comea 11 is placed (as indicated by the pair of parallel direction aπows) in contact with the comea 11 of an eye 10 of a patient. As further explained herein, the patient's eyelids are closed over the shield for about five minutes or until the reagent sufficiently penetrates the comea 11. The ideal system combines the administration of topical anesthetic, such as proparacaine, with the efficient administration of reagent by means of a collagen shield. Topical anesthetic is used to breach the epithelial barrier and to provide, simultaneously desensitization for the laser exposure. Presenting topical anesthesia by this method provides protracted desensitization for comfort following the laser exposure and, by way of the collagen shield, protects the surface epithelium preventing denudation. The reagent further provides the intermolecular stabilization necessary of the comeal stroma following the destabilizing laser treatment.
These and other possible ophthalmic dmg administration devices and methods (including mucoadhesive polymers, particulates, liposomes, ocular iontophoresis, ocular films, ocular inserts and comeal collagen shields), are described in A.K. Mitra, Ophthalmic Drug Delivery Systems, 58 Dmgs and the Pharmaceutical Sciences Series (February 1993). Fundamentals and applications of controlled dmg delivery systems are further provided in JR. Robinson and V.H.L. Lee, Controlled Drug Delivery, 29 Dmgs and the Pharmaceutical Sciences Series (2nd Edition). The comeal collagen shield is an ophthalmologic product available from such companies as Chiron and Bausch & Lomb. The shield resembles a translucent contact lens and is fabricated from bovine or porcine collagen tissue, which resembles the collagen molecule of the human eye.
All of the dmg reservoir systems and devices are effective in presenting adequate reagent to the stroma for inhibition of stromal remodeling and in the restoration of lamellar stability process.
EXAMPLE 1 - Biochemical method to maintain the structural integrity of the eye following photothermal keratoplasty surgery performed to correct ametropia.
Objective 1 :
1. Optimize riboflavin/synthetic cross-linking agent concentration, UV wavelength and time of UV exposure (if required); observe nature of cross-links
2. Determine if any necrosis and apoptosis is induced by riboflavin or other agent alone
3. Verify that riboflavin or other agent makes cross-links in cultured cells treated by LASER, and whether the necrosis and/or apoptosis induced by LASER is unaffected by subsequent riboflavin addition
Methods: Culture human comeal epithelial collagen-producing cells and determine the optimal cross-linking agent dose; preferably without inducing necrosis or apoptosis. The minimal cross-linking agent concentration, light exposure (visible and laser) and intensity required to activate the agents to induce cross-links will further be determined. The growth protocol for human comeal equivalents (i.e., containing the epithelium, stroma, and endothehum) described in Science. 286:2169-272. 1999, herein incoφorated by reference, can be used.
Cell culture (normal human comeal epithelium)
1. Add agent + UV (λ250 - 380) x 45 mn to ensure optimal activation (minimal time required will also be determined); wash extensively
2. Standard immunofluorescent methods to confirm riboflavin or other agent remains adherent to cells (e.g., primary anti-riboflavin antibodies + appropriate secondary)
3. Assess necrosis (colorimetric assay) and apoptosis (DNA fragmentation immunostain) 4. Repeat above sequence in cells with LASER exposure in middle of culture
We expect no further necrosis and/or apoptosis with riboflavin addition
Objective 2:
Rabbit eyes + LASER surgery 1. Add riboflavin and activate
2 Excise the eyes and use in one or more of the following: a. Immunohistology to identify riboflavin localization at the surgical site b. Determine cross-links via melting point analysis
Treat isolated, intact pig or rabbit eyes (generally acceptable models of study from the literature) in accordance with the methods determined in aim 1 above. The eyes are
subsequently dissected and mounted for standard immunohistochemical analyses. The basic tissue stmcture and cellular organization from control and treated eyes is compared and documented under light microscopy. The cross-linking agent localization is then determined to ensure that collagen cross-links do not occur outside of the surgical site.
Objective 3:
Repeat laser surgery on in vivo animal model (rabbits), but examine the eyes (per the above methods) to determine:
1. How long the riboflavin and cross-links persist in vivo (it may not matter- see #2 below)
2. How any further healing processes have altered the comea (histology).
3. The change in eye shape following surgery will be measured indirectly via intraocular pressure and other techniques for the determination of stmctural integrety.
Example 2 - Technique for the Confirmation of Change in the Corneal Rigidity
The stmctural integrity or rigidity inherent in the comea is imparted by the H+ bonding and intra- and inter-molecular cross-linking within its collagenous matrix. It is widely believed that surgical and/or laser intervention in attempts to modify the radius of curvature of the comea will, by its very nature, damage the cross-linking stability of the comeo-sclera. This paradoxical phase change results in an increase in distensibilty, thus predisposing to regression of the desired refractive endpoint making preoperative prediction of effect difficult, if not impossible.
The concept of remedial collagen cross-linking to post-operatively re-establish the stmctural integrity of the comea should inhibit this cause of refractive regression. The pre- clinical and clinical investigation of the efficacy of this method requires a technique for confirming the re-introduction of the comea rigidity.
Example 3 - Measuring Corneal/Ocular Rigidity
Comeal rigidity has been defined as the resistance offered by the envelope of the globe (comea) to a change in intra-ocular volume, this resistance being manifested as a change in intra-ocular pressure (IOP). It thus associates changes in volume with the pressure in the eye. It is related, in fact, to the distensibility of the globe and not to the usual connotation associated with "rigidity" or "inflexibility."
Even if the ocular rigidity were loosely regarded as a three dimensional manifestation of the essentially one-dimensional property of extensibility, it is clear that the complicating
factors at the one-dimensional level are sufficient to disturb the theoretical derivation of an adequate quantitative expression for rigidity. Nevertheless, it is to be expected that any alteration in the extensibility of the elements of the comeo-scleral envelope is likely to alter the rigidity of the eye as a whole. Thus, the ocular rigidity is reduced following intra-ocular surgery. It is raised when the globe is immersed in cross-linking reagents, such as formaldehyde, and varies with the state of hydration and temperature. Alterations in ocular rigidity of enucleated eyes is observed when immersed in solutions of different osmolarities and rigidity is increased when there is a fall in water content; the converse has been observed when the tissues were hydrated.
Previously, apart from the association with tonometry or tonography, the measurement of the rigidity is seldom of direct clinical value. With the introduction of preoperative artificial chemical cross-linking of the comea as a potential treatment of keratectasia, clinical measuring techniques for comeal stiffness are more important. A more pervasive application of the comeal stiffness measurement results from the remedial post-operative chemical cross-linking as a method to inhibit keratorefractive regression.
In its essentials, the determination of ocular rigidity involves altering the volume of the ocular contents by a known amount and measuring the accompanying change in intra- ocular pressure (IOP). Such measurements can be accurately recorded only by manometry, a technique employed for studies on the living eyes of experimental animals and on enucleated eyes, and only occasionally on living human eyes destined for enucleation.. Otherwise reliance must be placed upon tonometry.
Tonometric methods of estimating the rigidity differ merely in that the change in volume consists of a displacement of fluid from one part of the eye to another when the indentation tonometer (Schiotz) is applied and the comea is indented. "Differential tonometry" is one method whereby different degrees of indentation are made by applying different weights to the instmment. This tensile force varies with the elasticity of the comeo- sclera and has a component tangential to the walls of the globe; at the site of the comeal indentation it therefore acts outwards and upwards, tending to raises the indenting plunger integrated within the indentation tonometer.
In two eyes of the same intra-ocular pressure before tonometry, the weight of the instmment will raise the pressure to a higher level in the eye with the more rigid coat, and unless different eyes possess comparable rigidities, the tonometric readings will not accurately reflect the pressure in the undisturbed eye.
Thus, if the 5.5 gram and the 10.0 gram weights are used, a scale reading of 5.0 may be given with the first and of 10.0 with the second; according to the calibration tables of Friedenwald there has been a displacement of 12 cu mm of fluid and a rise in the intra-ocular pressure to 31.5 mm of Hg in the first case, and a displacement of 18.5 cu mm and a pressure rise to 41 mm of Hg in the second.
A combination of applanation tonometry with indentation tonometry provides an improvement upon differential tonometry. The principle of the method is the same as for differential tonometry, but the Goldmann applanation tonometer displaces a negligible amount of fluid so that the initial value is independent of ocular rigidity. This is the best available method for measuring the rigidity of the living human eye using readily available devices.
Example 4 - Confirmation of Change in Corneal Rigidity
Comeal rigidity in the steady-state should be determined and recorded. Using the differential indentation tonometry method requires the recording of the tonometric or scale units and coincident intra-ocular pressures from the Friedenwald tables for the tonometric weights of 5.5 gm and 10.0 gm. A similar measurement of intra-ocular pressure readings should be recorded using first fine indentation tonometer with a 10.0 gm. weight followed by the IOP determined by the applanation tonometer when using the combination method. The corneas should then be treated with the desired keratorefractive modality, such as laser thermal keratoplasty (LTK). Tonometric measurements should once again be recorded according to the same protocol. The appropriate remedial cross-linking reagents should then be instilled by the method of choice. The tonometric measurement protocol should then be repeated in 24 hours thus allowing remedial cross-linking to take place. Increased comeal rigidity (or the reciprocal finding of decreased comeal distensibility) determined by means of differential tonometry will be observed by reduced indentibility of the lower weighted tonometric instmment when compared with the greater weight as determined by the scale units and coincident estimated IOP (FIG. 10). This combination method merely substitutes applanation tonometric readings for that of the greater weighted indentation tonometer. A similar comparison is made. In this more accurate method, an indication of a higher IOP by the indentation tonomoter compared with the applanation method as related to the measurements determined before instillation of the cross-linking reagent following surgical intervention, clearly indicates a reduction of indentibility and, thus, an increase of the comeal rigidity. This would indicate the success of remedial cross-linking in the re-establishment comeal stmctural integrity.
As shown in FIG. 10, step (1), the indentation tonometer displaces a normal volume of intra-ocular fluid in steady-state. In step (2), LTK damages rigidity-causing collagenous cross-linking (especially the heat-labile H+ bonds) allowing greater indentation of tonometer plunger, and in step (3), chemical-induced remedial collagenous cross-linking yields normal tonometer plunger indentation confirming re-establishment of pre-laser rigidity.
Example 5 - Development of Corneal Model for Evaluation of Refractive Keratoplasty Trauma
A model to evaluate trauma has been developed in the rabbit comea to study the causes and evaluate remedies for refractive regression induced by refractive keratoplasty.
The goals: (1) to establish the parameters and model for studying the wound healing response in the rabbit comea, (2) to assess the degree of tissue damage and the wound healing response following laser refractive keratoplasty. While the major focus was placed upon the comea, the iris and trabecular meshwork has also been evaluated. Methods: Fifteen rabbits were divided into five groups. Groups I-IV were treated with LTK and then sacrificed at various periods.
1. Group I 1 hour
2. Group II 4 hours
3. Group III 24 hours 4. Group IV 48 hours
5. Group V Control
The corneas were removed and divided into 4 sections. Section B was placed into paraffin blocks, Section C was placed into OCT and Sections A & D were extracted for analysis by gelatin zymography. This study led to the following conclusions: 1) H & E staining shows marked swelling of the stroma that resolves by 4 hours with less swelling in the stromal matrix as a function of time. The epithelium has sloughed and is replaced by proliferating adjacent epithelium. By 4 hours the stroma is devoid of cells with PMN's invading through the epithelium into the stroma. By 24 hours and 48 hours, there are also additional macrophages/monocytes and a proliferation of keratocytes.
2) HSP70 antibody staining showed a definite staining pattern that was maximal at 4 hours and then gone by 48 hours adjacent to the laser treated areas. This provides a good marker for early changes.
3) Paraffin and OCT sections were stained with TGF-β antibodies and were either negative or inconclusive as wound healing markers.
) The gelatinase zymogram allowed us to follow the appearance of MMP-9. It provides an all or none marker since it is not present in the normal comea but appears after comeal wounding. It coπelated with the amount of laser trauma and resolves in 48 hours.
Standard H & E staining and TUNEL stains can be used to distinguish between necrosis and apoptosis. This is desirable in order to determine the specific inhibitory effects of various interventions in the comeal traumatic model.
Example 6 - Wound Healing in a Corneal Model
Goals:
1) Establish the parameters and model for studying the wound healing response in the rabbit comea
2) To assess the degree of tissue damage and the wound healing response following comeal coagulation using the holmium:YAG laser system.
Protocol:
1) 15 New England white rabbits were divided into 5 groups, with 3 rabbits per group.
2) Heplock catheters were placed in the ear vein, and approximately lcc of ketamine hydrochlonide (10 mg/kg) and xylazine (3 mg/kg) diluted 1 :3 with sterile salt solution was injected IV. A sterile drop of Ophthaine anesthetic was placed in the eye.
3) A lid speculum was placed in the eye, the rabbits were hand held at the HoNAG laser and each eye was treated. Groups I-IV were treated with LTK.
4) The laser settings for each eye are in table 1. 5) After treatment, 1 drop of Ciloxin was placed into each eye.
6) At at 1, 4, 24, and 48 hours, the rabbits in Groups I-IV respectively were sacrificed by subcutaneous injection of 2cc of ketamine/xylazine followed by intracardiac injection of 1.5cc of Euthasol (lcc/10 lbs).
7) Corneas were removed, rinsed in BSS and cut into 4 pieces (see below). 8) One central piece was placed into OCT and frozen in liquid nitrogen.
9) The other central piece was placed into 10% formalin for 2 hours and then processed for paraffin embedding.
10) The other pieces were frozen at -70 C.
11) Tissue is processed using TUNEL assay and immunochemistry with antibodies to specific proteins involved in the apoptosis pathway. Other tissue is prepared for TEM.
12) While major focus is on the comea, iris and trabecular meshwork are also evaluated.
The preceding description has been presented with reference to presently prefeπed embodiments of the invention. Workers skilled in the art and technology to which this invention pertains will appreciate that alterations and changes in the described stmcture may be practiced without meaningfully departing from the principal, spirit and scope of this invention.
Accordingly, the foregoing description should not be read as pertaining only to the precise stmctures described and illustrated in the accompanying drawings, but rather should be read consistent with and as support to the following claims which are to have their fullest and fair scope.