Expanding the Corneal Topical Therapy Arsenal

The eye care world is witnessing a surge in promising topical medications for various corneal conditions. Here, I focus on three topical medications that appear to be low risk options for their respective recalcitrant target conditions: 1) topical losartan for corneal haze/scarring; 2) topical insulin for persistent epithelial defects (PEDs); and 3) OK-101 for corneal neuropathic pain (in clinical trials). Although both topical losartan and insulin appear safe according to the existing literature, use of off-label medication requires clinical need, appropriate evidence, and practitioner experience in the condition being treated.¹

LOSARTAN

Losartan is an angiotensin-converting enzyme II receptor antagonist that disrupts intracellular transforming growth factor-beta (TGF-Β) signaling pathways to reduce corneal haze and scarring.² TGF-Β isotypes are perhaps the most pivotal growth factors for myofibroblast activity; they are responsible for generation of stromal fibrosis and corneal haze in wound healing and inflammation.³ The most effective way to eliminate corneal scarring (and its effect of scattering and blocking light rays) is corneal transplantation (including diffuse anterior lamellar keratoplasty and penetrating keratoplasty), while somewhat less invasive excimer laser phototherapeutic keratectomy can also be effective for more superficial scars.^4,5

However, topical losartan has shown good penetration to the posterior stroma through an intact corneal epithelial layer and basement membrane. Rabbit studies have illustrated that small losartan concentrations of 0.1 mg/mL to 0.8 mg/mL in a balanced salt solution dosed six times daily reduced corneal scarring fibrosis secondary to insults involving the anterior and/or posterior cornea.² Topical losartan is simple to compound and is already being used off label at 0.8 mg/mL by corneal specialists.⁶ It costs $90 to $150 per month in the United States, and pure losartan powder (vs crushed tablets) must be used to create the solution. Visual improvement may take place before a clinically detectable reduction in haze occurs,⁷ and treatment duration of 4 to 5 months is suggested before deeming the treatment ineffective.² Pereira-Souza et al published a case report of one eye with scarring post-LASIK and diffuse lamellar keratitis showing an increase in VA from 20/200 to 20/30 after 4.5 months of topical losartan 0.8 mg/mL six times daily.³

It seems plausible that topical losartan may help reduce scarring and haze associated with various etiologies such as PEDs, chemical burns, herpetic stromal keratitis, bacterial keratitis, traumatic or surgical lacerations or incisions, and scarring associated with keratoconus.⁸ TGF-Β–induced corneal dystrophies (including Reiss Bucklers, Thiel-Behnke, lattice type 1, and granular types 1 and 2) may also respond to topical losartan slowing TGF-Β–driven accumulation of mutant protein deposits in the cornea. Clinical trials are needed to demonstrate degree of efficacy in each of the above conditions.²

A Word of Caution

In my experience working with corneal specialists who use topical losartan, the particular “home-run” result such as the one achieved in a post-LASIK eye by Pereira-Souza et al should not be the expectation. Instead, partial rather than full clearing of the scarring and haze should be considered a success. Although corneal fibrosis that has been present for a period of months is still considered dynamic, we do not yet know whether treatment with losartan can help clear fibrosis and haze that has been present for years.

Compliance at six drops per day for several months is admittedly difficult for patients to follow through with. At my institution, many patients wear specialty contact lenses to improve vision through their scars, so being compliant with six drops per day (often on top of other topical medications for pre-existing conditions) when not wearing the lens is difficult.

Clinicians should avoid prescribing concentrations or drop schedules of losartan that have not yet been demonstrated as safe in animal models or human case reports.² One drop of 0.8 mg/mL six times a day (with eventual taper as discussed below) is the current recommendation. Excessive doses could interfere with TGF-Β’s role in the physiological functioning of other ocular structures, including the trabecular meshwork, retina, and lens. Note that 0.8 mg/mL equates to 0.08% when converted. Losartan should not be used in patients who are pregnant; pregnancy screening is thus recommended in applicable patients.²

It is possible that changes in the formulation of losartan drops to a pH other than the 6.7 to 7.0 range used in animal studies could influence treatment effect, lead to unforeseen side effects, or cause irritation upon instillation. Thus, stressing the importance to the pharmacy of using pure losartan powder and keeping pH within the above range is suggested. From myopia control studies with low-concentration topical atropine, it appears that formulation, including pH, can vary widely across different pharmacies.⁹ Finally, avoid initiating topical losartan within 1 month after traumatic or surgical lacerations, as some myofibroblast activity is needed for wound healing to reduce risk of perforation.²

Ideal Treatment Course

The best course of treatment for topical losartan remains unknown. Wilson et al recommended maintaining topical losartan at decreased dosing frequency for 6 months after significant clearing of the haze, given the dynamic nature of the corneal stromal scars. Otherwise, TGF-Β may resume reaching the stroma in a high enough quantity to drive myofibroblast activity after drop cessation. After 6 months of decreased frequency dosing, observation is recommended; if the haze returns, a more extended losartan topical treatment course may be needed. Lamellar or penetrating keratoplasty may be required for scars in corneas in which regeneration of the epithelial basement membrane or Descemet membrane has not occurred post-injury, allowing unguarded TGF-Β access to the corneal stroma.²

Soon, 0.8 mg/mL losartan may be a commonplace noninvasive option to offer before considering corneal transplantation in patients with scarred corneas not at risk of perforation. There is a long-term risk of rejection, failure, or graft dehiscence even after uneventful keratoplasties,¹⁰ and a long dose course of topical losartan might offer a less invasive approach than keratoplasty.² In addition, topical losartan may find a place in combination with other rehabilitation tools, such as a corneal or scleral gas permeable contact lenses, to help improve visual quality in eyes with residual scarring and haze.^11-13 Indeed, post-keratoplasty, there is still the decent chance patients will require rigid lenses anyway. Furthermore, there are etiologies of scarring that are ill-suited for keratoplasty due to increased risk of failure or rejection in the presence of neovascularization and/or limbal stem cell deficiency.

Research is also needed regarding whether topical losartan can penetrate the conjunctiva and partially help reverse conjunctival fibrotic changes in conditions such as trachoma, Stevens-Johnson syndrome, ocular cicatricial pemphigoid, and ocular graft-versus-host disease.²

INSULIN

Although used at high concentrations (25-100 IU/mL) in earlier decades, topical insulin is making a comeback at significantly lower concentrations to aid in healing PEDs in neurotrophic corneas.¹⁴ Several recent literature reviews have shown the efficacy of low-concentration topical insulin in helping to heal PEDs.^14-16 The higher insulin doses do not appear to confer greater treatment efficacy; therefore, a cautious approach to reduce any potential (albeit low) risk of toxicity is considered optimal.¹⁴

Insulin likely mediates an epithelial curative effect by binding to corneal receptors for both insulin and insulin-like growth factor. This receptor binding may lead to activation of ERK and PI-3K pathways within corneal epithelial cells that encourage repair and limit autophagy.^14,15 Insulin has also been shown to promote corneal epithelial cell migration to close wounds.¹⁴

The application of one 1 IU/mL drop of fast-acting insulin four times daily appears to be the most common dosing strategy.^16,17 Fast-acting insulin may be more effective than intermediate or long-acting insulin, as insulin has low conformational stability and could thus be easily degraded by enzymes present on the ocular surface.¹⁵ It is recommended that topical insulin be dissolved in polyethylene glycol (PEG) 400 instead of saline by the compounding pharmacy because PEG may increase tear film stability and help reduce the risk of insulin proteolysis.¹⁴ A preservative-free formulation is also preferred, as preservatives may hinder re-epithelialization. Storage in the refrigerator is preferred, as it can keep topical insulin stable for longer (about 28 days) compared with room temperature (about 2 weeks).

In the Literature

Case series and reviews have illustrated topical insulin contributing to PED resolution in patients with^16,18 and without diabetes,^19,20 while other studies have included both patients with and without diabetes.^17,21 Note that it is impossible to attribute all the success of healing PEDs to insulin alone because study participants with refractory PEDs had already undergone treatments with conventionally used agents such as artificial tears, antibiotics, antivirals, antiinflammatory medications, autologous serum, amniotic membranes, and tarsorrhaphy.

Soares et al conducted a retrospective study showing that 1 IU/mL of topical insulin four times daily helped achieved complete re-epithelialization in 19 of 21 patients with stage 2 or 3 neurotrophic keratitis and refractory epithelial defects between 7 and 45 days. Four of these patients had diabetes.²¹ Diaz-Valle et al prospectively showed that 17 of 21 patients with refractory PEDs re-epithelized by a median of 23 days after starting 1 IU/mL topical insulin four times daily. In the four lingering patients, an average reduction in PED size of 91.5% was still obtained.²²

Diaz-Valle et al later published a retrospective study comparing topical insulin (n = 61) with autologous serum control (n = 23) in the treatment of PEDs recalcitrant to other treatment.²³ Complete re-epithelization occurred in 51 patients (84%) treated with insulin and 11 patients (48%) with autologous serum (P = 0.002). The control group had a higher PED recurrence rate (43%) versus the topical insulin treatment group (11%).²³ In addition, amniotic membrane transplantation was ultimately required significantly more frequently in the autologous serum group.²³

More randomized control trials should be conducted to evaluate the efficacy of low-concentration topical insulin. However, because it is relatively inexpensive, easy to compound, and well-tolerated, it may emerge as a treatment option in patients with PEDs who are not responding to conventional treatments. Some investigators, such as Diaz-Valle et al, even recommend it as first-line therapy.²³

OK-101

OK-101 (OKYO Pharma) is the first drug to receive investigational new drug clearance by the FDA for clinical studies involving patients with corneal neuropathic pain (CNP),²⁴ a condition distinct from neurotrophic keratitis, for which cenegermin ophthalmic solution 0.002% (Oxervate Dompé) is already approved. A phase 2 clinical trial of OK-101 for the treatment of CNP patients is ongoing (n = 40). CNP is a rare but highly recalcitrant and quality-of-life-limiting condition, and as an orphan disease, it is possible the reviewal process may be expedited by the FDA.²⁵

While waiting for on-label medications for CNP, remember that sleep is crucial to address in patients with severe ocular surface symptoms, and success with scleral lenses (even when near perfectly fit) varies patient to patient.²⁶

THE FUTURE HOLDs PROMISE

Topical losartan and insulin each show promise in reducing corneal haze and helping heal recalcitrant persistent epithelial defects, respectively. Both are available for off-label use after compounding but still require further clinical study. To this point, OK-101 has reduced CNP in mouse models, as well as stinging and burning in human patients with dry eye.²⁷ Clinical trials in patients with CNP are ongoing and may offer an on-label option to address this very difficult-to-treat condition.