5FU vs 5FU With Viscoelastic Formulation for the Prevention of Scarring Post-trabeculectomy

Overview

Trabeculectomy is the most effective method of lowering intraocular pressure in glaucoma that is not well controlled with drop medication. The post-operative wound healing response remains the major barrier in surgical success. Scar tissue formation results in occlusion of the surgical drainage site created. As a consequence, the intraocular pressure rises once again. The current use of anti-scarring agents such as 5 Fluorouracil (5FU), is a well established method employed by ophthalmologists to clinically reduce the postoperative scarring response. However, in patients with a high risk of bleb failure, repeated subconjunctival injections of 5FU following bleb needling is often required resulting in increased clinical load and patient inconvenience. A slow release formulation of 5FU would be of benefit by releasing the 5FU into the subconjunctival space over an extended period thereby providing a prolonged therapeutic effect. In addition, hyaluronic acid is known to possess antifibrotic properties and from its gel-like physical state, would potentially act as a physical tissue spacer that may further limit active subconjunctival scarring at the site where it is injected. We hypothesise that injecting a mixture of 5FU with hyaluronic acid would give better outcomes than injecting 5FU alone. The aim of this study is to determine whether bleb needling with subconjunctival injection of Viscoelastic/5FU formulation is an effective method for prolonged drug delivery in preventing the post-operative scarring response. This will be a prospective case-controlled study involving patients who have already undergone trabeculectomy and who require bleb needling and subconjunctival 5FU injections.

Specific Aims The aim of this study is to determine whether needling of the conjunctival bleb after trabeculectomy with subconjunctival injection of a 5 Fluorouracil (5FU)/ viscoelastic formulation is an effective method for prolonged drug delivery in preventing post-operative scarring in the conjunctiva. Study Design Cell Culture Experiments Cell culture experiments were carried out to determine the effect of the hyaluronic acid, at varying commercially available concentrations, on human Tenon's fibroblasts, the target cell type of our proposed HA/ 5FU mixture. We also perform tests to examine the pharmacokinetic drug release profiles of the different 5FU/HA formulations. This allowed us to determine the most effective formulation to use for patients in the clinical arm of the study. Briefly, human Tenon's fibroblasts attained from tissue explants from cadaver eyes were grown until confluent in fetal calf serum containing the standard antibiotics for tissue culture, gentamicin and amphotericin, and placed in an incubator at 37°C under 95% humidity with 5% CO2. The cells were passaged when confluent and used for experiments between passages 3-6. Alpha smooth muscle actin expression Tenon's fibroblasts were seeded on 24 well plates and incubated in fetal calf serum (FCS) containing the standard antibiotics for tissue culture, gentamicin and amphotericin, and placed in an incubator at 37°C under 95% humidity with 5% CO2. The culture medium was removed after 24 hours and replaced with fresh medium described below. Collagen I production Standard 3-D collagen I gels seeded with fibroblasts for gel contraction assays were employed to quantify collagen production by fibroblasts treated with different HA concentrations. Gels treated with TGF beta (a known stimulator for collagen I synthesis and inducer of alpha smooth muscle actin expression ie myofibroblast phenotype) acted as positive control. Type I collagen derived from rat tail were mixed with culture medium and following confluency, trypsinised cells were resuspended in this mixture and cast in a 24 well plate and allowed to polymerise in an incubator at 37°C. After polymerisation, the gels were gently detached from the plates in which they are cast and suspended in fresh medium described below All experiments were conducted in triplicate and repeated 3 times. Human Tenon's fibroblasts were cultured in fresh media containing HA alone, 5-FU/HA (HA concentrations10, 14, 23 mg/ml) or cultured without HA or 5FU (controls). Fibrosis and scarring response of the myofibroblasts was assessed through Collagen I and alpha smooth muscle actin (ASMA) expression which were quantified by Real Time -Polymerase Chain Reaction (RT-PCR). WST-1 assay determined cell viability. 5-FU solution or powder was dispersed in HA and mechanically stirred and then vortex mixed. Our results showed that HA markedly reduced ASMA and collagen I mRNA expression by myofibroblasts compared to controls without HA. Treatment with HA following 5FU (10mg/ml) pretreatment resulted in a greater reduction in ASMA and collagen I expression compared to HA treatment alone, and statistical significance reached with 23mg/ml HA (p\<0.05). Cell proliferation and viability was not affected by HA. Our team of investigators has completed the experiments examining the release of 5FU from the 5FU/ HA mixture. 5FU solution or powder was mechanically stirred and vortex mixed in HA and the release of 5FU was quantified using high performance liquid chromatography (HPLC). The concentration of 5FU used was 10mg/ml and the concentration of HA used was 10, 14 and 23 mg/ml. The 5FU/HA mixture delayed drug release to 48 hours, with HA of 14mg/ml and 23mg/ml showing a greater delay in drug release than 10mg/ml, (p\<0.05). There was no significant difference in rate of drug release between powder and solution form of 5FU although the pH was notably different ie 6.6 and 10 respectively. The results of our in-vitro studies and drug release assays demonstrate that this 5FU/HA mixture has the desired therapeutic effect on Human Tenon's myofibroblasts. The optimum concentration for HA is 23mg/ml and it would be combined with 5FU 10mg/ml. We expect that 5FU would be released from the mixture over 48 hours; both HA and 5FU will have anti-scarring effects in the subconjunctival space. Patient Recruitment: Patients will be recruited and randomized at the first postoperative visit where a 5-FU injection is indicated according to clinical parameters indicating active scar formation. (usually after the first post-operative month when suturelysis has been completed). All post operative care apart from the 5FU/viscoelastic mixture injection in the control group will be unchanged from current treatment standards. At least 50 patients who have under gone trabeculectomy at the Singapore National Eye Centre will be recruited in the study from glaucoma clinics at Singapore National Eye Centre. This will consist of different types of glaucoma. 25 patients will receive subconjunctival injection of 5FU after bleb needling while 25 will receive injections of a 5FU/ viscoelastic formulation. Patient assessment and examination at baseline All patients (cases and controls) will have the following examination: Snellen visual acuity, slit-lamp examination of the anterior segment, bleb and intraocular pressure measurement with a Goldmann tonometer. Tonometry will also be performed pre injection using a pneumatonometer (Reichert) according to manufacturer's guidelines. Colour photographs of the bleb will be captured and stored digitally for masked bleb grading. The Visante OCT will be used to image the bleb morphology in dark conditions. It is a new non-contact instrument that rapidly obtains high-resolution images of the angle. The image capture scan takes less than 10 seconds and is akin to taking a photograph. The device allows qualitative and quantitative angle imaging, which is objective and reproducible. This will take place prior to the needling procedure. Bleb needling \& subconjunctival injection Having taken informed consent from the patient, the patient will be assigned into either the case or control group. In brief, needling of the conjunctival bleb will be carried out at the slit lamp using an aseptic technique. The doctor performing the procedure will use sterile gloves and clean the patient's eyelids and peri-orbital skin with povidone iodine 10%. The eye will be anaesthetised with Guttae Tetracaine 0.5% and a wire speculum inserted to separate the eyelids. Povidone iodine is the inserted into the conjunctival sac. 0.1ml of lignocaine 2% is injected into the subconjunctival space, temporal or nasal to the bleb before needling of the bleb is performed with a 27 guage needle attached to a 1ml syringe. 5FU solution (10mg/ml) (control arm) or 5FU/viscoelastic formulation (10mg/ml of 5FU in 23mg/ml HA, (50/50) volume/volume) (experimental arm) is injected posterior to the bleb after needling is complete. Injections are given within the same quadrant as the bleb. Post injection the conjunctival sac is thoroughly rinsed and re-examined on the slit-lamp. The IOP is checked at the end of the procedure using the Goldmann tonometer and the pneumotonometer. Patients will receive current standard of care with respect to intraoperative antifibrosis therapy and the timing and frequency of postoperative 5-FU. Further administration of subconjunctival 5-FU solution or 5FU/HA formulation, (depending on the treatment arm the patient was initially assigned to) at subsequent office visits will be administered when required according to current clinical practice for patients in both treatment arms. This will be recorded in the study. Follow-up examinations Bleb imaging with the Visante anterior segment OCT and IOP measurements will be recorded at every visit ie week 1, and 12. The need for further needling or other intervention eg massage/suture removal/lysis (if relevant) will be clinically decided by the ophthalmologist at each visit and bleb imaging will be performed before any additional procedures perfomed if it falls on a research visit ie week 1 or 12. Bleb grading from slit lamp photographs will be performed at baseline pre-injection and at week 12.