Efficacy and Safety of Processed Amniotic Fluid (pAF) Drops for the Treatment of Ocular Chronic Graft-Versus-Host Disease

Background: Ocular graft-versus-host disease (GVHD) affects more than 50% of patients with chronic GVHD and can lead to debilitating dry or painful eyes, impaired activities of daily living and poor quality of life. Management is mainly supportive, through control of surface inflammation, lubrication, and control of drainage and evaporation. Amniotic fluid (AF) is a rich source of nutrients, cytokines, and growth factors that has antimicrobial and immunomodulatory activities and may have reparative and regenerative roles in human disease. Investigators at the University of Utah developed protocols to collect AF from full-term pregnant women, process and sterile filter AF to separate cytokines and growth factors from residual cells and debris contained in the no-cellular fraction of AF, and treat patients with different conditions. We report the treatment effects and safety of processed AF (pAF) for the treatment of ocular GVHD.

Methods: In this randomized, double-blinded, placebo-controlled phase II trial (NCT# 03298815), patients age ≥ 18 years diagnosed within 5 years after allogeneic hematopoietic transplant with ocular GVHD causing dry eye symptoms requiring lubricant drops >3x/day, punctal plugs, thermally cauterized puncta or special eyewear to relieve pain; or inability to work because of ocular symptoms; or loss of vision due to keratoconjunctivitis sicca were enrolled. Patients with any other chronic GVHD treated with >3 lines of therapy beyond corticosteroids with or without calcineurin inhibitors or sirolimus; or had a difference in dryness between both eyes of >2 points by the International Dry Eye WorkShop (DEWS) 2007 report; or were not willing to discontinue the use of medicated lubricant eye drops prior to randomization were excluded. Randomization was performed within participants, with one eye receiving one drop of pAF and the other eye saline treatment (placebo) twice daily for 30 days. In patients wearing scleral lenses, the eye drops were applied prior to lens insertion in the morning and after removal at night. Patients were allowed to resume their medicated eye drops after the 30-day treatment period. The primary objectives were to determine the day 30 overall response (complete and partial response) and safety of pAF. Response type to treatment is defined in Table 1. Key secondary objectives included longitudinal changes in visual acuity and the corneal surface, and patient-reported outcomes.

Results: All fifteen patients' eyes were randomized to either pAF or saline treatment. The median age was 61 years (range, 35-72). At enrollment, six patients wore scleral lenses. The median baseline FACT-G score was 88 (range, 58-104). Baseline eye characteristics by treatment are shown in Table 2. Baseline NIH CC eye scores were similar between pAF and saline-treated eyes. The saline-treated group had more eyes with DEWS score 3 (n=7) compared to the pAF group (n=2). All 15 patients completed treatment. The overall response at day 30 in the pAF and saline-treated groups was 25% and 50%, respectively, p=0.38. All responding eyes had a partial response.

No treatment-related serious ocular adverse events were observed. Two minor events occurred in the pAF treated eye group, including a change in eyelid position (n=1) and stinging after study drop application (n=1). There was no significant change in visual acuity and corneal surface staining between treatment groups at 30, 60, and 100 days. There was no significant change in the FACT-G score at 30, 60, and 100 days, and there was no significant difference in patient reported eye pain between pAF vs. saline-treated eyes over time (p=0.21).

Conclusion: In this study, the primary endpoint of safety was met as there were few adverse events. One quarter of pAF-treated eyes responded. These findings justify evaluating pAF for ocular GVHD in a larger study. Limitations of this study included restricting patients to pAF application to twice a day and the study design of randomizing eyes within each subject as several patients had mild asymmetric disease between the two eyes. Other challenges included having delayed ocular assessments and/or altered perceptions of quality of life due to the COVID19 pandemic.