1Joanne Shen; 2Qing Wu; 3Nandita KheraDepartments of 1Ophthalmology, 2Research Biostatistics and 3Hematology/Oncology – Mayo Clinic – Scottsdale, Arizona

Purpose

Pilot Study of Intense Pulsed Light for Improvement of Severe Dry Eye Symptoms in Subjects with Ocular Rosacea Related to Inactive Graft-Versus-Host Disease

Results

• Ocular rosacea and meibomian gland disease is a major contributor to keratoconjunctivitis sicca after chronic graft-versus-host disease (GVHD).

• GVHD exacerbates the mucin defi ciency from conjunctival goblet cell destruction and aqueous defi ciency caused by damage to lacrimal and accessory lacrimal glands similar to the cholestasis damage in the biliary system.

• There is an extreme unmet need in the understanding and treatment of keratoconjunctivitis sicca which develops after conjunctival scarring from chronic ocular GVHD.

• Many chronic GVHD patients recover without signifi cant painful symptoms of irritation, chronic foreign body sensation and light sensitivity.

• However, a minority of patients are permanently disabled from their symptoms.

– Their cancer is gone, but their quality of life is signifi cantly altered.

• Traditional treatments are not suffi cient for these severely affected patients and include:

– Preservative free artifi cial tears – Moisture chambers – Punctal plugging and cautery – Topical cyclosporine drops – Oral tetracycline derivatives – Autologous serum tears – Ocular contact lens prosthesis – Lateral tarsorrhaphy – Conjunctival surgery – Corneal transplantation • Lipifl ow and intense pulse light (IPL)

combined with meibomian gland expression has been used for treatment for ocular rosacea dry eye at Mayo Clinic in Arizona and has been very promising for patients without systemic disease.

• IPL treatment under a research protocol was offered as a pilot study to see if dry eye syndrome could be improved for patients with quiescent ocular GVHD and to ensure there were no unexpected treatment sequelae.

Methods

Conclusions

• Average subject age: 49 years

• 5 women and 3 men

• 1 male subject developed active GI GVHD and had to drop out of the study at month 3

• All subjects had conjunctival sub epithelial fi brotic scarring from GVHD

• 6 out of 7 subjects had superior limbic keratoconjunctivitis

• Overall, the subjects experienced an 86% improvement in total number meibomian glands yielding liquid secretions on the lower eyelids which almost reached statistical signifi cance in this small pilot study

• The average SPEED2 score dropped 57.1% (p=0.003), and the OSDI also dropped 57.1% (p=0.027) after the IPL treatments

• Most of the subjects with severe meibomian gland loss and 1 subject with atrophic changes on glands from meibomography did not improve with IPL treatment

• 3 subjects that had signifi cant symptom improvement had mild, moderate and severe meibomian gland loss

• No vision loss or ocular side effects were observed during the course of the study

Dry eye in GVHD is a multifactorial ocular surface disease affecting all contributors to tear fi lm composition. A multifactorial approach for treating this severe form of dry eye is needed to improve quality of life for these bone marrow transplant patients. This pilot study shows symptomatic improvement of dry eye symptoms after IPL treatment in patients with GVHD after BMT. This pilot study is limited by small sample size, lack of case control, and single observer bias.

The symptom improvement may be confounded with use of punctal occlusion for severe fi lamentary keratitis during the course of the study. However, the meibomian gland evaluation improved over the treatment course, which would not be expected with punctal occlusion. Severe abnormalities on meibography is a negative prognostic indicator for symptom improvement. Long-term follow-up is planned to evaluate the duration of symptom and gland improvements.

References1. Blackie CA, Korb DR, Knop E, Bedi R, Knop N, Holland EJ. Nonobvious meibomian gland dysfunction. Cornea. 2010 (29): 1333-1345.2. Blackie CA, Korb DR. The diurnal secretory characteristics of individual meibomian glands. Cornea. 2010 (29): 34-38.3. Eom Y, Lee J, Kang S, Kim H, Song J. Correlation between quantitative measurements of tear fi lm lipid layer thickness and meibomian gland loss in patients with obstructive meibomian gland dysfunction and normal controls. Am J Ophthalmol. 2013; 155(6): 1104-10.4. Friedman NJ. Impact of dry eye disease and treatment on quality of life. Curr Opin Ophthalmol. 2010; 21: 310-16.5. Jabs DA, Wingard J, Green R, Farmer ER, Vogelsang G, Saral R. The eye in bone marrow transplantation. III. conjunctival graft-vs-host disease. Arch Ophthalmol. 1989; 107: 1343-48.6. Johnson DA, Jabs DA. The ocular manifestations of graft-versus-host disease. Int Ophthalmol Clin. 1997; 37(2): 119-33.7. McGinnigle S, Naroo S, Eperjesi F. Evaluation of dry eye. Surv Ophthalmol. 2012; 57(4): 293-316.8. Ngo W, Situ P, Keir N, Korb D, Blackie C, Simpson T. Psychometric properties and validation of the standard patient evaluation of eye dryness questionnaire. Cornea. 2013; 32(9): 1204-10.9. Tam PMK, Young AL, Cheng LL, Lam PTH. Topical 0.03% tacrolimus ointment in the management of ocular surface infl ammation in chronic GVHD. Bone Marrow Transplantation. 2010;45: 957-958.10. Toyos, R. Intense, pulsed light of dry eye syndrome. Cataract & Refractive Surgery Today. 2009 April: 71-73.

Support Detail

• 8 subjects status post bone marrow transplant with inactive GVHD with ocular rosacea, meibomian gland dysfunction, and severe dry eye symptoms were prospectively examined and treated with IPL and lid expression under approved institutional review board (IRB) protocol.

• Subjects' GVHD was deemed inactive by the treating hematologist.

• All subjects had failed to respond to conventional therapies.

• Subjects were evaluated by a single ophthalmologist with SPEED2 and OSDI symptom surveys, visual acuity, intraocular pressure, slit lamp examination with lissamine green and fl uorescein vital dye staining, Schirmers testing, lipid tear fi lm analysis, tear osmolarity, tear meniscus height, infrared meibomography,and tear break up time at baseline, 1, 2, and 3 months post treatment.

• Subjects received IPL and lid expression at baseline, months 1, 2, and 3.

• Patients who demonstrated fi lamentary keratitis and poor Schirmer’s scores underwent punctal cautery/hyfrecation to improve aqueous defi ciency.

• Paired t-test analysis was performed on SAS software.

Average Baseline ± SD 1 month p-value 3 months ± SD p-value

Total MGE 6.71 ± 3.99 11.57 0.120 12.71 ± 7.13 0.060

SPEED2 20.71 ± 3.04 18.14 0.060 15.71 ± 2.21 0.003

OSDI 57.61 ± 16.59 40.48 0.189 38.79 ± 12.90 0.027

Normal Glands

Mild Gland Drop Out

Moderate Gland Drop Out

Severe Gland Drop Out

Atrophic Glands

ProgramNumber2017

Sub Epithelial Conjunctival Fibrosis of Upper Tarsus

Obstructed Meibomian Glands

Expression of Turbid Meibum

Expression of Toothpaste-Like Meibum

Mayo Clinic Cancer CenterDiscretionary Funds, Mayo Clinic Department of Ophthalmology Discretionary Funds, Mayo Foundation for

Medical Education and Research