This double-masked randomised clinical trial compared the response of dry eye disease (DED) to treatment with topical steroids in patients with and without graft vs. host disease (GVHD). Forty-two patients in a single centre (n=21 with and n=21 without GVHD) who had moderate to severe DED were treated with loteprednolol etabonate 0.5% or artificial tears twice daily for four weeks. They were assessed using the Ocular Surface Disease Index (OSDI) and clinical examination corneal fluorescein staining (CFS), lissamine green staining, tear break up time and Schirmer test before and after treatment. At base line there was no significant difference between the groups. After four weeks the non GVHD patients who received steroid treatment OSDI score decreased by 34% where as in GVHD patients there was minimal change in the OSDI score. The average CFS score also decreased by 41% in those patients without GVHD and had minimal effect on those that had. In addition treatment with artificial tears resulted in 22% decreased OSDI and 32% decrease in CFS scores in non-GVHD patients, and had minimal effect in patients with GVHD. It was concluded that DED patients that have ocular GVHD have a poor response to low dose topical steroid compared to those without ocular GVHD even with similar baseline disease. This study only had a small sample size from one centre. Further research is needed using more frequent dosing, longer duration, stronger steroids and other anti-inflammatory agents.