Abstract
Purpose :
In a recent clinical trial (Pinto-Fraga, Ophthalmology 2015) we concluded that topical 0.1% fluorometholone therapy reduced ocular signs in dry-eye (DE) patients and prevented exacerbation caused by exposure to desiccating stress (DS). We now identify severity, activity (i.e. response to DS) and therapeutic tear biomarkers (tBM) in these patients using linear mixed effect statistical models.
Methods :
41 DE patients were recruited (visit 1, V1) and randomized to receive either topical 0.1% fluorometholone or its vehicle polyvinyl alcohol, 4 times daily for 21 days (V2). Immediately after, patients were exposed to DS in a controlled environment laboratory for 2h (V3) which increases DE activity as previously reported (Lopez-Miguel, Am J Ophthalmol 2014). Clinical evaluation and tear collection were performed at each visit. A panel of 18 potential tBM (EGF, IFNγ, IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-12, IL-13, IL-1RA, IL-17A, IP-10, RANTES, TNFα, MIP-1α, MMP-9, MCP-1) was analyzed (Luminex IS-100). Multidimensional Scaling (MDS) was performed using tBM levels at V1 to determine the pattern of similarities among patients. Linear Mixed Effect models were used to study the influence of disease severity, visit, treatment or their interactions.
Results :
MDS divided patients into 2 groups based on differences in EGF, IFNγ, IL-8, MMP-9, and RANTES levels (p≤0.013). These groups showed different clinical severity in Schirmer test (p=0.0028) and conjunctival (p=0.0022) or corneal staining (p<0.0001), thus separating moderate from severe DE. Significant (p≤0.0466) changes in EGF, IL-13, IL-1RA, IL-2, IL-4, IL-8, IP-10, MMP-9 and TNFα were found among visits and varied significantly when disease activity increased under DS (except IL-4). IL-1RA, IL-2 and TNFα were differentially affected by visit depending on treatment (p≤0.0244). IL-1RA and MMP-9 had different evolution depending on disease severity (p≤0.045). IL-13 increased (p=0.0049) in fluorometholone-treated patients.
Conclusions :
We identified several types of tBM in our setting: disease severity (IL-1RA, IL-8, EGF, IFNγ, MMP-9, RANTES), disease activity i.e. response to DS (EGF, IL-13, IL-RA, IL-2, IL-8, IP-10, MMP-9, TNFa) and therapeutic (IL-1RA, IL-2, IL-13, TNFa). This trial design using DS and tBM is useful to select target patients and to better define disease activity and evaluation end-points in clinical trials.
This is an abstract that was submitted for the 2016 ARVO Annual Meeting, held in Seattle, Wash., May 1-5, 2016.