New tools in dry eye disease researcha more efficient clinical trial design using controled environment and molecular biomarkers, and a new clinical questionnaire

Resumen

Objectives: The main aims of this doctoral thesis were 3: 1) to study the usefulness of a new two-step design of clinical trials using a controlled environment chamber to evaluate the safety and efficacy of new dry eye disease (DED) therapies; 2) to evaluate the effect of a common DED therapy on clinical symptoms and signs and tear inflammatory molecule levels at different time points (i.e. pre- and post-treatment, pre- and post-adverse environmental condition [ACE; 23°C temperature, 5% relative humidity, 0.43 m/s localized airflow] exposure), identifying different biomarkers (disease severity, therapeutic efficacy, and disease activity); 3) to analyze why clinical symptoms usually fail at translating what patients feel and to develop a new and simpler questionnaire that can detect changes in DED-related symptoms between two time-points in an easier and yet more accurate way than the current questionnaires. Methodology: To meet the first and second aims, a single-center, double masked, randomized, vehicle-controlled, phase II clinical trial was conducted, assessing the efficacy of topical 0.1%-fluorometholone in moderate-to-severe DED patients for ameliorating the worsening of the ocular surface when exposed to an adverse environment. A total of 41 patients randomly received one drop 4 times daily of either topical 0.1%-fluorometholone (FML group) or polyvinyl alcohol (PA group) for 22 days. During the 4 visits of the study (V1, day 0, baseline / V2; day 21, pre-ACE exposure / V3, day 21, post-ACE exposure / V4, day 22, 24h post-ACE exposure) DED signs and symptoms were evaluated. Also, tear samples were collected at the beginning of each visit for further analysis. An immune bead-based array analyzed the concentrations of 18 molecules (EGF, IFN-γ, TNF-α, IL-1β, IL-1RA, IL-2, IL-4, IL-6, IL-8/CXCL8, IL-10, IL-12, IL-13, IL-17A, IP10/CXCL10, MCP-1/CCL2, MIP-1α/CCL3, RANTES/CCL5 and MMP-9). Multidimensional scaling (MDS) used molecule concentrations at V1 to determine the pattern of similarities among patients. A linear mixed effect model analyzed the influence of visits, treatment, and severity on changes in molecule concentrations. Multiple comparisons analysis was used to eliminate the effects multiple statistical tests. Regarding the third aim, we carried out a prospective observational study in collaboration with the University of Cologne to perform a clinical-based evaluation of the Evaluation of Change in Symptoms-Questionnaire (ECS-Q), the new tool proposed. This study consisted on two visits (V1, baseline / V2, follow up after therapy) in which DED-related signs and symptoms were evaluated, the latest ones using different questionnaires, OSDI, SANDE II, and ECS-Q. The new ECS-Q had 2 parts: ECS-Q1, that had three possible answers as to how to grade change after therapy in V2 (better, worse, equal), and ECS-Q2 that quantified the change observed. Results: After 21-day treatment, the FML group had greater improvements in corneal and conjunctival staining, hyperemia, and tear break-up time than did the PA group. After ACE, the percentage of patients having a ≥1 grade increase in corneal staining was significantly higher in the PA group (63.1% vs 23.8%, respectively). Additionally, the FML group showed no significant changes in corneal staining, conjunctival staining, and hyperemia after the ACE exposure, while the PA group showed a significant worsening in corneal staining, conjunctival staining, and hyperemia. The FML group maintained V2 corneal staining scores, while the PA group did not recover the previous status 24h after the exposure. There were no adverse events. Regarding the tear samples analysis, MDS divided patients into two groups based on differences in EGF, IFN-γ, IL-8/CXCL8, RANTES/CCL5, and MMP-9 levels. These groups had different clinical severities based on Schirmer test, conjunctival staining, and corneal staining, thus, these groups were named moderate and severe DED groups. Both groups presented significant different levels of EGF, IFNγ, IL-2, IL-8/CXCL8, IL-10, IL-12, RANTES/CCL5, and MMP-9. IL-1RA, IL-2, and TNFα were differentially affected by time, depending on the treatment. Between V2 V3, there were significant changes in EGF, IL-1RA, IL-2, IL-8/CXCL8, IL-13, IP10/CXCL10, TNF-α, and MMP-9. The strongest biomarker candidates were IFN-γ, IL-2, IL-8, IL-12, RANTES/CCL5, and MMP-9 as DED severity biomarkers; IL-2 as a therapeutic (FML) biomarker; and EGF as a DED activity biomarkers. Finally, 36 DED patients were included in the observational study. Between V1 and V2, DED-related symptoms decreased with all the questionnaires used (OSDI, SANDE II, and ECS-Q). Some (5.5%) patients did not adequately complete SANDE II questionnaire, while all patient correctly interpreted ECS-Q. Patients responding “better” in ECS-Q presented a significantly lower corneal staining that those responding “worse”. Moreover, these patients had a significantly lower SANDE II score than those who responded “equal” or “worse”. Finally, a poor concordance between OSDI and ECS-Q was observed, while between ECS-Q and SANDE II the concordance was moderate. Conclusion: The new two-step clinical trial design proposed is useful to evaluate the efficacy of new DED therapies, both in the traditional way but also in the protective effect after an inflammatory desiccating stress. FML can be used as a positive control in future trials evaluating the effectiveness of new therapies for DED using this new clinical trial design. The following molecules (IFN-γ, IL-2, IL-12, RANTES/CCL5, and MMP-9) allow to classify patients according to their DED severity in an objective way. Thus, these biomarkers also can be useful to better select target patients for clinical trials. lL-2 is proposed as the strongest therapeutic biomarker for fluormetholone therapy that could be used as an objective therapeutic evaluation end-points in future clinical trials with that drug. EGF is proposed as a biomarker that could provide a better definition of DED disease activity. Due to its simplicity, the new questionnaire developed can be a useful tool for the evaluation of the patient´s perception regarding the evolution of their DED symptoms.