Role of Caspase-1 as a Biomarker of Ocular Surface Damage

Abstract

PURPOSE:

To examine the potential of caspase-1 as a biomarker for ocular surface damage.

DESIGN:

Cross-sectional study.

METHODS:

A total of 113 tear samples (64 subjects) were analyzed. Sixty-one samples were from individuals with dry eye disease (DED), defined as Ocular Surface Disease Index (OSDI) ≥13 and/or corneal staining (CS) ≥3; 32 were from individuals who used glaucoma medication, irrespective of DED metrics; and 20 were from controls (CS <3 and OSDI <13). All individuals completed a medical history form and underwent an ocular surface assessment. Protein levels of caspase-1 were determined by enzyme-linked immunosorbent assay off Schirmer’s strips. The primary analysis compared caspase-1 levels in individuals with signs of ocular surface damage (CS ≥3) in both case groups and controls. Secondary correlational analyses were conducted to examine relationships between caspase-1 levels and ocular signs and symptoms. Finally, area under the curve (AUC) analyses were performed to examine relationships between inflammatory markers and CS.

RESULTS:

The mean age of the population was 58±18 years; 70% were female. Tear samples from individuals with ocular surface damage presented higher caspase-1 levels than the control group. Caspase-1 levels showed a moderate positive correlation with CS (Spearman r = 0.31; P = .001) and eye redness (Spearman r = 0.39; P = .004), and a negative correlation with Schirmer’s (Spearman r = −0.46; P < .001) and tear break-up time (Spearman r = −0.33; P = .0006). Caspase-1 showed higher sensitivity and AUC for detecting ocular surface damage than InflammaDry, and its expression was not affected by anti-inflammatory agents.

CONCLUSION:

Caspase-1 levels were higher in the tears of individuals with ocular surface damage, suggesting its potential to be used as a biomarker and/or therapeutic target.

INTRODUCTION

Ocular surface damage, which manifests as punctate epithelial erosions in the conjunctiva and/or cornea, can be the end result of many disease processes, including dry eye disease (DED), medication use, and environmental exposure.^1,2 Dry eye disease itself is a multifactorial disorder characterized by a loss of homeostasis of the tear film, and accompanied by ocular symptoms, in which tear film instability and hyperosmolarity, ocular surface inflammation and damage, and neurosensory abnormalities play etiological roles.³ Due to this complexity, ocular surface damage in DED may present with or without symptoms^4,5 and can occur concomitantly with a number of comorbidities, including: autoimmune diseases,^6,7 low androgen levels,^8,9 anatomical abnormalities of the eyelids and conjunctivae,^10,11 history of refractive surgery,¹² and Meibomian gland dysfunction.¹³

Ocular surface damage can also be caused by a number of systemic¹⁴ and topical medications.^15,16 In particular, topical ocular hypotensive medications have been linked to ocular surface damage, both due to their preservatives and active agents.^17,18 In one study,¹⁹ investigators found significant correlations between corneal staining (CS) (graded via the Oxford scale) and the number of drops instilled per day. Specifically, the mean number of drops used by those with no CS was 2.76, which increased to 4 in those with the most severe staining (P = .01).¹⁹ Another study reported that individuals using 3 drops per day more frequently had features of DED, namely CS and decreased tear break-up time (TBUT), as compared with those using 1 drop per day (40% vs 5%; P = .01).²⁰ In addition, prolonged exposure to topical hypotensive medications can reduce corneal sensitivity, which is another cause of CS.^21,22

Inflammation is a key component of ocular surface damage both in DED and with the use of topical hypotensive medications.^16,23,24 Much of the focus in these diseases has been on the role of T cells in propagating inflammation, and thus most anti-inflammatories have focused on therapeutics that block T cells.^25,26 Pyroptosis, or inflammation-induced programmed cell death, is another inflammatory pathway that has found relevance in ocular surface diseases.²⁷ It is mediated by inflammasome formation, which is a multiprotein complex that activates caspase-1 and induces activation of interleukin(IL)-18 and IL-1β. Caspase-1 also cleaves gasdermin D, which oligomerizes and creates membrane pores, resulting in water influx, cell rupture, and the release of these two pro-inflammatory cytokines (IL-18 and IL-1β).^28,29

There are several sub-types of inflammasomes that have a role in the innate immune response present in a variety of diseases.^30,31 Relevant to ocular surface diseases, the nod-like receptor protein-3 (NLRP3) inflammasome has been implicated in DED.^27,32,33 In humans, individuals with DED (broadly defined as having DED-related symptoms, or a positive CS, or Schirmer < 5 mm, or TBUT < 5 seconds) showed elevated tear concentrations of several inflammasome pathway molecules, including caspase-1, a downstream mediator of inflammasome activation.²⁷ In rodents, desiccating stress triggered the expression of caspase-1, IL-1β, and IL-18 in the corneal epithelium, suggesting that this signaling pathway plays an important role in environment-induced DED.^32,33 Furthermore, other inflammasome sub-types, including NLRP12 and NLRC4, have been shown to contribute to the inflammatory response in environment-induced murine DED.³⁴

Inflammasomes have also been implicated in the pathophysiology of glaucoma. A study that evaluated inflammatory pathway components linked to glaucoma found that some inflammasome components, including caspase-1, were upregulated in human retinal protein samples in this population.³⁵ In a murine model of acute IOP-induced glaucoma, retinal ganglion cell death was noted to occur via a caspase-1-dependent pathway involving NLRP1 and NLRP3 inflammasomes.³⁶ However, the relationship between topical hypotensive medication and inflammasome activity on the ocular surface is unclear.

Given the paucity of data on relationships between inflammasome pathway activation and ocular surface diseases, the objective of this study was to evaluate relationships between caspase-1 and ocular surface metrics. Based on prior data in humans and animal models, it was hypothesized that caspase-1 levels would closely relate to ocular surface damage, irrespective of etiology. Furthermore, the study aimed to evaluate the potential of tear capsase-1 levels as a biomarker for ocular surface damage, as such a biomarker could be beneficial in settings where a slit lamp is unavailable (eg, primary health offices). Given the availability of InflammaDry, a point-of-care test that qualitatively assesses for ocular surface levels of matrix metalloproteinase-9 (MMP-9), its potential as a biomarker of ocular surface damage was also examined and the results were compared with tear capsase-1. Finally, given that chronic anti-inflammatory therapies are a common treatment strategy for ocular surface damage, and have been shown to decrease MMP-9 levels,^37,38 whether the use of anti-inflammatory therapies impacted tear caspase-1 levels was also examined.

MATERIALS AND METHODS

STUDY POPULATION:

This study was approved by the Institutional Review Board of the University of Miami Miller School of Medicine. The protocol (#20190334) conformed to the requirements of the United States Health Insurance Portability and Accountability Act and the tenets of the Declaration of Helsinki. All study patients were evaluated at the Bascom Palmer Eye Institute and were either recruited from the cornea or glaucoma divisions (cases) or optometry clinics (controls). With respect to cases, individuals with DED were recruited, which was defined as an Ocular Surface Disease Index (OSDI) ≥13³⁹ and/or CS ≥3.^40–42 These broad inclusion criteria were intentional because the study aimed to include a wide range of DED signs and symptoms. The second case group included any individuals who used topical hypotensive medication, irrespective of ocular surface signs and symptoms. Asymptomatic subjects (OSDI < 13 points) with no signs of ocular surface damage (CS <3 points on NEI scale) were enrolled into the control group. Exclusion criteria for all groups included pregnancy and ages <21 or >90 years. Tears from both eyes were sampled when possible, but eyes from the case groups with a history of penetrating keratoplasty (n = 3 eyes) were excluded. Eyes from the control group were excluded if CS ≥3 (despite the individual being asymptomatic). With these inclusion and exclusion criteria in mind, a total of 64 patients (113 eyes) were recruited for the study, including 33 individuals (62 eyes) in the DED group, 20 individuals (32 eyes) in the glaucoma groups, and 11 individuals (20 eyes) in the control group.

DATA AND TEAR SAMPLE COLLECTION:

After signing informed consent, all individuals first filled out the OSDI and then underwent a clinical examination with the following assessments, in the order performed: eye bulbar redness using the Effron bulbar redness scale (0 = no hyperemia to 4 = severe hyperemia);⁴³ MMP-9 (InflammaDry, graded as presence or absence of a pink strip);⁴⁴ and Schirmer’s test with anesthesia. After 5 minutes, the Schirmer strips were removed with sterile gloves and placed in sterile 1.5 mL Eppendorf containers; TBUT was performed using fluorescein strips; CS with fluorescein (NEI scale, range 0-15 points);⁴² and conjunctival staining with lissamine green (NEI score, range 0-18 points).⁴⁵

SAMPLE PROCESSING:

Samples were processed as described by Dermer and associates,⁴⁶ with a modified technique: briefly, after collecting the samples, the strips were immediately transferred from the 1.5 mL tubes to 0.5 mL sterile Eppendorf tubes, in which a small hole was previously cut towards the bottom; 60 μL of balanced saline solution were added to the 0.5 mL containers, which were placed inside 1.5 mL tubes; these were placed on a shaker at 400 rpm at room temperature (23 °C) for 30 minutes; then, another 60 μL of balanced saline solution were added and centrifuged at 10 000 rpm for 3 minutes; the supernatant was then collected from the 1.5 mL Eppendorf tube and kept at −80 °C until further processing with enzyme-linked immunosorbent assay (ELISA). As the tear volume collected through Schirmer’s strips differed among patients, the difference in volume was adjusted for by quantifying protein in each sample before running the ELISAs.

CASPASE-1 MEASUREMENT:

The concentration of caspase-1 in tears was determined using a commercially available ELISA kit (Abcam), according to manufacturer’s instructions. Briefly, tear supernatants and protein standards were loaded into different wells of a 96-well plate and mixed with an antibody cocktail, followed by incubation for 1 hour at room temperature on a plate shaker set to 400 rpm; the plate was then washed with wash buffer solution; TMB substrate was added to each well and incubated in the dark on the plate shaker for 10 minutes; finally, the stop solution was added to each well and the absorbance was measured at 450 nm in a SpectraMax M5 (Molecular Devices) spectrophotometer.

STATISTICAL ANALYSES:

R 3.6.1 (R Core Team) was used for statistical analyses. Analyses were completed using linear mixed models with random effects placed at the patient level to account for potential inter-eye correlations among those subjects with both eyes enrolled in the study. Data were adjusted for age and gender. Data were presented as mean±standard deviation (SD). P-values < .05 were considered significant. GraphPad Prism 9.1.0 (216) (GraphPad Software Inc.) was used to generate figures. Analysis of variance (ANOVA) and χ² were performed for comparison of demographic characteristics. Spearman correlations were used to examine correlations between caspase-1 levels and ocular surface signs, including CS. The area under the receiver operating characteristics curves (AUC) of caspase-1 and InflammaDry were then calculated to compare the biomarker potential of these two analytes in the study population. A linear mixed model was used to determine whether using anti-inflammatory medication (cyclosporine, lifitegrast, or steroids) had an effect on tear caspase-1 levels.

RESULTS

DEMOGRAPHIC CHARACTERISTICS AND CLINICAL PARAMETERS:

This study included a total of 64 patients (113 eyes): 33 patients (62 eyes) belonged to the DED group, 20 patients (32 eyes) to the glaucoma group, and 11 patients (20 eyes) to the control group. Demographic data are provided in Table 1. There were statistically significant differences in age and sex distributions between the three groups. Data on topical medication use (per individual eye) are listed in Table 2. Mean values and SD of all assessed parameters are provided in Supplementary Table 1. Overall, eyes receiving topical hypotensive medication had the most severe signs of ocular surface damage, with 21 (66%) eyes having CS ≥3 (Figure 1).

TABLE 1.

Demographic characteristics of the population and their comorbidities

	Controls n = 11	DED n = 33	Glaucoma n = 20	P Value
Eyes	20	61	32
Age (Mean±SD)	41±13	57±17	70±10	< .0001
Sex (female), n (%)	7 (64%)	28 (85%)	9 (45%)	.009
Race, n (%)				.65
White	9 (82%)	30 (91%)	18 (90%)
Black	2 (18%)	2 (6%)	1 (5%)
Asian	0 (0%)	1 (3%)	1 (5%)
Ethnicity, n (%)				.35
Non-Hispanic or Latino	1 (9%)	6 (18%)	6 (30%)
Hispanic or Latino	10 (91%)	27 (82%)	14 (70%)
Immunologic comorbidities, n (%)
Sjögren’s syndrome	0 (0%)	6 (18%)	0 (0%)
Rosacea	0 (0%)	4 (12%)	0 (0%)
Rheumatoid arthritis	0 (0%)	1 (13%)	1 (5%)
GVHD	0 (0%)	1 (3%)	0 (0%)
Bronchiectasis	0 (0%)	2 (6%)	0 (0%)
Fibromyalgia	0 (0%)	1 (3%)	0 (0%)

Abbreviations: DED = dry eye disease; GVHD graft versus host disease; SD = standard deviation.

TABLE 2.

Ocular topical medications (per individual eye)

Medication, n (%)	Controls	DED	Glaucoma
Cyclosporine	0 (0%)	11 (18%)	1 (3%)
Prednisolone	0 (0%)	2 (3%)	1 (3%)
Lifitegrast	0 (0%)	4 (7%)	0 (0%)
Number of hypotensive medications, n (%)
1	0 (0%)	0 (0%)	11 (34%)
2	0 (0%)	0 (0%)	5 (16%)
3	0 (0%)	0 (0%)	9 (28%)
4	0 (0%)	0 (0%)	7 (22%)

Abbreviation: DED, dry eye disease.

FIGURE 1.

Clinical presentation of the ocular surface in patients with dry eye disease (DED), glaucoma (GLC), and healthy controls (C). A) Patients using topical hypotensive medication presented with the highest corneal staining scores (NEI scale), followed by the DED group, which were statistically significant when compared with those of the controls. N = C: 19, DED: 61, GLC: 32. B) The conjunctival staining scores (NEI scale) were significantly higher in patients using topical hypotensive medication when compared with the controls. N = C: 19, DED: 49, GLC: 32. C) Patients using topical hypotensive medication showed the highest redness scores (Effron scale) when compared with the other two groups. N = C: 5, DED: 16, GLC: 32. D) Controls had the highest Schirmer’s test results (mm) when compared with the other two groups. N = C: 20, DED: 61, GLC: 32. E) Tear break-up time (TBUT) (sec) were longer in controls when compared with the group using topical hypotensive medication. N = C: 17, DED: 59, GLC: 32. F) The OSDI questionnaire scores were significantly higher in the glaucoma and DED group compared with the controls. N = C: 20, DED, 61, GLC: 32. Statistical significance was derived from linear mixed models. Data presented as boxplots with the 5th and 95th percentiles.

COMPARISON OF TEAR CASPASE-1 LEVELS BETWEEN CONTROLS, DED, AND GLAUCOMA GROUPS:

Caspase-1 was significantly elevated in eyes receiving topical hypotensive medication, followed by the DED and control eyes (109.20±42.59 pg/mL, 91.62±43.86 pg/mL and 54.88 ±23.04 pg/mL, respectively) with significant differences observed between the glaucoma and DED eyes, as compared with controls (P = .001 and P = .003, respectively Figure 2;). Given the between-group differences in age and sex, univariable and multivariable linear mixed models were conducted to examine caspase-1 levels among the study groups, while controlling for age, sex, and inter-eye correlations. Age and sex were not significantly associated with caspase-1 level. Even when adjusting for these potential confounders, the caspase-1 level remained significantly associated with the study group (Table 3). This study also evaluated whether caspase-1 level had a relationship with the number of topical medications used in the glaucoma group; while there was a positive trend, the association was not statistically significant (P = .25; data not shown).

FIGURE 2.

Caspase-1 in the tears of patients with dry eye disease (DED), glaucoma (GLC), and healthy controls (C). The population using topical hypotensive medication presented the highest caspase-1 concentration levels in their tears. Statistical significance was derived from linear mixed models. N = 113; C: 20, DED: 61, GLC: 32. Data presented as boxplots with the 5th and 95th percentiles.

TABLE 3.

Coefficients from univariable and multivariable linear mixed models predicting caspase-1 levels

	Univariable		Multivariable
	Coefficient	P Value	Coefficient	P Value
Age (y)	0.38	.21	−0.27	.43
Sex (male)	9.85	.39	8.32	.46
Study group
Control	55.89	.001	–	–
Dry eye disease	94.35	.005	108.23	.004
Glaucoma	109.28	.0005	123.51	.001

CASPASE-1 LEVELS ARE MODERATELY CORRELATED WITH LACK OF TEAR PRODUCTION, TEAR INSTABILITY, AND OCULAR SURFACE DAMAGE SIGNS BUT NOT WITH SYMPTOMS:

This study identified statistically significant correlations between caspase-1 level and the following variables: CS (Spearman r = 0.31; P = .001), TBUT (Spearman r = −0.33; P = .0006), Schirmer’s test (Spearman r = −0.46; P < .001), and bulbar redness (Spearman r = 0.39; P = .004). Neither conjunctival staining (Spearman r = 0.19; P = .06) nor symptoms assessed through OSDI (Spearman r = 0.08; P = .38) demonstrated a correlation with caspase-1 levels (Table 4).

TABLE 4.

Correlations between ocular surface damage signs and symptoms and caspase-1 levels in tears

Variable	Spearman r	P Value
OSDI	+0.08	.38
Bulbar redness	+0.39	.004
TBUT	−0.33	.0006
Corneal staining	+0.31	.001
Conjunctival staining	+0.19	.06
Schirmer’s test	−0.46	< .0001

Abbreviations: TBUT = tear break-up time; OSDI = Ocular Surface Disease Index.

CASPASE-1 AS A BIOMARKER OF OCULAR SURFACE DAMAGE:

The sensitivity and specificity of tear caspase-1 were then assessed to detect ocular surface damage (defined as CS ≥3) (Figure 3). An AUC of 0.7 was found (P = .0002, confidence interval: 0.62-0.81, SEM: 0.05). A cutoff point of 82.85 pg/mL maximized the sensitivity (73%) and specificity (64%). A similar analysis was performed to examine InflammaDry positivity compared with CS ≥3. Here, the AUC was 0.6 (P = .04), with a positive InflammaDry providing a sensitivity and specificity of 52% and 66%, respectively.

FIGURE 3.

Receiver operating characteristics (ROC) of caspase-1 for the diagnosis of ocular surface damage. AUC of caspase-1 in patients diagnosed with ocular surface damage as determined by whether patients presented corneal staining (>3 points on NEI scale) or not (<3 points on NEI scale). Using 82.85 pg/mL as the cut-off point, the AUC for caspase-1 was 0.7, with a sensitivity of of 73% and specificity of 64%.

RELATIONSHIPS BETWEEN CASPASE-1 IN TEARS AND TOPICAL ANTI-INFLAMMATORY MEDICATIONS:

Finally, relationships between the use of topical anti-inflammatory medications and tear caspase-1 levels were examined. After conducting a multivariable linear mixed model to adjust for age, sex, and potential inter-eye correlations, tear caspase-1 levels in eyes receiving anti-inflammatory eyedrops (corticosteroids, cyclosporine, or lifitegrast, n=18) were not statistically different from individuals not on topical anti-inflammatory agents (P = .13).

DISCUSSION

This study found that caspase-1, a molecule involved in the inflammasome cascade, was elevated in individuals using topical hypotensive medication and in those with a variety of ocular surface abnormalities, including CS, low tear production, and tear instability signs, as compared with controls. Correlations between caspase-1 and various clinical signs of ocular surface damage were significant, although moderate in strength. It also evaluated caspase-1 as a potential biomarker for CS, finding that tear caspase-1 levels ≥ 82.85 pg/mL (detected using an enzyme-linked immunoassay) had a 73% sensitivity of identifying individuals with clinically significant CS. In contrast, the sensitivity of InflammaDry (which detects MMP-9 levels ≥ 40 ng/mL using a rapid immunoassay test) was lower at 52%. These findings are consistent with prior studies that reported weak relationships between InflammaDry and other signs and symptoms of DED.^47,48 Overall, caspase-1 could potentially be a more reliable biomarker of ocular surface damage compared with InflammaDry.

These data introduce novel concepts regarding the pathophysiology, diagnosis, and treatment of ocular surface diseases (OSD). First, the data highlight that inflammasome pathway molecules may be involved in the pathophysiology of OSD, including glaucoma medication-associated disease and DED. This idea is supported by prior studies, which found elevated levels of caspase-1 in corneal epithelial cells of animals after desiccating stress³² and in the tears of individuals with DED.²⁷ Second, the involvement of the inflammasome pathway in OSD may have diagnostic implications. Inflammasome proteins have previously been studied as biomarkers in a variety of conditions, including: age-related macular degeneration,⁴⁹ psoriasis,⁵⁰ non-alcoholic steatohepatitis,⁵¹ Alzheimer’s disease,⁵² stroke,⁵³ multiple sclerosis,⁵⁴ traumatic brain injury,^55,56 and major depressive disorder.⁵⁷ For instance, high protein levels of caspase-1 in the cerebrospinal fluid of patients with traumatic brain injury predicted increased intracranial pressure and poor outcomes.⁵⁸ The current data suggest that caspase-1 has potential as a screening test for ocular surface damage in clinics that do not have access to a slit lamp, such as primary care offices, perhaps leading to earlier treatment intervention.

Third, the involvement of the inflammasome pathway in OSD may have therapeutic implications. Specifically, molecules that block the inflammasome pathway can be studied as novel therapies in DED. For example, NLRP3 inhibitors have been examined in in-vitro models and been shown to reduce hyperosmolar stress-induced inflammation, as well as the production of reactive oxygen species.⁵⁹ Caspase-1 inhibitors have been used in a mouse model of keratitis induced by Pseudomonas aeruginosa, showing a reduction in the clinical score of the keratitis after 3, 5, and 7 days of use.⁶⁰ In a similar manner, inhibition of caspase-1 or other inflammasome pathway molecules may be beneficial in a variety of OSD. This is especially needed, as the current study found that anti-inflammatory agents approved for DED, which mostly block T cell-mediated pathways, did not impact caspase-1 levels. In fact, it is known that cyclosporine and lifitegrast do not consistently stop disease progression or significantly improve signs and symptoms of ocular surface damage in all individuals.^61,62 This suggests that multiple inflammatory pathways, including inflammasome activation, may contribute to ocular surface damage and that blocking multiple pathways may be needed to minimize damage.

There is biologic plausibility that the inflammasome pathway is activated in OSD. Inflammasome, a key component of the innate immune response, is an intracellular multiprotein complex involved in the activation of the inflammatory molecule cysteine aspartase caspase-1,⁶³ processing of the IL-1 cytokines IL-1β and IL-18, and programmed death of a cell (ie, pyroptosis).⁶⁴ Its activation perpetuates a cycle of inflammation, where cell lysis and the release of cytosolic pro-inflammatory molecules to the extracellular compartment induce leukocyte chemotaxis that further amplifies the inflammatory cascade;^65,66 the main trigger of this cascade is infection.^67,68 However, inflammasome activation has also been shown to occur in various diseases in the nervous system,⁶⁹ heart,^70,71 liver,⁷² kidneys,⁷³ intestines,⁷⁴ and the eyes,⁷⁵ in the absence of infection. In the eye, several triggers may prompt inflammasome pathway activation, such as increased friction between the eyeball and eyelid, which occurs when the tear film is poor either in quantity or quality⁷⁶ and the use of topical medication, especially products that contain benzalkonium chloride.^15,16,77

As in all studies, these findings must be considered with study limitations in mind, including a specific population defined by the inclusion/exclusion criteria. First, a wide variety of DED subtypes were included and individuals were using a number of medications, both with and without preservatives. Given this large diversity, the study was not powered to examine caspase-1 levels by DED subtypes, by the use of specific topical hypotensive medications, or by the use of preserved vs non-preserved medication. Future studies with larger sample sizes will be needed to examine these questions. Second, the groups were not identical regarding age. Age-matching controls with cases was challenging, as the majority of older individuals had some degree of DED signs or symptoms. As an alternative, this study adjusted for age in the multivariable model and, interestingly, did not find that demographics impacted caspase-1 levels. Third, it did not examine caspase-1 levels in individuals with glaucoma who were not using topical hypotensives (ie, post-surgical patients). As such, whether elevated levels were due to the inherent disease process or due to iatrogenic damage from topical medications cannot be commented on. Fourth, it is acknowledged that the correlation coefficients between caspase-1 levels and DED signs were of weak-to-moderate magnitude, a finding frequently seen when examining relationships between various aspects of DED, most notably signs vs symptoms. It is believed that the multifactorial nature of DED, with multiple phenotypes, and the fact that different phenotypes are driven by a variety of factors contribute to the weak-to-moderate rho values.

Despite these limitations, this study demonstrated that the inflammasome pathway is active in individuals with signs of ocular surface damage, particularly those on topical IOP-lowering medications. A caspase-1 level of 82.85 pg/mL separated individuals with ocular surface damage (≥3 points in corneal staining) from those without, pointing to its potential as a diagnostic biomarker and therapeutic target. This test could also be combined with InflammaDry to examine multiple pathways of inflammation that may underlie ocular surface damage, and with the expanding selection of medication, help individualize treatment algorithms.

Supplementary Material

Supplemental Material available at AJO.com.