A total of 64 subjects were screened, of which 4 did not fulfil the eligibility criteria and were considered screen failures (Fig. 1). Of the 60 randomized subjects, 30 were allocated to the LCD group and 30 to the placebo group. At the end of the study, 59 subjects completed the study as per protocol requirements: 30 subjects in the LCD group (17 males and 13 females) and 29 in the placebo group (18 males and 11 females). One subject from the placebo group withdrew consent. The demographic characteristics are shown in Table 2. The mean age of subjects in the LCD group was 47.15 ± 12.12 years and in the placebo group was 45.43 ± 12.32 years. The mean body mass index (BMI) was 24.50 ± 2.86 for the LCD group and 24.06 ± 3.40 for the placebo group, and screen exposure time for subjects in the LCD group was 7.75 ± 2.76 h and in the placebo group was 7.86 ± 2.77 h at the screening visit. There was no statistical significance between the two groups for age, sex, BMI and screen exposure time at the baseline.
Table 2 Subject demographics Efficacy
The efficacy analysis was performed for 59 subjects who completed the study. The missing observations were imputed using the LOCF approach (the last observation carried forward).
Primary Efficacy Analyses
Schirmer’s Test In the LCD group, the baseline mean Schirmer’s strip wetness length as a measure of tear volume for the average of both eyes was 6.62 ± 1.73 mm, which increased to 9.90 ± 4.21 mm (↑51.95 ± 56.59%) on day 14, 12.23 ± 3.18 mm (↑92.60 ± 57.44%) on day 28 and 14.32 ± 2.51 mm (↑127.2 ± 59.70%) on day 56 when compared to the baseline value. In the placebo group, the baseline mean Schirmer’s strip wetness length for the average of both eyes was 7.09 ± 2.45 mm, which increased to 8.14 ± 2.55 mm (↑20.49 ± 35.92%) on day 14, 8.69 ± 2.90 mm (↑27.75 ± 35.47%) on day 28 and 7.43 ± 3.29 mm (↑7.70 ± 38.07%) on day 56 compared to the baseline value (Table 3a).
Table 3 Summary of efficacy endpoints between LCD and placebo groups There were no significant differences between the LCD and placebo groups before supplementation, suggesting homogeneity of subjects in the two groups at baseline as evaluated by between-group analysis using ANOVA. However, the LCD group showed a statistically significant increase in the tear volume over placebo on days 28 and 56 (P < 0.0001) while nearing significance on day 14 (P = 0.0580) (Fig. 2a).
OSDI The baseline mean OSDI scores for LCD and placebo groups were 30.10 ± 4.88 and 32.54 ± 5.06, respectively, and were not significantly different from each other. In the LCD group, the mean OSDI score reduced to 25.24 ± 4.74 (↓15.8 ± 9.94%), 19.16 ± 4.40 (↓35.5 ± 13.93%) and 17.09 ± 4.85 (↓42.7 ± 14.42%) on days 14, 28 and 56, respectively. In the placebo group, mean OSDI scores were reduced to 30.68 ± 5.48 (↓4.93 ± 13.69%) on day 14, 30.60 ± 5.91 (↓5.21 ± 14.71%) on day 28 and 31.00 ± 7.18 (↓3.93 ± 19.28%) on day 56 (Table 3b).
The LCD group showed a statistically significant decrease (P = 0.0001) in the mean OSDI score over the placebo group on days 14, 28 and 56 as per between-group analysis using ANOVA test (Fig. 2b).
Secondary Efficacy Analyses
TBUT In the LCD group, the baseline mean TBUT score for the average of both eyes was 8.18 ± 1.18 s, which increased to 9.97 ± 1.80 s (↑22.21 ± 17.33%) on day 14, 11.03 ± 1.32 s (↑36.56 ± 18.93%) on day 28 and 12.13 ± 1.38 s (↑51.10 ± 26.53%) on day 56. In the placebo group, the baseline mean TBUT score for the average of both eyes was 7.86 ± 1.25 s, which increased to 7.97 ± 1.48 s (↑2.60 ± 19.30%) on day 14, 7.93 ± 1.18 s (↑2.55 ± 17.66%) on day 28 and decreased to 6.86 ± 1.51 s (↓12.1 ± 16.53%) on day 56 (Table 3c).
No significant differences were observed between the LCD and placebo groups before supplementation as analyzed by between-group analysis using ANOVA test, suggesting homogeneity of subjects across the groups at baseline. However, the LCD group showed a statistically significant increase in the mean TBUT score over the placebo group on days 14, 28 and 56 (P < 0.0001) (Fig. 2c).
SPEED The baseline mean SPEED scores for the LCD and placebo groups were 11.63 ± 1.25 and 12.21 ± 1.78, respectively, and therefore were not significantly different from each other. The mean SPEED score gradually reduced to 8.97 ± 1.30 (↓22.4 ± 11.36%) on day 14, 7.07 ± 1.36 (↓38.9 ± 11.28%) on day 28 and 6.07 ± 1.86 (↓47.3 ± 16.39%) on day 56 in the LCD group. However, in the placebo group, the mean SPEED scores were 11.03 ± 2.23 (↓8.72 ± 17.13%) on day 14, 10.31 ± 2.16 (↓14.1 ± 19.87%) on day 28 and 12.41 ± 2.90 (↑3.64 ± 28.06%) on day 56 (Table 3d).
The LCD group showed a statistically significant decrease in mean SPEED score over the placebo group on days 14, 28 and 56 (P < 0.0001) (Fig. 2d) as per between-group analysis using ANOVA test.
Corneal Staining Scores There was a decrease in the mean corneal staining score for the average of both eyes from 1.14 ± 0.17 to 0.79 ± 0.25 with 31.1 ± 20.06% reduction for the LCD group whereas it increased from 1.10 ± 0.17 to 1.24 ± 0.20 with 13.98 ± 16.74% increase in the placebo group on day 56 (Table 3e).
In the between-group analysis using ANOVA test, data showed no significant differences between the LCD and placebo groups before supplementation, but the LCD group showed a statistically significant decrease (P < 0.0001) in mean corneal staining score compared to the placebo group on day 56 (Fig. 2e).
Conjunctival Staining Scores There was a decrease in the mean conjunctival staining score for the average of both eyes from 0.88 ± 0.22 at baseline to 0.58 ± 0.32 with 35.5 ± 33.34% reduction on day 56 for the LCD group whereas an increase in the mean conjunctival staining score was observed for the placebo group from 0.86 ± 0.20 at baseline to 1.03 ± 0.27 with 19.53 ± 18.84% increase on day 56 (Table 3f). While between-group analysis using ANOVA test showed no significant differences between the LCD and placebo groups before supplementation at baseline, it showed a statistically significant decrease (P < 0.0001) in the mean conjunctival staining score for the LCD group compared to the placebo group on day 56 (Fig. 2f).
Tear Osmolarity Mean tear osmolarity for the average of both eyes for the LCD group was 327.23 ± 12.42 mOsms/l at baseline, which decreased by 2.07 ± 1.87% to 320.42 ± 12.24 mOsms/l on day 56. However, mean tear osmolarity was increased in the placebo group from 327.34 ± 11.06 mOsms/l at baseline to 331.72 ± 11.20 mOsms/l at day 56 with a slight increase of 1.35 ± 1.71% (Table 3g).
The between-group analysis using ANOVA test showed no significant differences between LCD and placebo groups at baseline but the LCD group showed a statistically significant decrease (P < 0.0005) in tear osmolarity compared to the placebo group on day 56 (Fig. 2g).
MMP-9 Biomarker MMP-9 levels were measured at baseline and end of the study in the tears of the subjects using the InflammaDry® test. Sixteen subjects (53.33%) out of 30 from the LCD group and 15 subjects (51.72%) out of 29 from the placebo group showed positive MMP-9 results in the right eye at the baseline visit. Sixteen subjects (53.33%) out of 30 from the LCD group and 14 (48.28%) out of 29 subjects from the placebo group showed positive MMP-9 results in the left eye at the baseline visit (Table 4a, b; Fig. 3a, b). However, on day 56, only two subjects from the LCD group showed positive MMP-9 results for both eyes, which was a reduction of 87.50% from baseline. In the placebo group, 12 subjects showed positive MMP-9 results, which was a reduction of only 20% for the right eye and 14.29% for the left eye from the baseline.
Table 4 Number of subjects positive for MMP-9 biomarker between LCD and placebo groups Artificial Tear Use and Frequency Under ethical guidelines, the subjects were allowed to use artificial tears during the study period as a rescue medication. Fifteen (50%) subjects in the LCD group and 20 (68.97%) subjects in the placebo group used artificial tears between day 0 and 14. However, only 7 (23.33%) subjects in the LCD group and 20 (68.97%) subjects in the placebo group continued using artificial tears from day 15 to 56. There was a 53.33% reduction in the use of artificial tears from baseline to day 56 for the LCD group, and no change was observed for the placebo group (Table 5; Fig. 4a).
Table 5 Summary of the number of subjects who used artificial tears between LCD and placebo groups The between-group analysis using Z-test indicated a significant reduction in the number of subjects using artificial tears from day 15 to 28 (P = 0.0004) and day 29 to 56 (P = 0.0004) for the LCD group compared to the placebo group (Fig. 4a).
Furthermore, we observed that artificial tears were used 2.47 times/day in the LCD group and 2.65 times/day in the placebo group between day 0 to 14, which reduced to 1.07 times/day, a 64.40% reduction from baseline between day 15 to 28 and 0.73 times/day, and a reduction of 75.60% from baseline between day 29 to 56 in the LCD group. However, in the placebo group the use of artificial tears remained largely unchanged between baseline to day 56 with 2.45 times/day and 2.60 times/day, respectively, from day 15 to 28 and 29 to 56 (Table 6). The between-group analysis using the Kruskal-Wallis test showed a significant reduction in the frequency of artificial tears use from day 15 to 28 (P = 0.0009) and day 29 to 56 (P < 0.0001) in the LCD group compared to the placebo group (Fig. 4b).
Table 6 Summary of the mean frequency of artificial tears use between LCD and placebo groups Safety
The safety analysis was conducted for all the subjects who were randomized for the trial and consumed at least one dose of the study product. We found that the study product was safe and well tolerated with no incidence of supplementation-related adverse events. Though a total of four subjects reported adverse events with one subject from the LCD group reporting fever and three subjects from the placebo group reporting events such as body pain, headache and paresthesia, none of these adverse events were found to be related to the study supplements. No adverse events were severe; they were of mild intensity. There were no notable changes in urine analysis parameters, vital signs and physical examination. Hematology and biochemical laboratory results remained relatively unchanged.