Prism v7.0 for windows

Horses were grouped by laryngeal grade for each of the grading systems. The Lane grade 5 and Havemeyer grade IV groups were excluded from statistical analysis due to extremely small population size (n = 1). Normality was assessed by a Shapiro–Wilk test. Data following a normal distribution was reported as mean ± SD, whereas data not following a normal distribution was reported as median ± range. Categorical data were reported as counts of horses and compared using either a chi‐square test or a Fisher's exact test depending on expected counts. Statistically significant outliers detected by a robust regression and outlier removal (ROUT) method with Q = 1% were excluded from all analyses. Performance parameters were compared between groups using analysis of variance (ANOVA) with Dunn's multiple comparison post hoc test for normally distributed data and a Kruskal–Wallis test for nonnormally distributed data. Statistical analysis was performed using commercially available statistical software (GraphPad Prism v7.0 for Windows, GraphPad Software, La Jolla, CA 92108, USA) and statistical significance was set at P < 0.05.

For the statistical analysis, the median marker scores obtained from all available tumor measurements of each case were used. QIF signals between compartments were analyzed using linear regression and correlation functions and expressed as correlation coefficients. The marker levels in all compartments of interest were examined for differences in their associations with clinicopathologic features and compared across cohorts for validation. Differences between marker scores and patient characteristics were compared using the Student’s t-test/one-way analysis of variance for continuous variables and χ² test for categorical variables. Overall survival functions were compared using Kaplan-Meier estimates, and statistical significance was determined using the log-rank test. The optimal cutpoints for survival analysis of each target in each compartment of interest were obtained using the X-tile software previously described20 (link) in Cohort 1 and tested for validation in Cohort 2. All p values were based on two-sided tests, and all values under 0.05 were considered statistically significant. Statistical analyses were performed using JMP Pro software (V.13.0.0, 2016, SAS Institute, Cary, North Carolina, USA) and GraphPad Prism V.7.0 for Windows (GraphPad Software, San Diego, California, USA).