2.7. Statistical Analysis

Statistical analyses were performed using SAS/STAT^® version 9.4 (SAS Institute Inc., Cary, NC, USA). p-values > 0.05 were considered inconclusive.

Survival time was compared between treatments using the log-rank test. An overall test of survivor equality was conducted. Significant p-values were adjusted using the false discovery rate (FDR) method to control the type 1 error rate [22]. Adjusted p-values <0.05 were considered evidence that survival differed by pairwise comparison.

To evaluate tumor growth by treatment, random coefficient models were used, modeling tumor growth as a function of time (in days). A quadratic trajectory for each individual was assumed. A compound symmetric R-side covariance term was added to the tumor growth models to account for the fact that observations for each mouse at different timepoints are correlated. Whenever the models detected the difference between treatment groups, significance regions, i.e., intervals of days for which there are significant treatment differences, were computed.

To analyze the histological samples, the p-values for the number of infiltrated cells in various treatment groups were calculated using the Exact Savage Multisample test, a nonparametric test appropriate to use with >2 groups. Significance was found by adjusted p-value, and the Dwass–Steel–Critchlow–Fligner method was employed to determine which groups were different.