The first ten cases will be reviewed to assess the spectrum of OX-40 expression and to devise a semiquantitative scoring system. Each slide will be scanned in its entirety and then scored according to representative views with the greatest expression of immunostaining. Two independent viewers, who will be blinded to clinical data and outcome at the time of scoring, will review all slides. Scores from these independent readings will be compared, and differences were resolved upon mutual review of given cases. We will estimate the distribution of OX-40 IHC score by tumor type and tumor stage. OX-40 IHC score will be correlated with the IHC scores of other markers using Spearman's correlation coefficient. Kruskal-Wallis test will be used to test whether OX-40 IHC score is comparable between tumor type and tumor stage

5-micron sections will be cut, mounted on capillary gap slides and dried in a 60°C oven. The slides will then be then deparaffinized in xylene baths and gradually hydrated in graded alcohol. For epitope retrieval, the slides are placed in baths of 0.5 M Tris buffer at pH 10 and subjected to microwave irradiation for at least 19 minutes. The slides will be cooled and washed with deionized water and 0.05% TWEEN. They are then placed in a blocking solution of 0.3% bovine serum albumin, drained, and placed in wells of their primary antibody in a humidity chamber at room temperature overnight. Primary mouse antihuman antibodies will include anti-OX-40, anti-CD4, and anti-HLA class II. Anti-rat CD45 will be used as a negative control for each section. The slides will then be then washed and placed in their appropriate biotinylated anti-mouse secondary antibody for 24 minutes. Endogenous peroxidase in the samples is blocked using a solution of 3% hydrogen peroxide.

The first ten cases will be reviewed to assess the spectrum of OX-40 expression and to devise a semiquantitative scoring system. Each slide will be scanned in its entirety and then scored according to representative views with the greatest expression of immunostaining

Kaplan-Maier method will be used to estimate the disease-free survival, and a log-rank test will be used to compare the disease-free survival between high vs. low OX-40 groups