Over the past decades, immunohistochemistry has gained significance and already taken a crucial position in diagnosis of diseases and prognosis of patients. However, manual interpretation of immunohistochemistry and reproducibility of the scoring systems can be highly subjective. In the article, the immunohistochemical staining of survivin in 98 rectal cancers was analyzed by using Image Pro-Plus (IPP) [3 parameters: density mean, area sum, and integrated optical density (IOD)] and the results were compared with visual assessment (2 parameters: intensity and percentage). The correlations between the 2 methods were examined, significant correlations were observed between density mean and staining intensity (Spearman correlation coefficient, rs=0.806, P<0.001), IOD and staining intensity (rs=0.914, P<0.001), area sum and staining percentage (rs=0.883, P<0.001), IOD and staining percentage (rs=0.884, P<0.001). There was no significant difference between survivin expression and clinicopathologic variables (P>0.05) by visual assessment. However, by IPP analysis, both the density mean and IOD were higher in better-differentiated cancers than in worse differentiated ones (P=0.02 and 0.03). There was a substantial agreement between the 2 methods. Density mean and IOD of IPP were representative parameters to assess the immunostaining quantification, and increased sensitivity in scoring and provided a more reliable and reproducible analysis of protein expression, especially, more information of the protein expression in relation to clinicopathologic variables can be provided by IPP analysis.

Supplemental Digital Content is available in the text