50 Chapter 3 variable as well as multivariable Cox proportional hazards regression models were used to evaluate the prognostic value of MET immunoreactivity, demographical, clinical, and histopathological patient characteristics. Variables significantly associated with OS and DFS as well as potential confounders were included in backward selection procedures to select the final models. Calculations were done with SPSS Statistics (version 21; IBM) and R version 2.15.3 (version http://www.r-project.org). Unless otherwise mentioned, statistical significance was set at p < 0.05. Results Comparison of commercial antibodies directed against the C-terminus of MET As a guide, the Rimm Lab Algorithm for antibody validation (33) was used to check the specificity and sensitivity of the five purchased C-terminal MET antibodies (i.e., D1C2, CVD13, SP44, C-12 and C-28). In short, the algorithm states that the performance of antibodies should be as expected under all examined – reducing, native and FFPE – conditions in order to be found reliable. To properly asses the validity of the examined antibodies, their specificity and sensitivity was evaluated per examined condition based on the results described below. The details and properties of the used antibodies are described in the Materials and Methods section, paragraph antibodies (Table 1). To establish a baseline for MET expression, MET mRNA expression levels were determined in the MET antibody validation cell line panel (Supplementary table S1; Materials and Methods section, paragraph MET antibody validation cell line panel and culture conditions) by means of qRT-PCR. Although MET mRNA expression levels vary markedly between the cell lines (Figure 1A), ranging from very low (LNCaP) to high (HT-29), none of the cell lines are completely devoid of MET mRNA (i.e., truly negative). It should be mentioned here that we depicted LNCaP as negative for MET mRNA expression in Figure 1A because standardized MET fluorescence levels in this cell line are so low that they cannot be observed in the presented bar chart.
RkJQdWJsaXNoZXIy MTk4NDMw