Selecting immunohistochemical cut-off scores for novel biomarkers of progression and survival in colorectal cancer

Abstract

Background: Cut-off scores for determining positivity of biomarkers detected by immunohistochemistry are often set arbitrarily and vary between reports. Aims: To evaluate the performance of receiver operating characteristic (ROC) curve analysis in determining clinically important cut-off scores for a novel tumour marker, the receptor for hyaluronic acid mediated motility (RHAMM), and show the reproducibility of the selected cut-off scores in 1197 mismatch-repair (MMR) proficient colorectal cancers (CRC). Methods: Immunohistochemistry for RHAMM was performed using a tissue microarray of 1197 MMRproficient CRC. Immunoreactivity was scored using a semi-quantitative scoring method by evaluating the percentage of positive tumour cells. ROC curve analysis was performed for T stage, N stage, tumour grade, vascular invasion and survival. The score with the shortest distance from the curve to the point with both maximum sensitivity and specificity, i.e. the point (0.0, 1.0), was selected as the cut-off score leading to the greatest number of tumours correctly classified as having or not having the clinical outcome. In order to determine the reliability of the selected cut-off scores, 100 bootstrapped replications were performed to resample the data. Results: The cut-off score for T stage, N stage, tumour grade and vascular invasion was 100% and that for survival 90%. The most frequently selected cut-off score from the 100 resamples was also 100% for T stage, N stage, tumour grade, and vascular invasion and 90% for survival. Conclusions: ROC curve analysis can be used as an alternative method in the selection and validation of cutoff scores for determining the clinically relevant threshold for immunohistochemical tumour positivity.

Differential diagnostic and functional role of the multi-marker phenotype CDX2/CK20/CK7 in colorectal cancer stratified by mismatch repair status

“…To determine the association of positive protein expression with clinico-pathological features, cutoff scores for CK20, CK7, and CDX2 expression were determined by ROC curve analysis. 37 The ROC curve is a plot of the sensitivity and (1-specificity) for an outcome at each value of the protein expression score. A protein score can therefore be selected from the curve such that a cutoff at this value leads to the greatest number of patients correctly classified as with (maximizing sensitivity) and without (maximizing specificity) the clinical end point.…”

Combined analysis of VEGF and EGFR predicts complete tumour response in rectal cancer treated with preoperative radiotherapy

“…Selection of cut-off scores for protein positivity Relevant cut-off scores for tumour positivity for each protein marker were obtained by performing receiver-operating characteristic (ROC) curve analysis (Zlobec et al, 2006b). Briefly, plots of sensitivity and (1-specificity) for complete pathologic tumour response were obtained for each marker and the (0,1)-criterion was used to select the threshold value, or protein expression score, above which expression was to be considered 'positive' (Bewick et al, 2004).…”

“…The interpretation of immunoreactivity is underlined as a major source of contradictory findings (Goldstein and Armin, 2001;Resnick et al, 2004;Spano et al, 2005;Kim et al, 2006;. In order to avoid the use of predetermined and often arbitrarily set cut-off values, we have previously shown how ROC curve analysis in conjunction with a resampling procedure can be systematically used to evaluate the protein expression of immunohistochemical tumour markers (Zlobec et al, 2007a). Along with a reproducible semi-quantitative scoring system, ROC curve analysis is a powerful method for selecting cutoff scores to describe tumour marker positivity for a specific clinical endpoint, such as tumour response.…”

Multimarker phenotype predicts adverse survival in patients with lymph node‐negative colorectal cancer

“…24 This method identifies the point on the ROC curve that corresponds to the protein expression score above which sensitivity for the outcome is maximized while loss of specificity is minimized. 24 Tumors designated as ''negative'' for the protein were those with scores below the threshold value, whereas positive tumors were considered those with scores above the threshold value.…”