The generalized ROC criterion for the evaluation of several tumor markers

The main objective of this paper is to present a Method of evaluating sever al tumor markers by using the generalized ROC criterion. This criterion fin ds the best linear combination of the tumor markers such that the area unde r the ROC curve is maximized Confidence intervals for the generalized ROC c riteria are also presented. This methodology is applied to 51 patients with advanced colorectal cancer for whom the ACE tumor markers were measured be fore and during chemotherapy treatment. Two populations were defined accord ing to clinical response to chemotherapy. Each marker taken separately, whe ther on the raw scale or on the transformed scale, contained 0.5 in the con fidence interval and was thus non significant. This was also true for both markers on the raw, scale. However; the best linens combination on the loga rithms of ACE before and at evaluation gave a significantly better area und er the ROC curve. A weighted change in ACE measurements significantly disti nguishes between responders and non responders in patients with advanced co lorectal cancer. We propose that the methodology presented in this paper be used for the evaluation of several tumor markers.