Tumor markers (TM) are only tumor-associated probes reflecting tumor development. Being of protein, lipid and/or carbohydrate nature, they are expressed in or on tumor cells or induced in their environment. Most of them are circulating antigens appearing in serum or other body fluids, whose concentration results as sum of tumor marker expression, synthesis, release, catabolism, excretion as well as of tumor blood supply. TM concentration correlates in a high manner with the development, extention and therapy-induced tumor reduction. TM determination is performed by highly sensitive radio-or enzyme immunoassays, fluorescence and luminescence assays by means of poly and/or monoclonal antibodies. An appropriate interpretation of individual serial TM follow-up curves demands knowledge about intra-and inter-assay variation of TM test, the catabolism and excretion mode of TM and about pitfalls of transitory levels changes by acute treatment and unspecific influence of interfering human antibodies; in addition statistical parameters of the target TM (sensitivity, specificity, prevalence, positive/negative predictive value) should be known. According to TM indication, circulating TM are useless for screening and tumor localization, conditionally useful for surveillance of risk groups, staging and prognosis, but essentially useful as adjuncts in the follow-up care of tumor patients after operation or during radio-, chemo-or hormone therapy, where their level fluctation may anticipate the clinically detectable course of disease by a lead-time from 1 to 6 months. The indication for TM determination depends on therapeutic consequences and the frequency of serial measurements on different follow-up care recommendations, e. g. following extensive treatment, on change of therapy, on new staging and on an unclear alteration in the course of disease. Keywords: Tumor marker -tumor marker test -tumor marker statistic -cancer follow-up care Lab.med. 16: 13 (1992) 13 Brought to you by | University of Arizona Authenticated Download Date | 6/7/15 6:41 AM Spezifität RN RN + FP Effizienz Prävalenz RP + RN RP+FP+RN+FN Krank Krank+ nicht Krank Youden-lndex J = (Se+Sp-1) Abb. 1: Einfache statistische Parameter zur Beurteilung von Tumormarker-Tests. Brought to you by | University of Arizona Authenticated Download Date | 6/7/15 6:41 AM Lab.med. 16: 15 (1992) 15 Brought to you by | University of Arizona Authenticated Download Date | 6/7/15 6:41 AM 10 12 Abb. 6: Diagramm der positiv und negativ prädiktiven Werte von Serum-CEA bei Patienten mit Bronchialkarzinom (n = 38) versus Mesotheliom (n = 53) in Abhängigkeit vom Cutoff (Prävalenz *--16 Lab.med. 16: 16 (1992) Brought to you by | University of Arizona Authenticated Download Date | 6/7/15 6:41 AM Abb. 3: Verlaufsuntersuchungen bei Patientinnen mit Ovarialkarzinom (NED = no evidence for disease), Rem. '= Remission, NC -no change, Rez. = Rezidiv und Prog. -Progression nach GTMG (3). Lab.med. 16: 23 (1992) 23 Brought to you by |