Background: The availability of a targeted therapeutic antibody (trastuzumab) for HER2 positive tumours stresses the necessity to identify patients reliably who might benefit from trastuzumab treatment not only important for clinical reasons - positive clinical effects, but severe adverse events with grade 3 or 4 cardiotoxicity - but also for economic ones because of the high treatment costs. The research question is: What is the gold standard for diagnosing HER2 positive tumours? Which is the most accurate and reproducible method to find out the candidates for potential therapy with monoclonal antibodies, and are the applied tests reliable for selecting HER2 positive patients?

Method: Systematic review of studies on questions of validity and standardization of the 2 most commonly used methods IHC (immuno histochemistry) and FISH (fluorescence in situ hybridization), furthermore the inter-observer and the inter-laboratory concordance esp.in the diagnosis of borderline results.

Results: The review discusses - systematically - the HER2 testing results of more than 23 000 specimens, achieved in local, central or reference laboratories. IHC results show much greater variability than FISH test results, particularly for the FISH negative cases. The results of most presented studies indicate that high-level HER2 amplification and an IHC score of 3+ will identify HER2 positive breast carcinoma; the low-level amplification and/or IHC score of 2+ should be carefully interpreted. There is agreement that the most (cost-)effective testing strategy is to screen all patients with IHC, followed by FISH for IHC of 2+ (or of 2+ and 3+). Findings concerning different results from local/central laboratories point at moderate inter-observer and inter-laboratory reliability of test results. A volume/experience relationship is observed.