Receptor tyrosine kinases (PTKs) play key roles in the pathogenesis of numerous human diseases, including cancer, and therefore PTK inhibitors are currently under intense investigation as potential drug candidates. PTK inhibitor screening data are, however, poorly comparable because of the different assay technologies used. Here we report a comparison of ELISA-based assays for screening epidermal growth factor receptor (EGFR) tyrosine kinase (TK) inhibitory compound libraries to study interassay variations. All assays were based on the same protocol, except for the source of EGFR-TK enzymes. In the first protocol, the enzyme was isolated from A431 cells without affinity purification. In the second protocol, commercial EGFR-TK (Sigma) isolated from A431 cells by affinity-purification was employed. In the third protocol, an enzyme preparation obtained from a recombinant (Baculovirus transfected Sf9 cells) expression system was used. All assays employed the synthetic peptide substrate poly-(Glu,Tyr)l:4 and an ELISA-based system to detect phosphorylated tyrosine residues by a monoclonal antibody. We observed significant differences in both the activity of the enzymes and in the EGFR-TK inhibitory effect of our reference compound PD153035. The differences were significant in case of A431 cell lysate compared to affinity purified EGFR-TKs derived from either A431 cells or Baculovirus transfected Sf9 cells, whereas the latter two showed comparable results. Our data suggest that differences in terms of interassay variation are not related to the source of the enzyme but to its purity; changes in the mode of detection can markedly influence the reproducibility of results. In conclusion, normalization of the EGFR activity used for inhibitor screening and standardization of detection methods enable safe comparison of data.