For interpretation of quantitative gene expression measurements in clinical tumor samples, a normalizer is necessary to correct expression data for differences in cellular input, RNA quality, and RT efficiency between samples. In many studies, a single housekeeping gene is used for normalization. However, no unequivocal single reference gene (with proven invariable expression between cells) has been identified yet. As the best alternative, the mean expression of multiple housekeeping genes can be used for normalization. In this study, no attempt was made to determine the gold-standard gene for normalization, but to identify the best single housekeeping gene that could accurately replace the measurement of multiple genes. Expression patterns of 13 frequently used housekeeping genes were determined in 80 normal and tumor samples from colorectal, breast, prostate, skin, and bladder tissues with real-time quantitative RT-PCR. These genes included, large ribosomal protein, beta-actin, cyclophilin A, glyceraldehyde-3-phosphate dehydrogenase, phosphoglycerokinase 1, beta-2-microglobin, beta-glucuronidase, hypoxanthine ribosyltransferase (HPRT), TATA-box-binding protein, transferrin receptor, porphobilinogen deaminase, ATP synthase 6, and 18S ribosomal RNA. Principal component analysis was used to analyze these expression patterns, independent of the level of expression. Our approach identified HPRT as the single best reference gene that could be used as an accurate and economic alternative for the measurement of multiple housekeeping genes. We recommend this gene for future studies to standardize gene expression measurements in cancer research and tumor diagnostics until a definite gold standard has been determined.