More and more samples are obtained from biobanks for biomedical research; however, some of these samples may undergo thawing before processing. We aim to evaluate the reference gene expression stability in thawed renal cell carcinoma samples. Sixteen matched malignant and nonmalignant renal tissue samples were obtained and each sample was divided into 4 aliquots before being snap frozen and stored at −80°C. By quantitative real-time polymerase chain reaction, a time-course study was conducted on the thawed tissue to evaluate the expression stability of a panel of the 10 most frequently used reference genes in renal cell carcinom samples: ACTB, ALAS1, B2M, GAPDH, HMBS, HPRT, PPIA, RPLP0,TBP, and TUBB. As shown by geNorm M values, PPIA was the most stable gene at the 0-, 15-, and 30-minute time points (M=0.82, 0.85, and 0.76, respectively), whereas GAPDH was ranked last at the 5-, 15-, and 30-minute time points (M=1.38, 1.44, and 1.39, respectively). A positive correlation was found by linear regression between the thawing time and 2 to the power of crossing point values of all candidate reference genes (P<0.05). The mean coefficient of variance of all reference genes increased significantly at time points 5, 15, and 30 minutes compared with 0 minutes (P<0.01). In conclusion, using the geNorm algorithm, PPIA was identified as the most stably expressed gene between malignant and nonmalignant renal tissue samples that were thawed for similar time periods. All the reference genes showed high variations along with the thawing time; it should be recommended to use a combination of several candidate reference genes when comparing samples thawed for different time periods.
