RNAlater® Tissue Collection

Microarray analysis has become a powerful method for target gene discovery and molecular tumor characterization. It is also increasingly employed in clinical settings where gene expression profiling can be useful for predicting drug therapy response and patient outcome.

Sample preservation (i.e., stabilization of nucleic acids) has always been an important aspect of gene expression studies, but adapting sample collection protocols to a clinical setting can be difficult when the priority is the surgery and/or the pathologist’s assessment. Since liquid nitrogen is usually not available in most operating rooms, tissues obtained from surgery are often stored at ambient temperature or on ice for varying periods of time. While these samples may still have diagnostic and prognostic value for the pathologist, the RNA degradation that occurs during this period can result in assay variability and compromised reproducibility. The RNA modification and cleavage that occurs with formalin fixation: paraffin embedding (FFPE) protocols also complicates global gene expression profiling experiments.

An alternative to snap-freezing tissues in liquid nitrogen, RNAlater Tissue Collection: RNA Stabilization Solution provides an effective and convenient way to preserve samples. To confirm that RNA later Solution can preserve RNA in clinical samples, Dr. Chowdary and colleagues conducted microarray experiments comparing RNA from 51 paired samples from colon (Stage II, Dukes B) and breast (lymph node negative) cancer patients. A tissue section used for pathology verification was taken from the middle of the tissue sample, and one half of the remaining tissue was immersed in RNA later Solution, while the other half was snap-frozen. The results from these experiments were used in prognostic algorithms [2, 3] designed to predict recurrence of colon cancer or breast cancer.

Gene Expression Profiles: Based on analysis using an Agilent® bioanalyzer, no significant differences in RNA yield and quality were observed between RNA isolated from RNA later-treated or snap-frozen samples. Biotin labeled cRNA targets were amplified utilizing RNA from matched colon and breast tumor sample pairs and hybridized to Affymetrix U133A arrays. Strong correlations in global expression levels were observed in each matched sample pair (median Pearson correlation coefficients of 0.94 and 0.97 for colon tumor pairs and breast tumor pairs, respectively). Additional analyses were reported that examined false discovery rates, variability introduced by the sample acquisition protocol, overlap in differentially expressed genes between sample pairs, and hierarchical clustering to show relationships among gene expression patterns from each sample.

Prognostic Signatures: Using previously described algorithms [2, 3], a strong correlation in the Relapse Hazard Score (RHS) and prediction of cancer recurrence (96%) from the paired samples was observed. Not enough time has passed since sample collection to compare these results to actual relapse rates in these patient groups.