Impact of Single Freeze-Thaw Cycles on Serum Protein Stability: Implications for Clinical Biomarker Validation Using Mass Spectrometry

Abstract

BackgroundValidation of biomarkers for clinical diagnostics and therapy necessitates the availability of a substantial number of high-quality samples, along with a complete set of clinical data. Establishing standards for sample collection, storage, and quality control is essential to reduce the variability of sample quality. MethodsThis study evaluated the impact of a single freeze-and-thaw (FT) cycle on the quality of liquid nitrogen-stored clinical serum aliquots. Data-independent acquisition mass spectrometry (DIA-MS) was performed to measure serum protein abundance in a test cohort comprising 25 patients and 99 samples, and a validation cohort comprising 109 patients/samples. Abundance differences of paired fresh and FT samples were assessed by employing biostatistics and bioinformatics approaches, including clustering analyses, linear mixed models, machine learning, and functional annotation. ResultsFollowing the library-free data analysis and implementation of strict data preprocessing procedures, data on the relative abundance of 213 (test cohort) and 248 (validation cohort) human serum proteins was available. Overall, the proteomic data demonstrated high quality and reproducibility. Significant changes in abundance between fresh and FT samples was observed for 30 proteins in the test cohort (q-value [≤]0.05). Of these, 11 proteins (36.6%) were successfully validated in the validation cohort, including CP and IGHV6-1 which were identified as potential biomarker candidates for discriminating between patients with malignant or benign pancreatic disease. A correlation analysis of the measured protein intensity and storage duration revealed no significant association between the two variables. ConclusionsDistinct protein abundance patterns can be discerned between fresh serum samples and samples stored in liquid nitrogen. The findings of this study suggest that FT cycles are an important pre-analytical factor and should be addressed during the translation of protein-based biomarkers into clinical use.