Proteomic analysis is an important tool in classifying molecular architecture of tissue samples for subtyping and precision oncology applications. Tissues are commonly either flash frozen (FF) or embedded in optimal cutting temperature (OCT) compound for long-term storage. OCT provides a convenient specimen matrix for cryostat sectioning but contains glycols and resins which are incompatible with mass spectrometric analysis of proteins. In contrast, flash frozen preservation of tissue is historically considered to be more suitable and compatible with mass spectrometry applications. As a pilot study, we evaluated different tissue preservation methods, sample processing methods, and protein extraction methods, and their impact on quantitation and coverage of the proteome. In this pilot study, 30 breast cancer cases with matching flash frozen and OCT-embedded samples were used for proteomic analysis. Tissues were procured from the Clinical Breast Care Project (CBCP) in partnership with the Chan Soon-Shiong Institute of Molecular Medicine at Windber, and the Walter Reed National Military Medical Center (WRNMMC). Breast cancer subtypes were determined by clinical and expanded immunohistochemical panels. These subtypes were Her2, Luminal A (LA), Luminal B (LB) and Triple Negative (TN). Flash frozen tissues were cryosectioned to isolate 60-100 mg total tissue per case, and proteins were extracted using an 8M urea lysis buffer. The OCT specimens were laser microdissected to isolate primarily tumor cells from the samples. Proteins were extracted using the illustra TriplePrep kit. Digested proteins were then multiplexed using Tandem-Mass-Tag (TMT) 10plex isobaric labeling reagents, and chromatographically resolved using a Waters 2D nanoAcquity liquid chromatographer (LC) coupled to a Thermo Q Exactive Plus mass spectrometer (MS). Data was then analyzed using Proteome Discoverer. In total, 6130 proteins were identified using a 1% peptide FDR confidence filter in the 30 samples analyzed. Unsupervised hierarchical clustering of proteins across the samples resulted in two primary clusters, OCT and FF. Within the primary clustering, there were sub-clusters of Luminal, Her2, and Triple Negatives. At 2-fold threshold, 246 proteins were differentially expressed between OCT and FF tissues. However, the differential proteins between ER+ (LA and LB) and ER- (Her2 and TN) tumors correlate irrespective of the cryopreservation and processing methods. In conclusion, proteome analysis of samples stored under two different conditions provide subtyping information; however, there is differential expression of proteins between the two types of sample handling techniques which can impact biological interpretation of breast cancer precision oncology studies.

Citation Format: Kiki Panagopoulos, Punit Shah, Albert Kovatich, Michael A. Kiebish, Jeffrey Hooke, Leigh Campbell, Mary Lou Cutler, Audrey Kovatich, Hai Hu, Brenda Deyarmin, Lori Sturtz, Praveen Kumar Raj Kumar, Stella Somiari, Emily Y. Chen, Elder Granger, Viatcheslav Akmaev, Rangaprasad Sarangarajan, Niven Narain, Craig D. Shriver. Proteomic comparative assessment of flash-frozen and OCT embedded breast cancer tissues for utilization in precision oncology discovery studies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 5660.