Evaluation of changes in sequencing quality and transcriptomic profiles with increasing post-mortem interval: results from an optimization experiment within the UPTIDER tissue donation program

Abstract Background. Postmortem tissue donation programs can importantly enhance sample access for translational research on metastatic disease. However, this post-mortem setting poses logistical and technical challenges in terms of preserving nucleic acid quality and in particular RNA. Here we present the results of an experiment within our breast cancer tissue donation program UPTIDER (NCT04531696), aiming at assessing RNA degradation rates and expression profile changes in function of tissue type and sample-specific postmortem interval (ssPMI). Patients & Methods. For 7 patients, bulk RNA sequencing was performed using the Lexogen protocol on fresh frozen samples from healthy or tumour tissues taken repeatedly (at 1.5h time intervals) during the autopsy. ssPMI was defined as the time between the death of the patient and the freezing of the sample. Quality threshold was set at 0.5 million (M) of assigned reads (AR). Other quality metrics included number of expressed genes, evolution of proliferation-, hypoxia-, stromal- and immune-related transcriptional signatures (PMID:18698033, 20087356) with increasing ssPMI. Associations between quality metrics and ssPMI were assessed by linear regressions for longitudinal data, with quality metrics as dependent variable, time as independent variable and accounting for the clustering of the data by patient and organ using the generalized estimating equation method. Three nested models - with constant, linear and non-linear relationship - were compared using ANOVA testing strategy. Non-linearity was rendered by a restrict cubic spline with three knots. All tests were performed by the Wald test on regression coefficients. Results. Ninety samples (67 healthy, 23 tumour) were analyzed. Median ssPMI was 7.50 hours (range: 3.07-11.12). Most (87%) samples passed quality thresholds, with median AR being 1.70M (interquartile range: [0.70M-3.57M]). No association was found between quality metrics and time in healthy samples. In tumor samples, regarding sequencing quality, negative associations with increasing time were found for AR and for number of expressed genes with an average decay of 242308 reads per hour (95 confidence interval (95CI): [94415.62-390200.50], p-value=.001) and 251 genes per hour (95CI: [17.30-485], p-value=.035), respectively. At the transcriptomic level, potential subtle changes were observed regarding immune (e.g. STAT1 signature: -0.02, 95CI: [-0.05;0.00]) and hypoxia-related signatures (e.g. PGAM1 signature: -0.03, 95CI [-0.05;-0.01]), while no effect of time was seen for the proliferation and stromal-related signatures. Sample size precluded organ specific analyses. Conclusion. A decrease in the number of AR and number of expressed genes with increasing ssPMI was found in tumor samples leading to subtle changes in few transcriptional programs. Healthy samples showed stable quality metrics over time possibly explained by a lower cell activity in healthy as compared to tumour cells. Knowledge derived from this study will be integrated in the upcoming transcriptomic analyses of the samples collected within UPTIDER. Citation Format: François Richard, Tatjana Geukens, Giuseppe Marano, Wouter Van Den Bogaert, Maxim De Schepper, Marion Maetens, Amena Mahdami, Karen Van Baelen, Ha-Linh Nguyen, Anirudh Pabba, Sophia Leduc, Edoardo Isnaldi, Maysam Hajipirloo, Imane Bachir, Evy Vanderheyden, Bram Boeckx, Diether Lambrechts, Ann Smeets, Ines Nevelsteen, Kevin Punie, Patrick Neven, Hans Wildiers, Elia Biganzoli, Giuseppe Floris, Christine Desmedt. Evaluation of changes in sequencing quality and transcriptomic profiles with increasing post-mortem interval: results from an optimization experiment within the UPTIDER tissue donation program [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr P6-14-12.