Assessment of Telomere Length in Archived Formalin-Fixed, Paraffinized Human Tissue Is Confounded by Chronological Age and Storage Duration

Telomeres shorten with physiological aging but undergo substantial restoration during cancer immortalization. Increasingly, cancer studies utilize the archive of formalin-fixed, paraffin-embedded (FFPE) tissues in diagnostic pathology departments. Conceptually, such studies would be confounded by physiological telomere attrition and loss of DNA integrity from prolonged tissue storage. Our study aimed to investigate these two confounding factors. 145 FFPE tissues of surgically-resected, non-diseased appendixes were retrieved from our pathology archive, from years 2008 to 2014. Cases from 2013 to 2014 were categorized by patient chronological age (0–20 years, 21–40 years, 41–60 years, > 60 years). Telomere lengths of age categories were depicted by telomere/chromosome 2 centromere intensity ratio (TCR) revealed by quantitative fluorescence in situ hybridization. Material from individuals aged 0–20 years from years 2013/2014, 2011/2012, 2009/2010, and 2008 were compared for storage effect. Telomere integrity was assessed by telomere fluorescence intensity (TFI). Epithelial TCRs (mean ± SD) for the respective age groups were 4.84 ± 2.08, 3.64 ± 1.21, 2.03 ± 0.37, and 1.93 ± 0.45, whereas corresponding stromal TCRs were 5.16 ± 2.55, 3.84 ± 1.36, 2.49 ± 1.20, and 2.93 ± 1.24. A trend of inverse correlation with age in both epithelial and stromal tissues is supported by r = -0.69, p < 0.001 and r = -0.42, p < 0.001 respectively. Epithelial TFIs (mean ± SD) of years 2013/2014, 2011/2012, 2009/2010 and 2008 were 852.60 ± 432.46, 353.04 ± 127.12, 209.24 ± 55.57 and 429.22 ± 188.75 respectively. Generally, TFIs reduced with storage duration (r = -0.42, p < 0.001). Our findings agree that age-related telomere attrition occurs in normal somatic tissues, and suggest that an age-based reference can be established for telomere studies on FFPE tissues. We also showed that FFPE tissues archived beyond 2 years are suboptimal for telomere analysis.