We evaluated the long-term stability of thyroid peroxidase antibody (anti-TPO). In the Danish General Suburban Population Study (GESUS), serum samples were biobanked at -80 °C during 2010-2013. In a paired design with 70 subjects, we compared anti-TPO (30-198 U/mL) measured on fresh serum on Kryptor Classic in 2010-2011 (anti-TPOfresh) with anti-TPO remeasured on frozen serum (anti-TPOfrozen) on Kryptor Compact Plus in 2022. Both instruments used the same reagents and the anti-TPOn automated immunofluorescent assay, which was calibrated against the international standard NIBSC 66/387, based on the Time Resolved Amplified Cryptate Emission (TRACE) technology from BRAHMS. Values greater than 60 U/mL are regarded as positive in Denmark with this assay. Statistical comparisons included Bland-Altman, Passing-Bablok regression, and Kappa statistic. The mean follow-up time was 11.9 years (SD: 0.43). For anti-TPOfrozen vs. anti-TPOfresh, the line of equality was within the confidence interval of the absolute mean difference [5.71 (-0.32; 11.7) U/mL] and the average percentage deviation [+2.22% (-3.89%; +8.34%)]. The average percentage deviation of 2.22% did not exceed analytical variability. Passing-Bablok regression revealed both a statistically significant systematic and proportional difference: Anti-TPOfrozen=-22.6 + 1.22∗(anti-TPOfresh). Frozen samples were correctly classified as positive in 64/70 (91.4%; Kappa=71.8%). Anti-TPO serum samples in the range 30-198 U/mL were stable after 12-years of storage at -80 °C with an estimated nonsignificant average percentage deviation of +2.22%. This comparison is based on Kryptor Classic and Kryptor Compact Plus, which used identical assays, reagents, and calibrator, but for which the agreement in the range 30-198 U/mL is unclarified.