Background: The liquid chromatography-tandem mass spectrometry (LC-MS/MS) method was developed to overcome interference from thyroglobulin autoantibodies (TgAb); however, it has not yet been implemented in Korea. This study aimed to confirm the accuracy of LC-MS/MS compared to conventional methods and to identify its advantages in patients with thyroid carcinoma (TC). Methods: A total of 206 TC and 18 Hashimoto’s thyroiditis samples were collected. TgAb-positive (TgAb-P) was defined as TgAb >60 U/mL. Tg testing was performed using LC-MS/MS, immunoradiometric assay (Tg-IRMA), and chemiluminescence microparticle immunoassay (Tg-CMIA). The interference of TgAb in LC-MS/MS and CMIA methods was evaluated through an in vitro TgAb spiking experiment. Results: The frequency of TgAb-P in TC samples was 76.2%. Correlations between assays were as follows: Tg measurements made by LC-MS/MS (Tg-MS) and Tg-IRMA (R=0.93), Tg-MS and Tg-CMIA (R=0.96), and Tg-CMIA and Tg-IRMA (R=0.99), and it was lower in TgAb-P than TgAb-negative group. Clinical factors (total thyroidectomy, thyroid lobectomy, and Hashimoto’s thyroiditis) did not affect these correlations. In TgAb spiking experiments, Tg-CMIA showed false negatives in TgAb-P, whereas Tg-MS did not. Among 21 TC cases with highly suspicious disease recurrence but Tg-IRMA <1 ng/mL, Tg-MS detected Tg ≥0.5 ng/mL in six samples. However, there was no consistent pattern of recurrence or TgAb trends. Conclusion Correlations between assays were lower in TgAb-P cases. The spike test results show Tg-MS is less prone to false negatives in TgAb-P cases. Tg-MS may improve Tg detection in TgAb-P cases. However, we could not identify a distinct patient group with shared clinical features that would benefit from Tg-MS.

Immune-related adverse events, including immune hemolytic anemia, have been reported in patients treated with immune checkpoint inhibitors. In particular, RBC autoantibodies are important because they can cause hemolytic anemia and interfere with pre-transfusion tests. On the other hand, there are few reports on the characteristics of RBC autoantibodies induced by immune checkpoint inhibitors in Korea. The medical history and laboratory results, including pretransfusion tests of ten patients treated with immune checkpoint inhibitors, were reviewed retrospectively. The median interval from the first administration of immune checkpoint inhibitors to the development of autoantibodies was 12 weeks. In eight patients, only cold autoantibodies were developed. Both warm and cold autoantibodies developed in one patient, and warm autoantibodies alone were detected in one patient.Of seven patients tested by a direct antiglobulin test, two were negative, and the remaining five were positive for IgG and negative for C3d. In conclusion, this study presented ten cases of autoantibody developments in patients treated with immune checkpoint inhibitors and the possible relationship between the immune checkpoint inhibitors and RBC autoantibody development. Further comprehensive studies will be needed to elucidate this relationship.