In recent years, the incidence of DTC has increased globally [22, 23]. As a specific tumor marker for long-term follow-up in DTC, Tg can predict DTC persistence, distant metastasis or recurrence. However, the Tg level can be affected by TgAb, which made Tg measurement becomes unreliable and limited the role of Tg in prognosis in DTC patients’ follow-up. Therefore, TgAb has gained increasing attention as an essential indi-cator for the DTC surveillance. Prior studies of the clinical importance of TgAb in DTC have typically focused on the sequential changes in TgAb titers [12–16]. Many studies have reported that the stable, or a rising trend of TgAb titers after TT is an effective risk factor to predict persistent or recurrent of disease [24–26]. Pan et al. indicated that TgAb affected sTg measurements even if it was negative. They novely reported a negative correlation between TgAb and Tg levels, suggesting that sTg × TgAb product could be used to predict prognosis in patients with DTC [17]. But the majority of patients in that study were TgAb-negative patients (321/409). In view of that, the present study explored the prognostic value for Tg×paTgAb in TgAb-positive DTC patients since this part of patients were not rarely and accounted for almost 25% in DTC [6].
We totally enrolled 265 TgAb-positive DTC patients in the present study. The results revealed that preop-TgAb, paTgAb and sTg×paTgAb were significantly correlated with excellent response to RAIT. Some previous reports indicated that high pre-operative TgAb was an independent predictor of nodal metastases in DTC patients [27, 28]. Some studies [18, 29] indicated that paTgAb showed significant differences between different efficacy groups. Pan et al. [17] indicated sTg × paTgAb could be used to predict prognosis in patients with DTC. Those results were similar with our present study. Our data also suggested that maximal tumor diameter, primary lesions, tumor with ETE and N stage were significantly different in ER group and NER group, which consistent with previous reports [14, 30].
In order to confirm the predictive value of the three factors, we performed ROC curves for further analysis. The results demonstrated that the cut-off value of preop-TgAb, paTgAb, sTg×paTgAb was724.25IU/ml, 424.00 IU/ml and 59.73. According to the cut-off values, we divided the preop-TgAb, paTgAb and sTg×paTgAb product into two groups, respectively. The multivariate logistic regression analyses indicated that paTgAb, sTg×paTgAb and N stage were the independent risk factor of NER. The risk of NER increases by 2.640-fold for each 1 kU/L increase in the level of paTgAb and 2.846-fold for sTg×paTgAb. As mentioned before, Pan et al. [17] reported that the sTg×TgAb product cut-off value in TgAb-positive group was 466.36, which was much higher than our study. One reason for the difference may the definition of TgAb-positive was different, which in our study was 40IU/ml before initial RAIT, while in Pans’ study was 115 IU/ml. This difference led to the patients’ demographical natures were different. The other reason may be the differences of sample size in the studies, which the sample size for TgAb-positive patients in the Pan’s study was smaller than us (88 vs.265).
In addition, we found that the median DFS of patients with sTg×paTgAb<59.73 had a longer DFS than patients with sTg×paTgAb ≥ 59.73 (50.27months vs. 48.59months, p = 0.041), which confirmed the prognostic value for TgAb-positive patients. But the result of Kaplan-Meier analysis showed similar median DFS of patients with paTgAb<424IU/ml and TgAb ≥ 424IU/ml.
Nonetheless, there were still some limitations in our study. On the one hand, this study was a retrospective single-center study, and the sample size was relatively small, then the selection bias can’t be avoided. Therefore, further prospective multicentered with large-scale studies are required for validation. On the other hand, longer periods of follow-up are required to confirm the current findings.
In summary, we found that the sTg and paTgAb product can predict the efficacy as well as prognosis of RAIT for TgAb-positive DTC patients. Since sTg was interfered with TgAb before RAIT in these patients, the product of sTg and paTgAb can be an effective clinical reference indicator to recognize the NER patients in early time, which has a very positive and important role in optimizing treatment, improving prognosis and reducing the burden of patients.