The present multicenter prospective study has shown that serum TN-C can be a useful biomarker for treatment selection in the acute phase of KD. The serum level of TN-C upon admission of the patients who were categorized as high-risk according to their Kobayashi score was significantly higher than that of the low-risk patients. This finding is consistent with the results of our previous retrospective study, suggesting that serum TN-C alone could be a biomarker for identifying high-risk patients that is comparable with the Kobayashi score [8].
Histologically, coronary arteritis begins 6–8 days after KD onset and is characterized by inflammation consisting of a marked accumulation of monocytes/macrophages [37]. TN-C is expressed in the areas where inflammatory lesions form on the coronary arteries during the acute stage of KD, and the intensity of its expression correlates with the degree of inflammation [38]. Therefore, a high serum level of TN-C may reflect the severity of inflammation.
Japanese guidelines allow the stratification of KD patients by prediction scoring systems of IVIG resistance, treating the low-risk patients with IVIG and treating the high-risk patients with a steroid combined with conventional IVIG as the first-line therapy. However, unresponsiveness to this treatment protocol has become a major problem recently. In our present study, 28 out of the 114 low-risk patients were not responsive to the IVIG, and 10 out of the 37 high-risk patients were not responsive to the steroid combination therapy, consequently needing additional therapies. Importantly, in the high-risk patients, the serum TN-C levels upon admission for the initial treatment-resistant patients were significantly higher than those of the initial treatment-responsive patients. This finding suggests that the stratification of KD severity by Kobayashi score has limitations and that TN-C could be used as a biomarker to identify patients at greater risk of resistance to first-line treatment among the patients classified as high-risk based on their Kobayashi score.
Jone et al. demonstrated that the use of IVIG plus infliximab as initial therapy reduces the need for additional therapy in KD patients presenting with CALs [39]. Hamada et al. demonstrated that treatment with IVIG plus cyclosporine was safe and effective for favorable coronary artery outcomes in high-risk KD patients [40]. Notably, KD patients who are predicted to be at high risk require the use of a more potent treatment in addition to or instead of treatment with steroids/IVIG.
In the low-risk patients, the TN-C level upon admission was not significantly differenent between the initial treatment-responsive and initial treatment-resistant patients, suggesting that TN-C may not be a predictor of IVIG-resistance in this group. Thus, the TN-C level may be useful as a predictive biomarker in only the high-risk cases.
In both high-risk and low-risk patients, the first-line therapy significantly reduced the TN-C levels in the initial treatment-responsive patients but did not in the initial treatment-resistant patients. The patients with higher TN-C levels after the initial treatment needed a second-line therapy. These findings suggest that the TN-C level reflects the effectiveness of the treatment and could be used as a rationale for commencing additional therapy. Because steroids often mask the symptoms of inflammation, their use can make it difficult for physicians to accurately judge whether the inflammation has decreased or if further intervention is needed. Based on the findings of the present study, the TN-C level might be useful as an indicator of whether additional treatment is needed.
In this study, there was no difference in TN-C levels of patients with or without CALs. This could be because the number of CALs in our enrolled patients was too small for accurate evaluatation. However, it is known that treatment resistance is closely related to CAL onset [6–8]. If some biomarkers, such as TN-C, could predict a severe course of disease, it would be expected that more aggressive initial/additional treatment might reduce the incidence of CALs.
There are some limitations of this study. A sample size that might not be large enough to determine severe cases who need additional treatment. Among our enrolled subjects, there were 37 cases with high-risk patients, of which only 10 cases (5% of all subjects) did not respond to steroid combination therapy. Because only 5% of our subjects had such refractory cases, more subjects should be recruited for future studies. Second, a total of 199 patients (52%) were excluded because of not meeting the criteria. However, there were no significant differences in age, Kobayashi score and labolatory data between all 380 patients and 181 subjects. It seems no bias in patient distribution.