Intrahepatic cholangiocarcinoma frequently metastasizes to regional lymph node, which be associated with tumor recurrence after curative-intent resection. The association between lymph node metastasis and prognosis of ICC patients has not been fully investigated and validated. In the current study, lymph node metastasis was present in roughly 40% of ICC patients who underwent surgical resection. Even several studies have reported an higher incidence rate of lymph node metastasis in patients with ICC, as high as 45–62%[17, 18].
Therefore, we attempted to identify valuable predictor for lymph node metastasis in patients with ICC, as this information might be helpful in clinical practice. Several studies reported that tumor number and tumor diameter are significant predictive factors for LN metastasis in ICC patients[19, 20].Our study found that age, T stage, and tumor size were the best-fit predictor for LN metastasis in ICC patients. In addition, poorly differentiated ICC was also associated with a higher rate of lymph node metastasis [20].
Lymph node metastasis was reported to be one of the most significant prognostic factors for survival from ICC patients[9, 21]. However, despite the prognostic relevance of the LN status, several investigators reported that lymph node dissection (LND) does not seem to improve the OS of patients with ICC[22, 23].Actually, a retrospective large cohort study suggested that adjuvant chemotherapy could improve survival of ICC patients with lymph node metastasis, or advanced-stage[24–28]. Despite making a progress in understanding the pathophysiology of ICC and the emergence of novel treatment options[29], liver resection remains the only potential cure for ICC patients, whose survival has not improved significantly in recent decades [5, 30, 31].
Our research showed that ICC patients with LNM with high grade, absence of chemotherapy, and absence of surgery had unfavorable prognosis. Based on three independent prognostic factors of ICC patients with LNM, an individual nomogram for predicting OS was established and evaluated. The results indicated that the nomogram can serve as an effective tool to identify high-risk patients. It is well accepted that tumor grade could explain some of the heterogeneity associated with the expected course and clinical outcome in patients with various tumors[32, 33]. Tumor grade is a measure of the degree of differentiation of the tumor and is related to the prognosis of patients with tumor. In the present study, tumor grade was also observed as an independent prognostic predictor for OS in ICC patients with LNM. As far as we know, it is the first study to build a nomogram for predicting OS in ICC patients with LNM. Our nomogram, incorporating easily accessible factors in clinical practice, enabled easy calculation of individualized OS probabilities for ICC patients with LNM. Moreover, the nomograms showed relatively high accuracy with OS, and calibration curves display well-fitted in both the training set and testing set. In addition, high predictive accuracy does not mean that the predictive model has high clinical practicability [13, 15]. Decision curve analysis is one of the recommended methods in previous research on prognostic predictive models. It can quantify the net benefits of predictive models according to the threshold probability, so as to evaluate the clinical practical value of a predictive model[34–36]. Therefore, we introduced the decision curve analysis to examine the effectiveness of our nomogram in clinical practice. The decision curve analyses not only confirmed the clinical validity of our nomogram for OS, but also demonstrated that our nomogram had better clinical application value in ICC patients with LNM.
However, our research also has some limitations. First, limited patients (N = 381) of ICC with LNM may result the possible error in our study. Second, the information collected in the SEER database was about the ICC disease at the first diagnosis, which meant that the lymph node metastasis occur in the latter stage cannot be recorded. Third, this was a retrospective study in which patients’ selection bias existed inevitably, and the information about detailed treatment and progress after treatment was not available in the SEER database. However, notwithstanding these limitations, our study is the first to use a nomogram combined with clinically feasible variables including tumor grade, chemotherapy, and surgery of primary site to predict the OS of ICC patients with LNM, which has clinical significance for predicting prognosis of ICC patients with LNM.