Development and validation of a clinical prediction model for the risk of distal metastasis in intrahepatic cholangiocarcinoma: a real-world study

Cholangiocarcinoma (CCA) is the second most common primary hepatobiliary malignancy in the world and the most common malignancy of biliary tract [1‐3]. It is second only to hepatocellular carcinoma in terms of incidence, and is characterized by extremely insidious and highly aggressive onset. The highest incidence rates are in Asia, particularly in South Korea, Thailand, and Japan, and the incidence of ICC and ECC has increased in most countries between 1993 and 2012 [4]. Patients can only have a chance of long-term survival if they undergo R0 surgical resection, but most patients are already at an advanced stage of the disease or have developed metastases of distal organ when they are diagnosed in outpatient clinics, and are often deprived of the chance of operation, so the overall prognosis of CCA patients is extremely poor, with an overall survival(OS) rate of less than 10% at 5 years [1, 5‐8]. Additionally, it’s important to note that the risk factors for CCA include chronic liver diseases, such as hepatitis B and C, liver flukes infestation, and certain genetic disorders, which further complicate its global impact and the need for diverse research approaches [9]. On the other hand, even if patients receive surgical treatment, the high recurrence rate after surgery leads to a threat to long-term survival, and some reports show that the OS rate of CCA patients at 5 years after surgery is about 20-35% [10]. In recent decades, the incidence of ICC has gradually increased worldwide [11‐14]. Therefore, studies on the metastasis of CCA are really significant for the diagnosis and treatment of CCA, and will play an important role in the prevention and treatment of CCA.

In recent years, the establishment of statistical prediction models has become a major hot spot in clinical oncology research. Nomogram have been widely used as a highly individualized visual prediction tool for clinical prognosis prediction and decision making [15‐19]. The main feature of nomogram is the ability to create a convenient visual graph that can accurately calculate clinical survival probabilities based on the parameters of a statistical regression model [20‐22]. Based on the lack of good methods for predicting distal metastasis in CCA, we constructed a nomogram for predicting distal metastasis in CCA patients, hoping to improve the prediction of distal metastasis.

CCA is a malignant tumor originating from the epithelial cells of the bile duct system, which accounts for 10-15% of primary liver cancer, but is more malignant [12, 25‐28]. CCA is prone to intrahepatic and lymph node metastasis and lacks effective early diagnostic indicators, so patients are diagnosed at an advanced stage [29‐31]. In recent years, the establishment of statistical prediction models has become a hot topic in clinical research on cancer [32, 33]. We analyzed the clinical data of 345 patients with CCA to explore the factors that may affect the prognosis and to establish a distal metastasis prediction model to effectively evaluate the prognosis of patients. This model can aid clinicians in identifying high-risk patients early, allowing for more personalized and timely interventions, potentially improving the overall prognosis. At the same time, this study is the first to date to focus on predicting the distal risk of CCA.

At present, the diagnosis of CCA is mainly a comprehensive diagnosis combining medical history, imaging diagnosis and serological diagnosis; imaging diagnosis is mainly based on CT and MRI [34], serological diagnosis is mainly CA199 and carcinoembryonic antigen, and the sensitivity and specificity of CA199 for the diagnosis of CCA are 62 and 63% respectively [31, 35, 36]. Meanwhile, preoperative concentration of CA199 greater than 100 U/ml significantly shortens the CCA patients’ postoperative tumor-free survival time [37]. Another study pointed out that preoperative serum CA199 levels higher than 135 U/ml could predict the prognosis of patients with CCA to some extent, but lacked some accuracy [38]. Our findings were similar to our predecessors. The results showed CA199 to be an independent risk factor for the prognosis of distal metastases in diagnosing CCA. These insights can be integrated into clinical decision-making processes, helping clinicians to better predict and manage the course of the disease.

Related studies have shown that residual cancer at the incisal margin was an extremely important factor affecting prognosis [39], and our study also showed that surgical margin invasion was an independent risk factor for predicting distal metastasis from CCA. Multiple infiltration and metastasis could occur early along the peribiliary lymphatics, blood vessels, perineural spaces, and sparse fibrous connective tissue. In addition, the presence of abundant lymphatics, blood vessels, nerve fibers and loose connective tissues in and around the bile ducts provided an avenue for the “jumping” growth of bile duct cancer cells [40]. Understanding these factors can enhance surgical planning and postoperative management, tailoring treatment to individual patient profiles.

Our study showed the important role of hypertension in the prediction of distal metastasis in patients with CCA. About 38% of these patients with distal metastasis had hypertension in this study, whereas only 11% of these patients with non-distal metastasis had hypertension in this study. Further studies are needed to address the mechanisms associated with hypertension and distal metastasis in CCA. This finding suggests that monitoring and managing hypertension in CCA patients could be a key aspect of their overall treatment plan.

Currently, metabolic syndrome was considered a risk factor for CCA [41‐43]. Our study found that cholesterol was an important independent risk factor for the prediction of distal metastasis in patients with CCA. Although the exact role of lipids in CCA was unclear, we believed that cholesterol played a key role in the prediction of distal metastasis in patients with CCA. This observation opens avenues for exploring lipid-lowering strategies as part of the therapeutic approach for CCA.

Our study also has some limitations. This was a retrospective study with data from the single clinical center. The data lack detailed information in some patients’ conditions, such as the specifics of bile duct obstruction or dilatation, vascular invasion, and some biochemical indicators such as prealbumin and some hepatitis virus infections, making it difficult to conduct a more in-depth analysis. At the same time, it was difficult to obtain cases of intrahepatic cholangiocarcinoma, and there were subgroup imbalances in the data. In addition, asa single center clinical study, the results might have some bias.

We collected and processed the clinical data of a large number of patients with CCA to establish a nomogram of distal metastasis in patients with CCA and further evaluated the model. The model is important for predicting the risk of distal metastasis in patients with CCA, providing a risk assessment tool for clinicians to grasp the overall situation of patients, and providing theoretical support for the formulation of patients treatment plans.