Prognostic Prediction of Advanced Intrahepatic Cholangiocarcinoma Patients Receiving Chemotherapy Combined with Immunotherapy Based on Serum Lipid Profiles

Prognostic Prediction of Advanced Intrahepatic Cholangiocarcinoma Patients Receiving Chemotherapy Combined with Immunotherapy Based on Serum Lipid Profiles | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Prognostic Prediction of Advanced Intrahepatic Cholangiocarcinoma Patients Receiving Chemotherapy Combined with Immunotherapy Based on Serum Lipid Profiles Chengzhi Jiang, Kaijun Long, Tianyuan Fang, Wen Li, Liu Yang, and 3 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7803902/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background: Lipid metabolic reprogramming plays a critical role in tumor progression. Serum lipid levels have been associated with the prognosis of various malignancies. Aims: To develop a novel nomogram based on serum lipid parameters to predict overall survival in patients with intrahepatic cholangiocarcinoma. Methods: Serum lipid profiles and survival data were collected prior to the initiation of chemotherapy combined with immunotherapy. Survival analysis was performed to identify prognostic factors associated with ICC. Independent prognostic factors were used to construct a nomogram. The predictive performance of the nomogram was evaluated. External validation of the survival analysis and nomogram for serum lipids was conducted using a validation cohort. Results: Low-density lipoprotein cholesterol, high-density lipoprotein cholesterol, and apolipoprotein A1 were selected for further analysis. Survival analysis demonstrated that patients with low LDL-C, high HDL-C, and high ApoA1 levels exhibited significantly longer OS and PFS. A nomogram incorporating LDL-C and HDL-C was constructed to predict 1-, 2-, and 3-year survival probabilities. The nomogram exhibited favorable predictive performance. Discussion: P re-treatment serum levels of LDL-C, HDL-C, and ApoA1 exhibited significant prognostic value for advanced ICC. The nomogram constructed based on LDL-C and HDL-C effectively predicted survival outcomes, providing a theoretical basis to support treatment decision-making and individualized prognostic assessment in clinical practice. Serum lipid levels Intrahepatic Cholangiocarcinoma Immunotherapy Nomogram Figures Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6 Background Intrahepatic cholangiocarcinoma (ICC) is the second most common primary liver malignancy after hepatocellular carcinoma. Due to its insidious onset and lack of specific early symptoms, ICC is often diagnosed at an advanced stage. Although various therapeutic strategies—including chemotherapy, targeted therapy, and chemotherapy combined with immunotherapy—are currently available for advanced ICC, the overall clinical outcomes remain unsatisfactory, with a 5-year survival rate of less than 20% in most patients [1] . Notably, patients receiving chemoimmunotherapy exhibit longer overall survival (OS) compared to those receiving chemotherapy alone; however, the overall survival rates remain suboptimal, indicating substantial challenges in the immunotherapeutic management of advanced ICC [2,3] . Owing to factors such as low expression of immune checkpoints and intrinsic resistance of tumor cells, immune checkpoint inhibitors (ICIs) demonstrate limited efficacy in a subset of patients. Therefore, the identification of reliable biomarkers to stratify patients who are more likely to benefit from ICIs is essential for optimizing treatment strategies and guiding future therapeutic development [4] . Lipid metabolic reprogramming plays a pivotal role in various biological processes of tumor cells, including proliferation, survival, migration, invasion, and metastasis [5–12] . Compared to monolayer-cultured ICC cells, ICC stem-like cells exhibit enhanced de novo fatty acid synthesis activity and higher expression levels of key enzymes involved in fatty acid biosynthesis, such as fatty acid synthase (FASN). Clinically, ICC patients with high FASN expression are associated with poorer long-term survival outcomes [13] . Correspondingly, inhibition of the mTOR signaling pathway can suppress FASN-mediated fatty acid synthesis, thereby reducing fatty acid oxidation in tumor-associated macrophages and promoting antitumor immune responses [14] . Additionally, other studies have demonstrated that highly proliferative cholangiocarcinoma cells exhibit significantly higher lipid uptake compared to normal cholangiocytes. Treatment with fatty acid oxidation (FAO) inhibitors led to suppressed proliferation in several cancer cell types, including cholangiocarcinoma, suggesting that blockade of fatty acid catabolism may effectively inhibit tumor growth [15] . Moreover, lipid metabolic reprogramming is closely associated with the efficacy of tumor immunotherapy. For example, cPLA2α activity driven by tumor cells and regulatory T cells (Tregs) can induce lipid droplet accumulation in effector T cells, resulting in T cell senescence. Inhibition of cPLA2α has been shown to prevent effector T cell exhaustion and enhance the efficacy of immunotherapy [16] . Similarly, PD-L1-containing tumor-derived extracellular vesicles can also promote lipid droplet accumulation in T cells, leading to senescence and resistance to immunotherapy [17] . Thus, lipid metabolic reprogramming may exert critical regulatory effects on tumor immunotherapy by modulating the immune microenvironment [18] . With the widespread application of metabolomics, an increasing number of studies have revealed significant associations between alterations in serum lipid levels and the prognosis of various malignancies [19] . As common clinical biochemical indicators, serum lipids are essential components involved in energy storage, metabolism, and cellular signal transduction. Changes in serum lipid levels can indirectly reflect lipid alterations within the tumor microenvironment, thereby representing potential biomarkers for predicting the efficacy of immunotherapy. A retrospective study reported that low levels of triglycerides and high-density lipoprotein cholesterol (HDL-C) were strongly associated with recurrence in patients with thyroid cancer [20] . Earlier research also suggested that serum triglyceride and HDL-C levels were correlated with prostate cancer severity [21] . Sun et al. identified a causal association between elevated LDL-C and gastric cancer. Furthermore, triglyceride levels ≥2.2 mmol/L were found to increase the risk of gallbladder cancer in men over the age of 60 [9] . However, studies exploring the prognostic value and predictive significance of serum lipid levels in relation to immunotherapeutic outcomes in ICC remain scarce. This study analyzed the association between pre-treatment serum lipid levels and prognosis in patients with advanced ICC undergoing chemotherapy combined with immunotherapy. Independent prognostic risk factors were identified, and a nomogram-based prognostic model was constructed to provide a potential reference for clinical decision-making and prognostic assessment (Figure 1). Methods Study Population This study enrolled 263 patients diagnosed with unresectable advanced ICC at the Harbin Medical University Cancer Hospital between January 2014 and January 2024. Inclusion criteria were as follows: (1) Histopathological confirmation of ICC; (2) Presence of measurable lesions according to the Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST 1.1); (3) Effective radiological assessments performed prior to treatment and after every 2–3 treatment cycles; (4) Treatment regimen consisting exclusively of chemotherapy combined with immune checkpoint inhibitors (ICIs); (5) Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 2; and (6) Patient age between 18 and 80 years. Exclusion criteria were as follows: (1) Incomplete clinical data; (2) Absence of measurable lesions; (3) History of other histologically confirmed malignancies within the past five years; (4) Presence of severe organ dysfunction; (5) Diagnosis of autoimmune deficiency disorders; (6) Use of medications known to affect serum lipid levels during the treatment period, such as statins or fibrates; (7) Use of medications known to affect serum lipid levels during the treatment period, such as statins or fibrates; (8) History of prior locoregional liver therapies, including surgery, interventional procedures (e.g., ablation, embolization, or brachytherapy); (9) Patients who declined follow-up. Clinical Data Collection Clinical records were retrospectively reviewed to collect baseline clinical data from patients who met the inclusion and exclusion criteria. The collected variables included sex, age, smoking history, history of chronic alcohol consumption, Eastern ECOG performance status, American Joint Committee on Cancer (AJCC) staging, histological differentiation, Child–Pugh classification, and primary tumor size. Serum lipid profiles were collected within one week prior to the initiation of systemic therapy, including LDL-C, HDL-C, total cholesterol, triglycerides (TG), apolipoprotein B (APOB), apolipoprotein A1 (APOA1), and lipoprotein alpha (Lpα). Efficacy Evaluation and Follow-up Computed tomography (CT) and other imaging modalities were performed prior to treatment and subsequently every 2–3 treatment cycles to monitor and evaluate therapeutic response. Treatment efficacy was assessed according to the RECIST version 1.1 and categorized as complete remission (CR), partial remission (PR), stable disease (SD), or progressive disease (PD). The objective response rate (ORR) and disease control rate (DCR) were used as short-term efficacy endpoints. ORR was defined as the proportion of patients achieving CR or PR, while DCR was defined as the proportion of patients achieving CR, PR, or SD. PFS and OS were used as the primary indicators for evaluating long-term treatment efficacy. PFS was defined as the time from initiation of chemotherapy combined with immunotherapy for advanced disease to either documented disease progression or death from any cause. OS was defined as the time from initial diagnosis of ICC to death from any cause or the date of last follow-up. The primary endpoint was OS, while secondary endpoints included PFS, DCR, and ORR. Patient disease progression was monitored through regular hospital follow-up visits or telephone interviews, with the follow-up period ending on May 15, 2024. Survival Analysis ROC curve analysis was performed to calculate the AUC for serum lipid levels, including LDL-C, HDL-C, CHOL, TG, ApoB, ApoA-1, and Lpa. Parameters with AUC values greater than 0.7 were retained for subsequent analyses. The dataset was randomly divided into a training cohort and a testing cohort at a 7:3 ratio. The prognostic model was constructed using the training cohort and validated in the testing cohort. ROC curve analysis was conducted using the “pROC” package in R software to determine optimal cut-off values and corresponding AUCs. Serum lipid levels were then dichotomized into high and low groups based on these cut-off points. Clinical characteristics and short-term therapeutic efficacy between groups were compared using Fisher’s exact test. Survival curves for patients with high versus low serum lipid levels were generated using the “survival” package in R. Nomogram Univariate and multivariate Cox regression analyses were performed using the “survival” package in R to identify independent prognostic factors among patients’ clinical characteristics and serum lipid levels. Forest plots were generated using the “ggplot2” package. Independent prognostic factors were incorporated into a nomogram constructed with the “nomogram” package. Calibration curves were established using the “rms” package, and the concordance index (C-index) was calculated for internal validation of the nomogram. The predictive performance of the nomogram-based prognostic model was further assessed using ROC curves and DCA, generated via the “timeROC” and “Dcurves” packages, respectively. External validation was conducted using data from the testing cohort. Statistical Methods Data organization, statistical analyses, tabulation, and visualization were performed using R software version 4.3.2. Categorical variables were presented as counts (percentages). Normally distributed continuous variables were summarized as mean ± standard deviation, while non-normally distributed continuous variables were expressed as median (interquartile range). Comparisons of categorical variables between two groups were conducted using Fisher’s exact test. Ethics The study was conducted in accordance with the Declaration of Helsinki (6th revision, 2008). The study protocol was approved by the Ethics Committee of Cancer Hospital of Harbin Medical University (protocol number KY2023-18, approved on 01/11/2023). Results General Characteristics and Clinical Features A total of 263 patients meeting the inclusion and exclusion criteria were ultimately enrolled in this study (Table 1). All patients received first-line chemotherapy regimens consisting of GC (gemcitabine 1000 mg/m² intravenously on Days 1 and 8; cisplatin 25 mg/m² intravenously on Days 1 and 8; repeated every 3 weeks), GS (gemcitabine 1000 mg/m² intravenously on Days 1 and 8; S-1 administered orally twice daily from Days 1 to 14; repeated every 3 weeks), or GEMOX (gemcitabine 1000 mg/m² intravenously on Days 1 and 8; oxaliplatin 100 mg/m² intravenously on Day 1; repeated every 3 weeks). Upon disease progression, second-line chemotherapy consisted of mFOLFOX (oxaliplatin 85 mg/m² intravenously on Day 1; leucovorin 350 mg/m² intravenously on Day 1; 5-fluorouracil [5-FU] 400 mg/m² intravenous bolus on Day 1, followed by continuous infusion of 1200 mg/(m²·day) for 2 days; repeated every 2 weeks). Immunotherapy agents included camrelizumab (3 mg/kg intravenously) or sintilimab (200 mg intravenously). All patients were randomly divided into a training cohort and a testing cohort at a 7:3 ratio, including 184 patients in the training cohort and 79 patients in the testing cohort. Baseline characteristics between the training and testing cohorts showed no statistically significant differences except for primary tumor size (p = 0.022) (Table S1). Table 1 Clinical characteristics of the 263 patients with advanced ICC Characteristics N=263 Age 59.0 [51.0; 65.0] Gender Female 105 (39.9%) Male 158 (60.1%) Smoking No 234 (89.0%) Yes 29 (11.0%) Drinking No 240 (91.3%) Yes 23 (8.75%) ECOG 0 35 (13.3%) 1 222 (84.4%) 2 6 (2.28%) AJCC III 137 (52.1%) IV 126 (47.9%) Histological grade Well 48 (18.3%) Moderate-well 18 (6.84%) Moderately 68 (25.9%) Moderate-low 29 (11.0%) Poorly 100 (38.0%) Child-Pugh grade A 103 (39.2%) B 160 (60.8%) Tumor size (cm) 6.00 [4.00;8.00] CEA (µg/L) ＞4 125 (47.7%) ≤4 137 (52.3%) CA199 (U/mL) ＞20 100 (38.0%) ≤20 163 (62.0%) Determination of Optimal Cutoff Values for Serum Lipid Levels To elucidate the relationship between serum lipid levels and prognosis, ROC curve analysis was employed to determine the optimal cutoff values for serum lipid parameters—including LDL-C, HDL-C, TC, TG, ApoB, ApoA1, and Lpa—in order to stratify patients in the training and testing cohorts into high- and low-level groups. In the training cohort, the optimal cutoff value for LDL-C was 3.050 mmol/L, with an AUC of 0.87 (95% CI, 0.82–0.92). The sensitivity and specificity were 0.908 and 0.722, respectively. Based on this cutoff, patients were stratified into a high LDL-C group (n = 85) and a low LDL-C group (n = 99) (Figure 2A). For HDL-C, the optimal cutoff was 1.290 mmol/L, with an AUC of 0.78 (95% CI, 0.71–0.85), sensitivity of 0.776, and specificity of 0.657. Patients were divided into high HDL-C (n = 96) and low HDL-C (n = 88) groups accordingly (Figure 2B). The optimal cutoff for ApoA1 was 1.435 g/L, with an AUC of 0.75 (95% CI, 0.68–0.82), sensitivity of 0.645, and specificity of 0.806. Patients were categorized into high ApoA1 (n = 70) and low ApoA1 (n = 114) groups (Figure 2C). Similar results were observed in the validation cohort. The remaining indicators were excluded from further analysis due to AUC values below 0.7 and were therefore not grouped based on optimal cutoff values (Figure S1). Comparison of Clinical Characteristics among Different Levels of LDL-C, HDL-C, and APOA1 ROC analysis identified LDL-C, HDL-C, and ApoA1 as candidates for further stratified analysis. Subsequently, baseline clinical characteristics were compared among patient groups with different expression levels of LDL-C, HDL-C, and ApoA1. No significant differences were observed between the high and low LDL-C groups of advanced ICC patients with respect to sex, age, smoking history, long-term alcohol consumption, ECOG performance status, AJCC stage, histological differentiation, Child–Pugh classification, primary tumor size, carcinoembryonic antigen (CEA) level, and carbohydrate antigen 19-9 (CA19-9) level (Table 2). Table 2 Relationship between different LDL-C and clinical characteristics of patients with advanced ICC Characteristics High Group (n=85) Low Group (n=99) p- value Age 59.0 [53.0;65.0] 57.0 [51.0;65.0] 0.461 Gender 0.423 Female 35 (41.2%) 34 (34.3%) Male 50 (58.8%) 65 (65.7%) Smoking 0.535 No 72 (84.7%) 88 (88.9%) Yes 13 (15.3%) 11 (11.1%) Drinking 1 No 77 (90.6%) 90 (90.9%) Yes 8 (9.41%) 9 (9.09%) ECOG 0.865 0 10 (11.8%) 14 (14.1%) 1 74 (87.1%) 83 (83.8%) 2 1 (1.18%) 2 (2.02%) AJCC 0.75 III 45 (52.9%) 49 (49.5%) IV 40 (47.1%) 50 (50.5%) Histological grade 0.1 Well 13 (15.3%) 19 (19.2%) Moderate-well 9 (10.6%) 3 (3.03%) Moderately 15 (17.6%) 29 (29.3%) Moderate-low 11 (12.9%) 13 (13.1%) Poorly 37 (43.5%) 35 (35.4%) Child-Pugh grade 0.222 A 31 (36.5%) 46 (46.5%) B 54 (63.5%) 53 (53.5%) Tumor size (mm) 7.00 [5.00;8.00] 6.00 [4.00;8.00] 0.094 CEA (µg/L) 0.608 ＞4 44 (52.4%) 47 (47.5%) ≤4 40 (47.6%) 52 (52.5%) CA199 (U/mL) 0.171 ＞20 34 (40.0%) 29 (29.3%) ≤20 51 (60.0%) 70 (70.7%) There were no statistically significant differences between the high and low HDL-C groups of ICC patients in terms of sex, age, smoking history, long-term alcohol consumption, ECOG performance status, AJCC stage, Child–Pugh classification, primary tumor size, CEA level, and CA19-9 level ( p > 0.05). However, a statistically significant difference was observed in histological differentiation between the two groups ( p = 0.028, Table S2). No significant differences were observed between the high and low ApoA1 level groups of ICC patients regarding sex, age, smoking history, long-term alcohol consumption, ECOG performance status, AJCC staging, histological differentiation, Child-Pugh classification, CEA levels, and CA19-9 levels. However, a statistically significant difference was found in primary tumor size ( p = 0.01), with the median tumor size in the low ApoA1 group being larger than that in the high ApoA1 group (Table S3). Relationship between LDL-C, HDL-C, and APOA1 and Short-term Treatment Efficacy To investigate the relationship between LDL-C, HDL-C, and ApoA1 levels and patients’ short-term therapeutic responses, this study analyzed DCR and ORR, with the results as follows: A total of 85 patients exhibited low LDL-C levels (PR = 11, SD = 32, PD = 42), while 72 patients had high LDL-C levels (PR = 4, SD = 30, PD = 38). The DCR were 50.59% and 47.22% for the low- and high-LDL-C groups, respectively ( p = 0.889); the ORR were 12.94% and 5.56%, respectively ( p = 0.182) (Table 3). For HDL-C, 113 patients were classified as low-level (PR = 11, SD = 40, PD = 62) and 44 as high-level (PR = 4, SD = 22, PD = 18). The DCRs were 45.13% and 59.09% for the low- and high-HDL-C groups, respectively ( p = 0.367); the ORRs were 9.73% and 9.09%, respectively ( p = 1) (Table 3). Regarding ApoA1, 97 patients had low levels (PR = 7, SD = 38, PD = 52) and 60 patients had high levels (PR = 8, SD = 24, PD = 28). The DCRs were 46.39% and 53.33% for the low- and high-ApoA1 groups, respectively (p = 0.67); the ORRs were 7.22% and 13.33%, respectively ( p = 0.278) (Table 3). Data from the validation cohort are presented in Table S4. In summary, patients with low LDL-C, high HDL-C, and high ApoA1 levels exhibited better short-term therapeutic responses; however, none of these three indicators showed statistically significant differences in DCR or ORR. Table 3 Correlations of different levels of LDL-C、HDL-C and APOA1 of training cohort with recent efficacy Characteristics PR (n=15) SD (n=62) PD (n=80) DCR p value ORR p value LDL.C 0.889 0.1819 High Group 4 (26.7%) 30 (48.4%) 38 (47.5%) 47.22% 5.56% Low Group 11 (73.3%) 32 (51.6%) 42 (52.5%) 50.59% 12.94% HDL 0.3672 1 High Group 4 (26.7%) 22 (35.5%) 18 (22.5%) 59.09% 9.09% Low Group 11 (73.3%) 40 (64.5%) 62 (77.5%) 45.13% 9.73% APOA1 0.6701 0.2784 High Group 8 (53.3%) 24 (38.7%) 28 (35.0%) 53.33% 13.33% Low Group 7 (46.7%) 38 (61.3%) 52 (65.0%) 46.39% 7.22% Relationship between LDL-C, HDL-C, and ApoA1 and Long-term Therapeutic Outcomes Subsequently, OS was compared among advanced ICC patients stratified by LDL-C, HDL-C, and ApoA1 levels. In the training and validation cohorts, the median OS (mOS) for the low LDL-C groups were 35.3 months and 17 months, respectively, while the high LDL-C groups had mOS of 13.2 months and 15.2 months. In the training cohort, the low LDL-C group exhibited a significantly longer mOS than the high LDL-C group by 22.1 months ( p < 0.001, HR = 3.968, 95% CI 2.698–5.835) (Figure 3A). For HDL-C, in both the training and validation sets, the mOS in the low HDL-C groups were 14.9 months and 15.7 months, and in the high HDL-C groups were 32.4 months and 24.9 months, respectively. The training cohort’s low HDL-C group had a shorter mOS than the high HDL-C group by 17.5 months ( p < 0.001, HR = 0.364, 95% CI 0.248–0.533) (Figure 3B). Regarding ApoA1, in both the training and validation sets, the mOS for the low ApoA1 groups were 17.5 months and 16.1 months, and for the high ApoA1 groups were 12.88 months and 26.2 months, respectively. The low ApoA1 group in the training cohort showed a longer mOS than the high ApoA1 group by 4.62 months ( p < 0.001, HR = 0.463, 95% CI 0.311–0.689) (Figure 3C). Additionally, PFS was compared between patients with different levels of LDL-C, HDL-C, and ApoA1 in both the training and validation cohorts. In the training cohort, the mPFS was 8.8 months in the low LDL-C group versus 6.5 months in the high LDL-C group, with the low LDL-C group exhibiting a significantly longer mPFS by 2.3 months ( p = 0.027, HR = 1.549, 95% CI 1.021–2.348). In the validation cohort, the low LDL-C group had a 5-month longer mPFS compared to the high LDL-C group ( p = 0.029, HR = 2.04, 95% CI 1.01–4.14) (Figure 4A). In the training cohort, the mPFS was 6.1 months for the low HDL-C group and 9.17 months for the high HDL-C group, with the low HDL-C group showing a significantly shorter mPFS by 3.07 months ( p = 0.025, HR = 0.642, 95% CI 0.426–0.967). Similarly, in the validation cohort, the low HDL-C group had a 4.7-month shorter mPFS than the high HDL-C group ( p = 0.038, HR = 0.496, 95% CI 0.23–1.072) (Figure 4B). For ApoA1, the training cohort showed an mPFS of 7.83 months in the low ApoA1 group and 7.33 months in the high ApoA1 group, with no significant difference ( p = 0.64, HR = 0.906, 95% CI 0.601–1.367). However, in the validation cohort, the low ApoA1 group had a significantly shorter mPFS by 6.8 months compared to the high ApoA1 group ( p = 0.004, HR = 0.395, 95% CI 0.183–0.851) (Figure 4C). Independent prognostic factors affecting outcomes in advanced intrahepatic cholangiocarcinoma patients To further investigate independent prognostic factors, Cox proportional hazards regression models were employed. General clinical characteristics, along with varying levels of LDL-C, HDL-C, and ApoA1, were included in both univariate and multivariate analyses based on OS and PFS. Univariate Cox regression analysis based on the training cohort revealed that low levels of HDL-C, low levels of ApoA1, and larger primary tumor size were significantly associated with poor OS in ICC patients, whereas low LDL-C levels were correlated with favorable prognosis ( p < 0.05) (Figure 5A). Similarly, univariate Cox analysis in the validation cohort demonstrated that low HDL-C and low ApoA1 levels were adverse prognostic factors, while low LDL-C levels and well to moderately differentiated histology were associated with improved prognosis in ICC patients ( p < 0.05) (Figure 5B). Multivariate Cox regression analyses in both the training and validation cohorts demonstrated that low levels of LDL-C and HDL-C were independent prognostic factors for ICC patients ( p < 0.05). Specifically, low LDL-C was identified as an independent protective factor, whereas low HDL-C was an independent risk factor for poor prognosis in ICC patients (Figure 5C–D). Subsequently, Cox regression analysis was performed to evaluate the impact of LDL-C, HDL-C, and ApoA1 on PFS in ICC patients. Univariate Cox analysis in the training cohort demonstrated that well-differentiated tumors and low LDL-C levels were favorable prognostic factors for PFS, whereas low HDL-C was also identified as a favorable prognostic factor for PFS ( p < 0.05). Low ApoA1 was associated with poorer PFS, but this finding did not reach statistical significance (Figure S2A). In the validation cohort, the trends for all three markers were consistent with those observed in the training cohort; however, only the association with ApoA1 reached statistical significance (Figure S2B). Multivariate Cox regression analysis based on the training cohort indicated that low LDL-C was a protective factor for PFS in ICC patients, whereas low HDL-C was a risk factor; however, these results did not reach statistical significance (Figure S2C). In the validation cohort, multivariate Cox regression identified low ApoA1 as an adverse prognostic factor for PFS in ICC patients ( p < 0.05) (Figure S2D). Construction and Validation of the Nomogram Prediction Model Given the stronger association between the aforementioned serum molecules and OS in ICC patients, we focused exclusively on developing a predictive model for OS to forecast the prognosis of advanced ICC patients receiving chemotherapy combined with immunotherapy. Based on the results of the multivariate Cox regression analysis, LDL-C and HDL-C were selected as variables to construct a nomogram for predicting 1-, 2-, and 3-year survival probabilities. Patients with low HDL-C and high LDL-C scores exhibited higher risk scores, indicating poorer survival rates at 1, 2, and 3 years (Figure 6A). Subsequently, the nomogram prediction model was validated using calibration curves. Internal validation demonstrated that the C-index for predicting the prognosis of advanced ICC patients receiving immunotherapy was 0.68. The calibration curves for 1-, 2-, and 3-year survival closely approximated the ideal diagonal line (Figure 6B). DCA was performed to evaluate the clinical net benefit, revealing that the DCA models at 1-, 2-, and 3 years remained within the optimal range across certain threshold probabilities, indicating that the prediction model provides meaningful clinical net benefit and has practical clinical utility (Figure 6C). Time-dependent ROC analysis yielded AUCs of 0.716, 0.761, and 0.810 at 1-, 2-, and 3 years, respectively, further supporting the model’s predictive capability (Figure 6D). Moreover, the Hosmer-Leme show goodness-of-fit test resulted in a p -value of 0.1083 (Table S5), indicating no significant systematic bias and an acceptable model fit. To further validate the clinical utility of the prognostic model, external validation was performed using the testing cohort. Calibration of the model in the testing cohort demonstrated that the calibration curve closely approximated the ideal reference line (Figure 6E). The model’s discriminative ability for external data was assessed by ROC analysis, yielding an AUC of 0.844 with a 95% CI of 0.757–0.931 (Figure 6F). These results indicate that the model possesses robust predictive performance in external datasets. Discussion Various biomarkers, including gene mutation burden and tumor load, have been widely applied for prognostic assessment in cancer patients. However, due to high costs and challenges in sample collection, there remains a need to identify biomarkers that are more easily detectable while maintaining high specificity and sensitivity. Lipids play crucial roles in energy storage, metabolism, and signal transduction involved in cellular activities [22] . Studies have demonstrated that cancer cells within the tumor microenvironment require abundant nutrients to sustain tumor growth. Energy generated solely through glycolysis is insufficient to meet these demands; therefore, lipid metabolism is utilized to support rapid proliferation, survival, migration, invasion, and metastasis of tumor cells [23] . Liquid biopsy-based assessment of peripheral blood lipid profiles has emerged as a promising approach and has been applied in malignancies such as lung adenocarcinoma and breast cancer. Related investigations have also been conducted in intrahepatic cholangiocarcinoma [14,18,24,25] . This study utilized ROC curve analysis to select LDL-C, HDL-C, and ApoA1 as prognostic factors for patients with advanced ICC undergoing chemotherapy combined with immunotherapy. Survival analyses based on the training and validation cohorts demonstrated that patients with low serum LDL-C, high HDL-C, and elevated ApoA1 levels had significantly longer mOS. Regarding PFS, the training cohort showed that patients with low LDL-C and high HDL-C exhibited prolonged mPFS. Although the validation cohort results did not reach statistical significance, the overall trends for these two markers were consistent with the training cohort, likely due to the smaller sample size of the validation cohort. Consistent with our findings, Shu et al. reported that postoperative ICC patients with high serum HDL-C derived greater clinical benefit [26] . Additionally, Lin et al. found that cervical cancer patients exhibited higher LDL-C and lower HDL-C levels compared to healthy controls, and that elevated LDL-C and decreased HDL-C were adverse prognostic factors in this population [27] . These findings align with our results, suggesting that increased cancer risk and poor prognosis are associated with low HDL-C and high LDL-C levels. Conversely, another study observed that head and neck squamous cell carcinoma patients with high LDL-C and low ApoA1 levels experienced better PFS [28] . This discrepancy may reflect metabolic heterogeneity among different tumor types. Moreover, multiple studies have reported that elevated serum TG levels are associated with poor prognosis in lung cancer, thyroid cancer, rectal cancer, breast cancer, and prostate cancer [29–31] . However, in the present study, TG demonstrated an AUC of less than 0.7 in the ROC analysis, indicating suboptimal predictive performance. Therefore, TG was not included in further analyses. Future studies with larger sample sizes are warranted to clarify the potential association between TG and ICC. Subsequent Cox regression analyses based on the training and validation cohorts revealed that low serum LDL-C was a protective factor for OS in advanced ICC patients receiving chemotherapy combined with immunotherapy, whereas low HDL-C and ApoA1 were risk factors for OS. Multivariate Cox regression further identified low LDL-C and low HDL-C as independent protective and risk factors for OS, respectively. Similarly, Shu et al. reported that low HDL-C was an independent risk factor for postoperative ICC patients [22] , supporting the role of low HDL-C as an independent prognostic indicator in ICC. Additionally, Chen et al. identified LDL-C as an independent prognostic factor in non-esophageal squamous cell carcinoma, further confirming the prognostic value of LDL-C in cancer [32] . In another study on cervical cancer, patients exhibited elevated LDL-C levels and decreased HDL-C levels compared to healthy controls [27] . Conversely, a study on head and neck squamous cell carcinoma found that high LDL-C and low ApoA1 were protective factors [28] . These findings suggest heterogeneity in lipid metabolism among different cancer types, with the same serum lipid markers having distinct prognostic implications depending on the tumor context. Subsequently, LDL-C and HDL-C identified as predictive factors through Cox regression analysis were used to construct a nomogram for predicting 1-, 2-, and 3-year survival rates in advanced ICC patients receiving chemotherapy combined with immunotherapy. The nomogram was internally and externally validated using the training and validation cohorts, respectively, demonstrating robust predictive performance and clinical utility. These findings are consistent with previous studies in thyroid cancer, which also suggest that HDL-C can serve as a prognostic biomarker for cancer [20] . In summary, LDL-C and HDL-C may serve as reliable biomarkers for predicting prognosis in patients with advanced ICC. However, this study has several limitations. First, as a single-center study with a relatively small sample size due to the rarity of cholangiocarcinoma, the conclusions may not fully represent the broader population of advanced ICC patients in China. Future multicenter studies with larger cohorts are needed to validate these findings. Second, serum lipid levels can be influenced by factors such as patient age, sex, hormonal status, and underlying comorbidities; therefore, stricter study controls are required to minimize confounding effects. Third, this study did not investigate the metabolic reprogramming mechanisms underlying the observed alterations in serum lipid profiles, and the causes of lipid level changes remain unclear. Further research will be conducted to explore these underlying mechanisms. Conclusion In summary, we successfully constructed and validated a prognostic model to predict the OS of patients with advanced ICC, which provides a more accurate basis for the immunotherapy decision of such patients. The strategy of chemotherapy in combination with immunotherapy dominates in advanced ICC patients,and it is suggested that immunotherapy should be incorporated into clinical treatment protocols more frequently. Declarations Ethical considerations The study was conducted in accordance with the Declaration of Helsinki (6th revision, 2008). The study protocol was approved by the Ethics Committee of Cancer Hospital of Harbin Medical University (protocol number KY2023-18, approved on 01/11/2023). Consent to participate The ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants\’ legal guardians/next of kin because of the retrospective nature of the study. Consent for publication Not applicable. Conflict of interest The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construced as a potential competing of interest. Funding statement The author(s) received no financial support for the research, authorship, and/or publication of this article. Data availability The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author. Author contributions CZJ: Data curation, Methodology, Software, Visualization, Writing-original draft. KJL: Data curation, Methodology, Software, Validation, Writing-original draft. TYF: Data curation, Investigation, Writing-original draft. WL: Data curation, Investigation, Writing-original draft. LY: Data curation, Investigation, Writing-original draft. PCC: Data curation, Investigation, Writing-original draft. JT: Conceptualization, Project administration, Supervision, Writing-review & editing. KGL: Conceptualization, Project administration, Supervision, Writing-review & editing. Acknowledgements Not applicable. References Qurashi M, Vithayathil M, Khan S. Epidemiology of cholangiocarcinoma[J]. EJSO, 2025, 51(2). Piha-Paul S A, Oh D Y, Ueno M, Malka D, Chung H C, Nagrial A, Kelley R K, Ros W, Italiano A, Nakagawa K, Rugo H S, De Braud F, Varga A I, Hansen A, Wang H, Krishnan S, Norwood K G, Doi T. Efficacy and safety of pembrolizumab for the treatment of advanced biliary cancer: Results from the KEYNOTE-158 and KEYNOTE-028 studies.[J]. Int J Cancer, 2020, 147(8): 2190-2198. Kelley R, Bridgewater J, Gores G, Zhu A. Systemic therapies for intrahepatic cholangiocarcinoma[J]. J. Hepatol., 2020, 72(2): 353-363. Rui R, Zhou L, He S. Cancer immunotherapies: advances and bottlenecks.[J]. Front. Immunol., 2023, 14: 1212476. Raggi C, Taddei M L, Rae C, Braconi C, Marra F. Metabolic reprogramming in cholangiocarcinoma[J]. J Hepatol, 2022, 77(3): 849-864. Maurya S K, Chaudhri S, Kumar S, Gupta S. Repurposing of Metabolic Drugs Metformin and Simvastatin as an Emerging Class of Cancer Therapeutics.[J]. Pharm Res, 2025, 42(1): 49-67. Ruze R, Liu T, Zou X, Song J, Chen Y, Xu R, Yin X, Xu Q. Obesity and type 2 diabetes mellitus: connections in epidemiology, pathogenesis, and treatments.[J]. Front. Endocrinol., 2023, 14: 1161521. Tang H Y, Chen A Q, Zhang H, Gao X F, Kong X Q, Zhang J J. Vascular Smooth Muscle Cells Phenotypic Switching in Cardiovascular Diseases.[J]. Cells, 2022, 11(24). Yang J, Shay C, Saba N F, Teng Y. Cancer metabolism and carcinogenesis.[J]. Exp Hematol Oncol, 2024, 13(1): 10. Sun Y, Cao D, Zhang Y, Wu Y, Jia Z, Cui Y, Li D, Cao X, Jiang J. Appraising associations between signature lipidomic biomarkers and digestive system cancer risk: novel evidences from a prospective cohort study of UK Biobank and Mendelian randomization analyses.[J]. Lipids Health Dis, 2024, 23(1): 61. Wang S, Huang X, Zhao S, Lv J, Li Y, Wang S, Guo J, Wang Y, Wang R, Zhang M, Qiu W. Progressions of the correlation between lipid metabolism and immune infiltration characteristics in gastric cancer and identification of BCHE as a potential biomarker.[J]. Front Immunol, 2024, 15: 1327565. Duan Y, Deng M, Liu B, Meng X, Liao J, Qiu Y, Wu Z, Lin J, Dong Y, Duan Y, Sun Y. Mitochondria targeted drug delivery system overcoming drug resistance in intrahepatic cholangiocarcinoma by reprogramming lipid metabolism.[J]. Biomaterials, 2024, 309: 122609. Qian L, Wang G, Li B, Su H, Qin L. Regulation of lipid metabolism by APOE4 in intrahepatic cholangiocarcinoma via the enhancement of ABCA1 membrane expression[J]. PeerJ, 2024, 12: e16740. Lori G, Pastore M, Navari N, Piombanti B, Booijink R, Rovida E, Tusa I, Lewinska M, Andersen J B, Lottini T, Arcangeli A, Taddei M L, Pranzini E, Mancini C, Anceschi C, Madiai S, Sacco E, Rota S, Trapani A, Agrimi G, Ramazzotti M, Ostano P, Peraldo Neia C, Parri M, Carli F, Sabatini S, Gastaldelli A, Marra F, Raggi C. Altered fatty acid metabolism rewires cholangiocarcinoma stemness features[J]. JHEP Rep. : Innov. Hepatol., 2024, 6(10): 101182. Chen Y, Zhou Y, Ren R, Chen Y, Lei J, Li Y. Harnessing lipid metabolism modulation for improved immunotherapy outcomes in lung adenocarcinoma.[J]. J. ImmunoTher. Cancer, 2024, 12(7). Ruiz de Gauna M, Biancaniello F, González-Romero F, Rodrigues P M, Lapitz A, Gómez-Santos B, Olaizola P, Di Matteo S, Aurrekoetxea I, Labiano I, Nieva-Zuluaga A, Benito-Vicente A, Perugorria M J, Apodaka-Biguri M, Paiva N A, Sáenz de Urturi D, Buqué X, Delgado I, Martín C, Azkargorta M, Elortza F, Calvisi D F, Andersen J B, Alvaro D, Cardinale V, Bujanda L, Banales J M, Aspichueta P. Cholangiocarcinoma progression depends on the uptake and metabolization of extracellular lipids.[J]. Hepatol. (baltim. Md,), 2022, 76(6): 1617-1633. Liu X, Hartman C L, Li L, Albert C J, Si F, Gao A, Huang L, Zhao Y, Lin W, Hsueh E C, Shen L, Shao Q, Hoft D F, Ford D A, Peng G. Reprogramming lipid metabolism prevents effector T cell senescence and enhances tumor immunotherapy.[J]. Sci. Transl. Med., 2021, 13(587). Ma F, Liu X, Zhang Y, Tao Y, Zhao L, Abusalamah H, Huffman C, Harbison R A, Puram S V, Wang Y, Peng G. Tumor extracellular vesicle-derived PD-L1 promotes T cell senescence through lipid metabolism reprogramming[J]. Sci. Transl. Med., 2025, 17(785). Yang K, Wang X, Song C, He Z, Wang R, Xu Y, Jiang G, Wan Y, Mei J, Mao W. The role of lipid metabolic reprogramming in tumor microenvironment.[J]. Theranostics, 2023, 13(6): 1774-1808. Yuan J, Chen Z, Zhang J, Chen X, Chen S, Wang X, Zhang J, Su X, Zhu W, Sheng J, Teng L, Wang W. Preoperative Serum Lipids as Novel Predictors of Survival in 3575 Patients With Papillary Thyroid Cancer[J]. J. Clin. Endocrinol. Metab., 2024, 110(3): 668-676. Salgado-Montilla J, Soto Salgado M, Surillo Trautmann B, Sánchez-Ortiz R, Irizarry-Ramírez M. Association of serum lipid levels and prostate cancer severity among Hispanic Puerto Rican men.[J]. Lipids Health Dis., 2015, 14: 111. Yu S M, Chang X J, Gu Y Y, Jia X D, Gao X D, Huang J G, Dong J H, Zeng Z. Serum high-density lipoprotein cholesterol levels predict early recurrence and prognosis of intrahepatic cholangiocarcinoma after surgical resection[J]. Heliyon, 2024, 10(11): e32113. Li X, Jia Y, Li Y, Hei H, Zhang S, Qin J. Crosstalk between metabolic reprogramming and microbiota: implications for cancer progression and novel therapeutic opportunities[J]. FRONTIERS IN IMMUNOLOGY, 2025, 16. Bian X, Liu R, Meng Y, Xing D, Xu D, Lu Z. Lipid metabolism and cancer[J]. J. Exp. Med., 2021, 218(1). Yu X, Tong H, Chen J, Tang C, Wang S, Si Y, Wang S, Tang Z. CircRNA MBOAT2 promotes intrahepatic cholangiocarcinoma progression and lipid metabolism reprogramming by stabilizing PTBP1 to facilitate FASN mRNA cytoplasmic export[J]. Cell Death Dis, 2023, 14(1): 20. Zhu H, Hu H, Hao B, Zhan W, Yan T, Zhang J, Wang S, Hu H, Zhang T. Insights into a machine learning-based palmitoylation-related gene model for predicting the prognosis and treatment response of breast cancer patients[J]. Technol Cancer Res Treat, 2024, 23: 15330338241263434. Cheng L, Li Z, Zheng Q, Yao Q. Correlation study of serum lipid levels and lipid metabolism-related genes in cervical cancer[J]. FRONTIERS IN ONCOLOGY, 2024, 14. Wang S, Wang L, Li H, Zhang J, Peng J, Cheng B, Song M, Hu Q. Correlation analysis of plasma lipid profiles and the prognosis of head and neck squamous cell carcinoma[J]. Oral Dis., 2024, 30(2): 329-341. Li C, Wang F, Cui L, Li S, Zhao J, Liao L. Association between abnormal lipid metabolism and tumor.[J]. Front. Endocrinol., 2023, 14: 1134154. Chen J Y, Chi N H, Lee H S, Hsiung C N, Wu C W, Fan K C, Lee M R, Wang J Y, Ho C C, Shih J Y. Lipid Levels and Lung Cancer Risk: Findings from the Taiwan National Data Systems from 2012 to 2018.[J]. J Epidemiol Glob Health, 2025, 15(1): 11. Zhang W, Li Z, Huang Y, Zhao J, Guo S, Wang Q, Guo S, Li Q. Complex Role of Circulating Triglycerides in Breast Cancer Onset and Survival: Insights From Two-Sample Mendelian Randomization Study.[J]. Cancer Med., 2025, 14(4): e70698. Chen S, Li X, Wen X, Peng S, Xue N, Xing S, Liu Y. Prognostic nomogram integrated baseline serum lipids for patients with non-esophageal squamous cell carcinoma[J]. Ann. Transl. Med., 2019, 7(20): 548. Additional Declarations No competing interests reported. Supplementary Files TableS.docx FigureS1.tif FigureS2.tif FigureS3.tif Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7803902","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":543474021,"identity":"b39350c1-ed37-4f31-ab7e-3004afcb0988","order_by":0,"name":"Chengzhi Jiang","email":"","orcid":"","institution":"Third Affiliated Hospital of Harbin Medical University","correspondingAuthor":false,"prefix":"","firstName":"Chengzhi","middleName":"","lastName":"Jiang","suffix":""},{"id":543474022,"identity":"0d0953c5-cdbf-4dd8-ba24-956d37641250","order_by":1,"name":"Kaijun Long","email":"","orcid":"","institution":"The Second Affiliated Hospital of Zunyi Medical Unibersity","correspondingAuthor":false,"prefix":"","firstName":"Kaijun","middleName":"","lastName":"Long","suffix":""},{"id":543474023,"identity":"0cee61e0-9459-45dd-a07c-4dc7a39256c2","order_by":2,"name":"Tianyuan Fang","email":"","orcid":"","institution":"Third Affiliated Hospital of Harbin Medical University","correspondingAuthor":false,"prefix":"","firstName":"Tianyuan","middleName":"","lastName":"Fang","suffix":""},{"id":543474024,"identity":"8d883d51-3997-4224-9cd7-f555f1ccaf8b","order_by":3,"name":"Wen Li","email":"","orcid":"","institution":"Third Affiliated Hospital of Harbin Medical University","correspondingAuthor":false,"prefix":"","firstName":"Wen","middleName":"","lastName":"Li","suffix":""},{"id":543474025,"identity":"8d134dd1-64cc-4405-a082-46fed6b46fb7","order_by":4,"name":"Liu Yang","email":"","orcid":"","institution":"Third Affiliated Hospital of Harbin Medical University","correspondingAuthor":false,"prefix":"","firstName":"Liu","middleName":"","lastName":"Yang","suffix":""},{"id":543474026,"identity":"93e49325-d7aa-40a9-a339-957bcd42e6c4","order_by":5,"name":"Pengcheng Chai","email":"","orcid":"","institution":"Third Affiliated Hospital of Harbin Medical University","correspondingAuthor":false,"prefix":"","firstName":"Pengcheng","middleName":"","lastName":"Chai","suffix":""},{"id":543474027,"identity":"157a8855-9774-46ad-be73-6a9b0c8266c1","order_by":6,"name":"Ji Tao","email":"","orcid":"","institution":"Third Affiliated Hospital of Harbin Medical University","correspondingAuthor":false,"prefix":"","firstName":"Ji","middleName":"","lastName":"Tao","suffix":""},{"id":543474028,"identity":"63e7f0ec-92d2-4cc6-8a83-7566f5a8f920","order_by":7,"name":"Kaiguo Long","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAAu0lEQVRIiWNgGAWjYBACA4nk5x8+8NTIsbG3HyBWyzEzxhkyx4z5eM4kEKlFsoeNmcOGOXGehIMBkVqkedgeM+SwpbdJMCQw/KjYRoQWOR5244IzMrlt0o0HGHvO3CZKS4H0zB623DaZAwnMjG3EaJHmMZDm/cecziaRYECkFskeM2keHuYEErRIHDM2nMFzzLANGMgHifKL/Yzkhw+AUSkv395+8MGPCiK0oIADJKofBaNgFIyCUYALAADsNDepQvUg9gAAAABJRU5ErkJggg==","orcid":"","institution":"The Second Affiliated Hospital of Zunyi Medical Unibersity","correspondingAuthor":true,"prefix":"","firstName":"Kaiguo","middleName":"","lastName":"Long","suffix":""}],"badges":[],"createdAt":"2025-10-08 04:23:31","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7803902/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7803902/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":95807295,"identity":"4a9dbacb-2417-4dac-a419-224ff7dae71e","added_by":"auto","created_at":"2025-11-13 08:48:17","extension":"jpg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":166388,"visible":true,"origin":"","legend":"\u003cp\u003eWork flew of this study\u003c/p\u003e","description":"","filename":"1.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7803902/v1/f84efd87167f7f1dd9eb9f32.jpg"},{"id":95807289,"identity":"673248ee-5977-4ba2-8a02-165aa060fe41","added_by":"auto","created_at":"2025-11-13 08:48:16","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":49139,"visible":true,"origin":"","legend":"\u003cp\u003eThe ROC curve of LDL-C、HDL-C and APOA1. A The ROC curve of LDL-C. B The ROC curve of HDL-C. C The ROC curve of APOA1.\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-7803902/v1/38be415d8c6fd59cbbcc52af.png"},{"id":95807161,"identity":"31edcd4c-925f-4549-83f1-7c8b3ce91513","added_by":"auto","created_at":"2025-11-13 08:48:09","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":68444,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of Kaplan-Meier curves of OSbetween different levels of LDL-C, HDL-C and APOA1 for training cohort and testing cohort. A Kaplan-Meier curves of OS with LDL-C. B Kaplan-Meier curves of OS with HDL-C. C Kaplan-Meier curves of OS with APOA1. Notes: \u003cem\u003ep \u003c/em\u003e\u0026lt; 0.05 is statistically significant.\u003c/p\u003e","description":"","filename":"3.png","url":"https://assets-eu.researchsquare.com/files/rs-7803902/v1/1ccef3299964e4a7b2032124.png"},{"id":95807078,"identity":"446c8ca5-2b78-4b75-af47-df2061cf6ef5","added_by":"auto","created_at":"2025-11-13 08:48:05","extension":"png","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":75695,"visible":true,"origin":"","legend":"\u003cp\u003eComparison of Kaplan-Meier curves of PFS between different levels of LDL-C, HDL-C and APOA1 for training cohort and testing cohort. A Kaplan-Meier curves of PFS with LDL-C. B Kaplan-Meier curves of PFS with HDL-C. C Kaplan-Meier curves of PFS with APOA1. Notes: \u003cem\u003ep \u003c/em\u003e\u0026lt; 0.05 is statistically significant.\u003c/p\u003e","description":"","filename":"4.png","url":"https://assets-eu.researchsquare.com/files/rs-7803902/v1/5834daf551b812dbfbb20537.png"},{"id":95807309,"identity":"7d974a51-7c18-42fa-8411-fc4fc9663b58","added_by":"auto","created_at":"2025-11-13 08:48:19","extension":"jpg","order_by":5,"title":"Figure 5","display":"","copyAsset":false,"role":"figure","size":253483,"visible":true,"origin":"","legend":"\u003cp\u003eCOX regression analysis of OS with advanced ICC for training cohort and testing cohort A Univariate analysis of OS for training cohort. B Univariate analysis of OS for testing cohort. CMultivariate analysis of OS for training cohort. D Multivariate analysis of OS for testing cohort.\u003c/p\u003e","description":"","filename":"5.jpg","url":"https://assets-eu.researchsquare.com/files/rs-7803902/v1/1d8b9dd64d830fcecab61924.jpg"},{"id":95807193,"identity":"03afc03d-b1ff-463f-b87a-402c54440732","added_by":"auto","created_at":"2025-11-13 08:48:11","extension":"png","order_by":6,"title":"Figure 6","display":"","copyAsset":false,"role":"figure","size":79436,"visible":true,"origin":"","legend":"\u003cp\u003eNomogram construction and verification. A The nomogram model of OS. B Calibration of the model. C DCA of the model at 1-year, 2-year and 3-year. D A time-ROC curve of the model. E A ROC curve of testing cohort. F Calibration of the model by testing cohort.\u003c/p\u003e","description":"","filename":"6.png","url":"https://assets-eu.researchsquare.com/files/rs-7803902/v1/0e6dde3ddaac4db2da43017d.png"},{"id":99788303,"identity":"5f952e99-b5f8-412a-a77c-76a5b8536664","added_by":"auto","created_at":"2026-01-08 12:46:06","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":1680526,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7803902/v1/82354f78-29a9-48eb-85f7-f258cba8a073.pdf"},{"id":95807293,"identity":"6c490bb1-f319-440b-99e7-d324de92056a","added_by":"auto","created_at":"2025-11-13 08:48:17","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":31451,"visible":true,"origin":"","legend":"","description":"","filename":"TableS.docx","url":"https://assets-eu.researchsquare.com/files/rs-7803902/v1/27be8e1ba6af6a339536048b.docx"},{"id":95807291,"identity":"aed5d5f7-c6cb-4e7a-8777-c5d43bdf8ed5","added_by":"auto","created_at":"2025-11-13 08:48:16","extension":"tif","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":35407888,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS1.tif","url":"https://assets-eu.researchsquare.com/files/rs-7803902/v1/24c9d50b1279982bdc804f56.tif"},{"id":95807068,"identity":"b6c11b08-0d87-4774-b9ee-7a0d8c2e3dfd","added_by":"auto","created_at":"2025-11-13 08:48:05","extension":"tif","order_by":3,"title":"","display":"","copyAsset":false,"role":"supplement","size":35434420,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS2.tif","url":"https://assets-eu.researchsquare.com/files/rs-7803902/v1/9537da5f63655bac7ba3948e.tif"},{"id":95807086,"identity":"bd9c771b-9ac5-43eb-a4ec-3d54ae255ca9","added_by":"auto","created_at":"2025-11-13 08:48:05","extension":"tif","order_by":4,"title":"","display":"","copyAsset":false,"role":"supplement","size":19922352,"visible":true,"origin":"","legend":"","description":"","filename":"FigureS3.tif","url":"https://assets-eu.researchsquare.com/files/rs-7803902/v1/aa1ea4f43745e16249ee4a30.tif"}],"financialInterests":"No competing interests reported.","formattedTitle":"Prognostic Prediction of Advanced Intrahepatic Cholangiocarcinoma Patients Receiving Chemotherapy Combined with Immunotherapy Based on Serum Lipid Profiles","fulltext":[{"header":"Background","content":"\u003cp\u003eIntrahepatic cholangiocarcinoma (ICC) is the second most common primary liver malignancy after hepatocellular carcinoma. Due to its insidious onset and lack of specific early symptoms, ICC is often diagnosed at an advanced stage. Although various therapeutic strategies—including chemotherapy, targeted therapy, and chemotherapy combined with immunotherapy—are currently available for advanced ICC, the overall clinical outcomes remain unsatisfactory, with a 5-year survival rate of less than 20% in most patients\u003csup\u003e[1]\u003c/sup\u003e. Notably, patients receiving chemoimmunotherapy exhibit longer overall survival (OS) compared to those receiving chemotherapy alone; however, the overall survival rates remain suboptimal, indicating substantial challenges in the immunotherapeutic management of advanced ICC\u003csup\u003e[2,3]\u003c/sup\u003e. Owing to factors such as low expression of immune checkpoints and intrinsic resistance of tumor cells, immune checkpoint inhibitors (ICIs) demonstrate limited efficacy in a subset of patients. Therefore, the identification of reliable biomarkers to stratify patients who are more likely to benefit from ICIs is essential for optimizing treatment strategies and guiding future therapeutic development\u003csup\u003e[4]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eLipid metabolic reprogramming plays a pivotal role in various biological processes of tumor cells, including proliferation, survival, migration, invasion, and metastasis\u003csup\u003e[5–12]\u003c/sup\u003e. Compared to monolayer-cultured ICC cells, ICC stem-like cells exhibit enhanced de novo fatty acid synthesis activity and higher expression levels of key enzymes involved in fatty acid biosynthesis, such as fatty acid synthase (FASN). Clinically, ICC patients with high FASN expression are associated with poorer long-term survival outcomes\u003csup\u003e[13]\u003c/sup\u003e. Correspondingly, inhibition of the mTOR signaling pathway can suppress FASN-mediated fatty acid synthesis, thereby reducing fatty acid oxidation in tumor-associated macrophages and promoting antitumor immune responses\u003csup\u003e[14]\u003c/sup\u003e. Additionally, other studies have demonstrated that highly proliferative cholangiocarcinoma cells exhibit significantly higher lipid uptake compared to normal cholangiocytes. Treatment with fatty acid oxidation (FAO) inhibitors led to suppressed proliferation in several cancer cell types, including cholangiocarcinoma, suggesting that blockade of fatty acid catabolism may effectively inhibit tumor growth\u003csup\u003e[15]\u003c/sup\u003e. Moreover, lipid metabolic reprogramming is closely associated with the efficacy of tumor immunotherapy. For example, cPLA2α activity driven by tumor cells and regulatory T cells (Tregs) can induce lipid droplet accumulation in effector T cells, resulting in T cell senescence. Inhibition of cPLA2α has been shown to prevent effector T cell exhaustion and enhance the efficacy of immunotherapy\u003csup\u003e[16]\u003c/sup\u003e. Similarly, PD-L1-containing tumor-derived extracellular vesicles can also promote lipid droplet accumulation in T cells, leading to senescence and resistance to immunotherapy\u003csup\u003e[17]\u003c/sup\u003e. Thus, lipid metabolic reprogramming may exert critical regulatory effects on tumor immunotherapy by modulating the immune microenvironment\u003csup\u003e[18]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eWith the widespread application of metabolomics, an increasing number of studies have revealed significant associations between alterations in serum lipid levels and the prognosis of various malignancies\u003csup\u003e[19]\u003c/sup\u003e. As common clinical biochemical indicators, serum lipids are essential components involved in energy storage, metabolism, and cellular signal transduction. Changes in serum lipid levels can indirectly reflect lipid alterations within the tumor microenvironment, thereby representing potential biomarkers for predicting the efficacy of immunotherapy. A retrospective study reported that low levels of triglycerides and high-density lipoprotein cholesterol (HDL-C) were strongly associated with recurrence in patients with thyroid cancer\u003csup\u003e[20]\u003c/sup\u003e. Earlier research also suggested that serum triglyceride and HDL-C levels were correlated with prostate cancer severity\u003csup\u003e[21]\u003c/sup\u003e. Sun et al. identified a causal association between elevated LDL-C and gastric cancer. Furthermore, triglyceride levels ≥2.2 mmol/L were found to increase the risk of gallbladder cancer in men over the age of 60\u003csup\u003e[9]\u003c/sup\u003e. However, studies exploring the prognostic value and predictive significance of serum lipid levels in relation to immunotherapeutic outcomes in ICC remain scarce.\u003c/p\u003e\n\u003cp\u003eThis study analyzed the association between pre-treatment serum lipid levels and prognosis in patients with advanced ICC undergoing chemotherapy combined with immunotherapy. Independent prognostic risk factors were identified, and a nomogram-based prognostic model was constructed to provide a potential reference for clinical decision-making and prognostic assessment (Figure 1).\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cstrong\u003eStudy Population\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis study enrolled 263 patients diagnosed with unresectable advanced ICC at the Harbin Medical University Cancer Hospital between January 2014 and January 2024. Inclusion criteria were as follows: (1) Histopathological confirmation of ICC; (2) Presence of measurable lesions according to the Response Evaluation Criteria in Solid Tumors version 1.1 (RECIST 1.1); (3) Effective radiological assessments performed prior to treatment and after every 2–3 treatment cycles; (4) Treatment regimen consisting exclusively of chemotherapy combined with immune checkpoint inhibitors (ICIs); (5) Eastern Cooperative Oncology Group (ECOG) performance status of 0 to 2; and (6) Patient age between 18 and 80 years.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eExclusion criteria were as follows:\u0026nbsp;\u003c/strong\u003e(1) Incomplete clinical data; (2) Absence of measurable lesions; (3) History of other histologically confirmed malignancies within the past five years; (4) Presence of severe organ dysfunction; (5) Diagnosis of autoimmune deficiency disorders; (6) Use of medications known to affect serum lipid levels during the treatment period, such as statins or fibrates; (7) Use of medications known to affect serum lipid levels during the treatment period, such as statins or fibrates; (8) History of prior locoregional liver therapies, including surgery, interventional procedures (e.g., ablation, embolization, or brachytherapy); (9) Patients who declined follow-up.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eClinical Data Collection\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eClinical records were retrospectively reviewed to collect baseline clinical data from patients who met the inclusion and exclusion criteria. The collected variables included sex, age, smoking history, history of chronic alcohol consumption, Eastern ECOG performance status, American Joint Committee on Cancer (AJCC) staging, histological differentiation, Child–Pugh classification, and primary tumor size.\u003c/p\u003e\n\u003cp\u003eSerum lipid profiles were collected within one week prior to the initiation of systemic therapy, including LDL-C, HDL-C, total cholesterol, triglycerides (TG), apolipoprotein B (APOB), apolipoprotein A1 (APOA1), and lipoprotein alpha (Lpα).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEfficacy Evaluation and Follow-up\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eComputed tomography (CT) and other imaging modalities were performed prior to treatment and subsequently every 2–3 treatment cycles to monitor and evaluate therapeutic response. Treatment efficacy was assessed according to the RECIST version 1.1 and categorized as complete remission (CR), partial remission (PR), stable disease (SD), or progressive disease (PD). The objective response rate (ORR) and disease control rate (DCR) were used as short-term efficacy endpoints. ORR was defined as the proportion of patients achieving CR or PR, while DCR was defined as the proportion of patients achieving CR, PR, or SD.\u003c/p\u003e\n\u003cp\u003ePFS and OS were used as the primary indicators for evaluating long-term treatment efficacy. PFS was defined as the time from initiation of chemotherapy combined with immunotherapy for advanced disease to either documented disease progression or death from any cause. OS was defined as the time from initial diagnosis of ICC to death from any cause or the date of last follow-up. The primary endpoint was OS, while secondary endpoints included PFS, DCR, and ORR.\u003c/p\u003e\n\u003cp\u003ePatient disease progression was monitored through regular hospital follow-up visits or telephone interviews, with the follow-up period ending on May 15, 2024.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eSurvival Analysis\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eROC curve analysis was performed to calculate the AUC for serum lipid levels, including LDL-C, HDL-C, CHOL, TG, ApoB, ApoA-1, and Lpa. Parameters with AUC values greater than 0.7 were retained for subsequent analyses. The dataset was randomly divided into a training cohort and a testing cohort at a 7:3 ratio. The prognostic model was constructed using the training cohort and validated in the testing cohort. ROC curve analysis was conducted using the “pROC” package in R software to determine optimal cut-off values and corresponding AUCs. Serum lipid levels were then dichotomized into high and low groups based on these cut-off points. Clinical characteristics and short-term therapeutic efficacy between groups were compared using Fisher’s exact test. Survival curves for patients with high versus low serum lipid levels were generated using the “survival” package in R.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eNomogram\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eUnivariate and multivariate Cox regression analyses were performed using the “survival” package in R to identify independent prognostic factors among patients’ clinical characteristics and serum lipid levels. Forest plots were generated using the “ggplot2” package. Independent prognostic factors were incorporated into a nomogram constructed with the “nomogram” package. Calibration curves were established using the “rms” package, and the concordance index (C-index) was calculated for internal validation of the nomogram. The predictive performance of the nomogram-based prognostic model was further assessed using ROC curves and DCA, generated via the “timeROC” and “Dcurves” packages, respectively. External validation was conducted using data from the testing cohort.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eStatistical Methods\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eData organization, statistical analyses, tabulation, and visualization were performed using R software version 4.3.2. Categorical variables were presented as counts (percentages). Normally distributed continuous variables were summarized as mean ± standard deviation, while non-normally distributed continuous variables were expressed as median (interquartile range). Comparisons of categorical variables between two groups were conducted using Fisher’s exact test.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eEthics\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was conducted in accordance with the Declaration of Helsinki (6th revision, 2008). The study protocol was approved by the Ethics Committee of Cancer Hospital of Harbin Medical University (protocol number KY2023-18, approved on 01/11/2023).\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003e\u003cstrong\u003eGeneral Characteristics and Clinical Features\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eA total of 263 patients meeting the inclusion and exclusion criteria were ultimately enrolled in this study (Table 1). All patients received first-line chemotherapy regimens consisting of GC (gemcitabine 1000 mg/m\u0026sup2; intravenously on Days 1 and 8; cisplatin 25 mg/m\u0026sup2; intravenously on Days 1 and 8; repeated every 3 weeks), GS (gemcitabine 1000 mg/m\u0026sup2; intravenously on Days 1 and 8; S-1 administered orally twice daily from Days 1 to 14; repeated every 3 weeks), or GEMOX (gemcitabine 1000 mg/m\u0026sup2; intravenously on Days 1 and 8; oxaliplatin 100 mg/m\u0026sup2; intravenously on Day 1; repeated every 3 weeks). Upon disease progression, second-line chemotherapy consisted of mFOLFOX (oxaliplatin 85 mg/m\u0026sup2; intravenously on Day 1; leucovorin 350 mg/m\u0026sup2; intravenously on Day 1; 5-fluorouracil [5-FU] 400 mg/m\u0026sup2; intravenous bolus on Day 1, followed by continuous infusion of 1200 mg/(m\u0026sup2;\u0026middot;day) for 2 days; repeated every 2 weeks). Immunotherapy agents included camrelizumab (3 mg/kg intravenously) or sintilimab (200 mg intravenously).\u003c/p\u003e\n\u003cp\u003eAll patients were randomly divided into a training cohort and a testing cohort at a 7:3 ratio, including 184 patients in the training cohort and 79 patients in the testing cohort. Baseline characteristics between the training and testing cohorts showed no statistically significant differences except for primary tumor size (p = 0.022) (Table S1).\u003c/p\u003e\n\u003cp\u003eTable 1 Clinical characteristics of the 263 patients with advanced ICC\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"51%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eCharacteristics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eN=263\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eAge\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e59.0 [51.0; 65.0]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eGender\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e105 (39.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Male\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e158 (60.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eSmoking\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e234 (89.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e29 (11.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eDrinking\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e240 (91.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e23 (8.75%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eECOG\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e35 (13.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e222 (84.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e6 (2.28%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eAJCC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;III\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e137 (52.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;IV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e126 (47.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eHistological grade\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Well\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e48 (18.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Moderate-well\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e18 (6.84%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Moderately\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e68 (25.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Moderate-low\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e29 (11.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Poorly\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e100 (38.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eChild-Pugh grade\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e103 (39.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;B\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e160 (60.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTumor size (cm)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e6.00 [4.00;8.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eCEA (\u0026micro;g/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;＞4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e125 (47.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026le;4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e137 (52.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eCA199 (U/mL)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;＞20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e100 (38.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026le;20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e163 (62.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eDetermination of Optimal Cutoff Values for Serum Lipid Levels\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo elucidate the relationship between serum lipid levels and prognosis, ROC curve analysis was employed to determine the optimal cutoff values for serum lipid parameters\u0026mdash;including LDL-C, HDL-C, TC, TG, ApoB, ApoA1, and Lpa\u0026mdash;in order to stratify patients in the training and testing cohorts into high- and low-level groups.\u003c/p\u003e\n\u003cp\u003eIn the training cohort, the optimal cutoff value for LDL-C was 3.050 mmol/L, with an AUC of 0.87 (95% CI, 0.82\u0026ndash;0.92). The sensitivity and specificity were 0.908 and 0.722, respectively. Based on this cutoff, patients were stratified into a high LDL-C group (n = 85) and a low LDL-C group (n = 99) (Figure 2A). For HDL-C, the optimal cutoff was 1.290 mmol/L, with an AUC of 0.78 (95% CI, 0.71\u0026ndash;0.85), sensitivity of 0.776, and specificity of 0.657. Patients were divided into high HDL-C (n = 96) and low HDL-C (n = 88) groups accordingly (Figure 2B). The optimal cutoff for ApoA1 was 1.435 g/L, with an AUC of 0.75 (95% CI, 0.68\u0026ndash;0.82), sensitivity of 0.645, and specificity of 0.806. Patients were categorized into high ApoA1 (n = 70) and low ApoA1 (n = 114) groups (Figure 2C). Similar results were observed in the validation cohort.\u003c/p\u003e\n\u003cp\u003eThe remaining indicators were excluded from further analysis due to AUC values below 0.7 and were therefore not grouped based on optimal cutoff values (Figure S1).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eComparison of Clinical Characteristics among Different Levels of LDL-C, HDL-C, and APOA1\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eROC analysis identified LDL-C, HDL-C, and ApoA1 as candidates for further stratified analysis. Subsequently, baseline clinical characteristics were compared among patient groups with different expression levels of LDL-C, HDL-C, and ApoA1.\u003c/p\u003e\n\u003cp\u003eNo significant differences were observed between the high and low LDL-C groups of advanced ICC patients with respect to sex, age, smoking history, long-term alcohol consumption, ECOG performance status, AJCC stage, histological differentiation, Child\u0026ndash;Pugh classification, primary tumor size, carcinoembryonic antigen (CEA) level, and carbohydrate antigen 19-9 (CA19-9) level (Table 2).\u003c/p\u003e\n\u003cp\u003eTable 2 Relationship between different LDL-C and clinical characteristics of patients with advanced ICC\u003c/p\u003e\n\u003cdiv align=\"center\"\u003e\n \u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"99%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eCharacteristics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eHigh Group (n=85)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eLow Group (n=99)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003ep-\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003evalue\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eAge\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e59.0 [53.0;65.0]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e57.0 [51.0;65.0]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.461\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eGender\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.423\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e35 (41.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e34 (34.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Male\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e50 (58.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e65 (65.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eSmoking\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.535\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e72 (84.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e88 (88.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e13 (15.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e11 (11.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eDrinking\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;No\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e77 (90.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e90 (90.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Yes\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e8 (9.41%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e9 (9.09%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eECOG\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.865\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e0\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e10 (11.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e14 (14.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e74 (87.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e83 (83.8%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e2\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e1 (1.18%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e2 (2.02%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eAJCC\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.75\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;III\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e45 (52.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e49 (49.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;IV\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e40 (47.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e50 (50.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eHistological grade\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Well\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e13 (15.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e19 (19.2%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Moderate-well\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e9 (10.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e3 (3.03%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Moderately\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e15 (17.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e29 (29.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Moderate-low\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e11 (12.9%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e13 (13.1%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Poorly\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e37 (43.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e35 (35.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eChild-Pugh grade\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.222\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;A\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e31 (36.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e46 (46.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;B\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e54 (63.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e53 (53.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eTumor size (mm)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e7.00 [5.00;8.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e6.00 [4.00;8.00]\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.094\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eCEA (\u0026micro;g/L)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.608\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;＞4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e44 (52.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e47 (47.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026le;4\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e40 (47.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e52 (52.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u003cstrong\u003eCA199 (U/mL)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e0.171\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;＞20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e34 (40.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e29 (29.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026le;20\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e51 (60.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\n \u003cp\u003e70 (70.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n \u003c/table\u003e\n\u003c/div\u003e\n\u003cp\u003eThere were no statistically significant differences between the high and low HDL-C groups of ICC patients in terms of sex, age, smoking history, long-term alcohol consumption, ECOG performance status, AJCC stage, Child\u0026ndash;Pugh classification, primary tumor size, CEA level, and CA19-9 level (\u003cem\u003ep\u003c/em\u003e \u0026gt; 0.05). However, a statistically significant difference was observed in histological differentiation between the two groups (\u003cem\u003ep\u003c/em\u003e = 0.028, Table S2).\u003c/p\u003e\n\u003cp\u003eNo significant differences were observed between the high and low ApoA1 level groups of ICC patients regarding sex, age, smoking history, long-term alcohol consumption, ECOG performance status, AJCC staging, histological differentiation, Child-Pugh classification, CEA levels, and CA19-9 levels. However, a statistically significant difference was found in primary tumor size (\u003cem\u003ep\u003c/em\u003e = 0.01), with the median tumor size in the low ApoA1 group being larger than that in the high ApoA1 group (Table S3).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eRelationship between LDL-C, HDL-C, and APOA1 and Short-term Treatment Efficacy\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo investigate the relationship between LDL-C, HDL-C, and ApoA1 levels and patients\u0026rsquo; short-term therapeutic responses, this study analyzed DCR and ORR, with the results as follows:\u003c/p\u003e\n\u003cp\u003eA total of 85 patients exhibited low LDL-C levels (PR = 11, SD = 32, PD = 42), while 72 patients had high LDL-C levels (PR = 4, SD = 30, PD = 38). The DCR were 50.59% and 47.22% for the low- and high-LDL-C groups, respectively (\u003cem\u003ep\u003c/em\u003e = 0.889); the ORR were 12.94% and 5.56%, respectively (\u003cem\u003ep\u003c/em\u003e = 0.182) (Table 3). For HDL-C, 113 patients were classified as low-level (PR = 11, SD = 40, PD = 62) and 44 as high-level (PR = 4, SD = 22, PD = 18). The DCRs were 45.13% and 59.09% for the low- and high-HDL-C groups, respectively (\u003cem\u003ep\u003c/em\u003e = 0.367); the ORRs were 9.73% and 9.09%, respectively (\u003cem\u003ep\u003c/em\u003e = 1) (Table 3). Regarding ApoA1, 97 patients had low levels (PR = 7, SD = 38, PD = 52) and 60 patients had high levels (PR = 8, SD = 24, PD = 28). The DCRs were 46.39% and 53.33% for the low- and high-ApoA1 groups, respectively (p = 0.67); the ORRs were 7.22% and 13.33%, respectively (\u003cem\u003ep\u003c/em\u003e = 0.278) (Table 3). Data from the validation cohort are presented in Table S4.\u003c/p\u003e\n\u003cp\u003eIn summary, patients with low LDL-C, high HDL-C, and high ApoA1 levels exhibited better short-term therapeutic responses; however, none of these three indicators showed statistically significant differences in DCR or ORR.\u003c/p\u003e\n\u003cp\u003eTable 3 Correlations of different levels of LDL-C、HDL-C and APOA1 of training cohort with recent efficacy\u003c/p\u003e\n\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" width=\"99%\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eCharacteristics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePR (n=15)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eSD (n=62)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003ePD (n=80)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eDCR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003ep\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003evalue\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eORR\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u003cem\u003ep\u0026nbsp;\u003c/em\u003e\u003c/strong\u003e\u003cstrong\u003evalue\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eLDL.C\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.889\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.1819\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; High Group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4 (26.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e30 (48.4%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e38 (47.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e47.22%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e5.56%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Low Group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e11 (73.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e32 (51.6%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e42 (52.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e50.59%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e12.94%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eHDL\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.3672\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e1\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; High Group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e4 (26.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e22 (35.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e18 (22.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e59.09%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e9.09%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Low Group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e11 (73.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e40 (64.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e62 (77.5%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e45.13%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e9.73%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u003cstrong\u003eAPOA1\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.6701\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e0.2784\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; High Group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e8 (53.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e24 (38.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e28 (35.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e53.33%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e13.33%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; Low Group\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7 (46.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e38 (61.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e52 (65.0%)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e46.39%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\n \u003cp\u003e7.22%\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\"\u003e\u003cbr\u003e\u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cstrong\u003eRelationship between LDL-C, HDL-C, and ApoA1 and Long-term Therapeutic Outcomes\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eSubsequently, OS was compared among advanced ICC patients stratified by LDL-C, HDL-C, and ApoA1 levels. In the training and validation cohorts, the median OS (mOS) for the low LDL-C groups were 35.3 months and 17 months, respectively, while the high LDL-C groups had mOS of 13.2 months and 15.2 months. In the training cohort, the low LDL-C group exhibited a significantly longer mOS than the high LDL-C group by 22.1 months (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001, HR = 3.968, 95% CI 2.698\u0026ndash;5.835) (Figure 3A). For HDL-C, in both the training and validation sets, the mOS in the low HDL-C groups were 14.9 months and 15.7 months, and in the high HDL-C groups were 32.4 months and 24.9 months, respectively. The training cohort\u0026rsquo;s low HDL-C group had a shorter mOS than the high HDL-C group by 17.5 months (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001, HR = 0.364, 95% CI 0.248\u0026ndash;0.533) (Figure 3B). Regarding ApoA1, in both the training and validation sets, the mOS for the low ApoA1 groups were 17.5 months and 16.1 months, and for the high ApoA1 groups were 12.88 months and 26.2 months, respectively. The low ApoA1 group in the training cohort showed a longer mOS than the high ApoA1 group by 4.62 months (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001, HR = 0.463, 95% CI 0.311\u0026ndash;0.689) (Figure 3C).\u003c/p\u003e\n\u003cp\u003eAdditionally, PFS was compared between patients with different levels of LDL-C, HDL-C, and ApoA1 in both the training and validation cohorts. In the training cohort, the mPFS was 8.8 months in the low LDL-C group versus 6.5 months in the high LDL-C group, with the low LDL-C group exhibiting a significantly longer mPFS by 2.3 months (\u003cem\u003ep\u003c/em\u003e = 0.027, HR = 1.549, 95% CI 1.021\u0026ndash;2.348). In the validation cohort, the low LDL-C group had a 5-month longer mPFS compared to the high LDL-C group (\u003cem\u003ep\u003c/em\u003e = 0.029, HR = 2.04, 95% CI 1.01\u0026ndash;4.14) (Figure 4A). In the training cohort, the mPFS was 6.1 months for the low HDL-C group and 9.17 months for the high HDL-C group, with the low HDL-C group showing a significantly shorter mPFS by 3.07 months (\u003cem\u003ep\u003c/em\u003e = 0.025, HR = 0.642, 95% CI 0.426\u0026ndash;0.967). Similarly, in the validation cohort, the low HDL-C group had a 4.7-month shorter mPFS than the high HDL-C group (\u003cem\u003ep\u003c/em\u003e = 0.038, HR = 0.496, 95% CI 0.23\u0026ndash;1.072) (Figure 4B). For ApoA1, the training cohort showed an mPFS of 7.83 months in the low ApoA1 group and 7.33 months in the high ApoA1 group, with no significant difference (\u003cem\u003ep\u003c/em\u003e = 0.64, HR = 0.906, 95% CI 0.601\u0026ndash;1.367). However, in the validation cohort, the low ApoA1 group had a significantly shorter mPFS by 6.8 months compared to the high ApoA1 group (\u003cem\u003ep\u003c/em\u003e = 0.004, HR = 0.395, 95% CI 0.183\u0026ndash;0.851) (Figure 4C).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eIndependent prognostic factors affecting outcomes in advanced intrahepatic cholangiocarcinoma patients\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eTo further investigate independent prognostic factors, Cox proportional hazards regression models were employed. General clinical characteristics, along with varying levels of LDL-C, HDL-C, and ApoA1, were included in both univariate and multivariate analyses based on OS and PFS.\u003c/p\u003e\n\u003cp\u003eUnivariate Cox regression analysis based on the training cohort revealed that low levels of HDL-C, low levels of ApoA1, and larger primary tumor size were significantly associated with poor OS in ICC patients, whereas low LDL-C levels were correlated with favorable prognosis (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05) (Figure 5A). Similarly, univariate Cox analysis in the validation cohort demonstrated that low HDL-C and low ApoA1 levels were adverse prognostic factors, while low LDL-C levels and well to moderately differentiated histology were associated with improved prognosis in ICC patients (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05) (Figure 5B).\u003c/p\u003e\n\u003cp\u003eMultivariate Cox regression analyses in both the training and validation cohorts demonstrated that low levels of LDL-C and HDL-C were independent prognostic factors for ICC patients (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05). Specifically, low LDL-C was identified as an independent protective factor, whereas low HDL-C was an independent risk factor for poor prognosis in ICC patients (Figure 5C\u0026ndash;D).\u003c/p\u003e\n\u003cp\u003eSubsequently, Cox regression analysis was performed to evaluate the impact of LDL-C, HDL-C, and ApoA1 on PFS in ICC patients. Univariate Cox analysis in the training cohort demonstrated that well-differentiated tumors and low LDL-C levels were favorable prognostic factors for PFS, whereas low HDL-C was also identified as a favorable prognostic factor for PFS (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05). Low ApoA1 was associated with poorer PFS, but this finding did not reach statistical significance (Figure S2A). In the validation cohort, the trends for all three markers were consistent with those observed in the training cohort; however, only the association with ApoA1 reached statistical significance (Figure S2B).\u003c/p\u003e\n\u003cp\u003eMultivariate Cox regression analysis based on the training cohort indicated that low LDL-C was a protective factor for PFS in ICC patients, whereas low HDL-C was a risk factor; however, these results did not reach statistical significance (Figure S2C). In the validation cohort, multivariate Cox regression identified low ApoA1 as an adverse prognostic factor for PFS in ICC patients (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.05) (Figure S2D).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConstruction and Validation of the Nomogram Prediction Model\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eGiven the stronger association between the aforementioned serum molecules and OS in ICC patients, we focused exclusively on developing a predictive model for OS to forecast the prognosis of advanced ICC patients receiving chemotherapy combined with immunotherapy. Based on the results of the multivariate Cox regression analysis, LDL-C and HDL-C were selected as variables to construct a nomogram for predicting 1-, 2-, and 3-year survival probabilities. Patients with low HDL-C and high LDL-C scores exhibited higher risk scores, indicating poorer survival rates at 1, 2, and 3 years (Figure 6A).\u003c/p\u003e\n\u003cp\u003eSubsequently, the nomogram prediction model was validated using calibration curves. Internal validation demonstrated that the C-index for predicting the prognosis of advanced ICC patients receiving immunotherapy was 0.68. The calibration curves for 1-, 2-, and 3-year survival closely approximated the ideal diagonal line (Figure 6B). DCA was performed to evaluate the clinical net benefit, revealing that the DCA models at 1-, 2-, and 3 years remained within the optimal range across certain threshold probabilities, indicating that the prediction model provides meaningful clinical net benefit and has practical clinical utility (Figure 6C). Time-dependent ROC analysis yielded AUCs of 0.716, 0.761, and 0.810 at 1-, 2-, and 3 years, respectively, further supporting the model\u0026rsquo;s predictive capability (Figure 6D). Moreover, the Hosmer-Leme show goodness-of-fit test resulted in a \u003cem\u003ep\u003c/em\u003e-value of 0.1083 (Table S5), indicating no significant systematic bias and an acceptable model fit.\u003c/p\u003e\n\u003cp\u003eTo further validate the clinical utility of the prognostic model, external validation was performed using the testing cohort. Calibration of the model in the testing cohort demonstrated that the calibration curve closely approximated the ideal reference line (Figure 6E). The model\u0026rsquo;s discriminative ability for external data was assessed by ROC analysis, yielding an AUC of 0.844 with a 95% CI of 0.757\u0026ndash;0.931 (Figure 6F). These results indicate that the model possesses robust predictive performance in external datasets.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eVarious biomarkers, including gene mutation burden and tumor load, have been widely applied for prognostic assessment in cancer patients. However, due to high costs and challenges in sample collection, there remains a need to identify biomarkers that are more easily detectable while maintaining high specificity and sensitivity. Lipids play crucial roles in energy storage, metabolism, and signal transduction involved in cellular activities\u003csup\u003e[22]\u003c/sup\u003e. Studies have demonstrated that cancer cells within the tumor microenvironment require abundant nutrients to sustain tumor growth. Energy generated solely through glycolysis is insufficient to meet these demands; therefore, lipid metabolism is utilized to support rapid proliferation, survival, migration, invasion, and metastasis of tumor cells\u003csup\u003e[23]\u003c/sup\u003e. Liquid biopsy-based assessment of peripheral blood lipid profiles has emerged as a promising approach and has been applied in malignancies such as lung adenocarcinoma and breast cancer. Related investigations have also been conducted in intrahepatic cholangiocarcinoma\u003csup\u003e[14,18,24,25]\u003c/sup\u003e.\u003c/p\u003e\n\u003cp\u003eThis study utilized ROC curve analysis to select LDL-C, HDL-C, and ApoA1 as prognostic factors for patients with advanced ICC undergoing chemotherapy combined with immunotherapy. Survival analyses based on the training and validation cohorts demonstrated that patients with low serum LDL-C, high HDL-C, and elevated ApoA1 levels had significantly longer mOS. Regarding PFS, the training cohort showed that patients with low LDL-C and high HDL-C exhibited prolonged mPFS. Although the validation cohort results did not reach statistical significance, the overall trends for these two markers were consistent with the training cohort, likely due to the smaller sample size of the validation cohort. Consistent with our findings, Shu et al. reported that postoperative ICC patients with high serum HDL-C derived greater clinical benefit\u003csup\u003e[26]\u003c/sup\u003e. Additionally, Lin et al. found that cervical cancer patients exhibited higher LDL-C and lower HDL-C levels compared to healthy controls, and that elevated LDL-C and decreased HDL-C were adverse prognostic factors in this population\u003csup\u003e[27]\u003c/sup\u003e. These findings align with our results, suggesting that increased cancer risk and poor prognosis are associated with low HDL-C and high LDL-C levels. Conversely, another study observed that head and neck squamous cell carcinoma patients with high LDL-C and low ApoA1 levels experienced better PFS\u003csup\u003e[28]\u003c/sup\u003e. This discrepancy may reflect metabolic heterogeneity among different tumor types.\u003c/p\u003e\n\u003cp\u003eMoreover, multiple studies have reported that elevated serum TG levels are associated with poor prognosis in lung cancer, thyroid cancer, rectal cancer, breast cancer, and prostate cancer\u003csup\u003e[29–31]\u003c/sup\u003e. However, in the present study, TG demonstrated an AUC of less than 0.7 in the ROC analysis, indicating suboptimal predictive performance. Therefore, TG was not included in further analyses. Future studies with larger sample sizes are warranted to clarify the potential association between TG and ICC.\u003c/p\u003e\n\u003cp\u003eSubsequent Cox regression analyses based on the training and validation cohorts revealed that low serum LDL-C was a protective factor for OS in advanced ICC patients receiving chemotherapy combined with immunotherapy, whereas low HDL-C and ApoA1 were risk factors for OS. Multivariate Cox regression further identified low LDL-C and low HDL-C as independent protective and risk factors for OS, respectively. Similarly, Shu et al. reported that low HDL-C was an independent risk factor for postoperative ICC patients\u003csup\u003e[22]\u003c/sup\u003e, supporting the role of low HDL-C as an independent prognostic indicator in ICC. Additionally, Chen et al. identified LDL-C as an independent prognostic factor in non-esophageal squamous cell carcinoma, further confirming the prognostic value of LDL-C in cancer\u003csup\u003e[32]\u003c/sup\u003e. In another study on cervical cancer, patients exhibited elevated LDL-C levels and decreased HDL-C levels compared to healthy controls\u003csup\u003e[27]\u003c/sup\u003e. Conversely, a study on head and neck squamous cell carcinoma found that high LDL-C and low ApoA1 were protective factors\u003csup\u003e[28]\u003c/sup\u003e. These findings suggest heterogeneity in lipid metabolism among different cancer types, with the same serum lipid markers having distinct prognostic implications depending on the tumor context.\u003c/p\u003e\n\u003cp\u003eSubsequently, LDL-C and HDL-C identified as predictive factors through Cox regression analysis were used to construct a nomogram for predicting 1-, 2-, and 3-year survival rates in advanced ICC patients receiving chemotherapy combined with immunotherapy. The nomogram was internally and externally validated using the training and validation cohorts, respectively, demonstrating robust predictive performance and clinical utility. These findings are consistent with previous studies in thyroid cancer, which also suggest that HDL-C can serve as a prognostic biomarker for cancer\u003csup\u003e[20]\u003c/sup\u003e. In summary, LDL-C and HDL-C may serve as reliable biomarkers for predicting prognosis in patients with advanced ICC.\u003c/p\u003e\n\u003cp\u003eHowever, this study has several limitations. First, as a single-center study with a relatively small sample size due to the rarity of cholangiocarcinoma, the conclusions may not fully represent the broader population of advanced ICC patients in China. Future multicenter studies with larger cohorts are needed to validate these findings. Second, serum lipid levels can be influenced by factors such as patient age, sex, hormonal status, and underlying comorbidities; therefore, stricter study controls are required to minimize confounding effects. Third, this study did not investigate the metabolic reprogramming mechanisms underlying the observed alterations in serum lipid profiles, and the causes of lipid level changes remain unclear. Further research will be conducted to explore these underlying mechanisms.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, we successfully constructed and validated a prognostic model to predict the OS of patients with advanced ICC, which provides a more accurate basis for the immunotherapy decision of such patients. The strategy of chemotherapy in combination with immunotherapy dominates in advanced ICC patients,and it is suggested that immunotherapy should be incorporated into clinical treatment protocols more frequently.\u003c/p\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthical considerations\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe study was conducted in accordance with the Declaration of Helsinki (6th revision, 2008). The study protocol was approved by the Ethics Committee of Cancer Hospital of Harbin Medical University (protocol number KY2023-18, approved on 01/11/2023).\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe ethics committee/institutional review board waived the requirement of written informed consent for participation from the participants or the participants\\’ legal guardians/next of kin because of the retrospective nature of the study.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConflict of interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construced as a potential competing of interest.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding statement\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe author(s) received no financial support for the research, authorship, and/or publication of this article.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eData availability\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding author.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthor contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCZJ: Data curation, Methodology, Software, Visualization, Writing-original draft. KJL: Data curation, Methodology, Software, Validation, Writing-original draft. TYF: Data curation, Investigation, Writing-original draft. WL: Data curation, Investigation, Writing-original draft. LY: Data curation, Investigation, Writing-original draft. PCC: Data curation, Investigation, Writing-original draft. JT: Conceptualization, Project administration, Supervision, Writing-review \u0026amp; editing. KGL: Conceptualization, Project administration, Supervision, Writing-review \u0026amp; editing.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eQurashi M, Vithayathil M, Khan S. Epidemiology of cholangiocarcinoma[J]. EJSO, 2025, 51(2).\u003c/li\u003e\n\u003cli\u003ePiha-Paul S A, Oh D Y, Ueno M, Malka D, Chung H C, Nagrial A, Kelley R K, Ros W, Italiano A, Nakagawa K, Rugo H S, De Braud F, Varga A I, Hansen A, Wang H, Krishnan S, Norwood K G, Doi T. Efficacy and safety of pembrolizumab for the treatment of advanced biliary cancer: Results from the KEYNOTE-158 and KEYNOTE-028 studies.[J]. Int J Cancer, 2020, 147(8): 2190-2198.\u003c/li\u003e\n\u003cli\u003eKelley R, Bridgewater J, Gores G, Zhu A. Systemic therapies for intrahepatic cholangiocarcinoma[J]. J. Hepatol., 2020, 72(2): 353-363.\u003c/li\u003e\n\u003cli\u003eRui R, Zhou L, He S. Cancer immunotherapies: advances and bottlenecks.[J]. Front. Immunol., 2023, 14: 1212476.\u003c/li\u003e\n\u003cli\u003eRaggi C, Taddei M L, Rae C, Braconi C, Marra F. Metabolic reprogramming in cholangiocarcinoma[J]. J Hepatol, 2022, 77(3): 849-864.\u003c/li\u003e\n\u003cli\u003eMaurya S K, Chaudhri S, Kumar S, Gupta S. Repurposing of Metabolic Drugs Metformin and Simvastatin as an Emerging Class of Cancer Therapeutics.[J]. Pharm Res, 2025, 42(1): 49-67.\u003c/li\u003e\n\u003cli\u003eRuze R, Liu T, Zou X, Song J, Chen Y, Xu R, Yin X, Xu Q. Obesity and type 2 diabetes mellitus: connections in epidemiology, pathogenesis, and treatments.[J]. Front. Endocrinol., 2023, 14: 1161521.\u003c/li\u003e\n\u003cli\u003eTang H Y, Chen A Q, Zhang H, Gao X F, Kong X Q, Zhang J J. Vascular Smooth Muscle Cells Phenotypic Switching in Cardiovascular Diseases.[J]. Cells, 2022, 11(24).\u003c/li\u003e\n\u003cli\u003eYang J, Shay C, Saba N F, Teng Y. Cancer metabolism and carcinogenesis.[J]. Exp Hematol Oncol, 2024, 13(1): 10.\u003c/li\u003e\n\u003cli\u003e Sun Y, Cao D, Zhang Y, Wu Y, Jia Z, Cui Y, Li D, Cao X, Jiang J. Appraising associations between signature lipidomic biomarkers and digestive system cancer risk: novel evidences from a prospective cohort study of UK Biobank and Mendelian randomization analyses.[J]. Lipids Health Dis, 2024, 23(1): 61.\u003c/li\u003e\n\u003cli\u003e Wang S, Huang X, Zhao S, Lv J, Li Y, Wang S, Guo J, Wang Y, Wang R, Zhang M, Qiu W. Progressions of the correlation between lipid metabolism and immune infiltration characteristics in gastric cancer and identification of BCHE as a potential biomarker.[J]. Front Immunol, 2024, 15: 1327565.\u003c/li\u003e\n\u003cli\u003e Duan Y, Deng M, Liu B, Meng X, Liao J, Qiu Y, Wu Z, Lin J, Dong Y, Duan Y, Sun Y. Mitochondria targeted drug delivery system overcoming drug resistance in intrahepatic cholangiocarcinoma by reprogramming lipid metabolism.[J]. Biomaterials, 2024, 309: 122609.\u003c/li\u003e\n\u003cli\u003e Qian L, Wang G, Li B, Su H, Qin L. Regulation of lipid metabolism by APOE4 in intrahepatic cholangiocarcinoma via the enhancement of ABCA1 membrane expression[J]. PeerJ, 2024, 12: e16740.\u003c/li\u003e\n\u003cli\u003e Lori G, Pastore M, Navari N, Piombanti B, Booijink R, Rovida E, Tusa I, Lewinska M, Andersen J B, Lottini T, Arcangeli A, Taddei M L, Pranzini E, Mancini C, Anceschi C, Madiai S, Sacco E, Rota S, Trapani A, Agrimi G, Ramazzotti M, Ostano P, Peraldo Neia C, Parri M, Carli F, Sabatini S, Gastaldelli A, Marra F, Raggi C. Altered fatty acid metabolism rewires cholangiocarcinoma stemness features[J]. JHEP Rep. : Innov. Hepatol., 2024, 6(10): 101182.\u003c/li\u003e\n\u003cli\u003e Chen Y, Zhou Y, Ren R, Chen Y, Lei J, Li Y. Harnessing lipid metabolism modulation for improved immunotherapy outcomes in lung adenocarcinoma.[J]. J. ImmunoTher. Cancer, 2024, 12(7).\u003c/li\u003e\n\u003cli\u003e Ruiz de Gauna M, Biancaniello F, Gonz\u0026aacute;lez-Romero F, Rodrigues P M, Lapitz A, G\u0026oacute;mez-Santos B, Olaizola P, Di Matteo S, Aurrekoetxea I, Labiano I, Nieva-Zuluaga A, Benito-Vicente A, Perugorria M J, Apodaka-Biguri M, Paiva N A, S\u0026aacute;enz de Urturi D, Buqu\u0026eacute; X, Delgado I, Mart\u0026iacute;n C, Azkargorta M, Elortza F, Calvisi D F, Andersen J B, Alvaro D, Cardinale V, Bujanda L, Banales J M, Aspichueta P. Cholangiocarcinoma progression depends on the uptake and metabolization of extracellular lipids.[J]. Hepatol. (baltim. Md,), 2022, 76(6): 1617-1633.\u003c/li\u003e\n\u003cli\u003e Liu X, Hartman C L, Li L, Albert C J, Si F, Gao A, Huang L, Zhao Y, Lin W, Hsueh E C, Shen L, Shao Q, Hoft D F, Ford D A, Peng G. Reprogramming lipid metabolism prevents effector T cell senescence and enhances tumor immunotherapy.[J]. Sci. Transl. Med., 2021, 13(587).\u003c/li\u003e\n\u003cli\u003e Ma F, Liu X, Zhang Y, Tao Y, Zhao L, Abusalamah H, Huffman C, Harbison R A, Puram S V, Wang Y, Peng G. Tumor extracellular vesicle-derived PD-L1 promotes T cell senescence through lipid metabolism reprogramming[J]. Sci. Transl. Med., 2025, 17(785).\u003c/li\u003e\n\u003cli\u003e Yang K, Wang X, Song C, He Z, Wang R, Xu Y, Jiang G, Wan Y, Mei J, Mao W. The role of lipid metabolic reprogramming in tumor microenvironment.[J]. Theranostics, 2023, 13(6): 1774-1808.\u003c/li\u003e\n\u003cli\u003e Yuan J, Chen Z, Zhang J, Chen X, Chen S, Wang X, Zhang J, Su X, Zhu W, Sheng J, Teng L, Wang W. Preoperative Serum Lipids as Novel Predictors of Survival in 3575 Patients With Papillary Thyroid Cancer[J]. J. Clin. Endocrinol. Metab., 2024, 110(3): 668-676.\u003c/li\u003e\n\u003cli\u003e Salgado-Montilla J, Soto Salgado M, Surillo Trautmann B, S\u0026aacute;nchez-Ortiz R, Irizarry-Ram\u0026iacute;rez M. Association of serum lipid levels and prostate cancer severity among Hispanic Puerto Rican men.[J]. Lipids Health Dis., 2015, 14: 111.\u003c/li\u003e\n\u003cli\u003e Yu S M, Chang X J, Gu Y Y, Jia X D, Gao X D, Huang J G, Dong J H, Zeng Z. Serum high-density lipoprotein cholesterol levels predict early recurrence and prognosis of intrahepatic cholangiocarcinoma after surgical resection[J]. Heliyon, 2024, 10(11): e32113.\u003c/li\u003e\n\u003cli\u003e Li X, Jia Y, Li Y, Hei H, Zhang S, Qin J. Crosstalk between metabolic reprogramming and microbiota: implications for cancer progression and novel therapeutic opportunities[J]. FRONTIERS IN IMMUNOLOGY, 2025, 16.\u003c/li\u003e\n\u003cli\u003e Bian X, Liu R, Meng Y, Xing D, Xu D, Lu Z. Lipid metabolism and cancer[J]. J. Exp. Med., 2021, 218(1).\u003c/li\u003e\n\u003cli\u003e Yu X, Tong H, Chen J, Tang C, Wang S, Si Y, Wang S, Tang Z. CircRNA MBOAT2 promotes intrahepatic cholangiocarcinoma progression and lipid metabolism reprogramming by stabilizing PTBP1 to facilitate FASN mRNA cytoplasmic export[J]. Cell Death Dis, 2023, 14(1): 20.\u003c/li\u003e\n\u003cli\u003e Zhu H, Hu H, Hao B, Zhan W, Yan T, Zhang J, Wang S, Hu H, Zhang T. Insights into a machine learning-based palmitoylation-related gene model for predicting the prognosis and treatment response of breast cancer patients[J]. Technol Cancer Res Treat, 2024, 23: 15330338241263434.\u003c/li\u003e\n\u003cli\u003e Cheng L, Li Z, Zheng Q, Yao Q. Correlation study of serum lipid levels and lipid metabolism-related genes in cervical cancer[J]. FRONTIERS IN ONCOLOGY, 2024, 14.\u003c/li\u003e\n\u003cli\u003e Wang S, Wang L, Li H, Zhang J, Peng J, Cheng B, Song M, Hu Q. Correlation analysis of plasma lipid profiles and the prognosis of head and neck squamous cell carcinoma[J]. Oral Dis., 2024, 30(2): 329-341.\u003c/li\u003e\n\u003cli\u003e Li C, Wang F, Cui L, Li S, Zhao J, Liao L. Association between abnormal lipid metabolism and tumor.[J]. Front. Endocrinol., 2023, 14: 1134154.\u003c/li\u003e\n\u003cli\u003e Chen J Y, Chi N H, Lee H S, Hsiung C N, Wu C W, Fan K C, Lee M R, Wang J Y, Ho C C, Shih J Y. Lipid Levels and Lung Cancer Risk: Findings from the Taiwan National Data Systems from 2012 to 2018.[J]. J Epidemiol Glob Health, 2025, 15(1): 11.\u003c/li\u003e\n\u003cli\u003e Zhang W, Li Z, Huang Y, Zhao J, Guo S, Wang Q, Guo S, Li Q. Complex Role of Circulating Triglycerides in Breast Cancer Onset and Survival: Insights From Two-Sample Mendelian Randomization Study.[J]. Cancer Med., 2025, 14(4): e70698.\u003c/li\u003e\n\u003cli\u003e Chen S, Li X, Wen X, Peng S, Xue N, Xing S, Liu Y. Prognostic nomogram integrated baseline serum lipids for patients with non-esophageal squamous cell carcinoma[J]. Ann. Transl. Med., 2019, 7(20): 548.\u003c/li\u003e\n\u003c/ol\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Serum lipid levels, Intrahepatic Cholangiocarcinoma, Immunotherapy, Nomogram","lastPublishedDoi":"10.21203/rs.3.rs-7803902/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7803902/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground: \u003c/strong\u003eLipid metabolic reprogramming plays a critical role in tumor progression. Serum lipid levels have been associated with the prognosis of various malignancies.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAims:\u003c/strong\u003eTo develop a novel nomogram based on serum lipid parameters to predict overall survival in patients with intrahepatic cholangiocarcinoma.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods: \u003c/strong\u003eSerum lipid profiles and survival data were collected prior to the initiation of chemotherapy combined with immunotherapy. Survival analysis was performed to identify prognostic factors associated with ICC. Independent prognostic factors were used to construct a nomogram. The predictive performance of the nomogram was evaluated. External validation of the survival analysis and nomogram for serum lipids was conducted using a validation cohort.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults: \u003c/strong\u003eLow-density lipoprotein cholesterol, high-density lipoprotein cholesterol, and apolipoprotein A1 were selected for further analysis. Survival analysis demonstrated that patients with low LDL-C, high HDL-C, and high ApoA1 levels exhibited significantly longer OS and PFS. A nomogram incorporating LDL-C and HDL-C was constructed to predict 1-, 2-, and 3-year survival probabilities. The nomogram exhibited favorable predictive performance.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDiscussion: P\u003c/strong\u003ere-treatment serum levels of LDL-C, HDL-C, and ApoA1 exhibited significant prognostic value for advanced ICC. The nomogram constructed based on LDL-C and HDL-C effectively predicted survival outcomes, providing a theoretical basis to support treatment decision-making and individualized prognostic assessment in clinical practice.\u003c/p\u003e","manuscriptTitle":"Prognostic Prediction of Advanced Intrahepatic Cholangiocarcinoma Patients Receiving Chemotherapy Combined with Immunotherapy Based on Serum Lipid Profiles","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-11-13 08:01:35","doi":"10.21203/rs.3.rs-7803902/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"b38500df-62b5-4f6f-9149-d33a5d397196","owner":[],"postedDate":"November 13th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-12-30T12:55:00+00:00","versionOfRecord":[],"versionCreatedAt":"2025-11-13 08:01:35","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7803902","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7803902","identity":"rs-7803902","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}