Intrahepatic cholangiocarcinoma (ICC) is a malignant tumor that originates from the intrahepatic biliary epithelium, accounting for approximately 10%–20% of all cholangiocarcinoma cases. It is the second most common primary liver malignancy, constituting approximately 10% of all primary liver cancers (1, 2). The global incidence of ICC has been significantly increasing in recent years (3). The disease is highly aggressive and associated with a poor prognosis, with postoperative 3- and 5-year overall survival (OS) rates of only 30% and 18%, respectively (4). Complete surgical resection remains the primary curative treatment for ICC (5); however, only approximately 30% of patients are considered eligible for resection at initial diagnosis (6). Recent advances in neoadjuvant therapy, including chemotherapy, targeted therapy, and immunotherapy, have provided opportunities for conversion surgery in initially unresectable or borderline resectable cases, which significantly expand the pool of surgical candidates (7–10).

Due to its anatomical location, ICC resection frequently involves complex vascular and biliary reconstruction, thereby presenting considerable technical challenges and perioperative risks (11, 12). Surgical management has become more complicated due to the expanding use of neoadjuvant therapies. The dual aspects of antitumor efficacy and potential drug-induced liver injury contribute substantially to unclear surgical outcomes, underscoring the critical importance of precise surgical quality assessment in this context (13, 14). Several studies have indicated that surgical quality directly affects patient prognosis. For example, R0 resection has significantly improved survival rates (15, 16). However, current surgical outcome assessments often rely on isolated indicators (17), which lack integration and fail to provide a comprehensive reflection of overall treatment quality, despite their clinical relevance.

Textbook outcome (TO) is a multidimensional composite endpoint that incorporates multiple indicators such as R0 resection, no perioperative blood transfusion, no complications, no prolonged postoperative hospital stay, no unplanned readmission within 30 days, and no 30-day mortality (18, 19). TO was initially applied in colorectal cancer and is now increasingly adopted in ICC surgical evaluation and strongly correlated with improved long-term survival (18–21). Traditionally, chemotherapy-based neoadjuvant therapy was considered a negative predictor for TO, with studies indicating its potential to increase surgical risk (19, 22). However, with rapid advancements in targeted and immunotherapy, this perception warrants reevaluation. These novel therapies, characterized by higher response rates and improved safety profiles, may improve resectability, reduce surgical difficulty, and consequently facilitate TO achievement (23).

Therefore, using a cohort of patients with ICC and detailed neoadjuvant therapy information, this study aimed to investigate the association between TO and patient prognosis, identify clinical factors that are independently associated with TO achievement, and specifically assess the impact of neoadjuvant therapy—particularly regimens that incorporate targeted and immunotherapeutic agents—on TO. Furthermore, we developed and validated a TO-integrated prognostic prediction model to provide a practical tool to support clinical decision-making and surgical quality assessment.

This retrospective cohort study consecutively enrolled patients who underwent resection for ICC at the Second Affiliated Hospital of Zhejiang University School of Medicine from May 2011 to September 2024. Inclusion criteria were pathological confirmation of ICC. Exclusion criteria were concomitant other malignancies, incomplete follow-up data, and missing information on tumor characteristics or surgical procedures.

Data were systematically collected on patient demographics [age, sex, and body mass index (BMI]), underlying disease (e.g., diabetes and hypertension), liver function parameters, and American Society of Anesthesiologists classification, tumor characteristics (e.g., size, T stage, differentiation, vascular invasion, lymph node metastasis, and distant metastasis), laboratory values [e.g., serum albumin (ALB), total bilirubin, prothrombin time (PT), carcinoembryonic antigen (CEA), carbohydrate antigen 19-9 (CA19-9), and alpha-fetoprotein (AFP)], and surgical details (e.g., scope of surgery, surgical method, lymph node dissection, operation time, and intraoperative blood loss). Further, information on neoadjuvant and adjuvant therapies was recorded, along with surgical quality indicators and postoperative survival outcomes.

TO was defined as the simultaneous achievement of the following six criteria: R0 resection, no perioperative transfusion, no postoperative complications, no 30-day mortality, no unplanned readmission within 30 days, and postoperative length of stay not exceeding the 75th percentile of the cohort (18, 19). Neoadjuvant therapy refers to preoperative systemic treatment, including chemotherapy, chemotherapy with targeted therapy, or immunotherapy, for patients with borderline resectable tumors. It may be considered for those with a large tumor burden, multifocal disease, or major vascular involvement. Additionally, for patients with technically resectable intrahepatic tumors but high-risk features for recurrence, neoadjuvant therapy may also be considered. The decision to administer neoadjuvant therapy or proceed directly with surgery also depends on the patient’s overall health, individual preferences, and other factors (24). Resections were classified as major (≥3 Couinaud’s segments) or minor (<3 segments) (25). Intraoperative blood loss and operation time were dichotomized according to median values.

The Shapiro–Wilk test was used to test continuous variables for normality. Normally distributed variables were expressed as mean ± standard deviation and compared using analysis of variance; non-normal variables were reported as median (interquartile range) and compared using non-parametric tests. Categorical variables were expressed as frequencies and percentages, compared with χ² or Fisher’s exact test. Baseline characteristics and perioperative outcomes were compared between the TO and non-TO groups. Reasons for not achieving TO were analyzed. Univariate and multivariate logistic regression analyses, with additional in-depth evaluation focusing on neoadjuvant therapy, were conducted to identify factors associated with TO. Based on TO status, univariate and multivariate Cox regression analyses developed a prognostic model, visualized via a nomogram. Receiver operating characteristic (ROC) curves were used to assess discriminative ability. R (version 4.3.1) was used for analyses, with p-values of <0.05 indicating significance.

Comprehensive surgical quality assessment is crucial for patient outcomes due to the technical complexity of ICC resection and challenging postoperative recovery. TO, a composite endpoint that integrates multiple perioperative indicators, has gained broad acceptance in evaluating the quality of liver surgery (18, 19). Specific definitions vary, but numerous studies have consistently associated TO with significantly improved survival (19–21). This study further confirms that achieving TO markedly improves both OS and RFS in patients with ICC. Using multivariate logistic regression, we identified CEA level, intraoperative blood loss, ALB, BMI, and neoadjuvant therapy as independent factors affecting TO attainment. A prognostic nomogram that incorporates TO and other independent factors demonstrated favorable discrimination and calibration in internal validation.

A key finding of this study is that neoadjuvant therapy—particularly regimens that combine chemotherapy with targeted or immunotherapy—is a positive predictor for TO. Pawlik et al. analyzed patients with ICC who underwent surgery between 1993 and 2015 and revealed that neoadjuvant chemotherapy was associated with a lower TO rate (p = 0.04) (19). Separately, another study that incorporated data from 2000 to 2022 indicated that neoadjuvant therapy served as a negative factor for TO (22). This discrepancy may stem from evolving treatment paradigms. The majority of patients (77.19%) in our cohort received combination neoadjuvant therapy. Compared with chemotherapy alone, targeted and immunotherapies provide greater precision and an improved safety profile, which potentially facilitates tumor downstaging, reduces micrometastatic burden, and preserves liver function and overall condition, thereby enabling safer and more complete resection (26, 27). High-level clinical evidence, such as the KEYNOTE-966 and TOPAZ-1 phase III trials, which demonstrated survival benefits of immunotherapy–chemotherapy combinations in advanced biliary tract cancer, further support this perspective (28, 29). A phase II study by Guoming Shi et al. reported promising efficacy and safety for tislelizumab in combination with lenvatinib and GEMOX (gemcitabine and oxaliplatin) in the neoadjuvant setting (9). Further, neoadjuvant immunotherapy has been associated with higher TO rates in other malignancies, including esophageal cancer (23), which corroborates our findings. Notably, novel formulations (e.g., nab-paclitaxel and liposomal irinotecan) may improve both efficacy and safety by optimizing drug delivery (30, 31), potentially leading to better perioperative outcomes, even within the same class of chemotherapeutic agents.

Although neoadjuvant therapy significantly improved textbook outcomes for surgical patients in this study, some patients still experienced rapid disease progression shortly after treatment. In these cases, the tumors were often highly resistant to treatment and exhibited aggressive biological behavior. Preoperative disease progression can avoid unnecessary and ineffective surgery, thereby reducing the risk of early postoperative recurrence (32). This further emphasizes the clinical value of neoadjuvant therapy, as it aids in identifying patients who are unlikely to benefit from surgery and supports more informed decision-making.

Other factors associated with TO hold clinical relevance. CEA elevation is a biomarker of aggressive tumor biology and generally indicates an increased tumor burden, which may consequently elevate the complexity of surgical intervention (33, 34). ALB and BMI reflect nutritional and baseline status. Increased ALB levels have been associated with a higher rate of TO (35). Some studies have indicated that a BMI of ≥30 kg/m² may negatively affect TO (36); however, in the present cohort, the mean BMI was 22.88 kg/m², and no patient reached a BMI of 30 kg/m². These findings indicate that population characteristics, regional variations, and other factors may influence the association between BMI and TO (37, 38), which warrants further investigation for clarity. Intraoperative blood loss is a critical determinant of TO, closely associated with perioperative transfusion, which is itself associated with increased infection risk and worse survival (39).

The prognostic model developed herein integrated TO, PT, lymph node metastasis, and adjuvant therapy, which demonstrates good predictive performance. The role of adjuvant therapy (e.g., capecitabine and S1) is well-established from phase III trials such as BILCAP and JCOG1202 (40, 41). Lymph node metastasis is a recognized poor prognostic factor, and adequate dissection (≥6 nodes) is crucial for accurate staging (42). Prolonged PT may indicate impaired liver function, tumor burden, and systemic inflammation (43).

Compared with previous studies, our work highlights the evolving landscape of neoadjuvant therapy, particularly the positive impact of combination targeted/immunotherapy on TO. The inclusion of variables spanning preoperative patient status, tumor features, surgical details, and perioperative treatments improves the model’s comprehensiveness and utility. Limitations include its single-center, retrospective design, limited sample size, and potential selection bias. Thus, validation through multicenter, larger cohorts is warranted. Heterogeneity in neoadjuvant regimens (e.g., drug combinations, cycles, and molecular subtypes) may affect outcomes (44, 45), requiring subgroup analyses with expanded samples. Furthermore, the lack of a globally standardized TO definition and the potential influence of institutional policies and recovery pathways on metrics, such as postoperative length of stay (46, 47), underscore the need for consensus-building.

This study confirms that achieving TO significantly improves survival in patients with ICC and identifies CEA, intraoperative blood loss, ALB, BMI, and neoadjuvant therapy as independent influencing factors. Notably, this study is the first to report that neoadjuvant regimens that combine chemotherapy with targeted or immunotherapy improve TO attainment, which provides crucial evidence for neoadjuvant therapy strategies in ICC. The developed nomogram demonstrates robust prognostic predictive ability, thereby offering a practical tool for individualized clinical decision-making and surgical quality assessment.