Histopathological subtypes independently predict survival in ampullary carcinoma: a contemporary SEER nomogram study

Introduction

Ampullary adenocarcinoma (AAC), emerging from the ampullary complex distal to the bifurcation of the distal common bile duct and the pancreatic duct, represents a rare yet distinctive category of gastrointestinal malignancies, accounting for only 0.2% of all such cancers (1-3). Its rarity contributes to the challenges and complexities in both diagnosis and management. In 1994, Kimura et al. provided a pivotal classification of AAC into two histologic types: intestinal and pancreaticobiliary (PB), noting significant prognostic differences between them (4). This classification has been crucial for understanding the varied outcomes of these malignancies, as PB-AAC typically presents a poorer prognosis compared to intestinal AAC (5,6).

These histologic subtypes not only differ in their cellular origin but also in their clinical management (5-7). The intestinal type frequently resembles colon cancer (e.g., driven by APC mutations) and originates from the intestinal epithelium overlying the ampulla, evolving through an adenoma-dysplasia-adenocarcinoma sequence. In contrast, the PB-AAC originates from the endothelium of the distal common bile duct and distal pancreatic intraepithelial neoplasia in an analogous dysplasia-adenocarcinoma sequence, often with KRAS/TP53 mutations (8,9). According to the latest National Comprehensive Cancer Network (NCCN) guidelines (Version 2.2026), the treatment for intestinal AAC follows protocols similar to those for colon cancer, whereas PB-AAC and mixed types are treated following the guidelines for biliary tract cancers and pancreatic adenocarcinoma. Despite these distinctions, the American Joint Committee on Cancer (AJCC) tumor-node-metastasis (TNM) staging for AAC does not currently differentiate among these subtypes, employing a uniform staging system that fails to accurately reflect the prognosis for these diverse histologic types (10,11).

The prognostic factors for AAC remain a topic of active debate. Established adverse prognostic factors include advanced tumor stages (pT3–4), lymph node metastasis, and poor differentiation (12-14). Yet, the majority of studies exploring histopathologic predictors of survival in AAC are limited by relatively small sample sizes, leading to inconsistencies and a lack of reliable data to inform clinical, therapeutic, and prognostic strategies (15-17). These discrepancies underscore the need for more robust research to clarify the prognostic implications of intestinal and PB subtypes (2,3,18).

While previous studies, including some utilizing the Surveillance, Epidemiology, and End Results (SEER) database, have investigated prognostic factors in AAC, several limitations persist (12,13,19). These often include reliance on older data, smaller cohorts with definitive subtyping, a primary focus on overall survival (OS) rather than cancer-specific survival (CSS), or the lack of validated nomograms that specifically incorporate histopathological subtypes as key predictor variables for CSS. The current AJCC staging system also notably fails to differentiate outcomes based on these biologically distinct subtypes. We hypothesize that subtypes independently predict CSS, enabling a superior nomogram. Therefore, this study leverages the most recent and extensive SEER 17 registry data (2004–2021, N=726 typed cases) to provide a robust, contemporary analysis of the impact of histopathological subtypes on CSS. We specifically aimed to: (I) compare clinicopathological features and CSS between intestinal and PB/mixed subtypes; (II) identify independent prognostic factors for CSS in this large cohort; and (III) develop and validate one of the first prognostic nomograms specifically incorporating histopathological subtypes alongside other key clinicopathological variables to provide a more accurate, individualized prediction of CSS for patients with AAC. Through this approach, we aim to refine prognostic assessment and provide a practical tool to inform personalized management strategies for this challenging malignancy. We present this article in accordance with the TRIPOD reporting checklist (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0544/rc).

Methods

Patient selection

Patients for this study were selected from the SEER 17 Registries database (2000–2021), which is maintained by the National Cancer Institute of United States and released in April 2024 (available at: https://seer.cancer.gov/). The inclusion criteria for this study were stringently defined to ensure the selection of patients specifically diagnosed with primary AAC, with definitive histopathological classification. Eligible cases were identified based on the “Primary Site” coded as “C24.1-Ampulla of Vater”. Histological types were specified using the “ICD-O-3 Hist/behav, malignant” codes, which included “8144/3: Adenocarcinoma, intestinal type”, “8163/3: Pancreatobiliary type carcinoma”, and “8255/3: Adenocarcinoma with mixed subtypes”. Only patients diagnosed between the years 2004 and 2021 were included to capture recent trends and improvements in medical care. Exclusion criteria were applied. Cases coded as “8140/3: Adenocarcinoma, NOS (not otherwise specified)” were excluded because the NOS designation indicates that histopathological subtypes could not be determined, which conflicts with our primary research objective of evaluating subtype-specific prognosis. Patients with incomplete staging information (TX, NX, or MX) were also excluded. The detailed selection process is presented in Figure S1 and the Results section. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Variables definition

In this study, CSS was selected as the primary endpoint rather than OS to assess the impact of clinical factors more accurately on the prognosis of AAC, minimizing non-cancer mortality bias in elderly cohorts (median age >65 years). CSS is defined as the duration from the date of diagnosis to the date of death attributable specifically to AAC, excluding deaths caused by other factors such as heart diseases, pulmonary diseases, accidents, or other cancers. Our focus on CSS is particularly pertinent when comparing distinct biological subtypes like intestinal and PB/mixed, as it allows for a clearer assessment of tumor-intrinsic behavior and treatment efficacy related specifically to the cancer, minimizing confounding by non-cancer related mortality which can differ among patient groups or over long follow-up periods. We grouped pancreatobiliary and mixed subtypes together, a decision supported by sensitivity analyses showing no significant survival difference between pure pancreatobiliary and mixed tumors (log-rank P=0.42, Table S1), which was consistent with the classification approach outlined in NCCN guidelines (20,21).

The staging of AAC was based on the AJCC system, which underwent revisions during the study period. In SEER database, for patients diagnosed between 2004 and 2017, the AJCC 6th and 7th edition staging system were utilized. Subsequently, patients diagnosed from 2018 onward were staged according to the AJCC 8th edition. To maintain consistency in our analysis, T, N, and M staging for patients assessed using the AJCC 6th and 8th edition were converted to align with the criteria specified in the AJCC 7th edition. This approach ensured uniformity across all cases in the final analyses, regardless of the initial staging edition.

Regarding pathological grading, the categorization of tumor differentiation also saw changes over the study period. For diagnoses made from 2004 to 2017, the pathological grades were classified as well differentiated (Grade I), moderately differentiated (Grade II), poorly differentiated (Grade III), and undifferentiated (Grade IV). However, for patients diagnosed after 2018, the grading system was simplified to three levels: Grade I, Grade II and Grade III in SEER database. Consistent with SEER staging manual recommendations, pathological Grade IV assigned between 2004 and 2017 was converted to Grade III in our analyses to harmonize grading criteria across the entire study period.

Statistical analysis

Categorical variables are presented as frequencies and proportions. Categorical data were compared using the Chi-squared test. Univariate and multivariate Cox proportional hazard models were used to explore the association among various predictive variables and CSS, with proportional hazards assumption verified via Schoenfeld residuals. Median CSS was calculated using Kaplan-Meier curves, with subgroups being compared using the log-rank test. The study population was divided into a training set and a validation set in a ratio of 7:3 using the createDataPartition function from the caret package in R version 4.4.0 (R Foundation for Statistical Computing, Vienna, Austria), stratified by subtypes and stages to ensure balance. The nomogram was developed using the rms package, with backward stepwise selection for variables and 1,000 bootstrap resamples for internal validation to address overfitting. The predictive capacity of the nomogram was assessed using Harrell’s C-index (the concordance statistic, or C-statistic), which estimates the probability between the observed and predicted CSS, along with optimism-corrected values. Time-dependent receiver operating characteristic (ROC) curves were generated using the survival ROC package, and decision curve analysis (DCA) evaluated clinical utility for CSS at 1, 3, and 5 years in both cohorts. SPSS version 26.0 (IBM Corporation, Armonk, NY, USA) was used to conduct all statistical analyses. A P value <0.05 was considered statistically significant.

Results

Patient characteristics

The initial SEER database query identified 9,505 patients with primary site coded as “C24.1-Ampulla of Vater“ diagnosed between 2004 and 2021. Among these, 7,521 cases (79.1%) were coded as “8140/3: Adenocarcinoma, NOS”, indicating that histopathological subtypes could not be determined from available pathology reports. These cases were excluded to ensure that all included patients had definitive subtype classification. An additional 1,237 cases with other histological types (e.g., neuroendocrine tumors, squamous cell carcinomas) were also excluded. The remaining 747 patients had definitive histopathological classification: intestinal type (8144/3), pancreatobiliary type (8163/3), or mixed subtype (8255/3). After excluding 21 patients with incomplete staging information (TX, NX, or MX), the final study cohort comprised 726 patients (Figure S1). Of these, 385 patients (53.0%) had intestinal-type tumors, and 341 patients (47.0%) had pancreatobiliary or mixed-type tumors. The patient demographics and baseline characteristics are summarized in Table 1.

Regarding tumor characteristics, the PB/mixed type group had a significantly higher proportion of T3 tumors (57.5% vs. 34.5%, P<0.001) and lymph node involvement (62.8% vs. 51.4%, P=0.002) compared to the intestinal type group. The PB/mixed type group also had a higher proportion of poorly differentiated tumors (35.8% vs. 19.2%, P<0.001) and a lower proportion of well-differentiated tumors (6.2% vs. 11.4%, P<0.001) than the intestinal type group.

In contrast to other cancers of the PB system, AACs are more likely to be diagnosed in the early stage and to undergo curative resection. The proportion of patients who underwent surgery was similar between the two groups (93.3% vs. 96.4%, P=0.06). However, the PB/mixed type group had a significantly higher proportion of patients receiving chemotherapy (61.6% vs. 48.8%, P<0.001) compared to the intestinal type group. The proportion of patients receiving radiation therapy was not significantly different between the two groups (9.7% vs. 13.8%, P=0.09).

CSS stratified by histopathologic subtype

The Kaplan-Meier survival curves comparing CSS between intestinal type and PB/mixed type AACs across different AJCC stages are presented in Figure 1. For all patients, the intestinal type group had significantly better CSS compared to the PB/mixed type group [hazard ratio (HR) =1.60, 95% confidence interval (CI): 1.22–2.09, logrank P<0.001]. The median survival time for the intestinal type was 168 months (95% CI: 83 to unreached), whereas for the PB/mixed type group, it was 39 months (95% CI: 33–62). When stratified by AJCC stage, the intestinal type group consistently showed better CSS than the PB/mixed type group, although the difference was not statistically significant in stage I (log-rank P=0.08), stage IIA (log-rank P=0.04), stage IIB (log-rank P=0.86), and stage IV (log-rank P=0.16) disease. However, in stage III disease, the intestinal type group exhibited significantly better CSS compared to the PB/mixed type group (log-rank P=0.02). The median survival time was 44 months (95% CI: 30–63) for the intestinal type group and 27 months (95% CI: 23–34) for the PB/mixed type group. Multivariate-adjusted HRs confirm subtypes independence (see Table 2).

Figure 1 Kaplan-Meier survival curves for patients with ampullary carcinoma stratified by histopathologic subtypes according to a AJCC stages. (A) All patients, (B) stage I patients, (C) stage IIA patients, (D) stage IIB patients, (E) stage III patients, (F) stage IV patients. AJCC, American Joint Committee on Cancer.

Risk factors associated with CSS

Univariate and multivariate Cox proportional hazards analyses were performed to identify risk factors associated with CSS (Table 2). In the univariate analysis, histopathological type, T stage, N stage, M stage, AJCC stage, grade, and surgery were significantly associated with CSS (all P<0.05). Patients with intestinal type AAC had a significantly lower risk of cancer-specific mortality compared to those with PB/mixed type (HR =0.55, 95% CI: 0.41–0.72, P<0.001). Advanced T stage (T3–T4), lymph node involvement (N1), distant metastasis (M1), and higher AJCC stage (IIA–IV) were also associated with worse CSS (all P<0.05). Poorly differentiated tumors was associated with a higher risk of cancer-specific mortality compared to well-differentiated tumors (P=0.004).

In the multivariate analysis, histopathological type, T stage, N stage, M stage, surgery, and chemotherapy remained independent predictors of CSS (all P<0.05). Patients with intestinal type AAC had a 40% lower risk of cancer-specific mortality compared to those with PB/mixed type (HR =0.60, 95% CI: 0.45–0.81, P<0.001). Advanced T stage (T3–T4), lymph node involvement (N1), and distant metastasis (M1) were also independently associated with worse CSS (all P<0.05). Interestingly, while chemotherapy was not significantly associated with CSS in the univariate analysis (P=0.33), it emerged as an independent predictor in the multivariate analysis. Patients who did not receive chemotherapy or had unknown chemotherapy status had a 80% higher risk of cancer-specific mortality compared to those who received chemotherapy (HR =1.80, 95% CI: 1.32–2.44, P<0.001).

Variable selection for nomogram

Variable selection integrated statistical significance with clinical relevance. Although sex reached statistical significance in univariate analysis (P=0.02), it was not included in multivariate analysis because sex has not been consistently identified as an independent prognostic factor and the observed difference likely reflects confounding by other variables. Conversely, age did not reach significance in univariate analysis (P=0.21) but was included given its well-established prognostic importance across gastrointestinal malignancies.

Univariable analysis initially identified tumor grade as a significant factor. However, tumor grade was excluded from the final nomogram due to significant multicollinearity with histopathological subtype. PB type tumors are more frequently poorly differentiated, while intestinal-type tumors tend to be well-to-moderately differentiated. Since histopathological subtype was our primary variable of interest and already captured grade-related prognostic information, retaining both would introduce redundancy and compromise model stability.

The final nomogram incorporated seven independent prognostic factors (Table 2): age, histopathological subtype, T stage, N stage, M stage, surgery, and chemotherapy.

Prognostic nomogram and validation

The study population was randomly divided into a training set and a validation set in a ratio of 7:3. The patient demographics and baseline characteristics were comparable between the two cohorts (Table S2). A nomogram was developed based on the results of the multivariate Cox regression analysis to predict the CSS of patients with AAC (Figure 2, Table S3). The C-index of the nomogram was 0.751 (95% CI: 0.708–0.794), indicating good discriminative ability (compared to AJCC staging C-index ~0.65 from literature) (18).

Figure 2 Nomogram for predicting the 1-, 3-, and 5-year cancer-specific survival of patients with ampullary carcinoma. M, metastasis; N, node; T, tumor.

The predictive performance of the nomogram was assessed using ROC curves in both the training and internal validation cohorts (Figure 3). The area under the curve (AUC) values for predicting 1-, 3-, and 5-year CSS were 0.800, 0.787, and 0.799 in the training cohort, and 0.837, 0.823 and 0.812 in the internal validation cohort, respectively (superior to AJCC staging) (19). Calibration plots were used to evaluate the agreement among the predicted and observed CSS probabilities at 1, 3, and 5 years in both the training (Figures S2-S4) and validation cohorts (Figures S5-S7). DCA was performed to assess the clinical utility of the nomogram (Figures S8-S13, Table S4). These results suggest that the nomogram has good predictive accuracy for CSS in AAC patients.

Figure 3 ROC curves for the training (A) and validation (B) cohorts, predicting 1-, 3-, and 5-year cancer-specific survival. AUC, area under the curve; CI, confidence interval; ROC, receiver operating characteristic.

Discussion

Leveraging a large (N=726) contemporary SEER cohort, this study provides robust evidences that histopathological subtype independently predicts CSS in AAC, aligning with recent multicenter studies (22). Importantly, the 2026 NCCN guidelines recommend routine subtyping, yet conflicting evidence exists regarding its independent prognostic value. Our study addresses this gap, demonstrating that subtype retains independent significance (adjusted HR =0.60) even after rigorous adjustment.

The magnitude of our effect (HR for PB/mixed vs. intestinal of 1.67) is remarkably concordant with the Shin et al. meta-analysis (pooled HR =1.79) (6). While Pellegrini et al. found subtype not independent in a small surgical cohort, our larger population-based CSS focus likely enhances detection of tumor-intrinsic effects (22). This apparent discrepancy may reflect differences in study populations (surgically resected single-center cohort with 97 patients vs. population-based SEER cohort with 726 patients), outcome measures (OS vs. CSS), and the inclusion of non-resected or advanced-stage patients in our analysis. Importantly, our study’s focus on CSS—which excludes the substantial proportion (37.4%) of non-cancer deaths observed in this elderly cohort—may have enhanced the ability to detect subtype-specific, tumor-intrinsic prognostic effects. Additionally, Ozyigit Buyuktalanci et al. reinforced that pancreatobiliary phenotype clusters with adverse anatomic locations (ampullary-ductal/NOS)—precisely the groups associated with poorer OS (P=0.01) and higher rates of lymphovascular invasion and perineural invasion (23).

The biological aggressiveness of PB subtype, such as advanced tumor stage, lymph node metastasis, and poor differentiation, is associated with a higher frequency of adverse prognostic factors (24-26). Additionally, the PB subtype has been linked to a more aggressive tumor biology, with a higher prevalence of mutations in genes involved in cell cycle regulation and DNA repair, such as TP53 and SMAD4 (20,27). On the other side, the extremely prolonged median CSS observed in the intestinal subtype (168 months) should be interpreted with caution. This prolonged survival likely reflects the favorable natural history of intestinal-type AAC, but may also be influenced by evolving surveillance strategies, salvage treatments, and survivorship effects over the extended study period [2004–2021]. The unreached upper 95% CI for the median survival further underscores the need for cautious interpretation. These results further strengthen the consistency of clinicopathological features of histological phenotypes and the possibility of histological classification as a prognostic factor for AAC patients.

The 2026 NCCN Guidelines explicitly state that systemic therapy type may depend on the histologic subtype. However, the optimal selection of patients for adjuvant therapy remains controversial. A study on recurrence patterns after radical surgery identified lymphovascular invasion and histologic differentiation as significant predictors of early recurrence, and emphasized that patients with pancreatic invasion (T3) had significantly higher recurrence rates (28). These findings align with our observation that T3 stage was independently associated with worse CSS (HR =1.93) and support the rationale for more aggressive adjuvant strategies in patients with deep pancreatic invasion. Recent real-world data provide preliminary evidence that adjuvant chemoimmunotherapy may confer significant OS benefit in patients with PB subtype AAC harboring high-risk recurrence factors compared to surgery alone or adjuvant chemotherapy (29). Pellegrini et al. demonstrated that adjuvant chemotherapy was independently associated with improved OS specifically in the PB subtype, whereas no benefit was observed in the intestinal subtype (22). This finding corroborates our observation that PB tumors exhibit more aggressive biology and may derive greater benefit from systemic therapy. The TOPAZ-1 trial demonstrating survival benefit with durvalumab in biliary tract cancers, may extend to PB-AAC (30). While our findings support the clinical relevance of histopathological subtyping, treatment implications remain speculative and require prospective validation. Our SEER-based data cannot directly inform treatment decisions, and the observed survival differences between subtypes should not be interpreted as evidence for specific therapeutic superiority (31).

Despite NCCN guidelines recommending reporting of histologic subtypes, the distinction between intestinal and PB subtypes is made only in a minority of patients even in developed countries. From a Dutch nationwide pathology database of 5,246 AAC patients, merely 19.6% of patients received subtyping in daily clinical practice (32). Although the proportion had increased in recent years [2018–2020], only 37% of patients had a documented distinction between intestinal and PB subtypes. Worldwide efforts are required to standardize the pathological distinction of the various AAC subtypes, as the interobserver variability (concordance rates 70–85%) and overlapping immunophenotypes in 20–30% of cases remain challenges (18,33). Encouragingly, a meta-analysis suggested that routine histologic classification is feasible even in resource-limited settings without advanced immunohistochemical capabilities (6). On the other hand, a genomic classifier was developed as a new standard to better delineate the nature of AAC, defining biologically distinct phenotypes with greater accuracy than standard histologic classification (34).

Several prognostic models have been previously proposed for AAC. Li et al. developed a prognostic nomogram specifically in non-metastatic AAC patients after surgery, achieving a C-index of 0.70 (35). Tang et al. reported a SEER-based nomogram with a C-index of 0.78 (19). More recently, Huang et al. applied machine learning methods to predict survival in AAC using a large SEER cohort, achieved the highest C-index of 0.731 (36). Huang’s model acted as a “black box” with limited interpretability and cannot produce a simple scoring system or visual tool for bedside use, making it difficult for clinicians to directly apply in individual patient counseling. All these three models did not include histopathological subtype as a predictor. In contrast, our nomogram offers a transparent, user-friendly scoring system that can be easily calculated at the point of care, while specifically incorporating the clinically critical intestinal versus PB/mixed distinction. Besides, our model includes M1 patients (7.3% of our cohort), providing a more comprehensive tool applicable across all disease stages.

Several limitations exist. First, the retrospective nature of the study may introduce potential biases. The SEER database lacks detailed information on disease recurrence, specific chemotherapy regimens, and molecular characteristics such as microsatellite instability status—the latter being particularly relevant given that MSI-H tumors may respond differently to immunotherapy (29). Second, excluding 79.1% NOS cases introduces selection bias; our cohort represents patients with detailed pathological characterization (Table S5). Third, the year-of-diagnosis imbalance (PB/mixed more after 2016) reflects changing pathological practices, though era-adjusted analysis did not alter the subtype effect (HR =0.62). Fourth, chemotherapy status, although identified as an independent predictor in our multivariate model, should be interpreted with caution due to potential ‘indication bias’. In the SEER database, the decision to administer chemotherapy is not randomized and is likely influenced by unmeasured confounders including performance status, comorbidity burden, functional status, and patient preferences. Our model adjusts for age but cannot account for these unobserved factors. Therefore, the observed survival benefit associated with chemotherapy may partially reflect selection of healthier patients into treatment rather than a purely causal effect. Fifth, the lack of central pathology review for histopathological subtyping in SEER is a limitation, although the large sample size may mitigate subtype misclassification to some extent. Finally, external validation in an independent cohort from a different institution or geographic region is necessary before clinical implementation.

Conclusions

The PB/mixed subtypes of ampullary carcinoma are associated with significantly worse CSS compared to the intestinal subtype. Our validated nomogram provides a practical tool for individualized risk stratification, reinforcing the clinical necessity of routine histopathological subtyping as emphasized by NCCN guidelines.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the TRIPOD reporting checklist. Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0544/rc

Peer Review File: Available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0544/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jgo.amegroups.com/article/view/10.21037/jgo-2026-0544/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Gutmans J, Friedlaender A, Mechahougui H. Rare, Yet Targetable: New Perspectives on Ampullary Carcinomas. Int J Mol Sci 2026;27:1597. [Crossref] [PubMed]
2. Hautefeuille V, Williet N, Turpin A, et al. Ampullary tumors: French Intergroup Clinical Practice Guidelines for diagnosis, treatments and follow-up (TNCD, SNFGE, FFCD, UNICANCER, GERCOR, SFCD, SFED, ACHBT, AFC, SFRO, RENAPE, SNFCP, AFEF, SFP, SFR). Dig Liver Dis 2024;56:1452-60. [Crossref] [PubMed]
3. Abu Hilal M, Uijterwijk BA, Lemmers DHL, et al. The Brescia International Multidisciplinary Consensus Guidelines on the Optimal Pathology Assessment and Multidisciplinary Pathways of Non-Pancreatic Neoplasms in and Around the Ampulla of Vater (PERIPAN). United European Gastroenterol J 2025;13:1048-68. [Crossref] [PubMed]
4. Kimura W, Futakawa N, Yamagata S, et al. Different clinicopathologic findings in two histologic types of carcinoma of papilla of Vater. Jpn J Cancer Res 1994;85:161-6. [Crossref] [PubMed]
5. Moekotte AL, Lof S, Van Roessel S, et al. Histopathologic Predictors of Survival and Recurrence in Resected Ampullary Adenocarcinoma: International Multicenter Cohort Study. Ann Surg 2020;272:1086-93. [Crossref] [PubMed]
6. Shin DW, Kim S, Jung K, et al. Impact of histopathological type on the prognosis of ampullary carcinoma: A systematic review and meta-analysis. Eur J Surg Oncol 2023;49:306-15. [Crossref] [PubMed]
7. Pein RJ, Illyés I, Kenessey I, et al. Immunohistochemical profiling of ampullary carcinoma: Molecular subtypes, prognostic markers, and the role of Claudin-18. Virchows Arch 2026; Epub ahead of print. [Crossref]
8. Huang X, Hu W, Wu Y, et al. Transcriptomic-based molecular classification of ampullary adenocarcinoma. Cancer Lett 2026;657:218701. [Crossref] [PubMed]
9. Nappo G, Funel N, Laurenti V, et al. Ampullary Cancer: Histological Subtypes, Markers, and Clinical Behaviour-State of the Art and Perspectives. Curr Oncol 2023;30:6996-7006. [Crossref] [PubMed]
10. Chiorean EG, Chiaro MD, Tempero MA, et al. Ampullary Adenocarcinoma, Version 1.2023, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 2023;21:753-82. [Crossref] [PubMed]
11. Chun YS, Pawlik TM, Vauthey JN. 8th Edition of the AJCC Cancer Staging Manual: Pancreas and Hepatobiliary Cancers. Ann Surg Oncol 2018;25:845-7.
12. Wang SJ, Li YF, Liao S, et al. Proposal of a new T-stage classification system for ampullary carcinoma based on Surveillance, Epidemiology and End Result (SEER) database. Hepatobiliary Pancreat Dis Int 2021;20:568-73. [Crossref] [PubMed]
13. Feng J, Wu R, Zhang G, et al. Prognostic importance of numbers of retrieved lymph nodes and positive lymph nodes for ampulla of vater cancer (AVC) in 2347 patients from the Surveillance, Epidemiology, and End Results (SEER) database. PLoS One 2021;16:e0244987. [Crossref] [PubMed]
14. Anand U, Arora A, Bhadani P, et al. Recurrence patterns and predictors after pancreaticoduodenectomy for ampullary carcinoma. HPB (Oxford) 2026;28:657-65. [Crossref] [PubMed]
15. Zimmermann C, Wolk S, Aust DE, et al. The pathohistological subtype strongly predicts survival in patients with ampullary carcinoma. Sci Rep 2019;9:12676. [Crossref] [PubMed]
16. Misyura M, Selenica P, Ozcan K, et al. Germline Pathogenic Variants in Homologous Recombination Pathway Genes Are Frequent in Pancreatobiliary Ampullary Carcinoma. JCO Precis Oncol 2026;10:e2500596. [Crossref] [PubMed]
17. Furtado I, Gião N, Gonçalves A, et al. Ampullary Carcinoma: Prognostic Factors and a Literature Review. Cancers (Basel) 2026;18:707. [Crossref] [PubMed]
18. Reid MD, Balci S, Ohike N, et al. Ampullary carcinoma is often of mixed or hybrid histologic type: an analysis of reproducibility and clinical relevance of classification as pancreatobiliary versus intestinal in 232 cases. Mod Pathol 2016;29:1575-85. [Crossref] [PubMed]
19. Tang N, Chen ZY, Yang Z, et al. Development and verification of prognostic nomogram for ampullary carcinoma based on the SEER database. Front Oncol 2023;13:1197626. [Crossref] [PubMed]
20. Patel MA, Kratz JD, Carlson AS, et al. Molecular Targets and Therapies for Ampullary Cancer. J Natl Compr Canc Netw 2024;22:e237051. [Crossref] [PubMed]
21. Fabregat-Franco C, Castet F, Castillo G, et al. Genomic profiling unlocks new treatment opportunities for ampullary carcinoma. ESMO Open 2025;10:104480. [Crossref] [PubMed]
22. Pellegrini R, Del Chiaro V, Franco SR, et al. Histopathologic Subtype as a Determinant of Prognosis and Chemotherapy Benefit in Resected Ampullary Adenocarcinoma. J Hepatobiliary Pancreat Sci 2026;33:537-47. [Crossref] [PubMed]
23. Ozyigit Buyuktalanci D, Gun E, Dilek ON, et al. Histopathological and prognostic variability of ampullary tumors: A comprehensive study on tumor location, histological subtypes, and survival outcomes. Ann Diagn Pathol 2025;77:152476. [Crossref] [PubMed]
24. Akabane M, Kume A, Hashimoto M, et al. A Clinicopathological Exploration of Ampullary Adenocarcinoma. ANZ J Surg 2025;95:1764-73. [Crossref] [PubMed]
25. Park SJ, Shin K, Hong TH, et al. Histologic subtype-based evaluation of recurrence and survival outcomes in patients with adenocarcinoma of the ampulla of Vater. Sci Rep 2023;13:16547. [Crossref] [PubMed]
26. Narita M, Hatano E, Kitamura K, et al. Identification of patients at high risk for recurrence in carcinoma of the ampulla of Vater: Analysis in 460 patients. Ann Gastroenterol Surg 2024;8:190-201. [Crossref] [PubMed]
27. Liang H, Zhu Y, Wu YK. Ampulla of Vater carcinoma: advancement in the relationships between histological subtypes, molecular features, and clinical outcomes. Front Oncol 2023;13:1135324. [Crossref] [PubMed]
28. Zhang ZY, Liu DW, Hao DS, et al. Predictors and Recurrence Patterns After Radical Surgery in Ampulla of Vater Cancer: Comparative Analysis Between Early and Late Recurrence. Front Surg 2022;9:833373. [Crossref] [PubMed]
29. Xue J, Lu X, Shi X, et al. Real-world experience of postoperative adjuvant chemoimmunotherapy in patients with pancreatobiliary subtype ampullary adenocarcinoma. Oncologist 2025;30:oyaf104. [Crossref] [PubMed]
30. Oh DY, He AR, Bouattour M, et al. Durvalumab or placebo plus gemcitabine and cisplatin in participants with advanced biliary tract cancer (TOPAZ-1): updated overall survival from a randomised phase 3 study. Lancet Gastroenterol Hepatol 2024;9:694-704. [Crossref] [PubMed]
31. Uijterwijk BA, Lemmers DHL, Klei DS, et al. Adjuvant Chemotherapy After Curative Intent Resection of Ampullary Cancer (ADAPTA): Study Protocol for a Pan-European Prospective Multicenter Double Single Arm Trial. Am J Clin Oncol 2026;49:342-9. [Crossref] [PubMed]
32. de Bakker J, Sommeijer D, Besselink M, et al. The use of histopathological subtyping in patients with ampullary cancer: a nationwide analysis. World J Surg Oncol 2022;20:406. [Crossref] [PubMed]
33. Abraham UM, Ramkumar S. Correlation between Immunohistochemical and Histomorphological Features of Ampullary Carcinomas: A Study on 72 Cases from a Tertiary Health Care Center. Gastroenterol Res Pract 2020;2020:2080351. [Crossref] [PubMed]
34. Ecker BL, Seier K, Eckhoff AM, et al. Genome-Derived Ampullary Adenocarcinoma Classifier and Postresection Prognostication. JAMA Surg 2024;159:1365-73. [Crossref] [PubMed]
35. Li HB, Zhao FQ, Zhou J. Prognostic Nomogram for Disease-Specific Survival in Patients with Non-metastatic Ampullary Carcinoma After Surgery. Ann Surg Oncol 2019;26:1079-85. [Crossref] [PubMed]
36. Huang T, Huang L, Yang R, et al. Machine learning models for predicting survival in patients with ampullary adenocarcinoma. Asia Pac J Oncol Nurs 2022;9:100141. [Crossref] [PubMed]