Introduction
Pancreatic ductal adenocarcinoma (PDAC) and its variants and subtypes account for approximately 90% of adult pancreatic tumors.1,2 Surgical resection remains the only potentially curative treatment modality.3 The evaluation of pathologic features—such as extent of tumor invasion and margin status—is critical for prognostic assessment. Accurate evaluation depends on systematic and standardized sampling and reporting of pancreatic resection specimens. Such standardization not only facilitates individualized and high-quality patient management but also provides reliable and comparable pathologic data for clinical research. The increasing adoption of neoadjuvant therapy in pancreatic cancer may induce tumor regression, improve resectability and R0 resection rates, and thereby potentially improve disease-free survival and overall survival.4-10 However, there is currently a lack of standardized guidelines for pathologic examination and reporting of pancreatic cancer specimens after neoadjuvant therapy worldwide. This lack of standardization leads to significant variability in the assessment of key pathologic parameters—such as the extent of tumor regression and margin status—across different institutions.11,12 Such inconsistency not only hampers accurate evaluation of treatment response but also limits the comparability of clinical studies and the accumulation of comparable evidence.
Beyond its role in diagnosis and treatment, standardized pathologic evaluation is also important for screening-related research and practice. The current lack of uniform criteria for diagnosing precursor lesions—such as pancreatic intraepithelial neoplasia (PanIN) and intraductal papillary mucinous neoplasm (IPMN)—hampers systematic surveillance, risk stratification, and the clinical validation of early-detection biomarkers in high-risk populations (e.g., individuals with familial pancreatic cancer or chronic pancreatitis). Consistent pathologic standards would enable uniform diagnosis of these lesions and provide a pathologic basis for risk stratification in screening cohorts. Furthermore, a standardized tumor regression grading system could support objective evaluation of the response to neoadjuvant therapy, inform subsequent clinical management, and promote personalized treatment. Therefore, developing and implementing standardized guidelines for pathologic processing and reporting of pancreatic cancer specimens may improve clinicopathologic diagnostic quality and support consistent early screening, high-risk population surveillance, and treatment response evaluation in pancreatic cancer.
To standardize pathologic diagnosis of pancreatic cancer in China, the National Clinical Research Center for Digestive Diseases (Shanghai), the Pancreatic Disease Committee of the Chinese Medical Doctor Association, and the Chinese Journal of Pancreatology led the development of this evidence-based guideline. These efforts involved multidisciplinary experts in pathology, surgery, internal medicine, radiology, evidence-based medicine, and guideline methodology. The process followed the World Health Organization (WHO) Handbook for Guideline Development,13 the Institute of Medicine’s definition of clinical practice guidelines,14 the Chinese Principles for Developing/Revising Clinical Guidelines (2022),15 and the WHO International Classification of Diseases (11th Edition),16 and incorporated the Appraisal of Guidelines for Research & Evaluation II (AGREE II) instrument and the Reporting Items for Practice Guidelines in Healthcare (RIGHT) statement.17,18 Through a systematic literature review, nationwide expert consultation, and a modified Delphi method featuring iterative voting and panel discussions, 11 key recommendations were developed for standardized specimen sampling and diagnostic reporting. Drafted and revised by the Department of Pathology of the First Affiliated Hospital of Naval Medical University, the finalized “Evidence-Based Guidelines for Standardized Pathologic Sampling and Diagnostic Reporting of Pancreatic Cancer in China” was registered on the International Practice Guidelines Registry Platform (PREPARE-2023CN491; www.guidelinesregistry.cn) . Using the Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology,19 evidence certainty was classified as high (A), moderate (B), low (C), or very low (D), and recommendations were classified as strong or weak.
Methods
The guideline was developed through a systematic evaluation of pathologic sampling and diagnostic reporting standards for pancreatic cancer, in accordance with the WHO Handbook for Guideline Development.13 This process integrated the Institute of Medicine’s definition of clinical practice guidelines,14 the Chinese Principles for Developing/Revising Clinical Guidelines (2022) issued by the Chinese Medical Association,15 and the WHO International Classification of Diseases (11th Revision, ICD-11).16 Methodological rigor was further ensured by using the AGREE II instrument and following the RIGHT statement.17,18
Supporting institutions of the guideline
The guideline was led by the National Clinical Research Center for Digestive Diseases (Shanghai), the Pancreatic Disease Committee of the Chinese Medical Doctor Association, and the Chinese Journal of Pancreatology. The Department of Pathology at the First Affiliated Hospital of Naval Medical University served as the primary implementing body. Methodological support was provided by the GRADE China Center. This guideline is registered on the International Practice Guidelines Registry Platform (http://www.guidelinesregistry.cn/ ) under registration number PREPARE-2023CN491.
Target users and beneficiaries of the guideline
The guideline is intended primarily for pathologists and members of multidisciplinary teams involved in pancreatic disease management; patients with pancreatic cancer are the main beneficiaries.
Organizational framework for guideline development
The organizational structure for guideline development comprises five core components: Principal Clinical Expert, Lead Methodologist, Guideline Consensus Committee, External Review Panel, and Working Taskforce. The multidisciplinary team includes specialists in pathology, internal medicine, surgery, oncology, medical imaging, evidence-based medicine, and guideline appraisal methodologies.
Declaration of relationships and interests
The development of this guideline strictly adhered to the WHO Conflict of Interest Policy and ethical standards for guideline development. All core contributors, including invited experts and consultants participating in guideline meetings, completed declarations of interest. Following rigorous assessment using the WHO-standardized assessment instrument, no direct conflicts of interest related to this guideline were identified.
Formulation and prioritization of key questions
The guideline taskforce conducted a systematic literature search aligned with evidence-based medicine principles to identify key issues in pancreatic cancer specimen processing and diagnostic reporting. The search included published guidelines, systematic reviews, and original studies and identified 11 preliminary pathologic assessment questions. Through two modified Delphi rounds with institutional and national experts, followed by in-depth deliberations, the final questions evaluated in this guideline are summarized in Table 1.
| Key questions | Recommendation
strength | Evidence
quality | Recommendations |
|---|
| 1 | How should resection margins and specimen surfaces be defined in pancreatic cancer specimens? | Strong | Moderate | • Pancreatoduodenectomy: margins include: proximal and distal gastrointestinal, pancreatic transection, SMA, and bile duct or hepatic duct; circumferential surfaces: anterior/posterior peripancreatic and SMV groove
• Distal pancreatectomy: true margin includes pancreatic transection; specimen surfaces include anterior/posterior surfaces.
• Note: SMV groove and anterior/posterior surfaces are not true surgical margins.
• Complete circumferential sampling of the entire margin and surface planes is required. |
| 2 | How should sampling methods be selected for pancreatic cancer surgical specimens? | Strong | Moderate | • Pancreatoduodenectomy: serial transverse sections perpendicular to the long axis of the duodenum (for pancreatic head carcinoma only; not for ampullary or common bile duct tumors)
• Distal pancreatectomy: serial sections through the maximum tumor diameter in the anteroposterior plane.
• After neoadjuvant therapy: complete sampling mandatory when assessing pCR. |
| 3 | How should the histologic types and grades of pancreatic cancer be evaluated? | Weak | Low | • Histologic classification: use WHO 5th edition
• Differentiation grading: use CAP criteria for PDAC
• With precursor lesions: specify whether ductal adenocarcinoma arose from a precursor lesion; report histologic subtype and differentiation. |
| 4 | How should lymphovascular invasion be evaluated? | Strong | Moderate | • Record lymphatic and vascular invasion separately
• For named vessels (celiac trunk, SMA, CHA, portal vein, SMV, splenic artery/vein): document invasion depth (adventitia, media, intima) |
| 5 | How should perineural invasion be evaluated? | Weak | Low | • Report perineural invasion status
• Distinguish between intrapancreatic and extrapancreatic involvement when possible |
| 6 | How should the R1 resection margin be evaluated in pancreatic cancer? | Weak | very low | • R1 is defined as a microscopic tumor-to-margin distance of ≤ 1 mm |
| 7 | How should T stage be assessed? | Weak | Low | • Use AJCC/UICC 8th edition staging system
• For IPMN/IOPN/ITPN/MCN-associated invasive carcinoma: stage according to invasive-component size only (not whole lesion)
• After neoadjuvant therapy: large-section sampling recommended; document measurement method |
| 8 | How should N stage be assessed? | Strong | Moderate | • Use AJCC/UICC 8th edition staging system (Table 3)
• Count lymph nodes with direct tumor invasion or metastasis as positive
• Examine at least 12 lymph nodes for pancreaticoduodenectomy specimens |
| 9 | How should M stage be assessed? | Strong | Moderate | • M0: no distant metastasis
• M1: distant metastasis present
• Peritoneal seeding or positive ascites: M1 |
| 10 | How should tumor regression grading (TRG) be assessed after neoadjuvant therapy? | Strong | Low | • Use CAP or MD Anderson Cancer Center (MDACC) TRG systems |
| 11 | Should background lesions of pancreatic cancer be assessed? | Weak | Low | • Assess background lesions: PanIN, IPMN, ITPN, IOPN, MCN, chronic pancreatitis
• These assessments support risk stratification and understanding of tumorigenesis |
Evidence retrieval
Systematic literature searches were conducted across four English databases (PubMed, Cochrane Library, Embase, Web of Science) and five Chinese databases (Wanfang Data, China National Knowledge Infrastructure, China Biology Medicine, VIP Database, Yimaitong). Additionally, searches were performed on key international clinical guideline websites, including the National Guideline Clearinghouse, the Scottish Intercollegiate Guidelines Network, the WHO, and the Guidelines International Network. Google was searched as a supplementary source. The search included records published before December 31, 2023, and was limited to English or Chinese publications. Furthermore, the reference lists of all included studies were searched to identify additional relevant literature.
Evidence screening and data extraction
Evidence for the 11 histopathologic questions was derived from three clinical guidelines, eight newly developed or updated systematic reviews, and 169 additional studies, including clinical trials, cohort studies, case-control studies, case series, and case reports. After establishing inclusion and exclusion criteria through consensus among the guideline working group, methodologists, and pathology experts, comprehensive training was provided to all working group members. The group then synthesized evidence screened by domain experts and submitted it to the Guideline Steering Committee. Literature screening and data extraction were performed independently by two investigators, with disagreements resolved through discussion or consultation with a third investigator.
Evidence quality assessment and grading
The methodological quality of systematic reviews and meta-analyses was evaluated using the A Measurement Tool to Assess Systematic Reviews tool.20 For the included guidelines, the methodological quality was evaluated using AGREE II.21 Methodologically sound and relevant evidence was updated and used to formulate the guideline. Otherwise, the guideline working group updated or conducted systematic reviews or used other evidence. The same databases were searched when systematic reviews were updated or conducted. When systematic reviews were updated or conducted, the Cochrane risk-of-bias tool was used to assess randomized controlled trials.22 The GRADE approach was used to assess evidence certainty and recommendation strength.23 According to GRADE, evidence certainty was classified as high (A), moderate (B), low (C), or very low (D), and recommendations as strong (1) or weak (2).
Formation of guideline recommendations
Based on the working group’s synthesis of national and international evidence, the first questionnaire was developed by selecting critical questions and terminology from pathology reports, and classifying evidence levels and recommendation grades. This questionnaire was then distributed to an expert panel for a Delphi-based survey via Wenjuanxing. Each recommendation within the survey was evaluated using a 5-point Likert scale (“Strongly Agree,” “Mostly Agree,” “Partially Agree,” “Mostly Disagree,” “Strongly Disagree”). Recommendations achieving ≥75% consensus from experts (combining “Strongly Agree” and “Mostly Agree” responses) were adopted as formal guideline recommendations. The final guideline comprises 11 evidence-based recommendations with corresponding rationales.
External review
After consensus was reached on the recommendations, the initial guideline draft was prepared and submitted to the External Review Panel for evaluation. The external reviewers comprised pathology experts, clinical practitioners, and guideline methodology specialists.
Funding sources and their utilization
Primary funding was provided by the National Natural Science Foundation of China (NSFC) and the Shanghai Municipal Science and Technology Commission Innovation Action Plan, covering research expenditures, material costs, and activities required to develop the guideline.
Guideline dissemination and implementation
After publication, the guideline will be disseminated through three primary channels: (1) multimedia campaigns, (2) promotion at relevant academic conferences, and (3) targeted training on the guideline for pathologists and clinicians engaged in pancreatic disease management.
Results
The guideline development process yielded 11 formal recommendations. These encompass four critical domains: margin definition protocols, specimen sampling strategies, histopathologic tumor assessment standards, and pancreatic background pathology evaluation.
How should resection margins and specimen surfaces be defined in pancreatic cancer resection specimens?
Recommendation 1: For pancreatoduodenectomy specimens, surgical margins include (1) proximal and distal gastrointestinal margins, (2) pancreatic transection margin, (3) superior mesenteric artery (SMA) margin, and (4) bile duct or hepatic duct margin; circumferential surfaces comprise the anterior peripancreatic surface, posterior peripancreatic surface, and superior mesenteric vein (SMV) groove surface. For distal pancreatectomy specimens, the true surgical margin is the pancreatic transection margin, and the anterior and posterior peripancreatic surfaces are circumferential surfaces. The SMV groove and anterior and posterior peripancreatic surfaces are not classified as true surgical margins.24-27 Complete circumferential sampling of all margin and surface planes is required.
Quality of evidence: Moderate
Strength of recommendation: Strong
To standardize the nomenclature of resection margins in specimens from pancreaticoduodenectomy and distal pancreatectomy, and clarify the definitions of the true surgical margins and their corresponding circumferential surfaces, this guideline establishes a standardized margin assessment protocol. This will not only support assessment of resection completeness (R0 status) for early-stage pancreatic cancer but also provide a standardized pathologic basis for assessing postoperative recurrence.
Resection margins and specimen surfaces in pancreatoduodenectomy specimens
(1) The SMA (retroperitoneal/uncinate) margin, representing the most prognostically critical surgical margin, comprises a 3–4-cm soft-tissue zone in direct contact with the superior mesenteric artery. As shown in Figure 1, this region contains dense perineural plexuses and lies close to the SMA; it is a frequent site of local recurrence at the uncinate tumor bed and therefore requires precise margin assessment.27 This guideline recommends inking of the SMA margin plane followed by complete perpendicular sectioning of the entire inked surface for histologic examination, to enable accurate measurement of the closest tumor-to-margin distance.
(2) The pancreatic neck (transection) margin refers to the entire resection plane of the pancreatic stump. By performing complete sectioning parallel to the transected surface (with the cut surface facing down), carcinoma identified microscopically at the inked plane indicates direct tumor involvement of the inked resection plane (tumor-to-margin distance, 0 mm).24,25 This records a tumor-to-margin distance and does not establish a separate R1 threshold; the SMA margin protocol may be used in routine practice. When the tumor is grossly ≤ 1 cm from the margin, the entire transection plane should be inked, followed by perpendicular sectioning toward the inked surface. All inked tissue blocks should undergo histologic examination, and the tumor-to-margin distance should be recorded to the nearest millimeter.
(3) The bile duct or hepatic duct margin constitutes the entire transection plane of the ductal stump. By performing complete sectioning parallel to the transected surface (with the mucosal surface facing down), microscopic identification of carcinoma at the inked plane indicates direct tumor involvement of the inked resection plane (tumor-to-margin distance, 0 mm).24,25 Alternatively, perpendicular sectioning may be adopted in accordance with the SMA margin protocol.
(4) Proximal and distal gastrointestinal margins represent the entire transection planes of the digestive tract. With complete sectioning parallel to the transected surface (with the transected surface facing down), the initial histologic section displays the true resection margin. Microscopic identification of carcinoma at this inked plane indicates direct tumor involvement of the inked resection plane (tumor-to-margin distance, 0 mm).24,25 Alternatively, perpendicular sectioning may be performed according to the SMA margin protocol.
(5) Portal vein margin: The status of the portal vein margin and the depth of tumor invasion into the venous wall are prognostically significant.28 When a partial or complete segment of the portal vein is attached to the specimen, this must be explicitly documented. Both proximal and distal portal vein margins require histologic sampling. For blocks containing tumor-involved portal vein tissue, the entire venous wall thickness should be encompassed in the tissue section.
(6) Other circumferential surfaces
Posterior peripancreatic surface (excluding the SMA margin): This surface comprises the loose connective tissue spanning the entire posterior aspect of the pancreatic head, extending from the superior mesenteric artery (SMA) margin to the pancreatoduodenal groove. This surface requires complete sectioning (as a continuous plane rather than localized perpendicular sections) to determine the closest tumor-to-margin distance. When this surface is included with the SMA margin in the same tissue block, the two surfaces should be inked in different colors for precise margin orientation.
SMV groove: A smooth grooved impression (also termed the portal vein groove) on the posteromedial aspect of the pancreatic head, overlying the superior mesenteric vein (SMV). This surface should be completely sectioned as a continuous plane rather than localized perpendicular sections, consistent with the SMA margin protocol.
Anterior peripancreatic surface: This surface is defined as the entire ventral aspect of the pancreatic head extending from the SMV groove to the anterior pancreatoduodenal groove. Although not a true surgical resection margin, tumor invasion of this surface correlates with local recurrence and reduced survival, and therefore warrants histologic assessment.29-34 When adherent to adjacent resected structures, these should be processed as additional circumferential margins with reporting of the closest tumor distance.
Collectively, these margins and pancreatic surfaces constitute the circumferential surfaces of the pancreaticoduodenectomy specimen (Fig. 1). Applying different-colored inks to individual margins and surfaces facilitates microscopic orientation.24 Gross examination should document the closest tumor-to-margin distance, which should be confirmed histologically.
Resection margin and specimen surfaces in distal pancreatectomy specimens
(1) Proximal pancreatic (transection) margin: When the tumor is grossly ≤ 1 cm from this margin, the entire margin surface must be inked, followed by perpendicular sectioning toward the inked plane. All tissue blocks containing the inked margin require histologic examination with millimeter-level precision in documenting the tumor-to-margin distance. For tumors > 1 cm from the margin, sectioning parallel to the margin is permitted.
(2) Anterior peripancreatic surface: This surface reflects the relationship between the tumor and the soft tissue at the pancreatic anterior border. The tissue plane closest to the tumor must be sampled to assess surface involvement. The exact number of sections depends on the gross extent of tumor invasion.
(3) Posterior peripancreatic surface: This surface reflects the relationship between the tumor and the retroperitoneal soft tissue posterior to the pancreas. Tissue from the plane closest to the tumor must be sampled to assess surface involvement. The exact number of sections required depends on the gross extent of tumor invasion.
The distal pancreatectomy specimen should be evaluated for tumor involvement at the pancreatic neck margin and within the peripancreatic soft tissues surrounding the resection site (Fig. 1). Additionally, involvement of the splenic vessels (splenic artery and vein) and the spleen itself must be documented. The resection margins of the splenic vein and splenic artery should be sampled for histologic examination. The report should include the distance from the tumor to the closest resection margin or the specimen surface, and this distance must be confirmed microscopically.
How to select sampling methods for pancreatic cancer surgical specimens?
Recommendation 2: For pancreatoduodenectomy specimens, serial transverse sections perpendicular to the long axis of the duodenum are recommended for pancreatic head carcinomas; this method is not recommended for ampullary or common bile duct carcinomas. For distal pancreatectomy specimens, serial sections through the maximum tumor diameter in the anteroposterior plane (either perpendicular or parallel to the main pancreatic duct) should be used. Complete sampling of the entire lesion is mandatory when assessing pathologic complete response (pCR) in pancreatic cancer specimens after neoadjuvant therapy.
Quality of evidence: Moderate
Strength of recommendation: Strong
Currently, there are several approaches to sampling pancreaticoduodenectomy specimens.26,34-38 However, this guideline recommends sectioning perpendicular to the long axis of the duodenum using serial parallel sections (Fig. 2).34 This method is straightforward to perform, as it does not require opening the common bile duct or the main pancreatic duct. However, it displays critical anatomical structures (such as the ampulla of Vater, common bile duct, and main pancreatic duct) within the same plane. Consequently, it is recommended by several cancer centers.39,40 This sectioning technique typically dissects the pancreaticoduodenectomy specimen into 8–10 sections. This approach allows detailed examination of the tumor characteristics, including its relationship to surrounding key anatomical structures and surgical margins.35 A major advantage is that tumor extent can be measured in three dimensions and the relationship between the tumor and each surgical margin can be assessed in every section.41-44
In distal pancreatectomy specimens, the splenic artery and splenic vein run along the posterosuperior border of the pancreas. This anatomical relationship allows for clear identification of the anterior peripancreatic and posterior peripancreatic surfaces, which should be inked to facilitate specimen orientation. The entire specimen may be serially sectioned in the anteroposterior plane, either parallel to the main pancreatic duct (Fig. 3a) or perpendicular to the main pancreatic duct (Fig. 3b).
For total pancreatectomy specimens, the pancreatic head can be sectioned serially using transverse sections (i.e., perpendicular to the long axis of the duodenum). The pancreatic body and tail may be sectioned serially either parallel to or perpendicular to the main pancreatic duct.
Before specimen dissection, samples should be obtained from the pancreatic duct margin, common bile duct margin, and both resection margins of the alimentary tract. Tissue blocks must encompass both tumor-involved anatomical structures (e.g., duodenum, ampulla of Vater, common bile duct, and peripancreatic soft tissues) and show the relationship between the tumor and its respective surgical margins. However, gross visual identification of precise tumor infiltration boundaries is often unreliable; therefore, extensive sampling of the tumor and adjacent margins is essential.42,45,46 When feasible, large-section embedding may clarify spatial relationships between the tumor, surrounding structures, and margins, while enabling more accurate measurement of the maximum tumor diameter.39 Following neoadjuvant therapy, tumor tissue is frequently replaced by fibrosis, making it difficult to distinguish grossly residual tumor from treatment-related fibrotic areas and fibrosis associated with obstructive pancreatitis.47 Extensive sampling is imperative for accurately evaluating both the extent of residual tumor and its proximity to margins. Crucially, assessment for pCR after neoadjuvant therapy requires submission and examination of the entire pancreatic specimen.39,48 Furthermore, retaining sequential images of gross sections (Fig. 4) provides useful documentation for multidisciplinary discussions regarding tumor origin and other key pathologic features. Standardized sampling may improve the accuracy and consistency of pathologic diagnosis and characterization of disease extent and biological behavior.
How to evaluate the histologic type and differentiation of pancreatic cancer?
Recommendation 3: For the histologic classification of pancreatic cancer, the fifth edition of the World Health Organization (WHO) Classification of Pancreatic Tumors is recommended. For grading histologic differentiation of pancreatic ductal adenocarcinoma, the College of American Pathologists (CAP) evaluation criteria are recommended. When pancreatic ductal adenocarcinoma is associated with precursor lesions, the report should state whether the ductal adenocarcinoma arose from a precursor lesion and should specify the histologic subtype and grade.
Quality of evidence: Low
Strength of recommendation: Weak
Pancreatic cancers covered by this guideline include pancreatic ductal adenocarcinoma, acinar cell carcinoma, and carcinomas arising from precursor lesions. The fifth edition of the WHO Classification of Digestive System Tumors provides a detailed classification of the histologic types for these categories of pancreatic cancer (Table 2).1
| PDAC, NOS | 8500/3 |
|---|
| Colloid carcinoma | 8480/3 |
| Poorly cohesive carcinoma | 8490/3 |
| Signet-ring cell carcinoma | 8490/3 |
| Medullary carcinoma NOS | 8510/3 |
| Adenosquamous carcinoma | 8560/3 |
| Hepatoid carcinoma | 8576/3 |
| Large cell carcinoma with rhabdoid phenotype | 8014/3 |
| Carcinoma, undifferentiated, NOS | 8020/3 |
| Undifferentiated carcinoma with osteoclast-like giant cells | 8035/3 |
| Acinar cell carcinoma | 8550/3 |
| Acinar cell cystadenocarcinoma | 8551/3 |
| Mixed acinar-neuroendocrine carcinoma | 8154/3 |
| Mixed acinar-endocrine-ductal carcinoma | 8154/3 |
| Mixed acinar-ductal carcinoma | 8552/3 |
| Carcinoma arising from a precursor lesion |
| IPMN with associated invasive carcinoma | 8453/3 |
| ITPN with associated invasive carcinoma | 8503/3 |
| IOPN with associated invasive carcinoma | 8455/3 |
| MCN with associated invasive carcinoma | 8470/3 |
Pancreatic ductal adenocarcinoma and its specific subtypes account for approximately 90% of pancreatic malignancies, while acinar cell carcinoma accounts for < 2% of pancreatic malignancies in adults. Accurate diagnosis of the specific subtypes of pancreatic ductal adenocarcinoma is crucial, as they differ in biological behavior, prognosis, and treatment strategies. When precursor lesions such as intraductal papillary mucinous neoplasm (IPMN), intraductal oncocytic papillary neoplasm (IOPN), intraductal tubulopapillary neoplasm (ITPN), and mucinous cystic neoplasm (MCN) are present in the background of pancreatic ductal adenocarcinoma, diagnosis requires determining whether the invasive carcinoma component has arisen from an intraductal tumor (IPMN, ITPN, IOPN) or MCN.49,50 Patients with ductal adenocarcinoma arising from MCN have a significantly better 5-year survival rate than those with ductal adenocarcinoma not arising from MCN.49,50 Both ductal adenocarcinoma arising from IPMN and IPMN with associated invasive carcinoma exhibit a more favorable prognosis and are more likely to be diagnosed at an early clinical stage than conventional ductal adenocarcinoma. Among invasive carcinoma subtypes, colloid carcinoma arising from IPMN has a better prognosis than ductal adenocarcinoma arising from IPMN.51-54 Oncocytic carcinoma arising from IOPN has been associated with a particularly favorable prognosis, with a 5-year disease-specific survival rate of 100%.52,55,56 Patients with ITPN associated with invasive carcinoma have a 5-year survival rate of approximately 71%, which is significantly better than that of conventional ductal adenocarcinoma.57-60 Some rare subtypes have also been reported in the literature, including clear cell type,61-63 oncocytic subtype,64 ciliated cell type,65 vacuole-rich type,66 intestinal type,67 microglandular type,68 cystic type,69 large duct type,70 cystic papillary type,71 mucoepidermoid carcinoma,72 and non-mucinous glycogen-deficient cystadenocarcinoma.73 With the advancement of molecular pathology, pancreatic carcinoma with SMARCB1/INI1 loss has also been reported in recent literature.74,75 Given the rarity of these tumor types and the limited research data available, their clinicopathologic significance requires further investigation. Standardized diagnosis of special histologic subtypes (e.g., colloid carcinoma and adenosquamous carcinoma) may help define subtype-specific screening sensitivity and provide a pathologic basis for developing individualized screening strategies.
Histologic differentiation holds significant prognostic value for pancreatic ductal adenocarcinoma.76 WHO and CAP have established grading criteria that show high concordance and similar predictive value.1,77,78 The CAP system is simpler than the WHO criteria, as it does not require assessment of mucin production or mitotic figures, and is therefore more widely used. Accordingly, this guideline recommends using the CAP system for evaluating the histologic differentiation of pancreatic ductal adenocarcinoma: Grade 1, > 95% of the tumor forms glandular structures; Grade 2, 50–95% of the tumor forms glandular structures; Grade 3, < 50% of the tumor forms glandular structures.
How to evaluate lymphovascular invasion?
Recommendation 4: Lymphatic and vascular invasion should be recorded separately. For invasion of named vessels (celiac trunk, superior mesenteric artery, common hepatic artery, portal vein, superior mesenteric vein, splenic artery, splenic vein), the depth of invasion (adventitia, media, intima) should be documented.
Quality of evidence: Moderate
Strength of recommendation: Strong
Lymphatic and vascular invasion are associated with poor prognosis in patients with pancreatic cancer, regardless of whether they have received neoadjuvant therapy.79-81 According to the American Joint Committee on Cancer (AJCC)/Union for International Cancer Control (UICC) eighth edition of the tumor-node-metastasis (TNM) staging system, and the National Comprehensive Cancer Network (NCCN) guidelines, these pathologic features should be recorded separately.
It is difficult to distinguish between small intratumoral lymphatic vessels and veins on routine histologic sections; the presence of accompanying small arteries can help identify venous invasion. Special stains, particularly elastic fiber stains, or immunohistochemical markers such as CD31, CD34, and D2-40, are helpful for differential diagnosis. Additionally, immunohistochemistry or special stains can identify vascular invasion that may be mistaken for pancreatic intraepithelial neoplasia (PanIN). Invasion of large, named vessels may help predict postoperative recurrence and 5-year survival in patients with pancreatic cancer, especially for the celiac trunk, superior mesenteric artery, and common hepatic artery, involvement of which affects TNM stage. The report should clearly indicate the presence or absence of invasion and specify the depth of invasion for these vessels.27 For vessels not yet incorporated into the T stage, such as the portal vein, superior mesenteric vein, splenic artery, and splenic vein, it is also recommended to record the depth of invasion, because it may still have prognostic value for tumor metastasis, recurrence, and prognosis.28,82-89 Standardized documentation of lymphatic/vascular invasion, particularly precise assessment of the invasion depth in key named vessels, supports accurate staging and prognostic evaluation of advanced pancreatic cancer.
How to evaluate perineural invasion?
Recommendation 5: Perineural invasion should be reported, with intrapancreatic and extrapancreatic invasion distinguished when possible.
Quality of evidence: Low
Strength of recommendation: Weak
Perineural invasion is defined as tumor cell invasion into any of the three layers of a nerve or tumor cell encirclement of at least one-third of the nerve circumference.90 Perineural invasion is a significant histologic feature of pancreatic cancer and an adverse prognostic feature associated with postoperative recurrence.81,91-95 The peripheral nerves in the pancreas are abundant and are classified into intrapancreatic nerves (within the pancreatic capsule) and extrapancreatic nerves (outside the capsule). Intrapancreatic perineural invasion is an important factor influencing the prognosis and recurrence of pancreatic cancer patients after surgery,92,96,97 and is correlated with extrapancreatic perineural invasion.98 Intrapancreatic perineural invasion facilitates the spread of cancer cells along nerves to extrapancreatic nerves and soft tissues.99,100 Patients with extrapancreatic perineural invasion are more likely to experience recurrence and have a worse prognosis.101-103 Perineural invasion is also associated with poor prognosis in patients who have received neoadjuvant therapy.104,105 Standardized assessment of perineural invasion may inform treatment decisions and postoperative surveillance. Therefore, this guideline recommends separately recording intrapancreatic and extrapancreatic perineural invasion.
How to evaluate the R1 resection margin in pancreatic cancer?
Recommendation 6: R1 is defined as a microscopic tumor distance to the resection margin ≤ 1 mm. The ≤ 1-mm criterion applies uniformly to all assessed true surgical resection margins and circumferential specimen surfaces; a distance of 0 mm denotes direct tumor involvement of the inked plane or surface and is not a separate margin-specific R1 threshold.
Quality of evidence: very low
Strength of recommendation: Weak
Although current international guidelines assess R status primarily using true surgical margins, such as the SMA margin, pancreatic transection margin, bile duct margin, and proximal/distal gastrointestinal margins,24,25,77,106 they also recommend assessing all surfaces of the resection specimen, including the anterior peripancreatic surface, posterior peripancreatic surface, and the SMV groove surface.25,106 Tumor involvement of these surfaces increases the risk of local tumor recurrence.30,44,107,108 Positive resection margins are also associated with poorer survival.109-113 Numerous studies based on comprehensive and detailed pathologic examination protocols have found that evaluating all surfaces yields R1 rates > 70%, which have been associated with survival.31,38,42,43,114-118 Therefore, this guideline recommends comprehensive sampling of all resection margins and surfaces, with microscopic confirmation of the shortest distance from the tumor to the margin or surface.
The definition of R0 and R1 in pancreatic cancer remains controversial. The ≤ 1-mm criterion was originally derived from rectal cancer assessment protocols, where a tumor distance to the resection margin ≤ 1 mm was found to predict local recurrence and be associated with poorer survival.119-123 Considering the more dispersed growth pattern of pancreatic ductal adenocarcinoma, some studies suggest that a larger distance (e.g., 1.5 mm) may offer greater prognostic discrimination;123-125 however, this has not been applied in routine diagnostic practice. Standardized assessment of the retroperitoneal margin is also necessary.126 Furthermore, anatomically, the anterior surface of the pancreas is covered by peritoneum; therefore, involvement of the anterior surface should be recorded as direct surface penetration (0 mm).11
In rare cases, no definitive tumor is identified in the pancreatic parenchyma, interstitial fibrosis, or adipose tissue within 1 mm of the resection margin, but tumor is present in lymph nodes, lymphatic or blood vessels, or perineural spaces within this range. Evidence is insufficient to determine whether such margins should be classified as R1. The UICC TNM staging system proposes that if tumor cells are found only within the lumina of vessels within 1 mm of the margin, this is classified as R0, whereas if the tumor adheres to or invades the vessel wall, it is classified as R1.114 Currently, evidence is insufficient to define involvement of lymph nodes or peripheral nerves within 1 mm of the margin; it is recommended to record and describe these findings in the pathology report.
Additionally, the R1 definition of ≤ 1 mm is applicable only to pancreatic ductal adenocarcinoma. Evidence is insufficient to determine whether this definition applies to acinar cell carcinoma, which, unlike pancreatic ductal adenocarcinoma, typically does not exhibit a dispersed growth pattern. The closest distance should therefore be recorded from the tumor to the resection margin.
The definition of R1 margin distance in patients who have received neoadjuvant therapy remains uncertain.48 First, following neoadjuvant treatment, selective sampling may lead to false-negative results in margin assessment. Second, the distance between tumor cells and the resection margin used to define R1 is also controversial. Some studies suggest that a closest tumor-to-margin distance > 2 mm is associated with a better prognosis.127,128 Standardized 1-mm margin assessment currently provides a relatively uniform benchmark for defining “radical resection,” helps determine the completeness of surgical intervention, and lays the foundation for subsequent evidence collection. It should be noted that whether 1 mm represents the optimal cutoff value remains uncertain and requires further clinical validation. Consequently, for surgically resected specimens following neoadjuvant therapy, this guideline recommends comprehensive margin evaluation through complete specimen sampling and large-section histopathology, with detailed documentation of the closest distance of residual tumor to the resection margin.
How to assess T stage?
Recommendation 7: T staging should be performed according to the AJCC/UICC 8th edition staging system. For invasive carcinoma associated with IPMN, IOPN, ITPN, or MCN, the T1–T3 stage should be assessed based on the largest dimension of the invasive component, not the size of the entire lesion. For surgically resected pancreatic specimens following neoadjuvant therapy, large-section sampling is recommended, and the method for measuring tumor size should be documented.
Quality of evidence: Low
Strength of recommendation: Weak
T staging should follow the AJCC/UICC 8th edition staging system (Table 3).27 The staging of acinar cell carcinoma also follows the PDAC staging system. The designations “m,” “r,” and “y” indicate multiple primary tumors, recurrent tumors, and tumors assessed after neoadjuvant therapy, respectively.
| TNM stage | Details |
|---|
| Primary tumor (T) | TX: Primary tumor cannot be assessed |
| T0: No evidence of primary tumor |
| Tis: Carcinoma in situ
This includes high-grade pancreatic intraepithelial neoplasia (PanIN-3), intraductal papillary mucinous neoplasm with high-grade dysplasia, intraductal tubulopapillary neoplasm with high-grade dysplasia, intraductal oncocytic papillary neoplasm with high-grade dysplasia, and mucinous cystic neoplasm with high-grade dysplasia |
| T1: Tumor ≤ 2 cm in greatest dimension |
| T1a: Tumor ≤ 0.5 cm in greatest dimension |
| T1b: Tumor > 0.5 cm and < 1 cm in greatest dimension |
| T1c: Tumor 1–2 cm in greatest dimension |
| T2: Tumor > 2 cm and ≤ 4 cm in greatest dimension |
| T3: Tumor > 4 cm in greatest dimension |
| T4: Tumor involves the celiac axis, superior mesenteric artery, and/or common hepatic artery, regardless of size |
| Regional lymph nodes (N) | NX: Regional lymph nodes cannot be assessed |
| N0: No regional lymph node metastases |
| N1: Metastasis in 1–3 regional lymph nodes |
| N2: Metastasis in 4 or more regional lymph nodes |
| Distant metastasis (M) | M0: No distant metastasis |
| M1: Distant metastasis |
Tumor size reflects primary tumor burden and is an important prognostic factor.129 Because of the highly dispersed growth pattern of pancreatic cancer and pancreatic parenchymal atrophy and fibrosis, accurately defining the true tumor boundaries macroscopically is challenging.123 Currently, most guidelines recommend using macroscopic tumor size measurement with microscopic confirmation.11,39,77,130,131 The T stage is assessed based on the maximum tumor diameter, as tumor size provides better prognostic stratification compared to the extent of peripancreatic invasion.112,132-135 Based on the maximum diameter, tumors are classified as T1 (≤ 2 cm), T2 (> 2 cm and ≤ 4 cm), and T3 (> 4 cm). T4 is defined by tumor invasion of the superior mesenteric artery/celiac trunk/common hepatic artery, regardless of tumor size. T1 can be further subdivided into T1a (≤ 0.5 cm), T1b (> 0.5 cm and < 1 cm), and T1c (≥ 1 cm and ≤ 2 cm).27 Tis includes PanIN-3, IPMN with high-grade dysplasia, ITPN with high-grade dysplasia, IOPN with high-grade dysplasia, and MCN with high-grade dysplasia.
In T staging, if more than one tumor is present, the number, location, and size of each tumor should be recorded in detail. However, this applies only to multiple tumors identifiable macroscopically, and not to tumors that appear as a single mass grossly but are multifocal microscopically.77 For precursor lesions of PDAC with invasive carcinoma, including IPMN, IOPN, ITPN, and MCN with invasive carcinoma, the T1–T3 staging criteria are based on the maximum diameter of the invasive focus, not the size of the entire lesion.50,136 If the invasive foci are multifocal, it remains uncertain whether the T stage should be determined by the diameter of the largest invasive focus or the sum of multiple invasive foci. Therefore, in routine practice, a thorough evaluation is recommended. The number of invasive foci, the maximum diameter of each invasive focus, and the overall maximum diameter of the invasive foci should all be assessed to provide detailed pathologic data for future clinical research and development of more precise and standardized T-staging criteria.
For resected pancreatic cancer specimens following neoadjuvant therapy, current measurement methods are limited. After neoadjuvant treatment, tumor cells regress and are replaced by fibrous tissue. Particularly when tumor cells have largely regressed and the surrounding pancreatic tissue shows severe atrophy and fibrosis, it becomes more challenging to accurately identify the tumor boundaries macroscopically. Current T-staging criteria may not accurately stratify prognosis in these cases.137 Therefore, more robust methods are needed to assess the actual tumor size after neoadjuvant therapy. Some studies have used microscopic confirmation of tumor boundaries together with gross images to delineate tumor borders and measure tumor size for T-stage assessment.138 While this method is more accurate than direct macroscopic measurement, it is technically complex and still subject to significant error. Currently, the following two primary methods are recommended for tumor size assessment:
(1) Microscopic measurement of the maximum linear diameter of the entire area containing residual viable tumor cells, including non-neoplastic tissue (such as stroma and/or pancreatic parenchyma or other tissue structures) between tumor cells.
(2) Microscopic measurement of the maximum diameter of each tumor focus and summing these measurements.
Regardless of the measurement method, accurate assessment relies on precise sampling and adequate representation of the tumor’s maximum extent. Therefore, this guideline recommends using large-section sampling for surgically resected pancreatic specimens following neoadjuvant therapy to show the full extent of the tumor and its relationship to surrounding tissues, and to document the microscopic measurement method in the pathology report.
Tumor invasion into surrounding tissues includes involvement of peripancreatic soft tissues, the peritoneum (including the mesocolon, greater and lesser omenta), and the extrapancreatic biliary system. Tumors in the pancreatic head may also invade the duodenum (including the ampulla of Vater), while tumors in the pancreatic body and tail may invade the stomach, transverse colon, spleen, and left adrenal gland. Although tumor invasion into surrounding tissues does not affect tumor staging, it should still be recorded in the pathology report. Standardized staging supports assessment of tumor extent and evaluation of downstaging and tumor shrinkage after neoadjuvant therapy and may inform comprehensive early-intervention strategies.
How to assess N stage?
Recommendation 8: Follow the AJCC/UICC 8th edition staging system (Table 3). Both direct tumor invasion and metastases are counted as positive lymph nodes. As many lymph nodes as possible should be retrieved, with at least 12 examined in pancreaticoduodenectomy specimens.
Quality of evidence: Moderate
Strength of recommendation: Strong
Lymph node status is an important prognostic factor in pancreatic ductal adenocarcinoma. Based on prognostic data, the classification is as follows: N0 (no regional lymph node metastasis), N1 (metastasis in 1–3 regional lymph nodes), and N2 (metastasis in 4 or more regional lymph nodes). All lymph nodes from the resection specimen should undergo histologic examination. Separately submitted regional lymph nodes should be reported independently.
The number of positive lymph nodes detected has been shown to correlate with patient survival, while insufficient lymph node sampling may lead to understaging of the N stage.139-143 The International Collaboration on Cancer Reporting, CAP, and the AJCC/UICC 8th Edition recommend that at least 12 lymph nodes be retrieved from pancreaticoduodenectomy specimens. Based on data from multiple studies, the Royal College of Pathologists (UK) recommends that at least 15 lymph nodes should be retrieved from pancreaticoduodenectomy specimens.141,142,144,145 In patients who have received neoadjuvant therapy, the number of retrieved lymph nodes often decreases and may fall below 15.146 There is currently no consensus on the minimum number of lymph nodes to be retrieved from distal pancreatectomy specimens, and the benefit of extended lymphadenectomy remains uncertain.
Direct tumor invasion of lymph nodes is identified in up to 20% of pancreatic cancer resections. Some studies suggest that direct invasion does not represent true lymph node metastasis (i.e., spread via lymphatic vessels) and is prognostically equivalent to pN0.147-149 Other studies indicate that direct invasion is associated with outcomes similar to those of patients with true pN1 disease.150,151 In the UICC/AJCC 8th edition staging system, lymph nodes with direct tumor invasion are also counted as positive lymph nodes.11,106 Additionally, some studies suggest that the lymph node ratio (the ratio of positive lymph nodes to the total number of lymph nodes examined) is also an effective prognostic indicator, with a ratio > 20% associated with poorer survival rates.109,152-160 Standardized lymph node evaluation supports precise risk stratification and individualized treatment in pancreatic cancer.
In both the AJCC/UICC TNM and the Japan Pancreas Society (JPS) systems, lymph nodes around the common hepatic artery are considered regional lymph nodes,106 while para-aortic lymph nodes are not classified as regional. Survival is poorer in patients with positive para-aortic lymph nodes than in those without, and metastasis to para-aortic lymph nodes is considered distant metastasis (i.e., pM1)161-163; however, this classification is not uniformly applied in clinical practice.
How to assess M stage?
Recommendation 9: Absence of distant metastasis is classified as M0, whereas the presence of distant metastasis is classified as M1. Peritoneal seeding and positive ascites are classified as M1.
Quality of evidence: Moderate
Strength of recommendation: Strong
A key aspect of M staging is that peritoneal seeding and positive ascites are classified as M1.27,164 Pathologic identification of M1 disease may help determine whether curative-intent intervention remains appropriate in screening-detected cases.
How to assess tumor regression grading (TRG) after neoadjuvant therapy?
Recommendation 10: The CAP and MD Anderson Cancer Center (MDACC) assessment systems are recommended to evaluate TRG after neoadjuvant therapy. Pathologic response assessment following neoadjuvant therapy may provide feedback on early intervention strategies and may support optimization of preoperative regimens and outcomes for screening-eligible populations.
Quality of evidence: Low
Strength of recommendation: Strong
The role of neoadjuvant therapy in improving the prognosis of pancreatic cancer patients has become increasingly important. When evaluating pancreatic cancer resection specimens after neoadjuvant therapy, TRG is typically assessed to evaluate the tumor’s response to neoadjuvant therapy, thereby guiding subsequent adjuvant therapy and predicting patient prognosis. Currently, multiple TRG assessment systems are used for pancreatic cancer, including the Evans system, first proposed in 1992165; the CAP system,166 which is based on the grading system initially proposed by Ryan et al.167 for evaluating neoadjuvant therapy response in rectal cancer, also known as the modified Ryan scheme; the MDACC system, based on the work of Chatterjee et al.168; and the JPS system (Table 4).34 Although no consensus has been reached on the optimal assessment system, available evidence suggests that patients with complete or near-complete response have a better prognosis.168 The AJCC 8th edition ypT stage after neoadjuvant therapy has also demonstrated prognostic significance and correlation with tumor regression grade.169 The CAP assessment system is recommended by the NCCN guidelines. The Evans and JPS grading systems are specifically designed for PDAC and are commonly used in Japan.170 The MDACC system is similar to CAP but uses a three-tier classification instead of a four-tier one. All four TRG assessment systems mentioned above are semiquantitative. Although all systems define complete regression similarly, the thresholds and criteria for assessing the degree of tumor regression differ, making it difficult to compare them directly.
| Assessment system | Tumor regression grade | Definition |
|---|
| Evans | Grade I | < 10% or no tumor cell destruction |
| Grade IIa | Destruction of 10–50% of tumor cells |
| Grade IIb | Destruction of 51–90% of tumor cells |
| Grade III | Few (< 10%) tumor cells present |
| Grade IV | No viable tumor cells present |
| CAP | Score 0 | No viable cancer cells |
| Score 1 | Single cells or rare small groups of cancer cells |
| Score 2 | Residual cancer with evident tumor regression, but more than single cells or rare small groups of cancer cells |
| Score 3 | Extensive residual cancer with no evident tumor regression |
| JPS | Grade 1a | Estimated residual tumor ratio ≥ 90% |
| Grade 1b | Estimated residual tumor ratio ≥ 50%, < 90% |
| Grade 2 | Estimated residual tumor ratio ≥ 10%, < 50% |
| Grade 3 | Estimated residual tumor ratio < 10% |
| Grade 4 | No viable tumor cells |
| MDACC | Score 0 | No residual carcinoma |
| Score 1 | < 5% residual carcinoma |
| Score 2 | ≥ 5% residual carcinoma |
Complete tumor regression, near-complete regression, partial regression, and no regression are important outcome indicators of neoadjuvant therapy. However, the reliability of histologic assessment for these outcomes remains uncertain. The main challenges lie in establishing reliable diagnostic criteria for complete regression and achieving reproducible assessments of varying degrees of regression. Therefore, extensive and comprehensive tissue sampling is crucial. Most studies, however, lack information on the extent of sampling. The limited published data indicate significant variability both within and between studies. A study based on complete embedding of the pancreas and surrounding tissues showed that only 2.5% of patients achieved complete tumor regression.171 In contrast, other studies that did not employ extensive sampling reported complete tumor regression in 10% to 33% of patients.172-181 Consequently, accurate assessment of tumor regression requires extensive sampling of the lesional tissue. Particularly for evaluating complete tumor regression, it is essential to perform thorough sampling and histologic examination of the entire tumor bed and any adjacent macroscopically abnormal tissues.
Should background lesions of pancreatic cancer be assessed?
Recommendation 11: Background lesions of pancreatic cancer, such as PanIN, IPMN, ITPN, IOPN, MCN, and chronic pancreatitis, should be assessed.
Quality of evidence: Low
Strength of recommendation: Weak
In addition to focusing on the tumor itself, increasing attention is being given to background lesions associated with the tumor. PanIN is the most common precursor lesion of pancreatic ductal adenocarcinoma.182 Some studies suggest that the absence of accompanying PanIN may be associated with poorer postoperative survival.183 If IPMN, ITPN, IOPN, or MCN are present, they should be noted in the pathology report, along with clarification of the degree of epithelial dysplasia and the relationship to pancreatic cancer.136,184 The NCCN guidelines also recommend documenting background lesions, such as the presence or absence of PanIN or chronic pancreatitis, in the diagnostic report. Standardized assessment and documentation of background lesions in pancreatic cancer may support patient risk stratification and individualized management, improve understanding of tumor origin and biological behavior, and may inform early detection and prevention strategies.