Neoadjuvant chemotherapy or chemoradiotherapy versus upfront surgery alone for resectable or borderline resectable pancreatic cancer
Abstract
Objectives
This is a protocol for a Cochrane Review (intervention). The objectives are as follows:
To compare the efficacy and safety of neoadjuvant chemotherapy or chemoradiotherapy versus upfront surgery alone in patients with resectable or borderline resectable pancreatic cancer.
Background
Description of the condition
Pancreatic cancer is highly malignant, and nearly all patients diagnosed with pancreatic cancer are expected to die from the disease (Ryan 2014). In the United States, it is the fourth‐leading cause of cancer‐related death in both men and women (Siegel 2021).
Currently, surgical resection, especially complete (R0) resection, is the only potentially curative treatment to achieve long‐term survival for pancreatic cancer patients (Konstantinidis 2013; Mayo 2012; Shaib 2007). Most patients who cannot receive surgery die within one year (Hidalgo 2010; Sultana 2007). Based on the involvement of adjacent structures (local vessels, including the celiac artery, superior mesenteric artery and vein, common hepatic artery, and portal vein) and the presence of distant metastases, patients may or may not be selected for surgery. Several expert consensus groups have defined and categorised resectability of pancreatic cancer as ‘resectable’, ‘borderline resectable’, ‘locally advanced', and 'metastatic’ (Callery 2009; Katz 2016; NCCN 2021; Varadhachary 2006). Four commonly used criteria that define resectable, borderline resectable and locally advanced pancreatic cancer are listed in Table 1.
| Vascular | Definition | National Comprehensive Cancer Network (NCCN) (NCCN 2021) | MD Anderson Cancer Center (MDACC) (Varadhachary 2006) | American Hepato‐Pancreato‐Biliary Association/Society for Surgery of the Alimentary Tract/Society of Surgical Oncology (AHPBA/SSAO/SSO) (Callery 2009) | Alliance for Clinical Trials in Oncology (Alliance) (Katz 2016) |
| Celiac axis (CA) | Resectable | No arterial tumour contact | No extension | Clear fat planes | No radiographic interface |
| Borderline resectable | Solid tumour contact with the CA ≤ 180°, or > 180° without involvement of the aorta and with intact and uninvolved gastroduodenal artery (for body/tail) | No extension | Clear fat planes | Interface between tumour and vessel measuring < 180° of the circumference of the vessel wall | |
| Locally advanced | Solid tumour contact of > 180° (for head/uncinate process) Solid tumour contact with the CA and aortic involvement (for body/tail) | Encased and no technical option for reconstruction | Circumferential encasement | Interface between the tumour and the SMA measuring ≥ 180° of the circumference of the vessel wall | |
| Superior mesenteric artery (SMA) | Resectable | No arterial tumour contact | No extension | Clear fat planes | No radiographic interface |
| Borderline resectable | Solid tumour contact with the SMA ≤ 180° | Tumor abutment ≤ 180° | Tumor abutment ≤ 180° | Interface between tumour and vessel measuring < 180° of the circumference of the vessel wall | |
| Locally advanced | Solid tumour contact with the SMA > 180° | Encased (> 180°) | Circumferential encasement | Interface between the tumour and the SMA measuring ≥ 180° of the circumference of the vessel wall | |
| Common hepatic artery (CHA) | Resectable | No arterial tumour contact | No extension | Clear fat planes | No radiographic interface |
| Borderline resectable | Solid tumour contact with CHA without extension to CA or hepatic artery bifurcation allowing for safe/complete reconstruction | Short‐segment encasement/abutment of the common hepatic artery (typically at the gastroduodenal origin) | Short segment encasement or direct abutment | Reconstructable, short‐segment interface between tumour and vessel of any degree | |
| Locally advanced | Not mentioned | Encased and no technical option for reconstruction | Circumferential encasement | Long‐segment interface (of any degree) between the tumour and the CHA or its major tributaries with insufficient artery proximal and distal to the interface to perform reconstruction | |
| Superior mesenteric vein/portal vein (SMV/PV) | Resectable | No tumour contact with the SMV or PV or ≤ 180° contact without vein contour irregularity | Patent | No radiographic evidence of abutment, distortion, tumour thrombus, or venous encasement | Interface between the primary tumour and the SMV/PV measuring < 180° of the circumference of the vessel wall |
| Borderline resectable | Solid tumour contact with the SMV/PV of > 180°, contact of ≤ 180° with contour irregularity of the vein or thrombosis of the vein but with suitable vessel proximal and distal to the site allowing for safe and complete reconstruction Solid tumour contact with the inferior vena cava (IVC) | Short‐segment occlusion with suitable vessel above and below for reconstruction | Venous involvement of the SMV/PV demonstrating tumour abutment with or without impingement and narrowing of the lumen Encasement of the SMV/PV but without encasement of the nearby arteries Short segment venous occlusion resulting from either tumour thrombus or encasement but with suitable vessel proximal and distal to the area of vessel involvement allowing for safe resection and reconstruction | Interface between tumour and vessel measuring > 180° of the circumference of the vessel wall, and/or reconstructable occlusion | |
| Locally advanced | Unreconstructible SMV/PV due to tumour involvement or occlusion | Occluded and no technical option for reconstruction | Major venous thrombosis of the SMV/PV extending for several centimetres | Occlusion of the SMV/PV without a sufficient cuff of normal vein above and below the level of obstruction with which to perform venous reconstruction |
Abbreviations
CA: Celiac axis
CHA: Common hepatic artery
IVC: Inferior vena cava
PV: Portal vein
SMA: Superior mesenteric artery
SMV: Superior mesenteric vein
According to a guideline developed by the American Society of Clinical Oncology (ASCO) in 2016, resectable and borderline resectable tumours are defined as potentially curable pancreatic cancers for which surgical resection is an appropriate consideration (Khorana 2016). Unfortunately, most pancreatic cancer patients have either locally advanced or metastatic disease at diagnosis, and only 15% to 20% of patients can receive curative resection (Li 2004). Considering that nearly all patients who receive potential curative resection ultimately die due to local recurrence or distant metastasis, adjuvant treatment following resection has been established as the standard treatment option, following the results of several randomised controlled trials (RCTs) (Conroy 2018; Neoptolemos 2017; Oettle 2013; Uesaka 2016). However, the five‐year survival rate of patients who have undergone curative resection, followed by adjuvant treatment, is still lower than 20%.
Description of the intervention
Surgical resection is the only potentially curative treatment for pancreatic cancer. The standard operation for pancreatic cancer located in the head or uncinate process of the pancreas is pancreaticoduodenectomy (i.e. the Whipple procedure), and the operation for cancer located in the body or tail of the pancreas is distal pancreatectomy (usually combined with splenectomy). Despite significant advances in surgical techniques and perioperative management, pancreatic surgery, especially the Whipple procedure, is still one of the most challenging surgical undertakings for general surgeons. Although pancreatectomy in experienced high‐volume centres can be performed with a relatively low perioperative mortality of less than 4%, the perioperative morbidity remains 40% to 50%, and surgical complications can significantly increase the length and cost of hospital stay (Bakkevold 1993; Kawai 2010; Kneuertz 2012; Yeo 1997). Due to postoperative complications, delayed surgery recovery, deteriorating performance status, and early recurrence, nearly half of surgical patients fail to receive timely and designated adjuvant treatment (Aloia 2007; Mayo 2012a). In addition, a 2015 meta‐analysis reported that the R0 resection could only be achieved in about 40% of surgical patients (Chandrasegaram 2015). Unfortunately, even for patients undergoing R0 resection with adjuvant treatment, their prognosis is still poor, with a median overall survival (OS) of 24 months (Sinn 2017).
Neoadjuvant treatment for pancreatic cancer often refers to the administration of chemotherapy or chemoradiotherapy, prior to the definitive surgery. The low rate of margin‐negative resection, the poor long‐term outcome of pancreatectomy combined with adjuvant treatment, and the fact that prolonged surgery recovery prevents the delivery of adjuvant chemotherapy, have led to the investigation of neoadjuvant treatment in patients with pancreatic cancer. Neoadjuvant treatment can offer the potential to downstage tumours and sterilise the micrometastases at an earlier stage, which may increase the proportion of patients with surgically resectable status or margin‐free resection, and reduce the rate of recurrence. In addition, neoadjuvant treatment can assess the efficacy of chemotherapy prior to surgery, which may help to inform subsequent treatment. Patients who respond well to neoadjuvant treatment are ideal candidates for surgery. In contrast, patients who respond poorly or whose disease progresses are not likely to benefit from surgical resection, because their tumour has been proven to be chemotherapy‐resistant, and is at a high risk of recurrence after resection. Moreover, patients who have surgery after neoadjuvant treatment appear to exhibit similar morbidity and mortality compared with patients who have upfront surgery (Araujo 2013; Marchegiani 2018).
However, it is noted that the neoadjuvant approach also carries the risk that the initially resectable disease becomes unresectable. A meta‐analysis regarding neoadjuvant chemotherapy and/or radiotherapy, with particular emphasis on tumour response rates, demonstrated that 2.8% of patients treated with neoadjuvant treatment had complete response, 28.7% had partial response, 45.9% had stable disease, and 16.9% had progression of disease (Tang 2016). Considering that 16.9% of patients experienced tumour progression during neoadjuvant treatment, and that upfront surgery may be a better choice for such patients, the role of neoadjuvant treatment became controversial. The best means to evaluate chemo‐responsiveness before neoadjuvant treatment and separate patients to give individualised treatment is still an important issue.
Neoadjuvant chemotherapy or chemoradiotherapy in resectable pancreatic cancer
According to the National Comprehensive Cancer Network (NCCN) guideline, administering neoadjuvant chemotherapy or chemoradiotherapy is probably not warranted in patients with resectable pancreatic cancer, except for those with high‐risk features (e.g. markedly elevated CA19‐9; large primary tumours; large regional lymph nodes; excessive weight loss; extreme pain) (NCCN 2021). However, some studies have demonstrated the potential benefits of neoadjuvant treatment in patients with resectable pancreatic cancer. A retrospective propensity score‐matched analysis of 15,237 patients with clinical stage I or II pancreatic cancer (AJCC 2009) from the National Cancer Database (NCDB) showed that those who received neoadjuvant treatment had a lower proportion of positive lymph nodes (48% versus 73%, P < 0.01) and positive resection margin (17% versus 24 %, P < 0.01), and longer median OS (26 months versus 21 months, hazard ratio (HR) 0.72, 95% confidence interval (CI) 0.68 to 0.78; P < 0.01) than those who received upfront resection (Mokdad 2017). A 2019 Bayesian network meta‐analysis reported that neoadjuvant treatment could offer benefits in both R0 resection rate and overall survival over surgery‐first for patients with resectable disease (Bradley 2019). At this time, the role of neoadjuvant chemotherapy or chemoradiotherapy in patients with resectable pancreatic cancer is evolving, and there is limited evidence to administer specific regimens to these patients.
Neoadjuvant chemotherapy or chemoradiotherapy in borderline resectable pancreatic cancer
Although some patients with borderline resectable pancreatic cancer will be proven to be resectable in initial surgery, it has become increasingly apparent that patients with borderline resectable disease are at high risk for margin‐positive (R1) resection and are not considered to be good candidates for upfront surgery. There appears to be a trend toward more frequent use of neoadjuvant treatment in cases of borderline resectable tumours, and the use of neoadjuvant treatment in borderline resectable disease has been a highly debated topic. In a retrospective series from the MD Anderson Cancer Center at the University of Texas, 62 (39%) of 160 patients who underwent neoadjuvant chemoradiotherapy for borderline resectable pancreatic cancer achieved R0 resection (Katz 2008). A systematic review and meta‐analysis of 38 studies found that resectable or borderline resectable pancreatic cancer patients undergoing neoadjuvant treatment had improved OS (18.8 months versus 14.8 months) and higher R0 rate (86.8% versus 66.9%, P < 0.001), but had lower overall resection rate (66.0% versus 81.3%, P < 0.001) (Versteijne 2018). A combination fluorouracil, irinotecan, and oxaliplatin (FOLFIRINOX) chemotherapy regimen is a promising approach, and can lead to a significantly improved response rate and survival in patients with pancreatic cancer (Conroy 2011; Murphy 2018). In a patient‐level meta‐analysis, patients who had borderline resectable pancreatic cancer, treated with neoadjuvant FOLFIRINOX, showed a favourable median OS (22.2 months), resection rate (67.8%), and R0 resection rate (83.9%), which were significantly better than historical controls (Janssen 2019). Although the results of these studies are encouraging, there is not enough rigorous evidence supporting the clinical application of neoadjuvant treatment. The bulk of the available data are small sample single‐centre retrospective studies or single‐arm clinical trials that enrolled a mixture of patients with initially borderline resectable disease. In addition, the best regimen for borderline resectable disease in the neoadjuvant setting is still under exploration.
In conclusion, many studies have shown clinical benefits of neoadjuvant chemotherapy or chemoradiotherapy in patients with resectable or borderline resectable pancreatic cancer. However, rigorous evidence to support the application of neoadjuvant treatment in pancreatic cancer is limited.
How the intervention might work
The rationale for offering neoadjuvant chemotherapy or chemoradiotherapy to patients with pancreatic cancer includes the following considerations (NCCN 2021):
-
it provides early treatment and potential sterilisation for micrometastases of the tumour, which may improve the likelihood of R0 resection and reduce the recurrence;
-
it provides downstaging of some initially unresectable tumours to the point where they may then be removed surgically;
-
it provides an interval to assess the tumour biology, thereby selecting patients who may benefit from surgery; and
-
it provides better‐tolerated and more timely chemotherapy, since surgical recovery will not complicate the treatment delivery.
Why it is important to do this review
For patients with resectable or borderline resectable pancreatic cancer, radical surgical resection is the core of the standard therapy. As mentioned previously, the high rate of microscopic incomplete resection and the extremely poor prognosis of pancreatic cancer suggests that these patients may be appropriate candidates to undergo neoadjuvant chemotherapy or chemoradiotherapy. Over the past two decades, many clinical studies and meta‐analyses have shown the clinical value of neoadjuvant treatment in pancreatic cancer. Several international guidelines have considered neoadjuvant treatment to be a fundamental approach to the management of pancreatic cancer. However, adequately powered high‐level evidence, especially from multi‐centre phase III RCTs, is quite limited. The existing retrospective studies and non‐randomised controlled trials have selection bias; that is, patients undergoing neoadjuvant treatment may have less disease burden than those undergoing upfront surgery. Although recent RCTs have indicated the superiority of neoadjuvant treatment, there is no compelling evidence for the benefits of neoadjuvant treatment in pancreatic cancer. Since neoadjuvant treatment is increasingly applied in pancreatic cancer and has become a highly debated topic, we hope that this Cochrane Review, which will include only RCTs, could clarify the efficacy and safety of neoadjuvant chemotherapy or chemoradiotherapy for patients with resectable or borderline resectable pancreatic cancer.
Objectives
To compare the efficacy and safety of neoadjuvant chemotherapy or chemoradiotherapy versus upfront surgery alone in patients with resectable or borderline resectable pancreatic cancer.
Methods
Criteria for considering studies for this review
Types of studies
We will include only RCTs.
Types of participants
Eligible RCTs will include participants who are diagnosed with resectable or borderline resectable pancreatic cancer.
Types of interventions
Participants in intervention arms of eligible RCTs will have received neoadjuvant chemotherapy or chemoradiotherapy.
Participants in control arms of eligible RCTs will have received upfront surgery, without receiving neoadjuvant chemotherapy or chemoradiotherapy.
Types of outcome measures
Primary outcomes
Overall survival: the time interval from randomisation until death (from any cause)
Secondary outcomes
-
Resection rate: the proportion of patients who receive curative resection
-
R0 resection rate: the proportion of patients who have R0 (negative microscopic margin) resection
-
Disease‐free survival: the time interval from randomisation until disease progression in patients who receive curative resection
-
Toxicity: the incidence of grade ≥ 3 adverse events related to neoadjuvant treatment. Adverse events are graded by Common Terminology Criteria for Adverse Events (CTCAE)
-
Operative morbidity: the incidence of surgical complications
-
Operative mortality: the incidence of surgery‐related death
Search methods for identification of studies
We will conduct a comprehensive literature search to identify all published and unpublished randomised controlled trials. We will have no language restrictions.
Electronic searches
We will search the following electronic databases via OvidSP (from database inception to present):
-
Cochrane Central Register of Controlled Trials (CENTRAL) (Appendix 1);
-
MEDLINE (Appendix 2);
-
Embase (Appendix 3).
Searching other resources
We will search conference abstracts from the American Society of Clinical Oncology (ASCO) and the European Society of Medical Oncology (ESMO).
We will check the reference lists of included studies and review articles for additional references.
We will also search the following clinical trial registers:
-
International Standard Randomised Controlled Trials Number (ISRCTN) registry (www.controlled-trials.com);
-
World Health Organization International Clinical Trials Registry Platform (www.who.int/trialsearch);
-
US National Institutes of Health ongoing trials register (www.clinicaltrials.gov);
-
Chinese Clinical Trial Registry (www.chictr.org.cn);
-
European Union (EU) Clinical Trials Register (www.clinicaltrialsregister.eu);
-
Australian New Zealand Clinical Trials Registry (www.anzctr.org.au).
For ongoing trials and completed trials without data publication, principal investigators or trial sponsors will be contacted to obtain preliminary or unpublished data.
Data collection and analysis
Selection of studies
Two review authors (JRL and KLY) will independently screen titles and abstracts for inclusion, using EndNote (EndNote 2020). Duplicate records will be excluded at this stage. All potentially relevant studies will be coded as either 'potentially eligible' or 'not eligible'. We will retrieve the full texts of potentially eligible studies, and two review authors (JRL and KLY) will independently screen the full texts, identify studies for inclusion, and record exclusion reasons of the ineligible studies. We will identify and exclude duplicates and collate multiple reports of the same study so that each study, rather than each report, is the unit of interest in the review. In the event of a disputed qualification, an additional review author (CMB) will arbitrate through discussion. We will list studies excluded at the full text stage in the 'Characteristics of excluded studies' table, with reasons for exclusion. We will illustrate the selection process in a PRISMA flow diagram (Page 2021).
Data extraction and management
We will use a standardised data collection form for study characteristics and outcome data. One review author (JRL) will independently extract study characteristics from included studies. We will extract the following study characteristics:
-
Participant: median age, age range, sex, inclusion criteria, exclusion criteria, diagnostic criteria (definition of resectable or borderline resectable tumour used in trials);
-
Intervention: detailed treatment regimen;
-
Comparison: type of surgery performed;
-
Outcome: primary and secondary outcomes;
-
Study: author, design, sample size, country, date, source of funding.
Two review authors (JRL and KLY) will independently extract outcome data from included studies using Excel. In the 'Characteristics of included studies' table, we will note if the study authors reported outcome data in an unusable way. We will resolve disagreements by discussion or by involving a third review author (CMB). One review author (JRL) will copy across the data from the data collection form into RevMan 5.4 (Review Manager 2020). We will double‐check that the data is entered correctly by comparing the study reports with how the data are presented in the systematic review. A second review author (KLY) will spot‐check study characteristics for accuracy against the trial report.
Assessment of risk of bias in included studies
Two review authors (JRL and KLY) will use Cochrane’s original ‘Risk of bias’ tool (RoB 1) to independently assess the risk of bias for each included study (Higgins 2011). We will resolve any disagreement by discussion, or by involving a third review author (XL). We will assess the risk of bias according to the following domains:
-
Random sequence generation;
-
Allocation concealment;
-
Blinding of participants and personnel;
-
Blinding of outcome assessment;
-
Incomplete outcome data;
-
Selective outcome reporting;
-
Other bias.
Related information will be included in the 'Risk of Bias' table, constructed using RevMan 5.4 (Review Manager 2020).
We will use RevMan 5.4 (Review Manager 2020) to assess each bias domain in each included study to be at 'high', 'low', or 'unclear' risk of bias. We will provide a quote from the study report and justification for our judgement. We will summarise the 'Risk of bias' judgements across studies for each of the domains listed. If insufficient information is available to assess the risk of bias, we will contact the study authors to obtain or clarify the required information.
Assessment of bias in conducting the systematic review
We will conduct the review according to this published protocol and report any deviations from it in the 'Differences between protocol and review' section of the systematic review.
Measures of treatment effect
We will use the following measures of treatment effect.
-
For time‐to‐event outcomes (survival data), we will use the HR with its 95% CI.
-
For dichotomous outcomes (e.g. resection rates, adverse events, deaths), we will use the risk ratio (RR) with its 95% CI.
We will undertake the meta‐analysis only where this is meaningful (i.e. where the treatments, participants, and the underlying clinical question are similar enough for pooling to make sense).
Where multiple trial arms are reported in a single trial, we will include only the relevant arms. If two comparisons must be entered into the same meta‐analysis, we will halve the control group to avoid double‐counting.
Unit of analysis issues
We will only include individually randomised clinical trials where the unit of the analysis is the patient.
Dealing with missing data
We will contact investigators or study sponsors to verify key study characteristics and obtain missing numerical outcome data as indicated (e.g. when a study is identified as abstract only). If the HR is not reported by the authors, we will estimate it according to the practical guidance provided by Tierney et al. (Tierney 2007). If the dichotomous outcomes are not reported, we will include all randomised participants, assume that any dropouts do not have the event of interest, and use complete cases only.
Assessment of heterogeneity
We will use the I² statistic (I² > 50% indicates substantial heterogeneity) and the Chi² test (P > 0.10 indicates significant heterogeneity) to explore heterogeneity among the trials in each analysis (Higgins 2003). If we identify substantial or significant heterogeneity, we will explore it by pre‐specified subgroup analysis.
Assessment of reporting biases
If we are able to pool data from 10 or more trials, we will create and examine a funnel plot to explore possible publication biases. We will use Egger's test to determine the statistical significance of the reporting bias (Egger 1997). We will consider P < 0.05 to be a statistically significant reporting bias.
Data synthesis
Due to the clinical diversity, we will use random‐effects model to perform the meta‐analysis, using RevMan 5.4 (Review Manager 2020). For testing the robustness of our findings, regardless of which method was chosen, we will conduct sensitivity analyses for primary outcomes, using fixed‐effect models. In case of divergence between the two models, we will consider the results are not robust to the different assumptions of the two models, and will be cautious about our conclusions.
If meta‐analysis is not possible, we will undertake a narrative review of the findings.
Subgroup analysis and investigation of heterogeneity
We will carry out the following subgroup analyses:
-
neoadjuvant chemotherapy versus neoadjuvant chemoradiotherapy;
-
resectable pancreatic cancer versus borderline resectable pancreatic cancer;
-
tumour located in head of pancreas versus tumour located in body/tail of pancreas;
-
moderate to high risk of bias versus low risk of bias.
The following outcomes will be used in subgroup analysis:
-
overall survival;
-
resection rate;
-
R0 resection rate.
Sensitivity analysis
We will perform a sensitivity analysis (fixed‐effect model versus random‐effects model) to assess the robustness of our conclusions.
Reaching conclusions
We will only base our conclusions on findings from the quantitative or narrative synthesis of studies included in this review. We will avoid making recommendations for practice; our implications for research will give the reader a clear sense of the needed focus of future research and remaining uncertainties in the field.
Summary of findings and assessment of the certainty of the evidence
We will create a 'Summary of findings' table with the following outcomes:
-
overall survival;
-
resection rate;
-
R0 resection rate;
-
disease‐free survival;
-
toxicity;
-
operative morbidity;
-
operative mortality.
We will use the five GRADE considerations (study limitations, consistency of effect, imprecision, indirectness, and publication bias) to assess the certainty of the body of evidence for each outcome, based on the studies that contributed data to the respective meta‐analyses. We will classify the certainty of evidence as 'high certainty', 'moderate certainty', 'low certainty' or 'very low certainty'. We will use the GRADE methods and recommendations described in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2021). We will use the GRADEpro GDT software (GRADEpro 2015). We will use footnotes to justify all decisions to downgrade or upgrade evidence certainty of the included studies, and where necessary, will provide comments to aid the reader's understanding of the review. We will consider whether there is additional outcome information that was not incorporated into the meta‐analysis, note this in the comments, and state if it supports or contradicts the information from the meta‐analysis.
| Vascular | Definition | National Comprehensive Cancer Network (NCCN) (NCCN 2021) | MD Anderson Cancer Center (MDACC) (Varadhachary 2006) | American Hepato‐Pancreato‐Biliary Association/Society for Surgery of the Alimentary Tract/Society of Surgical Oncology (AHPBA/SSAO/SSO) (Callery 2009) | Alliance for Clinical Trials in Oncology (Alliance) (Katz 2016) |
| Celiac axis (CA) | Resectable | No arterial tumour contact | No extension | Clear fat planes | No radiographic interface |
| Borderline resectable | Solid tumour contact with the CA ≤ 180°, or > 180° without involvement of the aorta and with intact and uninvolved gastroduodenal artery (for body/tail) | No extension | Clear fat planes | Interface between tumour and vessel measuring < 180° of the circumference of the vessel wall | |
| Locally advanced | Solid tumour contact of > 180° (for head/uncinate process) Solid tumour contact with the CA and aortic involvement (for body/tail) | Encased and no technical option for reconstruction | Circumferential encasement | Interface between the tumour and the SMA measuring ≥ 180° of the circumference of the vessel wall | |
| Superior mesenteric artery (SMA) | Resectable | No arterial tumour contact | No extension | Clear fat planes | No radiographic interface |
| Borderline resectable | Solid tumour contact with the SMA ≤ 180° | Tumor abutment ≤ 180° | Tumor abutment ≤ 180° | Interface between tumour and vessel measuring < 180° of the circumference of the vessel wall | |
| Locally advanced | Solid tumour contact with the SMA > 180° | Encased (> 180°) | Circumferential encasement | Interface between the tumour and the SMA measuring ≥ 180° of the circumference of the vessel wall | |
| Common hepatic artery (CHA) | Resectable | No arterial tumour contact | No extension | Clear fat planes | No radiographic interface |
| Borderline resectable | Solid tumour contact with CHA without extension to CA or hepatic artery bifurcation allowing for safe/complete reconstruction | Short‐segment encasement/abutment of the common hepatic artery (typically at the gastroduodenal origin) | Short segment encasement or direct abutment | Reconstructable, short‐segment interface between tumour and vessel of any degree | |
| Locally advanced | Not mentioned | Encased and no technical option for reconstruction | Circumferential encasement | Long‐segment interface (of any degree) between the tumour and the CHA or its major tributaries with insufficient artery proximal and distal to the interface to perform reconstruction | |
| Superior mesenteric vein/portal vein (SMV/PV) | Resectable | No tumour contact with the SMV or PV or ≤ 180° contact without vein contour irregularity | Patent | No radiographic evidence of abutment, distortion, tumour thrombus, or venous encasement | Interface between the primary tumour and the SMV/PV measuring < 180° of the circumference of the vessel wall |
| Borderline resectable | Solid tumour contact with the SMV/PV of > 180°, contact of ≤ 180° with contour irregularity of the vein or thrombosis of the vein but with suitable vessel proximal and distal to the site allowing for safe and complete reconstruction Solid tumour contact with the inferior vena cava (IVC) | Short‐segment occlusion with suitable vessel above and below for reconstruction | Venous involvement of the SMV/PV demonstrating tumour abutment with or without impingement and narrowing of the lumen Encasement of the SMV/PV but without encasement of the nearby arteries Short segment venous occlusion resulting from either tumour thrombus or encasement but with suitable vessel proximal and distal to the area of vessel involvement allowing for safe resection and reconstruction | Interface between tumour and vessel measuring > 180° of the circumference of the vessel wall, and/or reconstructable occlusion | |
| Locally advanced | Unreconstructible SMV/PV due to tumour involvement or occlusion | Occluded and no technical option for reconstruction | Major venous thrombosis of the SMV/PV extending for several centimetres | Occlusion of the SMV/PV without a sufficient cuff of normal vein above and below the level of obstruction with which to perform venous reconstruction | |
| Abbreviations | |||||
