Pancreatic Cancer Resectability After Neoadjuvant Treatment: Understanding the Imaging Challenge

Table of Contents

• Key Points
• Background: Why This Research Matters
• How This Research Was Conducted
• Key Findings: What the Research Revealed
• Arterial vs. Venous Involvement: Different Behaviors
• New Imaging Signs of Treatment Response
• Why Tumor Size Changes Can Be Misleading
• Optimal Timing for Imaging Assessment
• Clinical Implications for Patients
• Limitations of Current Imaging
• Recommendations for Patients and Care Teams
• Frequently Asked Questions
• Source Information

Key Points

• After neoadjuvant therapy, CT often overestimates vessel involvement because scar tissue can look like tumor; in one study, specificity for R0 resectability fell from about 88% to 52%.
• In one series, only about one-third of artery segments still encased more than 180 degrees on CT had actual cancer invasion of the artery.
• In one study, partial regression of tumor contact with the superior mesenteric or portal vein was linked to R0 resection in 100% of cases.
• In reported series, only about 10% to 15% of patients had tumor shrinkage on CT, yet R0 resection rates among those operated on were often greater than 90%.
• A repeat multidetector CT is recommended within 25 days of any planned definitive operation to avoid finding metastatic disease unexpectedly at surgery.

Background: Why This Research Matters

Pancreatic ductal adenocarcinoma (PDAC), the most common type of pancreatic cancer, remains one of the most challenging cancers to treat. The disease has a rising incidence worldwide and continues to have a poor prognosis. The five-year survival rate remains below 10%, meaning fewer than 1 in 10 patients survive five years after diagnosis.

This poor outcome is primarily due to several factors. Pancreatic cancer tends to spread systemically early in the disease process. The cancer also invades nerves and frequently wraps around major blood vessels. These characteristics limit the number of patients who can undergo surgery that could potentially cure the disease.

At the time of diagnosis, most patients already have locally advanced cancer that has spread to nearby tissues or metastatic disease that has spread to distant organs. For patients with localized tumors that have not spread, surgical removal followed by chemotherapy offers the only potentially curative option. Achieving negative surgical margins, called R0 resection, is critical for long-term survival. This means no cancer cells are found at the edges of the removed tissue.

Initial Imaging and Staging

Contrast-enhanced multidetector CT (computed tomography) is the standard imaging method for initially staging pancreatic cancer. Medical teams use classification systems based on pancreatic protocol CT findings to categorize tumors into four groups:

• Resectable: The tumor shows no arterial or significant venous involvement, with clear fat planes around the superior mesenteric artery (SMA), celiac axis, and portal vein (PV)
• Borderline resectable (BRPC): The tumor shows limited vascular involvement, also called vascular abutment, meaning 180 degrees or less of circumferential contact with major arteries or short-segment venous narrowing that may allow vascular resection or reconstruction
• Locally advanced (LAPC): The tumor shows more extensive involvement, meaning more than 180 degrees of encasement of critical arteries or occlusion of the superior mesenteric vein/portal vein confluence that cannot be reconstructed, traditionally precluding immediate surgery
• Unresectable: The tumor has distant metastases or diffuse vascular encasement

Pancreatic Protocol CT Technique

High-quality multiphasic CT is crucial for optimal tumor detection and blood vessel assessment. A dedicated pancreatic CT protocol typically includes thin-section imaging of 3 millimeters or less with intravenous contrast during two phases. The late arterial phase, also called the pancreatic parenchymal phase, occurs approximately 40 to 50 seconds after injection. The portal venous phase occurs approximately 65 to 70 seconds after injection. Multiplanar reconstructions allow detailed evaluation.

This technique maximizes how prominently the tumor appears. Pancreatic ductal adenocarcinoma classically appears hypoenhancing, meaning it takes up less contrast than normal pancreas tissue. This allows accurate delineation of tumor-vessel interfaces and involvement of surrounding organs and structures. If initial CT images are not acquired using a pancreas protocol, repeating imaging with an appropriate technique is recommended to ensure precise staging. Using proper imaging protocols improves diagnostic accuracy and confidence in determining whether a tumor can be removed surgically.

Neoadjuvant Therapy and the Need to Reassess Resectability

The introduction of neoadjuvant therapy (NAT) has redefined the treatment plan for patients with borderline resectable and locally advanced pancreatic cancer. NAT includes multiagent chemotherapy with or without radiation before surgery. Neoadjuvant therapy aims to accomplish several goals:

• Shrink tumors to increase the likelihood of complete R0 resection by sterilizing margins
• Eliminate micrometastatic disease that may not be visible on imaging
• Allow only biologically favorable tumors to proceed to surgery

There is growing evidence that neoadjuvant therapy can convert initially unresectable or borderline tumors into candidates for curative resection. This is true even when arterial involvement is present, which was once considered a barrier to surgery. Surgical series have reported successful resections of tumors encasing the celiac axis or superior mesenteric artery following robust responses to neoadjuvant therapy. These cases historically would have been considered inoperable.

Despite these therapeutic advances, radiologic assessment of treatment response and post-neoadjuvant therapy resectability remains challenging. Therapy-induced fibrosis (scar tissue), necrosis (tissue death), and inflammation can persist as mass-like tissue on imaging. This obscures the distinction between residual tumor and scar-desmoplastic reaction, which is the fibrous tissue that forms around tumors.

Contrast-enhanced CT remains the primary imaging method for restaging after neoadjuvant therapy, but it has well-recognized limitations. Persistent soft-tissue encasement of more than 180 degrees around arteries or veins, a finding historically equated with unresectability, may actually represent fibrotic desmoplastic reaction rather than viable tumor after treatment.

Similarly, tumor size on CT may not change appreciably despite a major pathological response. This often occurs due to stromal collapse or therapy-related edema (swelling). Conventional radiographic criteria, such as absolute tumor diameter or static degrees of vessel contact, often fail to correlate with histopathologic response following treatment. These limitations underscore the need for a redefined post-neoadjuvant therapy imaging assessment that goes beyond pre-treatment anatomic criteria and incorporates both morphologic and dynamic features.

Complementary imaging modalities and techniques are being explored to improve post-therapy evaluation. Perfusion CT, diffusion-weighted MRI (DW-MRI), and PET/CT can provide functional insights into tumor biology, including vascular perfusion, cellularity, and metabolic activity. These techniques may help distinguish fibrosis from residual tumor.

Early studies have shown that changes in perfusion parameters or ADC values on DW-MRI can correlate with treatment response. PET with 18F-FDG can demonstrate metabolic tumor response that sometimes precedes morphologic change. However, these techniques lack standardized response criteria and can lead to variable results. CT remains the cornerstone of assessment, often supplemented with these modalities. Additionally, evolving surgical strategies, including planned vascular resections and reconstructions, have broadened the surgical candidacy of patients with residual vascular involvement following neoadjuvant therapy.

How This Research Was Conducted

The authors performed a comprehensive literature search to examine the evolving imaging criteria used to assess pancreatic cancer resectability following neoadjuvant therapy. They systematically queried the PubMed database for English-language articles published in the last 25 years focusing on post-neoadjuvant therapy radiologic evaluation.

Search terms included combinations of "pancreatic cancer," "neoadjuvant therapy," "CT," "resectability," "vascular involvement," and "imaging response." Both prospective and retrospective studies were included. Relevant meta-analyses and consensus guideline papers were also included.

Selection priority was given to studies directly evaluating the accuracy of CT in predicting surgical resectability or pathologic response after neoadjuvant therapy. Priority was also given to studies evaluating other imaging modalities like MRI or PET. Studies dealing with only pre-treatment imaging were excluded. Small case series or single case reports were also excluded to maintain robust evidence.

In total, the authors identified several relevant studies and extracted data on imaging findings such as tumor size and vessel involvement. They also extracted data on how these findings correlated with surgical and pathological outcomes, as well as any proposed new criteria or qualitative signs. These findings were synthesized qualitatively for this narrative review. No formal meta-analysis was performed given the heterogeneity in study designs and endpoints, but consensus trends across the literature were emphasized.

Key Findings: What the Research Revealed

The evidence synthesized from multiple studies reveals a paradigm shift in the radiologic assessment of pancreatic cancer resectability after neoadjuvant therapy. Collectively, these data demonstrate that traditional CT criteria are insufficiently reliable for predicting surgical resectability or true residual tumor status in the post-neoadjuvant therapy setting. These criteria include tumor size and the extent of tumor-vessel contact.

A recurring finding is a marked reduction in the specificity of CT for detecting actual tumor invasion of vessels after neoadjuvant therapy. Specificity refers to how well a test correctly identifies patients who do not have the condition. CT frequently overestimates residual tumor involvement of vessels, likely due to the presence of fibrosis following therapy that maintains a mass effect or encasement appearance.

Specific Study Results

In a study by Cassinotto and colleagues, the specificity of CT for R0 resectability dropped from approximately 88% before therapy to approximately 52% after neoadjuvant therapy. CT often overcalled vascular invasion that was not found at surgery. The researchers noted that 12 of 31 patients who achieved complete R0 resection after neoadjuvant therapy had been interpreted as "high risk of incomplete resection" on restaging CT. This interpretation was due to apparent vascular involvement.

Similarly, White and colleagues observed that restaging CT after chemoradiation tended to overstage disease. This excluded some patients from surgery who actually could have undergone R0 resection. These findings underscore that using static pre-treatment CT cutoffs, such as more than 180 degrees of arterial encasement, in the post-treatment context will often underestimate resectability.

A particularly illustrative result comes from Mayer and colleagues, who specifically evaluated arterial encasement greater than 180 degrees in post-neoadjuvant therapy CT. They found that only approximately one-third of arterial segments that still appeared encased more than 180 degrees on CT had actual histologic tumor invasion of the artery. In their series, many patients with persistent circumferential arterial contact on CT were nevertheless able to undergo successful R0 resections.

This suggests that a substantial proportion of what looks like artery encasement after therapy is actually due to desmoplastic reaction or inflammatory tissue rather than viable cancer.

Venous Involvement Findings

The same appears true for veins. Multiple studies reported that persistent vein narrowing or occlusion on CT after neoadjuvant therapy does not always signify unresectability. The vein can often be safely resected or even found patent at surgery with only fibrous adhesions remaining.

Cassinotto and colleagues noted that even when the superior mesenteric vein (SMV) or portal vein remained stenosed on post-therapy CT, this finding was not predictive of an R1 resection. R1 resection means cancer cells are found at the margin of the removed tissue. In their cohort, vein narrowing often did not equate to residual tumor at the margin.

In contrast, when they did observe any partial regression of the tumor-vessel interface on CT, it was a very positive sign. This was true even if the vessel was still contacted but to a lesser degree or over a shorter length. Partial regression of tumor contact with the superior mesenteric vein or portal vein was associated with R0 resection in 100% of cases in that study. Partial regression of contact with any major vessel, whether arterial or venous, predicted R0 resection in 91% of cases.

These statistics highlight how changes in the tumor-vessel interface are more prognostically meaningful than the static presence of contact itself.

Arterial vs. Venous Involvement: Different Behaviors

Post-neoadjuvant therapy imaging interpretation must distinguish between arterial and venous responses because they behave differently.

Arterial Involvement

CT signs such as more than 180 degrees of encasement are no longer definitive for unresectability. Desmoplastic tissue may encase arteries without actual invasion. Studies suggest that partial or incomplete arterial contact after neoadjuvant therapy is often resectable, especially if accompanied by halo signs or preserved vessel contour.

Venous Involvement

Veins such as the superior mesenteric vein and portal vein frequently recanalize after neoadjuvant therapy. Recanalization means the vein opens up again. Persistent occlusion may still be surgically manageable. The restoration of patency or improved caliber is a strong indicator of treatment response. Persistent narrowing alone is not predictive of R1 resections and should not preclude exploration.

New Imaging Signs of Treatment Response

Another key observation across the literature is that qualitative changes in perivascular tissue and vessel patency on CT correlate better with outcomes than do gross tumor size changes. Zins and colleagues have emphasized that simple metrics like tumor diameter or even CT attenuation change have poor correlation with pathologic response or tumor resectability.

Instead, radiologists are noting several features as more indicative of a favorable response:

• Reduction in the circumferential degree of vessel contact
• Restoration of a normal vessel contour or lumen caliber
• Development of a perivascular hypoattenuating "halo" replacing what was previously solid tumor around a vessel

The Halo Sign

The "halo sign" refers to the appearance of a low-density rim between the tumor and vessel on post-neoadjuvant therapy CT where previously the tumor was in direct contact. This is essentially an ill-defined fatty or scar tissue plane enveloping the vessel. Its appearance has been linked with tumor regression and the potential for a safe dissection plane surgically.

In one series, the presence of a low-contrast perivascular halo of approximately 46 Hounsfield units or less adjacent to an artery was identified. This was the unique CT feature distinguishing cases where the artery was not truly invaded. Hounsfield units are a measure of density on CT scans.

Why Tumor Size Changes Can Be Misleading

Importantly, even cases without vast radiologic tumor shrinkage can still be resectable. Several studies documented that objective RECIST partial responses on imaging are relatively uncommon with neoadjuvant therapy. RECIST stands for Response Evaluation Criteria in Solid Tumors, a standard way to measure tumor shrinkage. Often, the majority of cases are radiologically "stable disease," meaning the tumor size has not changed significantly.

Yet a substantial subset of these patients still achieve R0 resections and major pathologic tumor clearance. Marthey and colleagues and Xia and colleagues reported on borderline and locally advanced pancreatic cancer patients who completed neoadjuvant therapy and went to surgery. Only approximately 10% to 15% of these patients showed partial tumor shrinkage on CT. Approximately 70% had stable disease by size criteria.

Despite this, R0 resection rates in these surgical patients were high, often greater than 90% in responders. This indicates that the lack of a radiologic tumor size reduction does not equate to the lack of a treatment effect. Histopathologically, many of these "stable disease" tumors have extensive fibrosis or treatment-induced necrosis, explaining why they remained radiologically unchanged in size.

The meta-analysis of Suker and colleagues noted that many patients with stable or minimal response on imaging following neoadjuvant therapy with FOLFIRINOX still underwent successful resections. FOLFIRINOX is a combination chemotherapy regimen used for pancreatic cancer.

Meta-Analysis Results

Similar findings come from larger meta-analyses and series. A pooled analysis by Gillen and colleagues found that neoadjuvant therapy enabled surgical resection in roughly 26% of patients with initially unresectable or borderline tumors. The overall R0 resection rate was approximately 77% among those resected.

More recent multi-institution studies, such as those by Ferrone and colleagues, report even higher conversion rates. In Ferrone's series, approximately 30% of locally advanced pancreatic cancer patients were ultimately resected after FOLFIRINOX. Remarkably, 92% of those who reached the operating table had an R0 resection.

Notably, the majority, approximately 70%, of Ferrone's patients still had tumors appearing "borderline or unresectable" on post-neoadjuvant therapy CT. This underscores how imaging often did not reflect the true resectability that became evident intraoperatively. In their cohort, nearly half of the patients deemed unresectable by imaging after neoadjuvant therapy were found to be resectable at surgery, with successful margin-negative outcomes in most.

These outcomes collectively reinforce the idea that persistent CT findings of vessel involvement after neoadjuvant therapy, especially venous involvement, should not be viewed as an absolute contraindication to exploration. Instead, a comprehensive evaluation should guide the decision of whether to proceed to operation. This evaluation should incorporate interval imaging changes rather than static findings, tumor marker dynamics such as CA 19-9 drop, and clinical status.

Optimal Timing for Imaging Assessment

The optimal timing for imaging assessment after pancreatic cancer treatment depends on the treatment type and the specific goals of the imaging. Generally, imaging is performed 3 to 6 months after treatment to assess response and detect recurrence.

However, for patients with resectable disease, imaging may be performed closer to the surgical date. Patients undergoing neoadjuvant therapy may have imaging scans conducted more frequently to monitor treatment response.

Patients should undergo a repeat multidetector CT within 25 days of any planned definitive operative intervention for pancreatic cancer. This helps avoid unexpectedly finding metastatic disease at surgery.

Clinical Implications for Patients

The findings of this review have several important implications for patients with pancreatic cancer considering surgery after neoadjuvant therapy.

Imaging Alone Should Not Determine Surgical Candidacy

Persistent vessel contact on CT after neoadjuvant therapy does not necessarily mean a patient cannot have surgery. Many patients with what appears to be persistent vascular involvement on imaging are still able to undergo successful complete resections. The imaging findings often represent scar tissue rather than active cancer.

Changes Over Time Matter More Than Static Findings

How the tumor-vessel relationship changes between scans is more important than any single snapshot. Partial regression of tumor contact with blood vessels, even if some contact remains, is a strong predictor of successful complete resection. Patients should discuss with their care team how their imaging findings have evolved over time.

Multiple Factors Guide Surgical Decisions

The decision to proceed with surgery after neoadjuvant therapy should incorporate multiple factors beyond imaging alone:

• Interval changes on imaging rather than static findings
• Tumor marker dynamics, such as a significant drop in CA 19-9 levels
• Patient performance status and overall fitness for surgery
• Absence of metastatic disease
• Input from a multidisciplinary team including surgeons, radiologists, oncologists, and pathologists

The Role of Multidisciplinary Assessment

A collaborative decision-making process that incorporates clinical findings, imaging results, laboratory data, patient comorbidities, and preferences is essential. This process typically involves input from surgeons, radiologists, oncologists, pathologists, and other relevant specialists to ensure a well-rounded and individualized treatment plan.

Radiologic findings should be considered alongside serum tumor markers. For example, a dramatic drop in CA 19-9 strengthens the argument that a radiologically static mass may be largely fibrotic. The patient's overall condition also matters.

Surgeons experienced in pancreatic cancer should be directly involved in imaging review. Sometimes a surgeon may judge that even though a vessel looks involved, they have strategies to handle it. This is especially true if other indicators like lack of metastases and patient fitness are favorable.

The decision to proceed to surgery after neoadjuvant therapy often comes down to whether there is any chance for R0 resection and no evidence of metastatic disease. Given the disconnect between radiographic appearance and pathologic reality in many cases, tumor boards increasingly err on the side of surgical exploration. This occurs if imaging is equivocal but other factors are encouraging. This approach has led to improved outcomes for some patients who previously would have been deemed unresectable from imaging alone.

Limitations of Current Imaging

Despite improvements in understanding, post-neoadjuvant therapy imaging faces inherent limitations. CT cannot reliably distinguish live tumor cells from fibrotic tissue on a microscopic level. Some patients will have residual microscopic cancer in areas that appear "clean" on CT, leading to positive margins or early recurrence despite favorable imaging.

Conversely, CT can also underestimate response in some cases. For instance, a treated tumor may not shrink much but could be mostly necrotic. Unless there are calcifications or obvious necrotic patterns, CT might label it as stable disease when it has actually responded well to treatment.

These limitations mean that imaging is one piece of the puzzle rather than the complete answer. Ongoing research into imaging biomarkers and the adoption of advanced analytics hold promise for closing the gap between radiographic assessment and the ground truth of tumor viability.

Recommendations for Patients and Care Teams

Based on this comprehensive review, several recommendations emerge for patients with pancreatic cancer undergoing neoadjuvant therapy and their care teams.

For Patients

1. Understand that imaging findings after neoadjuvant therapy do not always reflect the true status of your cancer. Persistent vessel contact on CT may represent scar tissue rather than active cancer.
2. Ask your care team to explain how your imaging findings have changed over time, not just what the current scan shows.
3. Discuss your CA 19-9 levels and how they have changed during treatment, as this blood marker provides additional information about treatment response.
4. Ensure your case is reviewed by a multidisciplinary tumor board with experience in pancreatic cancer before final decisions about surgery are made.
5. Seek care at a center with experienced pancreatic cancer surgeons who can evaluate whether surgery is feasible even when imaging appears concerning.

For Care Teams

1. Transition from static morphologic criteria to response-based interpretation when assessing resectability after neoadjuvant therapy.
2. Focus on interval changes in tumor-vessel relationships rather than absolute thresholds.
3. Recognize the halo sign and vessel contour restoration as favorable indicators of treatment response.
4. Consider venous involvement differently from arterial involvement, as veins frequently recanalize and persistent narrowing alone should not preclude exploration.
5. Integrate radiologic, biochemical, and clinical findings in multidisciplinary evaluation to guide surgical decision-making.
6. Ensure repeat multidetector CT within 25 days of any planned definitive operative intervention.

In summary, post-neoadjuvant imaging requires a paradigm shift in interpretation. Radiologists must embrace a more dynamic assessment, noting how tumor-vessel relationships have changed rather than just whether they are present. Understanding the differences in arterial and venous responses and recognizing signs like the halo can provide critical guidance. However, imaging is one piece of the puzzle. The integration of multiple factors through multidisciplinary evaluation is essential to guide surgical decision-making and improve patient outcomes.

Frequently Asked Questions

If my CT still shows the tumor touching a blood vessel after chemotherapy, does that mean I can't have surgery?

Not necessarily. In one study, only about one-third of artery segments that still looked encased more than 180 degrees on CT had actual cancer invasion. Much of what looks like vessel involvement after treatment is scar tissue, so persistent contact alone should not rule out surgery.

Why does my scan report say 'stable disease' when the treatment is supposed to be working?

Tumor size often does not shrink much even when treatment works. In reported series, only about 10% to 15% of patients showed partial shrinkage on CT, while about 70% had stable disease by size. Many of these tumors were mostly scar or dead tissue, and R0 resection rates were still high.

What does a drop in my CA 19-9 level mean for my chances of surgery?

CA 19-9 is a blood marker that adds information beyond imaging. A significant drop strengthens the argument that a mass that looks unchanged on CT may be largely scar tissue rather than active cancer. It should be considered alongside your scans, fitness for surgery, and multidisciplinary review.

What is the halo sign on a CT scan after treatment?

The halo sign is a low-density rim seen between the tumor and a blood vessel after treatment, where the tumor was previously in direct contact. It is thought to be a scar or fatty tissue plane. In one series, a perivascular halo of about 46 Hounsfield units or less distinguished arteries that were not truly invaded.

How soon before surgery should I have another CT scan?

A repeat multidetector CT should be done within 25 days of any planned definitive operation for pancreatic cancer. This helps avoid unexpectedly finding metastatic disease at the time of surgery. Imaging after treatment is generally performed 3 to 6 months afterward, or more often during neoadjuvant therapy.

Who decides whether I go to surgery after neoadjuvant therapy?

A multidisciplinary team should decide, including surgeons, radiologists, oncologists, and pathologists. The decision should weigh interval changes on imaging rather than one snapshot, tumor marker trends such as CA 19-9, your performance status, absence of metastatic disease, and your preferences. Imaging alone should not determine surgical candidacy.

Can a vein that is still narrowed or blocked after treatment still be removed surgically?

Often yes. Multiple studies found that persistent narrowing or blockage of the superior mesenteric vein or portal vein on CT after neoadjuvant therapy does not always mean the cancer cannot be removed. Veins frequently reopen after treatment, and persistent narrowing alone was not predictive of a positive margin.

My pancreatic tumor still looks like it's touching a major blood vessel after chemotherapy — should I get a second opinion before accepting that surgery isn't possible?

Yes. Persistent vessel contact on CT after neoadjuvant therapy often represents scar tissue rather than active cancer. Only about one-third of arterial segments still appearing encased beyond 180 degrees had true tumor invasion, and many patients with persistent contact still achieved complete R0 resections. Partial regression of tumor-vessel contact predicted R0 resection in 91% of cases. Because imaging frequently overestimates residual involvement, a review of how your scans changed over time, alongside CA 19-9 trends and multidisciplinary input, can clarify whether surgery remains feasible. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original Article Title: Pancreatic Cancer Resectability After Neoadjuvant Treatment: An Imaging Challenge.

Journal: Diagnostics (Basel, Switzerland)

License: CC BY (open access)

Authors: Christofilis I, Triantopoulou C, Delis S.

Author Affiliations: Radiology Department, Konstantopouleio Hospital, Athens, Greece; HPB Unit, Konstantopouleio Hospital, Athens, Greece

Publication: Diagnostics 2025, Volume 15, Article 1810

Publication Date: Published July 18, 2025

DOI: https://doi.org/10.3390/diagnostics15141810

Academic Editor: Antonio Facciorusso

Article Type: Review

This patient-friendly article is based on peer-reviewed research. The original article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license.