What Is Borderline Resectable Pancreatic Cancer?

Borderline resectable pancreatic cancer sits in a gray zone between tumors that surgeons can remove cleanly and those too entangled with nearby blood vessels for safe operation. The designation matters enormously because it changes the entire treatment sequence: instead of going straight to surgery, most patients receive chemotherapy first, and sometimes radiation, before a surgical team reassesses whether the tumor can come out. A 2024 meta-analysis of randomized trials found that this “neoadjuvant-first” approach improved overall survival specifically for borderline resectable disease, bumping median survival from roughly 19 months to 29 months compared with upfront surgery. The category is relatively common, too, making up about a quarter of pancreatic cancers evaluated at referral centers.

What “Borderline Resectable” Actually Means

Pancreatic cancer is classified by how much the tumor touches or encases the major blood vessels running near the pancreas, particularly the superior mesenteric artery, the celiac artery, the superior mesenteric vein, and the portal vein. When the tumor has limited contact with these vessels but not enough to rule out surgery entirely, it falls into the borderline resectable category. An international consensus published in 2017 formalized this with three dimensions: anatomical, biological, and conditional. The anatomical criteria focus on the degree of vessel contact. Tumor touching less than half the circumference of a major artery, or narrowing a vein without extending too far, qualifies as borderline rather than unresectable.¹

The biological dimension adds a layer most people do not expect. Even if the anatomy looks favorable, a blood marker called CA 19-9 above 500 units per milliliter or suspicious-looking lymph nodes on imaging can push a case into borderline territory. The conditional dimension accounts for the patient’s overall fitness: someone too frail for a major operation (rated at a performance status of 2 or higher on a standard scale) is also categorized as borderline, regardless of what the scan shows.² The International Study Group of Pancreatic Surgery has endorsed the National Comprehensive Cancer Network criteria for this definition, which are broadly similar in their anatomical thresholds.³

This three-dimensional approach is important because it acknowledges that “can this tumor be cut out” is not purely a question of anatomy. Two patients with identical scans can end up in different treatment pathways depending on their bloodwork and physical condition.

Why Chemotherapy Comes Before Surgery

For borderline resectable disease, the standard approach at most major centers is to start with several months of chemotherapy before anyone picks up a scalpel. The rationale is straightforward: shrinking the tumor and treating any microscopic cancer cells that may have already spread gives the surgeon a better chance of removing the tumor completely while also filtering out patients whose cancer progresses despite treatment, since those patients would not have benefited from a major operation.

The survival benefit of this neoadjuvant-first strategy is backed by pooled data from randomized trials. A 2021 meta-analysis found that neoadjuvant therapy improved overall survival with a hazard ratio of 0.66, translating to a jump in median survival from about 19 to 29 months. The benefit was driven by the borderline resectable subgroup specifically, where the hazard ratio was 0.61.⁴ A more recent 2024 meta-analysis confirmed this pattern: neoadjuvant treatment improved overall survival for borderline resectable cancer (hazard ratio 0.60) but not for tumors already classified as clearly resectable.⁵

One subtlety worth understanding: the overall resection rate is actually lower with neoadjuvant treatment than with upfront surgery (about 66% versus 81%), because some patients progress during chemotherapy and never make it to the operating room. But among those who do get surgery, the rate of clear margins, where no cancer cells are found at the cut edge, is considerably higher: roughly 87% after neoadjuvant treatment versus 67% with upfront surgery. Positive lymph nodes also drop substantially.⁶ In other words, neoadjuvant therapy selects for patients who can truly benefit from surgery and gives them a cleaner operation when they get one.

Choosing a Chemotherapy Regimen

The two most common regimens are FOLFIRINOX (a combination of four drugs) and gemcitabine plus nab-paclitaxel. Both are aggressive, and which one a patient receives often depends on age, fitness level, and how well they tolerate side effects. In a study of 193 patients, those who received FOLFIRINOX had a median survival of about 39 months versus 29 months for gemcitabine/nab-paclitaxel, though the unadjusted comparison did not reach statistical significance. An adjusted analysis estimated that FOLFIRINOX added roughly 5 months of survival. Patients on FOLFIRINOX were younger and fitter on average, which complicates the comparison, but they also had fewer positive lymph nodes at surgery.⁷

A more recent real-world analysis from a single center found a more dramatic difference in borderline resectable cases: median survival of 37 months with FOLFIRINOX versus 16 months with gemcitabine/nab-paclitaxel, along with a higher rate of conversion to surgery (66% versus 39%).⁸ These are single-center data, so the numbers should be taken as directional rather than definitive, but the overall pattern across studies leans in FOLFIRINOX’s favor for patients who can handle it physically. For older patients or those with significant other health conditions, gemcitabine/nab-paclitaxel remains a reasonable choice.

The Role of Radiation

Whether to add radiation therapy after chemotherapy and before surgery is one of the more debated questions in this space. The idea is that a focused dose of radiation to the area where the tumor meets blood vessels can sterilize the surgical margin and improve the chance of a clean resection. A systematic review and meta-analysis comparing FOLFIRINOX alone to FOLFIRINOX followed by radiation found similar overall resection rates but a higher rate of clear margins when radiation was added (about 98% versus 88%).⁹ Median overall survival was roughly comparable between the two approaches (about 22 months in both groups in that pooled analysis), leaving open the question of whether the margin benefit translates to longer life.

Stereotactic body radiation therapy (SBRT), which delivers high doses in just a few sessions, has gained traction. In one series, induction chemotherapy followed by SBRT (a median dose of 35 Gy in five fractions) achieved a margin-negative resection rate above 97%, and patients who showed a pathologic response to the combined treatment had improved survival.¹⁰ Still, a small study tracking recurrence patterns after chemotherapy and SBRT found that while local control was generally good, every patient who recurred had a distant component to their recurrence, meaning the cancer came back in the liver, lungs, or elsewhere rather than at the original tumor site.¹¹ Radiation can help win the local battle, but the systemic war against distant spread remains the bigger challenge.

CA 19-9 as a Treatment Compass

The blood marker CA 19-9 is one of the few reliable tools clinicians have to track how borderline resectable pancreatic cancer is responding to treatment before surgery. It is not a perfect test, and some patients do not secrete it at all, but when it is elevated and then monitored over a course of chemotherapy, its trajectory tells a meaningful story.

In a study of 141 borderline resectable patients, those whose CA 19-9 normalized (dropped below 40 units per milliliter) after neoadjuvant therapy had markedly longer survival regardless of whether they went on to have surgery. Resected patients whose marker normalized had a median survival of 38 months versus 26 months for those whose marker stayed elevated. Among patients who never reached the operating room, normalization still predicted better outcomes (15 versus 11 months). Failure to normalize CA 19-9 roughly doubled the risk of death.¹²

A separate analysis reinforced this by defining an “optimal” CA 19-9 response and finding that patients who hit that benchmark had a median survival of about 39 months, while those with a suboptimal response lived a median of roughly 22 months. Strikingly, the suboptimal responders did no better than patients who went directly to surgery without any neoadjuvant therapy at all, and the only group with a reduced rate of liver metastases was the optimal responders.¹³ On the alarming end, a striking rise in CA 19-9 after neoadjuvant therapy (above 200 units per milliliter) predicted very poor outcomes even with surgery, with only about 41% of those patients alive at two years.¹⁴

Taken together, CA 19-9 trends are one of the most practical tools for deciding whether to proceed with surgery or pivot to a different treatment plan.

Why Scans Can Be Misleading After Chemotherapy

One frustrating reality of borderline resectable pancreatic cancer is that CT scans become less reliable after neoadjuvant treatment. Chemotherapy and radiation can cause scarring and inflammation around blood vessels that mimics tumor invasion on imaging, leading radiologists to overestimate how entangled the tumor still is. A study comparing CT accuracy before and after neoadjuvant therapy found that accuracy in predicting a clean resection dropped from 83% to 58%. Specificity plummeted from 88% to 52%, meaning that nearly half the time the scan suggested the tumor could not be removed completely, the surgeon actually achieved a clean resection.¹⁵

This has major practical implications. If a multidisciplinary team relies solely on post-treatment imaging to decide whether to operate, some patients with good biology (dropping CA 19-9, good performance status) will be denied a potentially curative surgery because their scan looks worse than the actual situation. Experienced centers weigh imaging alongside biomarker trends and clinical assessment rather than treating the scan as the final word.

Staging Laparoscopy and Hidden Metastases

Before committing to a major operation, many centers perform a staging laparoscopy: a minimally invasive procedure where a camera is inserted through a small incision to look for tiny tumor deposits in the abdomen that are invisible on CT or MRI. The yield of this step is surprisingly high. In an analysis of over 1,000 consecutive patients, 18% had findings on laparoscopy that changed their treatment plan, including visible metastases or positive peritoneal wash results. Among patients who had not yet received chemotherapy, the rate of occult disease was even higher at 23%.¹⁶

A separate cohort study found that a third of patients had radiologically invisible metastases detected at staging laparoscopy, including peritoneal deposits, liver metastases, and positive wash cytology.¹⁷ This procedure can spare patients the morbidity of an open operation that would have been futile. Factors associated with a positive finding include elevated CA 19-9, larger tumor size, and tumors in the body or tail of the pancreas.

The Surgery Itself

When a borderline resectable tumor involves veins, the surgeon often needs to remove and reconstruct a segment of the portal vein or superior mesenteric vein alongside the tumor. Vascular involvement in pancreatic cancer ranges from roughly 21% to 64% of cases, and at high-volume centers, vein resection and reconstruction accounts for 20% to 25% of all pancreatic head removals.¹⁸ A single-center series of 63 patients who underwent this combined procedure (either laparoscopic or open) reported no perioperative deaths and vessel patency above 92% in both groups. The laparoscopic approach had less blood loss and shorter hospital stays, though longer operating times.¹⁹ Laparoscopic vein resection and reconstruction appears safe and feasible in selected patients at experienced centers, with perioperative outcomes and long-term prognosis similar to the open approach.²⁰

Arterial resection is a more aggressive step and remains controversial. When the tumor encases an artery, some high-volume centers will resect and reconstruct it rather than declaring the patient unresectable. A study from one such center found that arterial resection carried acceptable morbidity comparable to vein-only resection, though it also noted that arterial involvement is often overestimated on imaging, meaning that less invasive techniques like “divestment” (carefully peeling the tumor off the artery wall) can sometimes be used instead.²¹ A decade-long comparative study found no fatalities among patients undergoing arterial resection, and complication rates improved over time with growing experience.²² These operations should only happen at centers that do them regularly.

Getting the Diagnosis Confirmed

Before starting chemotherapy, a tissue diagnosis is required. Most borderline resectable tumors are biopsied using endoscopic ultrasound-guided fine needle aspiration or biopsy (EUS-FNA or FNB). A population-based cohort study found that the first attempt at EUS-guided biopsy achieved a diagnostic accuracy of about 86%, and cumulative accuracy across all attempts reached 96%, with a false positive rate for malignancy under 1%.²³ This means the procedure is highly reliable, though a small minority of patients need a repeat biopsy before treatment can begin.

The Value of a Multidisciplinary Team

Borderline resectable pancreatic cancer is one of the diagnoses where a multidisciplinary tumor board review most clearly changes patient care. At a tertiary referral center that tracked 184 patients presented for resectability assessment, the tumor board changed the resectability classification in about a quarter of cases. Among patients initially called borderline, about 35% were reclassified as locally advanced, and a few were reclassified as metastatic after the full team reviewed imaging and clinical data. Critically, about 19% of patients had their overall disease management changed by the board discussion. When the board’s assessment was tested against what surgeons actually found in the operating room, it proved accurate in over 91% of cases.²⁴

This is not an abstract recommendation. A patient whose community hospital reads their scan as borderline resectable may be told by a specialized team that the disease is actually locally advanced and needs a different treatment strategy, or may be told it is actually operable sooner than expected. Either correction makes a meaningful difference.

Quality of Life After Treatment

Patients understandably want to know what life looks like on the other side of this journey. A study specifically tracking borderline resectable and locally advanced patients who completed neoadjuvant chemotherapy and then underwent surgery found that about a quarter experienced major complications. However, quality of life scores, including overall well-being, social functioning, and role functioning, improved significantly within three months of surgery. Pain, breathlessness, appetite loss, and diarrhea all improved after resection.²⁵

Broader studies of pancreatic surgery suggest that full recovery of quality of life typically takes somewhere between 3 and 12 months after the Whipple procedure, with the main ongoing issues being changes in digestive enzyme function and blood sugar control.²⁶ Compared with population averages, patients who have had pancreatic surgery report lower scores in role functioning but, interestingly, less social isolation. Patients who had surgery for cancer reported more fatigue, pain, and worry about the future compared with those who had surgery for benign disease, but they also expressed greater satisfaction with the care they received.²⁷

Circulating Tumor DNA and Emerging Biomarkers

Beyond CA 19-9, researchers are investigating whether circulating tumor DNA (ctDNA), fragments of tumor genetic material found in blood, can help guide treatment decisions. A study of 25 borderline resectable patients treated with modified FOLFIRINOX analyzed plasma ctDNA at baseline and found that 40% carried alterations in DNA damage repair genes such as BRCA1, BRCA2, and ATM. Those patients had substantially better progression-free survival (median about 27 months versus 14 months). Conversely, patients with KRAS mutations detected in their blood at baseline had dramatically worse overall survival (median around 9 months).²⁸

This kind of blood-based profiling is still in early stages, but it points toward a future where treatment can be further individualized. Patients with DNA repair gene alterations might, for instance, benefit from drugs that exploit that vulnerability, while those with detectable KRAS mutations in their blood might be steered toward more aggressive systemic therapy or clinical trials rather than surgery. For now, these tests are used primarily in research settings, but they are moving toward the clinic.