Pancreatic Cancer Treatment & Management

Updated: Oct 06, 2021
  • Author: Tomislav Dragovich, MD, PhD; Chief Editor: N Joseph Espat, MD, MS, FACS  more...
  • Print

Approach Considerations

There is consensus that surgery is the primary mode of treatment for pancreatic cancer. However, an important role exists for chemotherapy and/or radiation therapy in an adjuvant or neoadjuvant setting, and in the treatment of patients with unresectable disease.

Typically, extrapancreatic disease precludes curative resection, and surgical treatment may be palliative at best.

Historically, vascular involvement has been considered a contraindication to resective cure. However, the invasion of the superior mesenteric or portal vein is no longer an absolute contraindication. [61] These veins can be resected partially with as much as 50% narrowing of the lumen. In addition, complete reconstruction is possible, especially using native veins as replacement (ie, internal jugular, greater saphenous, or splenic).

Nonetheless, invasion of the superior mesenteric, celiac, and hepatic arteries still presents a barrier to resection. No evidence indicates that a vascular reconstruction, which permits an attempt at surgical resection, improves or contributes to survival.

After a thorough preoperative workup, the surgical approach can be tailored to the location, size, and locally invasive characteristics of the tumor. Curative resection options include pancreaticoduodenectomy, with or without sparing of the pylorus; total pancreatectomy; and distal pancreatectomy. Each procedure is associated with its own set of perioperative complications and risks, and these points should be taken into consideration by the surgical team and discussed with the patient when considering the goal of resection.

Guidelines from the National Comprehensive Cancer Network (NCCN) recommend that decisions about resectability involve input from a multidisciplinary group of specialists at a high-volume center. The NCCN panel also agreed that selecting patients for surgery should be based on the probability of cure as determined by resection margins. Other factors include comorbidities, overall performance, and age. [4]

Guidelines on pancreatic cancer from the European Society for Medical Oncology (ESMO) advise that complete surgical resection is the only potentially curative treatment available; however, 5-yr overall survival is only 10-20%, and long-term survival in patients with node-positive disease is rare. ESMO recommendations include the following [62] :

  • Optimal symptomatic treatment has a prime role in the management of metastatic disease; patients may require stenting or bypass surgery for obstructive jaundice or gastric outlet obstruction
  • The role of chemotherapy is limited
  • Gemcitabine has been associated with a small survival benefit compared with bolus 5-fluorouracil

In patients with unresectable locally advanced pancreatic cancer, local ablation has been explored as a treatment option. A systematic review concluded that the following strategies appear to be feasible and safe [63] :

  • Radiofrequency ablation (RFA)
  • Irreversible electroporation
  • Stereotactic body radiation therapy (SBRT)
  • High-intensity focused ultrasound (HIFU)
  • Iodine-125
  • Iodine-125–cryosurgery
  • Photodynamic therapy
  • Microwave ablation

Several of these ablative techniques have been shown to provide pain relief and improved survival. For example, medial survival of up to 25.6 months with RFA and 24.0 months with SBRT has been reported. Promising quality-of-life outcomes have been reported for SBRT. [63]

The poor response to therapy has spurred research into precision medicine for pancreatic cancer. The PRIMUS-004 trial will use cell lines and organoids generated from patients with pancreatic cancer to develop new molecular markers to predict response to agents that target DNA damage. [64]



Neoadjuvant Therapy

The use of chemotherapy and/or radiation therapy in the neoadjuvant setting has been a source of controversy. The rationale for using neoadjuvant therapy includes the following assertions:

  1. Pancreatic cancer is a systemic disease and should be treated systemically from the start.
  2. Patients will be able to tolerate the toxic effects of chemotherapy more readily before undergoing major pancreatic resection than after.
  3. The tumor will shrink with neoadjuvant therapy so resection will be less cumbersome, leading to improved overall survival.

Trials of preoperative chemoradiotherapy conducted at M.D. Anderson Cancer Center have shown median survival as high as 25 months. [65, 66] However, no form of neoadjuvant therapy in pancreatic carcinoma should be regarded as a standard form of therapy; this remains an area for clinical trial study.

The NCCN finds there is limited evidence to recommend specific neoadjuvant regimens off-study for patients with resectable or borderline resectable tumor, and practices vary with regard to the use of chemotherapy and radiation. The guidelines prefer consultation at a high-volume center when considering neoadjuvant therapy. If recommended, treatment should be at or coordinated through a high-volume center when possible. Participation in a clinical trial is encouraged. [4]

Possible regimens include the following [4] :

  • FOLFIRINOX/modified FOLFIRINOX, with or without subsequent chemoradiation
  • Gemcitabine + albumin-bound paclitaxel, with or without subsequent chemoradiation
  • Gemcitabine + cisplatin (≥2–6 cycles) followed by chemoradiation (only for known  BRCA1/2 or PALB2 mutations)

A study by Dhir et al in 193 patients with resectable  or borderline resectable pancreatic ductal carcinoma concluded that FOLFIRINOX and gemcitabine plus albumin-bound paclitaxel are viable options for neoadjuvant treatment. After adusting for covariates, however, overall survival was found to be 4.9 months longer with FOLFIRINOX than with gemcitabine-paclitaxel. [67]

In a retrospective study of 49 stage III locally advanced/borderline resectable patients who were initially unresectable, were downstaged through chemotherapy, and subsequently underwent surgical resection, prolonged preoperative chemotherapy was associated with excellent overall survival and high rates of lymph node–negative disease. [68, 69] A study by Loeherer et al found an improvement in overall survival from 9.2 months to 11.4 months with the addition of concurrent external beam radiation therapy to gemcitabine alone. [70]


Pancreaticoduodenectomy (Whipple Procedure)

Patients who will most likely benefit from this procedure have a tumor located in the head of the pancreas or the periampullary region. The Whipple procedure is not strictly the surgical approach for pancreatic head tumors. Pancreatic ductal tumors, cholangiocarcinoma (bile duct cancer), and duodenal masses will all require this resection. The operation traditionally involves removal of the pancreatic head, duodenum, gallbladder, and the antrum of the stomach, with surgical drainage of the distal pancreatic duct and biliary system, usually accomplished through anastomosis to the jejunum. The primary reason for removing so much of the intraabdominal structures is that they all share a common blood supply.

Pancreaticoduodenectomy has been shown to have an overall mortality rate of 6.6%. [71] Many forms of morbidity are associated with the operation. One of these is delayed gastric emptying. This occurs in approximately 25% of patients. This condition may require nasogastric decompression and will lead to a longer hospital stay. [72] Other morbidities include pancreatic anastomotic leak. This can be treated with adequate drainage. Postoperative abscesses are not uncommon.

Although preoperative biliary drainage was introduced to improve the postoperative outcome in patients with obstructive jaundice caused by tumors of the pancreatic head, van der Gaag et al found that routine use of this maneuver increases the rate of complications. In a multicenter, randomized trial, 202 patients with obstructive jaundice and a bilirubin level of 40–250 mmol/L (2.3-4.6 mg/dL) were assigned to undergo either preoperative biliary drainage for 4-6 weeks, followed by surgery, or surgery alone within 1 week after diagnosis. The rate of serious complications was higher in the biliary drainage group than in the early surgery group (74% vs 39%, respectively). No significant difference was noted in mortality or length of hospital stay between the 2 groups. [73]

Similarly, Limongelli et al found that preoperative biliary drainage predisposes patients to a positive intraoperative biliary culture, which in turn is associated with an increased risk of postoperative infectious complications and wound infection. [74]

The standard Whipple operation may be altered in order to include a pylorus-sparing procedure. This modification was previously incorporated to increase nutritional strength in these patients, because the increased gastric emptying associated with antrectomy caused nutritional deficiencies. Although many believe that delayed gastric emptying is worsened by this modification, studies have proven both resections to be equivalent in that regard.

Another source of controversy is the extent of lymphadenectomy that is necessary in a Whipple operation. In an elegant study, Pawlik et al found the ratio of positive nodes to total nodes removed was an important prognostic factor. [75] This was even more significant than margin positivity. [76]

Guidelines on pancreatic cancer from the European Society for Medical Oncology include the following treatment recommendations [62] :

  • Complete surgical resection is the only potentially curative treatment available. However, 5-yr overall survival is only 10-20%; long-term survival in node-positive tumors is rare
  • Optimal symptomatic treatment has a prime role in the management of metastatic disease; these patients may require stenting or bypass surgery for obstructive jaundice or gastric outlet obstruction
  • The role of chemotherapy is limited; gemcitabine has been associated with a small survival benefit compared with bolus 5-fluorouracil

Distal Pancreatectomy

This procedure possesses a lower mortality rate than the standard Whipple procedure does, at 3.5%, but its use in curative resection remains limited. [71] Essentially, a distal pancreatectomy may be an effective procedure for tumors located in the body and tail of the pancreas. Unfortunately, masses located in this area present later than the periampullary tumors and hence have a higher unresectability rate.

The procedure involves isolation of the distal portion of the pancreas containing the tumor, followed by resection of that segment, with oversewing of the distal pancreatic duct. The main complications for distal pancreatectomy involve pancreatic stump leak, hemorrhage, and endocrine insufficiency. [77] Once again, the best treatment for the pancreatic leak is adequate drainage.


Total Pancreatectomy

Although this procedure is the least commonly performed and has the highest associated mortality rate (8.3%), it may still be a valuable instrument in the surgical cure of pancreatic cancer. [71]

The indication for the use of total pancreatectomy is in cases in which the tumor involves the neck of the pancreas. This can either be a situation in which the tumor originates from the neck or is growing into the neck. These patients obviously get insulin-dependent diabetes. In some cases, the diabetes can be hard to control. Despite this, the morbidity of a total pancreatectomy is comparable to that of a Whipple procedure. [78]


Adjuvant Therapy

Several studies (including the GITSG, ESPAC, CONKO) suggested the possibility that chemotherapy, with or without radiation therapy, would significantly improve median survivals following surgical resection of operable disease. [79, 80] These studies were not definitive and not widely accepted as justification for offering either modality for adjuvant therapy.

However, a large, retrospective study supported the use of adjuvant chemoradiotherapy. Yang et al analyzed a registry of 2,877 patients who underwent surgical resection with curative intent for pancreatic adenocarcinoma; approximately half received no adjuvant therapy, and approximately a quarter received postoperative chemoradiotherapy. A significant survival benefit was found for the chemoradiotherapy patients. [81] In 2011, the NCCN panel recommended the measurement of serum CA 19-9 levels after surgery and before adjuvant therapy. [4]

A study by Neuhaus et al in 368 patients with resected pancreatic cancer found that adjuvant gemcitabine prolongs survival when compared with surgery alone. [82] The 3-year survival rates were 36.5% and 19.5% for the gemcitabine and surgery-only arms of the study, respectively. The 5-year survival rates were 21% and 9% for the gemcitabine and surgery-only arms, respectively.

This trial was definitive and transformative. Adjuvant therapy with gemcitabine is now accepted as standard therapy for surgically resected pancreatic cancer. [83]  However, the superior survival demonstrated with modified FOLFIRINOX (mFOLFIRINOX) may well make that regimen the standard of care for nonmetastatic pancreatic ductal adenocarcinoma in patients who have undergone surgical resection and have a good performance status (see Chemotherapy). [84, 4]

A systematic review and meta-analysis by Wan et al found that adjuvant therapy with metformin signficantly reduced the risk of death in Asian patients with pancreatic cancer (hazard ratio [HR]=0.74]), but not in whites. Mortality risk was reduced in patients with stage I-II disease treated with metformin (HR=0.76, 95% CI=0.68-0.86) as well as in those with stage I-IV disease (HR=0.88, 95% CI=0.79-0.99), but not in those with stage III-IV disease. [85]


Chemotherapy for Unresectable and Metastatic Pancreatic Cancer

In patients with metastatic disease, the combination of gemcitabine and erlotinib has led to a significantly higher median survival and 1-year survival than has the use of gemcitabine alone. [86] This has led to US Food and Drug Administration (FDA) approval of erlotinib for use in combination with gemcitabine in advanced, unresectable pancreatic cancer. The recommendation that this combination should now constitute standard therapy for metastatic or unresectable local disease is premature and problematic. The improvements in response rates seen, although significant, were not great and were obtained with no small amount of patient toxicity.

The combination should be used with considerable care, and the use of gemcitabine alone should still be considered as appropriate therapy for patients with metastatic disease. Gemcitabine alone should also be considered as appropriate therapy for patients with unresectable disease; there is no meaningful significant benefit obtained to adding radiotherapy in this situation. Such an addition simply increases toxicity. [87]

The combination of gemcitabine and capecitabine in advanced pancreatic cancer has been investigated by several groups. A randomized, multicenter, phase III clinical trial in 319 patients by the Central European Cooperative Oncology Group found that clinical response or quality of life was no better with the combination than with gemcitabine alone. [88]

This finding contrasts with the results of the phase III United Kingdom National Cancer Research Institute GEMCAP trial, an open-label, randomized study of gemcitabine alone versus gemcitabine combined with capecitabine in 533 patients. Compared with gemcitabine alone, treatment with the gemcitabine-capecitabine combination produced a significantly higher objective response rate (12.4% vs 19.1%, respectively) and progression-free survival and was associated with a trend toward improved overall survival.

In addition, a meta-analysis of two additional studies involving 935 patients showed a significant survival benefit in favor of the gemcitabine-capecitabine combination. Accordingly, these researchers recommended considering gemcitabine-capecitabine as one of the standard first-line options in locally advanced and metastatic pancreatic cancer. [89]

Results of the phase 3 Metastatic Pancreatic Adenocarcinoma Clinical Trial (MPACT) show that the addition of nanoparticle albumin-bound (nab)-paclitaxel to gemcitabine significantly improves overall survival in treatment-naive patients with metastatic pancreatic cancer compared with gemcitabine alone. [6] Overall survival was approximately 2 months longer in patients treated with combination therapy (8.5 vs 6.7 months). One-year and 2-year survival rates were also higher in the combination therapy group (35% vs 22% and 9% vs 4%, respectively).

The results of a European phase III trial (ACCORD/PRODIGE) that compared the nongemcitabine regimen FOLFIRINOX (leucovorin plus 5-fluorouracil [LV5-FU] plus oxaliplatin plus irinotecan) to gemcitabine in patients with metastatic pancreatic cancer were reported in May 2011. [90] The median survival on the FOLFIRINOX arm was 11.1 months, versus 6.8 months on the gemcitabine arm. Of note, the incidence of adverse events and febrile neutropenia was significantly higher on the FOLFIRINOX arm, despite the fact that only patients with ECOG performance status of 0-1 were included in this trial.

It remains to be seen how well this regimen will be integrated into the care of patients with pancreatic cancer and good performance status worldwide.

The phase III PRODIGE 24/CCTG PA.6 trial demonstrated that modified FOLFIRINOX (mFOLFIRINOX) also provides significantly longer survival than gemcitabine in patients with pancreatic ductal adenocarcinomas who have undergone R0 or R1 resection. [91]  This study is potentially practice changing. [84]  The NCCN now recommends FOLFIRINOX/modified FOLFIRINOX as a preferred first-line treatment for patients with metastatic or locally advanced unresectable disease with good performance status. [4]

At a median follow-up of 33.6 months, the median disease-free survival in PRODIGE 24/CCTG PA.6 was 21.6 months in the mFOLFIRINOX group versus 12.8 months in the gemcitabine group; median overall survival was 54.4 vs 35.0 months, respectively. Time until the appearance of metastases was a median of 30.4 months with mFOLFIRINOX versus 17.0 months with gemcitabine). Overall, more patients experienced severe side effects (mainly hematologic) in the mFOLFIRINOX group than in the gemcitabine group (76% vs 53%), but the side effects were manageable. [91]

Paclitaxel protein bound was approved by the FDA in September 2013 for metastatic pancreatic cancer. [7] The treatment regimen includes paclitaxel protein bound 125 mg/m2 plus gemcitabine 1000 mg/m2 IV over 30-40 min on days 1, 8, and 15 of each 28-day cycle. The NCCN also recommends the combination as a preferred first-line treatment for patients with metastatic or locally advanced unresectable disease with good performance status. [4] This regimen may be considered instead of FOLFIRINOX in patients unlikely to tolerate toxicities associated with FOLIRINOX. [6]

For patients with metastatic or locally advanced unresectable disease who have poor performance status, the NCCN recommends gemcitabine monotherapy. [4]

Capecitabine alone or capecitabine plus erlotinib may provide second-line therapy benefit in patient's refractory to gemcitabine. [8] There is no advantage to giving gemcitabine in any dose or time of infusion other than 1000 mg/m² over 30 minutes intravenously.

Combinations of gemcitabine with cisplatin, oxaliplatin, irinotecan, or docetaxel have in phase III trials not shown superior benefit over gemcitabine alone.

A new encapsulated form of irinotecan in a long-circulating nanoliposome (Onivyde) was approved by the FDA in 2015 for patients with advanced pancreatic cancer who have been previously treatment with gemcitabine-based chemotherapy. Irinotecan liposomal is indicated for use in combination with fluorouracil and leucovorin.

Approval of irinotecan liposomal was based on a 3-arm, randomized, open-label study (NAPOLI-1 trial), which was conducted in 417 patients with metastatic pancreatic adenocarcinoma whose cancer had progressed after treatment with gemcitabine alone or in combination with other agents. The regimen used in the trial was a combination of liposomal irinotecan (70 mg/m² IV infused over 90 min [dosage for free-base irinotecan]) administered prior to fluorouracil (2400 mg/m² infused over 46 h) and racemic leucovorin (400 mg/m² infused over 30 min) every 2 weeks.

Patients treated with this combination of liposomal irinotecan plus fluorouracil/leucovorin lived for an average of 6.1 months, compared with 4.2 months for those treated with only fluorouracil/leucovorin or 4.2 months for patients treated with irinotecan liposomal alone. Improvement in progression-free survival was also observed, to a median of 3.1 months with irinotecan liposomal plus fluorouracil/leucovorin compared with 1.5 months for fluorouracil/leucovorin alone. [92]


Palliative Therapy


Patients not undergoing resection for pancreatic cancer should have therapy focused on palliating their major symptoms. Pain relief is crucial in these patients. Narcotic analgesics should be used early and in adequate dosages. Combining narcotic analgesics with tricyclic antidepressants or antiemetics can sometimes potentiate their analgesic effects. In some patients, narcotics are insufficient and other approaches must be considered.

Neurolysis of the celiac ganglia may provide significant, long-term pain relief in patients with refractory abdominal pain. This can be performed transthoracically or transabdominally by invasive radiology or anesthesiology, transgastrically using EUS-guided fine-needle injection, or intraoperatively when assessing the patient's potential for resection.

Radiation therapy for pancreatic cancer can palliate pain but does not affect the patient's survival.

Some patients may experience pain from the obstruction of the pancreatic or biliary ducts, especially if the pain significantly worsens after eating. These patients may benefit from endoscopic decompression with stents.


Obstructive jaundice warrants palliation if the patient has pruritus or right upper quadrant pain or has developed cholangitis. Some patients’ anorexia also seems to improve after relief of biliary obstruction.

Biliary obstruction from pancreatic cancer is usually best palliated by the endoscopic placement of plastic or metal stents. The more expensive and permanent metallic stents appear to have a longer period of patency and are preferable in patients with an estimated lifespan of more than 3 months. Plastic stents usually need to be replaced every 3-4 months.

Patients can also undergo operative biliary decompression, either by choledochojejunostomy or cholecystojejunostomy, at the time of an operation for resectability assessment.

Duodenal obstruction

Approximately 5% of patients develop duodenal obstruction secondary to pancreatic carcinoma. These patients can be palliated operatively with a gastrojejunostomy or an endoscopic procedure.

Endoscopic stenting of duodenal obstruction is usually reserved for patients who are poor operative candidates. Some surgeons empirically palliate patients with a gastrojejunostomy at the time of an unsuccessful attempt at pancreatic resection in an effort to prevent the later need for this operation.



As with most patients with advanced cancer, patients with pancreatic carcinoma are often anorexic. Pharmacologic stimulation of appetite is usually unsuccessful, but it may be tried.

Patients may have some degree of malabsorption secondary to exocrine pancreatic insufficiency caused by the cancer obstructing the pancreatic duct. Patients with malabsorption diarrhea and weight loss may benefit from pancreatic enzyme supplementation. Their diarrhea may also be improved by avoidance of high-fat or high-protein diets.



The management of pancreatic carcinoma is a multidisciplinary process. Typically, the management of pancreatic cancer entails consultations with a gastroenterologist, medical oncologist, general surgeon or surgical oncologist, and, possibly, a radiation oncologist.

A gastroenterologist is usually involved either for evaluation of the cause of the patient's presenting symptoms (eg, abdominal pain, nausea, weight loss, diarrhea) or for a definitive diagnosis of the cause of jaundice by EUS and/or ERCP. Consultation with a gastroenterologist is also needed if an endoscopically placed stent is needed for palliation of obstructive jaundice.

Consultation with a medical oncologist is often needed to select and administer neoadjuvant, adjuvant, or primary chemotherapy for the disease. Consultation with a medical oncologist is also useful for the management of other common cancer symptoms, such as pain and nausea.

Consultation with a surgeon is needed when the patient's imaging studies suggest that operative resection may be feasible. The surgeon may perform diagnostic laparoscopy or even laparoscopic ultrasonography prior to an attempt at definitive resection.

If curative resection is not possible, consultation with a surgeon may still be useful to consider operative palliation of biliary and/or duodenal obstruction. Consult with a surgeon or surgical oncologist who is very experienced in performing pancreaticoduodenectomies.

Consultation with a radiologist may be needed for special issues, such as obstructive jaundice that is difficult to manage where percutaneous transhepatic cholangiography may be needed.

Consultation with a radiation oncologist is usually considered at the discretion of a medical oncologist when combined chemoradiation may be beneficial. This approach is only indicated when this combination therapy is the subject of a clinical trial.