Abdominal exploration is usually necessary when cytologic diagnosis cannot be made or if resection is to be attempted, (in up to 30% of patients with pancreatic carcinoma). In a patient with a localized mass in the head of the pancreas and without jaundice, laparoscopy may detect tiny peritoneal or liver metastases and thereby avoid resection in 4–13% of patients. Radical pancreaticoduodenal (Whipple) resection is indicated for cancers strictly limited to the head of the pancreas, periampullary area, and duodenum (T1, N0, M0). Five-year survival rates are 20–25% in this group and as high as 40% in those with negative resection margins and without lymph node involvement. Preoperative endoscopic decompression of an obstructed bile duct is often achieved with a plastic stent or short metal stent but does not reduce operative mortality and is associated with complications.
The best surgical results are achieved at centers that specialize in the multidisciplinary treatment of pancreatic cancer. Adjuvant chemotherapy with gemcitabine, 5-fluorouracil, or gemcitabine with capecitabine is superior to no adjuvant therapy. Gemcitabine with capecitabine and a modified FOLFIRINOX (5-fluorouracil, leucovorin, irinotecan, oxaliplatin) regimen have been found to be superior to gemcitabine alone. The role of adjuvant chemoradiation is controversial but often used in the United States. Neoadjuvant chemotherapy with or without radiation is increasingly being used to downstage patients and in those with resectable cancer. Common chemotherapy regimens for this purpose include FOLFIRINOX and gemcitabine with nanoparticle albumin-bound (nab)-paclitaxel. S-1, an oral combination fluoropyrimidine, is used in Japan and other Asian countries. Chemoradiotherapy downstages about 30% of patients with locally advanced disease to allow resection.
When resection is not feasible, endoscopic stenting of the bile duct is performed to relieve jaundice. A plastic stent is generally placed if the patient’s anticipated survival is less than 6 months (or if surgery is planned). A metal stent is preferred when anticipated survival is 6 months or greater. Whether covered metal stents designed to prevent tumor ingrowth offer an advantage over uncovered stents is uncertain because covered stents are associated with higher rates of migration and acute cholecystitis due to occlusion of the cystic duct. Surgical biliary bypass may be considered in patients expected to survive at least 6 months. Surgical duodenal bypass may be considered in patients in whom duodenal obstruction is expected to develop; alternatively, endoscopic placement of a self-expandable duodenal stent may be feasible. Chemoradiation may be used for palliation of unresectable cancer confined to the pancreas.
Chemotherapy has been disappointing in metastatic pancreatic cancer, although improved response rates have been reported with FOLFIRINOX and with the combination of gemcitabine and nab-paclitaxel. In patients who have received prior chemotherapy, 5-fluorouracil and leucovorin in combination with nanoliposomal irinotecan has resulted in improved survival compared with 5-fluorouracil and leucovorin alone. In patients with a BRCA1 or BRCA2 germline mutation, olaparib, a poly(adenosine diphosphate-ribose) polymerase inhibitor, has been reported to improve progression-free survival in metastatic pancreatic cancer. Celiac plexus nerve block (under CT or endoscopic ultrasound guidance) or thoracoscopic splanchnicectomy may improve pain control. Photodynamic therapy is under study.
Surgical resection is the treatment of choice for NETs, when feasible. Lesions that are less than 1 cm in diameter and nonfunctioning without evidence of local invasion or metastasis may be followed expectantly. Metastatic disease may be controlled with long-acting somatostatin analogs, interferon, chemotherapy, peptide-receptor radionuclide therapy, and chemoembolization.
There is a consensus that asymptomatic incidental pancreatic cysts 2 cm or smaller are at low risk for harboring invasive carcinoma. The cysts may be monitored by imaging tests (MRI) in 1 year and then every 2 years for 5 years and perhaps longer if no changes are observed, with EUS and FNA if a cyst enlarges to 3 cm and another high-risk feature (dilated main pancreatic duct, presence of a solid component) develops. The optimal approach is uncertain, however, and other guidelines have been proposed. Surgical resection is indicated for mucinous cystic neoplasms, symptomatic serous cystadenomas, solid pseudopapillary tumors (which have a 15% risk of malignant transformation), and cystic tumors larger than 2 cm in diameter that remain undefined after helical CT, EUS, and diagnostic aspiration. All intraductal papillary mucinous neoplasms of the main pancreatic duct should be resected, but those of branch ducts may be monitored with serial imaging if they (1) are asymptomatic and exhibit benign features; (2) have a diameter less than 3 cm (some authorities recommend a diameter 1.5 cm or smaller, but even lesions 3 cm or larger may be monitored in elderly persons with no other worrisome cyst features); and (3) lack nonenhancing mural nodules, a thick wall, an abrupt change in the caliber of the pancreatic duct with distal pancreatic atrophy, or possibly bile duct dilatation and gallbladder adenomyomatosis. Most such lesions with benign features remain stable on follow-up, but the risk of malignancy persists for at least 10 years. Moreover, the risk of pancreatic ductal carcinoma and of nonpancreatic cancers may also be increased in this group of patients. In the absence of locally advanced disease, survival is higher for malignant cystic neoplasms than for adenocarcinoma. Endoscopic resection or ablation, with temporary placement of a pancreatic duct stent, may be feasible for ampullary adenomas, but patients must be followed for recurrence.