Surgical Aspects of Pancreatic Cancer

Amanda B. Cooper¹, Ching-Wei D. Tzeng², and Matthew H.G. Katz¹

¹The University of Texas MD Anderson Cancer Center, Houston, TX, USA

²University of Kentucky, Lexington, KA, USA

Case study 110.1

A jaundiced 64-year-old woman presents with a computed tomographic (CT) scan showing a 3 cm hypoattenuating pancreatic head mass with a tumor–portal vein interface measuring 270°. The superior mesenteric vein (SMV) appears patent. The tumor also abuts the hepatic artery at the level of the gastroduodenal artery (GDA). Peripancreatic lymph nodes are not enlarged. Endoscopic ultrasound (EUS)–guided biopsy of the mass is consistent with pancreatic adenocarcinoma.

• What would you tell her about her stage and prognosis?

Several clinical staging systems exist for pancreatic adenocarcinoma. The American Joint Commission on Cancer (AJCC) system stages patients’ disease based on physical examination findings and the results of cross-sectional imaging. According to the AJCC system, this patient has a potentially resectable cancer. The final pathologic AJCC stage might be either T3N0 (IIA) or T3N1 (IIB) following resection, however, because CT may underestimate lymph node involvement. Under the current (7th edition) system, stages I–II denote potentially resectable disease (irrespective of lymph node status), whereas stages III–IV indicate unresectable and/or metastatic cancer. A large study using the National Cancer Database validated the current AJCC staging system’s ability to predict 5-year survival and showed that patients with pathologic stage IIA disease have a 15.7% 5-year survival rate after successful resection.

According to National Comprehensive Cancer Network (NCCN) guidelines, this tumor would be anatomically staged as borderline resectable based on portal vein encasement (with reconstruction potential) and GDA abutment up to the hepatic artery. At the MD Anderson Cancer Center, clinical staging assesses the primary tumor anatomy (“A”), the tumor biology (“B”), and the patient’s condition (“C”) (Figure 110.1). Within this system, this woman’s disease would be borderline resectable type A unless she has an elevated CA 19-9 level or extensive comorbidities.

Figure 110.1 The MD Anderson clinical-staging and stage-specific treatment algorithm for pancreatic cancer patients. Patients with “clinically resectable” tumors have resectable tumor anatomy, no elevated suspicion for metastatic disease, and a performance status suitable for a major abdominal operation. Either surgery or neoadjuvant therapy is a rational treatment option for such patients. Three categories of patients have “borderline resectable” tumors: type A patients have borderline tumor anatomy, typically consisting of venous occlusion or arterial abutment; type B patients have borderline biology due to a significantly elevated CA 19-9 level and/or radiologic findings indeterminate for metastatic disease; and type C patients have borderline physiology, such as frailty or potentially reversible comorbidities requiring either prolonged work-up or intervention prior to major surgery. Patients with borderline resectable tumors are routinely treated with neoadjuvant therapy before resection. Patients with locally advanced or metastatic disease are treated palliatively.
*Previously defined MDACC (not AJCC or NCNN) criteria for potentially resectable and borderline resectable anatomy.

By definition, patients with anatomic borderline resectable disease have a high risk of positive surgical margins and early treatment failure with a surgery-first approach. The general consensus is that these patients should receive chemotherapy and/or chemoradiation before surgical resection. In a retrospective study of borderline resectable patients treated with a neoadjuvant approach, 38% completed multimodality therapy and resection with a 40-month median overall survival (OS), comparable to that of patients with “potentially resectable” anatomy who received multimodality therapy and resection. Figure 110.1 illustrates the distinction between borderline resectable and potentially resectable tumors using the MD Anderson system.

Case study 110.2

A 68-year-old man presents with a 2.1 cm hypodense mass in the uncinate process of the pancreas and a bilirubin level of 13.8 mg/dl. A CT scan shows no obvious extrapancreatic disease or vascular involvement.

• What would you recommend to this patient? Would you obtain a tissue biopsy? Refer to a gastrointestinal specialist for endobiliary decompression? Refer to a pancreatic surgeon? Start chemotherapy immediately?

If not already performed, a “pancreatic protocol” CT scan (triple-phase—noncontrast, portal venous, pancreatic parenchymal, and arterial phase—with water as an oral contrast agent with 2–3 mm cuts through the pancreas) allows for optimal assessment of resectability. This patient appears to have resectable tumor anatomy by NCCN criteria and should be referred to a multidisciplinary group that includes a pancreatic surgeon for consideration of resection. Although administration of neoadjuvant therapy to patients with resectable cancers is a valid option at many centers, particularly within a clinical trial, primary surgical resection remains the standard of care. If the surgeon believes the patient is an appropriate candidate for resection, confirmatory biopsy is unnecessary if clinical evidence favors a diagnosis of pancreatic cancer.

Multiple randomized controlled trials (RCTs) have explored whether preoperative biliary stenting benefits patients with obstructive jaundice scheduled for imminent surgery. The best designed of these was a 2010 multicenter Dutch trial, limited to patients with pancreatic head cancer and bilirubin levels of 2.3–14.6 mg/dl. Patients randomized to preoperative biliary drainage underwent drainage 4–6 weeks before surgery. Drainage was performed endoscopically, with percutaneous transhepatic drainage reserved for endoscopic failures. No significant difference was found between the two groups in mortality, length of hospital stay, or postoperative complications, but a significantly higher rate of serious complications occurred in the biliary drainage group (74% vs. 39% in the immediate surgery group) when including procedure-associated complications. The authors concluded that routine preoperative biliary drainage for jaundiced pancreatic cancer patients undergoing surgery-first sequencing increases complication rates. Two meta-analyses reported similar findings.

A 2012 Cochrane Review summarizing the results of RCTs of preoperative biliary drainage included six trials with 520 patients with biliary obstruction from benign or malignant causes. Four trials included percutaneous transhepatic biliary drainage, and two used exclusively endoscopic biliary stenting. Although the Cochrane group felt all included trials had a high risk of bias, they found no significant difference between the drained and undrained groups with respect to mortality or length of stay. Preoperatively drained patients, however, had a higher overall serious morbidity rate. The Cochrane group concluded that with the limitations of the existing studies, the evidence was insufficient to recommend routine preoperative biliary drainage in jaundiced patients.

Although these studies suggest preoperative biliary drainage is unnecessary for jaundiced pancreatic cancer patients who undergo immediate surgery, in practice, patients often present after diagnostic endoscopic retrograde cholangiopancreatography and have had a biliary stent placed as prophylaxis against cholangitis due to instrumentation of an obstructed biliary system.

< div class='tao-gold-member'>

Only gold members can continue reading. Log In or Register to continue

Oncohema Key

Fastest Oncology & Hematology Insight Engine

110: Surgical Aspects of Pancreatic Cancer

Related

Stay updated, free articles. Join our Telegram channel

Full access? Get Clinical Tree