Palliative treatment of pancreatic and periampullary tumors

Introduction

Pancreatic and periampullary tumors are a common cause of cancer death, with rising incidence in the Western world. In 2018 the incidence of pancreatic cancer in the United States (US) was 13.7 per 100,000 persons, compared with 11.6 per 100,000 persons in 2000. Pancreatic cancer is the fourth most common cause of cancer death in men and women in the US (see Chapters 61 and 62 ). Periampullary tumors, in particular pancreatic adenocarcinoma, confer an extremely poor prognosis. Unfortunately, the vast majority of patients with pancreatic or periampullary cancer (approximately 75%–85%) are not candidates for curative-intent surgery at diagnosis because of extensive local disease or metastases. ^, According to the National Cancer Institute 2009 to 2018 records, only about 12% of patients with pancreas cancer present with localized disease confined to the pancreas, 30% present with regional locally advanced disease (including resectable and unresectable disease), and half have distant metastases at diagnosis. Furthermore, even patients undergoing curative-intent resections have a relatively poor prognosis secondary to rapid disease progression or recurrence. The corresponding estimated 5-year survival rates by stage of diagnosis are 37%, 12%, and 3% for localized, regional, and metastatic disease. These data clearly demonstrate that noncurative intent care represents a substantial portion of care for patients with pancreatic and periampullary tumors because the vast majority of patients will most likely require palliation at some point. Therefore surgeons should be knowledgeable in various aspects of palliative options, including both surgical and nonsurgical strategies that have evolved over the years.

Palliation is performed with the intent to improve the patient’s quality of life through prevention and relief of suffering. ^, The goal of treatment transitions from an attempt to cure or prolong survival to that of offering durable solutions for debilitating symptom relief. Considering operative intervention for palliation is complex; possible benefits of intervention must be weighed against the inherent risks of the intervention, while accounting for longevity associated with advanced disease. In pancreatic and periampullary cancer, surgical palliation should be considered specifically for three commonly debilitating entities: biliary obstruction, gastric outlet obstruction, and tumor-related pain.

In the last decade, several advances in care have resulted in moderate improvements in survival for patients with pancreatic cancer. Most importantly, treatment with combinations of modern systemic agents (such as gemcitabine plus a taxane agent or oxaliplatin, irinotecan, fluorouracil, and leucovorin [FOLFIRINOX]) have demonstrated superior survival benefit compared with conventional gemcitabine mono-chemotherapy, ^, with reported median survival of up to 54 months for patients in resected pancreatic cancer (see Chapter 62 ). Other advances in care include use of neoadjuvant therapy for borderline/locally advanced disease, introduction of targeted therapy (biological agents and immunotherapy), improved multidisciplinary management of chemotherapy-associated toxicities, and optimization of surgical technique, including use of minimal invasive surgery and enhanced postoperative care and recovery pathways. As patients live longer with incurable disease, the need of palliative treatment becomes increasingly pertinent. This is especially true in locally advanced disease given the potentially locally obstructive and destructive nature of periampullary or pancreatic head tumors. In general, surgical palliation is mostly used to relieve the effects of advanced locoregional disease. Patients with distant metastases may be candidates for palliative surgical intervention. Metastatic spread, however, is associated with more aggressive disease, fewer local signs and symptoms, and generally a shorter overall life expectancy. In a National Cancer Database (NCDB) analysis of patients with metastatic pancreatic adenocarcinoma diagnosed between 2003 to 2011, only 19% of the patients underwent operative palliation, with the majority receiving nonoperative palliation (chemotherapy, radiation, or pain management alone; see Chapter 66 ).

This chapter focuses on palliative surgical treatment of biliary obstruction, gastric outlet obstruction, and tumor-related pain. We also discuss local ablative therapies, palliative pancreatectomy, and strategies for end-of-life care for terminally ill patients. The role of chemotherapy and radiotherapy is discussed in Chapter 66 . It is critical that readers recognize that palliative treatment in pancreatic and periampullary tumors should be tailored to the individual patient with multidisciplinary support. Ideally, the patient’s clinical presentation, functional status, comorbidities, tumor stage, tumor biology, and patient and tumor genetics should all be accounted for. Finally, any care should be consistent with the patient’s goals of care, which should be thoroughly and expertly evaluated with shared decision making to determine the role of any interventions.

Biliary obstruction

Jaundice secondary to extrinsic mechanical obstruction of the biliary tree is the presenting symptom in 50% to 70% of patients with pancreatic head adenocarcinoma and develops in as many as 80% during the course of the disease ^, (see Chapters 61 and 62 ). Obstructive jaundice can manifest clinically with jaundiced skin and sclera, pruritis, nausea, vomiting, cachexia, dark colored urine, and clay-colored stool. Bile stasis secondary to obstruction can eventually lead to recurrent ascending cholangitis, fat malabsorption, progressive malnutrition, hepatic dysfunction and failure, and eventually death. In patients with reasonable life expectancy, relieving the obstruction is of paramount importance because successful biliary drainage has been shown to improve quality of life (see Chapter 29, Chapter 30, Chapter 31 ). Decompression of the biliary obstruction relieves the patient from the often-debilitating symptoms previously described and prevents rapid clinical deterioration that can accompany cholangitis in previously instrumented settings. Treating biliary obstruction is associated with improved liver function and nutritional status, and, on a cellular level, has been shown to reverse proinflammatory immune responses in both animal and human models. ^, Finally, in patients who are candidates for chemotherapy, unobstructed bile flow and resolution of jaundice are a prerequisite for successful and safe administration of systemic chemotherapy to avoid toxicity secondary to inadequate biliary excretion of metabolites (see Chapter 66 ).

Importantly, patients report improved quality of life after biliary decompression. In a prospective study of 50 patients with malignant biliary obstruction and unresectable disease, 1 month after successful endoscopic biliary drainage, patients with reduction in plasma bilirubin levels reported substantial improvement in social function (relative risk [RR]: 0.11; 95% confidence interval [CI]: 0.03–0.19) and mental health (RR: 0.04; 95% CI: 0.01–0.08) compared with pre-decompression. Another small single institution study found that patients with locally advanced or metastatic disease ( n = 64) who underwent biliary or enteral stent placement reported an increase in quality of life at 6 months compared with baseline.

Interventions for biliary drainage in obstructive jaundice include operative biliary bypass, endoscopic stent placement, percutaneous external drainage, and endoscopic ultrasound (EUS)-guided biliary drainage (the most novel technique). These various tools in the armamentarium of the multidisciplinary treatment team have ensured the cases of biliary obstruction that are completely unreachable for drainage are minimized. In the last several years, the pendulum has shifted away from operative palliation of the biliary tree, largely because of enhanced durability of decompressive endoscopic metal stents (see Chapters 30 and 31 ). In a large case series from Johns Hopkins Hospital of patients with unresectable pancreatic adenocarcinoma, between the years 1996 to 2010 there was a consistent temporal decrease in the use of palliative surgical bypass ( Fig. 67.1 ); in 1996 to 2001, 10% of the patients received hepaticojejunostomy bypass, whereas since 2002 under 4% of patients underwent this procedure.

Biliary stenting

The current standard of care is biliary drainage via endoscopic self-expandable metallic stents (see Chapter 30 ). Endobiliary stent placement has been shown to be associated with lower complication rates and faster recovery compared with operative palliative biliary drainage and results in durable relief with similar success rates (80%–100%) and an even higher postintervention quality of life. ^, The effectiveness of endoscopic biliary stenting has improved thanks to the usage of self-expanding metal stents introduced in the 1990s that are far superior to previously relied-on Teflon/polyethylene plastic stents. Interestingly, covered metal stents have been shown to possess longer patency compared with uncovered stents in three randomized controlled trials (RCTs). Importantly, while easier than major surgical interventions, endoscopic retrograde cholangiopancreatography (ERCP) and biliary stent placement carry a risk for complications, such as perforation, cholecystitis, cholangitis, hemorrhage, and acute pancreatitis. Furthermore, endobiliary stents can fail because of obstruction or migration, necessitating complex reinterventions and readmissions, which, in turn, increase the risk of complications. Five RCTs have compared surgical biliary drainage (see Chapter 32 ) to endoscopic drainage ( Table 67.1 ). The majority of these studies were conducted before 1994, with the most recent one published in 2006. The studies enrolled small patient cohorts and included a variety of stent types (in size and material). In a 2007 meta-analysis, the pooled plastic stent treatment arm showed fewer complications (RR 0.6; 95% CI 0.45–0.81) but a higher risk of recurrent biliary obstruction compared with surgery (RR 18.59; 95% CI 5.33–64.86). An updated meta-analysis from 2014 reported similar technical success rates (RR 0.99; 95% CI 0.93–1.05), complications, and mortality (RR 1.54; 95% CI 0.87–2.71) between surgical biliary bypass and endoscopic biliary stent placement but with consistently lower rates of recurrent jaundice after surgical bypass (RR 0.14; 95% CI 0.03–0.63; Fig. 67.2 ). Furthermore, patients with prolonged survival in the stent group had more hospital days (a 2-fold increase for stent patients in the total number of hospital days from the index procedure until death). A 2015 meta-analysis comparing the same 5 studies concluded that stenting has lower procedure-associated complications (risk difference of –0.19 to –0.24 favoring stenting). These pooled estimates should be interpreted with caution. In the two most recent meta-analyses, 4 of the 5 studies included were conducted before 1994 and used plastic stents. In addition, the meta-analyses did not include a more recent small RCT from 2003, with similar procedure-related morbidity (7% in the stent group [ n = 14] and 8% in the surgical bypass group [ n = 13]). Although contemporary high-quality evidence is lacking, these data allow for some general conclusions to be drawn (i.e., that endoscopic stent placement is associated with slightly lower morbidity). However, surgical bypass offers a more durable long-term solution.

TABLE 67.1

Prospective Randomized Controlled Trials Comparing Surgical Biliary Drainage With Percutaneous Drainage or Endoscopic Drainage

	BORNMANN ET AL., 1986		SHEPERD ET AL., 1988		ANDERSEN ET AL., 1989		SMITH ET AL., 1994		NIEVEEN VAN DIJKUM ET AL., 2003		ARTIFON ET AL., 2006
	BYPASS	STENT a	BYPASS	STENT	BYPASS	STENT	BYPASS	STENT	BYPASS	STENT	BYPASS	STENT
No. patients	25	25	25	23	25	25	101	100	13	14	15	15
Success (%)	76	84	92	80	88	96	93	95	100	100	100	100
Morbidity (%)	32	28	56	30	20	36	58	29	8	7	47	33
30-day mortality (%)	20	8	20	9	24	20	14	3	0	0	0	0
Hospital stay (days)	24	18	13	5	27	26	26	19	12	3	19	5
Readmission(s) (%)	-	-	12	43	-	-	-	-	11	14	40	26
Recurrent jaundice or cholangitis (%)	16	38	0	30	0	0	2	36	-	-	0	27
GOO (%)	0	14	4	9	0	0	11	19	-	-	7	0
Survival (weeks)	15	19	18	22	14	12	26	21	27	17	29	23

GOO , Gastric outlet obstruction; No., number.

a Percutaneous transhepatic biliary drainage

Cholecystenteric bypass

The gallbladder can serve as a conduit for biliary-digestive bypass, with minimal alteration to the normal anatomy. In the setting of biliary obstruction, the gallbladder is often distended (Courvoisier’s sign), further strengthening the technical appeal of this approach as the anastomosis becomes simpler. The bypass can be done to the stomach, duodenum, or jejunum. With cholecystojejunostomy, the procedure is more extensive as a jejunal loop needs to be approximated to the gallbladder in an anticolic or retrocolic configuration, and an additional Roux limb or Braun enteroenterostomy created to avoid reflux. The anastomosis can be safely performed handsewn or using a staple device ( Fig. 67.3 ). Cholecystogastrostomy and cholecystoduodenostomy are procedures that are largely of historic interest. Tumors involving the proximal gastroduodenal tract at the time of surgery preclude these procedures. If pursued, they require mobilization of the gallbladder fundus from the cystic plate. With gallbladder bypass, the surgeon must rule out tumor involvement in the cystic duct/common hepatic duct juncture to ensure effective drainage of the biliary tree. An intraoperative cholecystogram is recommended to ensure free-flowing bile into the gallbladder (see Chapter 24 ).