Pancreatic cancer is the fifth leading cause of cancer death in the United States.In the year 2005, an estimated 32,180 new cases will be diagnosed, and 31,800deaths will be ascribed to this cancer.
Pancreatic cancer is the fifth leading cause of cancer death in the United States.In the year 2005, an estimated 32,180 new cases will be diagnosed, and 31,800deaths will be ascribed to this cancer.
Incidence and epidemiology
The incidence of pancreatic cancer is slightly higher in males than infemales. These gender differences are most prominent among youngerindividuals.
The peak incidence of pancreatic carcinoma occurs in the seventh decadeof life. Two-thirds of new cases occur in people > 65 years old.
The incidence is higher in the black population, with an excess risk of 40%-50% over whites. Perhaps more importantly, black males probably have thehighest risk of pancreatic cancer worldwide.
Cancer of the pancreas is a highly lethal disease historically, with fewreports of 5-year survivors. However, more recent series have shown a decreasein both operative mortality and overall morbidity. There has also been asignificant increase in 5-year survival after curative resection (21%-25%). Factorsthat appear to be important in predicting long-term survival after resectioninclude clear surgical margins, negative lymph nodes, and reduced perioperativemortality.Adenocarcinoma of the pancreas, the most common histologic type, has a mediansurvival of 9-12 months and an overall 5-year survival rate of 3% for allstages. At the time of diagnosis, over 50% of patients with pancreatic adenocarcinomahave clinically apparent metastatic disease. Among patients whosedisease is considered to be resectable, 50% will die of recurrent tumor within2 years.
Etiology and risk factors
The specific risk factors for pancreatic cancer are not as striking as those forother GI malignancies, such as esophageal and gastric carcinomas. There does,however, appear to be a significant relationship between pancreatic cancer andenvironmental carcinogens.
Cigarette smoke is one of the carcinogens directly linkedto the causation of pancreatic malignancies. Heavy cigarette smokers have atleast a twofold greater risk of developing pancreatic carcinoma than nonsmokers.In Japan, cigarette smoking carries an even greater risk, which can be asmuch as 10-fold in men smoking one to two packs of cigarettes daily.
found particularly in processed meat products, reliablyinduce pancreatic cancer in a variety of laboratory animals. No study hasdirectly linked dietary carcinogens to pancreatic cancers in humans.
The contribution of caffeine consumption to the development of pancreaticcarcinoma is controversial. A case-controlled study showed a correlationbetween caffeine consumption and pancreatic cancer. However, other studieshave been unable to confirm this relationship.
A clear-cut relationship between alcohol use and pancreatic carcinomahas not been shown.
does not seem to be a risk factor for pancreatic cancer. However,10% of all patients with pancreatic carcinoma present with new-onset diabetes.
Cancer of the pancreas is a genetic disease. To date, morethan 80% of resected pancreatic cancers have been found to harbor activatingpoint mutations in K-
. In addition, the tumor-suppressor genes
are all frequently inactivated in this cancer.Familial pancreatic carcinoma has been associated with the following geneticsyndromes: hereditary pancreatitis, ataxia-telangiectasia, hereditarynonpolyposis colorectal cancer (HNPCC), familial atypical mole melanoma(FAMM) syndrome, Peutz-Jeghers syndrome, and familial breast cancer. Familieswith
germline mutations may be at higher risk of developing pancreaticcancer than those without these mutations.
Signs and symptoms
The initial clinical features of pancreatic carcinoma include anorexia, weightloss, abdominal discomfort or pain, and new-onset diabetes mellitus or thrombophlebitis.The vague nature of these complaints may delay diagnosis forseveral months.
Specific symptoms usually relate to localized invasion of peripancreaticstructures. The most common symptom is back pain, which stems from tumorinvasion of the splanchnic plexus and retroperitoneum or pancreatitis. Thispain is described as severe, gnawing, and radiating to the middle of the back.Pain can also be epigastric or in the right upper quadrant if bile duct obstructionis present.
In a majority of cases, patients with pancreatic cancer present withepigastric or back pain and/or jaundice. Painless or sometimes painless jaundiceoccurs with early lesions near the intrapancreatic bile duct.
Tumor invasion of the duodenum or gastric outlet may give riseto nausea or vomiting as a presenting symptom. This symptom is rare early inthe course of the disease. Changes in bowel habits related to pancreatic insufficiencymay also be present, along with associated steatorrhea.
Recent onset of glucose intolerance in an elderly patientassociated with GI symptoms should alert physicians to the possibility of pancreaticcarcinoma.
A palpable gallbladder
occurring in the absence of cholecystitis or cholangitissuggests malignant obstruction of the common bile duct until provenotherwise. This so-called Courvoisier's sign is present in about 25% of all pancreaticcancer patients.
Other physical findings
include Trousseau's syndrome (migratory superficialphlebitis), ascites, Virchow's node (left supraclavicular lymph node), or a periumbilicalmass (Sister Mary Joseph's node).
Screening and diagnosis
Early diagnosis of pancreatic carcinoma is difficult but essential if surgical resectionand cure are to be improved. Defining early lesions at a resectable stageremains a diagnostic challenge. To date, leading medical organizations havenot recommended routine screening of asymptomatic individuals for pancreaticcancer.
The use of serologic tumor markers for pancreatic carcinoma,such as CA19-9, was originally thought to be appropriate as a screening tool.However, since the prevalence of pancreatic carcinoma in the general populationis extremely low (0.01%), many false-positive screening results are generated.Also, the sensitivity of CA19-9 is not high (20%) in stage I cancers. Nevertheless,CA19-9 may be a useful marker for diagnosing patients at high riskwith the appropriate symptoms, such as smokers, recent-onset diabetics, thosewith familial pancreatic cancer, or those with unexplained weight loss or diarrhea.This marker also is useful in following disease and in assessing the adequacyof resection or therapy.No currently available serum marker is sufficiently accurate to be consideredreliable for screening asymptomatic patients.
is useful for staging patients with pancreatic carcinoma and forformulating treatment plans. Approximately 10%-15% of patients thought tohave resectable disease are found to have distant metastases at laparoscopy.The false-negative rate of laparoscopy is < 10%. The strongest indications forlaparoscopy are locally advanced disease and tumors of the body and tail ofthe pancreas.
also is being explored for the diagnosis of pancreatic carcinoma.Cytology is positive in 5%-10% of patients who are thought to havelocalized disease. There are anecdotal cases of long-term survival after resectionwhere positive cytology of peritoneal washings was noted. However, theclinical/prognostic value of this test is not yet known.
Imaging for pancreatic carcinoma is best performed with conventional ultrasonographyand CT.
The limit of sonographic resolution for early pancreatic carcinomais a diameter on the order of 1.0-1.5 cm. A mass located in the pancreatichead will produce dilatation of the common bile duct and pancreatic duct. Theactual sensitivity of ultrasonography in the diagnosis of pancreatic carcinoma is~70%.
provides better definition of the tumor and surrounding structures thandoes ultrasonography and is operator-independent. CT correctly predictsunresectable tumors in 85% of patients and resectable tumors in 70% of patients.Findings of tumor unresectability on CT scanning include distant lymphadenopathy,encasement or occlusion of the superior mesenteric artery (SMA) orceliac artery, occlusion of the portal vein or superior mesenteric vein (SMV),and distant metastases.
More recently, spiral CT has emerged as a preferred technique forincreasing the accuracy of detecting pancreatic carcinoma in general and vesselencasement in particular. This technique permits rapid data acquisition and computer-generated three-dimensional (3D) images of the mesenteric arterial andvenous tributaries in any plane. Spiral CT is quicker and less expensive and usesless contrast medium than angiography.
The use of positron emission tomography with
fluorodeoxyglucose (FDGPET)in the evaluation of patients with pancreatic cancer is expanding. A recentstudy of 126 patients with focal, malignant, or benign pancreatic lesionsshowed high sensitivity of FDG-PET for detection of small pancreatic neoplasms.Lack of focal glucose uptake excludes pancreatic neoplasms (sensitivity85.4%, specificity 60.9%).
At present, MRI is not as accurate as CT in diagnosing and stagingpancreatic carcinoma. MRI may be as useful as CT in staging and can providemagnetic resonance angiography and magnetic resonance cholangiopancreatography(MRC) images if needed. As yet, MRC is not a standard testfor the diagnosis of pancreatic carcinoma, but it may become helpful inthe future.
Endoscopic ultrasonography (EUS)
is a newer modality for the diagnosisof pancreatic carcinoma, with an overall diagnostic accuracy rate of approximately85%-90%. For the assessment of regional lymph node metastases, theaccuracy of EUS is 50%-70%. This technique is also important in the evaluationof portal vein/SMV involvement by tumor. In addition, EUS-guidedfine-needle cytology of periampullary tumors may yield new informationwith respect to the diagnosis of pancreatic cancer and may be less risky inspreading cells by needle tracking than percutaneous biopsies.In a comparison of EUS and spiral CT, both techniques showed comparableefficacy in detecting tumor involvement of lymph nodes and the SMVs andportal veins. However, EUS is less helpful in the evaluation of the SMA.
Endoscopic retrograde cholangiopancreatography (ERCP)
may somedaybe supplanted as a diagnostic tool by EUS, although, at present, ERCP is usedin many clinics. Also, if a patient presents with jaundice and the CT scan revealsdilatation of the common bile duct without an obvious mass, ERCP maybe complementary to spiral CT. ERCP findings of pancreatic cancer includean abrupt or tapered cutoff of either or both the main pancreatic and commonbile ducts.
arising from the exocrine gland ductal system is the mostcommon type of pancreatic cancer, accounting for 95% of all cases. Two-thirdsof these cancers originate in the pancreatic head, and the remainder arise in thebody or tail. Most ductal carcinomas are mucin-producing tumors and usuallyare associated with a dense desmoplastic reaction.Although most pancreatic adenocarcinomas arise from the ductal epithelium,pancreatic acinar carcinomas and cancers arising from mucinous cystic neoplasmsare also found.
which is usually microscopic, is not unusual.
Perineural invasion occurs in the majority of patients withpancreatic carcinoma. In addition, pancreatitis distal to and surrounding thetumor is usually present. Most patients present with lymph node metastases inthe region of the pancreaticoduodenal drainage basins. Subpyloric and inferiorpancreatic head, SMA, and para-aortic lymph node groups also may be involved.
Staging and prognosis
Pancreatic adenocarcinoma is staged according to local spread of disease, nodalstatus, and distant metastatic involvement using the American Joint Committeeon Cancer (AJCC) TNM system (Table 1). The T staging of the primarytumor includes an analysis of direct extension of disease to the duodenum, bileduct, or peripancreatic tissues. A T4 advanced cancer may extend directly tothe SMA or celiac axis, meaning that the cancer is unresectable.
Independent prognostic factors
Lymph node metastases and tumor size anddifferentiation have independent prognostic value in patients with pancreaticcarcinoma. Significantly improved survival is seen in patients with smaller lesions,lymph node-negative tumors, and tumors in which the surgical marginsare not involved.
Lymph node and margin status
Prior to the age of adjuvant therapy, lymphnode status was the most dominant prognostic factor (Figure 1). It is nowrivaled by surgical margin status in series where surgical margins have beenmeticulously examined.
SURGICAL TREATMENT OF RESECTABLE DISEASE
The rate of resection for curative intent ranges from 10% to > 75%, with thehigher percentage resulting from both a more aggressive approach and betterpreoperative staging for resectability. Also, there is growing evidence that patientswith potentially resectable pancreatic cancer have a shorter hospital stay,reduced surgical mortality, and an overall better outcome if the surgery is performedat "high-volume" medical centers staffed by experienced surgeons(approximately 16 operable cases per year).Extended resections may include portal or superior mesenteric vessels, colon,adrenal, or stomach. If resection of adjacent organs or tissues results in theconversion of a positive to a negative resection margin, it is of great potentialbenefit to the patient.
Determination of resectability
The initial approach to surgery for pancreatic carcinoma includes a determinationof resectability. This determination should be first made preoperativelywith high-quality CT or MRI and perhaps EUS. Operative determination ofresectability includes careful examination of the liver, porta hepatis, and portaland superior mesenteric vessels. The head of the pancreas and uncinate processare mobilized by an extensive Kocher maneuver to evaluate the head ofthe pancreas. The SMA is palpated, and its relationship to the tumor is assessed.The hepatic artery and celiac trunk are examined to make certain there isno vascular encasement.Criteria for unresectability include distant metastases and involvement of theSMA and celiac axis.An analysis of 200 patients who underwent resection of pancreatic adenocarcinomain the era prior to adjuvant therapy found that the most important factorsinfluencing long-term survival were the diameter of the primary tumor,status of the resected lymph nodes, and status of the resected margins. Patientswith tumors < 3 cm in diameter had significantly longer median survival and5-year survival rates (21 months and 28%, respectively) than those with tumors≥ 3 cm (11.5 months and 15%). Patients with no lymph node involvement hada 5-year survival rate of 36%, as compared with < 5% for those with positivenodes. Patients who underwent resections with negative margins had a 5-yearsurvival rate of 26%, vs 8% for those with positive margins. The type of resection(pylorus-preserving vs standard Whipple procedure) did not influencesurvival.
Extent of resection
Whipple vs pylorus-preserving procedure
If the tumor is deemed to beresectable, a standard pancreaticoduodenectomy (Whipple procedure) orpylorus-preserving Whipple procedure (PPW) is performed. The PPW theoreticallyeliminates the nutritional problems caused by a reduced gastric reservoirand gastric dumping, but this finding has not been shown to alter longtermnutritional status. If there is any doubt about cancer proximity or bloodsupply to the pylorus, an antrectomy should be performed. If the tumor approachesthe pylorus or involves the subpyloric nodes, classic antrectomy ispreferred.
Most patients with resectable periampullary tumorscan successfully undergo pancreaticoduodenectomy without an intraoperativebiopsy. A time-consuming frozen section interpretation may not be informative,and histologic confirmation may be impossible with small lesions associated withperitumoral pancreatitis. Most large series of pancreaticoduodenectomy for carcinomainclude resections of benign pathology based on clinical judgment. Anegative fine-needle cytology should not deter an experienced surgeon fromproceeding with resection. However, medicolegal considerations may prompta biopsy.
The most common reconstruction technique aftera Whipple resection requires a single retrocolic jejunal loop to complete thepancreaticojejunostomy, which is followed by a cholangiojejunostomy andgastrojejunostomy. A duct-mucosal anastomosis is preferred to the pancreaticojejunostomy.Pancreaticogastrostomy is also an effective and safe means of creatingthe anastomosis.
Operative mortality of pancreaticoduodenectomyis currently < 6% in major surgical centers. The leading causes of postoperativemortality include postoperative sepsis, hemorrhage, and cardiovascular events.Most of the septic complications arise from pancreaticojejunostomy leaks.In many series, early delayed gastric emptying is the leading cause of morbidityfor pylorus-preserving procedures. The number-two cause of morbidity,seen in 5%-15% of all patients, is a leak or fistula from the pancreatic anastomosis.Today, most fistulas close spontaneously with the addition of somatostatinanalog treatment and adequate drainage. Pancreatic fistulas heal with conservativemeasures in more than 95% of patients.
Surgical palliation is also considered in patients undergoing exploration withcurative intent. Jaundice, gastric obstruction, and pain may be alleviated bysurgical palliation.
Biliary tract obstruction
Either a choledochojejunostomy or cholecystojejunostomycan be used to bypass the biliary obstruction. Recurrent jaundiceand cholangitis are less likely to develop when the common duct is used fordecompression.
Although duodenal obstruction is rare as a presentingsymptom, duodenal involvement may occur eventually in 25% of patients. Someauthors believe that prophylactic bypasses are safe and should be performed inall patients. One phase III trial supports prophylactic bypass, but anotherdoes not.
Severe back pain may be an incapacitating symptom. Pain reliefmay be achieved by chemoablation of the celiac plexus or by alcohol injection,which may be performed intraoperatively or percutaneously. An intraoperativeinjection of 25 mL of ethanol (95%) on both sides of the celiac axis will ablatetumor pain. (For further discussion of these techniques, see chapter 37 on painmanagement.)
NEOADJUVANT AND ADJUVANT THERAPY
Even with apparently adequate surgical resection,pancreatic cancer has a high risk oflocoregional recurrence. Moreover, most lesionsare unresectable, even when there is noapparent distant metastatic disease. Thus,there is a theoretical rationale for the adjunctiveuse of radiation therapy, either before orafter surgery, in almost all patients. Preoperative(neoadjuvant) radiation therapy may helprender locally advanced lesions resectablewith negative margins (RO resection). Postoperative(adjuvant) radiation therapy mayhelp eliminate suspected residual microscopicdisease in the tumor bed and/or regionallymphatics. Alternative radiation techniques,including intensity-modulated radiotherapy(IMRT) and 3D conformal radiation therapy,are being explored.With an effective chemotherapeutic agent, there is greater potential for adequatelocoregional cytotoxicity?as well as control of subclinical distant disease?than could be obtained with limited doses of adjuvant radiation therapyalone.
Preoperative chemoradiation therapy
Several single-institution studies haveevaluated the role of preoperative irradiation in conjunction with fluorouracil(5-FU)- and gemcitabine (Gemzar)-based chemotherapy. In these studies, 60%-80% of the lesions were completely resected 1.0-1.5 months after the completionof chemoradiotherapy. Median survival has ranged from 16 to 36 months, butno phase III trials have been conducted to evaluate preoperative therapy vspostoperative sequencing.Preoperative radiation therapy, to 4,500-5,000 cGy, in conjunction withchemotherapy should be considered for patients with pancreatic adenocarcinomawho are medically fit but who have marginally resectable disease. Thereare research initiatives to further address the role of neoadjuvant chemotherapy.For example, a new Eastern Cooperative Oncology Group (ECOG)study will evaluate gemcitabine plus radiotherapy vs gemcitabine, 5-FU, andcisplatin followed by radiotherapy and 5-FU for patients with locally advanceddisease. Other phase II studies involved high-dose gemcitabine and high-dosegemcitabine and cisplatin with short-term radiation therapy to locally advancedcancer.
Postoperative chemoradiation therapy
A small Gastrointestinal Tumor StudyGroup (GITSG) trial demonstrated a significant prolongation of survival (mediansurvival increase, from 11 to 20 months) among patients with pancreaticadenocarcinoma who received irradiation plus bolus 5-FU chemotherapy aftercurative resection, as compared with those given no adjuvant treatment. Animprovement in the long-term cure rate was also observed among those givenchemoradiation therapy.The European Organization for Research and Treatment of Cancer (EORTC)completed a trial of 218 patients similar to that of the GITSG trial but withoutmaintenance chemotherapy. Reported data suggest no significant differencebetween split-course radiation therapy with bolus 5-FU and observation onlyafter curative resection (two-tailed
value = .099); however, there was a trendtoward benefit in median survival favoring those who received treatment. TheEuropean trial is difficult to interpret because 20% of patients randomized toreceive postoperative treatment were not treated, and the study was inadequatelypowered for survival.The GITSG study utilized 4,000 cGy of radiation delivered in a split-coursefashion-with a planned 2-week break midway through the treatment. However,single-institution studies indicate that 4,500-5,000 cGy can be safely deliveredin 5.0-5.5 weeks without a treatment break.Careful attention to field size is important. The GITSG trial allowed portals aslarge as 20 x 20 cm. However, ports that are approximately 12 x 12 cm areusually sufficient to cover the tumor bed with a 2- to 3-cm margin. The use ofmultiple beams and high-energy photons is also important.A total of 541 patients were enrolled in a trial conducted by the EuropeanStudy Group for Pancreatic Cancer (ESPAC). This study evaluated the benefitsof adjuvant therapy. The design was complex, attempting to assess severaloptions. It included no further therapy after surgery, chemoradiation therapy(bolus 5-FU with split-course radiotherapy), chemotherapy (5-FU with leucovorin),and chemoradiation therapy followed by chemotherapy.Interpretation of the results is confounded by the fact that some institutionsopted for a full 2 x 2 randomization (all four options), whereas others allowedonly two options (no further therapy vs chemotherapy or no further therapy vschemoradiation therapy). Patients receiving these two options could also havetherapy other than that prescribed in the randomization. Furthermore, no datawere collected regarding time to recurrence or whether treatment was givenafter recurrence. Curiously, the median survival of those in the control groupwas more than 17 months, much longer than those of the control groups fromthe GITSG and EORTC trials.Only the 5-FU with leucovorin arm would be considered a state-of-the-art approach,and it was demonstrated to improve survival significantly (
= .0005).This finding would suggest a strong benefit to postoperative chemotherapy. Ifradiation therapy is included, it would probably best be given after 1-2 monthsof full-dose chemotherapy. Most practitioners would recommend continuouscourseradiation therapy rather than split-course treatment.The ESPAC is now conducting a postoperative trial comparing various chemotherapyregimens with a control group. In the United States, the findings ofseveral phase II trials of postoperative regimens as well as a phase III trial willbe available soon. In addition, the GI Intergroup is considering a randomizedphase II trial to explore new combinations incorporating the monoclonal antibodiesbevacizumab (Avastin) and cetuximab (Erbitux), each given withgemcitabine; irradiation will be given with oral capecitabine (Xeloda).
Locally advanced but potentionally resectable lesions
These lesions comprise 10%-15% of the cases presenting to physicians. Datafrom preoperative chemoradiation series indicate that trimodality therapy isessential for long-term survival in these patients. There is now a randomizedphase II ECOG trial (E1200) evaluating gemcitabine with radiation therapy(500 mg/m
/50 min/wk during radiation therapy of 5,040 cGy with tightenedfields) plus postresectional gemcitabine or preoperative chemotherapy, with adose of two cycles of gemcitabine, 5-FU, and cisplatin, followed by 5-FU/radiationtherapy, then surgery, and then postoperative gemcitabine. End pointsare margin-free resectability and survival.
TREATMENT OF UNRESECTABLE DISEASE
can prolong and/or improve quality of life in some patients withunresectable adenocarcinoma of the pancreas. It is better combined with chemotherapy.Long-term survival is, unfortunately, highly unusual.
The addition of chemotherapy to radiation therapyhas been shown to improve the survival of patients with unresectable pancreaticadenocarcinoma, with moderate doses of radiation only slightly less effectivethan higher doses. In a GITSG trial of unresectable disease, moderatedoseradiation (4,000 cGy) with 5-FU chemotherapy significantly improvedsurvival, as compared with higher doses of radiation (6,000 cGy) and no chemotherapy(median survival, 9.6 vs 5.2 months). The GITSG has also comparedchemotherapy plus irradiation with chemotherapy alone and demonstrateda significant improvement with combined-modality therapy (mediansurvival, 42 vs 32 weeks).
Based on these data, except in a protocol setting, the palliative management ofa patient with unresectable pancreatic adenocarcinoma who has significant localsymptoms should probably consist of moderate doses of radiation (4,000-5,000 cGy) in conjunction with 5-FU-based chemotherapy. As in adjuvant treatment,carefully shaped portals approximately 12 x 12 cm should be used.
Approaches under investigation
At present, numerous trials are exploring a varietyof chemoradiation therapy approaches, including single-agent or combinationtherapy with oral or infusional 5-FU, paclitaxel, cisplatin, gemcitabine, docetaxel(Taxotere), and oxaliplatin (Eloxatin). Trials with combined gemcitabine andirradiation are of particular interest due to the activity of this drug in pancreaticcancer and the fact that it is a potent radiosensitizer. The benefit of irradiationfor patients with locally advanced disease, however, remains a research questionbecause of toxicity concerns and the relatively brief survival rates. Therefore,a new ECOG trial will evaluate gemcitabine alone vs gemcitabine andirradiation for this group of patients.If gemcitabine is given either before or after a course of radiation therapy, fulldoses of 1,000 mg/m
are possible. If irradiation and gemcitabine are given concurrently,doses of either modality must be sharply reduced. A current phase IItrial is combining "full-dose" gemcitabine (1,000 mg/m
) with radiation therapydirected at the primary tumor alone (36 Gy). In addition, a Radiation Therapydirected at the primary tumor alone (36 Gy). In addition, a Radiation TherapyOncology Group (RTOG) randomized trial evaluated radiation therapy(50.4 Gy) and weekly gemcitabine and paclitaxel with or without the farnesyltransferase inhibitor R11577 for locally advanced pancreatic cancer. In thattrial, the gemcitabine dose was 75 mg/m
/wk and the paclitaxel dose was 40mg/m
/wk. Radiation was conventionally fractionated to a dose of 50.4 Gy.Both these trials are now completed and being analyzed.The dose of gemcitabine that can be given concurrently with irradiation dependson the volume and dose of radiation. If full doses of gemcitabine (1,000mg/m
/wk) are given concurrently with irradiation, the dose of radiation mustbe makedly reduced to avoid unacceptable GI toxicity.
TREATMENT OF METASTATIC ADENOCARCINOMA
Pancreatic adenocarcinoma is still one of the most frustrating, resistant solid neoplasmsto treat, and therapy for metastatic disease remains palliative. Few agentshave demonstrated activity of > 10%. Moreover, most of the reported series havebeen small, and not all encouraging results have been duplicated.
As metastatic pancreatic carcinoma is incurable, the anticipated risks ofchemotherapy, which are often substantial, must be balanced against the gainsthat may be achieved; unfortunately, they are few. Patients who are debilitateddue to their underlying or comorbid disease should not be offered chemotherapy,as their likelihood of deriving any benefit is exceedingly slim. However, patientswho desire therapy and who, while symptomatic, still have a good performancestatus may be offered "standard" chemotherapy (Table 2) or, if possible,should be encouraged to participate in a clinical trial.
Historically, single-agent 5-FU has been associated with a response rateof 25% in pancreatic cancer. FAM (5-FU, Adriamycin [doxorubicin], andmitomycin [Mutamycin]) and 5-FU plus doxorubicin offer no advantage over5-FU alone. 5-FU plus leucovorin appears to be ineffective.
is indicated for the treatment of locally advanced or metastaticpancreatic adenocarcinoma. Gemcitabine was compared with 5-FU in a groupof 126 previously untreated patients and showed a small, but statistically significant,improvement in response rate. Median survival in the gemcitabinegroup was 5.7 months, with 18% of patients alive at 12 months, as comparedwith a median survival of 4.4 months in the group receiving 5-FU, with 2% ofpatients alive at 12 months. Perhaps more important, clinical benefit response(a composite measurement of pain, performance status, and weight) occurred in23.8% of the gemcitabine-treated group, as compared with 4.8% of the 5-FU-treated group. Due to its palliative potential, gemcitabine has become the standardof care for patients with unresectable pancreatic adenocarcinoma.A recent randomized, phase II trial of dose-intense gemcitabine administeredby standard infusion vs a fixed-dose rate (10 mg/m
/min) suggested an improved1-year survival with the fixed-dose rate.
There have been anumber of recent attempts to improve thetherapeutic outcome for patients with metastaticpancreatic cancer by comparing promisingcombinations of agents in randomizedclinical trials. Unfortunately, the results havebeen disappointing. The ECOG comparedgemcitabine with or without 5-FU, demonstratinga median survival of 5.4 months forgemcitabine vs 6.7 months for the combination;however, this difference was not statisticallysignificant. Another trial explored theaddition of irinotecan to gemcitabine. Therewas no survival benefit when this regimen wascompared with gemcitabine alone, although the combination did increase tumorresponse rate (16.1% vs 4.4%,
< .001),Combination trials were presented during ASCO 2004. A European study ofthe topoisomerase inhibitor exatecan (DX-89511) vs gemcitabine included 339patients, showing no significant difference in survival. Furthermore, pain, qualityof life, and time to tumor progression were worse in the exatecan arm.Another trial evaluated exatecan and gemcitabine vs gemcitabine alone in 349patients. Efficacy parameters were similar for both arms of the trial.A phase III study of 565 patients compared gemcitabine with the combinationof gemcitabine plus the multitargeted antifolate pemetrexed (Alimta) and demonstrateda significant response benefit with the combination (14.8% vs 7.1%,
= .004). However, survival and disease progression-free survival were comparable.There was increased hematologic toxicity with the combination.A fourth trial evaluated standard-dose gemcitabine vs a fixed-rate infusion ofgemcitabine plus oxaliplatin. The trial accrued 326 patients and showed a superiorresponse rate for the combination (26.8% vs 11.3%) and superior diseaseprogression-free survival (5.8 vs 3.7 months). Although these results were notstatistically significant, a trend toward increased survival was shown with thegemcitabine/oxaliplatin combination. Furthermore, during ASCO 2003, a trialexploring the combination of cisplatin plus gemcitabine showed no survivaladvantage for the combination.
Agents with marginal activity
include mitomycin, doxorubicin, ifosfamide(Ifex), streptozocin (Zanosar), and docetaxel. To date, monoclonal antibodytherapy and hormonal manipulation have been ineffective. A phase II study ofanti-epidermal growth factor receptor (EGFR)-antibody IMC-C225(cetuximab) combined with gemcitabine has shown a 12% partial response rateand 39% stable disease in advanced pancreatic cancer. Side effects includedrash/folliculitis and fatigue. A phase III trial of the combination is accruingpatients (see boxed item).The ECOG is also exploring EGFR-directed therapy in pancreatic cancer witha new randomized phase II trial comparing docetaxel and irinotecan (CPT-11,Camptosar) with or without cetuximab. The trial will correlate EGFR expressionwith outcome after therapy. Other "targeted" therapies are underinvestigation.
A progressively better understanding of the molecular biologyof pancreatic cancer has revealed numerous new therapeutic targets. Areasof active current research include attempts to replace tumor-suppressor genes(ie,
53) and to inhibit K-
protein function.Many patients seek "complementary" or "alternative" treatment strategies. TheNCI (National Cancer Institute) has activated a phase III study of gemcitabinevs intensive pancreatic proteolytic enzyme therapy with ancillary nutritionalsupport for pancreatic cancer patients based on phase II data.
PANCREATIC ENDOCRINE TUMORS
Pancreatic endocrine tumors (PETs) cover a spectrum of neoplasms. Many,although not all, originate from the pancreatic islets of Langerhans.PETs are not rare. Autopsy studies have documented an incidence as high as1.5%. Most of these lesions are clinically silent.The normal islet contains α,β,γ cells and enterochromaffin cells, which primarilysecrete glucagon, insulin, somatostatin, and serotonin, respectively. All ofthese hormones may be secreted in excess byPETs. Other hormones that may be secretedby these tumors include vasoactive intestinalpeptide (VIP), gastrin, pancreatic polypeptide(PP), and calcitonin. The aggressiveness of aPET in terms of its metastatic potential appearsto be due to the cell of origin.
Types of tumors
Insulinomas are β-cell tumors of the pancreaticislets that produce insulin. Four-fifths ofinsulinomas occur as a solitary lesion, and < 10% of these tumors demonstratemalignant potential (in terms of invasiveness or the development of metastases).In patients with the multiple endocrine neoplasia type 1 (MEN-1)syndrome, insulinomas are multicentric (10% of patients). In addition, a smallgroup of insulinomas are associated with diffuse islet-cell hyperplasia ornesidioblastosis.
are gastrin-secreting tumors associated with the Zollinger-Ellisonsyndrome (ZES). These tumors can be either sporadic or familial. Sporadicgastrinomas do not have associated endocrinopathies, whereas hereditarygastrinomas occur in patients with MEN-1 syndrome. Patients with thesporadic form of ZES may have single or multiple gastrinomas. This findingcontrasts with patients with hereditary MEN-1 PETs, who generally have amore diffuse tumor process within the pancreas.It is known that 80%-90% of gastrinomas are located within the "gastrinomatriangle,"defined as the junction of (1) the cystic and common duct, (2) thesecond and third portions of the duodenum, and (3) the neck and body of thepancreas. Although tumors most characteristically are located within the pancreas,a significant percentage of patients with ZES demonstrate primary tumorsof the duodenal wall. Extrapancreatic and extraintestinal locations occurin approximately 10% of patients.More than 90% of gastrinomas are malignant. The spectrum of clinical diseaseprogression includes localized tumors, regional lymph node metastases, andwidespread metastatic disease.
Approximately three-quarters of VIPomas and approximatelyhalf of all glucagonomas and somatostatinomas are malignant.
Although many PETs cause considerable morbiditydue to the inappropriately elevated levels of the hormones that they secrete,even "nonfunctional" PETs, ie, those without an associated demonstrable hormone-related syndrome (such as PPomas, neurotensinomas, and nonsecretoryPETs), may be aggressive. Nonfunctional tumors account for up to 30% of allPETs. Two-thirds of these nonfunctional tumors will demonstrate metastaticlesions at some point during the patient's lifetime.
Signs and symptoms
The symptom complex that is observed depends on which hormone or hormonesare secreted in excess.
are associated with symptoms of recurrent hypoglycemia. Diagnosisof these tumors is made by the demonstration of inappropriately elevatedlevels of insulin, proinsulin, and C peptide at the time of hypoglycemiaand an elevated insulin-glucose ratio (> 0.3).
Symptoms of gastrinoma-ZES are due to the effect of elevatedlevels of circulating gastrin. Ulceration of the upper GI tract is seen in > 90% ofpatients. Diarrhea is the second most common symptom. Approximately 25%of gastrinomas occur in the context of MEN-1 and are associated with parathyroidhyperplasia and hypercalcemia.The diagnosis of ZES is established by the demonstration of hypergastrinemia(fasting serum gastrin concentration > 1,000 pg/mL) and gastric acid hypersecretionin a patient with ulcerative disease.
VIP excess causes a profuse, watery diarrhea, hypokalemia,hypophosphatemia, and hypochlorhydria (WDHA syndrome).
are associated with a rash (described as a necrotizing migratoryerythema), glossitis, cheilosis, constipation and ileus, venous thrombosis,and hyperglycemia. Not all of these manifestations are secondary to elevatedglucagon levels alone. The etiology of these signs and symptoms remains unknown,but some patients respond to supplemental zinc and amino acid infusions.
which are rare, are associated with elevated blood glucoselevels, achlorhydria, cholelithiasis, and diarrhea.
Ultrasonography, CT, MRI, and selective arteriography withportal vein sampling have been utilized for the preoperative localization ofinsulinomas. The sensitivity of these preoperative imaging tests ranges fromapproximately 30% to 60%. This is because 40% of insulinomas are ≤ 1 cm andtwo-thirds of these tumors are < 1.5 cm.Because the success of preoperative localization tests is disappointing and 90%of these tumors will be found and successfully resected by an experienced endocrinesurgeon, there is a general trend toward performing fewer tests. Somecenters utilize preoperative ultrasonography if the patient has not undergoneprior pancreatic surgery. Other centers still routinely employ portal vein catheterizationand angiography.More recently, intraoperative sonography has been shown to aid the surgeon.In one series, 84% of tumors not localized preoperatively were correctly locatedby surgical exploration and intraoperative sonography. Many lesions notdiscovered by surgical palpation may be found by this technique. At present,there is much less reliance on blind distal resection than was previously advocated.Obviously, the technique of intraoperative ultrasonography may not beas helpful in the MEN-1 syndrome, in which multiple small insulinomas maybe found.
CT, ultrasonography, selective abdominal angiography, selectivevenous sampling of gastrin, intraoperative ultrasonography, EUS, and intraoperativeendoscopy have all been reported to be useful in localizinggastrinomas. More recently, somatostatin receptor scintigraphy (SRS) has becomea valuable tool for PET localization; several studies have suggested greatersensitivity and specificity when compared with other diagnostic tests.
Surgery for insulinomas
For larger insulinomas in the body or tail of the pancreas, a distal pancreatectomymay be preferable to enucleation. For tumors in the head of the pancreas,enucleation of the tumor is usually possible. Patients with MEN-1 or islet-cellhyperplasia may benefit from an 80% distal pancreatectomy. If the insulinomais not found at surgery, a blind pancreatectomy is not warranted. Further imagingand venous sampling studies may reveal the exact location of the tumor.A surgical cure results in normal values on subsequent provocative testing,during which blood insulin and glucose concentrations are measured simultaneously.Some insulinoma recurrences actually represent persistent diseaseafter incomplete tumor excisions or overlooked secondary multiple tumors.
Surgery for gastrinoma-ZES
The ideal treatment for gastrinoma-ZES is surgical excision of the gastrinoma.However, this approach is possible in only 20% of patients, most of whom havea sporadic tumor. With the development of effective antisecretory agents andpreoperative localization with octreotide scanning, the majority of patients demonstratingwidespread metastatic disease can be identified and spared surgicalexploration. In addition, some series report that patients with nonmetastaticsporadic gastrinoma may have a higher incidence of extrapancreatic sites thanwas previously thought. One series has reported that two-thirds of gastrinomasare extrapancreatic.
Patients with sporadic gastrinoma
All patients with sporadic gastrinomashould undergo localization studies and be considered for exploratorylaparotomy, with the goal of potential cure of ZES. Recent evidence suggeststhat resection of primary gastrinoma decreases the incidence of liver metastasesand ZES. Overall, surgery produces complete remission in approximately 60%of patients with sporadic ZES, and subsequent survival is excellent.
Patients with ZES and MEN-1
Some experts believe that surgery should notbe used in the management of patients with MEN-1 and ZES. Instead, theyrecommend treatment with antisecretory medications. This approach is somewhatcontroversial, as some authors believe that all patients without demonstratedliver metastases should undergo surgery to remove duodenal andpancreatic gastrinomas.Moreover, since many patients with ZES and MEN-1 die of metastaticgastrinoma at a young age, a surgical approach may be warranted. Surgeryshould be performed only if imaging studies localize the tumor. Although radicalsurgery may not provide a cure, removal of large tumors may decreasemetastatic potential and increase survival.
During surgery, the entire pancreas should be mobilizedand scanned ultrasonographically to permit a thorough examination of the pancreatichead, duodenum, stomach, mesentery, liver, and splenic hilum. Intraoperativeendoscopy with transillumination of the bowel wall may also be usefulin identifying duodenal lesions. In general, enucleation is the treatment of choice,except for lesions within the duodenal wall, which may require pancreaticoduodenectomy.If no tumor is found, blind distal pancreatectomy should beavoided, since 90% of gastrinomas are located within the gastrinoma triangle.Surgical resection of liver metastases is controversial. However, several authorshave demonstrated meaningful survival in patients with small, isolated lesions.The use of ablative procedures, with open, laparoscopic, or percutaneous techniques,can reduce the neurohormonal tumor burden.
Radiation therapy for PETs
The role of adjuvant radiation therapy for PETs of the pancreasis unclear. Because of the rarity of these lesions and their often indolentbehavior, the role of this therapy will probably never be demonstrated. However,postoperative irradiation can be considered for patients with positive nodesor microscopically close margins. Concurrent chemotherapy with such agentsas 5-FU and/or streptozocin also can be considered. Radiation doses are thesame as are used in adjuvant treatment of pancreatic cancer.
Anecdotal reports indicate that pancreatic PETs may respondto palliative doses of irradiation. Long-term control of unresectable diseasehas been reported.
Chemotherapy for PETs
PETs are more sensitive to chemotherapy than are carcinoid tumors.
Agents that have demonstrated antitumor activity includerecombinant human interferon alfa-2a and alfa-2b (Roferon-A, Intron A, respectively),5-FU, doxorubicin, dacarbazine (DTIC-Dome), and streptozocin.
Combination chemotherapy is often more effectivethan monotherapy. For example, in an ECOG study, the combination of5-FU and streptozocin demonstrated a higher response rate than streptozocinalone (63% vs 36%) in PETs, as well as a better complete response rate (33% vs12%) and median survival duration (26.0 vs 16.5 months). Therapy with doxorubicinplus streptozocin was superior to therapy with both 5-FU plusstreptozocin and single-agent chlorozotocin in terms of response and survivaland is the combination most widely used in the United States. Etoposide combinedwith cisplatin is active in poorly differentiated neuroendocrine malignanciesbut is marginally effective in well-differentiated lesions.
TREATMENT OF SYMPTOMS
Octreotide (Sandostatin) is often successful in palliating symptoms in patientswith PETs, although this success depends somewhat on the cell type. For example,insulinomas are marginally responsive to octreotide, whereas gastrinomasand VIPomas often respond. However, compared with carcinoid tumors, themedian duration of response of PETs to octreotide is significantly shorter (~10weeks).As discussed more fully in the section on carcinoid tumors below, a promisingexperimental approach for patients whose tumors express somatostatin receptorsis the use of octreotide conjugated to a therapeutic radioisotope.
Omeprazole (Prilosec), an inhibitor of the function of the parietal cell hydrogenpump, is more effective than H
-receptor antagonists in blocking gastricacid production and is useful in the symptomatic management of gastrinomas.Other agents available for symptomatic treatment of insulinomas includediazoxide (Hyperstat), an insulin-release inhibitor, and, more recently, glucagon,by continuous infusion through a portable pump. Both of these agents areused in conjunction with frequent high-carbohydrate meals.Patients with the glucagonoma syndrome are treated symptomatically withinsulin, high-protein meals, supplemental zinc, amino acid infusions, andanticoagulants.
Hepatic arterial embolization
Hepatic arterial embolization, with or without chemotherapy (chemoembolization),is an alternative palliative therapy for patients with either carcinoidtumors or a PET who have predominant liver metastases or who are symptomatic.Embolization is best reserved for patients with < 75% tumor involvementof the liver, bilirubin level < 2 mg/dL, and an ECOG performance statusof ≤ 2.
CARCINOID TUMORS OF THE GI TRACT
Carcinoid tumors typically arise from components derived from the primitivegut, lungs, and, rarely, the gonads. Approximately 85% of all carcinoidsoriginate from the gut, predominantly the appendix, followed by the smallbowel and rectum.These tumors have the propensity to cause considerable morbidity by virtue ofcreating a syndrome of hormonal excess. For example, although the majorityof carcinoids are hormonally inert, these neoplasms may produce excessiveamounts of serotonin (from dietary tryptophan), prostaglandins, kinins (secondaryto kallikrein release), and a variety of other hormones, which mayaccount for the "carcinoid syndrome."
Signs and symptoms
The most common sign of the carcinoid syndrome is flushing, whichis often triggered by alcohol, catecholamines, or emotional stress. It ranges inseverity from a minor annoyance to profound vasodilatation with near syncopeand hypotension.
is also common and is due to GI hypermotility. It usually occursafter meals and is rarely voluminous, bulky, or foul-smelling.
Diarrhea may be associated with crampy pain, althoughother etiologies for the pain must be considered, including bowel obstructiondue to tumor or mesenteric fibrosis.
Patients may also develop bronchospasm, which may be mediatedby histamine. This problem is often associated with (although less commonthan) flushing.
Valvular heart disease
A late finding is right-sided valvular heart disease,although left-sided lesions may be noted occasionally. The fibrous deposits maylead to tricuspid insufficiency and/or pulmonary stenosis. Valve replacement israrely necessary, however.
If there is sufficient shunting of dietary tryptophan from niacinto serotonin synthesis, patients may develop diarrhea, dermatitis, anddementia, although this symptom triad is rare if patients maintain adequateintake of a balanced diet.
Diagnostic studies include CT/MRI of the abdomen and a 24-hour urine testfor 5-hydroxyindoleacetic acid (5-HIAA). Some radiologists prefer to obtain atriple-phase CT scan of the liver to detect these highly vascular liver metastases.
Indium-111 octreotide scintigraphy (OctreoScan) hasbeen shown to have a higher sensitivity for detecting pancreatic tumors and issuperior to CT or MRI for detecting metastatic disease, particularly extrahepaticdisease. One study suggests that indium-111 octreotide scintigraphy can reducecosts by avoiding unnecessary surgeries. Also, a positive scan may predict whichpatients may benefit from treatment with somatostatin analogs (eg, octreotide).Initial studies with a new peptide tracer, indium-111 DOTA-lanreotide, suggesthigh tumor uptake and a more favorable dosimetry than is seen with indium-111 DTPA-D-Phel-octreotide.
Site and size of tumor
The site of tumor origin is potentially prognostic, asmost appendiceal carcinoids (75%) are < 1 cm when found and are usuallycured by resection. Similarly, rectal carcinoids are usually small and completelyresectable for cure.In contrast, small bowel carcinoids tend to present at a more advanced stage,and approximately one-third have multicentric primary lesions. However, ifthe disease is completely resectable, patients have a 20-year survival rate of80%; patients with unresectable intra-abdominal or hepatic metastases havemedian survival durations of 5 and 3 years, respectively.
The management of carcinoid tumors focuses not only on treating bulky disease,in common with other solid malignancies, but also on treating the complicationsof hormonal excess.
TREATMENT OF BULKY DISEASE
For tumors that are found incidentally in the appendixand that are probably between 1 and 2 cm, appendectomy is the treatmentof choice. For tumors > 2 cm, a right hemicolectomy and lymph node dissectionare appropriate.
Small intestine and rectal carcinoids
should be resected with a wedgelymphadenectomy to evaluate nodal disease. Duodenal lesions should be locallyexcised if small (< 2 cm), with radical resection reserved for larger tumors.
Liver resection or ablation of liver metastases with cryotherapyor radiofrequency techniques is useful in patients with limited extrahepaticdisease and/or asymptomatic carcinoid syndrome. Tumor debulking canprotect liver functional reserve and improve quality of life.
may be of benefit in selected patients without extrahepaticdisease whose cancer progresses after other therapeutic interventions.
Carcinoid tumors are responsive to radiation therapy and frequently are wellpalliated with this modality. Overall, treatment with higher radiation doses (29-52 Gy) has been associated with higher response rates (40%-50%) than treatmentwith lower doses (10%).
Since carcinoid tumors tend to be resistant to most chemotherapeutic agents,there are no standard regimens for the treatment of unresectable tumors.
Agents that have reported activity include 5-FU, doxorubicin,and recombinant human interferon alfa-2a and alfa-2b. However, the responserate with these agents is in the range of 10%-20%, the response duration is< 6 months, and complete remission is rare.
Combination chemotherapy regimens represent littleimprovement over single-agent therapy, with response rates ranging from 25%to 35%, response durations < 9 months, and rare complete remissions.
TREATMENT OF SYMPTOMS
The most active agent is the somatostatin analog octreotide. Eventhough native somatostatin is effective in controlling many symptoms, due toits short half-life (< 2 minutes), this agent would have to be administered viacontinuous infusion to be clinically useful. However, octreotide may be administeredsubcutaneously every 8-12 hours, facilitating outpatient therapy. Theinitial dose of octreotide is 100-600 μg/d in 2-4 divided doses, although theeffective dose varies between patients and must be titrated to the individualpatient's symptoms.Octreotide not only is useful in managing the chronic problems of the carcinoidsyndrome but also is effective in treating carcinoid crisis (volume-resistanthypotension), which may be precipitated by surgery or effective antitumortreatment.Octreotide is well tolerated, although chronic treatment may be associated withcholelithiasis, increased fecal fat excretion, fluid retention, nausea, and glucoseintolerance. Occasional objective antitumor responses have been observed inpatients who have received octreotide; the median duration of symptomaticimprovement is 1 year. One report evaluating the cost-effectiveness of octreotidesuggested that it may double survival time. Other somatostatin analogs, includinglanreotide and vapreotide, are under investigation.
SMS 201-995 pa LAR
is a long-acting somatostatin analog that allows formonthly dosing, avoiding the need for three daily injections. This new agentimproves quality of life while apparently maintaining the same activity seenwith daily octreotide. The usual monthly dose is 20 or 30 mg.Patients who demonstrate disease resistance with somatostatin analog treatmentalone may benefit from combination therapy with interferon-? and thissomatostatin analog.
Radiolabeled somatostatin analogs
A promising experimental treatmentapproach involves the use of octreotide or other somatostatin analogs conjugatedto radioisotopes (eg, indium-111 or yttrium-90) in patients whose tumorsexpress somatostatin receptors (eg, those with a positive OctreoScan result).This approach allows targeted in situ radiotherapy by taking advantage of internalizationof the radioligand into the cell to produce DNA damage and celldeath, with little effect on normal tissue. Initial reports have shown favorableresults with this technique.
Other agents that have been used for symptomatic management include H
-receptor antagonists, methoxamine (Vasoxyl), cyproheptadine, anddiphenoxylate with atropine. The symptom complex of diarrhea, dermatitis,and dementia may be prevented or treated with supplemental niacin.
Hepatic arterial embolization
Hepatic arterial embolization with such agents as Ivalon or Gelfoam, with orwithout chemotherapy (chemoembolization), is an option for patients with eithera carcinoid tumor or an islet-cell carcinoma who have predominant livermetastases or who are symptomatic. These lesions often are hypervascular,and, thus, peripheral hepatic embolization may provide symptomatic relief insome patients. It is unclear whether this therapy has any effect on patient survival.
Adrenocortical carcinoma is a rare, highly malignant neoplasm that accountsfor about 0.2% of cancer deaths. Long-term survival is dismal overall; the survivalrate is 23% at 5 years and 10% at 10 years.
The etiology of adrenocortical cancer is unknown, but some cases have occurredin families with a hereditary cancer syndrome.
Signs and symptoms
Approximately half of adrenocortical neoplasms produce hormonal and metabolicsyndromes of hormone hypersecretion (such as Cushing's syndrome, virilizingor feminizing syndromes, and hyperaldosteronism). In children,Cushing's syndrome is rare but is often due to adrenal carcinoma. Mixed syndromes,such as Cushing's syndrome and virilization, strongly suggest adrenalcarcinoma. The combination of hirsutism, acne, amenorrhea, and rapidlyprogressing Cushing's syndrome in a young female is a typical presentation. Inmen, estrogen-secreting tumors are associated with gynecomastia, breast tenderness,testicular atrophy, impotence, and decreased libido.Often the diagnosis of adrenocortical carcinoma is not evident until the discoveryof metastases or until the primary tumor becomes large enough to produceabdominal symptoms. Smaller tumors may be discovered incidentally, whenunrelated abdominal complaints are investigated radiographically.
Complete surgical resection is the treatment of choice in patients with localizeddisease, as it offers the best chance of extending the disease-free intervaland survival.
is one of only a few effective agents; it exerts a specificcytolytic effect on adrenocortical cells and has been used to treat unresectableor metastatic adrenocortical carcinoma. Only 15%-30% of patients experienceobjective tumor regression, with a median duration of about 7 months. Mitotaneis given at a dose of 4-8 g/d as tolerated, although the dose is variable.
Doxorubicin has been of benefit in a limited number of patients,and combination chemotherapy is under investigation.Suramin (Metaret), a sulfonated drug that is cytotoxic to human adrenocorticalcarcinoma cell lines, has been evaluated but has not proven useful in inoperableadrenocortical cancer. Innovative chemotherapy programs are clearlyneeded for this disease.
Controlling hormone hypersecretion
Hormone hypersecretion can becontrolled medically in most cases. Agents that are effective in reducing steroidproduction and in palliating associated clinical syndromes include the antifungaldrug ketoconazole (Nizoral), 800 mg/d; aminoglutethimide (Cytadren),1-2 g/d; and metyrapone (Metopirone), 1-4 g/d or higher as needed to controlcortisol levels. These agents may be used alone or with mitotane.
Pheochromocytomas are catecholamine-secreting tumors that arise from chromaffincells in the adrenal medulla or extra-adrenal sympathetic ganglia. Thesetumors constitute a surgically correctable cause of hypertension in 0.1%-1.0%of hypertensive persons.Only about 10% of pheochromocytomas are considered to be malignant. Thevast majority (90%) of pheochromocytomas are found in the adrenal medulla,and 97% are located below the diaphragm. Approximately 10% each ofpheochromocytomas are bilateral, malignant, multifocal, extra-adrenal, foundin children, or associated with a familial syndrome.Pheochromocytomas in patients with familial syndromes, such as MEN-2 andvon Hippel-Lindau syndrome (VHL), are less likely to be malignant than otheradrenal lesions. In contrast, pheochromocytomas in patients with a family historyof malignant pheochromocytoma are more apt to be malignant.
Epidemiology and etiology
Pheochromocytomas occur in all age groups, but the incidence peaks in thethird to fifth decades of life. Most pheochromocytomas (90%) are sporadic.Approximately 10% of cases are inherited as an autosomal-dominant trait, eitherindependently or as a part of the MEN-2 syndrome; bilateral tumors are morecommon in this setting.Both MEN-2A and MEN-2B include medullary thyroid carcinoma and pheochromocytoma.MEN-2A includes hyperparathyroidism, whereas MEN-2Bincludes ganglioneuromas and marfanoid habitus. In MEN-2 families, pheochromocytomaoccurs in 5.5%-100% (mean, 40%), depending on the kindredstudied. Bilateral medullary hyperplasia is almost always present. Pheochromocytomasare bilateral in 70% of cases and usually multicentric, but they arerarely extra-adrenal or malignant.
Signs and symptoms
Patients can present with various symptoms, ranging from mild labilehypertension to hypertensive crisis, myocardial infarction, or cerebral vascularaccident, all of which can result in sudden death. The classic pattern ofparoxysmal hypertension occurs in 30%-50% of cases; sustained hypertensionmay also occur and resembles essential hypertension. A characteristic presentationincludes "spells" of paroxysmal headaches, pallor or flushing, tremors,apprehension, palpitations, hypertension, and diaphoresis.
The diagnosis of pheochromocytoma relies on an appropriate history and documentationof excessive catecholamine production.
Measurement of 24-hour urinary catecholaminesand their metabolites, vanillylmandelic acid and metanephrine, iscommonly used; the metanephrine level is considered to be the most specificsingle test. Serum catecholamine measurements are more susceptible to falseelevations due to stress-related physiologic fluctuations. The evaluation of serumcatecholamines after clonidine suppression, however, provides a usefuldiagnostic tool that is more convenient than urine collection. Dynamic provocativetests are rarely indicated.
Almost all pheochromocytomas are localized in the abdomen,mostly in the adrenal medulla; other locations include the posterior mediastinumor any distribution of the sympathetic ganglia. After the diagnosis isestablished biochemically, radiologic methods may be needed for preoperativelocalization of the lesion; CT and MRI are most widely used. Iodine methyliodobenzylguanidine (MIBG) and SRS provide a "functional" image; they aremost helpful in the detection of occult contralateral or extra-adrenal lesions.
Differentiating benign from malignant tumors
The histologic differentiationbetween benign and malignant lesions is extremely difficult and often impossibleto make; this distinction may require the development of lymph node,hepatic, bone, or other distant metastases. Recurrent symptoms of pheochromocytoma,often emerging many years after the original diagnosis, are suggestiveof malignancy. Biochemical confirmation of recurrent catecholaminehypersecretion and localization of metastatic lesion(s) with iodine-131-MIBGscan constitute diagnostic proof.
PREOPERATIVE MEDICAL MANAGEMENT
Phenoxybenzamine (Dibenzyline), an oral, long-acting, noncompetitive?-adrenoceptor blocker, is a widely used, very helpful first drug; it is given at adose of 10-40 mg/d. Propranolol, a β-blocker (20-80 mg/d), is usually addedafter a few days to prevent tachycardia or arrhythmia. The use of β-blockersalone is hazardous because they may precipitate a paradoxical rise in bloodpressure. The tyrosine hydroxylase inhibitor metyrosine (Demser) may be addedin patients whose blood pressure is not well controlled with the combination ofan α- and a β-blocker.
The principles of pheochromocytoma resection are complete tumor resection,avoidance of tumor seeding, and minimal tumor manipulation. Adrenalectomycan be performed by means of an open anterior transabdominal, open posteriorretroperitoneal, laparoscopic lateral transabdominal, or laparoscopic posteriorretroperitoneal approach. In the past, an open anterior approach was the standardbecause it allowed for complete exploration and inspection for potentialtumor foci. However, with the improved accuracy of preoperative imagingand increased experience with laparoscopic procedures, there is little need forexploration in areas in which a tumor has not been identified.Except for in tumors < 6 cm, the laparoscopic approach to pheochromocytomais probably the technique of choice. In the absence of obvious local tumorinvasion or metastatic disease, a laparoscopic procedure is acceptable to manyexperienced endocrine surgeons.The most critical intraoperative aspect of surgery is control of blood pressureimmediately after removal of the tumor, when all agonistic effects are abolishedand the effects of α- and β-blockers are still present. Close cooperationwith the anesthesiologist to expand fluid volume and prepare the appropriateinfusions of agonists to support vascular stability is critical.
TREATMENT OF METASTATIC MALIGNANT
The treatment of choice for metastatic malignant pheochromocytoma remainsproblematic.
Medical and radiation therapy
Medical therapy with α- or β-blockers, as well as metyrosine, is almost alwaysrequired to maintain hemodynamic stability. Chemotherapy utilizingstreptozocin-based regimens or the combination of cyclophosphamide (Cytoxan,Neosar), vincristine, and dacarbazine has yielded promising responses. Treatmentwith iodine-131-MIBG or (in Europe) with radiolabeled somatostatinhas met with only limited success. In most cases, uncontrolled catecholaminehypersecretion eventually escapes biochemical blockade, and fatal hypertensivecrisis ensues.
In those cases in which limited and resectable lesions can be identified, surgerycan effect complete and lasting remission of the disease.
ON PANCREATIC CANCER
Berlin JD, Catalano P, Thomas JP, et al:
Phase III study of gemcitabine in combinationwith fluorouracil versus gemcitabine alone in patients with advanced pancreatic carcinomas.Eastern Cooperative Oncology Group Trial E2297. J Clin Oncol 1:3270â3275,2002.
Buchsbaum DJ, Bonner JA, Grizzle WE, et al:
Treatment of pancreatic cancer xenograftswith erbitux (IMC-C225) anti-EGFR antibody, gemcitabine, and radiation. IntJ Radiat Oncol Biol Phys 15:1180â1193, 2002.
Crane CH, Wolff RA, Abruzzese JL, et al:
Combining gemcitabine with radiation inpancreatic cancer: Understanding important variables influencing the therapeutic index.Semin Oncol 28(suppl 10):25â33, 2001.
Huang JJ, Yeo CJ, Sohn TA, et al:
Quality of life and outcomes after pancreaticoduodenectomy.Ann Surg 231:890â898, 2000.
Maisey N, Chau I, Cunningham D, et al:
Multicenter randomized phase III trial comparingprotracted venous infusion (PVI) fluorouracil (5-FU) with PVI 5-FU plus mitomycinin inoperable pancreatic cancer. J Clin Oncol 15:3130â3136, 2002.
Milano MT, Chmura SJ, Garofalo MC, et al:
Intensity-modulated radiotherapy in treatmentof pancreatic and bile duct malignancies: Toxicity and clinical outcome. Int J RadiatOncol Biol Phys 1:445â453, 2004.
Neoptolemos JP, Stocken DD, Friess H, et al:
A randomized trial of chemoradiotherapyand chemotherapy after resection of pancreatic cancer. N Engl J Med 18:1200â1210,2004. Erratum in: N Engl J Med Aug 12:726, 2004.
Rocha Lima CM, Green MR, Rotche R, et al:
Irinotecan plus gemcitabine results in nosurvival advantage compared with gemcitabine monotherapy in patients with locallyadvanced or metastatic pancreatic cancer despite increased tumor response rate. J ClinOncol 15:3776â3783, 2004.
Rose DM, Delbeke D, Beauchamp RD, et al:
1-Fluorodeoxyglucose-positron emissiontomography in the management of patients with suspected pancreatic cancer. Ann Surg229:729â737, 1999.
Tempero M, Plunkett W, Ruiz Van Haperen V, et al:
Randomized phase II comparisonof dose-intense gemcitabine: Thirty-minute infusion and fixed dose rate infusion inpatients with pancreatic adenocarcinoma. J Clin Oncol 15:3402â3408, 2003.
Willett CG, Clark JW:
Update on combined-modality treatment options for pancreaticcancer. Oncology (Huntingt) 17:29â36, 2003.
Wong GY, Schroeder DR, Carns PE, et al:
Effect of neurolytic celiac plexus block onpain relief, quality of life, and survival in patients with unresectable pancreatic cancer: Arandomized controlled trial. JAMA 3:1092â1099, 2004.
The Johns Hopkins experience with pancreaticoduodenectomy with or withoutextended retroperitoneal lymphadenectomy for periampullary carcinoma. J GastrointestSurg 4:231â232, 2000.
ON NEUROENDOCRINE GI TUMORS
Ducreux M, Ruszniewski P, Chayvialle JA, et al:
The antitumoral effect of the longactingsomatostatin analog lanreotide in neuroendocrine tumors. Am J Gastroenterol95:3276â3281, 2000.
Faiss S, Pape UF, Bohmig M, et al:
Prospective, randomized, multicenter trial on theantiproliferative effect of lanreotide, interferon alfa, and their combination for therapyof metastatic neuroendocrine gastroenteropancreatic tumors. The InternationalLanreotide and Interferon Alfa Study Group. J Clin Oncol 15:2689â2696, 2003.
Fazio N, Oberg K:
Prospective, randomized, multicenter trial on the antiproliferativeeffect of lanreotide, interferon alfa, and their combination for therapy of metastatic neuroendocrinegastroenteropancreatic tumors. J Clin Oncol 1:573â574, 2004.
Kulke MH, Mayer RJ:
Carcinoid tumors. N Engl J Med 340:858â868, 1999.
Kvols LK, Krenning EP, Pauwels S, et al:
Phase I study of 90Y-SMT487 (Octreother)in patients with somatostatin receptor (SS-R) positive neuroendocrine (NE) tumors(abstract). Proc Am Soc Clin Oncol 19:207, 2000.
Moertel CG, Lefkopoulo M, Lipsitz S, et al:
Streptozocin-doxorubicin, streptozocinfluorouracil,or chlorozotocin in the treatment of advanced islet-cell carcinoma. N EnglJ Med 326:519â523, 1992.
Nilsson O, Kolby L, Wangberg B, et al:
Comparative studies on the expression ofsomatostatin receptor subtypes, outcome of octreotide scintigraphy and response tooctreotide treatment in patients with carcinoid tumours. Br J Cancer 77:632â637, 1998.
Norton JA, Alexander HR, Fraker DL, et al:
Comparison of surgical results in patientswith advanced and limited disease with multiple endocrine neoplasia type 1 and Zollinger-Ellison syndrome. Ann Surg 234:495â506, 2001.
Raderer M, Kurtaran A, Leimer M, et al:
Value of peptide receptor scintigraphy using(123)I-vasoactive intestinal peptide and (111)In-DTPA-D-Phel-octreotide in 194 carcinoidpatients: Vienna University Experience, 1993 to 1998. J Clin Oncol 18:1331â1336,2000.
Ramanathan RK, Cnaan A, Hahn RG, et al:
Phase II trial of dacarbazine (DTIC) inadvanced pancreatic islet cell carcinoma. Study of the Eastern Cooperative OncologyGroup-E6282. Ann Oncol 12:1139â1143, 2001.
Rubin J, Ajani J, Schirmer W, et al:
Octreotide acetate long-acting formulation versusopen-label subcutaneous octreotide acetate in malignant carcinoid syndrome. J ClinOncol 17:600â606, 1999.
Sarmiento JM, Que FG:
Hepatic surgery for metastases from neuroendocrine tumors.Surg Oncol Clin N Am 12:231â242, 2003.
Solorzano CC, Lee JE, Pisters PW, et al:
Nonfunctioning islet cell carcinoma of thepancreas: Survival results in a contemporary series of 163 patients. Surgery 130:1078â1085, 2001.
Virgolini I, Patri P, Novotny C, et al:
Comparative somatostatin receptor scintigraphyusing in-111-DOTA-lanreotide and in-111-DOTA-Tyr3-octreotide versus F-18-FDG-PETfor evaluation of somatostatin receptor-mediated radionuclide therapy. Ann Oncol12(suppl 2):S41âS45, 2001.
Waldherr C, Pless M, Maecke HR, et al:
The clinical value of [90Y-DOTA]-D-Phel-Tyr3-octreotide (90Y-DOTATOC) in the treatment of neuroendocrine tumours: A clinicalphase II study. Ann Oncol 12:941â945, 2001.
Yao KA, Talamonti MS, Nemcek A, et al:
Indications and results of liver resection andhepatic chemoembolization for metastatic gastrointestinal neuroendocrine tumors. Surgery130:677â682, 2001.
ON ADRENAL NEOPLASMS
Jossart GH, Burpee SE, Gagner M:
Surgery of the adrenal glands. Endocrinol MetabClin North Am 29:57â68, 2000.
Kendrick ML, Lloyd R, Erickson L, et al:
Adrenocortical carcinoma: Surgical progressor status quo? Arch Surg 136:543â549, 2001.
Sosa JA, Udelsman R:
Imaging of the adrenal gland. Surg Oncol Clin N Am 8:109â127,1999.
Vassilopoulou-Sellin R, Schultz PN:
Adrenocortical carcinoma: Clinical outcome atthe end of the 20th century. Cancer 92:1113â1121, 2001.
Wajchenberg BL, Albergaria Pereira MA, Medonca BB, et al:
Adrenocortical carcinoma:Clinical and laboratory observations. Cancer 88:711â736, 2000.
Goede AC, Winslet MC: Surgery for carcinoid tumours of the lower gastrointestinaltract. Colorectal Dis 5:123â128, 2003.
Kolby L, Persson G, Franzen S, et al:
Randomized clinical trial of the effect of interferonalpha on survival in patients with disseminated midgut carcinoid tumours. Br JSurg 90:687â693, 2003.
Krenning EP, de Jong M:
Therapeutic use of radiolabelled peptides. Ann Oncol 11(suppl3):267â271, 2000.
Carcinoid tumors: Molecular genetics, tumor biology, and update of diagnosisand treatment. Curr Opin Oncol 14:38â45, 2002.
Schell SR, Camp ER, Caridi JG, et al:
Hepatic artery embolization for control of symptoms,octreotide requirements, and tumor progression in metastatic carcinoid tumors. JGastrointest Surg 6:664â670, 2002.
Soreide JA, van Heerden JA, Thompson GB, et al:
Gastrointestinal carcinoid tumors:Long-term prognosis for surgically treated patients. World J Surg 24:1431â1436, 2000.
van der Horst-Schrivers AN, Wymenga AN, de Vries EG:
Carcinoid heart disease. NEngl J Med 348:2359â2361, 2003.
Baghai M, Thompson GB, Young WF Jr, et al:
Pheochromocytomas and paragangliomasin von Hippel-Lindau disease: A role for laparoscopic and cortical-sparing surgery.Arch Surg 137:682â689, 2002.
Eisenhofer G, Lenders JW, Linehan WM, et al:
Plasma normetanephrine andmetanephrine for detecting pheochromocytoma in von Hippel-Lindau disease and multipleendocrine neoplasia type 2. N Engl J Med 340:1872â1880, 1999.
Goldstein RE, O’Neill JA Jr, Holcomb GW III, et al:
Clinical experience over 48 yearswith pheochromocytoma. Ann Surg 229:755â764, 1999.
Vaughan ED Jr:
Diseases of the adrenal gland. Med Clin North Am 88:443â466, 2004.
Zendron L, Fehrenbach J, Taverna C, et al:
Pitfalls in the diagnosis ofphaeochromocytoma. BMJ 13:629â630, 2004.