Japanese Journal of Clinical Oncology 33:246-253 (2003)
© 2003 Foundation for Promotion of Cancer Research
Report of the 16th International Symposium of the Foundation for Promotion of Cancer Research: Recent Advances in Pancreatic Cancer
1 University of Texas, M.D. Anderson Cancer Center, TX, USA, 2 National Cancer Center, Tokyo and 3 Hepatobiliary and Pancreatic Surgery Division, National Cancer Center Hospital, Tokyo, Japan
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION SESSION 1: EPIDEMIOLOGY AND...SESSION 2: PATHOLOGY AND...SESSION 3: DIAGNOSIS OF...SESSION 4: STAGING OF...SESSION 5: SURGICAL TREATMENT...SESSION 6: NON-SURGICAL...ACKNOWLEDGMENTS |
|---|
The 16th International Symposium of the Foundation for Promotion of Cancer Research, entitled ‘Recent Advances in Pancreatic Cancer’, was held in Tokyo on January 22–24, 2003. The symposium was organized by Drs Tadao Kakizoe, James L. Abbruzzese, Hiroyasu Esumi and Tomoo Kosuge. Dr Takashi Sugimura was the adviser.
The prognosis of pancreatic cancer remains dismal despite the continuous efforts of investigators. It appears as if the recent evolution of life science has left pancreatic cancer behind. However, beneath the surface, new technology and knowledge are accumulating. This symposium was intended to help them to take shape by providing outstanding researchers from Japan, Europe and the United States with an opportunity to discuss and to exchange information.
|
SESSION 1: EPIDEMIOLOGY AND ETIOLOGY OF PANCREATIC CANCER |
|---|
|
TOPINTRODUCTIONSESSION 1: EPIDEMIOLOGY AND...SESSION 2: PATHOLOGY AND...SESSION 3: DIAGNOSIS OF...SESSION 4: STAGING OF...SESSION 5: SURGICAL TREATMENT...SESSION 6: NON-SURGICAL...ACKNOWLEDGMENTS |
|---|
Chairperson: Randall E. Brand
Dr Albert B. Lowenfels reviewed the epidemiology of pancreatic cancer. He pointed out that pancreatic cancer, one of the most lethal human cancers, is now a common cause of cancer death in many countries, including the USA and Japan. Persons at high risk of pancreatic cancer, in addition to the elderly, include blacks, smokers and patients with certain pre-existing diseases such as pancreatitis and long-term diabetes mellitus. Smoking doubles the risk of pancreatic cancer and causes early onset of disease in high-risk populations. Inherited diseases that are associated with increased risk of pancreatic cancer include familial pancreatic cancer, patients with BRCA2 mutation, familial atypical multiple-mole melanoma (FAMM) syndrome, hereditary non-polyposis colorectal cancer (HNPCC) syndrome, Peutz–Jeghers syndrome, familial adenomatous polyposis (FAP) and hereditary pancreatitis. Family history is a risk factor for pancreatic cancer. The risk of pancreatic cancer is increased by 18-fold if a first-degree relative had pancreatic cancer and by 57-fold if two family members had pancreatic cancer. Dietary factors are also important and there is evidence that a high-calorie diet and obesity increase the risk of pancreatic cancer, whereas consumption of fruits and vegetables and physical activity have protective roles. Studies have shown that ~30% of pancreatic cancers can be attributed to smoking, 20% to diet, 10% to hereditary diseases, 4% to benign pancreatic disorders and the remaining 36% to unknown etiology.
In view of the high mortality rate for pancreatic cancer and lack of effective treatment, an important strategy for reducing the burden of this cancer is to recognize risk factors and, if possible, to reduce exposures to these risk factors. High priority should be given to efforts to control smoking. At the present time, reducing the prevalence of smoking is the single most effective measure to prevent these lethal tumors. Rates of pancreatic cancer in the USA are currently declining in males but not females, presumably because of changes in smoking habits, the increasing incidence of pancreatic cancer in Japanese males may also be due to smoking.
Dr Shinichi Egawa focused his presentation on the pancreatic cancer registry in Japan. Since 1981, the Japan Pancreas Society (JPS) and the National Cancer Center have jointly sponsored a nationwide registry system for pancreatic cancer. By the end of 2001, they had collected 23 302 cases from 350 hospitals. The JPS and UICC have independently developed classifications for pancreatic cancer. Both systems are based on TNM classifications, but the definitions of each factor are somewhat different. The pros and cons of both systems were compared and the JPS system was found frequently to offer better separation of mortality between stages. Dr Egawa pointed out that coordinated efforts and better communication between JPS and UICC are needed. To reduce human errors in the process of cancer registration and to compare various factors internationally, JPS developed an electronic documentation system to permit online registration of patients with pancreatic cancer. Each patient’s record consists of 260 parameters, from which JPS and UICC factors are automatically calculated. Factors include conclusive findings from physical examination, imaging, surgical exploration and pathological studies. Each institute is asked to follow up with patients to update survival data. In this series the overall 5-year survival rate for patients with pancreatic cancer is 9.4% and median survival duration is 8.4 months. The favorable pancreatic tumors are cystic neoplasms, intraductal papillary neoplasms and endocrine tumors, with 5-year survival rates of 47.5, 59.1 and 67.2%, respectively. Approximately half of the patients in the registry have no histological confirmation of their cancer and this group of patients has the worst prognosis, with median survival duration of 4.3 months and a 5-year survival rate of 1.6%. Patients who undergo pancreatectomy have a 5-year survival rate of 13.4% and median survival duration of 11.7 months, whereas those who undergo palliative surgery, exploratory laparotomy or no surgery have a 5-year survival rate of 0.5% and median survival duration of 4.2 months.
|
SESSION 2: PATHOLOGY AND TUMOR BIOLOGY OF PANCREATIC CANCER |
|---|
|
TOPINTRODUCTIONSESSION 1: EPIDEMIOLOGY AND...SESSION 2: PATHOLOGY AND...SESSION 3: DIAGNOSIS OF...SESSION 4: STAGING OF...SESSION 5: SURGICAL TREATMENT...SESSION 6: NON-SURGICAL...ACKNOWLEDGMENTS |
|---|
Chairpersons: Roland M. Schmid and Albert B. Lowenfels
Dr Kiyoshi Mukai presented his lecture on the pathology of pancreatic cancer. The aggressive nature of pancreatic cancer is difficult to explain on the basis of morphology alone. More than 80% of pancreatic cancers are invasive ductal carcinoma of the pancreatic head and are histologically classified as well differentiated to moderately differentiated tubular adenocarcinomas. Two histological classification schemes of pancreatic tumors have been proposed: the Histological Typing of Tumors of the Exocrine Pancreas (1996) and the Japanese Classification in General Rules for the Study of Pancreatic Cancer (2001).
Several studies have been performed on histological prognostic factors of invasive ductal carcinomas. Certain reports have stated that the differentiation of adenocarcinomas is closely correlated with clinical behavior, while other reports did not confirm this finding. The discrepancy is probably caused by the fact that the diagnostic criteria are defined by descriptive terms, such as mild, moderate or severe atypia/dysplasia. Another reason for the discrepancies in diagnosis between Western and Japanese pathologists is the difference in methods used to determine the degree of differentiation. Western pathologists assign the histological grade by the least differentiated area, whereas Japanese pathologists classify carcinomas on the basis of the predominant area. This difference causes a higher incidence of well-differentiated adenocarcinomas in Japan. Dr Mukai stressed that these differences in histological diagnosis make international comparison of the biological behavior and treatment results of pancreatic cancer difficult and that there is great need to establish a unified histological grading system.
Dr Raul A. Urrutia presented a lecture entitled ‘Building a Bioinformatic Model for Better Understanding of Pancreatic Cancer’. Genomic bioinformatics is the large-scale processing of genomic information to gain a better understanding of biological processes. Dr Urrutia reported the establishment of BPRP, a bioinformatic platform for processing and archiving genomic information as it relates to normal pancreatic cell functions and pancreatic cancer. The building of BPRB provides information that (1) confirms previous knowledge regarding genotypic/phenotypic features of this organ in normal conditions and in pancreatic cancer, thus validating the veracity of the information that can be gathered through bioinformatics, and (2) generates a wealth of novel observations that expand our understanding of the function of pancreatic cells and the biological behavior of their cancerous counterparts. Dr Urrutia then reviewed the ongoing Pancreatic Proteomics project at the Mayo Clinic. The aims of the project are to determine human genome information specific for pancreatic cancer, build a bioinformatic-based model that explains the behavior of pancreatic cancer and build translational efforts based on the information in the database.
Dr Michiie Sakamoto focused his discussion on orthotopic models of pancreatic cancer. Dr Sakamoto and co-workers established orthotopic models of pancreatic cancer from both human pancreatic cancer tissue and cell lines. Using this model system, Dr Sakamoto demonstrated that orthotopically implanted tumors metastasize to lymph nodes as observed clinically in individual patients, but they rarely metastasize to the liver. In contrast, implanted cancer cell lines metastasize widely to sites including the liver and lungs. Dr Sakamoto raised several antibodies against xenograft tumors and cell lines, one of which was identified as an anti-dysadherin monoclonal antibody. Using the anti-dysadherin antibody as a research tool, his team found that dysadherin was overexpressed in moderately to poorly differentiated primary pancreatic cancers and metastatic lesions. Multivariate analysis revealed that dysadherin and E-cadherin were independent prognostic factors.
Dr Roland M. Schmid presented his lecture on a murine tumor progression model of pancreatic cancer. Dr Schmid and co-workers showed that transforming growth factor alpha (TGF-
) transgenic mice developed ductal pancreatic cancer at ages >1 year. The tumor cells were positive for cytokeratin 8/18 and 19. This tumor model allows study of tumor progression from normal ductal structure to premalignant lesions to invasive cancer. In premalignant lesions, activation of Ras and Erk was observed and also induction of cyclin D1-Cdk4 without an increase of cyclin E or PCNA. In addition, Bcl-xl was up-regulated and Bax down-regulated in early lesions. The up-regulation of Bcl-xl might be a consequence of activation of Stat3 and NF-
B. Inhibition of Stat3 and NF-B activation causes apoptosis of pancreatic cancer cells. These findings indicate that apoptosis resistance precedes formation of invasive pancreatic cancer. Moreover, Stat3 and NF-B signaling are important for tumor maintenance.
Crossbreeding with p53-null mice accelerates tumor development in TGF- transgenic mice. However, it still takes 120 days for tumors to develop, indicating that loss of p53 alone is not sufficient and that additional genetic changes are required. By using comparative genomic hybridization and loss of heterozygosity (LOH) analysis, two other genetic changes were identified. These are gene amplification of part of chromosome 15, including the myc locus, and loss of the distal part of chromosome 14, including the Rb1 locus. These data imply that dysregulation of the G1/S transition by up-regulation of c-myc or by loss of Rb1 in a p53-mutant background is required for pancreatic cancer development. In addition, LOH analysis revealed biallelic deletions of ink4a/arf or LOH of the Smad4 locus in some tumors, suggesting that multiple loci are involved in antitumor activities. In summary, this model recapitulates pathomorphological features of human pancreatic cancer. It will allow analyses that may increase understanding of epigenetic and genetic events important for pancreatic tumorigenesis.
Dr Hiroyasu Esumi presented a lecture entitled ‘Tumor Microenvironment and Cancer Treatment’. Tumors grow in a stressful environment because of uncontrolled proliferation of tumor cells and inadequate supply of oxygen and nutrients. Tumor cells have developed strategies to adapt to this stressful environment, including increased glycolysis, erythropeosis and angiogenesis. Tumor cells are confronted, however, with insufficient supplies of both oxygen and nutrients, and the switch to glycolysis seems inadequate to explain the adaptation. Dr Esumi hypothesized that increased glycolysis might not be sufficient to maintain cell energy production and that there might be a previously unknown biological response to insufficient blood supply and other stresses. To make his point clear, Dr Esumi used ‘austerity’ to describe tumor tolerance to nutrition. He subsequently reviewed laboratory evidence that supports his hypothesis. When HepG2, a human hepatoma cell line, was subjected to glucose starvation, cells started to die within 12 h and more than 80% of cells died within 24 h under normoxic conditions, whereas more than 80% of cells survived under hypoxic conditions (1% oxygen). The result was unexpected, because cell survival was expected to be much more dependent on glucose considering the increase in glycolysis during hypoxia. The molecular and biochemical mechanisms responsible for this hypoxia-enhanced cell survival were analyzed and found to be dependent on the presence of certain amino acids, PKB/Akt activity and MAPK activity. Since normal tissues are, theoretically, never exposed to a sustained insufficient blood supply, Dr Esumi hypothesized that tolerance of glucose deprivation may serve as a cancer tissue-selective target for treatment. To test his hypothesis, an assay to select agents that block the tolerance of PANC-1 cells to nutrient starvation but not ordinary culture was developed; using this assay system, two candidates, pyrvinium pamoate and kigamicin, were identified. Both of these agents showed clear antitumor activity and less toxicity in nude mice. These results indicate that modulating tumor tolerance of nutrient starvation is an effective target for cancer treatment.
|
SESSION 3: DIAGNOSIS OF PANCREATIC CANCER |
|---|
|
TOPINTRODUCTIONSESSION 1: EPIDEMIOLOGY AND...SESSION 2: PATHOLOGY AND...SESSION 3: DIAGNOSIS OF...SESSION 4: STAGING OF...SESSION 5: SURGICAL TREATMENT...SESSION 6: NON-SURGICAL...ACKNOWLEDGMENTS |
|---|
Chairpersons: Raul A. Urrutia, John P. Neoptolemos and Henry Q. Xiong
Dr Sachiko Tanaka presented the results of a study designed to determine the effectiveness of early detection of pancreatic cancer with periodic ultrasound (US) examinations. US is non-invasive, safe and convenient. It has high sensitivity for the detection of pancreatic cancer. US can also efficiently identify the dilation of the main pancreatic duct (MPD) that is an early sign of pancreatic cancer. Dr Tanaka also reviewed different strategies for improving the sensitivity of US, including fluid intake and contrast-enhanced dynamic US.
Dr Tanaka reported the results of using US for detection of pancreatic cancer in a large number of patients. A total of 1092 patients were referred for US checkup and 488 were registered for the study. Three ductal adenocarcinomas and one IPMA were detected by the periodic checkups. The common features of the three adenocarcinoma cases included pancreatic cysts, MPD dilation and hypo-echoic lesions. Dr Tanaka concluded by remarking that slight dilation of MPD and cysts were considered to presage pancreatic cancer, that special US with liquid intake was effective for the early detection of pancreatic cancer, and that contrast-enhanced dynamic US is capable of providing useful information, visualizing the fine vascular events in a small pancreatic cancer.
Dr Randall E. Brand gave a lecture on the use of US for the early detection of pancreatic cancer. Dr Brand pointed out that early detection and diagnosis was defined as the identification of an advanced precursor lesion in an asymptomatic individual or pancreatic cancer in a symptomatic individual. The ideal would be to offer pancreatectomy to patients with advanced precursor lesions (PANIN-3 or carcinoma in situ). Three US modalities are currently used for pancreatic imaging: transabdominal (TUS), endoscopic (EUS) and intraductal (IUS). EUS is believed to be the most sensitive imaging technique for detection of pancreatic masses that are smaller than 3 cm. Dr Brand reviewed the results of a surveillance study of patients with family history of pancreatic cancer. In these patients, irregular ducts, poor filling of pancreatic ducts, narrowing or dilation of ducts and formation of cystic lesions were detected by ERCP and found to be indicative of precancerous dysplastic alterations. Furthermore, every patient with an abnormal ERCP finding in this study also had abnormal EUS findings, including echogenic foci, hyperechoic nodules or an echogenic main duct. Dr Brand mentioned that, in an ongoing study of individuals with high risk of pancreatic cancer, EUS was performed, followed by fine-needle aspiration (FNA) of any mass. Secretin stimulation was used in the study to collect pancreatic juice for early detection of tumor markers. If there were abnormal findings, more invasive studies such as ERCP were performed.
Dr Masao Tanaka presented a lecture entitled ‘ERCP and cytology: the Only Method of Detecting "In Situ" Carcinoma of the Pancreas’. Dr Tanaka and co-workers have developed balloon endoscopic retrograde pancreatography with a compression study (BERP-CS) for detailed visualization of the fine branches and the MPD. To elucidate its diagnostic utility in the differentiation of benign and malignant stenosis of the MPD, the BERP-CS technique was applied to 21 patients with pancreatic cancer and 27 patients with chronic pancreatitis. They found statistically significant differences in the length of the stenosis and the ratio of the ductal diameter of the stenotic segment against that of the prestenotic segment, although the prevalence of branch findings such as stenosis, dilation and/or disappearance of ducts, cyst formation and extravasation did not differ between the two groups.
BERP-CS in combination with pancreatic juice cytological studies would contribute to more specific diagnosis of pancreatic cancer. BERP-CS is associated with a significant risk of complications, however, and is not safe for screening purposes. In patients with a high risk of pancreatic cancer, BERP-CS may offer diagnostic value.
Dr Noriyuki Moriyama focused his lecture on the evaluation of pancreatic cancer by computed tomography (CT) scanning. The advent of multi-slice helical CT has greatly enhanced the ability of CT to detect pancreatic cancer. The key CT findings in the diagnosis of pancreatic cancer are changes in the borders of the pancreas, local tumefaction of the organ, dilation of the pancreatic duct and multiple irregular areas of low attenuation within the pancreas. These findings are frequently observed in patients with unresectable pancreatic cancer. Detecting tumors smaller than 2 cm by CT has been a challenge. Helical CT has made it possible to conduct early-phase scanning of the whole pancreas with a slice thickness of 5 mm. Furthermore, multi-slice CT has made it possible to perform early-phase scanning of the whole pancreas with a slice thickness of 1–2 mm. Moreover, multi-planar reconstruction allows the visualization of a lesion in the horizontal, frontal and curved planes and improved diagnostic precision.
Dr Chusilp Charnsangavej presented his lecture titled ‘Diagnosis and Staging of Pancreatic Cancer: State-of-the-Art CT’. In pancreatic cancer, multidetector helical CT following IV contrast administration at 5 ml/s is the preferred modality. Multiphasic scanning, which includes pancreatic parenchymal phase and portal venous phase using a scanning collimation of 2.5 mm, is required to optimize lesion detection, anatomical information and staging evaluation. The multiphasic scanning technique allows the lesion to be distinguished from the normal parenchyma because the pancreatic parenchyma enhances maximally during the pancreatic parenchymal phase, whereas the tumor does not enhance as well. Lesions that enhance earlier during this phase include intraductal papillary tumor and neuroendocrine tumor. During this phase, the arterial anatomy is well defined and arterial involvement by the tumor is easily identified. The portal venous phase is useful for the detection of hepatic metastasis and venous involvement. Using this technique, the sensitivity of lesion detection was about 87–93%, the specificity was 71–88% and the accuracy of resectability prediction was about 85% at The University of Texas M.D. Anderson Cancer Center. Moreover, image data from this scanning technique can be used for 3-D image display for demonstration of vascular anatomy, the extent of vascular involvement and images of the bile duct in the imaging plane; such displays are easier for clinicians to understand and use for surgical planning.
Dr Yukio Muramatsu presented his lecture entitled ‘Roles of MRCP in the Diagnosis of Pancreatic Cancer’. Magnetic resonance cholangiopancreatography (MRCP) is one of the recent great advances in MRI. In the diagnosis of pancreatic cancer, MRCP is used to evaluate pancreatic duct morphology, to determine extra pancreatic and/or intraductal cancerous extension and to differentiate between intraductal papillary mucinous cancer and other cystic conditions. The T1-weighted images with and without contrast medium are useful to detect pancreatic cancer and its extra pancreatic extension; T2-weighted images are useful to evaluate tumor characteristics and the relationship between the tumor and the dilated duct. Cystic components and liquefied necrotic degeneration within the tumor usually can be seen as high signal intensities on T2-weighted images.
Pancreatic cystic lesions and dilatation of the pancreatic duct are not rare and are easily depicted by MRCP. Therefore, MRCP offers an excellent screening tool for differential diagnosis. For example, intraductal mucinous cancers can usually be detected by segmental dilation of the MPD and multilocular cystic lesions by diffuse dilation of the MPD. In conclusion, Dr Muramatsu speculated that more powerful MRI that uses a QD array coil is expected to be a breakthrough in detecting small pancreatic cancers.
Dr Koji Murakami lectured on clinical applications of PET for pancreatic cancer. Several researchers have reported the usefulness of [18F]FDG-PET for pancreatic cancer. In general, the aims of PET for cancer management are considered to be (1) detection of small tumors, (2) characterization of tumors, (3) staging of cancers, (4) monitoring of therapeutic effects and (5) early diagnosis of recurrences. Studies from Dr Murakami and others have shown that PET is useful in the characterization of tumors, monitoring of therapeutic effects and diagnosis of recurrences. For example, pancreatic cancer can be discriminated from pancreatitis by PET study and FDG activity correlates with clinical response to radiation. Differences in the FDG uptake ratio between low- and high-grade malignancies may possibly be an important clue to predicting prognosis.
The role of PET in the detection and staging of pancreatic cancer is less defined and is controversial. PET fails to identify lesions smaller than 1 cm and adds no additional information regarding lymph node metastasis. Dr Murakami then reviewed the other limitations of PET. He pointed out that respiratory artifacts interfere with the detection of small tumors. Dr Murakami and co-workers adopted a custom-made respiratory gating system to decrease respiratory artifacts in the upper abdomen. By using this system, the standardized uptake value increases by ~14% on average with pancreatic cancer. The results indicated that the influence of respiratory movement is greater for smaller tumors and that partial volume effect should not be ignored.
|
SESSION 4: STAGING OF PANCREATIC CANCER |
|---|
|
TOPINTRODUCTIONSESSION 1: EPIDEMIOLOGY AND...SESSION 2: PATHOLOGY AND...SESSION 3: DIAGNOSIS OF...SESSION 4: STAGING OF...SESSION 5: SURGICAL TREATMENT...SESSION 6: NON-SURGICAL...ACKNOWLEDGMENTS |
|---|
Chairperson: Chusilp Charnsangavej
Dr Jeffrey E. Lee presented his lecture on preoperative staging of pancreatic cancer. Staging workup usually includes CT scan, ERCP and EUS. Helical CT performed with contrast enhancement and a thin-section technique can accurately assess the relationship of low-density tumors to the celiac axis, superior mesenteric artery (SMA) and superior mesenteric–portal vein (SMPV) confluence. Accurate clinical staging with objective CT criteria can maximize resectability rates and avoid unnecessary surgery for those who have unresectable disease. The CT criteria used to define a potentially resectable pancreatic cancer are (1) absence of extrapancreatic disease, (2) patent SMPV confluence (assuming the technical ability to resect isolated involvement of the superior mesenteric vein or SMPV confluence) and (3) no direct tumor extension to the celiac axis or SMA. EUS provides additional information for the primary pancreatic mass and its relationship with blood vessels and lymph node metastasis. EUS-guided FNA is the preferred procedure for biopsy of pancreatic masses for confirmation of diagnosis. For the past decade, laparoscopy has been used in patients who have radiological evidence of localized pancreatic cancer to detect extrapancreatic tumors not seen on CT scans, thereby allowing laparotomy to be limited to patients who truly have localized disease. Recent investigations suggest that occult extrapancreatic disease is found in ~10% of patients with tumors deemed resectable following high-quality CT. Laparoscopy is reasonable to consider prior to laparotomy (during the same anesthesia induction) in patients for whom a decision has been made to proceed with the procedure. However, laparoscopy as a staging procedure under a separate anesthesia induction is not routinely performed at the M.D. Anderson Cancer Center.
Dr Shuji Isaji presented his lecture entitled ‘Staging Classifications of Pancreatic Cancer: Comparison of the Japanese and UICC Classifications’. Dr Isaji stated that the different classification systems for pancreatic cancer used by the JPS and UICC have hampered exchange of data between Japan and Western countries. Dr Isaji then reviewed the pros and the cons of the JPS and UICC classification systems and pointed out that although the JPS system predicts outcome much better than the UICC system, it is very difficult to apply to universal use and has poor reproducibility. In April 2002, the Committee published the fifth edition of the General Rules, using the UICC classification scheme as a reference, with the goal of making the JPS classification simple, easy to understand and acceptable by international standards while not sacrificing any of its merits. In these general rules, the T category is a minor modification of UICC, N category is a modification of JPS with much simpler grouping and the stage grouping is devised to reflect survival accurately. Unfortunately, at almost the same time as the JPS fifth edition was published, the UICC published its sixth edition, which is very different from the JPS fifth edition. Comparative study using the data from the National Pancreatic Cancer Registry reveals again that the JPS stage grouping predicts outcome much better than the UICC system. Therefore, mutual discussion is needed before preparing future editions.
|
SESSION 5: SURGICAL TREATMENT OF PANCREATIC CANCER |
|---|
|
TOPINTRODUCTIONSESSION 1: EPIDEMIOLOGY AND...SESSION 2: PATHOLOGY AND...SESSION 3: DIAGNOSIS OF...SESSION 4: STAGING OF...SESSION 5: SURGICAL TREATMENT...SESSION 6: NON-SURGICAL...ACKNOWLEDGMENTS |
|---|
Chairpersons: Philip A. Philip and Theodore S. Lawrence
Dr Tomoo Kosuge focused his discussion on surgical treatment of pancreatic cancer. Dr Kosuge briefly reviewed the history of total pancreatectomy and pancreatoduodenectomy. Early attempts at pancreatectomy were associated with inordinate risk. The postoperative morbidity rate was more than 50% and the mortality rate was about 20%. Hemorrhage, sepsis and malnutrition secondary to overt or silent anastomotic leak were the major causes of the postoperative deaths. Currently, the perioperative mortality rate in high-volume dedicated centers is <5%. Improvements in a wide range of fields have contributed to the lower mortality rate, such as refined surgical technique, better understanding of vascular anatomy, effective drainage of anastomotic sites, development of percutaneous and angiographic interventional techniques and adequate nutritional support.
The extent of surgery has been a controversy. Some advocate bypass surgery, whereas others recommend radical resection. Dr Kosuge and co-workers have aggressively managed pancreatic cancer with extended resection and adjuvant use of intraoperative radiation. The actuarial 5-year survival rate for 116 patients who underwent radical pancreatectomy between 1990 and 1998 was 20.1%, with a perioperative mortality rate of 2.6%.
Dr Kevin C. Conlon presented his experience with surgical resection of pancreatic cancer. Dr Conlon stated that complete resection of the primary disease remains the only hope of cure. Resectability rates vary from institution to institution. During the period 1983–2002, only 23% of the patients admitted to Memorial Sloan-Kettering Cancer Center with adenocarcinoma of pancreas underwent a potentially curative resection for their disease. Dr Conlon discussed important issues in the surgical management of patients with resectable adenocarcinoma of the pancreas, including the type and extent of resection, whether extended lymphadenectomy is warranted, optimal management of the pancreatic duct and the role, if any, of perioperative suppression of pancreatic exocrine secretion by drugs such as octreotide. Published data suggest that there is no difference in overall survival or perioperative morbidity and mortality rates between standard or pylorus-preserving pancreaticoduodenectomy. Recent prospective randomized trials have demonstrated that extended resection does not appear to prolong disease-specific survival.
Although the operative mortality rate in most experienced centers is <3%, considerable morbidity still exists. Failure of the pancreatic anastomosis accounts for a significant proportion of the complications (>30%) following pancreaticoduodenectomy. A number of studies have attempted to address this issue. Randomized trials have suggested that pancreaticogastric reconstruction is equivalent to pancreaticojejunal reconstruction as a safe technique with a somewhat lower anastomotic leak rate than end-to-end/invaginating pancreaticojejunostomy. The use of somatostatin analogs to decrease exocrine pancreatic secretion and thus aid healing of the pancreatic anastomosis has been suggested, but a recent randomized trial showed that this does not reduce pancreas-specific complications.
Dr Osamu Ishikawa presented his lecture on adjuvant therapy for pancreatic cancer. Dr Ishikawa reported the results of extended resection (D2) of pancreatic cancer. Surgically removing a wide range of lymphatic and connective tissue resections increased the 5-year survival rate by 25% by decreasing the incidence of local recurrence from 62 to 30%. Addition of radiation to extended pancreatectomy, however, failed to improve the survival rate because of the high incidence of liver metastasis, although the incidence of local recurrence decreased to 7%. To decrease the rate of liver metastases after extended pancreatectomy, Dr Ishikawa and co-workers developed liver perfusion chemotherapy, which consists of continuous perfusion of a low-dose of 5-fluorouracil (FU) via both the hepatic artery and portal vein (two-channel chemotherapy). This therapy improved the 5-year survival rate from 23 to 40% in the extended surgery-alone arm and the incidence of liver metastasis decreased to 4% compared with 21% in the surgery-alone arm. Although adjuvant two-channel chemotherapy improved the 5-year survival rate for patients who underwent extended pancreatectomy, the local recurrence rate for T3 cancer remains as high as 72%. Hence it is important to develop an effective adjuvant therapy strategy for such locally advanced cancers.
Dr John Neoptolemos presented his lecture on chemoradiation as adjuvant therapy for pancreatic cancer. Dr Neoptolemos stated that the first trial from the USA reported 43 patients with clear resection margins randomized to undergo either surgery or surgery followed by 40 Gy radiotherapy. The median survival duration in the radiotherapy group was 20 months compared with 11 months in the surgery-only group and the 2-year survival rates were 42 and 15%, respectively. Another trial randomized 47 patients to receive FAM chemotherapy or observation. No long-term survival benefit to the chemotherapy was shown. The EORTC trial compared chemoradiotherapy with surgery alone in 218 patients with pancreatic or ampullary cancers. The survival improvement in the pancreatic cancer group (median survival 17.1 vs 12.6 months for observation) was not statistically significant.
The ESPAC-1 trial recruited nearly 600 patients. Of these, 285 entered a 2x2 factorial design involving randomization to receive chemotherapy vs no chemotherapy and radiotherapy vs no radiotherapy; the rest were randomized to receive a single option only. The median survival duration in patients who received chemotherapy was significantly longer at 19.7 months than that of the no chemotherapy group (14 months). The median survival duration of patients randomized to receive chemoradiotherapy was 15.5 months, comparable to those in many other studies, but was not different from that of patients randomized to receive no chemoradiotherapy. The survival benefit of chemotherapy applied to patients with R1 resection margins and also those with R0 resection margins.
|
SESSION 6: NON-SURGICAL TREATMENT OF PANCREATIC CANCER |
|---|
|
TOPINTRODUCTIONSESSION 1: EPIDEMIOLOGY AND...SESSION 2: PATHOLOGY AND...SESSION 3: DIAGNOSIS OF...SESSION 4: STAGING OF...SESSION 5: SURGICAL TREATMENT...SESSION 6: NON-SURGICAL...ACKNOWLEDGMENTS |
|---|
Chairpersons: Kevin C. Conlon and Jeffrey Lee
Dr Hideki Ueno presented his lecture on chemoradiotherapy for locally advanced pancreatic cancer. In patients with locally advanced disease, prospective randomized trials have demonstrated that the combination of 5-FU with radiation therapy yielded significantly longer survival than either radiotherapy alone or chemotherapy alone. Dr Ueno and co-workers first tested protracted infusion of 5-FU with concurrent external beam radiation therapy (EBRT) to intensify the antitumor effect of chemotherapy. This therapy did not reduce the rate of distant metastasis, however, and the median survival duration (10.3 months) was identical with that achieved with conventional chemoradiotherapy using bolus 5-FU infusion. Next, the combination of cisplatin with radiation was tested. Cisplatin has a synergistic effect with radiation therapy and it also has some activity against pancreatic cancer. However, the result of this trial was disappointing with a short median survival duration of 7.7 months. Moreover, the toxicity of this regimen was considerable. Recently, results of two phase I trials of chemoradiotherapy were reported; one consisted of weekly gemcitabine with concurrent EBRT and the other involved hyperfractionated radiation therapy with protracted infusion of 5-FU. The trial determined the recommended dose of gemcitabine (250 mg/m2 weekly) and hyperfractionated radiation (64.8 Gy).
Dr Theodore S. Lawrence focused his discussion on gemcitabine as a radiation sensitizer. Standard therapy for locally advanced pancreatic cancer has been based on the combination of 5-FU and radiation. Gemcitabine is more effective than 5-FU for patients with metastatic disease and gemcitabine is a potent radiation sensitizer in the laboratory. Radiosensitization is produced under conditions in which dATP pools are depleted and the cells are simultaneously redistributed into S phase. A 1 h exposure to clinically achievable concentrations of gemcitabine produces radiosensitization that can last for days. Another potential radiation sensitizer is capecitabine. Capecitabine can be given orally and may be a more effective radiation sensitizer than 5-FU since the conversion to 5-FU in tumors is more efficient owing to induction of thymidine phosphorylase by radiation. In addition to the biological mechanism of sensitization, other factors that influence the design of clinical trials involving radiation sensitizers have included the natural history of the illness and the radiation treatment volume. For instance, given the high chance of metastatic spread, it seems logical to administer gemcitabine or capecitabine using a dose and schedule anticipated to produce both radiosensitization and systemic therapeutic effects. Furthermore, in the case of gemcitabine, clinical trial results suggest that the full benefit of gemcitabine can be realized only if conformal therapy is delivered to the tumor with minimal margins, as toxicity increases dramatically if attempts are made to treat clinically uninvolved lymph nodes.
In addition to chemotherapeutic agents, molecular targeted therapies, such as monoclonal antibodies against growth factor receptors and small molecule tyrosine kinase inhibitors, are being evaluated preclinically. Although no clinical trials in pancreatic cancer patients combining molecular targeted therapies with radiation have been conducted, this approach would be anticipated to improve the therapeutic index of treatment.
Dr Masayuki Imamura presented the findings of a multicenter randomized trial comparing surgery and radiochemotherapy for locally advanced pancreatic cancer. The objective of this trial was to compare radical resection with radiochemotherapy for locally advanced disease without distant metastasis. Patients whose cancer invaded the pancreatic capsule without any involvement of the SMA or the common hepatic artery and patients without distant metastasis were randomized to undergo radical surgery or radiochemotherapy. Twenty-two patients were randomized to undergo surgery and 20 to receive chemoradiation. The mean survival duration after surgery was significantly longer than that after chemoradiation, with a smaller hazard ratio (HR 0.46, 95% CI: 0.22–0.97). The median survival duration for patients who underwent surgery was 13 months, whereas for patients who received chemoradiation it was 8.9 months. There was no significant difference in the quality of life between the two treatment groups. The results suggest that locally advanced pancreatic cancer without distant metastasis should be treated by radical surgery.
Dr Takuji Okusaka presented his lecture on chemotherapy for advanced pancreatic cancer. Dr Okusaka outlined strategies for improving treatment for patients with advanced pancreatic cancer, including combination chemotherapy and the discovery of new agents. Dr Okusaka then reported several clinical trials conducted at the National Cancer Center Hospital. A phase I trial of methotrexate plus 5-FU enrolled 21 patients. With a partial response rate of 13% and a median survival duration of 4 months, this study clearly showed that this combination has limited activity against pancreatic cancer. A phase II trial of UFT revealed a median survival duration of 4.2 months with no response in patients with advanced pancreatic cancer. S1 showed promising activity in a pilot phase II trial and a large-scale phase II trial of S1 is ongoing in Japan. Nineteen patients have been enrolled in the study as of the date of the presentation and four patients (21%) had a partial response.
Despite worldwide agreement about the role of gemcitabine as first-line treatment in pancreatic cancer, therapies that can provide a more significant survival advantage are needed because the prognosis for patients with this disease remains very poor.
Dr Philip A. Philip presented his lecture on the treatment of metastatic and recurrent pancreatic cancer. Dr Philip stated that the goals of chemotherapy in advanced disease are to improve symptoms and quality of life, delay disease progression and prolong survival. Major research strategies in advanced pancreatic cancer have included the combination of gemcitabine with other cytotoxic agents. Several such combinations have resulted in higher response rates, longer time to progression and modest prolongation of survival. Combinations of gemcitabine plus a platinum compound (cisplatin or oxaliplatin) are of clinical interest. Available data on combination chemotherapy in advanced pancreatic cancer is based on the results of phase II clinical trials. Interpretation of such results must take into consideration study design and characteristics of patients enrolled in such trials. Inclusion of patients with localized disease and good performance status favorably affects the outcome of clinical trials. The CALGB is conducting a study to evaluate several chemotherapy regimens, including gemcitabine given as a fixed dose rate (FDR) infusion, gemcitabine plus taxotere, gemcitabine plus CPT-11 and gemcitabine plus cisplatin. ECOG is comparing gemcitabine, gemcitabine FDR and gemcitabine plus oxaliplatin.
Dr Henry Q. Xiong focused his discussion on targeted therapy for pancreatic cancer. Dr Xiong stated that epidermal growth factor receptor (EGFR) contributes to the development of pancreatic cancer. Elevated expression of EGFR is detected in more than 90% of human pancreatic cancer cases. Co-expression of EGFR and epidermal growth factor (EGF) or TGF- is associated with greater tumor size, advanced clinical stage and decreased survival duration. In addition to its effect on cell proliferation, EGFR is involved in tumor angiogenesis, a crucial step in the progression of pancreatic cancer.
The mechanism of action of EGFR-targeted therapies has been studied by using orthotopic pancreatic cancer models. IMC-225 and PKI 166 have been shown to inhibit effectively the growth of orthotopically implanted pancreatic tumors. These inhibitory effects were enhanced when IMC-225 or PKI-166 was combined with gemcitabine. A histological study of tumor specimens revealed that these EGFR-targeted therapies induced apoptosis and suppressed proliferation of tumor cells. Moreover, these therapies induced apoptosis of endothelial cells, which are not thought to be direct targets of EGFR inhibition. These data suggest that, in addition to anti-proliferative activity, antiangiogenic activity contributes significantly to the antitumor effect of EGFR inhibitors.
Most EGFR inhibitors are in the early phases of clinical development. A phase II trial of IMC-225 in combination with gemcitabine for patients with advanced pancreatic cancer showed promising results. The partial response rate was 12.5% (5/40) and the stable disease rate was 63.4%. The 1 year survival rate was 32% and the median survival duration was 7 months.
Dr Toshio Tahashi presented results of non-myeloablative allogeneic stem cell transplantation (NST) for unresectable pancreatic cancer. Graft vs tumor effect (GVT) is the hypothesis behind NST for pancreatic cancer. Five patients with unresectable pancreatic cancer received NST. All five had successful engraftment of stem cells from HLA-identical siblings. Two patients had GVT effect, showing remarkable tumor reduction and/or decreases in serum levels of carcinoembryonic antigen (CEA) and CA 19-9. One patient experienced remarkable regression of the primary pancreatic cancer following NST, but the patient died of graft vs host disease.
Dr James Abbruzzese gave the keynote lecture entitled ‘Pancreatic Cancer: a Challenge for the 21st Century’. Despite extensive efforts over the past century, conventional therapeutic approaches such as surgery, radiation, chemotherapy or combinations of these modalities have had a minimal impact on the course of this aggressive neoplasm. Tremendous progress has been made in understanding the biology of pancreatic cancer during the last decade. However, much more effort is needed to link the molecular properties of pancreatic cancer with the clinical behavior of this aggressive disease. Dr Abbruzzese outlined the areas of focus for future research efforts. First is the identification of individuals at risk of familial and sporadic pancreatic cancer. Second is the development of effective screening tools, including endoscopy and endosonography, CT and MRI and genomic and proteiomic approaches. Third is the development of risk-adapted early interventions, including surgery, chemoprevention and accurate staging for patient selection for surgery and multimodality management. Fourth is improvement in the management of advanced pancreatic cancer. Progress in molecular biology has facilitated the identification of novel molecular targets. Clinical evaluation and credentialing of these molecular targets and molecular targeted agents are crucial in the development of molecular targeted therapy.
|
ACKNOWLEDGMENTS |
|---|
|
TOPINTRODUCTIONSESSION 1: EPIDEMIOLOGY AND...SESSION 2: PATHOLOGY AND...SESSION 3: DIAGNOSIS OF...SESSION 4: STAGING OF...SESSION 5: SURGICAL TREATMENT...SESSION 6: NON-SURGICAL...ACKNOWLEDGMENTS |
|---|
The members of the organizing committee greatly appreciate the contributions of all the speakers and participants at this international symposium and their contribution to its ultimate success. The symposium was supported by the Foundation for Promotion of Cancer Research to promote the program of the Second Term Comprehensive 10 Year Strategy of Cancer Control by the Ministry of Health, Labor and Welfare, Japan.
|
FOOTNOTES |
|---|
+ For reprints and all correspondence: Tomoo Kosuge, Hepatobiliary and Pancreatic Surgery Division, National Cancer Center Hospital, 5-1-1 Tsukiji, Chuo-ku, Tokyo 104-0045, Japan. E-mail: tkosuge{at}ncc.go.jp
Received April 7, 2003; accepted April 8, 2003
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||