YEAR 2006

Screening for Early Pancreatic Neoplasia in High-Risk Individuals
Dr. Marcia Canto and colleagues from Johns Hopkins report the results of the "CAPS 2" screening for early pancreatic cancer program in the June issue of Clinical Gastroenterology and Hepatology. Dr. Canto screened 72 individuals with a strong family history of pancreatic cancer, 6 patients with the Peutz-Jeghers Syndrome, and 149 controls using a combination of endoscopic ultrasound (EUS) and computerized tomography (CT scanning). If something abnormal was identified the patients also underwent endoscopic retrograde cholangiopancreatography (ERCP). All of the patients were asymptomatic. Remarkably, 8 of the patients were found to have a tumor in their pancreas (10% yield of screening); 6 patients had 8 benign intraductal papillary mucinous neoplasms (IPMNs), 1 had an IPMN that progressed to invasive ductal adenocarcinoma, and 1 had pancreatic intraepithelial neoplasia. Endoscopic ultrasound (EUS) and computerized tomography (CT scanning) also diagnosed 3 patients with 5 extrapancreatic neoplasms. Endoscopic ultrasound (EUS) and endoscopic retrograde cholangiopancreatography abnormalities suggestive of chronic pancreatitis were more common in high-risk patients than in control subjects. From these studies, Dr. Canto concluded that screening EUS and CT diagnosed significant asymptomatic pancreatic and extrapancreatic neoplasms in high-risk individuals (people with a strong family history of pancreatic cancer and people with the Peutz-Jeghers Syndrome). Dr. Canto also concluded that abnormalities suggestive of chronic pancreatitis are identified more commonly in high-risk individuals. Dr. Canto plans to start "CAPS 3." More information about this research screening protocol will be posted on this web site as it becomes available.

DNA methylation alterations in the pancreatic juice of patients with suspected pancreatic disease
The biggest challenge of pancreatic cancer is to try to detect the cancer at an early stage when it can be surgically removed. Dr. Michael Goggins' Early Detection Laboratory is dedicated to identifying new markers for the early detection of pancreatic cancer. Just as there is mammography for breast cancer and the PSA test for prostate cancer, so too do we need a test for early pancreatic cancer. Scientists working in The Early Detection Laboratory at Johns Hopkins examined the DNA in pancreatic juice samples that were collected from pancreatic cancer patients to determine if the DNA from these samples showed an abnormal amount of methylation. Methylation is the addition of an additional carbon to specific areas of the DNA. Hypermethylation (or too much methylation) of certain genes will stop the genes from working properly. In particular, hypermethylation of genes whose normal function is to protect a cell from developing into a cancer (tumor suppressor genes) or whose function is to repair damaged DNA (mismatch repair genes) have been associated with the development of cancer. The scientists found that there was more methylation in the pancreatic juice samples collected from pancreatic cancer patients than there was in the samples collected from chronic pancreatitis patients and patients with no history of pancreatic disease. This finding is important because it suggests that the detection and quantification of methylated DNA in pancreatic juice may provide a promising approach to the diagnosis of pancreatic cancer.

YEAR 2005

Early Results Using Therapeutic Pancreatic Cancer Vaccine Show Promise
Researchers in the Sol Goldman Pancreatic Cancer Research Center in the Johns Hopkins Kimmel Cancer Center are encouraged by early results of a treatment vaccine for pancreatic cancer. At about two years into a study of 60 patients, the researchers report that 88 percent survived one year and 76 percent are alive after two years.
"Even though our results are preliminary, the survival rates are an improvement over most published results of pancreatic cancer treatment studies," says Daniel Laheru, assistant professor at the Johns Hopkins Kimmel Cancer Center. Laheru is expected to present his findings in a press briefing at a joint meeting of the American Association for Cancer Research/National Cancer Institute/European Organization for Research and Treatment of Cancer in Philadelphia on November 15.
Until recently, most studies have shown pancreatic cancer survival rates at about 63 percent one year after diagnosis and 42 percent at two years. The long-term outlook is more grim - only 15 to 20 percent of patients with local disease are alive at five years. "Since there is no universal standard for treating pancreatic cancer, it is difficult to make direct comparisons between all the studies," says Laheru.
In the current study, his team combined an immune-boosting vaccine with surgery and conventional postoperative chemotherapy and radiation. The vaccine, originally developed at Johns Hopkins, uses irradiated pancreatic cancer cells incapable of growing, but genetically altered to secrete a molecule called GM-CSF. The molecule acts as a lure to attract immune system cells to the site of the tumor vaccine where they encounter antigens on the surface of the irradiated cells. Then, these newly armed immune cells patrol the rest of the patient's body to destroy remaining circulating pancreatic cancer cells with the same antigen profile.
Patients get one vaccine injection eight to ten weeks after surgery, then four booster shots after chemotherapy and radiation. Laheru and his team completed enrolling patients in the study this past January. The average follow-up time is 32 months.
Jaffee and Laheru hope to begin multi-institutional studies in about a year. They are working with Hopkins pathologists from the Sol Goldman Pancreatic Cancer Research Center to analyze proteins from pancreatic cancer cells that may help them refine the vaccine's targets.
Proceedings, AACR-NCI-EORTC International Conference on Molecular Targets and Cancer Therapeutics, November 2005. (Abstract #2229, A Safety and Efficacy Trial of Lethally Irradiated Allogeneic Pancreatic Tumor Cells Transfected with the GM-CSF Gene in Combination with Adjuvant Chemoradiotherapy for the Treatment of Adenocarcinoma of the Pancreas).

Reaction of the Normal Pancreas to an Adjacent Pancreatic Cancer Provides Clues about Tissue Invasion and Detection.
Two papers from the Hopkins group help define the genes that are expressed (made) in the tissues adjacent to pancreatic cancer. In the first paper by Ricci and colleagues, the technique of in situ hybridization was used to determine the genes that are turned on in the body's response (called a stromal reaction) to an infiltrating cancer (analogous to the body trying to heal a wound). Importantly, Dr. Ricci found that the genes that are made in this stromal response to an invasive pancreatic cancer are related to how aggressively the tumor invades the surrounding pancreas, and not to the underlying biology of the tumor. Understanding how pancreatic cancers invade the normal pancreas and spread to other organs is a critical step to understanding how to interfere with this process. In the second paper by Fukushima and colleagues, gene expression profiling was used to better understand the gene expression patterns of pancreatic tissue adjacent to infiltrating pancreatic cancers as compared to pancreatic tissue adjacent to chronic pancreatitis. Dr. Fukushima found 20 different genes were overexpressed in pancreatic tissue adjacent to an invasive cancer compared to normal pancreatic tissue adjacent to chronic pancreatitis. These results demonstrate that some of the molecular alterations in normal pancreatic tissues that occur in response to adjacent infiltrating pancreatic ductal adenocarcinoma can provide a rich source of markers for detecting pancreatic cancer.

New Potential Target for Therapy:
Dr. Anirban Maitra and colleagues at Johns Hopkins have identified a genetic change in pancreatic cancers that has potential therapeutic implications. MTAP is a gene on chromosome 9 and novel chemotherapeutic strategies exploiting the selective loss of MTAP function in cancers have been proposed. The MTAP gene is adjacent to the p16 gene and MTAP and p16 are frequently deleted from the DNA of pancreatic cancers. Dr. Maitra has found these deletions of the MTAP and p16 genes in 30% of pancreatic cancers, suggesting that selected patients with pancreatic cancer may benefit from therapies targeting this loss. Studies are now underway in animal models to test this potential new treatment approach.

Precursor Lesions of Pancreatic Cancer
Catching the Horse before It has Fled the barn

Three articles by Johns Hopkins scientists have expanded our knowledge of the precursor lesions which are thought to develop into invasive pancreatic cancers. Understanding the biology of these precursor lesions is of critical importance if we are to detect, and potentially treat, pancreatic cancer before it spreads. In the first article, Anirban Maitra and collegues comprehensively reviewed the various subtypes of precursor lesions that are known to progress to invasive pancreatic cancer. While Pancreatic Intraepithelial Neoplasia or "PanIN" is the prototype precursor lesion associated with the "usual" cancers (ductal adenocarcinomas), other larger precursor subtypes such as intraductal papillary mucinous neoplasms (IPMNs) and mucinous cystic neoplasms (MCNs) are being increasingly recognized with better imaging and screening techniques. The pancreatic cancers that arise in the context of these latter precursor lesions, particularly IPMNs, can have a different biology and outcome than the usual PanIN-associated invasive cancers. Therefore, it is important that pathologists who examine surgically removed pancreata are familiar with the histology of the various non-invasive precursors to invasive pancreatic cancers.

The second and third articles identify potential therapeutic targets in PanINs, the most common precursor lesion of pancreatic cancer. Inactivation of function of a critical gene that controls the cell cycle – p16 (CDKN2A) – is extremely frequent in invasive pancreatic cancers. In about a third of these cases, loss of p16 function occurs via deletion of both p16 gene copies (called "homozygous deletion" in genetic terminology) and is large enough to include a neighboring gene known as MTAP in the deletions. Complete loss of MTAP function can be exploited for therapy using drugs that selectively affects MTAP-negative pancreatic cancer cells without damage to normal tissues (see March 2005 What's New). Now, Dr. Hustinx and collegues have shown that even subsets of non-invasive precursor lesions (PanINs) harbor deletions of both copies of the MTAP gene. This is the first demonstration of a homozygous deletion in a PanIN lesion, and the authors have described a simple assay that will enable determining MTAP gene status in limited tissue materials, such as biopsy specimens. Why is this important? Compounds that selectively target MTAP-negative pancreatic cancers are already in clinical trials; if successful, one can envision extrapolating these trials to treat the precursors to invasive cancer before a cancer develops. The challenge will be in identifying patients with non-invasive, MTAP-negative precursor lesions who might potentially benefit from this therapy. With advances in molecular imaging and biopsy techniques, scientists are hopeful that this strategy will be actualized in the future.

In the third article, Dr. Prasad and colleagues performed the first large scale gene expression profiling of PanIN lesions using "gene chips". Johns Hopkins scientists were one of the first to comprehensively determine the gene expression profile of invasive pancreatic cancer, and the group at Hopkins has now extended this knowledge to precursor lesions as well. In order to isolate these tiny ductal lesions from surrounding normal tissues, the scientists used a technique called "laser capture microdissection". They found 49 genes that were expressed differentially compared to normal ductal epithelium. Of note, they found many of the overexpressed genes in PanINs are normally turned on by activation of the sonic hedgehog pathway during development (See What's New, September 2003), thereby confirming that abnormal activation of this pathway plays a role in both early and advanced pancreatic cancers. Hopkins scientists had previously demonstrated that the sonic hedgehog pathway is a powerful therapeutic target in invasive pancreatic cancers, and this current work provides rationale for investigating this line of therapy in the prevention of pancreatic cancers as well.

Why is this important to me? The studies of the MTAP gene are important because compounds that selectively target MTAP-negative pancreatic cancers are already in clinical trials; if successful, one can envision extrapolating these trials to treat the precursors to invasive cancer before a cancer develops. The challenge will be in identifying patients with non-invasive, MTAP-negative precursor lesions who might potentially benefit from this therapy. With advances in molecular imaging and biopsy techniques, scientists are hopeful that this strategy will be actualized in the future.

YEAR 2004

New Technology Developed for the Early Detection of Pancreatic Cancer:
There are no early warning signs of pancreatic cancer and there are no early detection tests. As a result, most patients are not diagnosed until after the cancer has spread. Just as there is a PSA test for prostate cancer, so too do we urgently need an early detection test for pancreatic cancer. Dr. James Eshleman at Hopkins has recently developed a new technology that can detect rare DNA mutations (alterations in the DNA sequence) even when these mutations are admixed with much larger numbers of normal DNA sequences (Nat Methods. 2004 Oct 21;1(2):141-147 ). The technology, called "LigAmp" detected the most common mutation found in pancreatic cancer (KRAS2 gene mutations) even when a single mutant KRAS2 gene was admixed with 1,000 normal genes. Dr. Eshleman and colleagues have further shown that LigAmp can be used to detect DNA mutations shed from pancreatic cancers in pancreatic juice samples.

Applying new bioinformatics technology to discover new markers of pancreatic cancer (November and December 2004):
The cloning of the human genome has opened the door to global analyses of gene expression in pancreatic cancer on a scale not imaginable a few years ago. These analyses can help identify literally hundreds of genes that are made at high levels by pancreatic cancer cells. All too often however, the analyses produce potentially interesting gene sequences but the identity of the gene for which these sequences code for has not been identified. Akhalesh Pandey at Johns Hopkins has used cutting edge bioinformatics tools to identify a number of exciting genes that are overexpressed in pancreatic cancer. These genes are potential targets for new therapies and for the early detection of pancreatic cancer.

New proteomic markers of pancreatic cancer found (September 2004):
Proteomics is the study of all of the proteins in a tissue or fluid. Dr. Akhelesh Pandey carried out a comprehensive characterization of the "pancreatic juice proteome" in patients with pancreatic adenocarcinoma. A total of 170 unique proteins were identified including known pancreatic cancer tumor markers (e.g., CEA, MUC1) and proteins overexpressed in pancreatic cancers (e.g., hepatocarcinoma-intestine-pancreas/pancreatitis-associated protein (HIP/PAP) and lipocalin 2). In addition, he identified a number of proteins that have not been previously described in pancreatic juice (e.g., tumor rejection antigen (pg96) and azurocidin). The proteins identified in this study will be further assessed for their potential as biomarkers for pancreatic cancer by quantitative proteomics methods or immunoassays.

Exploiting BRCA2 gene mutations to treat pancreatic cancer (August 2004):
Nine years ago, the Kern Laboratory found mutations of a new gene, called BRCA2. Soon, this laboratory and others working in other tumor systems found that mutations of the gene were often inherited, raising the risk for pancreatic, ovarian, and breast cancer when an individual inherits one bad copy of the gene. This was the second gene found to cause inherited breast cancer, thus leading to the gene name, BRCA2. Dr. Kern's postdoctoral fellow Michiel van der Heijden followed up on this earlier discovery and showed that that pancreatic cancer cells with BRCA2 gene mutations are especially susceptible to the anticancer drugs mitomycin and cis-platin. Based on this exciting finding, it may be possible in the future to recommend individualized therapeutic regimens for patients with these mutations. More research, and possibly even a clinical trial, in this new area are underway.

Immune target for the treatment of pancreatic cancer discovered (August 2004):
Dr. Elizabeth Jaffee has developed a novel whole cell vaccine to treat patients with pancreatic cancer. She has previously found that this vaccine treatment produces an anti-tumor immune response in some patients. In the August issue of the Journal of Experimental Medicine, Dr. Jaffee reports that she has discovered a protein, mesothelin that appears to be responsible for the anti-tumor effect seen with the whole cell vaccine. Mesothelin is an antigen demonstrated previously by gene expression profiling to be up-regulated in most pancreatic cancers, and Dr. Jaffee found the consistent induction of CD8(+) T cell responses to mesothelin in patients who responded to the vaccine. This finding not only provides insight into the immune mechanisms underlying anti-tumor responses, but it is also hoped that it will lead to new "peptide" based vaccines that are cheaper and easier to administer than the whole cell vaccine.

Using endoscopic ultrasound (EUS) to screen for early pancreatic cancer in asymptomatic patients (July 2004):
In the July issue of Clinical Gastroenterology Hepatology, Dr. Marcia Canto and colleagues from Johns Hopkins report the results of screening for pancreatic cancer in asymptomatic (patients without symptoms) individuals known to be at increased risk because of their family history. Endoscopic ultrasound (EUS) was used. In this technique a small tube with an ultrasound device at the end is inserted through the patient's nose and into the patient's stomach and duodenum. The ultrasound device can then be used to image the pancreas. Thirty-eight patients were screened in Dr. Canto's study and six pancreatic masses were found. Several of these patients went to surgery and one was found to have an early cancer and another a precancerous tumor (intraductal papillary mucinous neoplasm). Therefore, 5.3% or 1 in 20 patients was found to have a clinically important pancreatic mass. This study therefore represents the first step towards demonstrating that screening for early pancreatic cancer is possible.

The risk of pancreatic cancer increases if you have more relatives with pancreatic cancer (April, 2004):
Dr. Klein and colleagues followed 838 families who participate in the National Familial Pancreatic Tumor Registry at Johns Hopkins to determine how if new pancreatic cancers are more likely to develop in families in which more than one family member has been diagnosed with pancreatic cancer. Twenty-two new pancreatic cancer cases developed in these families after the family had entered the registry. Dr. Klein found that individuals with three or more first-degree relatives with pancreatic cancer (brothers, sisters, parents or children) had a 32-fold increased risk of developing pancreatic cancer. Individuals with two first-degree relatives with pancreatic cancer had a 6.5-fold increased risk. This increase in risk was even higher among family members who smoked cigarettes. The results of this study further establish that pancreatic cancer does cluster in families. Additionally, these results help to identify individuals who may benefit from screening for the early signs of pancreatic cancer once reliable screening tests are developed.

Two potential new blood markers of pancreatic cancer (March and April 2004):
New blood markers for the early detection of pancreatic cancer are urgently needed. Just as there is a PSA test for prostate cancer, so too do we need a test for pancreatic cancer. Dr. Goggins' lab is dedicated to the discovery of these markers and in 2004 he reported the discovery of two new promising markers, osteopontin and MIC-1. These markers were identified in previous studies of gene expression using "gene chip" analysis of surgically resected cancers, and Dr. Goggins went on to show that both MIC-1 and osteopontin are released into the blood and that the blood levels of these markers are higher in patients with pancreatic cancer than they are in patients without cancer. Both markers were significantly better than the previous "gold standard" CA 19-9.

Improved understanding of intraductal papillary mucinous neoplasms (IPMNs) of the pancreas (March 2004):
Intraductal papillary mucinous neoplasms (IPMNs) of the pancreas are being diagnosed with growing frequency, but these distinctive tumors of the pancreas have not been well-characterized. Dr. Goggins and coworkers used cutting edge oligonucleotide microarrays to analyze the genes made ("the gene expression profile") by a series of IPMNs. They identified four genes that appeared to be highly associated with the presence of an invasive adenocarcinoma. Notably, the expression of at least two of the four genes was observed in 73% of 22 invasive IPMNs but in none of 16 noninvasive IPMNs (P < 0.0001). These findings suggest that preoperative assessment of gene expression profiles may be able to differentiate invasive from noninvasive IPMNs. They have also performed a similar analyses on another type of tumor in the pancreas called "mucinous cystic neoplasm."

Series of advances made by Dr. Iacobuzio-Donahue's lab (Spring 2004):
Dr. Iacobuzio-Donahue recently published two exciting articles in Cancer Research. The first article, published in the December 15 issue (Cancer Research 2003; 63:8614-22) describes the largest, most comprehensive study to date of gene expression (the genes made) in pancreatic cancer. Close to 100 samples were analyzed using the most current gene chip (Affemetrix U133). Dr. Iacobuzio-Donahue and colleagues discovered 142 potential new markers of pancreatic cancer. Already, Dr. Koopmann has shown that one of these markers, called "osteopontin," is elevated in the blood of patients with pancreatic cancer (Cancer Epi Biomarkers Prev, 2004; 13:487-91) and Dr. Nichols has demonstrated that another of these markers, claudin 4, may be a useful therapeutic target (Am J Clin Pathol. 2004; 121:226-30). In the second paper published by Dr. Iacobuzio-Donahue (Cancer Research 2004; 64:871-875) she describes the largest genetic analysis conducted to date of the DNA changes in pancreatic cancer. Dr. Iacobuzio-Donahue and colleagues conducted a large-scale "allelotypes" (an analysis of DNA losses in a cancer) on a series of pancreatic cancers and they discovered a number of hot spots of DNA alterations in pancreatic cancer. These hot spots will help other scientists identify the genes that are targeted for inactivation in pancreatic cancer. An understanding of these genes, in turn, may lead to a better understanding of why pancreatic cancer develops and how to treat it.

YEAR 2003

Mutations in the BRAF gene found in pancreatic cancer (October 2003):
It has become clear that pancreatic cancer is a disease caused by damage to the DNA (called mutations). The identification of which genes are mutated in pancreatic cancer provides insight into the fundamental nature of the disease. In the October issue of the American Journal of Pathology, Dr. Kern and colleagues report mutations in the BRAF and in the FBXW7 genes. It is hoped that a better understanding of the effects of these mutations will provide insight into why pancreatic cancer is so aggressive.

Determining the stage at which selected genes are made in tumors of the pancreas (September 2003):
Just as colon polyps give rise to invasive colon cancer, so too has it become clear that small non-invasive lesions in the pancreas (called PanINs) can give rise to invasive pancreatic cancer. These small non-invasive lesions in the pancreas are exciting, because they represent a curable stage of pancreatic cancer. Dr. Anirban Maitra at Johns Hopkins has studied a large series of "PanINs" and he has defined the stages at which different genes are made by the tumor cells. This finding is exciting because it identifies which genes are the best targets for the development of early detection tests, and which genes are the best targets for the treatment or prevention of early tumors of the pancreas.

Improving the interpretation of small biopsies of the pancreas (September 2003):
It can be very difficult to biopsy the pancreas, and when the pancreas is biopsied often only a few cells are obtained. As a result, the diagnostic interpretation of pancreatic biopsies can be very difficult. Scientists at Johns Hopkins have identified a group of genes that are made at high levels by pancreatic cancer cells. They then showed that staining small biopsies for two of these markers, PSCA and mesothelin, can greatly improved the accuracy of diagnosis and therefore improve patient care.

New cellular pathway found to be active in pancreatic cancer (September 2003):
Scientists at Johns Hopkins have discovered a cellular pathway that is activated in pancreatic cancer, a finding that provides a potential new way to treat pancreatic cancer. These findings are reported in the journal Nature (the September 14 advanced online publication can be accessed at www.nature.com/nature). This work was performed in the laboratory of Dr. Philip Beachy, who is a Member of the National Academy of Sciences, and a renowned developmental biologist at Johns Hopkins. Dr. Beachy and colleagues demonstrated that the "Hedgehog" pathway is abnormally turned on in many digestive tract tumors, including those of the pancreas. In addition, they have demonstrated that the administration of a drug known as cyclopamine that specifically blocks this pathway in cancer cells results in dramatic reduction of cancer growth. In a mouse model they showed complete and sustained reduction of the tumor following only two weeks of therapy with cyclopamine; most importantly, the mice did not appear to suffer any side effects from the therapy. Although application to humans is years away, the results of this study have potential implications for future treatment options in pancreatic cancers, and demonstrate how knowledge of underlying molecular abnormalities in tumors can lead to new therapies.

New Targets for Aberrant Methylation in Pancreas Cancer (July 2003):
Research by Norihiro Sato M.D., Ph.D. in the laboratory of Dr. Michael Goggins has led to the discovery of multiple genes that undergo silencing by DNA methylation (the addition of a carbon group to DNA) in pancreatic cancer. Knowledge of these genes provides us with a better understanding of the role of DNA methylation in pancreatic cancer development. The doctors also showed that these abnormally methylated genes can be detected in pancreatic juice from patients with pancreatic cancer and raising hopes that their detection could aid in the early diagnosis of pancreatic cancer.

Pancreatic Cancer Linked to Errant Reactivation of Embryo Cell Pathway (June 2003):
Research by Johns Hopkins Kimmel Cancer Center specialists has uncovered a novel pathway in the origin of pancreatic cancers. The "Notch" pathway, normally turned off in adults, can be turned on after injury to the pancreas.

Fanconi Gene Abnormalities in Pancreatic Cancer (May 2003):
Dr. Michiel van der Heijden and colleagues studied two of the Fanconi genes: FANCC and FANCG. Inherited and new mutations were found in a number of pancreatic cancers. Some of these mutations are inherited, meaning that these individuals had an increased risk of developing pancreatic cancer because they were born with these mutations in their DNA (they had inherited them from one of their parents. Dr. van der Heijden and colleagues had another interesting finding: three of the nine persons whose pancreas cancer had young onset (less than 50 years of age) had such mutations.

There are no easy tests for the kinds of Fanconi gene mutations now being studied by the researchers, but such tests may become available in the future. Such testing is likely to be of clinical importance. Cells that are defective in the Fanconi genes are known from other research to be highly sensitive to certain chemicals. If may be possible in the future to recommend a different therapeutic regimen for patients with these mutations. More research in this exciting new area is needed.

Activin Abnormalities in Pancreatic Cancer (March 2003):
Activin receptors play an important "controlling" roll in normal pancreatic cells. Dr. Byungwoo Ryu and colleagues in the Kern Laboratory therefore studied the genes that respond to activin signals to uncover the ways in which the cells are regulated by activin. Using a high-density gene expression screen, they studied gene expression changes characteristic of activin. Some of the genes regulated include genes that directly control cell division. This work was published in the journal Cancer Biology & Therapy, (Volume 2, pages 164-70), and represents the largest study of gene responses to activin published to date.

Also working in Dr. Kern's lab, Dr. Paula Hempen and colleagues found mutations of another form of the activin receptor in pancreatic cancer cells, extending the numbers of tumors known to have abnormalities in the activin system. The newly discovered mutations are in the ACVR2 gene, the activin type 2 receptor. The ACVR2 gene mutations were found in nearly all gastrointestinal tumors that had defects in a DNA-repair pathway involved in familial forms of cancer, including families at high risk of colorectal, pancreatic, and endometrial cancers.

Anti-Cancer Drug Can Lead to Cancer Invasion (February 2003):
Dr. Sato and colleagues reported an unexpected adverse effect of an anti-cancer drug that is used to remove "methyl" groups (carbon-hydrogen) from DNA. The authors found that this drug can inadvertently switch on certain genes (matrix metalloproteinases) and that the switching on of these genes can promote cancer invasion. This study shows that each therapy needs to be carefully evaluated, because therapies can have unanticipated side effects.

YEAR 2002

Target of chemopreventive agents (August 2002):
One of the more exciting approaches to saving lives is preventing pancreatic cancer through the use of "chemopreventive" agents. Chemopreventive agents are drugs that are taken regularly and reduce one's risk of cancer. Cox-2 (also known as cyclooxygenase 2) is an enzyme made by cells that is a target for some chemopreventive drugs. In the August issue of the American Journal of Clinical Pathology, Dr. Maitra and colleagues from Johns Hopkins report that cyclooxygenase 2 is expressed (made by) small lesions in the pancreas called "pancreatic intraepithelial neoplasia." This is important because pancreatic intraepithelial neoplasia is believed to be the precursor to pancreatic cancer. The expression of cyclooxygenase 2 in pancreatic intraepithelial neoplasia lesions suggests that cyclooxygenase 2 inhibitors could be used to prevent the development of pancreatic cancer. Further work is needed, but clearly chemopreventive agents, such as cyclooxygenase 2 inhibitors, offer an exciting new approach to fight pancreatic cancer before it develops.

BRCA2 gene is important in familial pancreatic cancer (July 2002):
Investigators from Johns Hopkins found that 1 in 6 (17%) patients with a very strong family history of pancreatic cancer have inherited ("germline") mutations (changes) in the BRCA2 gene. The BRCA2 is also known as the second breast cancer gene. Although it has been known for years that BRCA2 was important in familial breast cancer, the importance of BRCA2 in familial pancreatic cancer is only now coming to light. Inherited mutations in BRCA2 are particularly common in the Ashkenazi Jewish population. The identification of a gene responsible for the familial aggregation of pancreatic cancer is important because at-risk family members can now be tested to see if they carry this gene.

Genetic alterations in solid-pseudopapillary tumors of the pancreas (April 2002):
Solid-pseudopapillary tumors are rare tumors of the pancreas that arise primarily in young women. They have a distinct appearance under the microscope and a much better prognosis than the more common ductal adenocarcinomas of the pancreas. Scientists at Johns Hopkins studied the fundamental genetic (DNA) changes in a series of solid-pseudopapillary tumors of the pancreas and found that they almost all have mutations (DNA changes) in a particular gene called Beta-catenin. By contrast, usual ductal adenocarcinomas of the pancreas almost never have Beta-catenin mutations.

It can sometimes be difficult in a small biopsy or even in a resected specimen to distinguish between the various types of tumors that arise in the pancreas. This discovery suggests that the presence or absence of Beta-catenin mutations in a tumor can be used to distinguish between solid-pseudopapillary tumors (which have a very good prognosis) and usual ductal adenocarcinomas (which do not have a good prognosis).

Genetic alterations in acinar cell carcinomas of the pancreas (March 2002):
Acinar cell carcinomas are rare malignant tumors of the pancreas. They are microscopically different from the more common "ductal adenocarcinomas" of the pancreas. The fundamental DNA (genetic) changes that underlie the development of acinar cell carcinomas have not yet been elucidated. Scientists at Johns Hopkins studied 21 acinar cell carcinomas of the pancreas and found a distinct pattern of DNA changes in these tumors. These DNA changes included loss of chromosome arm 11p and changes in the "APC/beta-catenin pathway." These results indicate that acinar cell carcinomas are genetically distinct from pancreatic ductal adenocarcinomas, but some cases contain genetic alterations common to another rare tumor type in the pancreas called "pancreatoblastomas." An understanding of the variants of pancreatic cancer helps us understand why some tumors occur in different patients. In addition, it should form a basis for targeting specific therapies to specific tumor types.

Discovery of a panel of genes made by pancreatic cancers (February and March 2002):
Scientists at Johns Hopkins used two different techniques―"SAGE" and gene chips―to discover a panel of over 100 genes that are made at high levels in pancreatic cancer but not in normal tissues. They used the recently developed technology "SAGE" (serial analysis of gene expression) technology as well as cutting-edge gene chips. These technologies were applied to a panel of pancreatic cancers and normal pancreas cells and the investigators discovered a novel panel of over genes that appear to be made ("expressed") at high levels in pancreatic cancer. Several genes that may be involved in the fundamental nature of malignant changes in pancreatic ductal epithelium were identified. Some genes, such as S100A4, prostate stem cell antigen, and carcinoembryonic antigen-related cell adhesion molecule 6, suggest potential use as diagnostic markers. Others suggest potential novel therapeutic targets. These two studies provide insight into the fundamental nature of pancreatic cancer and each of the more than 100 genes discovered to be made at high levels in pancreatic cancer may serve as a new marker for the early detection of pancreatic cancer, or as a target for the development of new chemotherapies.

Target of chemopreventive agents made in pancreatic cancer precursors:
One of the more exciting approaches to saving lives that would otherwise be lost to pancreatic cancer is preventing pancreatic cancer through the use of "chemopreventive" agents. Chemopreventive agents are drugs that are taken regularly and reduce one's risk of cancer. For example, a recent study suggested that regular aspirin use may reduce the risk of developing pancreatic cancer. Cox-2 (also known as cyclooxygenase 2) is an enzyme made by cells that is a target for some chemopreventive drugs. In the August issue of the American Journal of Clinical Pathology Dr. Maitra and colleagues from Johns Hopkins report that cyclooxygenase 2 is expressed (made by) small lesions in the pancreas called "pancreatic intraepithelial neoplasia".

Why is this important?
Pancreatic intraepithelial neoplasia is believed to be the precursor to pancreatic cancer. The expression of cyclooxygenase 2 in pancreatic intraepithelial neoplasia lesions suggests that cyclooxygenase 2 inhibitors could be used to prevent the development of pancreatic cancer. Further work is needed, but clearly chemopreventive agents, such as cyclooxygenase 2 inhibitors, offer an exciting new approach to fight pancrewatic cancer before it develops.

BRCA2 gene important in familial pancreatic cancer In the July 2002 issue of Cancer research K. Murphy and colleagues from Johns Hopkins reported that 1 in 6 (17%) patients with a very strong family history of pancreatic cancer have inherited ("germline") mutations (changes) in the BRCA2 gene. The BRCA2 is also known as the second breast cancer gene. Although it has been known for years that BRCA2 was important in familial breast cancer, the importance of BRCA2 in familial pancreatic cancer is only now coming to light. Inherited mutations in BRCA2 are particularly common in the Askenazi Jewish population.

Why is this important?
The identification of a gene responsible for the familial aggregation of pancreatic cancer is important because at-risk family members can now be tested to see if they carry this gene. If you are interested in learning more about genetic testing visit: http://www.pbs.org/gene/findout/3_findout.html. To find a genetic counselor near you visit: http://www.nsgc.org/GeneticCounselingYou.asp. To join the National Familial Pancreas Tumor Registry at Johns Hopkins, contact Alison Klein: aklein1@jhmi.edu.

Genetic alterations in solid-pseudopapillary tumors of the pancreas:
Solid-pseudopapillary tumors are rare tumors of the pancreas that arise primarily in young women. They have a distinct appearance under the microscope and a much much better prognosis than the more common ductal adenocarcinomas of the pancreas. Dr. Susan Abraham and colleagues studied the fundamental genetic (DNA) changes in a series of solid-pseudopapillary tumors of the pancreas and found that they almost all have mutations (DNA changes) in a particular gene called Beta-catenin. By contrast, usual ductal adenocarcinomas of the pancreas almost never have Beta-catenin mutations.

Why is this important?
It can sometimes be difficult in a small biopsy or even in a resected specimen to distinguish between the various types of tumors that arise in the pancreas. This discovery by Dr. Abraham suggests that the presence or absence of Beta-catenin mutations in a tumor can be used to distinguish between solid-pseudopapillary tumors (which have a very good prognosis) and usual ductal adenocarcinomas (which do not have a good prognosis)

Genetic alterations in acinar cell carcinomas of the pancreas
Acinar cell carcinomas are rare malignant tumors of the pancreas. They are microscopically different from the more common "ductal adenocarcinomas" of the pancreas. The fundamental DNA (genetic) changes that underlie the development of acinar cell carcinomas have not yet been elucidated. Scientists at Johns Hopkins studied 21 acinar cell carcinomas of the pancreas and found a distinct pattern of DNA changes in these tumors. These DNA changes included loss of chromosome arm 11p and changes in the "APC/beta-catenin pathway". These results indicate that acinar cell carcinomas are genetically distinct from pancreatic ductal adenocarcinomas, but some cases contain genetic alterations common to another rare tumor type in the pancreas called "pancreatoblastomas".

Why is this important?
An understanding of the variants of pancreatic cancer helps us understand why some tumors occur in different patients. In addition, it should form a basis for targeting specific therapies to specific tumor types.

Discovery of a panel of genes made by pancreatic cancers
Scientists at Johns Hopkins used two different techniques- "SAGE" and gene chips to discover a panel of over 100 genes that are made at high levels in pancreatic cancer but not in normal tissues. Dr. Ryu and colleagues used the recently developed technology "SAGE" (serial analysis of gene expression) technology while Dr. Christine Iacobuzio-Donahue and colleagues used gene chips. These technologies were applied to a panel of pancreatic cancers and normal pancreas cells and Drs. Ryu and Iacobuzio-Donahue discovered a novel panel of over genes that appear to be made ("expressed") at high levels in pancreatic cancer. Several genes that may be involved in the fundamental nature of malignant changes in pancreatic ductal epithelium were identified. Some genes, such as S100A4, prostate stem cell antigen, carcinoembryonic antigen-related cell adhesion molecule 6, and mesothelin, suggest potential use as diagnostic markers. Others suggest potential novel therapeutic targets.

Why is this important?
These two studies provide insight into the fundamental nature of pancreatic cancer and each of the more than 100 genes discovered to be made at high levels in pancreatic cancer may serve as a new marker for the early detection of pancreatic cancer, or as a target for the development of new chemotherapies.

Extent of surgery for pancreatic cancer:
The extent of surgery appropriate for patients with pancreatic cancer has long been debated. Some surgeons perform a "standard" whipple resection (also called a standard pancreaticoduodenectomy) while others have suggested that more extensive ("radical") surgery is needed. Dr. C. Yeo and colleagues from Johns Hopkins therefore performed a randomized study of 294 patients surgically treated at Johns Hopkins. Patients were randomized to either a "standard" whipple or a more "radical" whipple procedure. In the September 2002 issue of the Annals of Surgery, Dr. Yeo reports that the radical (extended) whipple can be performed with similar mortality (death rate) but some increased morbidity (complications) compared to standard whipple. No long-term benefit was found in the more radical surgery.

Why is this important?
This study suggests that although more radical surgery can be performed safely, it may not provide any long-term benefits.

Ann Surg 2002 Sep;236(3):355-68

YEAR 2001

New marker for pancreatic cancer (December 2001):
Researchers at Johns Hopkins identified a new marker for pancreatic cancer. This marker was discovered using "SAGE" a technology developed at Johns Hopkins to help scientists determine which genes are expressed (made) by a cancer. Dr. Argani found that almost all pancreatic cancers express the gene called "mesothelin" at levels much higher than those found in normal, non-cancerous, tissues. The discovery that mesothelin is made at high levels in pancreatic cancer has potential diagnostic, imaging, and therapeutic implications. For example, scientists in Dr. Liz Jaffee's lab are already conducting studies in the laboratory to see if mesothelin can be used as an immune target to treat patients with pancreatic cancer.

New prognostic marker for pancreatic cancer (December 2001:
It can be very hard to predict the prognosis for patients with pancreatic cancer. Dr. M. Tascilar and colleagues at Johns Hopkins studied 249 patients with pancreatic cancer who underwent a Whipple resection for pancreatic cancer. Patients with pancreatic cancers that expressed (made) the SMAD4 protein had significantly longer survival (19.2 months) than did patients whose cancers did not express (make) the SMAD4 protein (14.7 months). This SMAD4 survival benefit persisted after adjustment for known prognostic factors including tumor size, margins, lymph node status, pathological stage, blood loss, and use of adjuvant chemoradiotherapy. From this, Dr. Tascilar was able to conclude that patients undergoing Whipple resection for pancreatic cancer survive longer if their cancers express SMAD4.

This study helps confirm the importance of the SMAD4 gene in pancreatic cancer.

Unraveling the genetic changes in pancreatoblastomas (August 2001):
While scientists have made great strides in advancing our understanding of pancreatic ductal adenocarcinoma, little is known about rare tumors that arise in the pancreas. Pancreatoblastoma is a rare pancreatic tumor with a distinctive microscopic appearance that generally affects infants and young children. Researchers at Johns Hopkins analyzed a series of nine pancreatoblastomas for genetic alterations (changes in the DNA sequence of the tumors). They found three interesting things. First, pancreatoblastomas are genetically very different from the more common ductal adenocarcinomas of the pancreas. Pancreatoblastomas show alterations (mutations) in the beta-catenin/APC genes. Second, they also showed that chromosome 11p is frequently altered in pancreatoblastomas. Chromosome 11p is frequently altered in hepatoblastomas (a rare pediatric tumor in the liver), suggesting that pancreatoblastomas are more closely related to hepatoblastomas than they are to pancreatic ductal adenocarcinomas. Finally, one of the patients included in the series of pancreatoblastomas had the clinical syndrome called "familial adenomatous polyposis" or "FAP." Patients with familial adenomatous polyposis develop numerous polyps in their colon at an early age and this study demonstrates that they can also develop pancreatoblastomas. An understanding of the variants of pancreatic cancer helps us understand why some tumors occur in adults and some in children. In addition, it should form a basis for targeting specific therapies to specific tumor types.

Markers of cancer invasion (July 2001):
There is a great effort underway to identify new ways to detect cancers early. A major approach is to identify "tumor-specific" and "tissue-specific" markers. For example, useful markers can be substances found to be produced by cancer cells in both tissue culture (cancer cells grown artificially outside the body) and in patient samples. Such markers would not be normally found in normal tissues at a high level. These are the invasion-specific markers. Dr. Ryu and colleagues in Dr. Kern's laboratory for pancreatic cancer research at Johns Hopkins searched for such genes. Dozens of invasion-specific markers were identified in invasive cancers obtained from patient samples. Many of these were new markers not previously considered as cancer markers and many of the genes are expressed not by the tumor cells but instead by the patient's response to the tumors. Some of these markers are known to be secreted and to be detectable in simple blood samples. A strong effort is underway to examine these candidates and develop markers for use in the early detection of cancer, to aid medical imaging and to serve as targets for the development of invasion-specific anticancer therapy.

Activin receptors-A new anticancer signal in human tumors (June 2001):
The major problem with human tumors is that they do not obey the signals from their surrounding cells that should restrain their growth. To date, very few of such signals have been defined, and this limits our ability to understand and counter this basic abnormality. Because we need to understand these signals, there has been a great effort to identify genes that are mutated and turned off in tumors. These are the "tumor-suppressor genes." The inactivation of these genes allows tumors to escape from the normal growth controls that the surrounding cells and tissue are trying to place on them. Activin is a protein secreted by normal cells. To exert its action, it must bind receptors on a cell.

The receptors propagate a signal to the cell but it was not previously known that these signals were able to suppress tumor growth. Mutations within the activin receptor gene were found in some pancreatic and biliary cancers by Dr. Gloria Su and colleagues in Dr. Kern's laboratory for pancreatic cancer research at Johns Hopkins. In tumors that lack the mutations, someday it might be possible to administer activin as a therapeutic strategy. It might also be possible to mimic the effects of activin on tumor cells by a precise molecular targeting using specially designed new drugs that directly activate the signal pathway without the need for intact receptors. This is a new idea that was previously unknown, but now can be explored. It is our hope that one can design a therapy to attack the most vulnerable components of pancreatic cancer.

Hopkins' scientists use molecular tool to discover new markers of hepatopancreatico-biliary cancer (June 2001):
Dr. Argani and colleagues from Johns Hopkins discover a new marker of pancreatic cancer. This new marker called "prostate stem-cell antigen" (PSCA) was discovered by using a technique developed at Johns Hopkins called "serial analysis of gene expression" (SAGE). Since the original description of SAGE, a group of cooperating scientists from a number of institutions have created an online database of gene expression that includes SAGE data on a variety of tissues and cancers (http://www.ncbi.nlm.nih.gov/SAGE/ ). The investigators at Johns Hopkins used this database to compare the gene expression levels in pancreatic cancer tissues with those seen in non-cancerous pancreatic tissues. The goal was to identify genes selectively "turned-on" in the cancers. One of the genes found using this approach is called "prostate stem-cell antigen." Prostate stem-cell antigen is a gene originally thought to be largely restricted to prostate cells. Dr. Argani and colleagues demonstrate that prostate stem-cell antigen (PSCA) is, in fact, highly overexpressed in approximately 60% of primary pancreatic cancers. It is not expressed in the normal pancreas. These findings are exciting for several reasons. First, they demonstrate the power of new technologies such as SAGE to discover new tumor markers. Second, PSCA, because it is selectively overexpressed in pancreatic cancers, might be a useful marker for pancreatic cancer. Third, other groups have shown that PSCA can be an immune target and therefore PSCA is being explored as a target for the immune treatment of cancers. The demonstration of PSCA expression in pancreatic cancer suggests a new avenue for treating pancreatic cancers, immunotherapy directed at cells expressing PSCA.

Molecular and immunohistochemical analysis of small cell carcinoma of the gallbladder: An unusual entity (May 2001):
Dr. Argani and colleagues published the largest series of cases examining small cell carcinomas of the gallbladder, a highly unusual neoplasm that has been described only recently. Dr. Maitra characterized the clinical, histopathologic, immunohistochemical and molecular features of 12 small cell carcinomas of the gallbladder. It was discovered that only half of these rare cancers are "pure" and half are combined with other neoplasms (e.g., adenocarcinoma, squamous carcinoma, and rarely, carcinosarcoma). These cancers were studied using molecular and immunohistochemical techniques. We found that the molecular changes in small cell carcinomas were similar to those of adenocarcinomas occurring at this site, with a high frequency of p53 and p16INK4a abnormalities, and a low frequency of deleted in pancreatic carcinoma-4 (Dpc-4) inactivation and K-ras codon 12 mutations. In contrast to small cell carcinomas of the lung, p16INK4a function appears to be abrogated more frequently in these carcinomas. We hope these results will help us develop a rational approach to the diagnosis and therapy of these unusual tumors.

Molecular analysis of bile duct carcinomas (April 2001):
Using immunohistochemical labeling, Dr. Argani and colleagues were able to show that the DPC4, a tumor suppressor gene discovered at Johns Hopkins that is known to play a major role in pancreatic cancer, is also targeted in bile duct carcinomas. Loss of Dpc4 protein was identified in a significant percentage of bile duct carcinomas. However, not all bile duct carcinomas were equal: we were able to demonstrate that distal common bile duct carcinomas (those located near the pancreas) were far more likely to demonstrate loss of DPC4 than proximal bile duct cancers (Klatskin tumors and cholangiocarcinomas of the liver). In fact, the frequency of DPC4 loss that we demonstrated in distal bile duct carcinomas (55%) is identical to that which was demonstrated in pancreatic cancer. Similarly, we were able to show that the p53 gene product was abnormally expressed far more frequently in distal bile duct cancers than proximal ones. These results show that distal common bile duct cancers have some of the same genetic alterations as pancreatic cancers, while other bile duct cancers are biologically distinct. We hope that these results will allow us to develop more rational therapies for these tumors.

Familial pancreatic cancer (March 2001):
For years, isolated reports in the medical literature have suggested that pancreatic cancer runs in some families. For example, it has been reported that former President Jimmy Carter lost his father, two sisters and brother to pancreatic cancer. A. Tersmette and colleagues from Johns Hopkins report that first-degree relatives (brothers and sisters, parents and children) of patients with "familial pancreatic cancer" have a significantly increased risk of developing pancreatic cancer. Tersmette and colleagues followed 341 families enrolled in the National Familial Pancreas Tumor Registry (NFPTR) and found that the first-degree relatives of familial pancreatic patients had an 18-fold increased risk of developing pancreatic cancer when compared to the general population (the "SEER" database). In this study, familial pancreatic cancer was defined as at least a pair of first-degree relatives with pancreatic cancer in a family. Remarkably, if there were three or more family members with pancreatic cancer when the family enrolled in the NFPTR, then the risk of other family members developing pancreatic cancer jumped to 57-fold greater than the general population. This study firmly establishes that "Familial Pancreatic Cancer" is a real entity and it provides a quantitative measure of the risk of pancreatic cancer in these families. Studies such as this will form the basis for identifying individuals at-risk for developing pancreatic cancers who might benefit from new screening tests as they are developed.

New vaccine to treat pancreatic cancer (January 2001):
Dr. Elizabeth Jaffee and colleagues at Johns Hopkins report the result of a phase I clinical trial of a novel vaccine treatment for patients with a pancreatic cancer. The vaccine was produced by genetically altering pancreatic cancer cells growing in culture so that the cells would produce large quantities of an immune activating factor called "Granulocytic-macrophage colony-stimulating factor" (or GM-CSF for short). Dr. Jaffee treated 14 patients with this vaccine in a phase I dose escalation trial. The patients underwent surgery at Johns Hopkins after which they received various doses of the vaccine. No dose-limiting toxicities were encountered. Instead, Dr. Jaffee was able to demonstrate that the vaccine induced an anti-tumor immune response in three patients who received the highest dose of the vaccine (>10x10⁷ vaccine cells). Remarkably, these three patients remained alive and free of disease more than 36 months after diagnosis. Based on these results, Dr. Jaffee and her team will conduct phase II trials of the GM-CSF vaccine. These trials are scheduled to begin in the late summer.

Consistent overexpression of Fatty Acid Synthase (FAS) in biliary tract carcinomas: A novel target for anti-biliary tract cancer drug development (January 2001):
Fatty Acid Synthase (FAS) is the primary enzyme involved in the breakdown of fats. FAS has been demonstrated to be overexpressed in several human cancers (breast, endometrial, prostate, colon). In some cancers, high levels of FAS expression have been associated with poor prognosis, suggesting that FAS expression may promote tumor growth and virulence. Recently synthesized inhibitors of FAS have demonstrated antitumor activity without concurrent toxicity to normal tissues, and hence hold promise as therapy for tumors that overexpress FAS (Cancer Res 2000; 60: 213-218) (PNAS 2000; 97: 3450-3454). FAS expression had not been studied in biliary tract carcinomas, which are highly aggressive and often do not respond to conventional therapy. Dr. Argani and colleagues recently examined 107 biliary tract carcinomas for FAS overexpression using an immunohistochemical assay on formalin-fixed, paraffin-embedded tissue. FAS was overexpressed in 93% of carcinomas of the biliary tract. Therefore, FAS inhibitors hold promise as a new therapy for biliary tract carcinomas.

New Marker for pancreatic cancer
In the December 2001 issue of Clinical Cancer Research Dr. Argani and colleagues from Johns Hopkins reported the identification of a new marker for pancreatic cancer. This marker was discovered using "SAGE" a technology developed at Johns Hopkins to help scientists determine which genes are expressed (made) by a cancer. Dr. Argani found that almost all pancreatic cancers express the gene called "mesothelin" at levels much higher than those found in normal, non-cancerous, tissues.

Why is this important?
The discovery that mesothelin is made at high levels pancreatic cancer has potential diagnostic, imaging, and therapeutic implications. For example, scientists in Dr. Liz Jaffee's lab are already conducting studies in the laboratory to see if mesothelin can be used as an immune target to treat patients with pancreatic cancer.

New prognostic marker for pancreatic cancer
It can be very hard to predict the prognosis for patients with pancreatic cancer. Dr. M. Tascilar and colleagues at Johns Hopkins studied 249 patients with pancreatic cancer who underwent a Whipple resection for pancreatic cancer. Patients with pancreatic cancers that expressed (made) the SMAD4 protein had significantly longer survival (19.2 months) than did patients whose cancers did not express (make) the SMAD4 protein (14.7 months). This SMAD4 survival benefit persisted after adjustment for known prognostic factors including tumor size, margins, lymph node status, pathological stage, blood loss, and use of adjuvant chemoradiotherapy. From this, Dr. Tascilar was able to conclude that patients undergoing Whipple resection for pancreatic cancer survive longer if their cancers express SMAD4.

Why is this important?
This study helps confirm the importance of the SMAD4 gene in pancreatic cancer. While SMAD4 may be useful as a marker for prognosis, one should always keep in mind, as stated by Stephen Jay Gould, that "the median isn't the message".(www.cancerguide.org)

Unraveling the genetic changes in pancreatoblastomas
While scientists have made great strides in advancing our understanding of pancreatic ductal adenocarcinoma, little is known about rarer tumors that arise in the pancreas. Pancreatoblastoma is a rare pancreatic tumor with a distinctive microscopic appearance that generally affects infants and young children (see the FAQ section of this Web site for more information on pancreatoblastomas and other rarer variants of pancreas cancer- http://pathology2.jhu.edu/pancreas/typtable.cfm). Dr. Abraham analyzed a series of nine pancreatoblastomas for genetic alterations (changes in the DNA sequence of the tumors). She found three interesting things. First, pancreatoblastomas are genetically very different from the more common ductal adenocarcinomas of the pancreas. Pancreatoblastomas show alterations (mutations) in the beta-catenin/APC genes. 2) Dr. Abraham also showed that chromosome 11p is frequently altered in pancreatoblastomas. Chromosome 11p is frequently altered in hepatoblastomas (a rare pediatric tumor in the liver), suggesting that pancreatoblastomas are more closely related to hepatoblastomas than they are to pancreatic ductal adenocarcinomas. 3) Finally, one of the patients included in Dr. Abraham's series of pancreatoblastomas had the clinical syndrome called "familial adenomatous polyposis" or "FAP". Patients with familial adenomatous polyposis develop numerous polyps in their colon at an early age and Dr. Abraham demonstrates that they can also develop pancreatoblastomas.

Why is this important?
An understanding of the variants of pancreatic cancer helps us understand why some tumors occur in adults and some in children. In addition, it should form a basis for targeting specific therapies to specific tumor types.

Markers of Cancer Invasion
In an effort to identify new ways to identify cancers that otherwise would remain undetected for too long, Dr. Ryu and colleagues in Dr. Kern's laboratory for pancreatic cancer research at Johns Hopkins searched for genes ("markers") produced by the invasive tumor or the body's reaction to it. Dozens of invasion-specific markers were identified in invasive pancreatic cancers obtained from patient samples. Many of these were new markers not previously considered as cancer markers, and many of the genes are expressed not by the tumor cells but instead by the patient's response to the tumors. Some of these markers are known to be secreted and to be detectable in simple blood samples.

Why is this important?
A strong effort is underway to examine these candidate genes and to develop markers for use in reliable assays for cancer that can be done on serum, to aid medical imaging, and to serve as targets for the development of invasion-specific anticancer therapy.

Activin Receptors - A New Anticancer Signal in Human Tumors
The major problem with human tumors is a social one. Tumor cells do not obey the signals from their surrounding cells that should restrain their growth. To date, very few of such signals have been defined, and this limits our ability to understand and counter this basic abnormality.

Because of the need to understand these signals, there has been a great effort to identify genes that are mutated and turned off in tumors. These are the tumor-suppressor genes. The inactivation of these genes allows tumors to escape from the normal growth controls that the surrounding cells and tissue are trying to place on them.

Activin is a protein secreted by normal cells. To exert its action, activin must bind receptors on a cell. The receptors propagate a signal to the cell, but it was not previously known that these signals were able to suppress tumor growth. Mutations within the activin receptor gene were found recently in some pancreatic cancers by Dr. Gloria Su and colleagues in Dr. Kern's laboratory for pancreatic cancer research at Johns Hopkins.

Why is this important?
In tumors that lack the mutations, someday it might be possible to administer activin as a therapeutic strategy. It might also be possible to mimic the effects of activin on tumor cells by a precise molecular targeting using specially designed new drugs that directly activate the signal pathway without the need for intact receptors. This is a new idea that was previously unknown, but now can be explored. It is our hope that one can design a rational therapy that would specifically attack the most vulnerable components of pancreatic cancer.

Discovery of New Markers of Pancreatic Cancer
In the June 1^st issue of Cancer Research Dr. Argani and colleagues from Johns Hopkins described the discovery of a new marker of pancreatic cancer. This new marker called "prostate stem-cell antigen" (PSCA) was discovered by using a technique developed at Johns Hopkins called "serial analysis of gene expression" (SAGE, see the What's New May 28, 1997). Since the original description of SAGE, a group of cooperating scientists from a number of institutions have created an online database of gene expression that includes SAGE data on a variety of tissues and cancers ( http://www.ncbi.nlm.nih.gov/SAGE/). The investigators at Hopkins used this database to compare the gene expression levels in pancreatic cancer tissues with those seen in non-cancerous pancreatic tissues. The goal was to identify genes that were selectively "turned-on" in the cancers. One of the genes the Hopkins found using this approach coded for a protein called "prostate stem-cell antigen." Prostate stem-cell antigen is a gene that was originally thought to be largely restricted to prostate cells. Dr. Argani and colleagues demonstrate that prostate stem-cell antigen (PSCA) is, in fact, highly overexpressed in approximately 60% of primary pancreatic cancers. It is not expressed in the normal pancreas.

Why is this important?
These findings are exciting for several reasons. First, they demonstrate the power of new technologies such as SAGE to discover new tumor markers. Second, PSCA, because it is selectively overexpressed in pancreatic cancers, maybe a useful marker for pancreatic cancer. Third, other groups have shown that PSCA can be an immune target and therefore PSCA is being explored as a target for the immune treatment of cancers. The demonstration of PSCA expression in pancreatic cancer suggests a new avenue for treating pancreatic cancers. That is, immunotherapy directed at cells expressing PSCA.

On an important side note, this work was supported, in large part, by generous donations from the friends and family of Michael Rolfe demonstrating the power of private giving to advance pancreatic cancer research.

Familial Pancreatic Cancer
For years, isolated reports in the medical literature have suggested that pancreatic cancer runs in some families. For example, it has been reported that former President Jimmy Carter lost his father, brother and two sisters from pancreatic cancer.

Tersmette and colleagues from Johns Hopkins report that first-degree relatives (brothers and sisters, parents and children) of patients with "familial pancreatic cancer" have a significantly increased risk of developing pancreatic cancer. Tersmette and colleagues followed 341 families enrolled in the National Familial Pancreas Tumor Registry (NFPTR) at Johns Hopkins and found that the first-degree relatives of familial pancreatic patients had an 18-fold increased risk of developing pancreatic cancer when compared to the general population (the "SEER" database). In this study, familial pancreatic cancer was defined as at least a pair of first-degree relatives with pancreatic cancer in a family. Remarkably, if there were three or more family members with pancreatic cancer when the family enrolled in the NFPTR, then the risk of other family members developing pancreatic cancer jumped to 57-fold greater than the general population.

Why is this important?
This study firmly establishes that "Familial Pancreatic Cancer" is a real entity and it provides a quantitative measure of the risk of pancreatic cancer in these families. Importantly, studies such as this will form the basis for identifying individuals at-risk for developing pancreatic cancer who might benefit from new screening tests as they are developed.

If you have a strong family history of pancreatic cancer and would like to join the research studies currently underway at Hopkins, please consider joining the NFPTR. If you would like to join, please contact the Coordinator of the NFPTR, Mirian Tillery (mtillery@jhmi.edu ).

New Vaccine to Treat Pancreatic Cancer
In the January issue of The Journal of Clinical Oncology (volume 19; 2001: pages 145-156), Dr. Elizabeth Jaffee and colleagues at Johns Hopkins reported the result of a phase I clinical trial of a novel vaccine treatment for patients with a pancreatic cancer. The vaccine was produced by genetically altering pancreatic cancer cells growing in culture so that the cells would produce large quantities of an immune activating factor called "Granulocytic-macrophage colony-stimulating factor" (or GM-CSF for short). Dr. Jaffee treated 14 patients with this vaccine in a phase I dose escalation trial. The patients underwent surgery at Hopkins after which they received various doses of the vaccine. No dose-limiting toxicities were encountered. Instead, Dr. Jaffee was able to demonstrate that the vaccine induced an anti-tumor immune response in three patients who received the highest dose of the vaccine (>10x10⁷ vaccine cells). Remarkable, these three patients remained alive and free of disease more than 25 months after diagnosis.

Why is this important?
Based on these results, Dr. Jaffee and her team are conducting phase II trials of the GM-CSF vaccine. As much as Dr. Jaffee would like to offer the vaccine to everyone, eligibility criteria had to be established for this study. Patients with adenocarcinoma of the pancreas who have surgery Johns Hopkins Hospital to remove their pancreas cancer and who have no clinical evidence of spread of the cancer outside the pancreas will be eligible for this study. Patients with bile duct cancer or neuroendocrine tumors or islet cell cancer are not eligible. Please contact Dr. Elizabeth Jaffee (ejaffee@jhmi.edu) or Barbara Biedrzycki, R.N. (biedrba@jhmi.edu) for more information on eligibility criteria.

YEAR 2000

Genes isolated from Tumor Vessels:
Dr. St. Croix in Bert Vogelstein and Kern Kinzler's laboratory identified 46 genes which appear to be specifically elevated (or "turned on") in tumor-associated blood vessels.

Why is this important?
The identification of these genes is exciting because they are potential targets for drugs aimed at shrinking tumors by starving them of their blood supply.

Thousands of Chemical Compounds Tested
Drs. Gloria Su and Taylor Sohn in Scott Kern's laboratory tested over 16,000 chemical compounds looking for compounds that can "turn on" on a specific function that is often "broken "in pancreatic cancer (DPC4 signaling pathway). They identified half a dozen promising compounds, one of which turned out to be a novel and specific inhibitor of a protein called HDAC (histone deacetylase).

Why is this important?
This finding represents a new method to develop drugs to treat pancreatic cancer - "fixing" the cancer cells by screening for compounds that turn on specific functions lost in the cancer.

Gene Inactivation in Precursors to Invasive Pancreatic Cancer
Drs. Michael Goggins and Robb Wilentz studied the genetic changes in the very earliest lesions that give rise to invasive pancreatic cancer. They found that the BRCA2 and DPC4 genes are inactivated in some of these early lesions (called "Pancreatic Intraepithelial Neoplasia").

Why is this important?
These studies provide an important first step in the development of novel screening tests to detect early, and therefore potentially curable, pancreatic cancers.

DNA Methylation in Pancreatic Cancer
Dr. T. Ueki in Michael Goggins' laboratory discovered that a number of genes are selectively "hypermethylated" in pancreatic cancer. Methylation refers to the addition of an extra carbon atom to DNA and it is a common mechanism by which cancer preventing genes (tumor-suppressor genes) are inactivated in cancers.

Why is this important?
These findings provide a novel target for the development of a new screening test for pancreatic cancer.

Quality of Life After a Whipple Drs. Taylor Sohn and Charles Yeo studied patients' self-reported quality of life after Whipple surgery (pancreaticoduodenectomy). Over 192 patients were studied and over 30 quality of life measurements assayed. Remarkably, quality of life scores for patients who underwent a Whipple at Johns Hopkins were comparable to those patients who had their gallbladder removed for stones.

Why is this important?
These data demonstrate that surviving Whipple patients as a group have near normal quality of life scores. This corrects the misimpression that Whipple patients have severely impaired quality of life, and cannot return towards normal activities.

New Immunohistochemical Stain for the DPC4 Gene Product
In 1996, Dr. Scott Kern and colleagues at Hopkins discovered a new gene which appeared to be selectively inactivated (deleted) in the development of cancer of the pancreas. Dr. Kern and his colleagues named this gene "DPC4" for Deleted in Pancreas Cancer locus 4 (see Science 1996, vol. 271:350-353). The discovery of this gene was exciting because DPC4 is inactivated in a large number of cancers of the pancreas, and because its inactivation appears to be relatively specific for cancer of the pancreas. That is, DPC4 appears to be only rarely inactivated in other tumor types (see Cancer Research 1996, vol 56:2527-2530).

Robb Wilentz and colleagues from Johns Hopkins developed a new immunohistochemical stain for DPC4. This new stain can detect DPC4 in tissue sections (such as biopsies), and the staining pattern directly mirrors the DPC4 gene status.

Why is this important?
Because of its simplicity and availability, the immunohistochemical staining technique Dr. Wilentz developed for DPC4 has a number of direct clinical applications. For example, staining for DPC4 may help to distinguish benign chronic pancreatitis (which should express DPC4) from cancer of the pancreas (half of which will not express DPC4). Thus, DPC4 staining will add a valuable tool to the interpretation of needle biopsies of the pancreas. In addition, the immunohistochemical assay reported by Wilentz and colleagues for DPC4 may lead to answers in the investigative area. For instance, the immunohistochemical study of early lesions in the pancreas may help determine the stage at which DPC4 inactivation contributes to neoplastic progression and thereby help in the development of new screening tests for early pancreatic cancer.

YEAR 1999

Careful Pathologic Examination Can Be Used to Predict Outcome for Patients with Mucinous Cystic Tumors:
Mucinous cystic tumors are rare neoplasms of the pancreas characterized by the presence of large cysts (fluid filled cavities) lined by mucin producing cells. Some investigators have suggested that all mucinous cystic neoplasms are malignant (capable of spreading to other organs) and that all mucinous cystic tumors should therefore be designated as cancers - "mucinous cystadenocarcinomas."

Robb Wilentz at Johns Hopkins studied over 60 of these rare tumors and found that with careful examination they could accurately separate mucinous cystic tumors into two groups - those that are entirely benign (the tumors never recurred) if completly resected ("mucinous cystadenomas") and those that had a malignant or cancerous potential ("mucinous cystadenocarcinomas"). Importantly, two-thirds of the mucinous cystic tumors in Dr. Wilentz's study fell into the entirely benign group.

Why is this important?
This study demonstrates that simply lumping all mucinous cystic tumors into the malignant category would have incorrectly labeled two-thirds of the patients as having cancer when they didn't!

The study is also important because it highlights the impact private donations can have on research. Dr. Wilentz conducted this research during a research fellowship year he spent in the laboratory of Dr. S. Kern and this fellowship was supported by the Helen S. Heller and Daniel Kim Memorial Funds for pancreatic cancer research at Hopkins. Without this private support, Dr. Wilentz would not have been able to do his study. Private giving makes a difference!

Peutz-Jeghers Gene Plays a Role in Pancreatic Cancer:
Twelve years ago doctors at Johns Hopkins reported that patients with a rare syndrome, called "The Peutz-Jeghers Syndrome," had an increased risk of developing pancreatic cancer. The reason for this increased risk remained a mystery until now.

The Peutz-Jeghers Syndrome is rare inherited syndrome in which affected patients develop dark pigmented spots on their lips ("mucocutaneous melanin macules") and polyps in their intestinal tract ("hamartomas"). These patients have an increased risk of cancer, expecially pancreatic cancer. Gloria Su, Ph.D. and colleagues from Johns Hopkins uncovered the genetic basis for this association.

The STK11/LKB1 gene in chromosome 19 is responsible for the Peutz-Jeghers Syndrome. Gloria Su and colleagues examined the status of the STK11/LKB1 gene in a large series of pancreatic cancers and in pancreatic cancer resected from patients with the Peutz-Jeghers Syndrome. They found that the STK11/LKB1 gene was inactivated in 4-6% of the pancreatic cancers. While the inactivation of this gene plays a role in only a small percentage of pancreatic cancers, Gloria Su and colleagues made a second, quite remarkable discovery. They found that the inactivation of the STK11/LK1 gene in patients with the Peutz-Jeghers Syndrome explained the development of pancreatic cancer in these patients.

Why is this important?
These findings solve a long-standing mystery. The inheritance of a defective copy of the STK11/LKB1I gene causes the Peutz-Jeghers Syndrome and the inactivation of this gene in these patients explains their increased risk of pancreatic cancer.

Radical Whipple is Safe
Surgical resection is currently the most effective treatment for cancer of the pancreas. The extent of the surgery which should be performed is, however, controversial. Some have argued that the standard pancreaticoduodenectomy (Whipple procedure) should be extended to include the removal of the distal stomach (distal gastrectomy) as well as the removal of additional lymph nodes (retroperitoneal lymphadenectomy). The more extended surgery is called a "radical pancreaticoduodenectomy". The controversy over standard versus radical Whipple has been difficult to resolve, because most centers do one surgery or the other and data between institutions may not be comparable.

In the May 1999 issue of the Annals of Surgery, Drs. Yeo, Cameron and colleagues from Johns Hopkins report a study that may finally help resolve this controversy. They reported a randomized single-institution trial in which patients were randomized to receive either a standard or a radical Whipple. Of the 114 patients randomized, 56 underwent a standard Whipple and 58 a radical Whipple. Dr. Yeo and colleagues found that the two procedures can be performed with similar morbidity and mortality. Importantly, the one-year survival rate for both groups was similar (~80%).

Why is this important?
This important study will continue and the patients enrolled will be followed to determine if there are any long-term benefits to doing a radical Whipple. For now, part of the controversy in the radical versus standard Whipple debate has been answered. Both can be performed with equal morbidity and mortality, but the radical Whipple does not provide any improvement in survival at one year.

DPC4 Pathway Targeted in Pancreatic Cancer
Drs. Michael Goggins and Jaile Dai in Scott Kern's laboratory extensively studied the DPC4 pathways in pancreatic cancer using modern genetic techniques. Dr. Goggins found that the "TGF-ß receptors" can be targeted in pancreatic cancer and Dr. Dai discovered that if he took cells that made no DPC4, and then genetically engineered the cells so that they would produce DPC4, then the cells grew much slower. He also showed that activated DPC4 can kill selected cancer cells.

Why is this important?
These studies provide a clearer understanding of just how pancreatic cancer cells come to misbehave. An understanding of "what's broken" in pancreatic cancers should, in turn, lead to new methods to treat the disease.

YEAR 1998

New Type of Pancreatic Cancer
Michael Goggins identified a new type of pancreatic cancer called "Medullary Cancer". This new type of pancreatic cancer has a very specific appearance under the microscope, and Mike has discovered that medullary cancers have an usual pattern of genetic changes called "microsatellite instability".

Why this is important?
Patients with this newly recognized type of pancreatic cancer may have a better prognosis than patients with typical pancreatic cancer, and they may also respond to different types of chemotherapy. Furthermore, because this newly recognized type of cancer has an unusual genetic change associated with it, it may be possible to detect these cancers using genetic tests.

Familial Pancreatic Cancer
Ralph Hruban established the National Familial Pancreas Tumor Registry at Hopkins in 1994. Over 350 families have enrolled in this registry. These include a family in which five brothers and sisters died of pancreatic cancer and seven families in which three generations have been affected. Analysis of the families enrolled in this registry has demonstrated that genetic changes cause pancreatic cancer to run in some families. Just as children inherit their hair color and eye color from their parents, so too do some inherit an increased risk of developing pancreatic cancer.

Why is this important?
As the genetic changes responsible for familial pancreatic cancer are uncovered, family members can be tested to see if they are at risk. Furthermore, because they have such a high risk of developing pancreatic cancer, these families will be the first to benefit from new screening tests for pancreatic cancer.

Surgery For Pancreatic Cancer
Surgeons at Hopkins have worked hard to improve the surgical approach to tumors of the pancreas. Thanks to their efforts the Whipple operation (removal of the head of the pancreas) is now a safe option for many patients. In fact more Whipple procedures are performed every year at Hopkins than any other hospital in the world.

Why is this important?
The impact of improved surgery on the pancreas has been dramatic. Operative mortality rates have dropped from over 20% to less than 2% at Hopkins. Remarkably, because more patients are now coming to Hopkins, these improvements at Hopkins have led to a significant decrease in hospital mortality in the entire state of Maryland for pancreatic surgery.

YEAR 1997

Pancreas Cancer Vaccine
Elizabeth Jaffee at Hopkins has developed a novel vaccine for the treatment of pancreatic cancer. Dr. Jaffee has used two cancers resected from patients at Hopkins to develop this vaccine. After the cancers were grown and cultured, she used a virus to introduce a new gene into the cancer cells. This new gene called "Granulocyte - Macrophage Colony Stimulating Factor" was incorporated into the cancer cells and it codes for a factor which actively stimulates the body's immune system. These cultured cancer cells are then used to produce a live anti-pancreas cancer vaccine. This vaccine has already been tested on 15 patients with pancreatic cancer in a Phase I clinical trial. Although early in the testing process, Dr. Jaffee has already seen positive skin reactions in vaccinated patients.

Why is this important?
This is a completely new approach to the treatment of pancreatic cancer. Should it prove effective, it will harness the body's immune system against the cancer without causing the side effects of chemotherapy. Importantly, Dr. Jaffee has been studying the immune reaction of patients vaccinated with this vaccine and hopes to use these studies to develop a second generation of pancreatic cancer vaccines.

Genetic Profile of Pancreatic Cancer
Ester Rosenblum in Scott Kern's laboratory has characterized the genetic profile of pancreatic cancer. This includes mutation (DNA changes) in multiple genes, including the K-ras, p16, p53, DPC4, BRCA2, LKB1, MKK1 and TGF Beta receptor genes.

Furthermore, Robb Wilentz and Carlos Caldas have demonstrated, using the technique of PCR (polymerase chain reaction - the equivalent to a genetic xerox machine), that some of these genetic changes (such as K-ras) can be detected in the duodenal fluid and/or stool of patients with pancreatic cancer. Why is this important?
An understanding of the genetic profile of pancreatic cancer will allow scientists to develop new drugs to treat pancreatic cancer, and, because these genetic changes can be detected in stool these discoveries will lead the way to new screening tests to detect early pancreatic cancers. Just as colon cancer can now be detected by testing stool for blood, so too do we envision detecting pancreas cancer by testing stool samples for genetic changes.

Precursors to Pancreatic Cancer
Dan Brat and Robb Wilentz have examined a large series of resected pancreatic cancers and they have identified the precursor lesions to cancer of the pancreas. Called "Duct Lesions" these precursors arise in the smallest pancreatic ducts. They probably arise years before the patient develops an infiltrating cancer and many years before they develop any symptoms. Hopkins scientists have characterized these under the microscope, and they have also studied these duct lesions at the genetic level. This two-pronged approach to the study of duct lesions has established that duct lesions are indeed the elusive precursor to pancreatic cancer.

Why is this important?
This important findings represents a first step towards the development of new tests to detect pancreatic cancer at very early stages. If we are to be effective in treating pancreatic cancer we must detect it before the "Horse out of the barn."

YEAR 1996

DPC4
Stephen Hahn in Scott Kern's laboratory at Hopkins discovered the DPC4 (Deleted in Pancreas Cancer 4) gene on chromosome 18 and he demonstrated that this gene is mutated (missing) in approximately half of all pancreatic cancers. In the other half of the cancers, DPC4 is harder to turn on, due to other gene defects. Remarkably, the mutation of DPC4 appears to be relatively specific for pancreatic cancer.

Why is it important?
DPC4 is the first gene which appears to specifically altered in pancreatic cancer. Now that it has been identified scientists at Hopkins are working to understand how DPC4 works so that they can develop new drugs to "replace" the missing DPC4 function in pancreatic cancers.

The Breast Cancer Gene (BRCA2)
The second breast cancer gene (called "BRCA2") was discovered because of a remarkable advance made by the Johns Hopkins pancreas cancer research team. Mieke Schutte in Scott Kern's lab used a revolutionary technique called "RDA" to study a pancreas cancer removed by surgeons at Johns Hopkins. She discovered that this cancer was missing a small piece of DNA from chromosome 13. This piece of DNA was the second breast cancer gene. Michael Goggins extended these and found that as many as 7% of patients with pancreas cancer get their cancer because they inherit had a defective copy of the BRCA2 gene. These patients were born with a defective copy of the BRCA2 gene and they inherited this defective copy from one of their parents.

Why is this important?
This discovery is quite remarkable because it suggests: (1) that there is a link between pancreas and breast cancer; and (2) that patients with a family history of breast and pancreas cancer can now be tested to see if they carry this gene.

YEAR 1995

Chromosome Changes Associated with Pancreas Cancer
Connie Griffin has applied "classical cytogenetics" (a test performed on fluids obtained by amniocentesis from pregnant women to determine the health of their babies) to visualize and examine individual chromosomes in pancreatic cancers. She examined a large number of pancreatic cancers using this technique and has identified recurring chromosome changes in these tumors. These include losses of specific chromosome, and gains of other portions of chromosomes. These studies provide an important advance in our understanding of the genetic changes responsible for the development of cancer of the pancreas.

Why is it important?
Because genes sit on chromosomes, the cytogenetic analysis of cancers can lead to the discovery of genes important of the development of the cancer. It is our hope that further analysis of the recurrent chromosome changes identified by Connie Griffin, will lend to the identification of additional yet undiscovered genes responsible for cancer of the pancreas. Once discovered an understanding of the function of these genes may lead to new ways to diagnose and treat pancreatic cancers.

SAGE
Victor Velculescu in Ken Kinzler's laboratory developed a revolutionary technology called "SAGE" (serial analysis of gene expression). SAGE is a powerful technique which can be used to identify and quantify all genes expressed in a tissue. When the investigators applied SAGE to pancreatic cancer they were able to identify a total of 404 "transcripts" expressed at high levels in pancreatic cancers (transcripts are pieces of RNA which code for proteins (also called antigens) that might be released into the blood). The identification of these transcripts is an important advance, because any one of the 404 may form the basis for a new screening test. In an effort to share this exciting data the Hopkins investigators have established a Web page on which they freely share all of their SAGE findings (http://welchlink.welch.jhu.edu./~molgen-g/home.htm).

Why is this important?
This is the first step in developing a completely new screening test for pancreatic cancer. For example, prostate cancers make "prostate specific antigen" which is detectable in the blood and which forms the basis for the current screening test for the early detection of prostate cancers. The application of SAGE technology to pancreatic cancers is the first step in developing such a test for the pancreas. We hope to find a "pancreas specific antigen".

YEAR 1994

Role of p16
Carlos Caldas in Scott Kern's laboratory was the first to demonstrate the role of p16 gene in the development of pancreatic cancer. Christopher Moskaluk extended these studies and demonstrated that some cases of inherited pancreatic cancer, particularly those associated with melanoma, are caused by inherited mutations (defects) in the p16 gene. Dr. Moskaluk also showed that p16 is defective in many of the earliest forms of pancreatic cancer, providing a clue as to how the disease develops.

Why this is important?
Family members from families in which there is a strong history of pancreatic cancer and melanoma can now be tested for inherited defects in the p16 gene. Carriers of the p16 defects can be screened more carefully for pancreatic cancer and melanoma, while those found not to carry the p16 defect will be relieved of their anxiety.