The Sol Goldman Pancreatic Cancer Research Center

What's New 1997

Hopkins Surgeons Report Their Experience With 650 Whipples Performed in the 1990's
December 18, 1997

Dr. Charles Yeo, a Professor of Surgery at Johns Hopkins with a particular interest in cancer of the pancreas, and colleagues reported their experience with 650 consecutive Whipple resections performed at The Johns Hopkins Hospital in the 1990's in the September issue Annals of Surgery (1).

Between January 1990 and July 1996 650 patients underwent a Whipple resection at The Johns Hopkins Hospital. Dr. Yeo followed these patients to determine the pathology, complications and outcomes. Most of the Whipple resections were performed for cancer of the pancreas (43%), however, Whipples were also performed for ampullary cancers (11%), distal common bile duct cancers (10%), duodenal cancers (4%), chronic pancreatitis (11%), neuroendocrine tumors (5%), periampullary adenomas (3%), cystadenocarcinomas (2%), cystadenomas (4%) and other conditions (7%). The median number of units of red cells transfused was 0 and the median operative time was 7 hours. During this period, 190 consecutive were performed without a mortality. The overall operative mortality rate for this group of 650 patients was only 1.4%. The most common complications included early delayed gastric emptying, pancreatic fistula, and wound infection. The median post-operative length of stay was 13 days.

In multivariate analysis the most powerful independent predictors favoring long term survival included a pathologic diagnosis of duodenal adenocarcinoma, tumor diameter less than 3 cm, negative resection margins, absence of lymph node metastases, well differentiated histology, and no reoperation. This single institution, high volume experience demonstrates that Whipple procedures can be performed safely for a variety of malignant and benign disorders of the pancreas. Overall survival is determined largely by the pathology within the resection specimen.

Yeo CJ, Cameron JL, Sohn TA, Lillemoe KD, Pitt HA, Talamini MA, Hruban RH, Ord SE, Sauter PK, Coleman J, Zahurak ML, Growchow LB, Abrams RA. Six hundred fifty consecutive pancreaticoduodenectomies in the 1900s. Pathology, complications and outcomes. Annals of Surgery 226(3):248-260, 1997.

Hopkins Scientists Develop New Tests to Analyze Liver Lesions Biopsied from Patients with Pancreas Cancer
October 21, 1997

Patients with cancer of the pancreas often have nodules in their livers. In some patients these nodules are perfectly harmless lesions called "benign bile duct proliferations". In other patients, however, the nodules may be nodules of cancer which have spread from the patients pancreas to the liver. These latter nodules are called "metastatic carcinoma". These different types of nodules can make a big difference in treatment. Surgery to remove a pancreas cancer is effective only in patients in which the disease has not spread beyond the pancreas (called "localized disease"). In contrast, chemotherapy without resection of the primary pancreatic cancer is generally the treatment of choice for patients with liver metastases. Unfortunately, it can be very difficult to distinguish benign bile duct proliferations from well differentiated metastatic pancreatic cancer in the liver. In this month's issue of the American Journal of Pathology research scientists from Johns Hopkins, in collaboration with a group of scientists at the Academic Medical Center in Amsterdam, The Netherlands, report a new gene based test that may help make this distinction.

Most primary pancreatic cancers of the pancreas have mutations in the K-ras oncogene (1). In this study of over 100 liver nodules, the scientists found that when the primary pancreatic cancer had a K-ras mutation, the liver metastases also always had the same mutation. In contrast, the vast majority (94%) of benign bile duct proliferations did not harbor K-ras mutations.

These results suggest that K-ras mutations may be useful in distinguishing benign bile duct proliferations from metastases in the liver, and importantly, they demonstrate a potential clinical application of our improved knowledge of genetics of carcinoma of the pancreas. Other potential applications of tests for K-ras include blood or stool tests for rare pancreatic cancer cells and genetic tests as adjuncts to cytologic diagnoses (1,2).

Hruban RH, van Manseld ADM, Offerhaus GJA, van Weering DHJ, Allison DC, Goodman SN, Kensler TW, Bose KK, Cameron JL, Bos JL. K-ras oncogene activation in adenocarcinoma of the human pancreas: a study of 82 carcinomas using a combination of mutatenriched polymerase chain reaction analysis and allele-specific oligonucleotide hybridization. Am J Pathol 1993, 143:545-554.

Hruban RH, Sturm PDJ, Slebos RJC, Wilentz RE, Musler AR, Yeo CJ, Sohn TA, van Velthuysen MLF, Offerhaus GJA. Can K-ras codon 12 mutations be used to distinguish benign bile duct proliferations from metastases in the liver? A molecular analysis of 101 liver lesions from 93 patients. Am J Pathol 1997, 151:943-949.

Caldas C, Hahn SA, Hruban RH, Redston MS, Yeo CJ, Kern SE. Detection of K-ras mutations in the stool of patients with pancreatic adenocarcinoma and pancreatic ductal hyperplasia. Cancer Res 54:3568-3573, 1994.


Hopkins Researchers Demonstrated the Value of Post-Operative Chemo and Radiation Therapy
August 8, 1997

Should patients with pancreas cancer get any additional therapy after the tumor has been surgically removed? This important question has finally been answered. In the May 1997 issue of Annals of Surgery (Vol 225, 631-636) Dr. Charles Yeo and his colleagues report the Johns Hopkins experience treating patients with radiation and chemotherapy after they have had a Whipple procedure for pancreas cancer. 174 patients were enrolled in this study and the patients received one of three post-operative treatments. These included: 1) standard therapy [external beam radiation therapy to the pancreatic bed given with two 3-day fluorouracil (5-FU) courses followed by weekly bolus 5-FU for 4 months]; 2) intensive therapy [external beam radiation therapy to the pancreatic bed with prophylactic hepatic radiation given with and followed by infusional 5-FU plus leucovorin for 4 months]; or 3) no therapy (no post-operative radiation therapy or chemotherapy). The Hopkins researchers found that the use of adjuvant chemo radiation therapy was associated with significantly longer patient survival. Patients who received post-operative chemo-radiation therapy had a median survival of 19.5 months compared to 13.5 months without therapy. The intensive therapy group had no survival advantage compared to the standard therapy group. From this study the Hopkins researchers conclude that adjuvant chemo radiation therapy significantly improves survival after Whipple procedure for adenocarcinoma of the head, neck or uncinate process. Indeed, based on these survival data, standard adjuvant chemo radiation therapy appears to be indicated for patients treated by Whipple procedure.

Yeo CJ, Abrams RA, Grochow LB, Sohn TA, Ord SE, Hruban RH, Zahurak ML, Dooley WC, Coleman J, Sauter PK, Pitt HA, Lillemoe KD, and Cameron JL. Pancreaticoduodenectomy for pancreatic adenocarcinoma: Postoperative adjuvant chemoradiation improves survival. Annals of Surgery 225(5):621-636, 1997.


A Genetic Fingerprint Proves the Origin of Pancreatic Cancer
July 1, 1997

The best way to fight pancreatic cancer would be to detect it in its earliest stages. But there has been some uncertainty as to what the earliest form might look like. The pancreases of elderly persons often have ducts with abnormal lining cells, but for years these lesions were largely overlooked as trivial. In the vast majority of cases, the microscopic appearance of these lesions does not readily suggest that they might be worrisome. Indeed, one could calculate that if these ductal lesions were the precursor of pancreatic cancer, then only about 1% of them appear to progress to that stage. These small numbers, and the rather benign microscopic appearance of most duct lesions, have raised interesting questions about the true origin of pancreatic cancer.

In a recent issue of the scientific journal, Cancer Research (June 1, 1997), Drs. Chris Moskaluk, Ralph Hruban, and Scott Kern at Johns Hopkins seem to have finally nailed the culprit. And it is indeed the common, ordinary-looking duct lesion. They used a rare mutation in the p16 cell cycle gene, found in infiltrating pancreas cancer, as a marker to determine if any of the nearby duct lesions were related to the infiltrating cancer. This mutation changed only one of the hundreds of nucleotides (building blocks of DNA) in this gene, and could serve as a genetic "fingerprint" for the origin of the cancer. Remarkably, the same p16 mutation was found in the infiltrating pancreas cancer and in a neighboring duct lesion. Another type of mutation, this time in the K-ras gene, was also shared between the duct and the cancer. The proof was now in: a rather unimpressive duct lesion had, over what was probably the span of many years or decades, evolved into the cancer which finally invaded through the duct wall to become an aggressive tumor. Evidence from other patients was also identified to support this scenario.

A small proportion of duct lesions, about 20%, evolve to become more aggressive-looking, as judged by the microscope. Previous work by the Hopkins investigators and by researchers in Japan (Cancer Research 1994; 54:3568; Cancer Research 1993; 53:953) had indicated that mutations of the K-ras gene, common in pancreatic cancers, also appeared more often in the duct lesions with the most aggressive microscopic appearances. K-ras gene mutations were found predominantly in the ducts having the features of "dysplasia" and "papillary architecture". But a large variety of pancreatic abnormalities harbor mutations of the K-ras gene, and therefore they did not seem to be a very specific fingerprint for the most aggressive lesions. The current work extends this type of analysis by using a more specific marker, the p16 gene. Now that the origin of pancreatic cancer is established, we can work harder to determine how these lesions make the transition to cancer, and which individuals might be at greatest risk for progression of their duct lesions.

Moskaluk CA, Hruban RH, Kern SE. p16 and K-ras gene mutations in the intraductal precursors of human pancreatic adenocarcinomas. Cancer Research 57:2140-2143, 1997.


Pancreas Cancer Vaccine Approved For Study
June 10, 1997

Hopkins clinicians/scientists involved in pancreas cancer research are pleased to announce that the FDA has just approved a new vaccine trial for the treatment of pancreas cancer. We are now enrolling patients into this pancreatic tumor vaccine study. Patients who have newly diagnosed stage I, II, or III adenocarcinoma of the pancreas are eligible. Patients must be able to undergo resection of their pancreatic tumor at The Johns Hopkins Hospital. For more information, please contact one of the following co-investigators:

Pat Sauter, R.N. (410)-955-5718
JoAnn Coleman, R.N. (410)-955-5718
Charles Yeo, M.D. (410)-955-7496
John Cameron, M.D. (410)-955-5166
Keith Lillemoe, M.D. (410)-955-7495

This trial will use an allogeneic irradiated pancreas cancer vaccine generated by ex vivo granulocyte-macrophage colony-stimulating factor gene transfer. Such an approach has already proven "feasable, safe and bioactive" in patients with renal cell carcinoma (see Cancer Research vol 57, 1537-1546, 1997).


Gene Expression In Pancreas Cancer
May 28, 1997

The first step in developing a convenient screening test for a cancer is to identify the proteins that the cancer makes. Once these proteins are identified one can then test blood samples from patients with cancer to see if these proteins are released into and detectable in the blood. For example, prostate cancers make a "prostate specific antigen" which is detectable in the blood and which forms the basis for the current screening test for the detection of early prostate cancers.

In this week's issue of Science Magazine, (Science Vol 276:1268-1272, 1997) Lin Zhang, Wei Zhou, Victor E. Velculescu, Scott E. Kern, Ralph H. Hruban, Stanley R.Hamilton, Bert Vogelstein and Kenneth W. Kinzler from Johns Hopkins report the results of an analysis of a series of pancreas cancers using a new technique called SAGE (serial analysis of gene expression). SAGE is a powerful technique which can be used to identify and quantify all of the genes expressed in a tissue. When the investigators applied SAGE to pancreas cancer they were able to identify a total of 404 transcripts (transcripts are pieces of RNA which code for proteins which might be released into the blood of patients) that were expressed at higher levels in pancreatic cancers as compared to normal. The majority (268) of these transcripts were pancreas-specific; suggesting that this technique could be used to identify new pancreas cancer specific markers which could be used to develop screening tests for pancreas cancer.

L. Zhang, W. Zhou, V. E. Velculescu, S.E. Kern, R.H. Hruban, S.R. Hamilton, B. Vogelstein, K.W. Kinzler. Gene Expression Profiles in Normal and Cancer Cells. Science 276:1268-1272, 1997.

V.E. Velculescu, L. Zhang, B. Vogelstein, K.W. Kinzler. Serial Analysis of Gene Expression. Science 270:484-487, 1995.


Hopkins Researchers Show A Genetic Picture of Pancreatic Cancer
May 1, 1997

For centuries, physicians and the public have wondered about the nature of cancer: What is it? How does it arise? Why do most cancers appear in the later years of life? By the turn of the century, many authorities were already convinced that a cancer represented a highly related aberrant group of cells, the errant offspring of a single original cell of the body. This cell presumably had, through uncontrolled growth and division, given rise to many generations of progeny cells which eventually formed a much larger and ever-proliferating lesion.

By the late 1940s, the mutational theory was gathering considerable excitement, even though the techniques to find mutations in the tumors had not yet been developed. One author in 1949 noted, "The recent work on carcinogenesis and mutagenesis is of ... importance since the rational control of a disease should be based upon its true biological nature" (Yale J Biol Med 21:293, 1949). By the early 1950s, there were attempts to estimate the number of mutations needed. A number of investigators noted that the sharp rise in cancer rates in the elderly could be matched to a mathematical model where the age of a person, when multiplied by itself 4 to 6 times, seemed to match their expected cancer risk. This suggested that nearly half a dozen independent mutations, each with some small chance of occurrence during any one year of life, were needed to form the common cancers (Cancer 4:916, 1951 and Br J Cancer 7:68, 1953, and others). By the late 1980s, there was some success in confirming these estimates, even though the particular mutations and genes involved still were largely unknown.

We now have made great strides in determining what the mutations are and which genes they affect. In the latest issue of the scientific journal Cancer Research (May 1, 1997), Dr. Ester Rozenblum and colleagues report a high-resolution picture of the genetic abnormalities of 40 pancreatic cancers as determined in the Kern Laboratory at Johns Hopkins. Her study probably represents the most precise and exhaustive description of the prevalence of various mutations in a cancer type to date. She finds that most of the cancers involve mutations in at least three to four different genes. Many genes sustained mutations or loss of both of their normal copies, and the number of gene abnormalities was therefore seen to be as high as nine distinct mutations. Each mutation is thought to have aided the evolution of the expanding population of cells into its eventual invasive, cancerous form.

Dr. Rozenblum also found that mutations of any one gene could co-exist with mutations of any other gene she studied. For example, an inherited mutation of the BRCA2 gene was clearly not adequate to cause the pancreatic cancer of one patient. This particular tumor, in order to become fully cancerous, had to develop eight additional mutations in key genes. This is important information, especially for the BRCA2 gene involved in inherited susceptibility to breast, pancreatic, and other tumors. It means that the important tumor-suppressive action of BRCA2 is distinct from the suppressive action of the other genes previously known to involve pancreatic cancer.

When combined with the discovery of the importance of the BRCA2 and DPC4 genes, the past fourteen months have provided discoveries of two totally new avenues for understanding the "true biological nature" of pancreatic cancer. Like the myriad pixels in a digital picture, each mutated gene helps us to construct a clearer picture of this cancer at ever-higher resolution.


Johns Hopkins Awarded Specialized Program of Research Excellence (SPORE) in Gastrointestinal Cancer
February 12, 1997

The National Institutes of Health have awarded the Johns Hopkins Medical Institutions a Specialized Program of Research Excellence (SPORE) in gastrointestinal cancer. This 5 year award supports a highly interactive multidisciplinary program of transitional research directed at reducing the incidence of and mortality from pancreatic and colorectal cancer. The SPORE includes 4 research programs involving 8 research projects which build upon successful existing research programs. In addition, 4 core laboratories are funded. The programs include:

Project 1 - Translational Application of Molecular Genetics for the Early Detection of Pancreatic and Colorectal Cancer

  • Project 1A - "Mutant genes in thesis for early detection of colorectal neoplasms" (Bernard Levin, M.D, MD Anderson)
  • Project 1B - "New Genetic Markers for Pancreatic Cancer" (Scott E. Kern, M.D.) This project aims to continue the work of Dr. Kern in identifying new cancer causing genes responsible for the development of pancreatic cancer.
  • Project 1C - "Targets for screening in pancreatic neoplasia" (Scott E. Kern, M.D.) The overall goal of this project is to define biologically "high risk" precursor lesions in the pancreas and explore the applicability of gene detection as a clinical screening tool.

Project 2 - "Molecular markers for metastasis in colorectal cancer" (Stanley R. Hamilton, M.D.)

Project 3 - "Prevention of pancreatic and colorectal cancer in genetically destined patients by molecular genetic approaches"

  • Project 3A - "Genetic analysis of hereditary colorectal cancer syndromes"(Kenneth Kinzler, M.D.)
  • Project 3B - "Agents and mechanisms of tumor prevention in the MIN model of familial adenomatous polyposis" (Stanley R. Hamilton, M.D.)
  • Project 3C - "Markers for risk in familial pancreas cancer" (Ralph H. Hruban, M.D.) This project will study inheritance patterns in familial pancreatic cancers. It will take advantage of the database and patients cooperating in the National Familial Pancreatic Tumor Registry . These studies should provide a rational basis for predictive gene testing, and for counseling and screening individuals at risk for the development of pancreatic cancer.

Project 4 - "Novel therapies for pancreatic and colorectal cancer - Identification of T-cell recognition targets on pancreatic adenocarcinoma" (Elizabeth Jaffee, M.D.) . This project will identify tumor antigens expressed by human pancreatic adenocarcinomas. Both CD8+ and CD4+ pancreatic tumor specific T-lymphocyte lines and clones will be generated from lymphocytes isolated from patients with adenocarcinoma of the pancreas following vaccination with a GMCSF secreting pancreatic tumor vaccine. A genetic approach will be used to identify genes encoding for MHC class I restricted antigens. The identified antigens will then be evaluated for recognition by CD4+ T-cell lines and clones. Ultimately, the identification of common pancreatic antigens will allow for the development of generalized gene therapy vaccine approaches that can generate immune responses potent enough to treat adenocarcinoma of the pancreas.

Four cores will support these research programs. These include:

  • Core 1 - SPORE administration and communication (Stanley R. Hamilton, M.D.)
  • Core 2 - Human tissue resource and logistic core (Stanley R. Hamilton, M.D.)
  • Core 3 - Colorectal and pancreatic patient registry (Frank Giardiello, M.D. and Ralph H. Hruban, M.D.)
  • Core 4 - Biostatistics (Stephen N. Goodman, M.D., Ph.D.)

Finally, and importantly, the SPORE grant will support a developmental research program headed by Bert Vogelstein, M.D. This program will help fund pilot projects. The SPORE also provides funding for a career development program which will allow for the development of young investigators interested in cancer research.

This SPORE is designed to develop areas of basic science with potential impact on pancreatic and colorectal cancer and to move these promising areas into clinical evaluation including clinical trials. The SPORE is also designed to communicate important findings rapidly into the research community to stimulate investigation, and to bring validated transitional findings into the medical community where the research can ultimately reduce the incidence and mortality of these common cancers.