pancreas cancer : How important is finding an experienced surgeon?

Pancreatic cancer is the fourth leading cause of cancer death in the United States. Recent data from the National Cancer Database indicate that pancreaticoduodenectomy (the Whipple procedure: designed to remove the head, neck and uncinate process of the pancreas as well as the majority of the duodenum) is the most commonly performed cancer-directed operation for pancreatic cancer, although it is used in only 9% of patients. In this large national database, the five-year survival rate for patients treated by pancreaticoduodenectomy in 1985 was 3%. In contrast to these national figures, specialized centers have reported decreasing operative mortality rates and improving long-term survival rates after pancreaticoduodenectomy for pancreatic cancer. For example, the five-year survival rate for patients treated surgically at Johns Hopkins now exceeds 20%. Many factors are likely to be responsible for the improving safety of pancreaticoduodenal resection, including improvements in intensive and critical care, improved surgical experience with decreases in operative time and less need for blood replacement, and regionalization of patient care to specialized "Centers of Excellence," such as Johns Hopkins.

In the April 11, 2002 issue of the New England Journal of Medicine, Dr. Birkmeyer and colleagues from Department of Veterans Affairs, Vermont, report their analysis of surgical mortality in the United States (N Engl J Med 2002 Apr 11;346(15):1128-37). Using information from the national Medicare claims database and the Nationwide Inpatient Sample they examined the relationship between hospital volume (total number of procedures performed each year) and mortality (death in hospital or within 30 days) for a variety of surgical procedures. The mortality rate for Whipple procedures (pancreaticoduodenectomy) at low-volume centers (16.3%) was much greater than the mortality rate at high-volume centers (3.8%). From their analyses the authors conclude that patients "can significantly reduce their risk of operative death by selecting a high-volume hospital." High-volume centers were defined in this study as centers that perform more than 16 whipples per year. Last year close to 240 whipples were performed at Johns Hopkins.

A recent analysis performed at The Johns Hopkins Hospital has determined the factors which favor long-term survival after pancreaticoduodenectomy. Between April 1970 and April 1994, 208 patients underwent a standard pancreaticoduodenectomy for adenocarcinoma of the head of the pancreas at The Johns Hopkins Hospital. The preference of surgeons at Johns Hopkins is to perform partial pancreatectomy whenever possible, leaving the pancreatic body and tail in place. Distal gastrectomy is typically reserved for tumor involvement of the distal stomach or first portion of the duodenum. Multiple factors were analyzed including patient demographics, intraoperative factors, tumor characteristics, and post-operative use of adjuvant therapy. The primary outcome variable analyzed was survival. The results of this review, which is the largest single institution experience reported to date, allowed assessment of 201 of the 208 patients. Seven patients had incomplete outcome data and were excluded. For the group of 201 patients comprising the study population, the overall postoperative in-hospital mortality rate was low, with the current figure being 0.7% (for the last 149 patients). This means that survival data for patients treated at Johns Hopkins reflect an operative mortality rate of less than 1%. The mean age of the patients was 63 years, with a slight male predominance (108 men and 93 women). There were no differences in survival based on age, gender, or race. The actuarial one, three and five-year survival rates for all 201 patients were 57%, 26%, and 21% respectively, with a median survival of 15.5 months. There were 11 five-year survivors, 7 six-year survivors, and one fifteen-year survivor. By univariate statistical analysis, a significant improvement in survival has been observed from the decade of the 1970s, through the decade of the 1980s, to the decade of the 1990s. Patients resected in the 1970s had a median survival of 7.5 months and a three-year survival of only 14%, while patients undergoing resection in the 1990s had a median survival of 17.5 months and a three-year survival of 36%.

Multiple intraoperative factors were analyzed in order to determine their impact on survival. Using multivariate analysis, the type of pancreatic resection, the intraoperative blood loss, the number of packed red blood cell transfusions and the operative time were not found to be significant prognostic factors. However, multivariate analysis revealed that many tumor characteristics were important in predicting patient survival. Patients with tumors less than 3 cm in diameter had significantly longer median survival and better five-year survival compared with patients with tumors 3 cm or more in diameter. The status of the lymph nodes in the resected specimen proved to be a highly significant factor predicting survival. The status of the resection margins also proved to be a highly significant factor, with patients undergoing resection with negative margins having a median survival of 18 months and a five-year survival of 26%, whereas those resected with positive margins had a median survival of only 12 months and a five-year survival of 8%. The DNA content of the pancreatic cancer cells, as determined by image cytometry, proved also to be a highly significant determinant of survival. Patients with diploid tumors had a median survival of 24 months and a five-year survival of 39%, significantly better than the median survival of 11.5 months and five-year survival of 8% observed in patients with aneuploid tumors.

Since the completion of this review of 201 patients with pancreatic cancer, nearly 400 additional pancreatic cancer patients have been treated by pancreaticoduodenectomy at The Johns Hopkins Hospital. It is likely that more pancreas cancer surgery is performed at Johns Hopkins than at any other hospital in the world.
Summary
In recent years the surgical treatment of adenocarcinoma of the head, neck or uncinate process of the pancreas via pancreaticoduodenectomy has been associated with falling postoperative morbidity and mortality rates and improving long-term survival. The results from our recent single institution experience demonstrate an actuarial five-year survival of 21% for all patients undergoing pancreaticoduodenal resection for adenocarcinoma of the pancreas. Importantly, the actuarial five-year survival is improved for patients resected with tumors less than 3 cm in diameter (28%), negative margins (26%), negative nodal involvement (36%), or diploid tumor DNA content (39%). Multivariate analysis indicated that the parameters that serve as the strongest independent predictors of favorable outcome are tumor DNA content, tumor diameter, status of resected lymph nodes, margin status, and decade of resection (resection in the 1990s being most favorable). The increasing use of postoperative combined modality chemoradiation therapy appears to be another factor favoring long-term survival.
This large series from The Johns Hopkins Hospital provides room for cautious optimism in the treatment of pancreatic adenocarcinoma. The development of more promising adjuvant therapies, such as strategies combining chemoradiation with immunotherapy, may further enhance survival. Additionally, developments in the field of molecular genetics hold promise for the earlier detection of pancreatic cancer, possibly using gene-based diagnostic modalities.

Copyright ©1996-2006 The Johns Hopkins University, Baltimore, Maryland