The course of 196 patients with proven carcinoma of the pancreas seen at Yale New Haven Hospital from 1972 to 1982 was analyzed. Only 73% of the patients were preoperatively expected to have cancer of the pancreas. The patients who underwent resection had the longest mean survival but also the longest total hospital stay. Twenty-seven patients survived 1 year or more, but nonresected patients constituted 81.5% of this group. The only 5-year survivor did not undergo resection. Forty-seven percent of patients who survived 1 year and had not undergone gastroduodenal bypass, developed duodenal obstruction. It was not possible to identify a subset of patients with a favorable prognosis. A review totaling approximately 37000 patients, of whom 4100 had undergone resections, revealed only 156 survivors, 12 of whom had not been resected, for an overall survival rate of only 0.4%. No author had more than 3.4% of the total number of patients as 5-year survivors.

We have reviewed the natural history, reliability of diagnosis, and survivorship of 100 patients with adenocarcinoma of the pancreas, in the context of a thorough review of the literature on survival after therapy for adenocarcinoma of the pancreas. There is 40-62.5% error in the histologic confirmation of the diagnosis of pancreatic cancer. The error by inspection and palpation alone at the time of surgery may be as great as 25%. The absolute 5 year survival rate calculated from 61 clinical studies representing approximately 15,000 patients is 0.4%. The best series in the current literature has only 3% 5 year rate based upon the total population of pancreatic cancer patients. 12.3% of 5 year survivors from the world literature did not have curative surgery. This study shows the necessity for standardization of reporting methods. The same patients and survivors should not be used repeatedly in different reports. Some authors who claim the most effective palliation by pancreatic resection have the highest mortality rates.Cancer 42:2494-2506, 1978.ANCER OF THE PANCREAS, a malignancy C difficult to recognize or treat, has apparently so increased in frequency as to account for 2 1,800 deaths in the United States in 1977' and is now the fourth leading cause of death from cancer among men. The age adjusted mortality rate from pancreatic cancer in the United States has risen from 2.9 per 100,000 in 1920 to 9 per 100,000 in 1970,57,58 an increase of over 300%. In our own state, the Connecticut Tumor Registry has shown an incidence of pancreatic cancer of 6.9 per 100,000 in men in 1935 to 1939,19 9.9 in 1968,20 and 12 Accepted for publication February 3, 1978. dence and any increased survival after therapy. In order to have a base for our clinical impressions, we reviewed the natural history, diagnosis and survivorship of patients with pancreatic cancer in the state of Connecticut. MATERIALS A N D METHODST o establish a convenient data base, we investigated the fate of 100 patients with histologically proven adenocarcinoma of the pancreas, and we selected the years 1960-1971 for our review. The criteria for selection were: 1) histologic proof of the lesion, and 2) operative or autopsy localization of the primary tumor in the pancreas. In some patients with a pancreatic mass, in whom the diagnosis depended upon lymph node or liver biopsy, the site of pancreatic cancer was substantiated at laparotomy or confirmed by autopsy. In order to find 100 records which fulfilled the foregoing criteria, 197 records of patients with a discharge diagnosis of pancreatic cancer had to be scrutinized. The implication of this will be discussed later. Follow-up, ranging from one month to eight years and eleven months, was possible in 97 of the 100 histologically proven cases. The findings in this series were then compared to published results in a thorough literature review.

After adjusting for calculations and duplications, the total number of 5-year survivors can hardly be more than 700-800.

Resections for pancreatic cancer have been performed for 65 years, with approximately 20,000 reported. A number of authors claim a 5-year survival rate of 30% to 58%. Review of the literature reveals only about 1,200 5-year survivors; however, 10 times as many individual resected survivors have been reported (in various countries), and nonresected survivors are overlooked. This high survival percentage is obtained by reducing the subset on which calculations are based and by using methods such as the Kaplan-Meier method, which produces higher figures as increasing numbers of patients are lost to follow-up. After adjustments, hardly more than 350 resected survivors could be found. Revision of statistical methods is urgently needed.

Objective. The incidence of pancreatic cancer is estimated to be 48,960 in 2015 in the US and projected to become the second and third leading causes of cancer-related deaths by 2030. The mean costs in 2015 may be assumed to be $79,800 per patient and for each resection $164,100. Attempt is made to evaluate the results over the last 80 years, the number of survivors, and the overall survival percentage. Methods. Altogether 1230 papers have been found which deal with resections and reveal survival information. Only 621 of these report 5-year survivors. Reservation about surgery was first expressed in 1964 and five-year survival of nonresected survivors is well documented. Results. The survival percentage depends not only on the number of survivors but also on the subset from which it is calculated. Since the 1980s the papers have mainly reported the number of resections and survival as actuarial percentages, with or without the actual number of survivors being reported. The actuarial percentage is on average 2.75 higher. Detailed information on the original group (TN), number of resections, and actual number of survivors is reported in only 10.6% of the papers. Repetition occurs when the patients from a certain year are reported several times from the same institution or include survivors from many institutions or countries. Each 5-year survivor may be reported several times. Conclusion. Assuming a 10% resection rate and correcting for repetitions and the life table percentage the overall actual survival rate is hardly more than 0.3%.