Role of serum markers and or various clinical parameters in the diagnosis of pancreatic carcinoma

This study was performed to ascertain the role of serum markers and simple clinical data in detecting pancreatic cancer and in distinguishing this malignancy from chronic pancreatitis and other gastrointestinal diseases. Serum CA 19-9, tissue polypeptide antigen and carcinoembryonic antigen were measured in 38 control subjects, 37 patients with pancreatic cancer, 39 with chronic pancreatitis and 44 with extra-pancreatic diseases mainly of gastrointestinal origin. Clinical data recorded included age, sex, presence of pancreatic calcifications, weight loss, pain, jaundice, alcohol abuse, diabetes mellitus. Serum markers gave a correct allocation of the subjects in 48.1% of the cases with pancreatic cancer patients correctly predicted in 62.2%. Clinical data correctly diagnosed 74.2% of subjects. Chronic pancreatitis was identified in 84.6% of the cases and pancreatic cancer in 64.9%. The first clinical variables selected were pain and age. The addition of serum markers to clinical data did not enhance accuracy of the results. We conclude that the diagnosis of chronic pancreatic diseases should first be suspected on the basis of accurately recorded simple clinical data; serum markers seem to be only occasionally useful. Since indicative clinical data and serum markers become positive in the advanced phases of pancreatic cancer, early diagnosis of this malignancy still remains an objective to reach.