The incidence of cancer of the pancreas has increased steadily over the past 25 years. In males it is the fourth commonest cancer causing death, exceeded only by cancers of the lung, colon and rectum, and prostate. In females it is the fifth commonest cause of death, with only cancers of the breast, colorectum, lung, and ovary/uterus being more frequent. The incidence is higher in males, with a sex ratio of two males to each female; peak incidence occurs in the fifth through seventh decade.

The overall five-year survival rate is less than 3%, and most patients who develop carcinoma of the pancreas die within six months of diagnosis. The poor prognosis in this condition is secondary to the inability to diagnose the carcinoma at an early stage. When symptoms present, the tumor is far advanced and often has metastasized to regional lymph nodes and to adjacent and distant organs, as shown in Table 7.

Ductal cell adenocarcinoma accounts for 90% of pancreatic tumors. Approximately 5% of pancreatic carcinomas are of islet cell origin; the rest consist of cystadenocarcinoma, giant cell carcinoma and epidermoid carcinoma. The head of the pancreas is the commonest site of involvement, accounting for 70% of the cases, whereas the body and tail account for 20% and 10% of the cases, respectively.

Several etiological agents have been invoked in the pathogenesis of pancreatic carcinoma, although most of the studies have not yielded consistent results. Epidemiologically, long-term cigarette smoking is a well-established risk factor. The mechanism causing carcinoma is not proven and has been explained on the basis of the presence of a carcinogen in tobacco smoke. High-fat or high-protein diets tend to stimulate CCK release from the duodenum, which in turn can cause pancreatic hypertrophy and may predispose to carcinoma, although the evidence is not convincing. Diabetics are at twice the risk of developing carcinoma of pancreas as the general population. The mechanism of this is not known. There is no evidence to suggest that alcoholic chronic pancreatitis predisposes to carcinoma. A recent study has shown a four- to five-fold increase in pancreatic carcinoma in individuals exposed to DDT (dichlorodiphenyltrichloroethane).

Some epidemiological studies have suggested an increased rate of pancreatic carcinoma in patients who drank chlorinated water; this remains to be proven. Genetic defects have also been invoked, such as oncogenes and 53 p tumor suppressor mutation, in addition to some evidence suggesting an increase in epidermal growth factor receptor activity.

Ultrasonography is the procedure of choice for detecting pancreatic cancer. Its usefulness is dependent on the examiner's expertise (Figure 8). Examination may be less than optimal in the presence of increased bowel gas. The sensitivity of this test in pancreatic cancer is reported to be 76-94%, with a specificity of 96%. Once a lesion is detected, a guided biopsy may be helpful in establishing the diagnosis. When obstructive jaundice is present, ultrasound may reveal the presence of hepatic lesions or obstruction of the biliary tree. This procedure is simple and involves no radiation exposure.

Computerized tomography (CT) is more accurate and gives more information than ultrasonography (Figure 9). However, it is also more expensive and involves low radiation exposure; also, small centers may not have the equipment. With this technique, bowel gas does not interfere with the visualization of the pancreas. There is better identification of any retroperitoneal invasion. A guided biopsy of the lesion is also possible. However, small early lesions (especially in the head of the pancreas) may be missed on CT scan; hence, its usefulness in early diagnosis is not clear.

When there is a clinical suspicion of a pancreatic lesion and the ultrasound or CT scan is normal, an ERCP is helpful. It has the advantage of combining gastroduodenoscopy, cholangiography and pancreatography. The papilla may also be examined and cytologic sampling may be obtained. When obstruction is present, therapeutic drainage via stents may be attempted. Angiography is no longer used for diagnosing pancreatic carcinoma, but is still useful to evaluate patients who have known carcinoma for resectability, outlining vascular anatomy. Newer diagnostic tools such as endoscopic ultrasound may further improve selection of patients who might benefit from curative surgery. Magnetic resonance imaging has no apparent advantage over CT.