Receiving any cancer diagnosis is depressing, but a diagnosis of cancer of the pancreas is one of the most depressing of all. Its most common form, adenocarcinoma of the pancreatic duct has the lowest survival rate of any of the common tumours, with five-year survival rates of only 2-3%. Furthermore, this very poor prognosis has scarcely changed since the 1970s, over a period in which survival rates for some other tumours have improved dramatically.

The treatment of this most challenging of tumours was the chosen topic for the 2011 Stevens Lecture at the Royal Society of Medicine (RSM). The RSM is a membership-based general medical society based in London, and one of the largest accredited medical education providers in the UK. This annual lecture was endowed as a means of building connections between the medical profession and the general pubic; it provides an accessible introduction to a medical topic of current interest. It is often given by a senior clinician, but not always; previous lecturers have included MPs, peers, the Princess Royal and environmentalist Sir Crispin Tickell. The 2011 lecturer, however, was an extremely distinguished doctor: Professor Robin Williamson, a consultant surgeon at Hammersmith Hospital, London, UK with a specialist interest in hepatopancreatobiliary (HPB) surgery, and also the immediate past president of the RSM. He was introduced by his successor, RSM President Professor Parveen Kumar.

Professor Williamson began his lecture by introducing the general features of the disease that he has spent his long career fighting, and explaining some of the reasons for its poor prognosis. Like many cancers, its initial symptoms may be diffuse and it is often only diagnosed when it has spread beyond the original organ; there are no useful markers for screening and, even if there were, there is no obvious target population to screen. Although people with both Type I and Type II diabetes have a pancreatic cancer risk about twice as high as that of people without the disease, this would be an enormous population to screen and could only ever identify a few cases. Furthermore, this is also often an aggressive tumour, and one that arises close to important blood vessels, metastasises quickly and presents a complex challenge to surgeons.

Quoting Oscar Wilde – "I can resist anything except temptation" – Williamson explained that there are two temptations for any lecturer required to speak about a particularly intractable disease: to discuss similar conditions with a better prognosis, and to concentrate on the most promising subset of cases. As a surgeon, he was almost obliged to give in to the latter temptation, as only 20% of pancreatic cancer cases are surgically resectable and these are almost by definition the ones with the best prognosis. Before discussing treatment, however, he briefly covered other malignant and non-malignant diseases with similarities to adenocarcinoma of the pancreas. Other, rarer tumours that arise in the pancreas, including neuro-endocrine and cystic tumours, present with similar symptoms and can be mistaken for pancreatic carcinoma, but have rather better – although not good – prognoses. Chronic pancreatitis can also be mistaken for malignant disease.

In about three-quarters of cases, adenocarcinoma of the pancreas arises within the head of the organ. These tumours fairly rapidly obstruct the bile ducts, and patients present with painless obstructive jaundice and a palpable gallbladder. When the cancer arises in the neck, body or tail of the pancreas the symptoms – pain in the abdomen and back, weight loss and loss of appetite – are slower to develop, and diagnosis is typically made many months after the first symptoms appear. The tumours with the best prognosis tend to be those that develop in the pancreas head and close to the bile duct, so jaundice will develop before the cancer has spread.

Williamson went on to describe what he referred to as "para-neoplastic effects": physical or psychological symptoms that arise only indirectly from a pancreatic tumour. These include cachexia (wasting); glucose intolerance or diabetes; venous thrombosis, which is common in cancer patients and particularly so in those with pancreatic disease; and, most interestingly, depression. Anecdotally, depression and anxiety are thought to be particularly closely associated with pancreatic cancer, possibly due to its "evil reputation". However, statistical analysis of the association between these symptoms and the complete range of gastro-intestinal cancers has revealed only a small increase that can probably be explained by the fact that a larger proportion of pancreatic cancer patients will have advanced disease.

Patients presenting with symptoms consistent with pancreatic cancer are investigated using liver function tests, ultrasound and CT scans, and endoscopic ultrasound and biopsies. A cholangiogram – an image of the bile duct using X-ray – is usually performed to delineate the level of bile-duct stricture, and it may be appropriate to decompress the obstructed biliary tree at the same time by passing a fine catheter. A biopsy is also only essential if surgery is not performed. Williamson told a story of one patient whose chronic pancreatitis was misdiagnosed as inoperable pancreatic cancer and who spent his life savings on a world cruise, recovered, and sued his doctors for bankruptcy on his return.

Once a diagnosis of adenocarcinoma of the pancreas is confirmed, the first step in treatment is to assess whether the tumour can be resected, as this provides the only chance of a cure. A tumour will generally be resectable if it has not spread to the liver and if the portal vein is clear. Borderline cases can be assessed using laparoscopy, which is thorough and accurate but invasive or by endoscopic ultrasound to show the all-important relationship to the portal vein. Tumours not considered suitable for surgery will be treated with drugs, most often gemcitabine either alone or in combination with other drugs. Just occasionally, chemotherapy will shrink a tumour to the extent that surgery becomes possible.

The first significant development in modern surgery for pancreatic cancer was the Whipple procedure (or pancreatoduodenectomy), introduced in 1935, in which the head of the pancreas is removed together with the gall bladder, the common bile duct, the duodenum and part of the stomach. During the last 30 years, a modification termed pylorus-preserving pancreatoduodenectomy (PPPP) has become increasingly popular. The main advantage of this technique is that the retained stomach allows better postoperative nutrition. Randomised trials have concluded that there is little difference in long term prognosis between the two procedures, and that, not surprisingly, outcomes improve in specialist centres where more operations are performed.

Williamson concluded his lecture by returning to his title, presenting four challenges that still remain in dealing with this tumour and turning them round into hopes for the future. Ongoing research is beginning to address the challenges of our limited knowledge of the molecular development of pancreatic cancer, the need for earlier diagnosis and safer surgery and the short-term nature of most "cures". It may take twenty or thirty years, but with increased knowledge of the molecular biology and genetics of this disease, biomarkers and screening techniques, improved surgical techniques and better chemotherapy drugs, Williamson is hopeful that the prospects for patients with this intractable disease should improve markedly.