UK Pancreatic Cancer incidence statistics

This page presents pancreatic cancer incidence statistics including, age and sex, trends over time and histology. The ICD code for pancreatic cancer is ICD9 157, ICD10 C25.

Pancreatic cancer is the 11th most common cancer in the UK, with an average of 21 cases diagnosed every day.

The pancreas is situated in a deep location, flat against the back of the abdomen behind the stomach and small intestine. It is part of the digestive system and is a large gland that makes digestive juices and insulin ( Figure 1.1)

Download this chart (32KB)

 

The prognosis for people diagnosed with pancreatic cancer is poor, partly due to non-specific symptoms which leads to the disease being diagnosed at a late stage, and partly because of the location of the pancreas.

Pancreatic cancer incidence by age and sex

In 2006 there were 7,660 newly diagnosed cases of pancreatic cancer in the UK. Although there are a similar number of cases in males and females, the age-standardised rates are higher for males. ( Table 1.1) 1-4.

It has been estimated that the lifetime risk of developing pancreatic cancer is 1 in 86 for men and 1 in 86 for women in the UK. These were calculated on February 2009 using incidence and mortality data for 2001-2005 5

Download this table (41KB)

Figure 1.21-4 shows that the age-specific incidence rates increase with age, rates start to rise significantly in people aged 45 and over. Around three-quarters (74%) of pancreatic cancer cases occur in people aged 65 years or over.

Download this chart (57KB)

Pancreatic cancer incidence trends

The poor prognosis for pancreatic cancer means that trends in incidence and mortality follow a very similar pattern. Figure 1.3 shows the incidence and mortality rates for males and females from 1975 to 2007. 1-4,6,7,10.

Download this chart (42KB)

The age-standardised rates for men have been declining steadily since the late 1970s, whereas the rates for women have remained at a fairly constant level. The decrease for males partly reflects the decline in smoking prevalence, as smoking accounts for up to 20% of pancreatic cancer cases in the UK.

The pancreas cancer incidence trend for the UK is shown in Figure 1.4.

Download this chart (36KB)

Pancreatic cancer histology

Around 95% of pancreatic tumours are adenocarcinomas, originating from the exocrine (digestive enzyme-producing) part of the pancreas. Nearly all of these are ductal adenocarcinomas.

There are also three rarer types of exocrine pancreatic cancer, for which the treatment followed is more or less the same as for ductal adenocarcinoma. Endocrine tumours of the pancreas exist too, arising from the islets of Langerhans (which produce several hormones including insulin), but are rare.

UK Pancreatic Cancer mortality statistics

This section on pancreatic cancer contains information by age and sex, geographic variation and trends in mortality.

Pancreatic cancer is the fifth most common cause of cancer death in the UK, causing more than 7,700 deaths in 2007.

It is possible for mortality to exceed incidence for cancer sites such as the pancreas due to the very poor survival rates. Mortality follows incidence very closely so a drop in incidence may leave mortality rates higher as proportions of patients diagnosed in previous years also contribute to the mortality data.

Pancreatic cancer mortality by age and sex

The numbers of deaths and mortality rates for pancreatic cancer in the UK ( Table 2.11-,3) closely reflect the numbers of people being diagnosed.

Download this table (34.0KB)

Patterns by age and sex ( Figure 2.11-,3) are therefore very similar to those described for incidence.

Download this chart (36.0KB)

Geographic variation in pancreatic cancer mortality

Mortality rates in the European Union (EU) are consistently higher for males than females. Central and Eastern European countries have the highest mortality rates and Portugal and Cyprus have the lowest ( Figure 2.24).

Download this chart (44.5KB)

The UK ranks 21st in the EU for pancreatic cancer mortality and has a lower than average mortality rate for both males and females.

Worldwide there were an estimated 232,306 new cases of pancreatic cancer diagnosed in 2002, accounting for 2% of all new cancer cases. Figure 2.3 shows that the more developed countries have higher mortality rates for both men and women compared with the less developed countries4.

Download this chart (46.0KB)

Rates also vary by ethnic group, with the highest rates in the black population in the United States, where the annual incidence is about 12 per 100,000 population 5.

Trends in pancreatic cancer mortality

There has been little improvement in survival over the last few decades, which has remained exceptionally low. Therefore the mortality trend has matched the gradual decline seen in the incidence rates very closely. The fall in mortality rates is largely due to falling male rates, with a decrease of 20% over the last 30 years. Female rates have remained fairly stable over the same period, while rates in persons have fallen by 8%. ( Figure 2.4).

Download this chart (35.5KB)

Pancreatic Cancer survival statistics

This page contains information on survival rates for pancreatic cancer by sex, age, stage and comparisons with European survival rates.

Pancreatic cancer survival rates by sex

The proportion of people surviving pancreatic cancer is very low, and the length of time between diagnosis and death is typically short, at usually less than six months.

The most recent data for patients diagnosed in England and Wales show that around 13% of people with pancreatic cancer survive beyond 12 months after diagnosis and only 2-3% beyond five years ( Figure 3.11-,3).

Download this chart (24.0KB)

Improvements in palliative care have resulted in better short-term survival for both men and women; one-year relative survival rates have doubled since the early 1970s (see Figure 3.1 above).

Improvements include the use of endoscopic stent placement, effective pain relief and pancreatic enzyme supplementation. However, there is no evidence of an improvement in five-year relative survival rates for patients diagnosed with pancreatic cancer over the last 30 years.

Pancreatic cancer survival rates by age

As for many cancers, survival rates are higher in younger patients ( Figure 3.23). The one- and five-year survival rates for those aged under 50 are around 26% and 9% respectively, and 7% and 2% for those over 80.

Download this chart (19KB)

Pancreatic cancer survival rates by stage

Median survival following surgical resection for pancreatic cancer is of the order of 11-20 months. The five-year survival ranges from 7-25% 4,5. Patients with irresectable locally advanced disease (Stage III) have a median survival of 6-11 months 6. Patients who have metastatic disease have a median survival of only 2-6 months 7.

European comparisons of pancreatic cancer survival rates

The EUROCARE-3 study 8 provides  age-standardised relative survival rates across many European countries for patients diagnosed during 1991-1994. One-year survival rates for pancreatic cancer in England, Scotland and Wales are lower than the European average and this is statistically significant in England for men and in all three countries for women. The five-year survival rates are lower than the European average but these differences are not significant.

Pancreatic cancer risk factors

This page presents information on the risk factors for pancreatic cancer, including cigarette smoking and tobacco, pancreatitis , diabetes, bodyweight and physical activity, family history, non-steroidal anti-inflammatory drugs, diet and alcohol andother factors.

Cigarette smoking,tobacco and pancreatic cancer risk

Smoking is estimated to cause up to 20% of pancreatic cancers in the UK. 1. A British study reported that ex-smokers were 1.4 times more likely to die of pancreatic cancer compared with men who had never smoked. Current smokers of less than 25 cigarettes a day had 1.8 times the risk of death and smokers of 25 or more cigarettes a day were at 3.1 times the risk of never smokers 2.

Prospective studies have reported a positive trend with number of years of smoking among men but not women 3. Smokers are diagnosed on average 10 years younger than non-smokers 4. Studies have reported that risk returns to that of a never smoker from between 10-20 years after cessation 3,5. While the chemical cause is unclear, it is possible that N-nitroso compounds in tobacco are carried to the pancreas in the blood 6.

Regular cigar and/or pipe smokers have a 50% increased risk of pancreatic cancer 1. There is also evidence that risk increases with the use of snus although this may be limited to current smokers. 7,39

Pancreatitis and risk of pancreatic cancer

Chronic pancreatitis (CP), diagnosed in about one in 100,000 people in the UK each year, is a condition where the pancreas is inflamed. It is caused in the majority of cases by alcohol abuse. Risk estimates for pancreatic cancer in people suffering from CP range from 2 to 32 8-10.

People suffering from hereditary pancreatitis, which is characterised by onset before the age of 30 and a family history, have a risk ratio of at least 50 11,12, and an estimated lifetime risk of pancreatic cancer of 40-70% 13.

Diabetes and risk of pancreatic cancer

A meta-analysis reported a relative risk of 1.8 for pancreatic cancer in people with type II (non-insulin dependent) diabetes 14. A nested case-control study has recently reported results supporting this association, showing a statistically significant two-fold risk increase in people with biochemically diagnosed type II diabetes or with the highest fasting blood insulin levels 15.

Another meta-analysis showed that people with type I and early-onset diabetes have double the risk of pancreatic cancer 40

Bodyweight, physical activity and pancreatic cancer risk

Obese individuals have a significant increase in risk of pancreatic cancer of 19% compared with individuals with a healthy body mass index (BMI) 16. Evidence points to a slightly stronger risk increase in men than women 3,17-21.

Evidence is inconsistent about the effect of physical activity on pancreatic cancer risk. A number of large cohort studies have reported no effect 21,22 while other cohort and case-control studies report a protective effect in people doing the greatest amount of moderate physical activity 17,23-26. Evidence suggests that physical activity has no effect in people with a healthy BMI and obesity has no effect in people with the highest levels of physical activity 23.

Family history and risk of pancreatic cancer

Studies show that people with at least one first-degree relative diagnosed with pancreatic cancer have almost double the risk of someone without a family history 41.

Diet, alcohol and risk of pancreatic cancer

Overall, the contribution of dietary factors to pancreatic cancer risk is probably small. High dietary folate has been linked to a 51% reduced risk, but supplements have not been shown to have an effect. 45.

Chronic heavy alcohol consumption is a risk factor for pancreatitis, but evidence for an association with pancreatic cancer is inconsistent. Overall, research suggests an increased risk in heavy drinkers 27-29 , but no increased risk for people consuming up to 30g of alcohol a day 30.

Non-steriodal anti-inflammatory drugs and risk of pancreatic cancer

Use of NSAIDs probably has no effect on risk of pancreatic cancer 46.

Other risk factors for pancreatic cancer

Significant reduction in risk has been reported in people with a history of allergies and risk decreases with increasing number of allergies and severity of symptoms 31. The relevant allergies are those affecting the skin, such as eczema or hives and reaction to insect bites, hay fever and respiratory allergies other than asthma 32.

There is some evidence that a history of peptic ulcer increases the risk of pancreatic cancer, although since smoking is related to peptic ulcer development, there is the chance for confounding by this factor 33. It has been suggested that the bacteria that populate the stomach post-operatively cause formation of carcinogenic nitrosamines 34. An increased risk of pancreatic cancer has been shown in male smokers testing positive for Helicobacter pylori, although another study showed no association, and any mechanism is uncertain 35,42. One study has shown a reduced risk of pancreatic cancer in people taking statins 43.

Findings relating to a number of other medical conditions, including gallbladder disease, or cholecystectomy, and hypertension, are inconsistent 36,37, 3,38.

Pancreatic Cancer molecular biology and genetics

Pancreatic cancer can run in families; it is estimated that around 10% of cases are associated with an inherited gene, which raises the risk of developing cancer 1. Risk increases in relatives of those diagnosed before age 60 2,3 and with the number of affected relatives 1,4-,7.

Molecular stages in the development of pancreatic cancer

The molecular stages in the development and progression of pancreatic cancer are not fully understood but some of the key genetic alterations have been identified 8.

Many of these changes disturb normal cell cycle regulation and promote growth of cancer cells. Some of the earliest genetic alterations observed in cancer progression occur in the proto-oncogene, K-RAS. Mutations in K-RAS become more common as cancer progresses and are reported in almost 100% of cases. Genes that are thought to influence the later stages of disease progression include CDKN2A (which encodes two  tumour suppressors, ARF and INK4A), TP53 and DPC4/SMAD4. Faults in CDKN2A have been reported in 80-95% of cases, while faults in TP53 occur in around 50% of cases. Changes in TP53 are commonly missense mutations in the DNA binding domain. Loss of TP53 is associated with aneuploidy (having one or more chromosomes above or below the normal chromosome number), which is common in pancreatic cancer. Deletion of 18q21 containing the DPC4/SMAD4 gene occurs in around 30% of pancreatic cancers and is a marker for poor clinical outcome.

Pancreatic cancer and inherited risk

Around 10% of cases of pancreatic cancer are associated with an inherited faulty gene. A small proportion of these cases are linked to cancer syndromes that increase the risk of the disease 8,9. In a European study of 26 families with at least two first-degree relatives with pancreatic cancer, 19% of cases were found to be due to mutations in the BRCA2 gene 10. Inherited mutations in BRCA2 increase the risk of pancreatic cancer, but this risk is lower than that of developing breast and ovarian cancer. In sporadic cancers, BRCA2 mutations are observed in up to 10% of cases and are thought to occur late in tumour development 8,11.

A number of other cancer syndromes also carry an increased risk of pancreatic cancer, including Peutz-Jeghers ( LKB1/STK11), Lynch syndrome, (mutations in mismatch repair genes) and familial atypical multiple mole syndrome, or FAMM ( CDKN2A). Unusually, pancreatic cancer linked to these syndromes does not have an early onset but occurs at a similar age to sporadic cases. Pancreatic cancer can occur as a feature of other inherited diseases including hereditary pancreatitis, cystic fibrosis and ataxia-telangiectasia 9.

The majority of familial cases are not associated with a cancer syndrome or other inherited disease. Of these, many families exhibit an autosomal dominant pattern of inheritance and the susceptibility gene in one large USA pedigree has been localised to chromosome 4q32-34 12, although the gene itself has not been identified.

The future of pancreatic cancer diagnosis and  treatment

Progress in diagnosis and treatment will benefit from the results of large-scale genome analysis using microarrays to identify prognostic markers and possible drug targets. For example, high levels of the calcium binding protein, calcyclin, are associated with poor survival 13. Reduced levels of the protein BNIP3, from the Bcl-2 family, have been observed in 90% of cancer tissue samples and could provide a useful marker of sensitivity to the drug gemcitabine 14.

Pancreatic Cancer symptoms, diagnosis and treatment

This section contains information on the symptoms and diagnosis of pancreatic cancer and its treatment, both for early stage disease and advanced disease.

Symptoms and diagnosis of pancreatic cancer

While most patients present with obstructive jaundice, weight loss and back pain, pancreatic cancer often has vague presenting symptoms, developing at a late stage of disease 1-,5. Patients may also present with late onset diabetes mellitus and acute pancreatitis 6. While there is currently insufficient evidence to support the introduction of population screening for pancreatic cancer 7 there may be a place for secondary (targeted) screening for high risk groups in the future 8,9.

Carbohydrate antigen 19-9 (CA 19-9) levels are often raised in people with obstructive jaundice, chronic pancreatitis and pancreatic, biliary and gastrointestinal cancers 10,11. While people with early tumours often have normal levels of CA 19-9, it remains the most accurate and frequently used biomarker for pancreatic cancer. It is particularly useful in assessing treatment response in advanced cases 12,13 and identifying early recurrence following surgery 14.

While trans-abdominal ultrasound is often used as the first line of patient investigation and can detect most tumours larger than 2cm in size 15. Contrast enhanced computerised tomography (CE-CT scan) is the current gold standard and can achieve diagnostic rates of 97% 16. Magnetic resonance imaging (MRI) is similarly effective 17 but less easy to interpret. It is usually reserved for patients who cannot receive CT contrast. Endoluminal ultrasound (EUS) is valuable in imaging early pancreatic tumours as small as 2-3mm 18,19.

For diagnostic biopsy, percutaneous fine needle aspiration (FNA) has good sensitivity and specificity 20 but there are concerns about its role in intra-peritoneal and needle tract seeding 21. The incidence of peritoneal carcinomatosis has been shown to be less following EUS guided biopsy 22. Laparoscopy can be used to assess patients prior to surgery and also to obtain biopsy material. Laparoscopic ultrasound (LUS) enables intra-operative scanning of the liver and pancreas, correctly predicting resectability in >90% of cases. However, recent improvements in multislice CT scanning mean that diagnostic laparoscopy is now used less 23.

Major pancreatic centres use a combination of multislice CT with EUS or Endoscopic Retrograde Cholangio-Pancreatography (ERCP) or Magnetic Resonant Cholangio-Pancreatography (MRCP) and selective use of laparoscopy combined with LUS to stage tumours. The pathological staging of pancreatic ductal adenocarcinoma is based on the 2002 International Union Against Cancer (UICC) tumour node metastasis (TNM) classification 24.

Treatment of pancreatic cancer

Cancer can occur in the head, the body, or the tail of the pancreas. Around 70% of tumours are in the head of the pancreas and these are often easier to remove.

The only potentially curative treatment for pancreatic cancer is surgery. Patients with liver or peritoneal metastases and distant lymph node metastases are unlikely to be considered for resection 25.

Treatment of early stage pancreatic cancer

The classical Kausch-Whipple procedure and the pylorus-preserving procedure are the most common approaches for cancer of the head of pancreas. Tumours of the body and tail require left-sided resections. Total pancreatectomy is not routinely performed for cancer of the pancreatic head, but may be used for more widespread disease or intra-ductal papillary mucinous neoplasms. The majority of patients develop disease recurrence within 1 to 2 years after resection 26-,28.

While evidence for benefits of adjuvant therapy has been lacking, chemoradiotherapy and follow-on chemotherapy have been used commonly in America for a number of years 29-,31. Few randomised studies have looked at these treatments and, with the exception of ESPAC-1 (a European randomised controlled trial to assess the roles of adjuvant chemotherapy and adjuvant chemoradiation in resectable pancreatic cancer), most studies have been underpowered. The ESPAC-1 trial showed a survival advantage for adjuvant chemotherapy but not adjuvant chemoradiotherapy 32. Early results from a randomised trial reported an almost doubling of median disease free survival for patients receiving adjuvant gemcitabine compared with surgery alone 33.

The ESPAC-3 trial is comparing gemcitabine and 5-Flouroracil & Folinic Acid (5FU/FA). A recent meta-analysis of adjuvant therapy concluded that chemotherapy is an effective adjuvant therapy for pancreatic cancer, with a survival advantage for chemoradiotherapy in patients with positive margins 34.

Treatment of advanced pancreatic cancer

The majority of patients with pancreatic cancer present with inoperable disease. Symptom control and quality of life are extremely important in these patients as well as therapies aimed at extending survival. The main symptoms which require palliation are pain, obstructive jaundice and duodenal obstruction.

Randomised trials have demonstrated a survival benefit in patients who receive chemotherapy 35-,38. While 5FU has been the most widely used chemotherapy drug in patients with advanced pancreatic cancer, the improved survival and clinical benefit demonstrated for gemcitabine 39 has led to its adoption as standard first-line treatment. With the exception of erlotinib 40, no survival advantage has been demonstrated for gemcitabine in combination with a variety of cytotoxic and novel agents. 41-,44

Capecitabine has shown encouraging results in phase II trials 45. The Cancer Research UK GEM-CAP phase III trial compared capecitabine and gemcitabine with gemcitabine alone and the results are expected to change clinical practice. They show a significant improvement in overall survival with the addition of capecitabine to gemcitabine with acceptable levels of toxicity.

Benefits for combination therapy compared to chemoradiotherapy or chemotherapy alone are not widely established. Several small randomised studies have shown slight or no improvements in median survival with chemoradiotherapy plus follow-on chemotherapy vs. chemotherapy alone 30,46 and results from a number of small randomised studies of radiotherapy vs. radiotherapy plus follow-on chemotherapy have shown inconsistent results 47-,49.