Purpose of Study
The purpose of the study was to identify whether new cases of PSC with IBD (PSC-IBD) mirror new cases of IBD without PSC across England, observe how different the ‘patient journey’ is between diseases, and whether basic factors (such as age, sex, race) affect the need for a liver transplant, influence the risks of developing cancer, or affect the chances of dying for a particular reason.
Glossary
Colorectal cancer – bowel cancer
Hepatopancreatobiliary (HPB) – any cancer associated with the liver, bile ducts or pancreas
Colonic resection – bowel surgery
Colectomy – removal of the colon
What the study measured
This study looked at ‘clinical events’ found in the patient records. These include whether a person:
- died (and the reason for this, including whether or not it was related to PSC);
- had a liver transplant;
- was diagnosed with bowel cancer;
- had bowel surgery;
- was diagnosed with cancers of the bile duct (cholangiocarcinoma, CCA), pancreas, gallbladder, or liver;
- had their gallbladder removed, and the reasons for this.
Strengths
The paper looked at NHS health records of 284,560 people who were diagnosed with IBD in England over 10 years, of which 2,588 developed PSC. This is a very good timescale and a very good sample size; as it is estimated that approximately 620,000 people are currently living with IBD in the United Kingdom.
The paper also provides a balanced view of PSC, looking at patients around the country, not just those being treated at transplant centres. This is important, as people with more aggressive forms of PSC are more likely to be looked after in transplant hospitals, whereas older individuals may be less likely to attend due to other conditions.
Because of the unique healthcare system in England, the research team was able to look at detailed (albeit anonymised) data for each patient, right down to scan records and routine healthcare appointments. This means that they were able to analyse the data thoroughly without accessing any identifiable personal information.
Weaknesses
The paper only looks at PSC-IBD patients, not PSC patients who do not have IBD. This was because most people with PSC have IBD (estimated 70-80%), and the investigators want to be as certain about the diagnosis of both conditions as possible. Also, the aims of the study were to identify how different (or similar) PSC-IBD behaves to IBD without PSC.
The paper also excluded patients diagnosed below the age of 18 (due to restrictions in the way NHS data access is guarded), so its findings cannot be used to make judgements about ‘paediatric PSC’.
Similarly, laboratory blood test results and medications were not available to the investigators.
Key Findings
- New cases of IBD have broadly stayed the same across England over 10 years, but those for PSC-IBD appear to be rising.
- Areas in England that had the most new diagnoses of PSC-IBD were different to parts of the country with the most IBD.
- In people with IBD, the diagnosis of PSC was linked to an (approximately) 3-fold increased risk of death, colon cancer (2-3 fold), and a lower average age at bowel cancer diagnosis, compared to IBD patients. The risk of other events was also increased, including liver transplantation, cholangiocarcinoma (approx. 30-fold), gallbladder cancer (>5-fold), pancreatic cancer (5-fold), and the need for surgery to remove the colon or gallbladder.
- Black patients had roughly double the risk of needing a liver transplant or experiencing a PSC-related death compared to white
- Patients of male sex had an increased risk of needing liver transplantation or experiencing a PSC-related death compared to those of female sex (women had a 25% lower risk).
- If PSC-IBD was diagnosed in people younger than 40 years old, the risk of colorectal cancer was 4 times greater than patients with IBD alone. If PSC-IBD was diagnosed at 60 years old or higher, the risk of bowel cancer was very similar to people with IBD alone.
- Approximately 50% of all deaths in the PSC-IBD group were not related to PSC or liver disease.
- Liver-related events (transplantation or death from PSC) made up 75% of all clinical events recorded amongst people diagnosed with PSC below the age of 40; whereas in the group diagnosed above the age of 60 years old, only 31% of all events were related to PSC (most causes of death were not PSC related).
- The risks of death following a diagnosis of hepatopancreatobiliary cancer (bile duct cancer or pancreatic or liver cancer collectively) was lower in the group of patients who underwent annual scans (i.e. any combination of a yearly ultrasound or MRI or CT scan).
Other Points of Interest
- Nationwide data on what happens to PSC patients (i.e. cancers, transplants etc) is needed because patients studied at transplant centres can skew data, making the prognosis for people diagnosed with PSC appear worse than it may be.
- Routine surveillance for hepatopancreatobiliary cancers is not standardised across the UK, but annual imaging (e.g. MRCP, ultrasound, CT) is practiced at some centres. PSC increases the risk of these cancers, and this study highlights the importance of regular surveillance.
- Most commonly, bowel surgery was performed for non-cancer related reasons (probably active colitis).
- Most colorectal cancers and bowel surgeries happened before, or within the first year of PSC diagnosis.
- The majority of cholangiocarcinomas occurred within a year of PSC diagnosis.
- Cholangiocarcinoma was uncommon in people diagnosed with PSC below the age of 40 years, but rates increased in those of older age.
Take home messages
-
The fact that PSC diagnoses are going up in areas where IBD rates stay the same suggests that PSC-IBD is a different disease to IBD on its own.
-
PSC-related liver disease tends to be more aggressive in younger individuals, men, and Black people.
-
Annual imaging with a combination of either an ultrasound scan or MRI or CT scan of your liver and bile ducts may allow cancers being picked up earlier, and if diagnosed, a longer survival time following the cancer being identified.
Acknowledgements
Many thanks to Geraint Roberts