Rates are climbing in under 50s. Researchers are still seeking reasons, but we do know more about risk factors.
Young onset pancreatic cancer rates are rising, especially in the Western world, and we don’t entirely know why. Lifestyle is definitely a factor, and the usual suspects – smoking and tobacco – are still implicated, but they’re not alone.
“Yes, there is an increase; but there is no cause for alarm. None of this data should cause paranoia or anxiety. But it should cause awareness,” Associate Professor George Barreto, a researcher and gastrointestinal, HPB and liver transplant surgeon at Flinders Medical Centre in South Australia, tells TMR.
“It’s important to be armed with the information because the eye doesn’t see what the mind does not expect to see. The aim of this data that is being shared is to be more vigilant and to consider, is it a possibility?”
Professor Barreto is lead author on a paper showing that incidence, death and disability adjusted life years increased significantly from 1990 to 2019 in people under the age of 50 without a family history of pancreatic cancer, up 30%, 25% and 11%, respectively. The rates were higher for males, highest in Eastern Europe, and lowest in Central Sub-Saharan Africa. Incidence went down in a few places too – notably Sweden and Finland.
Incidence, death and disability adjusted life years related to young onset pancreatic cancer increased with age and sociodemographic index. However, the authors note that data, let alone of high quality, is not uniformly available across the globe, especially in low income areas, which might explain some but not all of the disparity in numbers between economically disparate regions.
Late onset pancreatic cancer (in people aged 50 and over) still occurs at twice the rate of young onset disease, but the rates of change in incidence, death and disability adjusted life years are expected to be on par.
Just under 37,000 people were diagnosed with young onset pancreatic cancer globally in 2019. Most cases of pancreatic cancer, of any kind, are fatal, with the number of deaths in a year almost the same as the number of new diagnoses (94%). The five-year survival rate is just 11%. In those patients diagnosed in time to be operated on, the late onset overall five-year survival rate is still only 27% and an even lower 18% in young onset disease, which has a different, more aggressive tumour biology to late-onset disease (over the age of 50). Delayed presentation is also linked to worse outcomes in this age group.
Young onset pancreatic cancer has different molecular characteristics to older onset disease, leading the authors “to question the use of traditional treatment algorithms in patients with YOPC,” the paper says.
Risk factors – the usual suspects?
In 2019, smoking and tobacco use were still the leading risk factors, but their impact has been decreasing since 1990, especially in females. Meanwhile BMI and fasting plasma glucose are becoming more significant than they used to be, particularly in high income countries.
The increased risk associated with being overweight is still relatively small, says Professor Rachel Neale, a senior group leader at the QIMR Berghofer Medical Research Institute who wrote the current guidelines (2020) on pancreatic cancer decision support and is in the process of updating them for Cancer Australia.
However, Type 2 diabetes or increased blood glucose increase the risk by around two-fold, she tells TMR.
“Diabetes and pancreatic cancer have a complex relationship,” says Professor Neale.
“A small proportion of cases of diabetes, about 0.5-1% is actually caused by an underlying pancreatic cancer. We’re doing a lot of work on this at the moment – how do we distinguish cases of new onset diabetes caused by underlying pancreatic cancer from a standard Type 2 diabetes situation?”
A few models for have been published to try and narrow down the field. One, developed by Professor Neale’s team, uses a patient’s medication data, diabetes severity and age at diagnosis to predict the risk of a female patient with new onset Type 2 diabetes having pancreatic cancer. Another model uses changes in weight and blood glucose levels and age at onset of diabetes.
Testing and screening
Despite the significant rate of change, the experts say population level screening is not the answer because pancreatic cancer is so rare.
“Screening is not necessarily going to reduce mortality. The ovarian cancer trial in the UK is a very sobering example, where they screened at large scale population level and did not find a reduction in mortality from ovarian cancer. Part of the problem is that ovarian and pancreatic are both very fast growing very, very aggressive tumours. And screening programs, by definition, tend to pick up the slower growing, less aggressive tumours.”
Currently the only mechanism available for investigating the possibility of pancreatic cancer is imaging.
“If we send people with diabetes for testing, we’re sending 99+ people to find one case. We don’t know the economics of that,” says Professor Neale.
The other problem with imaging is that it finds things that may not ever cause harm, but it sets people on a path of invasive testing that may itself be harmful, she says.
“This is particularly problematic for the pancreas,” says Professor Neale.
Further investigations required MRI or endoscopic ultrasound, where the system was already overwhelmed, or even a tissue acquisition. This can cause acute pancreatitis, a condition that could itself be fatal.
Any hope of preventing late presentation of young onset pancreatic cancer lies in having a much better blood test, says Professor Neale.
“But even if it was a perfect blood test that was 100% sensitive and specific, we still would not use that to screen the entire population, because pancreatic cancer is too rare. But we may end up using it in subgroups of the population.”
Using the tumour marker CA 19-9 to triage people for imaging is currently controversial and not recommended, because it can be elevated in people who don’t have pancreatic cancer.
Awareness of the symptoms is still the best option available she says.
“If you had someone who had new onset diabetes, who also happened to have upper abdominal pain that radiated through to the back epigastric pain, then you would have a higher index of suspicion. If they also had a family history, you would have a markedly high index of suspicion. Diabetes that does not respond to treatment would be something that might make you think twice, particularly for someone who doesn’t look like they should have diabetes.”
New recommendations for older people will be released by Cancer Australia in the next 12 months.
“In people with 60 or over with new onset diabetes where the diabetes didn’t respond to management despite compliance with medication would raise the index of suspicion about somebody having a potential pancreatic cancer.
“And in those people, it may be appropriate to discuss with the patient whether or not they would like to have investigations of their pancreas, noting that it may actually have untoward sequalae. In these people the risk of pancreatic cancer will still be low, and you might find something not worth finding, but a shared decision-making approach with those patients would be appropriate.”
