select search filters
briefings
roundups & rapid reactions
Fiona fox's blog
Scientists comment on a press release and analysis about cancer cases in the under 50s in England from the World Cancer Research Fund.
Prof Rhian Gabe, Professor of Biostatistics and Clinical Trials, Queen Mary University of London, said:
“This gives an important overall picture with respect to trends in incidence for different cancers and I agree that further epidemiological investigation needs to be prioritised and funded. Investigation of incidence of individual cancers by stage or other markers that indicate risk of progression could give clues as to how much of this might relate to increases in awareness, testing opportunities and improved detection methods and how much may be due to increase in causes and exposures, some of which that may be preventable. This, of course could vary by the different cancer types and in the UK, for example, there has been increased awareness and testing for prostate cancer generally and that might hold for men in their forties, especially perhaps for men at increased risk of prostate cancer. Further investigation could help us improve preventive strategies and model how this may impact future mortality rates.”
Dr Jason Oke, Lecturer in Medical Statistics and Cancer Research, Nuffield Department of Primary Care Health Sciences and Centre for Evidence-Based Medicine, University of Oxford, said:
“We should be cautious when interpreting the reported rise in cancer incidence among people under 50, particularly for prostate cancer. Although this may represent a genuine increase in clinically significant early-onset disease, it may also reflect greater diagnostic scrutiny in younger men. PSA testing in men aged 40-49 has increased significantly since 2000, from 3.3 per 1,000 person-years in 2000 to 21 per 1,000 person-years in 2017 (Collins et al), increasing the likelihood that prostate cancers will be detected earlier – and more indolent cancers that might never have become clinically apparent being diagnosed. Rising incidence, therefore, does not always imply a comparable rise in the underlying burden of clinically meaningful disease in this age-group. What appears to be a burgeoning epidemic of prostate cancer may, at least be an epidemic of testing.”
Dr Adam Jacobs, Retired Statistician, said:
“The researchers found that the recorded incidence of several types of cancer had become more common in people under 50 between the years of 2001 to 2023. However, it is not clear how much of this increase is due to a genuine increase in cancer incidence and how much is due to better detection through screening programmes or other means of detection. If this is a genuine rise in incidence, then the researchers are right to flag this as a genuine cause of concern, but if it’s simply because of better detection, there would be less cause for concern. It is hard to disentangle these effects from the data they present. It is also worth noting that the data they present focus only on those cancer types that became more common over the study period. They also investigated other types of cancer, some of which had shown significant decreases in recorded incidence, notably cancers of the oeophagus, lung, or bladder. It is noteworthy that those 3 cancers are strongly associated with smoking, and the decrease in incidence may reflect the decrease in population smoking rates over the study period.”
Prof Amy Berrington, Professor of Clinical Cancer Epidemiology, Institute of Cancer Research (ICR), said:
“We published a similar analysis on this topic earlier this year using the same dataset of cancer registrations in England. The approach is reasonable and very straightforward – they have just taken the cancer registration numbers per year and looked at the differences in diagnoses and rates between 2001 and 2023.
“They didn’t evaluate the changes for the over 50s and so this analysis is missing some important context – are these cancers also increasing in over 50s? And how do the increases in the over 50s compare to the under 50s?
“This comparison is important to guide the scientific research and also to put the increases in perspective for policy or public health planning.”
Prof Kevin McConway, Emeritus Professor of Applied Statistics, Open University, said:
“Unfortunately I have had very little time to look at this. Here are a few comments and questions that I have, though it would take me a long time to answer them properly.
“The table in the press release presents a partial and to some extent misleading picture of the increases. That’s because it presents (in its last column) only the percentage increase in the absolute numbers of new cases between 2001 and 2023. But this takes no account of the fact that the population of people aged under 50 increased over this period by somewhere around 10%, so if nothing else had changed, you’d expect the number of cases to have increased because of the population increase alone.
“It would be better to look at the percentage increases in the rate of new cases per 100,000 population, because that takes some account of the changes in population size. Generally those percentage increases are rather smaller than those stated in the last column of the table in the press release, though they are generally still large.
“I do not understand from the ‘methodology explainer’ how the rates of new cases per 100,000 were obtained. It may all be fine, but I can’t check without being able to access the data myself, and I cannot access it from the NHS source given in the data spreadsheet that was attached. I get into the data source, but at a later point it won’t let me have the data. This may be because I’m doing it wrong, or because I don’t have the right permissions to use the NHS data. Unfortunately I have no time to check that.
“However, the ‘methodology explainer’ says that these rates of new cases are age-specific incidence rates. The trouble is that one has to have a method of putting together the numbers of new cases at different ages that takes into account the fact that the pattern of different ages of people under 50 may have changed between 2001 and 2003. If, on average, people under 50 were older, that is, closer to 50, in 2023 than in 2001, one would expect increases in those rates of new cases even if nothing has really changed, simply because people close to 50 have always been more likely to have a cancer diagnosis than people a long way under 50. It’s of course quite possible that WCRF dealt with this properly, but I can’t see what they did exactly and have no more time to check it myself. I’m a statistician so need to have a good idea of how numbers are actually calculated.
“That said, there may well still be good evidence of increases in cancer rates in people under 50 for the cancers considered here – I’m simply being suitably sceptical, which is my job as a statistician. But an increase in the numbers of cancers recorded in cancer registries does not necessarily correspond well to how the number of cancers would have changed if we could somehow, miraculously, know how many cancers, diagnosed or undiagnosed people aged under 50 actually have in their bodies. A cancer is only recorded in a registry if it is diagnosed and if the diagnosis is properly recorded. Improvements in diagnosis methods would change the number of recorded cancers, even if the actual number of cancers in people’s bodies did not change. So actually using figures like these to consider what might have caused the increases in new cases is difficult, I’m afraid.
“I’m not being complacent and saying that we shouldn’t look further at why the rates of new cases appear to have gone up – only that to do so is a serious and major piece of work, that needs to be done. I do agree with the press release where it says “Understanding why these increases are happening will be key to providing improvements in prevention, screening and treatment.” But that work hasn’t yet been done, and is likely to be complicated. The press release makes this clear in several places.”
Prof Prabhakar Rajan, Professor of Urology and Robotic Surgery, Barts Cancer Institute, Queen Mary University of London, said:
“These data showing that prostate cancer diagnoses among men under 50 years have more than trebled since 2001 may seem alarming. Although this cancer is less common in younger men than in older age groups, the sustained increase in diagnoses is notable. The reasons are likely to be complex and multifactorial. Greater awareness, testing and changes in diagnostic practices may be identifying more cases, while longer-term changes in diet, body weight and physical activity, environmental exposures and genetic or tumour biology may also play a role.
“However, these data should be interpreted with caution: an increase in cases does not necessarily mean an increase in cancer risk, as changes in awareness, testing, access to healthcare and recording may also contribute. The analysis identifies important trends but cannot establish their causes, and further research is needed to understand what is driving these changes and how best to improve early detection and prevention.”
Rachel Richardson, Methods Support Unit Manager, The Cochrane Collaboration, said:
“These data from the World Cancer Research Fund highlight a concerning trend and it will be important to understand what is causing the rise in cases of certain cancers in younger people.
“However, there are a couple of things to bear in mind. The number of people under 50 who are diagnosed with cancer is still very low. The press release states that cases of prostate cancer have more than trebled since 2001: this is correct, but there were only 158 new cases in 2001 and 640 in 2023. There were also decreases in the number of cases of other cancers, for example oesophageal cancer.
“The second important factor is that we do not know what is driving this increase. There may be a genuine increase in the incidence of these cancers, but the figures may also be driven by increased screening and testing. For example, there have been several high-profile public awareness campaigns to encourage testing for (for example) prostate cancer.”
Press release title: ‘Cancer cases increasing in under 50s in England, with prostate cancer showing steepest rise, says charity’, from the World Cancer Research Fund International.
There is no full report or journal paper.
Declared interests
Prof Rhian Gabe: “None, except I am the Co-lead of TRANSFORM trial of prostate cancer screening.”
Dr Jason Oke: “No conflicts.”
Dr Adam Jacobs: “No conflicts of interest to declare.”
Prof Amy Berrington: “No COIs to declare.”
Prof Kevin McConway: “I have no conflicts of interest.”
Prof Prabhakar Rajan: “Prabhakar Rajan is a director and founder of healthXY Ltd, has received reimbursement for consultancy and advisory boards from Antev Ltd and Medtronic Ltd, speaker fees from Janssen-Cilag Ltd (Johnson & Johnson), and educational support from Angiodynamics UK Ltd, EDAP TMS S.A., HC 21 Healthcare Ltd, Medtronic Ltd, and Janssen-Cilag Ltd (Johnson & Johnson).”
Rachel Richardson: “No interests to declare.”
For all other experts, no reply to our request for DOIs was received.