Although initially considered rare tumours, recent data indicates that the incidence of NETs has increased exponentially over the last 4 decades and they are as common as Myeloma, Testicular Cancer, and Hodgkin’s Lymphoma. In terms of prevalence, NETs represent the second commonest gastrointestinal malignancy after colorectal cancer. In fact, this graph and other reports below indicate the rate of incidence is potentially rising faster than any other Cancer on the planet. The rise is mainly attributed to lung, small intestine, and rectal NETs. However, since the World Health Organisation’s revised classification of Neuroendocrine Neoplasms in 2010, there would be a significant increase if this data exercise was run again. This is due to the abandonment of the division between benign and malignant NETs following the 2010 declaration that all NETs now have malignant potential and are graded accordingly. The SEER data compiled below did not take into account benign NETs.
Incidence and Prevalence
Before I continue, it’s important to understand the difference between incidence and prevalence. In the crudest of terms, incidence is the number of new cases of a disease being diagnosed (normally aligned to a specific quota of the population per year, generally 100,000). Prevalence normally indicates an amount of people living at any one time with a disease. It’s also important to note that different nations or groups of nations classify ‘rare’ in different ways – not really helpful when looking at worldwide statistics.
So why the increase? I suspect the reasons include (but are not limited to), more awareness (population and medical staff), better detection techniques and probably more accurate reporting systems, at least in USA, Norway, Canada and now in the UK i.e. a mixture of underdiagnoses and misreporting. The Canadian study is important as it also noted the proportion of metastases at presentation decreased from 29% to 13%. This is the first study that suggests an increased incidence of NETs may be due to an increased (and earlier?) detection. This has the knock on effect of increasing prevalence as most NET Cancer patients will normally live for longer periods. Add to this the plethora of better treatments available today, you have a highly prevalent cancer.
US SEER – The Trigger and Turning Point
In the largest study of its kind, well-known Neuroendocrine Cancer expert James C. Yao researched the Surveillance, Epidemiology and End Results (SEER) database. His team studied 35,825 cases of Neuroendocrine Cancers in the United States covering data between 1973 and 2004. The report concluded that in 2004 there were 5.25 new cases of NETs per 100,000 people, compared with 1.09 per 100,000 in 1973 . This is in contrast to the overall incidence of malignancies, which has remained relatively constant since 1992 (see the yellow line on the graph). The study also pointed out that due to increased survival durations over time, NETs are more prevalent than previously reported. This is an astonishing set of statistics – particularly as they are based on data which is now 12 years old. If you analyse the NET data for 1994 (10 years before the end of the study period), you will see an incidence rate of approx 3.25/100,000. In 2004, the incidence rate had risen to 5.25/100,000. Although not an exact science, it does suggest the potential incidence rate at 2014 (10 years after the study period) might possibly have climbed well beyond 6/100,000 and even further if the same rate of increase displayed by the study had continued. You can see evidence of this extrapolation in the diagram below which has been extracted from cited article 2a. This study also confirmed a prevalence of 103,000 NET patients as at 2004. As this is regarded as the most accurate NET statistic ever produced, it is interesting to note that was at a time when the prognostics for NET were not as good as they are today indicating there must be a very significant increase by 2017. Moreover, this was prior to the WHO 2010 reclassification of NETs so more diagnoses will be counted today than were not counted in 2004. See below to see the significance of this figure (see section ‘Do the math’).
Meanwhile in Norway ……
Data from the Norwegian Registry of Cancer showed a similar incidence of Neuroendocrine Cancers with a 72% increase between 2000 and 2004 compared with 1993–1997 . Also in Norway, an article published in 2015 entitled “Epidemiology and classification of gastroenteropancreatic neuroendocrine neoplasms using current coding criteria” indicated a high crude incidence of GEP-NEN, at 5·83 per 100 000 inhabitants over the period 2003-2013 (adjusting to 7.64 for Europe in 2013 – see diagram below extracted from cited article 2a). It was also noted together with the statement “….a significant increasing trend over time”. [2a] Citation [2b]
Meanwhile in Canada …….
CNETs have highlighted an article published in the magazine ‘Cancer’, February 15, 2015, showing that the incidence of Neuroendocrine Tumours has markedly increased in Canada over the course of 15 years (1994-2009). The results showed that the incidence of Neuroendocrine Tumours has increased from 2.48 to 5.86 per 100,000 per year.  
Meanwhile in UK …….
The latest figures from Public Health England (PHE) indicate the incidence of NETs has risen to 8/100,000 using the latest International Classification of Diseases for Oncology (ICD-O) methodology version 3 – ICD-O-3 (i.e. not rare). That means a new NET diagnosis every 2 hours. You can see a summary of the report here which is based on 8,726 cases: NEW: Public Health England release new incidence data for Neuroendocrine Cancer
Meanwhile in New Zealand …….
Do the Math
So it seems that Neuroendocrine Cancer is not only the fastest growing cancer in incidence terms but as a group of cancers, it may no longer be considered rare in statistical terms if you roughly extrapolate the US SEER data graph above to 2015 and overlay the other studies quoted. In fact, this picture below from Novartis US does have an extrapolation via a regression analysis of the SEER figures as at 2004 taking it up to what it might have been at 2013. You can see clearly from this chart that it goes way beyond the incidence threshold for rare status and there are a further 3 years to add to that. Although it says ‘Carcinoid’ I suspect they mean all NETs and this is part of the nomenclature confusion that still pervades the internet. Even if they only meant so-called ‘Carcinoid’ tumours, this would increase the incidence even further than shown below.
Whilst reporting has been improved, it is clearly still not 100% accurate. Therefore, even the figures above may be understated due to an incorrect cause of death reporting and incorrect diagnosis/recording of the wrong cancers (e.g. pNETs recorded as Pancreatic Cancer, Lung NETs recorded as Lung Cancer, etc). This is certainly still happening in UK and I suspect in most other countries. Add to that the regular reports of Neuroendocrine Tumours being found during autopsies and you have the potential for an even further unrecorded increase had these been found prior to death.
The issue is also complicated by the method used in USA for naming a disease ‘rare’. Rather than use incidence rates, the USA uses the number of people living with the disease at any one time (i.e. essentially the prevalence). This is currently 200,000 as a threshold – anything below that is considered rare.
The top graphic is an extract from a slideshow published by Dana Farber. You will see they cite the “greater than 100,000” figure; and they also estimate an annual incidence of 16,000 (a very conservative figure most likely based on the 5/100,000 incidence rate from 2004).
When I first started researching NETs back in 2010, the US figure (which varies from source to source) was around 125-150,000. Many people still quote this today. However, at ENETS 2017 Dr James Yao provided an updated figure of 175,000 – I believe this is still conservative. It was Dr Yao’s work on the SEER database which confirmed the four-fould increase in incidence as at 2004.
However, you will also see that Dana Farber is estimating more than 200,000 people are as yet undiagnosed. Even if that were 50% accurate, it would put the current prevalence figure in US over 200,000 – ergo NOT RARE.
Let’s cut to the chase – NETs are not rare, they are just less common
Are we shouting loud enough about this? I don’t think so. ‘Rare’ is very frequently used within our community almost to the point of being a status symbol leading to denial. However, based on these figures, this looks like an outdated approach along with its associated icons. Statistically, rare cancers don’t normally do well in comparison to the more common types and they certainly don’t attract the same resources and research.
“A neoplasm on the rise. More prevalent than you may think. Incidence increased dramatically during past 3 decades” (Novartis)
“it’s less rare than we used to think. It’s more malignant than we previously thought” (Dr Richard Warner)
“…..it is one of the most rapidly increasing cancers in the U.S. There has been a 500-percent increase in the last 30 years” (Dr Edward Wolin)
“Estimated more than 200,000 undiagnosed cases in the US” (Dana Farber)
“Neuroendocrine Cancer – it’s not rare, it’s less common. We should be talking more about this” (Ronny Allan)
Thanks for reading
I’m also active on Facebook. Like my page for even more news.
Tweeters ……have you retweeted this yet?