By Cancer Rose, 19/10/2025
A rumor has taken hold, greatly exaggerated by a media little caring about checking the facts: cancers are “exploding” and those of the pancreas are affecting young people and have “tripled”.
A study published in BMJ Oncology in 2023[1] caused considerable controversy and was widely covered by the press without much nuance.
Several previous studies [2] suggested that the incidence (number of new cases of cancer occurring in a population within a year) of cancers of various organs diagnosed in adults aged 50 years had increased in many parts of the world since the 1990s. Risk factors were little studied.
According to this study, since 1990, the incidence (number of new cases) and the number of deaths from so-called early cancers* have increased significantly worldwide. Early-stage cancers of the breast, trachea, bronchi, and lung, stomach, and colorectal cancers had the highest mortality rates and burden on the population in 2019.
Countries with a medium to high socio-demographic index and people aged 40 to 49 were particularly affected, according to the authors. Dietary risk factors (diet rich in red meat, low in fruit, high in sodium, and low in milk, etc.), alcohol consumption, smoking, and a sedentary lifestyle are the main risk factors highlighted for these findings.
* The so-called “early-onset cancers ” referred to in these studies refer to cancers occurring in people aged 50, people considered young compared to the average age of onset of most cancers. In fact, 47% of cancer cases occur in men over 70, 30% over 75, and 14% in men over 80. This proportion is 42% in women over 70, 30% over 75, and 17% in women over 80.
The authors of the BMJ study suggested that further prospective cohort studies on the life course are needed to explore the causes of cancer, while encouraging healthy lifestyles that could reduce the burden of early-onset cancer (around the age of 50).
Detailed examination, and nuances…
Some precautions should be taken.
The overall figure for cancer cases is one thing, but it does not take into account two key demographic factors: aging and population growth. A third factor is screening, which increases detections and, therefore, de facto, the incidence figure.
If a population is growing in size and aging, over 30 years of data analysis, it is normal to see a proportionally increased raw number of cancers.
The correct measure is the incidence rate adjusted for demographic factors per 100,000 people.
Below is Figure 2 from the BMJ study, which shows the global incidence and mortality rates of 29 early cancers in 1990 and 2019, by sex, as well as the DALY indicator, i.e., life expectancy adjusted for disability induced by the disease (which we will return to later).
As can easily be seen in this graph from the BMJ study, the cancers whose incidence has literally and truly “exploded” are the cancers…screened. But not pancreatic cancer.
If we refer to the data from NIH (National Cancer Institute): “Using statistical models for analysis, age-adjusted rates for new cases of pancreatic cancer increased by an average of 0.9% each year in 2013–2022. Age-adjusted mortality rates were stable in 2014–2023.” So there is certainly a slight increase even when adjusting for age, but it is far from an explosion and a “tripling”.
Let’s take the case of breast cancer. Its incidence has increased considerably with the expansion of mammography machines over the decades. However, while it remains a major public health problem, causing the deaths of nearly 12,000 women each year, its mortality rate, after adjusting for population aging, remains roughly the same as it was in 1960. Example in France :
Illustration taken from the book ‘Breast Cancer Screening: The Great Illusion,’ B. Duperray, published by T. Souccar
The incidence of breast cancer is indeed skyrocketing, it can be said, due to unrestrained screening activity, without any reduction in mortality. This shows that in this case, it is indeed human activity that is artificially creating the disease, without solving the real problem, since it does not reduce cases of advanced-stage cancer, as expected by screening….
Between 1980 and 2000, the incidence rate increased by an average of 2.7% per year. The increase affected all age groups but was more marked among women aged 50 to 75. This age group is the same as the one for women for whom systematic mammographic screening is carried out (10 pilot departments in 1980).
And mortality has not been significantly reduced, as expected.
In fact, the authors of the BMJ article from 2023 that we’re quoting even add some nuance to their own words (which the press sadly doesn’t do, preferring to alarm readers):
“the study has several limitations due to the intrinsic weaknesses of the GlobalBurden of Diseases (GBD) 2019 study.
- First, the accuracy of GBD data has been compromised by the quality of cancer registry data in different countries. Thus, underreporting and underdiagnosis in developing countries may lead to underestimation of incidence and deaths from early-onset cancers.
- Second, the increasing trend in the burden of early-onset cancers remains unclear, which could be related to early detection and early exposures.
- Third, the estimation of exposure to risk factors was conducted using data with rare survey dates and from different sources, which may affect the precision of their influence and introduce potential measurement bias.
- Fourth, it is inevitable that implementing a dichotomy at age 50 has drawbacks, as pathological, molecular, and biological characteristics are unlikely to change significantly at this age.
For the authors, an important first conclusion is that dietary risk factors, alcohol consumption, and tobacco consumption were the main risk factors for the main early-onset cancers in 2019.
Secondly, the authors note that the study primarily demonstrated the following: the global morbidity (all effects subsequent to a disease, or sequelae) of early-onset cancer (in “young” people) increased between 1990 and 2019, while mortality decreased slightly, as did the “disability-adjusted life year (DALY),” i.e., life expectancy adjusted for disability due to disease (which is therefore reduced).
We can therefore legitimately ask the question of the benefit of early detection for populations.
The incidence and mortality rates (according to the study) varied considerably depending on the region, country, and type of cancer.
The most affected areas are those with medium to high socio-demographic levels. The most common cancers are breast cancer, lung cancer, colorectal cancer, and stomach cancer, which mainly occur in areas with high standards of healthcare and where certain cancers are included in screening programs.
It was expected that low-grade cancers would increase due to screening enabling early diagnosis of cancer, but is this detection beneficial if the duration of “ living with the disease ” and the associated morbidity are increased, without extending total life expectancy, and without reducing the severity of certain cancers that are extensively screened (breast, prostate, thyroid)?
An answer in JAMA
September 2025
Increase in early cancers: more apparent than real
https://jamanetwork.com/journals/jamainternalmedicine/article-abstract/2839347
Vishal R. Patel, MD, MPH; Adewole S. Adamson, MD, MPP; H. Gilbert Welch, MD, MPH
Harvard Medical School, Boston, Massachusetts (Patel); Brigham and Women’s Hospital, Boston, Massachusetts (Patel, Welch); Dell Medical School, Austin, Texas (Adamson).
The increase in early-onset cancers is not necessarily due to an increase in clinically significant cancers, that is, cancers that will manifest and actually impact the person, the researchers argued, but may be due to increased diagnostic testing and overdiagnosis. Overdiagnosis is an unnecessary detection of indolent cancers that would never have caused harm if they had remained unrecognized.
Of the eight cancers with the fastest-rising incidence, only two—colorectal cancer and endometrial cancer—showed a slight increase in early mortality. For the other six cancers, stable or declining mortality rates and an increase in the number of diagnoses suggest that higher detection (rather than more disease) may explain the trend.
What we observe
Overall, the 8 cancers with the fastest increasing incidence (number of new cases occurring in the population within a year) (>1% per year) among US adults younger than 50 years (thyroid, anal, kidney, small bowel, colorectal, endometrium, pancreas, and myeloma) have doubled in incidence since 1992, while aggregate mortality for these cancers has remained stable. Colorectal and endometrial cancers showed a slight increase in mortality; for the others, stable or declining mortality alongside increasing diagnoses suggests that greater detection (rather than more disease) accounts for this trend.
In some cancers, such as thyroid and kidney cancer, overdiagnosis is well documented.
Although not one of the fastest growing cancers (0.6% per year), breast cancer remains the most common early-onset cancer and, despite the increase in diagnoses in women under 50, mortality has fallen by about half.
Analysis
“Overall, the rise in early cancer appears to be less of an epidemic of disease and more of an epidemic of diagnosis,” Welch and colleagues wrote.
Much of the increase in cancer appears to reflect increased diagnostic testing, favoring irrelevant detections and overdiagnosis —that is, the detection of lesions that would never have caused harm.
Interpreting the rising incidence as an epidemic of disease results in a vicious cycle of more screening (the illusion of screening effectiveness) followed by equally useless treatment, while diverting attention and resources from other, more urgent threats, particularly in the health of young adults, the authors say.
Legend: Combined rates for cancers whose incidence increased by more than 1% per year on average between 1992 and 2022 in adults under 50 years of age: thyroid cancer, anal cancer, kidney cancer, small bowel cancer, colorectal cancer, myeloma, endometrial cancer (limited to women), and pancreatic cancer.
Incidence in 2020 is defined as a light blue dot because delays in health services during the COVID-19 pandemic were associated with reduced incidence rates for most cancer sites. Mortality refers to the combined mortality of the same 8 cancers. Metastatic incidence refers to the combined rate at which patients were first diagnosed and found to have metastatic disease (it does not include those who were first diagnosed with local disease and subsequently developed metastases).
Incidence and mortality data were obtained from surveillance, epidemiology, end results, and the national vital statistics system, respectively
How to interpret this figure?
Since the 1990s, the overall early incidence of these 8 cancers (thyroid cancer, anal cancer, kidney cancer, small bowel cancer, colorectal cancer, myeloma, endometrial cancer -limited to women, and pancreatic cancer) has approximately doubled, while their overall early mortality has remained remarkably stable, with the rate in 2022 being identical to that in 1992 (5.9 deaths per 100,000 in both years; Figure 1).
In other words, these cancers are diagnosed more frequently in young adults, but without a corresponding change in the most feared outcome: the mortality rate. The pattern suggests that, overall at least, the increased incidence of early cancers appears to be due less to an increase in the incidence of clinically significant cancers than to the rise in overdetection.
In a 2024 article, Welch explained that the disconnect between the increase in the incidence of a cancer and its specific mortality remaining stable indicates the presence of overdiagnosis.
While we would expect a decrease in mortality, since theoretically, the more cases of cancer are detected at the so-called “early” stage, the more fatal cancers we would avoid. This is not true for the cancers taken as examples (kidney, melanoma, thyroid), nor for breast or prostate cancer.
Chart- Legend
Incidence and mortality trends are shown for cancers whose incidence increased by more than 1% per year from 1992 to 2022 in adults under 50 years of age: thyroid cancer, anal cancer, kidney cancer, small bowel cancer, colorectal cancer, myeloma, endometrial cancer (females), and pancreatic cancer.
Three-year moving averages were applied to the rates, and the y-axes were scaled by panel to facilitate visualization of trends. Data were obtained from the National Vital Statistics System and the Surveillance, Epidemiology and End Results database.
When assessing cancer trends individually by site, the authors demonstrate heterogeneity among the eight cancers studied.
When examining mortality (Figure 2), the most important criterion for policymakers and patients, two cancers stand out due to increasing mortality: colorectal cancer and endometrial cancer.
Specifically, mortality from colorectal cancer has increased by about 0.5% per year since 2004, “suggesting some increase in the occurrence of clinically significant cancers,” the authors note; “However, its incidence has increased by about 2% per year, raising the possibility that the increase in incidence does not concern clinically significant cancers, but also includes overdiagnosis .” »
For endometrial cancer, both incidence and mortality rates have increased by about 2% per year, a trend the authors suggest is likely explained by rising rates of obesity and falling rates of hysterectomy.
For the other six cancers, stable or declining mortality rates alongside increasing diagnoses suggest that greater detection (rather than more disease) explains the trend in early cancer incidence. For example, Welch and colleagues pointed out that diagnoses for thyroid cancer have “exploded” despite stable mortality. Since 1992, there have been more than 200,000 excess diagnoses of thyroid cancer in young adults, while the number of deaths has been virtually unchanged. They
described this as “a classic signature of overdiagnosis .”
Kidney cancer diagnoses have also increased despite the drop in mortality, likely reflecting incidental detection due to increased use of abdominal imaging, the authors say.
Overdiagnosis in both thyroid and kidney cancers is well documented; for kidney cancer, it is often due to CT and ultrasound imaging and incidental findings during investigations for other pathologies.
The incidence of anal cancer has increased and then decreased, probably reflecting changing patterns of human papillomavirus and HIV infection.
For the remaining gastrointestinal cancers (small bowel and pancreas), overdiagnosis probably explains the low incidence and stable mortality, much of which reflects incidental detection of small, indolent neuroendocrine tumors on cross-sectional imaging (CT scans) or endoscopy. Finally, the increasing incidence of multiple myeloma without an increase in mortality may be explained by more widespread use of serum protein electrophoresis in biology, with approximately one-third of asymptomatic patients presenting with abnormal results on routine laboratory tests.
CONCLUSION
Overall, the rise in early-onset cancer appears to be less an epidemic of the disease and more an epidemic of diagnoses.
The lack of a substantial increase in deaths, despite the rising incidence, underscores the need to provide context to the early-onset cancer narrative and the myth of an epidemic of cancers.
First, there is unequivocal good news regarding cancer: mortality from all cancers combined in adults under 50 has fallen by nearly half since the 1990s [i].
Second, other causes of death are equally relevant: cancer deaths account for only 10% of all deaths in adults under 50. Suicides and non-cancer deaths (accidents or overdoses) are more than 4 times higher than cancer deaths, and both are increasing.
Unnecessary cancer diagnoses ( overdiagnoses, that is, those that are unlikely to lead to symptoms or death) should be a wake-up call, the authors warn in JAMA.
A cancer diagnosis can fundamentally disrupt the lives of young adults, turning those who may feel perfectly healthy into lifelong patients. (See here “How Overdiagnosis Wastes Years of Life”)
The physical toll of cancer treatment is most significant in young adults and can include infertility, long-term organ damage, and secondary cancers. The emotional fallout is substantial, as the long-term pain of anxiety and depression that comes with being labeled a cancer survivor can ripple through a person’s family and community. [ii] [iii] [iv]
Financially, the costs do not stop at treatment; monitoring, follow-up care, and managing adverse effects create long-term expenses. For young adults already dealing with limited savings and childcare responsibilities, the financial pressure can be devastating. This is why our collective also warns against screening too early, particularly for breast cancer from the age of 40, which exposes people to overdiagnosis and false alarms. We also warn against the practice of routine tests ” to reassure themselves,” sometimes requested by the public and promoted by the press.
Policymakers, researchers, and the media must be careful, the authors say, not to overinterpret the rising incidence of early-onset cancers. Searching for biological causes for rising cancer incidence without evidence of an increase in “true” cancers—that is, cancers that manifest clinically—is bound to be unproductive. Chasing potential risks is not only a waste of time but also diverts funding and attention from addressing more critical medical problems that particularly affect younger populations.
This alarms the public and perpetuates the idea that something in our environment or lifestyle is triggering more cancers, while doctors are simply detecting more cases of cancers that were always present, but without clinical translation, and indolent. Worse, it encourages the belief that young, healthy individuals could benefit from low-value screening interventions, such as whole-body imaging and
multi-cancer early detection tests. (Read: https://cancer-rose.fr/en/2022/09/15/liquid-biopsies-the-grail/)
The authors urge caution in the use of early detection and screening.
More diagnoses do not necessarily mean more deaths, but it does mean more lives that will be profoundly impacted unnecessarily.
The challenge is to refine the diagnosis to detect and treat only the factors that truly matter.
While some of the increase in early-onset cancer is likely real, it is small and confined to a few cancer sites (colon, uterus). The “epidemic” narrative not only exaggerates the problem, but may also exacerbate it. While more testing is often seen as the solution to an epidemic, it can just as easily be the cause.
In an accompanying editor’s note in the article, Ilana B. Richman, MD, MHS, and Cary P. Gross, MD, both of the Yale School of Medicine in New Haven, Connecticut, noted that Welch and colleagues’ findings have “important implications,” and that these results underscore the importance of choosing meaningful indicators for cancer control and prevention efforts.
“The goal of cancer screening and treatment should not simply be to detect cancer, but rather to reduce its morbidity and, ultimately, mortality,” they write. “Focusing on changes in mortality is therefore a more reliable way to identify policy priorities, prioritize areas for study, and justify changes in practice.”
References
[i] NationalCancerInstituteSurveillanceResearch Program. Surveillance, Epidemiology, and End Results Program. Accessed September 28, 2022. https://seer.cancer.gov/
[ii] LeeARYB,LowCE,YauCE,LiJ,HoR,HoCSH. Lifetime burden of psychological symptoms, disorders, and suicide due to cancer in childhood, adolescent,andyoungadultyears:asystematic review and meta-analysis. JAMA Pediatr. 2023;177 (8):790-799. doi:10.1001/jamapediatrics.2023.2168
[iii] HoffmanKE,McCarthyEP,RecklitisCJ,NgAK. Psychological distress in long-term survivors of adult-onset cancer: results from a national survey. Arch Intern Med. 2009;169(14):1274-1281. doi:10. 1001/archinternmed.2009.179
[iv] Carreira H, Williams R, Müller M, Harewood R, Stanway S, Bhaskaran K. Associations between breast cancer survivorship and adverse mental health outcomes: a systematic review. J Natl Cancer Inst. 2018;110(12):1311-1327. doi:10.1093/jnci/djy177
🛈 Nous sommes un collectif de professionnels de la santé, rassemblés en association. Nous agissons et fonctionnons sans publicité, sans conflit d’intérêt, sans subvention. Merci de soutenir notre action sur HelloAsso.
🛈 We are an French non-profit organization of health care professionals. We act our activity without advertising, conflict of interest, subsidies. Thank you to support our activity on HelloAsso.
