The Epidemiology of Colorectal Cancer in Younger and Older Patients

Abstract

Background

Recent studies have shown that the incidence of colorectal cancer among younger persons is rising. We investigated incidence trends and survival in the German federal state of North Rhine–Westphalia.

Methods

Cancer registry data from the period 2008–2019 were classified into two age groups (15–54 and 55–99). In each age group, the standardized incidence, average annual percent change (AAPC), and relative 5-year survival (RS) were calculated and stratified according to the site, histology, size, and grade of the colorectal tumor.

Results

167 919 cases were included, with adenocarcinoma accounting for 86.4%. In 2019, the age-standardized incidence per 100 000 person-years was 13.8 and 10.3 among men and women (respectively) in the younger age group, compared with 197.9 and 126.3 in the older age group. Over the study period from 2008 to 2019, the incidence declined among older men and women (AAPC –2.6% and –2.9%) but remained nearly constant among younger men and women (-0.5% and –0.4%). The incidence of neuroendocrine, T1, and G2 tumors rose in both age groups (AAPC range: 2.3%–12.2%; 2.2%–8.3%, 6.3%–8.8%). Younger patients have a better RS, with the largest difference between age groups being found for neuroendocrine tumors (88% and 83% in younger men and women, 65% and 61% in older men and women).

Conclusion

The incidence of colorectal tumors has remained constant in persons under age 55 and declined in persons aged 55 and older. Nonetheless, the incidence of neuroendocrine tumors and of small and well-differentiated tumors has risen in both age groups. The trends among younger persons and the rise in neuroendocrine tumors merit further study.

Colorectal cancer (CRC) includes malignancy of the appendix, colon, and rectum and is the third most common form of cancer in men and the second most common form in women (1, 2). The incidence of CRC is rising in many low- and middle-income countries, whereas it has stabilized or declined in high-income countries (3). Studies show that the available screening measures such as colonoscopy play an effective role in reducing CRC incidence (4, 5). Colonoscopy makes it possible to detect and remove asymptomatic-stage tumors as well as precursor lesions.

In Germany, individuals with statutory health insurance are offered the fecal occult blood test (FOBT) in addition to colonoscopy for colorectal cancer screening. Since 2002, anyone aged between 50 and 54 years is entitled to one FOBT a year and, from the age of 55 years, to two colonoscopies at an interval of at least 10 years. In 2019, the hitherto opportunistic screening for colorectal cancer was replaced by organized screening, and since then, men have been entitled to a colonoscopy from the age of 50 years.

Seen as a whole, this is why the incidence of CRC is falling or stabilizing in high-income countries; however, studies show differences in this trend in terms of the age of disease onset. For example, incidence among people aged over 50 years has declined or remained constant, whereas the incidence among those aged under 50 years has risen (6– 12). This excludes the incidence of neuroendocrine tumors, which has risen among younger and older people (13– 15).

There is evidence that CRC in younger patients differs on the clinical, pathologic, and molecular level from CRC in older patients (16). The majority of CRCs diagnosed in young adults are sporadic and of unclear etiology (17). They are often found as a result of symptoms and not through cancer screening (18, 19). Although genetic components and factors such as sedentary lifestyle, obesity, and diabetes mellitus increase the risk of developing CRC, they cannot on their own explain this rise in younger people (16). Not even the increased number of screening tests among individuals aged under 50 years is able to account for this rise (20).

In contrast, an analysis of nationwide German registry data showed a decline in incidence in both younger and older individuals (21). The aim of that particular analysis was to analyze the trend in incidence and relative survival for younger and older patients with CRC in North Rhine-Westphalia (NRW) for the period 2008–2019 on the basis of cancer registry data, stratified by tumor localization, histological groups, tumor size, and histopathological grading.

Methods

The database of the NRW federal state cancer registry represents a population of approximately 18 million inhabitants. Cancer has been a notifiable disease since 2005. The coverage of cancer registration, which is regularly estimated by the Center for Cancer Registry Data at the Robert Koch Institute, is put at > 90% for the study period. The data basis for the investigation was all malignant CRCs diagnosed in the period 2008–2019 (ICD-10: C18–C20) with an age of onset of 15–99 years. Cases were classified into two groups based on age of onset: the younger group comprised 15- to 54-year-olds and the older group, 55- to 99-year-olds. The age threshold between groups was set at 55 years due to the fact that, during the study period, screening colonoscopy was offered from 55 years (since 2019, it has been available to men aged 50 years and above).

Malignant CRC was classified and evaluated in accordance with the International Classification of Diseases for Oncology (ICD-O-3) (e1) based on topography in the proximal colon (C18.0, C18.2–C18.4), distal colon (C18.5–C18.7), unspecified colon (C18.9), and rectum (C19–C20), as well as on histology in adenocarcinomas, neuroendocrine tumors, and other unspecified malignant neoplasms (see Table 1 for morphology codes). In addition, analyses were stratified according to tumor size (T stage) and histopathological grading. Due to the low case number, C18.8 coded tumors (overlapping sites of colon, 859 cases, 0.7%) and other histologically specific malignant neoplasms (538 cases, 0.4%) were not taken into consideration for the stratification according to localization or histology. Tumors of the appendix (C18.1) were excluded since they differ from other CRCs in terms of their biology and etiology (22, 23).

Statistics

Age-specific incidence rates (ASR) were calculated using the old European standard population (24). Furthermore, the average annual percentage changes (AAPCs) were estimated using joinpoint regression. This enables analysis of trends even when the scale and direction of the trend change.

Joinpoint regression fits a polyline of piecewise linear segments to the incidence rates, with the number and positions of joinpoints determined by permutation tests. By using a log link function, the polyline maps out a relative (percentage) change. The AAPC is not expressed by the slope of a single regression line but rather by the mean of the slope coefficients weighted by the number of years in a segment. This enables a precise estimation of AAPC (25).

Relative 5-year survival (RS) was calculated for the calendar period 2017–2019. RS is a computational method to determine cancer-specific survival and is defined as a quotient of the observed survival of patients and the expected survival in the general population of the same age, sex, and calendar period (26). Expected survival was calculated based on the mortality tables for NRW using the Ederer-II method (27, 28). RS was estimated with the period approach (29). The period approach yields more up-to-date survival estimates compared to the traditional cohort approach, since it uses only the information on survival from a recent calendar period for which mortality data is available (30). Thus, the period approach makes it possible to promptly identify changes in the prognosis of cancer patients. The RS was age-standardized with the International Cancer Survival Standard (ICSS) (31). Death-certificate-only (DCO) cases were excluded from the calculation of RS. Calculations were performed with R 3.6.2 (e2), RS was estimated with the periodR package, version 1.0–6 (32), and Joinpoint Regression 4.9.0.0, which is made available by the US National Cancer Institute, was used for the AAPCs.

Results

Incidence

Table 2 shows the incidences of CRC according to age and sex for CRC overall and for the stratifications. A total of 167,919 cases were included, 90% of which were in the older age group, and 86.4% of all tumors were adenocarcinomas. The median age of disease onset was 50 and 49 years in younger men and women, respectively, and 73 and 77 years in older men and women, respectively.

Table 2. Absolute and relative incidence of colorectal cancer in North Rhine-Westphalia, Germany, 2008–2019.

	Younger patients				Older patients				Total
	Males		Females		Males		Females
	n	%	n	%	n	%	n	%	n	%
Total	8749	100	7047	100	80,225	100	71,898	100	167,919	100
ICD-O-3 localizations
Proximal colon	1869	21.4	1475	20.9	21,996	27.4	26,293	36.6	51,633	30.7
Distal colon	2162	24.7	2070	29.4	20,691	25.8	15,762	21.9	40,685	24.2
Rectum	4122	47.1	2986	42.4	29,529	36.8	20,306	28.2	56,943	33.9
Overlapping sites of colon	44	0.5	30	0.4	462	0.6	528	0.7	1064	0.6
Not otherwise specified	552	6.3	486	6.9	7547	9.4	9009	12.5	17,594	10.5
Histological groups
Adenocarcinomas	7730	88.4	6189	87.8	70,639	88.1	60,442	84.1	145,000	86.4
Neuroendocrine tumors	375	4.3	348	4.9	978	1.2	925	1.3	2626	1.6
Other unspecified malignant neoplasms	596	6.8	437	6.2	8404	10.5	10,225	14.2	19,662	11.7
Other specified malignant neoplasms	48	0.5	73	1.0	204	0.3	306	0.4	631	0.4
Tumor size
T1	1009	11.5	864	12.3	9481	11.8	7003	9.7	18,357	10.9
T2	844	9.6	744	10.6	8629	10.8	7591	10.6	17,808	10.6
T3	3320	37.9	2647	37.6	30,292	37.8	25,455	35.4	61,714	36.8
T4	1082	12.4	922	13.1	8252	10.3	9377	13.0	19,633	11.7
Unknown	2494	28.5	1870	26.5	23,571	29.4	22,472	31.3	50,407	30.0
Histopathological grading
G1	795	9.1	649	9.2	5844	7.3	4216	5.9	11,504	6.9
G2	5197	59.4	4205	59.7	50,580	63.0	41,163	57.3	101,145	60.2
G3/4	1802	20.6	1483	21.0	12,963	16.2	14,398	20.0	30,646	18.3
Unknown	955	10.9	710	10.1	10,838	13.5	12,121	16.9	24,624	14.7

The percentage of DCO was 10% for older patients (2008: 19%; 2019: 7%) and 3% for younger patients (2008: 6%; 2019: 1%).

Table 3 shows the age-standardized incidence rates for 2008 and 2019. The incidence of tumors in the rectum was—with the exception of older women in 2019—highest in all age groups (2008 and 2019: younger men 6.7 and 6.2, respectively; younger women 4.7 and 4.7 respectively; older men 103 and 73, respectively, per 100,000). In older women, the incidence of tumors in the proximal colon was highest in 2019 (2008, rectum 54; 2019 proximal colon, 51 per 100,000).

Table 3. Age-standardized incidence rates (ASR) of colorectal cancer per 100,000 person-years (old European standard) for 2008 and 2019 by age, sex, and tumor characteristics in North Rhine-Westphalia, Germany.

	Younger patients				Older patients
	Males		Females		Males		Females
	2008	2019	2008	2019	2008	2019	2008	2019
Total	13.8	13.9	11.2	10.3	274.1	197.9	175.4	126.3
ICD-O-3 localizations
Proximal colon	2.6	3.4	2.1	1.9	61.3	58.4	50.6	51.2
Distal colon	3.2	3.8	3.1	3.2	69.2	54.6	41.5	28.2
Rectum	6.7	6.2	4.7	4.7	103.4	73.1	53.7	36.9
Overlapping sites of colon	0.1	0.0	0.0	0.0	1.5	0.8	1.4	0.5
Not otherwise specified	1.2	0.5	1.2	0.4	38.7	11.1	28.1	9.7
Histological groups
Adenocarcinomas	12.2	12.0	9.9	8.9	227.8	176.3	142.5	109.6
Neuroendocrine tumors	0.3	0.7	0.2	0.6	2.3	3.0	1.5	1.8
Other unspecified malignant neoplasms	1.3	1.1	1.0	0.7	43.6	17.8	30.9	14.0
Tumor size
T1	1.0	2.0	0.7	1.5	23.3	31.5	13.9	17.3
T2	1.2	1.2	1.0	1.2	29.6	22.0	18.4	13.8
T3	5.2	5.1	4.3	3.5	97.8	68.7	58.4	41.7
T4	1.6	1.7	1.4	1.2	24.0	19.5	19.1	16.8
Unknown	4.8	3.8	3.7	2.9	99.4	56.3	65.7	36.8
Histopathological grading
G1	0.7	2.0	0.3	1.3	10.9	22.6	6.3	12.3
G2	7.9	8.0	6.7	6.4	162.3	124.7	98.7	73.8
G3/4	3.3	2.5	2.7	1.6	47.7	26.9	34.7	23.0
Unknown	2.0	1.4	1.4	1.1	53.1	23.6	35.6	17.2

With regard to tumor size, T3 tumors had the highest incidence in both age groups (2008 and 2019: younger men, 5.2 and 5.1, respectively; younger women, 4.3 and 3.5, respectively; older men, 98 and 69, respectively; older women, 58 and 42, respectively, per 100,000).

Grade G2 tumors had the highest incidence in all age groups (2008 and 2019: younger men, 7.9 and 8.0; younger women, 6.7 and 6.4; older men, 162 and 125; older women, 99 and 74 per 100,000).

Average annual percentage change (AAPC)

Within the period analyzed, incidence fell among older men and women (AAPC: -2.6% and -2.9%, respectively), while it remained constant among younger men and women (AAPC: -0.5% and -0.4%, respectively (table 4).

Table 4. Average annual percentage change (AAPC) with confidence interval (CI) in colorectal cancer by age, sex, and tumor characteristics in North Rhine-Westphalia, Germany, 2008–2019.

	Younger patients				Older patients
	Males		Females		Males		Females
	AAPC	[95% CI]	AAPC	[95% CI]	AAPC	[95% CI]	AAPC	[95% CI]
Total	−0.5	[−1.3; 0.3]	−0.4	[−1.1; 0.3]	−2.6	[−3.3; −1.9]	−2.9	[−3.9; −2]
ICD-O-3 localizations
Proximal colon	1.4	[0; 2.8]	0.5	[−1.6; 2.8]	−0.5	[−1.3; 0.3]	0	[−0.6; 0.5]
Distal colon	1.1	[−0.3; 2.6]	0.8	[−0.6; 2.3]	−2.1	[−2.6; −1.6]	−3.1	[−4; −2.2]
Rectum	−0.9	[−2.1; 0.3]	−0.2	[−1.3; 0.9]	−3.6	[−2.2; −8.7]	−3.2	[−3.8; −2.6]
Not otherwise specified	−9.6	[−11.7; −7.5]	−9.7	[−11.7; −7.7]	−10.7	[−12; −9.5]	−9.7	[−11.2; −8.2]
Histological groups
Adenocarcinomas	−0.5	[−1.5; 0.4]	−0.7	[−1.3; −0]	−2.5	[−3; −2.1]	−2.4	[−2.8; −2]
Neuroendocrine tumors	8.4	[−1.6; 19.6]	12.2	[3.9; 21.2]	2.3	[0.5; 4.2]	3.7	[−1; 8.6]
Other unspecified malignant neoplasms	−3.1	[−10.6; 4.9]	−1.2	[−7; 4.9]	−6	[−9.3; −2.7]	−6.7	[−9.1; −4.2]
Tumor size
T1	4.6	[0.8; 8.5]	8.3	[1.5; 15.6]	2.9	[1.7; 4.1]	2.2	[0.4; 3.9]
T2	−1.4	[−3.7; 0.9]	0.3	[−2; 2.6]	−3	[−4.1; −1.8]	−2.6	[−3.5; −1.7]
T3	−0.9	[−2.5; 0.7]	−1.4	[−2.6; −0.3]	−3.4	[−4.5; −2.2]	−3.2	[−4.2; −2.3]
T4	−0.4	[−2.2; 1.4]	−0.2	[−2.1; 1.9]	−1.1	[−2.9; 0.7]	−1.5	[−3.3; 0.4]
Unknown	−1.5	[−4.6; 1.8]	−2.1	[−5.8; 1.8]	−3.8	[−6; −1.6]	−4.8	[−6.2; −3.4]
Histopathological grading
G1	8.8	[5.7; 12]	8.7	[4.7; 12.8]	6.5	[5.5; 7.5]	6.3	[4.6; 8.1]
G2	−0.4	[−1.5; 0.7]	0.1	[−0.9; 1.1]	−2.2	[−3.6; −0.8]	−2.7	[−3.1; −2.2]
G3/4	−3.9	[−5.3; −2.6]	−4.7	[−5.8; −3.6]	−5.3	[−6.1; −4.5]	−4.0	[−4.6; −3.4]
Unknown	−2.6	[−7.5; 2.4]	−2.2	[−5.3; 1.1]	−5.9	[−8.1; −3.6]	−6.2	[−9.1; −3.3]

In younger men, the incidence of tumors in the colon rose slightly (proximal AAPC: 1.4%, distal AAPC: 1.1%). In older men and women, the incidence of tumors in the distal colon and rectum fell (AAPC range: -3.1% to -2.8%).

The incidence of neuroendocrine tumors, tumor size T1, and grade G1 tumors rose in both age groups (neuroendocrine tumors, AAPC range: 2.3–12.2%; T1, AAPC range: 2.2–8.3%, G1, AAPC range: 6.3–8.8%).

Relative 5-year survival

The 5-year RS was higher in the younger age group compared to the older (younger men and women, 74% and 75%, respectively; older men and women, 64% and 67%, respectively; Figure 1, eFigure); this also applied to most stratifications.

Figure.

Relative 5-year survival in colorectal cancer patients with 95% confidence interval and difference in percentage points between younger and older patients according to sex and tumor characteristics in North Rhine-Westphalia, Germany, calendar period 2017–2019.

a, Total; b, localization; c, histological group;

NET, neuroendocrine tumors

eFigure.

Relative 5-year survival in colorectal cancer patients with 95% confidence interval and difference in percentage points between younger and older patients according to sex and tumor characteristics in North Rhine-Westphalia, Germany, calendar period 2017–2019.

a, Tumor size; b, histopathological grading

UNK, unknown

Exceptions to this were tumor size T1 and T2 as well as grade G3/4 tumors in women; there were no differences here in 5-year RS between younger and older patients.

The most notable difference was in the 5-year RS for neuroendocrine tumors (younger men and women, 88% and 83%, respectively; older men and women, 65% and 61%, respectively).

Discussion

The overall incidence of colorectal cancer in North Rhine-Westphalia declined in the 2008–2019 period. However, the analysis shows that the incidence trends between younger and older patients differ. CRC incidence fell in the older age group, while it remained constant in the younger age group. Neuroendocrine tumors as well as prognostically favorable stage T1 tumors and well differentiated tumors (G1) rose in both age groups, whereas the incidence of larger stage-T2 to -T4 tumors, poorly or undifferentiated tumors (G3/G4), and adenocarcinomas declined in particular in the older age group. In addition, a slight rise in the incidence of tumors in the proximal and distal colon was observed in younger men. Regarding relative 5-year survival, there was a trend toward a better prognosis for younger patients. The greatest differences between age groups was seen for neuroendocrine tumors, where survival was significantly better for younger patients.

The trend in incidence in the older age group is presumably affected by colorectal cancer screening (21). Recent results on the effect of screening programs on the incidence of colorectal cancer in Germany and other European countries show a rise in UICC stage-1 cancer (33). The incidence of tumors in the proximal colon in the older age group has remained constant, which could be due to the proven lower effectiveness of colonoscopy in the proximal colon (5).

In the younger age group, it is less clear that incidence trends can be attributed to colorectal cancer screening. Unfavorable diet, lack of exercise, and obesity already at a young age may also contribute to the earlier onset of CRC. Over the study period, 35–40% of CRC cases in both age groups were found at an advanced stage (T3); however, an increase in prognostically favorable stages was seen in the younger age group. It is possible that the age at colonoscopy uptake went down during the study period, meaning that tumors and precursor lesions were detected earlier even in younger people. Furthermore, the fecal occult blood test (FOBT) was available for screening in 50- to 54-year-olds during the study period. Colonoscopy at a younger age is also recommended in the case of familial CRC. Weigl et al. (34) estimate that 21% of 40- to 44-year-olds, 28% of 45- to 49-year-olds, and 35% of 50- to 54-year-olds in Germany have ever undergone colonoscopy.

The observed rise in the incidence of neuroendocrine tumors tallies with the results of an analysis of SEER (Surveillance, Epidemiology and End Results) data, which sees the more intensive use of endoscopy, alongside improved classification of these tumors, as the reason for this (15). Neuroendocrine CRC has a highly variable clinical presentation. Neuroendocrine tumors (NET) of the colon are often detected late and at advanced stages, whereas NET of the rectum tend to be small and low-grade at diagnosis (35). Therefore, we investigated the extent to which the rise in T1 and G1 tumors could be due to the likewise increased rate of NET diagnosis and, even after excluding NET, found a continued but lower rise in the incidence of T1 and G1 tumors.

The difference in relative 5-year survival between the younger and old age groups may be due to fewer comorbidities and more aggressive treatment. Abdelsattar et al. (36) also observed a higher cancer-specific survival in younger patients, despite the fact that disease stage was more often advanced at diagnosis compared to that in older patients. Bahadoer et al. (37) attributed the differences in prognosis in a number of European countries to the higher mortality rate among older patients in the first year after surgery. If patients survived the first year, there were no differences compared to younger patients.

One of the strengths of this work lies in the extensive database of the federal state cancer registry of NRW, which makes it possible to also analyze rarer tumors and tumor characteristics.

Limitations of the study include the period for which data are available, which does not map incidence before and immediately after the introduction of colorectal cancer screening in 2002. In addition, the cancer registry data do not include any information as to whether a tumor was detected through screening or at symptom onset. Therefore, it is not possible to investigate the direct effect of screening colonoscopy. The often still comparatively high number of DCO cases in recent registers also needs to be taken into consideration particularly when calculating survival time. DCO cases are excluded from survival analysis since no information on date of diagnosis is available. Since DCO cases become more likely at a more advanced age and with a worse prognosis (38), this leads to an overestimation of prognosis particularly in the older age group (39). Therefore, the difference in prognosis between younger and older patients is likely to be even greater in that analysis than the difference presented here.

This paper investigated differences in incidence and probability of survival between the younger and older population aged from 55 years that has been eligible to take part in colorectal cancer screening with colonoscopy since 2002. Vuik et al. (12) found a sharper rise in incidence in Europe in the period 2004–2016 for 20- to 29-year-old adults (7.9%) compared to 30- to 39-year-olds (4.9%) and 40- to 49-year-olds (1.6%). An evaluation of screening colonoscopy (40) also showed that 15.4% of men and 14.9% of women took advantage of this screening option in the period 2010–2019. The use of screening colonoscopy rose by 14.3% in 2019 due to the lowering of the age limit for men to 50 years and the organized invitation process. Therefore, incidence trends by localization, tumor stage, and grading, particularly in the younger age group, should be investigated in further studies in a more differentiated manner according to age and time period. The observed rise in the incidence of neuroendocrine tumors in both age groups should also be the subject of further research regarding the varying clinical picture according to tumor localization. In contrast to Bahadoer et al. (37), the prognosis for older patients in NRW after the first postoperative year could be compared to that of younger patients. For a more detailed analysis, one could also use the treatment data recorded in the statewide clinical cancer registry that has been in existence in NRW since 2016.

In summary, this analysis has shown that the incidence of CRC is overall in decline and that the effects of screening colonoscopy are becoming apparent in the over-55 age group. Incidence trends in the younger age group and the rise in neuroendocrine tumors in both age groups should be further investigated.

Table 1. Assignment of ICD-O-3 morphology codes to histological groups (only grade 3 malignancy was taken into account).

Histological group	Morphology code
Adenocarcinomas	8140–8149, 8160–8163, 8190–8221, 8260–8337, 8350–8552, 8570–8576, 8940–8941
Neuroendocrine tumors	8013, 8041–8045, 8150–8156, 8240–8246, 8248–8249
Other unspecified malignant neoplasms	8000–8005, 8010–8011, 8014–8015, 8020–8022, 9970, 9975
Other specified malignant neoplasms	All others