Colorectal cancer incidence trends in younger versus older adults: an analysis of population-based cancer registry data

Abstract

Background: Previous studies have shown that colorectal cancer incidence is increasing among younger adults (aged <50 years) in multiple high-income western countries in contrast with stabilising or decreasing trends in incidence in older adults (aged ≥50 years). This study aimed to investigate contemporary colorectal cancer incidence trends in younger adults versus older adults.

Methods: Colorectal cancer incidence data, including year of diagnosis, sex, and 5-year age group for 50 countries and territories, were extracted from the WHO-International Agency for Research on Cancer Cancer Incidence in Five Continents Plus database. The Human Development Index 2022 was retrieved from the United Nations Development Programme and grouped into very high (>0·80), high (0·70-0·79), medium (0·55-0·69), and low (<0·55) categories. Age-standardised incidence rates (ASR) per 100 000 person-years of early-onset (diagnosed between ages 25 to 49 years) and late-onset (diagnosed between ages 50 to 74 years) colorectal cancer (ICD 10th revision, C18-20), diagnosed between 1943-2003 and 2015-17, were calculated using the direct method and Segi-Doll world standard population). The primary study objective was to examine contemporary colorectal cancer incidence trends in younger adults versus older adults using data until 2017 from 50 countries and territories. Temporal trends were visualised and quantified with joinpoint regression, stratified by age at diagnosis (25-49 years or 50-74 years). Average annual percentage changes (AAPC) were estimated.

Findings: In the most recent 5 years (2013-17 for all countries analysed, except for Japan [2011-15], Spain [2012-16], and Costa Rica [2012-16]), the incidence rate of early-onset colorectal cancer was highest in Australia (ASR 16·5 [95% CI 16·1-16·9]), the USA (Puerto Rico; 15·2 [14·2-16·2]), New Zealand (14·8 [14·0-15·6]), the USA (14·8 [14·7-14·9]), and South Korea (14·3 [14·0-14·5]) and lowest in Uganda (4·4 [3·6-5·2]) and India (3·5 [3·3-3·7]). The highest incidence rates among older adults were found in the Netherlands (168·4 [166·9-170·0]) and Denmark (158·3 [155·8-160·9]) and the lowest were in Uganda (45·9 [38·5-51·4]) and India (23·5 [22·8-24·3]). In terms of AAPC, in the most recent 10 years, incidence rates of early-onset colorectal cancer were stable in 23 countries, but increased in 27 countries with the greatest annual increases in New Zealand (AAPC 3·97% [95% CI 2·44-5·52]), Chile (3·96% [1·26-6·74]), Puerto Rico (3·81% [2·68-4·96]), and England (3·59% [3·12-4·06]). 14 of the 27 countries and territories showed either stable (Argentina, France, Ireland, Norway, and Puerto Rico) or decreasing (Australia, Canada, Germany, Israel, New Zealand, Slovenia, England, Scotland, and the USA) trends in older adults. For the 13 countries with increasing trends in both age groups, the average annual percentage increase in younger compared to older adults was higher in Chile, Japan, Sweden, the Netherlands, Croatia, and Finland; lower in Thailand, France (Martinique), Denmark, and Costa Rica; and similar in Türkiye, Ecuador, and Belarus. The rise in early-onset colorectal cancer was faster among men than women in Chile, Puerto Rico, Argentina, Ecuador, Thailand, Sweden, Israel, and Croatia, whereas faster increase among women compared to men was in England, Norway, Australia, Türkiye, Costa Rica, and Scotland.

Interpretation: Early-onset colorectal cancer incidence rates are rising in 27 of 50 countries and territories examined, with the rise either exclusive to early-onset disease or faster than the increase in older adults in 20 of the 27 countries. The findings underscore the need for intensified efforts to identify factors driving these trends and increase awareness to help facilitate early detection.

Funding: Intramural Research Program of the American Cancer Society, Cancer Grand Challenges, and National Institutes of Health.

Figure 1

Trends in colorectal cancer incidence rates in younger (25–49 years) and older adults (50–74 years) *Subnational data. Population coverage is detailed in the table and details on population-based cancer registries are in the appendix (pp 5–7). Joinpoint (version 5.2.0) was used to fit a series of joined straight lines on a logarithmic scale of age-standardised incidence rates (dots) using the grid search method and the permutation test with the maximum number of joinpoints set at two (appendix pp 2–4). All annual percentage changes and corresponding 95% CIs in each segment can be found in the appendix (pp 8–11).

Figure 1

Trends in colorectal cancer incidence rates in younger (25–49 years) and older adults (50–74 years) *Subnational data. Population coverage is detailed in the table and details on population-based cancer registries are in the appendix (pp 5–7). Joinpoint (version 5.2.0) was used to fit a series of joined straight lines on a logarithmic scale of age-standardised incidence rates (dots) using the grid search method and the permutation test with the maximum number of joinpoints set at two (appendix pp 2–4). All annual percentage changes and corresponding 95% CIs in each segment can be found in the appendix (pp 8–11).

Figure 1

Trends in colorectal cancer incidence rates in younger (25–49 years) and older adults (50–74 years) *Subnational data. Population coverage is detailed in the table and details on population-based cancer registries are in the appendix (pp 5–7). Joinpoint (version 5.2.0) was used to fit a series of joined straight lines on a logarithmic scale of age-standardised incidence rates (dots) using the grid search method and the permutation test with the maximum number of joinpoints set at two (appendix pp 2–4). All annual percentage changes and corresponding 95% CIs in each segment can be found in the appendix (pp 8–11).

Figure 1

Trends in colorectal cancer incidence rates in younger (25–49 years) and older adults (50–74 years) *Subnational data. Population coverage is detailed in the table and details on population-based cancer registries are in the appendix (pp 5–7). Joinpoint (version 5.2.0) was used to fit a series of joined straight lines on a logarithmic scale of age-standardised incidence rates (dots) using the grid search method and the permutation test with the maximum number of joinpoints set at two (appendix pp 2–4). All annual percentage changes and corresponding 95% CIs in each segment can be found in the appendix (pp 8–11).

Figure 2

The most recent 10-year average annual percentage change in age-standardised incidence rates in colorectal cancer in younger (25–49 years) and older (50–74 years) adults from 2008 to 2017* AAPC=average annual percentage change. *Exceptions are Costa Rica and Spain (2005–16) and Japan (2006–15). †Subnational data. Population coverage is detailed in the table and details on population-based cancer registries are in the appendix (pp 5–7). Dots represent the 10-year AAPC, and lines indicate the corresponding 95% CIs (appendix pp 8–9). Sex-specific results are shown in the appendix (pp 14–15, 18–19).