Physical Activity, Genetic Susceptibility, and Risk of Colorectal Cancer in Type 2 Diabetes

A Large Population-Based Cohort Study

Background: It is crucial to identify modifiable preventive measures to mitigate the risk of colorectal cancer (CRC) among patients with type 2 diabetes (T2D), a high-risk group for CRC. We conducted a study to investigate the potential association between various levels of physical activity (PA) and the incidence of CRC, taking account of genetic susceptibility, in a T2D population cohort.

Methods: The study was based on UK Biobank (UKB) data on persons diagnosed with T2D; the participants were tracked until 2022. Hazard ratios (HR) and 95% confidence intervals (95% CI) for CRC were calculated using Cox regression models.

Results: The 33 733 patients with T2D were followed up for a median of 13.62 years. During this time, 551 patients were diagnosed with CRC. Compared with low PA, the multivariable adjusted HR for CRC among T2D patients was 0.81, 95% CI [0.66; 0.98] and 0.74 [0.58; 0.94] in the groups with moderate and high PA, respectively. In right colon cancer, moderate and high levels of PA were associated with 31% (HR 0.69; 95% CI [0.51; 0.93]) and 42% (HR 0.58; 95% CI [0.40: 0.85]) reductions in the risk of CRC. High PA (HR 0.54; 95% CI [0.35; 0.84] was associated with a lower risk of CRC even in patients with a high polygenic risk score (PRS). Persons with low PRS and high PA had the lowest risk of CRC.

Conclusion: Our study suggests that moderate to high PA helps to reduce the risk of CRC in T2D patients and that joint consideration of PA level and PRS could provide valuable insights for personalized strategies to prevent CRC.

Cite this as: Wu Y, Meng M, Liu Y, Zeng R, Feng J, Lian Q, Ma Y, Zhang L, Huang W, Leung FW, Duan C, Sha W, Chen H: Physical activity, genetic susceptibility, and risk of colorectal cancer in type 2 diabetes: a large population-based cohort study. Dtsch Arztebl Int 2025; 122: 186–92. DOI: 10.3238/arztebl.m2025.0013

Colorectal cancer (CRC) is globally ranked as the third most widespread form of cancer and the most frequently occurring malignant tumor of the digestive tract. CRC accounts for more than 10% of all cancers worldwide and 9.4% of all cancer-related deaths (1). The etiology of CRC is multifactorial, including a variety of lifestyle factors (2, 3). Of particular interest is type 2 diabetes (T2D), which accounts for nearly 90% of the estimated 537 million diabetes cases worldwide (4, 5).

Multiple epidemiological studies have demonstrated that persons with T2D have an approximately 30% higher risk of CRC than persons without diabetes, which may be attributable to chronic inflammation, insulin resistance, hypertriglyceridemia, and obesity (3). Since the risk of CRC may increase with the rising prevalence and earlier onset of T2D, the risk of CRC in patients with T2D requires particularly close attention.

In recent years, numerous studies have demonstrated that physical activity (PA) is associated with reduced morbidity and mortality of persons with cancer, including CRC, most likely through mechanisms such as improved glucose metabolism and heightened insulin sensitivity (6, 7, 8, 9). It is now generally acknowledged that PA is associated with a reduced risk of gastrointestinal cancers, including CRC (10). However, among T2D patients, who are considered a special high-risk group for the occurrence of CRC, the relationship between PA and CRC remains controversial (11, 12). The absence of a combined approach using the polygenic risk score (PRS) has left a gap in our understanding of the association between PA and CRC risk at different genetic risk levels.

To close this knowledge gap, we used data on T2D patients from the UK Biobank (UKB) for a study to investigate the potential association between various levels of PA—expressed in terms of metabolic equivalent task (MET-min/week)—and the incidence of CRC. Additionally, we incorporated the PRS to enhance our understanding of the association between PA and CRC incidence among persons with high genetic susceptibility. The hypotheses presented in this study formed part of a broader analysis conducted using UKB data (project number: 83339).

Materials and methods

Study design and study population

The UKB is a large, population-based cohort, with over 500 000 participants aged 40 to 69 years, recruited between 2006 and 2010. The Biobank project approval number for our study is 83339. This analysis is exploratory in nature, as it does not involve predefined primary or secondary endpoints. Details of the T2D population from the UKB can be found in eSupplement-Methods. eSupplement-Figure 1 presents an overview of the study design in the T2D population.

Assessment of physical activity

Baseline PA was evaluated using the self-reported International Physical Activity Questionnaire (IPAQ) (13). From the responses, the total MET-min/week for all PA, including walking, moderate, and vigorous PA, were computed (eSupplement-Methods) (14). Participants were classified into three exclusive groups according to their PA, converted into MET-min/week: low (<  600), moderate (600–3000), and high (≥  3000), following the IPAQ standard scoring criteria (6).

Diagnosis of colorectal carcinoma

The diagnosis of CRC was defined by the corresponding International Classification of Diseases (ICD) codes (ICD-10: C18–20), obtained from hospital inpatient records. Participant follow-up began at the time of their recruitment to the cohort and continued until the first cancer registration, death, or the final follow-up visit, whichever occurred first. Furthermore, we conducted an in-depth evaluation of left-sided colon cancer (LSCC), right-sided colon cancer (RSCC) and rectal cancer (RC) as outcomes (eSupplement-Methods) (15).

Definition of covariates

The following covariates were selected for our study: age, sex, ethnicity, Townsend Deprivation Index, household income, educational attainment, body mass index (BMI), smoking status, alcohol consumption, vegetable consumption, fruit consumption, vitamin or mineral supplementation and other dietary supplements, consumption of processed meats, consumption of bread and cereals, family history of CRC, blood vitamin D level, HbA1c, comorbidities, comedications, and PRS (eSupplement-Methods). The PRS from the UKB was calculated on the basis of combined statistics derived from previous genome-wide association studies (GWAS) of CRC (16). A directed acyclic graph (DAG), based on the existing literature and expert knowledge, was used to identify potential confounders (eSupplement-Figure 2) (17, 18, 19, 20).

Statistical analysis

The baseline characteristics of the participants were displayed as percentages for categorical variables and as mean with standard deviation (SD) for continuous variables. Covariates with missing data counts exceeding 20% were excluded, while those with missing data counts under 20% were addressed using multiple imputation (eSupplement-Methods). Detailed information on the number of missing covariates can be found in eSupplement-Table 1.

Cox proportional hazards models were used to assess the association between PA and CRC in persons with T2D. Specifically, we examined three distinct models, with model III serving as the primary analytical model. Three incremental models were fitted (eSupplement-Methods).

Multiple hypothesis testing correction was performed using the FDR (false discovery rate) method to assure credibility of the results. The cumulative hazard plot was utilized to analyze the cumulative incidence of CRC across the three groups with different PA. The continuous variable of PA was categorized into multiple levels, and its relationship with the incidence of CRC was illustrated using a smooth spline curve.

Subgroup analyses were conducted to evaluate potential heterogeneity in the associations between PA and endpoints across strata of key confounders, including age, BMI, and insulin usage (eSupplement-Methods). The likelihood ratio test was applied post stratification to evaluate interactions between PA and factors associated with CRC. Moreover, the PRS from the UKB was categorized into three genetic risk groups: low (bottom quintile), intermediate (quintiles 2–4), and high (top quintile). The subgroups of PRS were also further evaluated (eSupplement-Methods).

All analyses were conducted using R (version 4.2.0, https://www.r-project.org/) within RStudio (version 4.2.0).

Results

Population characteristics

For this study, the baseline characteristics of eligible individuals were classified on the basis of their PA (Table 1, eSupplement-Table 2). The study included 33 733 participants with T2D. During a median follow-up duration of 13.62 years (interquartile range [IQR] 12.94–14.30), 27.1% (n = 9132) engaged in low PA, 46.9% (n = 15 819) in moderate PA, and 26.0% (n = 8782) in high PA. A total of 551 participants subsequently developed CRC during the observation period. Details on the characteristics of the T2D population can been found in eSupplement-Results.

Table 1

Selected baseline characteristics of UK Biobank participants by PA in the T2D population

Association between physical activity and the risk of CRC

In model II, we found that inverse associations with the risk of CRC in T2D patients for moderate and high PA (P_FDR < 0.050). In model III, the risk of CRC was 19% lower for the moderate PA group (hazard ratio [HR] 0.81; 95% confidence interval [0.66; 0.98]) and 26% lower for the high PA group (HR 0.74 [0.58; 0.94] than in the group with low PA (Table 2). Figure 1 indicates an inverse correlation between high PA and CRC occurrence in the T2D population. Moderate PA is well established as being associated with reduced risk of CRC. Further increasing the amount of PA (beyond 3000 MET-min/week) continues to decrease the incidence rate, but the decrease flattens out and tends to stabilize. Moreover, the cumulative hazard function supported the association between PA and CRC incidence in T2D patients (eSupplement-Figure 3).

Figure 1

Restricted cubic spline

Table 2

Association of PA with CRC at different intestinal sites in the T2D population

Physical activity and the risk of CRC in different bowel segments (eSupplement-Results)

The results for the different sites at which CRC occurred show the same trend as the results of the main analysis above. Especially for RSCC, moderate and high levels of PA were associated with a strong risk reduction after full adjustment of 31% (HR 0.69 [0.51; 0.93]) and 42% (HR 0.58 [0.40; 0.85]) compared with the lower PA in T2D group. There is no effect of PA on the risk of developing RC.

Subgroups and genetic susceptibility

No multiplicative interaction was detected between PA and key confounders in the subgroup analyses (Figure 2). However, we observed a monotonic association between lower PRS and higher PA with a lower risk of CRC. Persons with the lowest PRS and highest PA had the least risk of CRC (HR 0.23 [0.12; 0.41]) (Figure 3). eSupplement-Table 3 shows the association noted between the PRS and CRC risk in the T2D population and illustrates a progressive correlation in this regard. Persons in the T2D population with a high PRS exhibit a higher proportion of white ethnicity and are predominantly male (eSupplement-Table 4). eSupplement-Table 5 displays the association between moderate or high PA and CRC in the PRS-stratified analysis with low PA as reference. Analyses within each genetic risk category defined using the PRS showed that high PA in the group with the highest PRS was associated with a lower risk for CRC , but there was no evidence of interactions.

Figure 2

Stratified analysis

Figure 3

Association between polygenie risk score, physical activity, and colorectal cancer

Sensitivity analysis

See eSupplement-Results.

Discussion

In this study, the fully adjusted model showed that the participants with T2D who had moderate and high levels of PA were at less risk of CRC than those with low PA. Moreover, both moderate and high PA showed protective trends for different sites of CRC. These findings underscore the importance of regular PA in lowering the risk of CRC in persons with T2D, regardless of their genetic predisposition. When the participants were stratified by PRS, high PA was associated with a reduced CRC risk in the subgroup with T2D and high PRS. Persons with low PRS and high PA had the lowest risk of CRC. This shows how important it is not only to consider genetic susceptibility but also to integrate PA-related interventions when developing personalized prevention strategies for CRC in T2D patients.

In recent years, cancer has become the primary factor in diabetes-related mortality (21). The associations between daily habits and CRC in T2D patients have not yet been explored in full. In this large cohort study, we have demonstrated the association between moderate to high PA and a reduced risk of CRC, compared with low PA, in persons with T2D. Our study diverges, in particular, from the findings of Schmid D et al., who observed no association between PA and colon cancer risk in the T2D population without investigating genetic susceptibility (11). Our findings not only confirm and expand upon the conclusions of Han Hee Lee et al., but also highlight the importance of evaluating PA in conjunction with genetic susceptibility (12). Moreover, the results of our analysis show that both moderate and high PA were associated with reduced CRC risk in different bowel segments, with RSCC being the most pronounced. In our experience, the right colon, the proximal colon, is more susceptible to muscular contractions at the level of the ileum and displays more intense propulsion (22). This may reduce the contact time of possible carcinogens with the colonic mucosa (23). However, some studies have indicated that there may be no significant difference between proximal and distal colon cancer. These inconsistent results may be due to differences in research methods (24).

We found an association between PA and CRC in persons with T2D, independent of genetic disease risk. Recent studies support this notion, demonstrating that PA is associated with a reduced risk of CRC regardless of the individual genetic risk (25, 26).

Despite the considerable sample size, we did not find a noteworthy interaction between PRS and PA. This is consistent with recent studies by Chen et al. and Yang et al. (27, 28). Furthermore, our findings indicate that the combination of lower PRS and higher PA level was associated with the lowest risk of CRC. Combining genetic risk assessment with PA interventions enables the provision of more accurate and personalized CRC prevention strategies for high-risk persons.

Various mechanisms for the impact of PA (sex hormones, metabolic hormones, and others) have been proposed (29, 30), with the effects and their magnitudes potentially differing with training intensity (31). The positive effects of PA are attributable to stimulation of the metabolism and remodeling of the molecular structure of skeletal muscle (32). As a result, PA influences both HbA1c and BMI by enhancing insulin sensitivity, increasing glucose uptake in muscles, and promoting fat breakdown through the secretion of lipolytic hormones (33, 34). Relatively vigorous PA can modulate intra-tumor redox homeostasis, thereby reducing tumor cell growth, and is associated with decreased incidence of CRC (35). Especially in patients with T2D, high-intensity PA enhances islet β-cell function, increases insulin sensitivity, and reduces insulin resistance (36, 37, 38).

The notable advantages of this study include the large sample and the comprehensive data resources, enabling a thorough exploration of associations. In-depth evaluation of cancers in different bowel segments permits a more nuanced understanding of how PA impacts CRC development in specific regions of the colon and rectum. Furthermore, this study is the first to examine the effect of the interaction between PA and the PRS on CRC in the T2D population. Nonetheless, certain limitations require consideration. First, given the observational design of this research, no causal conclusions can be drawn. Second, all PA data came from self-reports, which may have introduced a degree of bias. In future research, incorporation of data from wearable devices could enhance the reliability of PA assessments and further validate our findings. Despite detailed documentation of PA at different intensities, further potentially relevant areas of PA were not included in our study, which partly constrains comparability with research focused on those specific domains (39). Additionally, subjective PA data from the IPAQ may deviate from objective energy consumption measures, underlining the need for caution in interpreting the results (40). Third, the predominance of Caucasian participants in the UKB may limit the generalizability of the results to other ethnic populations; however, the large sample size enables reliable estimation of the association between exposure and disease. Fourth, despite the adjustment for numerous confounders, there may still be unmeasured factors that could lead to residual bias. Finally, our sensitivity analysis can only partially reduce the impact of reverse causation, as other health impairments may also lead to changes in PA.

Acknowledgments

The UK Biobank resource is open to all researchers (https://www.ukbiobank.ac.uk). The authors thank the UK Biobank for the access to the data. This research was conducted under Application Number 83339.

Funding

This research was supported by the National Natural Science Foundation of China Regional Innovation and Development Joint Foundation (U23A20408), the National Natural Science Foundation of China (82171698, 81300279, 81741067, 82170561, 82202058, 8217060280), the Program for High-level Foreign Expert Introduction of China (G2022030047L), the Natural Science Foundation for Distinguished Young Scholars of Guangdong Province (2021B1515020003), the Natural Science Foundation of Guangdong Province (2022A1515012081), the Guangzhou Basic and Applied Basic Research Scheme-Project for Pilot Voyage (2024A04J6573), the Foreign Distinguished Teacher Program of Guangdong Science and Technology Department (KD0120220129), the Climbing Program of Introduced Talents and High-level Hospital Construction Project of Guangdong Provincial People’s Hospital (DFJH201803, KJ012019099, KJ012021143, KY012021183). Partial funding was provided by VA Clinical Merit and ASGE clinical research funds (FWL).

Ethical approval
UK Biobank database received ethical clearance from the North West Multi-Center Research Ethics Committee: The approval numbers are: 11/NW/0382, 16/NW/0274, and 21/NW/0157. All participants gave written informed consent before enrolment in the study, which was conducted in accordance with the tenets of the Declaration of Helsinki.

Conflict of interest statement
The authors declare that no conflict of interest exists.

Received on 10 July 2024, revised version accepted on: 16 January 2025

Corresponding authors
Prof. Felix W Leung, MD, [e-mail];
Prof. Chongyang Duan, PhD, [e-mail];
Prof. Weihong Sha, PhD, [e-mail];
Prof. Hao Chen, PhD, [e-mail] 

References

1. Sung H, Ferlay J, Siegel RL, et al.: Global Cancer Statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021; 71: 209–49

2. Kerr J, Anderson C, Lippman SM: Physical activity, sedentary behaviour, diet, and cancer: an update and emerging new evidence. Lancet Oncol 2017; 18: e457–71

3. Yu GH, Li SF, Wei R, Jiang Z: Diabetes and colorectal cancer risk: clinical and therapeutic implications. J Diabetes Res 2022; 2022: 1747326

4. Saeedi P, Petersohn I, Salpea P, et al.: Global and regional diabetes prevalence estimates for 2019 and projections for 2030 and 2045: results from the International Diabetes Federation Diabetes Atlas, 9th edition. Diabetes Res Clin Pract 2019; 157: 107843

5. Ahmad E, Lim S, Lamptey R, Webb DR, Davies MJ: Type 2 diabetes. Lancet 2022; 400: 1803–20

6. Chudasama YV, Khunti KK, Zaccardi F, et al.: Physical activity, multimorbidity, and life expectancy: a UK biobank longitudinal study. BMC Med 2019; 17: 108

7. Pollak M: The insulin and insulin-like growth factor receptor family in neoplasia: an update. Nat Rev Cancer 2012; 12: 159–69

8. Zhang AMY, Wellberg EA, Kopp JL, Johnson JD: Hyperinsulinemia in obesity, inflammation, and cancer. Diabetes Metab J 2021; 45: 285–311

9. Yang IP, Tsai HL, Huang CW, et al.: High blood sugar levels significantly impact the prognosis of colorectal cancer patients through down-regulation of microRNA-16 by targeting Myb and VEGFR2. Oncotarget 2016; 7: 18837–50

10. Xie F, You Y, Huang J, et al.: Association between physical activity and digestive-system cancer: an updated systematic review and meta-analysis. J Sport Health Sci 2021; 10: 4–13

11. Schmid D, Behrens G, Matthews CE, Leitzmann MF: Physical activity and risk of colon cancer in diabetic and nondiabetic US adults. Mayo Clin Proc 2016; 91: 1693–705

12. Lee HH, Lee KN, Kim JS, et al.: Association between regular physical activity and lower incidence of colorectal cancer in patients with diabetes mellitus: a nationwide cohort study. Colorectal Dis 2023; 25: 1588–97

13. Guidelines for data processing and analysis of the International Physical Activity Questionnaire (IPAQ). 2005. www.IPAQ.ki.se-Request PDF. www.researchgate.net/publication/267932370_Guidelines_for_ data_processing_and_analysis_of_the_International_Physical_ Activity_Questionnaire_IPAQ2005_URL_httpwwwIPAQkise (last accessed on 23 September 2024).

14. Craig CL, Marshall AL, Sjöström M, et al.: International physical activity questionnaire: 12-country reliability and validity. Med Sci Sports Exerc 2003; 35: 1381–95

15. Yang LP, Wang ZX, Zhang R, et al.: Association between cigarette smoking and colorectal cancer sidedness: a multi-center big-data platform-based analysis. J Transl Med 2021; 19: 150

16. UK Biobank release and systematic evaluation of optimised polygenic risk scores for 53 diseases and quantitative traits | medRxiv. www.medrxiv.org/content/10.1101/2022.06.16.22276246v2 (last accessed on 15 September 2024).

17. O’Sullivan DE, Sutherland RL, Town S, et al.: Risk factors for early-onset colorectal cancer: a systematic review and meta-analysis. Clin Gastroenterol Hepatol 2022; 20: 1229–40.e5

18. Cheng WJ, Chiang CC, Peng MT, et al.: Chronic obstructive pulmonary disease increases the risk of mortality among patients with colorectal cancer: a nationwide population-based retrospective cohort study. Int J Environ Res Public Health 2021; 18: 8742

19. Bradbury KE, Murphy N, Key TJ: Diet and colorectal cancer in UK biobank: a prospective study. Int J Epidemiol 2020; 49: 246–58

20. Lytras T, Nikolopoulos G, Bonovas S: Statins and the risk of colorectal cancer: an updated systematic review and meta-analysis of 40 studies. World J Gastroenterol 2014; 20: 1858–70

21. Pearson-Stuttard J, Bennett J, Cheng YJ, et al.: Trends in predominant causes of death in individuals with and without diabetes in England from 2001 to 2018: an epidemiological analysis of linked primary care records. Lancet Diabetes Endocrinol 2021; 9: 165–73

22. Moran BJ, Jackson AA: Function of the human colon. Br J Surg 1992; 79: 1132–7

23. Boyle T, Keegel T, Bull F, Heyworth J, Fritschi L: Physical activity and risks of proximal and distal colon cancers: a systematic review and meta-analysis. J Natl Cancer Inst 2012; 104: 1548–61

24. Zhang Y, Cantor KP, Dosemeci M, Lynch CF, Zhu Y, Zheng T: Occupational and leisure-time physical activity and risk of colon cancer by subsite. J Occup Environ Med 2006; 48: 236–43

25. Cho YA, Lee J, Oh JH, et al.: Genetic risk score, combined lifestyle factors and risk of colorectal cancer. Cancer Res Treat 2019; 51: 1033–40

26. Carr PR, Weigl K, Jansen L, et al.: Healthy lifestyle factors associated with lower risk of colorectal cancer irrespective of genetic risk. Gastroenterology 2018; 155: 1805–15.e5

27. Chen X, Guo F, Chang-Claude J, Hoffmeister M, Brenner H: Physical activity, polygenic risk score, and colorectal cancer risk. Cancer Med 2023; 12: 4655–66

28. Yang T, Li X, Farrington SM, et al.: A systematic analysis of interactions between environmental risk factors and genetic variation in susceptibility to colorectal cancer. Cancer Epidemiol Biomarkers Prev 2020; 29: 1145–53

29. Brown JC, Winters-Stone K, Lee A, Schmitz KH: Cancer, physical activity, and exercise. Compr Physiol 2012; 2: 2775–809

30. Song M, Chan AT, Sun J: Influence of the gut microbiome, diet, and environment on risk of colorectal cancer. Gastroenterology 2020; 158: 322–40

31. Morelli C, Avolio E, Galluccio A, et al.: Impact of vigorous-intensity physical activity on body composition parameters, lipid profile markers, and irisin levels in adolescents: a cross-sectional study. Nutrients 2020; 12: 742

32. Papadimitriou N, Dimou N, Tsilidis KK, et al.: Physical activity and risks of breast and colorectal cancer: a mendelian randomisation analysis. Nat Commun 2020; 11: 597

33. Umpierre D, Ribeiro PAB, Kramer CK, et al.: Physical activity advice only or structured exercise training and association with HbA1c levels in type 2 diabetes: a systematic review and meta-analysis. JAMA 2011; 305: 1790–9

34. Pritzlaff CJ, Wideman L, Blumer J, et al.: Catecholamine release, growth hormone secretion, and energy expenditure during exercise vs. recovery in men. J Appl Physiol (1985) 2000; 89: 937–46

35. Assi M, Dufresne S, Rébillard A: Exercise shapes redox signaling in cancer. Redox Biol 2020; 35: 101439

36. Kirwan JP, Solomon TPJ, Wojta DM, Staten MA, Holloszy JO: Effects of 7 days of exercise training on insulin sensitivity and responsiveness in type 2 diabetes mellitus. Am J Physiol Endocrinol Metab 2009; 297: E151–6

37. Heiskanen MA, Motiani KK, Mari A, et al.: Exercise training decreases pancreatic fat content and improves beta cell function regardless of baseline glucose tolerance: a randomised controlled trial. Diabetologia 2018; 61: 1817–28

38. Kanaley JA, Colberg SR, Corcoran MH, et al.: Exercise/physical activity in individuals with type 2 diabetes: a consensus statement from the American College of Sports Medicine. Med Sci Sports Exerc 2022; 54: 353–68

39. Mahmood S, MacInnis RJ, English DR, Karahalios A, Lynch BM: Domain-specific physical activity and sedentary behaviour in relation to colon and rectal cancer risk: a systematic review and meta-analysis. Int J Epidemiol 2017; 46: 1797–813

40. Speakman JR, Selman C: Physical activity and resting metabolic rate. Proc Nutr Soc 2003; 62: 621–34