
==== Front
Am J Gastroenterol
Am J Gastroenterol
AJGAST
ACG
The American Journal of Gastroenterology
0002-9270
1572-0241
Wolters Kluwer Philadelphia, PA

38704818
AJG-24-0638
10.14309/ajg.0000000000002853
00030
3
Article
Colon
Excess Weight, Polygenic Risk Score, and Findings of Colorectal Neoplasms at Screening Colonoscopy
Fu Ruojin MMed 12ruojin.fu@dkfz-heidelberg.de

https://orcid.org/0000-0003-2476-7110
Chen Xuechen PhD 13xuechen.chen@dkfz-heidelberg.de

https://orcid.org/0000-0001-7990-3437
Niedermaier Tobias PhD 1tobias.niedermaier@ibe.med.uni-muenchen.de

https://orcid.org/0009-0003-4475-0583
Seum Teresa MSc 12teresa.seum@dkfz-heidelberg.de

https://orcid.org/0000-0002-8307-3197
Hoffmeister Michael PhD 1m.hoffmeister@Dkfz-Heidelberg.de

https://orcid.org/0000-0002-6129-1572
Brenner Hermann MD, MPH 145
1 Division of Clinical Epidemiology and Aging Research, German Cancer Research Center (DKFZ), Heidelberg, Germany;
2 Medical Faculty Heidelberg, Heidelberg University, Heidelberg, Germany;
3 College of Pharmacy, Jinan University, Guangzhou, China;
4 German Cancer Consortium (DKTK), German Cancer Research Center (DKFZ), Heidelberg, Germany;
5 Division of Preventive Oncology, German Cancer Research Center (DKFZ) and National Center for Tumor Diseases (NCT), Heidelberg, Germany.
Correspondence: Hermann Brenner, MD, MPH. E-mail: h.brenner@dkfz-heidelberg.de.
9 2024
03 5 2024
119 9 19131920
12 3 2024
25 4 2024
Copyright © 2024 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of The American College of Gastroenterology
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article distributed under the terms of the Creative Commons Attribution-Non Commercial-No Derivatives License 4.0 (CCBY-NC-ND), where it is permissible to download and share the work provided it is properly cited. The work cannot be changed in any way or used commercially without permission from the journal.

INTRODUCTION:

Excess weight is an established risk factor of colorectal cancer (CRC). However, evidence is lacking on how its impact varies by polygenic risk at different stages of colorectal carcinogenesis.

METHODS:

We assessed the individual and joint associations of body mass index (BMI) and polygenic risk scores (PRSs) with findings of colorectal neoplasms among 4,784 participants of screening colonoscopy. Adjusted odds ratios (aORs) for excess weight derived by multiple logistic regression were converted to genetic risk equivalents (GREs) to quantify the impact of excess weight compared with genetic predisposition.

RESULTS:

Overweight and obesity (BMI 25–<30 and ≥30 kg/m2) were associated with increased risk of any colorectal neoplasm (aOR [95% confidence interval, CI] 1.26 [1.09–1.45] and 1.47 [1.24–1.75]). Obesity was associated with increased risk of advanced colorectal neoplasm (aOR [95% CI] 1.46 [1.16–1.84]). Dose-response relationships were seen for the PRS (stronger for advanced neoplasms than any neoplasms), with no interaction with BMI, suggesting multiplicative effects of both factors. Obese participants with a PRS in the highest tertile had a 2.3-fold (95% CI 1.7–3.1) and 2.9-fold (95% CI 1.9–4.3) increased risk of any colorectal neoplasm and advanced colorectal neoplasm, respectively. The aOR of obesity translated into a GRE of 38, meaning that its impact was estimated to be equivalent to the risk caused by 38 percentiles higher PRS for colorectal neoplasm.

DISCUSSION:

Excess weight and polygenic risk are associated with increased risk of colorectal neoplasms in a multiplicative manner. Maintaining normal weight is estimated to have an equivalent effect as having 38 percentiles lower PRS.

KEYWORDS:

colorectal neoplasm
body mass index
polygenic risk score
genetic risk equivalent
Deutsche ForschungsgemeinschaftBR1704/16-1 Hermann BrennerBundesministerium für Bildung und Forschung01GL1712 Hermann BrennerDeutsche Krebshilfe70113330 Hermann BrennerOPEN-ACCESSTRUE
SDCT
==== Body
pmcINTRODUCTION

In the past decades, the prevalence of obesity has escalated, reaching a pandemic level (1,2). Being overweight or obese increases the risk of multiple chronic diseases including multiple cancers and is associated with lower quality of life and a reduced life span (3). Excess weight is an established risk factor of colorectal cancer (CRC) (4–6), the third most commonly diagnosed cancer and the second leading cause of cancer-related mortality worldwide (7). The slow progression of most colorectal neoplasms via the adenoma-carcinoma sequence, typically spanning at least 10 years (8), offers opportunities of prevention through lifestyle modifications and early detection and intervention.

Studying how genetic and environmental factors interact with each other can help reveal the mechanisms of colorectal carcinogenesis (9). Furthermore, joint consideration of both factors may be helpful for targeted efforts of prevention. In recent years, polygenic risk scores (PRSs), integrating information from genetic variants found in genome-wide association studies (GWASs), have become useful tools to measure individuals' genetic risk of CRC (10,11). It has been shown that PRSs may complement and substantially improve CRC risk classification based on known factors such as lifestyle and family history (12–14). Besides CRC, PRSs have also been shown to exhibit associations with the prevalence of advanced adenomas, the precursor of most CRCs (15). Previous work has explored individual and joint associations of excess weight and a PRS with the risk of CRC (16). However, it is still unclear how excess weight is associated with the risk of different stages of colorectal carcinogenesis when further considering PRS levels.

We aimed to assess the independent and joint associations of excess weight and a PRS with the prevalence of colorectal neoplasms, including non-advanced and advanced adenomas, among participants of a large screening colonoscopy study. Furthermore, we used the recent measure of genetic risk equivalent (GRE) (17) to quantify the impact of excess weight in terms of the comparable impact of background genetic risk.

METHODS

Study design and population

Our analysis is based on data from the BliTz (Begleitende Evaluierung innovativer Testverfahren zur Darmkrebsfrüherkennung) study. BliTz is an ongoing study conducted among screening colonoscopy participants in southwestern Germany with the primary aim to evaluate diagnostic performance of noninvasive CRC screening tests. In the German screening colonoscopy program, screening colonoscopies are performed by licensed and experienced gastroenterologists, mostly in an outpatient setting. The BliTz study was approved by the Ethics Committees of the responsible state physicians' chambers (Baden-Wuerttemberg, M118-05-f; Rhineland-Palatinate, 837.047.06(5145); Saarland, 217/13; Hesse, MC 254/2007) and Ethics Committees of the Medical Faculty Heidelberg (178/2005) and was registered at the German Clinical Trials Register (DRKS-ID: DRKS00008737). The participants of the current analysis were enrolled between November 2005 and January 2019.

Detailed information on the BliTz study has been reported elsewhere (18,19). In short, participants have been recruited in 20 gastroenterology practices since 2005. Information on socioeconomic status, demographics, lifestyle variables, medical history, and family history of CRC was collected using standardized questionnaires before the colonoscopy. Colonoscopy and histology reports were collected from the collaborating colonoscopists and their affiliated local pathology institutes. Two trained data extractors who were blinded to questionnaire and genetic data independently retrieved findings from these reports. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology guideline for cross-sectional studies.

Assessment of body weight

Body weight was quantified by body mass index (BMI). BMI is defined as a person's weight (kilogram) divided by the square of height (meters). According to the World Health Organization classification, BMI <18.5 kg/m2 is defined as underweight, 18.5–<25 kg/m2 as normal weight, 25–<30 kg/m2 as overweight, and ≥30 kg/m2 as obesity (20). In this study, BMI was derived from self-reported information on height and weight provided in participants' questionnaires. BMI <18.5 kg/m2 was very rare (<0.5%) and combined with the normal weight category (18.5–<25 kg/m2) in our analyses. Eventually, participants were categorized according to BMI levels into 3 groups: <25, 25–<30, and ≥30 kg/m2.

Derivation of PRSs

Genotyping for the included participants was performed by the Global Screening Array (Illumina, San Diego, CA) for 3,882 participants and the Illumina OncoArray-500k V1.0 BeadChip (Illumina) for 902 participants. A PRS was calculated by combining information from 140 CRC-related single nucleotide polymorphisms (SNPs) that were identified in a recent large-scale GWAS (11). The PRS was calculated as the weighted sum of the number of risk alleles for each SNP (0, 1, or 2 copies of the risk allele and the imputed dosages for imputed SNPs, with weight as their GWAS reported beta coefficients for each risk allele in Supplementary Table S1, see Supplementary Digital Content 1, http://links.lww.com/AJG/D282) and was further categorized by tertiles according to the distribution of PRS among the participants with normal colonoscopy result.

Statistical analysis

Participants with complete genotype data and BMI information were included in our analyses. To focus the analyses on associations with relevant neoplastic findings, we excluded participants with hyperplastic polyps, nondefined polyps, or serrated adenomas/polyps <1 cm as the most advanced finding at screening colonoscopy. To enhance the representativeness of the results for a screening population with average risk and to minimize the possibility of overlooked neoplasms, we further excluded participants younger than 50 or 80 years and older, with a history of CRC or inflammatory bowel disease, with a colonoscopy in the preceding 5 years, with inadequate bowel preparation, or with incomplete colonoscopy (cecum not reached).

Participants were first divided into 2 groups according to the most advanced finding in colonoscopy: any neoplasm and no finding. Within the group of participants with any neoplasm, a further distinction was made between those with and without advanced neoplasms. Advanced precancerous lesions included advanced adenomas (i.e., adenomas with at least one of the characteristics: ≥1 cm in size, tubulovillous or villous components, or high-grade dysplasia) and sessile serrated polyps ≥1 cm in size.

The distribution of characteristics between the 2 groups (with and without neoplasm) was described and compared using the χ2 test. Multiple imputations were performed for the missing values of the following variables using the R package mice, version 3.7.0 (21): years of schooling, smoking status, alcohol consumption, physical activity, consumption of red and processed meat, history of hormone replacement therapy in women, history of diabetes, and use of nonsteroidal anti-inflammatory drugs.

Multivariable logistic regression models were used to assess the independent association of BMI and PRS levels with the presence of any colorectal neoplasms and any advanced colorectal neoplasms. Two levels of covariate adjustment were used. All models were adjusted for age and sex. More comprehensively adjusted models additionally included other potentially CRC-related risk factors (8): years of schooling (<10/10–11/≥12), smoking status (never/former/current smoker), physical activity (<30/≥30 min/d), alcohol consumption (none/0–<12/12–<25/25–<50/≥50 g per day), red and processed meat consumption (≥1/<1 time per day), history of hormone replacement therapy in women (yes/no), history of diabetes (yes/no), CRC family history in a first-degree relative (yes/no), use of nonsteroidal anti-inflammatory drugs (yes/no), history of colonoscopy (yes/no), and PRS (for the analysis of BMI) or BMI (for the analysis of PRS). Interaction analyses included a cross-product term of PRS and BMI along with the main effect terms in multivariable regression models. Furthermore, the joint association of PRS and BMI levels on the risk of colorectal neoplasms was assessed using BMI <25 kg/m2 in PRS tertile 1 as the uniform reference.

In addition to odds ratios (ORs) for BMI and PRS, the corresponding GREs were calculated as the ratios of the regression coefficients for both variables in the logistic regression models. Details are provided in the Supplementary Materials (see Supplementary Digital Content 1, http://links.lww.com/AJG/D282). The concept of GREs was developed in analogy with the concept of risk and rate advancement periods (22) and has been used in several recent articles on colorectal cancer risk (17,23). GREs enable us to directly compare risks associated with lifestyle factors and genetic predisposition, and to quantify the extent of genetic risk, expressed in percentiles of the PRS, that may be compensated for by avoiding the risk factor under investigation. Subgroup analyses were conducted for subgroups of participants defined by age, sex, history of colonoscopy, and family history of CRC.

R software (version 4.1.3; R Foundation for Statistical Computing, Vienna, Austria) was used for all analyses. Statistical tests were 2-sided, and P < 0.05 was considered as statistically significant.

RESULTS

Characteristics of the study population

A total of 4,784 participants were included after selection (Figure 1), including 2,224 individuals who exhibited colorectal neoplasms as the most advanced finding in colonoscopy screening (867 with advanced neoplasms, including 811 with advanced precancerous lesions and 56 with CRC) and 2,560 individuals without abnormal findings at screening colonoscopy. The distribution of participants' characteristics is presented in Table 1. Of those with and without neoplasm, respectively, 62.7% and 43.8% were male. The median age for individuals carrying neoplasms was 62 years, 2 years higher than among those without neoplasms. Compared with the participants without neoplasms, those with colorectal neoplasms were more likely to have lower levels of education, to be overweight or obese, to smoke, to have a higher PRS, to consume more alcohol, to consume more red and processed meat, and to have diabetes. In addition, fewer among the participants carrying neoplasms had a history of colonoscopy.

Figure 1. Flowchart of selection of study participants.

Table 1. Baseline characteristics of the study population

Associations of BMI and PRS with colorectal neoplasm risk

The individual associations of BMI and PRS with the presence of colorectal neoplasms are presented in Table 2. Overweight and obesity were associated with 26% and 47%, respectively, increased risk of colorectal neoplasm compared with BMI <25 kg/m2 (BMI 25–<30 vs <25 kg/m2: aOR [95% confidence interval, CI] 1.26 [1.09–1.45]; BMI ≥30 vs <25 kg/m2: aOR [95% CI] 1.47 [1.24–1.75]). Participants with obesity showed a comparable increased risk of carrying advanced neoplasms. Individuals in the highest tertile of PRS exhibited an aOR (95% CI) of 1.93 (1.67–2.24) for colorectal neoplasms compared with those in the lowest PRS tertile, and the corresponding risk was even higher for advanced neoplasm. No statistically significant interaction was observed between BMI categories and PRS levels.

Table 2. Individual association of BMI and polygenic risk score with colorectal neoplasms risk

Table 3 presents the association of PRS with the risk of colorectal neoplasms, stratified by BMI levels. Across all comparisons, higher PRS tertiles were associated with increased risk compared with the lowest PRS tertile. The joint association of BMI and PRS with colorectal neoplasm risk, using the combination of low BMI and low PRS as reference, is presented in Table 4. Those with a BMI ≥30 kg/m2 in the highest PRS tertile exhibited 2.3- and 2.9-fold increased risk of carrying any colorectal neoplasm or any advanced colorectal neoplasm, respectively. The corresponding risk of carrying an advanced neoplasm was generally higher when comparing the advanced neoplasm group with no finding group.

Table 3. Joint association of BMI and polygenic risk score with colorectal neoplasms risk

Table 4. Odds ratios (95% confidence intervals) of the joint association of BMI levels and polygenic risk score with colorectal neoplasms risk, using BMI <25 in the lowest PRS tertile as the reference group

Genetic risk equivalents for comparisons between BMI levels

Table 5 presents the estimates of GREs for overweight and obesity compared with normal weight. The estimated effects of obesity on carrying any neoplasm and advanced neoplasm were equivalent to the effect of having 38 or 28 percentiles higher PRS, respectively (GRE [95% CI] 38.4 [19.8–57.0]; and 28.1 [10.2–46.1], respectively). To further explore the GREs in different subgroups, a stratified analysis was conducted considering factors such as sex, age, history of colonoscopy, and family history of CRC. The results for the risk of carrying any neoplasm are shown in Supplementary Figure S1 and Supplementary Tables S2–S5 (see Supplementary Digital Content 1, http://links.lww.com/AJG/D282). The GREs of obesity were constantly higher than that of overweight in most subgroups, except for the participants with a history of colonoscopy. The interactions between BMI and the relevant stratification factors were not statistically significant in any of the subgroup analyses.

Table 5. Genetic risk equivalent for comparisons between BMI categories

DISCUSSION

In this large CRC screening study, we found that higher BMI and PRS levels were independently associated with an elevated risk of colorectal neoplasms. Associations of elevated BMI were approximately equally strong with any neoplasms and advanced neoplasms. The PRS showed particularly strong associations with the presence of advanced neoplasms. Although no significant interactions between BMI and PRS were observed, their multiplicative effects resulted in strongly increased risks among participants with both high BMI and high PRS in models jointly considering both factors. Joint consideration of BMI and PRS levels could thereby help better stratify the risk of colorectal neoplasms. The association of obesity with carrying colorectal neoplasms translated into a GRE of 38, meaning that the impact of obesity compared with being normal weight was equivalent to having a 38 percentile higher polygenic risk, which further underlines the large potential of reducing genetically determined risk by maintaining a normal body weight.

Our results of the individual association of overweight and obesity with the risk of colorectal neoplasms are largely in line with those of previous studies. A meta-analysis based on 23 studies published before 2012 derived summary odds ratios (95% CI) of 1.21 (1.07–1.38) and 1.32 (1.18–1.48) for the association between overweight and obesity and colorectal adenomas, respectively, with a significant dose-response relationship (P < 0.01) (24). A later meta-analysis published in 2018, which focused on studies with investigator-measured BMI found somewhat stronger associations, with summary odds ratios of 1.44 (1.30–1.61) and 1.42 (1.24–1.63), respectively, possibly due to more precise measurements of height and weight (25).

Even though excess weight is an established risk factor of CRC, the mechanisms on how it promotes colorectal carcinogenesis are not yet fully understood. Potential carcinogenic mechanisms include insulin resistance and high levels of insulin-like growth factor 1, sex steroids biosynthesis, high levels of leptin and low circulating levels of adiponectin, and chronic inflammation (26,27). There is increasing evidence that visceral adiposity, which may be better reflected in other measures of excess weight, such as the waist-to-hip ratio, may be particularly relevant for some of these mechanisms, especially among women (28). Like in most previous studies assessing the association between overweight and risk of colorectal neoplasms, such measures were not available in our study, suggesting that the impact of excess weight on the development of colorectal neoplasms may even be larger than suggested by previous studies and our study.

The PRS used in our study has been derived and validated for CRC by a large colorectal cancer GWAS consortium (11). In agreement with previous findings (15,29), the PRS was found to be also strongly related to the full range of colorectal neoplasms, the vast majority of which are colorectal adenomas, the precursors of most CRCs. This suggests that genetic predisposition plays a major role throughout the process of carcinogenesis. Nevertheless, the somewhat stronger association found with advanced colorectal neoplasms may also suggest the role of genetic predisposition not only for initial development of neoplasms but also for their further progression.

To our knowledge, our study is the first to evaluate the individual and joint impact of both excess weight and PRS levels on the prevalence of colorectal neoplasms in a colonoscopy screening population and to derive GREs as a comparative measure of their impact which may be helpful for risk communication. Our analysis demonstrates that the joint consideration of both determinants enables enhanced risk stratification compared with risk stratification by PRS alone. In particular, the strongly increased risk in the presence of both obesity and a high PRS suggests that maintaining a normal weight would be particularly important and beneficial for those with high PRS levels to reduce their CRC risk. The high GREs derived in our study may be helpful to communicate this message. They imply that even a substantially increased genetic risk can be compensated to a large extent by maintaining a normal weight, ideally in combination with other healthy lifestyle factors. Our study shows that such preventive effects are not only relevant in later stages of carcinogenesis, but already play a major role early in the early development of colorectal neoplasms. Our study thereby complements and expands evidence on GREs derived for CRC risk in recent work (23,30,31). Deriving and communicating GREs for CRC precursors may help to strengthen awareness of maintaining a healthy lifestyle as a key for preventing CRC (besides many other chronic diseases), in particular in case of an unfavorable genetic predisposition.

Our study has several major strengths. We were able to include a large cohort of participants all of whom underwent screening colonoscopy, allowing a clear distinction between participants who were free of colorectal neoplasms and those who carried neoplasms at various stages of colorectal carcinogenesis. Quantification of genetic risk by the PRS was available for the 4,784 participants included in the study. The study had not been included in the GWAS by which the PRS was derived, and therefore, risk prediction by the PRS is not subject to the winner's curse. All associations were carefully adjusted for the established CRC risk factors by multivariable analyses. Nevertheless, a number of limitations also need to be kept in mind. Despite the multivariable adjustment, we cannot rule out residual confounding by unmeasured or imperfectly measured confounders. Self-reported weight and height at a single point of time most likely led to major imprecision in quantifying the relevant lifetime exposure to excess weight and its impact on colorectal carcinogenesis. The derived ORs and GREs should, therefore, be considered as conservative estimates of a likely stronger impact of excess weight. Our study was furthermore conducted in an almost exclusively White population, which limits generalizability.

Despite these limitations, our study provides information on the individual and joint contributions of excess weight and genetic predisposition which may be valuable for guiding preventive efforts and to communicate the potential of maintaining a normal weight in CRC prevention. Further research should aim for refined measurements of the relevant exposure during the life course to more fully disclose the potential for prevention and for better targeting preventive efforts which may not only be beneficial for CRC prevention but also for the prevention of a broad range of other chronic diseases whose risk is increased by excess weight.

CONFLICTS OF INTEREST

Guarantor of the article: Hermann Brenner, MD, MPH.

Specific author contributions: H.B.: study concept and design. R.F. and X.C.: analysis of data. R.F., X.C., T.N., T.S., M.H. and H.B.: interpretation of data. R.F. and H.B.: drafting the manuscript. All authors provided comments, revised the draft and approved the final version of the manuscript.

Financial support: This study was partly funded by grants from the German Research Council (DFG, grant no. BR1704/16-1), the Federal Ministry of Education and Research (BMBF, grant no. 01GL1712), and the German Cancer Aid (no. 70113330). The funding source had no role in any aspect pertinent to the study.

Potential competing interests: The authors declare no conflict of interest.Study Highlights

WHAT IS KNOWN

✓ Excess weight and polygenic risk are established risk factors of colorectal cancer.

WHAT IS NEW HERE

✓ Excess weight and polygenic risk independently contribute to the risk of different stages of colorectal carcinogenesis.

✓ Joint consideration of both body mass index and polygenic risk score may enhance risk stratification for colorectal cancer screening.

✓ Obesity is estimated to have an equivalent effect of having a 38 percentile higher polygenic risk score.

Supplementary Material

SUPPLEMENTARY MATERIAL accompanies this paper at http://links.lww.com/AJG/D282
==== Refs
REFERENCES

1. Blüher M . Obesity: Global epidemiology and pathogenesis. Nat Rev Endocrinol 2019;15 (5 ):288–98.30814686
2. Chong B Jayabaskaran J Kong G Trends and predictions of malnutrition and obesity in 204 countries and territories: An analysis of the Global Burden of Disease Study 2019. EClinicalMedicine 2023;57 :101850.36864983
3. MacEwan JP Chiu K Ahmad NN Clinical, economic, and health-related quality of life outcomes in patients with overweight or obesity in the United States: 2016–2018. Obes Sci Pract 2024;10 (1 ):e726.38263999
4. Hanyuda A Cao Y Hamada T Body mass index and risk of colorectal carcinoma subtypes classified by tumor differentiation status. Eur J Epidemiol 2017;32 (5 ):393–407.28510098
5. Li X Jansen L Chang-Claude J . Risk of colorectal cancer associated with lifetime excess weight. JAMA Oncol 2022;8 (5 ):730–7.35297997
6. Mandic M Safizadeh F Niedermaier T . Association of overweight, obesity, and recent weight loss with colorectal cancer risk. JAMA Netw Open 2023;6 (4 ):e239556.37083659
7. Ferlay J Ervik M Lam F . Global Cancer Observatory: Cancer Today. International Agency for Research on Cancer: Lyon, France, 2024. (https://gco.iarc.who.int/today). Accessed February 23, 2024.
8. Keum N Giovannucci E . Global burden of colorectal cancer: Emerging trends, risk factors and prevention strategies. Nat Rev Gastroenterol Hepatol 2019;16 (12 ):713–32.31455888
9. Hunter DJ . Gene-environment interactions in human diseases. Nat Rev Genet 2005;6 (4 ):287–98.15803198
10. Huyghe JR Bien SA Harrison TA Discovery of common and rare genetic risk variants for colorectal cancer. Nat Genet 2019;51 (1 ):76–87.30510241
11. Thomas M Sakoda LC Hoffmeister M Genome-wide modeling of polygenic risk score in colorectal cancer risk. Am J Hum Genet 2020;107 (3 ):432–44.32758450
12. Weigl K Chang-Claude J Knebel P . Strongly enhanced colorectal cancer risk stratification by combining family history and genetic risk score. Clin Epidemiol 2018;10 :143–52.29403313
13. Jeon J Du M Schoen RE Determining risk of colorectal cancer and starting age of screening based on lifestyle, environmental, and genetic factors. Gastroenterology 2018;154 (8 ):2152–64.e19.29458155
14. Carr PR Weigl K Edelmann D Estimation of absolute risk of colorectal cancer based on healthy lifestyle, genetic risk, and colonoscopy status in a population-based study. Gastroenterology 2020;159 (1 ):129–38.e9.32179093
15. Weigl K Thomsen H Balavarca Y . Genetic risk score is associated with prevalence of advanced neoplasms in a colorectal cancer screening population. Gastroenterology 2018;155 (1 ):88–98.e10.29574091
16. Chen X Li H Mandic M . Assessment of body mass index, polygenic risk score, and development of colorectal cancer. JAMA Netw Open 2022;5 (12 ):e2248447.36547977
17. Chen X Li H Guo F . Alcohol consumption, polygenic risk score, and early- and late-onset colorectal cancer risk. EClinicalMedicine 2022;49 :101460.35747198
18. Gies A Cuk K Schrotz-King P . Direct comparison of diagnostic performance of 9 quantitative fecal immunochemical tests for colorectal cancer screening. Gastroenterology 2018;154 (1 ):93–104.28958859
19. Peng L Balavarca Y Weigl K . Head-to-head comparison of the performance of 17 risk models for predicting presence of advanced neoplasms in colorectal cancer screening. Am J Gastroenterol 2019;114 (9 ):1520–30.31464746
20. González-Muniesa P Mártinez-González MA Hu FB Obesity. Nat Rev Dis Primer 2017;3 :17034.
21. van Buuren S Groothuis-Oudshoorn K . mice: Multivariate imputation by chained equations in R. J Stat Softw 2011;45 :1–67.
22. Brenner H Gefeller O Greenland S . Risk and rate advancement periods as measures of exposure impact on the occurrence of chronic diseases. Epidemiol Camb Mass 1993;4 (3 ):229–36.
23. Chen X Jansen L Guo F . Smoking, genetic predisposition, and colorectal cancer risk. Clin Transl Gastroenterol 2021;12 (3 ):e00317.33646204
24. Okabayashi K Ashrafian H Hasegawa H Body mass index category as a risk factor for colorectal adenomas: A systematic review and meta-analysis. Am J Gastroenterol 2012;107 (8 ):1175–85; quiz 1186.22733302
25. Wong MCS Chan CH Cheung W Association between investigator-measured body-mass index and colorectal adenoma: A systematic review and meta-analysis of 168,201 subjects. Eur J Epidemiol 2018;33 (1 ):15–26.29288474
26. Bardou M Barkun AN Martel M . Obesity and colorectal cancer. Gut 2013;62 (6 ):933–47.23481261
27. de Andrade Mesquita L Wayerbacher LF Schwartsmann G . Obesity, diabetes, and cancer: Epidemiology, pathophysiology, and potential interventions. Arch Endocrinol Metab 2023;67 (6 ):e000647.37364149
28. Bull CJ Bell JA Murphy N Adiposity, metabolites, and colorectal cancer risk: Mendelian randomization study. BMC Med 2020;18 (1 ):396.33327948
29. Obón-Santacana M Díez-Villanueva A Alonso MH Polygenic risk score across distinct colorectal cancer screening outcomes: From premalignant polyps to colorectal cancer. BMC Med 2021;19 (1 ):261.34743725
30. Chen X Ding J Li H . The power of a healthy lifestyle for cancer prevention: The example of colorectal cancer. Cancer Biol Med 2022;19 (11 ):1586–97.36476570
31. Chen X Guo F Chang-Claude J . Physical activity, polygenic risk score, and colorectal cancer risk. Cancer Med 2023;12 (4 ):4655–66.35891576
