Body shape, fat distribution and sarcopenia and risk of liver cancer in European population: a Mendelian randomization study

doi:10.21203/rs.3.rs-5312354/v1

Download PDF

Research Article

Body shape, fat distribution and sarcopenia and risk of liver cancer in European population: a Mendelian randomization study

https://doi.org/10.21203/rs.3.rs-5312354/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Body shape, fat and muscle are likely associated with risk of liver cancer. Evidence for the impact of these factor is limited and controversial. Because conventional observational studies cannot sidestep the effect of confounding and reverse causality, it remains unclear whether there is a causal relationship of body shape, composition and its distribution with risk of liver cancer.

Methods

In this study, a two-sample Mendelian randomization (MR) approach was applied to evaluate the potential causal association of 17 risk factors of body shape, fat distribution and sarcopenia with risk of liver cancer in European population. Summary genetic association estimates for 17 exposures and liver cancer were obtained from MRCIEU GWAS database.

Results

MR analysis indicated that genetically predicted body mass index (BMI) and waist-to-hip ratio (WHR) were associated with liver cancer risk [OR: 1.0005, P = 0.037; OR: 1.0014, P = 0.004, respectively]. Whole body fat mass, trunk fat mass, arm fat mass and leg fat mass were associated with liver cancer risk, while the corresponding fat-free mass were not associated with liver cancer risk. Genetically determined sarcopenia was not associated with liver cancer risk, as either. There did not suggest evidence of potential heterogeneity or directional pleiotropy.

Conclusion

Our study found genetically predicted BMI, WHR and fat mass were associated with liver cancer risk, and the positive association between fat mass and liver cancer risk did not change with changes in fat distribution. However, fat-free mass and sarcopenia associated factors were not associated with liver cancer risk.

genome-wide association study

fat

sarcopenia

liver cancer

Mendelian randomization

single-nucleotide polymorphisms

The incidence of liver cancer is growing rapidly and it remains a global health challenge [1, 2]. Although age, gender, hepatitis virus infection, non-alcoholic steatohepatitis (NASH) and lifestyle such as alcohol consumption are established risk factors for liver cancer, its etiology and pathogenesis remain to be fully elucidated.

In the past decades, obesity has been suggested to be associated with many cancer risk and skeletal muscles have been considered as endocrine organs which can release myokines to inhibiting cancer progression [3–5]. Body mass index (BMI), a representative indicator of body shape and the most commonly measured marker, shown strong correlation with many cancers risk including liver cancer [6]. However, BMI cannot distinguish whether the body is composed of muscle or fat, and cannot reflect the fat type and distribution, as either. Additionally, muscle strength is not equivalent to muscle mass to some degree. Therefore, more factors including body shape, fat/muscle mass and distribution should be included to assess comprehensively the association with liver cancer risk. Moreover, because conventional observational studies are susceptible to potential bias such as unmeasured confounders and reverse causality, it remains unclear whether the association of body index with liver cancer risk is causal or not.

Mendelian randomization (MR) analysis is a genetic epidemiological method. It used instrumental variables (IVs) such as single-nucleotide polymorphisms (SNPs) as proxies for a risk factor, and thus exploring the evident for a causal effect of a given exposure on an outcome [7]. There are three key hypotheses that must be met: 1) a strong correlation between IVs and exposure; 2) IVs are independent of confounding factors between exposure and outcome; 3) IVs influence outcome directly through exposure [8, 9]. MR studies are less vulnerable to reverse causality bias and are generally uncorrelated with confounding factors such as socioeconomic position and lifestyle factor [10]. Therefore, in this study, we aimed to evaluate potential causal association of 17 body index relevant exposures including body size, fat and muscle with risk of liver cancer, using the two-sample MR study design.

2.1 Outcome data source

Summary level statistics of genetic association estimates for liver & bile duct cancer was obtained from a genome-wide association study (GWAS) including 350 cases and 372,016 controls of European ancestry conducted by the UK Biobank consortium. Full details of the study are available from MRCIEU GWAS database (ieu-b-4915, https://gwas.mrcieu.ac.uk/). The study was approved by the relevant ethics committee, and informed consent was obtained from all participants.

2.2 Exposures data source

There are seventeen exposures which can be classified into four types including body shape exposures, fat associated exposures, fat free associated exposures and sarcopenia associated exposures. Body mass index, waist-to-hip ratio (WHR), body fat/fat-free mass, hand grip strength and appendicular lean mass were analyzed. Summary genetic association estimates for above exposures were obtained from MRCIEU GWAS database, which were listed in Table 1.

Table 1

Characters of exposure
Exposure	GWAS ID	Consortium	Sample size	Population
Body shape exposure
Body mass index	ieu-b-40	GIANT	681,275	European
Waist-to-hip ratio	ieu-a-79	GIANT	210,082	European
Fat associated exposure
Whole body fat mass	ukb-b-19393	MRC-IEU	454,137	European
Trunk fat mass	ukb-b-20044	MRC-IEU	454,588	European
Leg fat mass (left)	ukb-b-7212	MRC-IEU	454,823	European
Leg fat mass (right)	ukb-b-18096	MRC-IEU	454,846	European
Arm fat mass (right)	ukb-b-6704	MRC-IEU	454,757	European
Arm fat mass (left)	ukb-b-8338	MRC-IEU	454,684	European
Fat-free associated exposure
Whole body fat-free mass	ukb-b-13354	MRC-IEU	454,850	European
Trunk fat-free mass	ukb-b-17409	MRC-IEU	454,508	European
Leg fat-free mass (right)	ukb-b-12828	MRC-IEU	454,835	European
Leg fat-free mass (left)	ukb-b-16099	MRC-IEU	454,805	European
Arm fat-free mass (right)	ukb-b-19520	MRC-IEU	454,753	European
Arm fat-free mass (left)	ukb-b-19925	MRC-IEU	454,672	European
Muscle mass associated exposure
Appendicular lean mass	ebi-a-GCST90000025	NA	450,243	European
Hand grip strength (right)	ukb-b-10215	MRC-IEU	461,089	European
Hand grip strength (left)	ukb-b-7478	MRC-IEU	461,026	European
Abbreviations: GWAS, genome-wide association study.

The population used in above studies was European. Each participating study was approved by the relevant ethics committee, and informed consent was obtained from all participants.

2.3 Statistical analysis

Statistical analyses were performed using “TwoSampleMR” in R version 4.2.3. SNPs from independent loci associated with exposure at genome-wide significance threshold of P < 5×10⁸ were included as IVs for the MR analysis. The linkage disequilibrium (LD) threshold was set to r² = 0.001 within a distance of 10,000 kb. To assess the potential causal relationship of above exposures with risk of liver cancer, we combined the ratio

estimates using the inverse-variance weighted (IVW) method in a random-effects model [8]. The heterogeneity was also tested, and if the P value < 0.05, it indicates that there exists potential heterogeneity. To assess the presence of pleiotropy, we performed MR Egger regression. In MR-Egger regression, if the P value for the MR-Egger intercept < 0.05, it indicates that there exist directional pleiotropic effects [10]. The MR leave-one-out analysis was also used for pleiotropy.

An overall design of the present study is shown in Fig. 1. In body shape exposures, MR analysis indicated that genetically predicted BMI and WHR were associated with liver cancer risk [odd ratio (OR): 1.0005, 95% CI: 1.0000 to 1.0009, and P = 0.037; OR: 1.0014, 95% CI: 1.0004 to 1.0023, and P = 0.004, respectively] using the IVW method. There did not exist potential heterogeneity (P = 0.895 and 0.601, respectively) and MR-Egger regression analysis did not suggest evidence of potential directional pleiotropy (P value for intercept = 0.454 and 0.997, respectively).

In fat associated exposures, whole body fat mass was associated with liver cancer risk (OR: 1.0006, 95% CI: 1.0002 to 1.0011, and P = 0.008) using the IVW method. There did not suggest evidence of potential heterogeneity or directional pleiotropy. The MR results of trunk fat mass, arm fat mass and leg fat mass were consistent with the result of whole body fat mass. The scatter plots of body shape and fat-associated exposures were shown in Fig. 2. In fat-free associated exposures, MR analysis indicated that genetically predicted whole body fat-free mass were not associated with liver cancer risk (OR: 1.0005, 95% CI:0.9999 to 1.0010, and P = 0.106). The MR results of trunk fat-free mass, arm fat-free mass and leg fat-free mass were not associated with liver cancer risk, as either.

Finally, sarcopenia associated exposures including appendicular lean mass and hand grip strength (right/left) were analyzed and the IVW method showed that above exposures were not causally associated with liver cancer risk (P = 0.362, 0.422 and 0.481, respectively). The detail results were shown in Table 2.

Table 2

The effect estimates, test of heterogeneity and test of pleiotropy of 17 exposures on liver cancer
Exposure	nSNP	OR (95% CI)	P for association	P for heterogeneity test	P for MR-Egger intercept
Body shape exposure
Body mass index	491	1.0005 (1.0000-1.0009)	0.037	0.895	0.454
Waist-to-hip ratio	36	1.0014 (1.0004–1.0023)	0.004	0.601	0.997
Fat associated exposure
Whole body fat mass	374	1.0006 (1.0002–1.0011)	0.008	0.297	0.609
Trunk fat mass	360	1.0006 (1.0001–1.0010)	0.011	0.620	0.514
Leg fat mass (left)	365	1.0009 (1.0003–1.0014)	0.004	0.470	0.607
Leg fat mass (right)	359	1.0009 (1.0003–1.0015)	0.003	0.240	0.676
Arm fat mass (right)	377	1.0007 (1.0003–1.0012)	0.002	0.646	0.931
Arm fat mass (left)	367	1.0007 (1.0002–1.0012)	0.003	0.715	0.942
Fat-free associated exposure
Whole body fat-free mass	473	1.0005 (0.9999–1.0012)	0.106	0.648	0.383
Trunk fat-free mass	479	1.0004 (0.9999–1.0010)	0.150	0.357	0.377
Leg fat-free mass (right)	429	1.0002 (0.9996–1.0008)	0.474	0.705	0.392
Leg fat-free mass (left)	430	1.0003 (0.9997–1.0009)	0.336	0.562	0.043
Arm fat-free mass (right)	442	1.0004 (0.9997–1.0010)	0.262	0.426	0.979
Arm fat-free mass (left)	450	1.0004 (0.9998–1.0010)	0.177	0.626	0.590
Muscle mass associated exposure
Appendicular lean mass	149	0.9995 (0.9984–1.0007)	0.422	0.892	0.397
Hand grip strength (right)	140	0.9996 (0.9985–1.0007)	0.481	0.887	0.211
Hand grip strength (left)	546	1.0002 (0.9998–1.0005)	0.363	0.069	0.085
Abbreviations: SNP, single-nucleotide polymorphisms; OR, ddd ratio; CI, confidence interval; MR, Mendelian randomization.

In this two-sample MR study, we used 17 exposures related with body shape, fat and muscle, and we found that genetically determined BMI and WHR were associated with liver cancer risk. The exposures of fat all over the body mass were associated with liver cancer risk, while there did not show evident for a causal effect of body fat-free mass on liver cancer risk. Moreover, our study did not provide evidence for the causal effect of sarcopenia on liver cancer risk. Sensitivity analyses using alternative MR methods produced similar results.

In past decades, body shape phenotype was an acknowledged factor associated with cancer risk [11]. BMI, a representative indicator of obesity and the most commonly and easily measured marker, was widely studied and regarded as an established risk factor for several malignancies [12]. According to a population-based cohort study of 5.24 million UK adults, BMI was positively associated with liver risk (HR: 1.19, 95CI: 1.12 to 1.27) [6]. And BMI had a linear positive association with incidence of liver cancer in men based on a prospective cohort of 54,725 Finns [13]. MR studies can evaluate the potential causal effect of exposure on outcome, and are less vulnerable to reverse causality bias. Genetically predicted BMI was associated with an increased risk of liver cancer using MR method (OR: 1.13, 95% CI: 1.03 to 1.25, p = 0.012) [14], which was consistent with our result. However, BMI cannot reflect the fat type, fat distribution and muscle strength. Therefore, exposures related to fat/fat-free mass and distribution were also analyzed in our study. WHR is an important index to determine central obesity. Prospective cohort studies found WHR was related with liver cancer incidence [5, 15]. A recent study showed that fat accumulation throughout the body including waist, trunk, arm and leg was all associated with increased liver cancer risk [16]. Previous public health recommendations have reported that obesity was a risk factor for cancer prevention [17], and the result of our study was consistent with this prevailing view and first conclude that the accumulation of fat anywhere in the body was a genetic risk factor for liver cancer risk.

Additionally, current physical activity guidelines aim to increase and maintain muscle strength for cancer prevention [18]. Reduced muscle strength was associated with an increased risk of mortality in many studies [19, 20]. However, the number of studies of the association between muscle strength on risk of cancer was limited and the association was still controversial. A prospective cohort study found thar absolute grip strength was inversely and linearly associated with liver cancer risk (HR: 0.86, 95% CI: 0.79; 0.93, P < 0.001) [21]. The Prospective Urban-Rural Epidemiology (PURE) study, a longitudinal population study done in 17 countries, found the association between cancer risk and grip strength in high-income countries (0.916, 0.880–0.953; p < 0.0001), while this association was not found in middle-income and low-income countries [22]. However, a meta-analysis which included 42 studies did not show an association between hand grip strength and overall cancer [23]. In our study, that genetically predicted sarcopenia was not associated with liver cancer risk, suggesting that sarcopenia is unlikely to be a causal factor in the development of liver cancer.

The main strength of our study compared with the conventional observational study was that those studies cannot sidestep the confounding and reverse causation bias, while using MR approach can minimize such influence. However, there are some limitations. Firstly, because the ancestry of participants in this study was restricted to European populations, our results may not be suitable to be extrapolated to other populations. However, this may also minimize the bias caused by population stratification. Secondly, although the UK Biobank is a large prospective cohort, the number of liver cancer cases is relevant limited compared with other cancer types, which awaits the future expansion of this cohort and new GWAS. Another limitation is canalization, which means the genetic variations on normal development will be damped or buffered by compensatory developmental processes [7]. Therefore, longitudinal studies were also needed to clarify the exact role of body shape, fat and muscle in the development of liver cancer.

In conclusion, we found that genetically predicted BMI, WHR, whole fat mass, trunk fat mass, arm fat mass and leg fat mass were associated with liver cancer risk. This correlation between fat mass and liver cancer risk did not show fat distribution specificity. However, fat-free mass and sarcopenia associated factors were not associated with liver cancer risk.

Conflict of interest

The authors declare that they have no competing interests.

Ethics approval and consent to participate

Not applicable.

Author Contribution

Participated in research design: Chiyu He, Peiru Zhang, Huigang Li, Di Lu , Zuyuan LinParticipated in the writing of the paper: Xinyu YangParticipated in the performance of the research: Zhihang Hu, Zuyuan Lin,Wei Shen ,Hao ChenParticipated in data analysis: Chiyu He, Jinyan Chen,Reviewing / editing the paper: Di Lu, Shusen Zheng, Xiao Xu, Xuyong Wei, Li Zhuang

Acknowledgements

This work was supported by National Key Research and Development Program of China (No. 2021YFA1100500), the Young Program of National Natural Science Funds (No. 82000617). We thank all members of Key Laboratory of Integrated Oncology and Intelligent Medicine of Zhejiang Province for helpful suggestions regarding the manuscript.

Forner A, Reig M, Bruix J. Hepatocellular carcinoma. Lancet 2018;391:1301-1314.
Llovet JM, Kelley RK, Villanueva A, Singal AG, Pikarsky E, Roayaie S, et al. Hepatocellular carcinoma. Nat Rev Dis Primers 2021;7:6.
Avgerinos KI, Spyrou N, Mantzoros CS, Dalamaga M. Obesity and cancer risk: Emerging biological mechanisms and perspectives. Metabolism 2019;92:121-135.
Huang Q, Wu M, Wu X, Zhang Y, Xia Y. Muscle-to-tumor crosstalk: The effect of exercise-induced myokine on cancer progression. Biochim Biophys Acta Rev Cancer 2022;1877:188761.
Parra-Soto S, Cowley ES, Rezende LFM, Ferreccio C, Mathers JC, Pell JP, et al. Associations of six adiposity-related markers with incidence and mortality from 24 cancers-findings from the UK Biobank prospective cohort study. BMC Med 2021;19:7.
Bhaskaran K, Douglas I, Forbes H, dos-Santos-Silva I, Leon DA, Smeeth L. Body-mass index and risk of 22 specific cancers: a population-based cohort study of 5.24 million UK adults. Lancet 2014;384:755-765.
Smith GD, Ebrahim S. 'Mendelian randomization': can genetic epidemiology contribute to understanding environmental determinants of disease? Int J Epidemiol 2003;32:1-22.
Burgess S, Butterworth A, Thompson SG. Mendelian randomization analysis with multiple genetic variants using summarized data. Genet Epidemiol 2013;37:658-665.
Lawlor DA, Harbord RM, Sterne JA, Timpson N, Davey Smith G. Mendelian randomization: using genes as instruments for making causal inferences in epidemiology. Stat Med 2008;27:1133-1163.
Bowden J, Davey Smith G, Burgess S. Mendelian randomization with invalid instruments: effect estimation and bias detection through Egger regression. Int J Epidemiol 2015;44:512-525.
Sedlmeier AM, Viallon V, Ferrari P, Peruchet-Noray L, Fontvieille E, Amadou A, et al. Body shape phenotypes of multiple anthropometric traits and cancer risk: a multi-national cohort study. Br J Cancer 2023;128:594-605.
Renehan AG, Tyson M, Egger M, Heller RF, Zwahlen M. Body-mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies. Lancet 2008;371:569-578.
Song X, Pukkala E, Dyba T, Tuomilehto J, Moltchanov V, Mannisto S, et al. Body mass index and cancer incidence: the FINRISK study. Eur J Epidemiol 2014;29:477-487.
Vithayathil M, Carter P, Kar S, Mason AM, Burgess S, Larsson SC. Body size and composition and risk of site-specific cancers in the UK Biobank and large international consortia: A mendelian randomisation study. PLoS Med 2021;18:e1003706.
Yan LJ, Yang LS, Yan YC, Tan SY, Ding ZN, Liu H, et al. Anthropometric indicators of adiposity and risk of primary liver cancer: A systematic review and dose-response meta-analysis. Eur J Cancer 2023;185:150-163.
Rask-Andersen M, Ivansson E, Hoglund J, Ek WE, Karlsson T, Johansson A. Adiposity and sex-specific cancer risk. Cancer Cell 2023;41:1186-1197 e1184.
Lauby-Secretan B, Scoccianti C, Loomis D, Grosse Y, Bianchini F, Straif K, et al. Body Fatness and Cancer--Viewpoint of the IARC Working Group. N Engl J Med 2016;375:794-798.
Rock CL, Thomson C, Gansler T, Gapstur SM, McCullough ML, Patel AV, et al. American Cancer Society guideline for diet and physical activity for cancer prevention. CA Cancer J Clin 2020;70:245-271.
Celis-Morales CA, Welsh P, Lyall DM, Steell L, Petermann F, Anderson J, et al. Associations of grip strength with cardiovascular, respiratory, and cancer outcomes and all cause mortality: prospective cohort study of half a million UK Biobank participants. BMJ 2018;361:k1651.
Lopez-Bueno R, Andersen LL, Koyanagi A, Nunez-Cortes R, Calatayud J, Casana J, et al. Thresholds of handgrip strength for all-cause, cancer, and cardiovascular mortality: A systematic review with dose-response meta-analysis. Ageing Res Rev 2022;82:101778.
Parra-Soto S, Pell JP, Celis-Morales C, Ho FK. Absolute and relative grip strength as predictors of cancer: prospective cohort study of 445 552 participants in UK Biobank. J Cachexia Sarcopenia Muscle 2022;13:325-332.
Leong DP, Teo KK, Rangarajan S, Lopez-Jaramillo P, Avezum A, Jr., Orlandini A, et al. Prognostic value of grip strength: findings from the Prospective Urban Rural Epidemiology (PURE) study. Lancet 2015;386:266-273.
Wu Y, Wang W, Liu T, Zhang D. Association of Grip Strength With Risk of All-Cause Mortality, Cardiovascular Diseases, and Cancer in Community-Dwelling Populations: A Meta-analysis of Prospective Cohort Studies. J Am Med Dir Assoc 2017;18:551 e517-551 e535.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Body shape, fat distribution and sarcopenia and risk of liver cancer in European population: a Mendelian randomization study

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

1 Introduction

2 Materials and methods

2.1 Outcome data source

2.2 Exposures data source

2.3 Statistical analysis

3 Results

4 Discussion

5 Conclusion

Declarations

Conflict of interest

Ethics approval and consent to participate

Author Contribution

Acknowledgements

References

Additional Declarations

Status:

Version 1