1. Genetics and Genomics

Evaluating the effect of metabolic traits on oral and oropharyngeal cancer risk using Mendelian randomization

  1. Mark Gormley  Is a corresponding author
  2. Tom Dudding
  3. Steven J Thomas
  4. Jessica Tyrrell
  5. Andrew R Ness
  6. Miranda Pring
  7. Danny Legge
  8. George Davey Smith
  9. Rebecca C Richmond
  10. Emma E Vincent
  11. Caroline Bull
  1. MRC Integrative Epidemiology Unit, Population Health Sciences, Bristol Medical School, University of Bristol, United Kingdom
  2. Bristol Dental Hospital and School, University of Bristol, United Kingdom
  3. University of Exeter Medical School, RILD Building, RD&E Hospital, United Kingdom
  4. University Hospitals Bristol and Weston NHS Foundation Trust National Institute for Health Research Bristol Biomedical Research Centre, University of Bristol, United Kingdom
  5. Translational Health Sciences, Bristol Medical School, University of Bristol, United Kingdom
https://doi.org/10.7554/eLife.82674
Share this article
Cite this article
  1. Mark Gormley
  2. Tom Dudding
  3. Steven J Thomas
  4. Jessica Tyrrell
  5. Andrew R Ness
  6. Miranda Pring
  7. Danny Legge
  8. George Davey Smith
  9. Rebecca C Richmond
  10. Emma E Vincent
  11. Caroline Bull
(2023)
Evaluating the effect of metabolic traits on oral and oropharyngeal cancer risk using Mendelian randomization
eLife 12:e82674.
https://doi.org/10.7554/eLife.82674
  2. Download BibTeX
  3. Download .RIS
2 figures, 1 table and 3 additional files

Figures

Figure 1
Download asset Open asset
Directed acyclic graph (DAG) depicting Mendelian randomization applied to this study.

Genetic variants (G) can act as proxies or instruments to investigate if an exposure (X) is associated with a disease outcome (Y). Causal inference can be made between X and Y if the following conditions are upheld. (1) The genetic variants which make up the instrument are valid and reliably associated with the exposure (i.e., the ‘relevance assumption’); (2) There is no measured or unmeasured confounding of the association between the genetic instrument and the outcome (i.e., the ‘exchangeability’ assumption); (3) There is no independent pathway between the genetic instrument and the outcome, except through the exposure (i.e., the ‘exclusion restriction principle’).

Figure 2 with 19 supplements
Download asset Open asset
Mendelian randomization results of genetically proxied metabolic disorders with risk of oral and oropharyngeal cancer including sensitivity analyses in GAME-ON.

Oral and oropharyngeal cancer combined n = 6034, oral cancer n = 2990 and oropharyngeal cancer n = 2641. Abbreviations: IVW, inverse variance weighted; OR, odds ratio with 95% confidence intervals; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; FG, fasting glucose; FI, fasting insulin; HbA1c, glycated haemoglobin; SBP, systolic blood pressure; DBP, diastolic blood pressure. OR are expressed per 1 standard deviation (SD) increase in genetically predicted BMI (4.81 kg/m2), WC (0.09 unit), WHR (0.10 unit), T2D (1-log unit higher odds of T2D), FG (1-log unit increase in mmol/l fasting glucose), FI (1-log unit increase in mmol/L fasting insulin), HbA1c (1-log unit % higher glycated haemoglobin), SBP (1 unit mmHg increase), and DBP (1 unit mmHg increase).

Figure 2—figure supplement 1
Download asset Open asset
Power calculations for oral and oropharyngeal analyses in GAME-ON.

α set at 0.05.

Figure 2—figure supplement 2
Download asset Open asset
Forest plots showing Mendelian randomization results for genetically proxied body mass index (BMI) with risk of combined oral and oropharyngeal cancer in GAME-ON.

Effect estimates on oral and oropharyngeal cancer are reported on the log odds scale.

Figure 2—figure supplement 3
Download asset Open asset
Forest plots showing Mendelian randomization results for genetically proxied waist–hip ratio (WHR) with risk of combined oral and oropharyngeal cancer in GAME-ON.

Effect estimates on oral and oropharyngeal cancer are reported on the log odds scale.

Figure 2—figure supplement 4
Download asset Open asset
Forest plots showing Mendelian randomization results for genetically proxied waist circumference (WC) with risk of combined oral and oropharyngeal cancer in GAME-ON.

Effect estimates on oral and oropharyngeal cancer are reported on the log odds scale.

Figure 2—figure supplement 5
Download asset Open asset
Forest plots showing Mendelian randomization results for genetically proxied type 2 diabetes mellitus (T2D) with risk of combined oral and oropharyngeal cancer in GAME-ON.

Effect estimates on oral and oropharyngeal cancer are reported on the log odds scale.

Figure 2—figure supplement 6
Download asset Open asset
Forest plots showing Mendelian randomization results for genetically proxied glycated haemoglobin (HbA1c) with risk of combined oral and oropharyngeal cancer in GAME-ON.

Effect estimates on oral and oropharyngeal cancer are reported on the log odds scale.

Figure 2—figure supplement 7
Download asset Open asset
Forest plots showing Mendelian randomization results for genetically proxied fasting glucose (FG) with risk of combined oral and oropharyngeal cancer in GAME-ON.

Effect estimates on oral and oropharyngeal cancer are reported on the log odds scale.

Figure 2—figure supplement 8
Download asset Open asset
Forest plots showing Mendelian randomization results for genetically proxied fasting insulin (FI) with risk of combined oral and oropharyngeal cancer in GAME-ON.

Effect estimates on oral and oropharyngeal cancer are reported on the log odds scale.

Figure 2—figure supplement 9
Download asset Open asset
Forest plots showing Mendelian randomization results for genetically proxied systolic blood pressure (SBP) with risk of combined oral and oropharyngeal cancer in GAME-ON.

Effect estimates on oral and oropharyngeal cancer are reported on the log odds scale.

Figure 2—figure supplement 10
Download asset Open asset
Forest plots showing Mendelian randomization results for genetically proxied diastolic blood pressure (DBP) with risk of combined oral and oropharyngeal cancer in GAME-ON.

Effect estimates on oral and oropharyngeal cancer are reported on the log odds scale.

Figure 2—figure supplement 11
Download asset Open asset
Scatter plot for body mass index (BMI) with risk of combined oral and oropharyngeal cancer in GAME-ON.
Figure 2—figure supplement 12
Download asset Open asset
Scatter plot for waist circumference (WC) with risk of combined oral and oropharyngeal cancer in GAME-ON.
Figure 2—figure supplement 13
Download asset Open asset
Scatter plot for waist–hip ratio (WHR) with risk of combined oral and oropharyngeal cancer in GAME-ON.
Figure 2—figure supplement 14
Download asset Open asset
Scatter plot for type 2 diabetes mellitus (T2D) with risk of combined oral and oropharyngeal cancer in GAME-ON.
Figure 2—figure supplement 15
Download asset Open asset
Scatter plot for glycated haemoglobin (HBA1c) with risk of combined oral and oropharyngeal cancer in GAME-ON.
Figure 2—figure supplement 16
Download asset Open asset
Scatter plot for fasting glucose (FG) with risk of combined oral and oropharyngeal cancer in GAME-ON.
Figure 2—figure supplement 17
Download asset Open asset
Scatter plot for fasting insulin (FI) with risk of combined oral and oropharyngeal cancer in GAME-ON.
Figure 2—figure supplement 18
Download asset Open asset
Scatter plot for systolic blood pressure (SBP) with risk of combined oral and oropharyngeal cancer in GAME-ON.
Figure 2—figure supplement 19
Download asset Open asset
Scatter plot for diastolic blood pressure (DBP) with risk of combined oral and oropharyngeal cancer in GAME-ON.

Tables

Table 1
Mendelian randomization results of genetically proxied metabolic traits with risk of oral and oropharyngeal cancer in GAME-ON.
IVWWeighted medianWeighted modeMR-Egger
 ExposureOutcomeExposure/outcome sourceOutcome
N		Number of
SNPs		OR (95% CI)pOR (95% CI)pOR (95% CI)pOR (95% CI)p
 BMIOral and oropharyngeal cancer combinedPulit et al.
GWAS/GAME-ON		60342720.89 (0.72, 1.09)0.260.71 (0.50, 1.00)0.050.63 (0.37, 1.04)0.070.66 (0.40, 1.10)0.11
Oral cancer29902720.92 (0.71, 1.19)0.530.83 (0.55, 1.28)0.400.79 (0.38, 1.62)0.520.75 (0.39, 1.41)0.37
Oropharyngeal cancer26412720.89 (0.68, 1.15)0.360.75 (0.50, 1.13)0.170.53 (0.27, 1.03)0.060.56 (0.29, 1.07)0.08
 WCOral and oropharyngeal cancer combinedShungin et al.
GWAS/GAME-ON		6034430.73 (0.52, 1.02)0.070.64 (0.40, 1.05)0.080.67 (0.36, 1.26)0.220.43 (0.17, 1.08)0.08
Oral cancer2990430.82 (0.53, 1.26)0.360.66 (0.36, 1.21)0.180.67 (0.32, 1.39)0.290.54 (0.17, 1.76)0.31
Oropharyngeal cancer2641430.66 (0.43, 1.01)0.060.56 (0.30, 1.05)0.070.37 (0.17, 0.83)0.020.30 (0.09, 0.98)0.05
 WHROral and oropharyngeal cancer combinedPulit et al. GWAS/GAME-ON60341760.98 (0.74, 1.29)0.880.98 (0.64, 1.49)0.920.95 (0.45, 2.00)0.891.80 (0.87, 3.71)0.11
Oral cancer29901761.18 (0.84, 1.65)0.351.00 (0.58, 1.73)0.990.69 (0.29, 1.67)0.412.49 (1.02, 6.12)0.05
Oropharyngeal cancer26411760.83 (0.59, 1.14)0.250.88 (0.51, 1.50)0.630.93 (0.37, 2.30)0.871.19 (0.50, 2.86)0.70
 T2DOral and oropharyngeal cancer combinedVujkovic et al. GWAS/GAME-ON60342540.92 (0.84, 1.01)0.090.85 (0.74, 0.97)0.020.84 (0.71, 1.01)0.060.91 (0.77, 1.09)0.31
Oral cancer29902540.94 (0.84, 1.05)0.270.84 (0.72, 0.99)0.040.82 (0.66, 1.02)0.080.88 (0.71, 1.08)0.22
 Oropharyngeal cancer26412540.94 (0.84, 1.05)0.270.89 (0.73, 1.10)0.291.02 (0.80, 1.30)0.881.00 (0.81, 1.24)0.99
 HbA1c Oral and oropharyngeal cancer combinedWheeler et al.
GWAS/GAME-ON (Lesseur et al., 2016)		6034370.56 (0.32, 1.00)0.050.52 (0.23, 1.20)0.120.54 (0.24, 1.21)0.140.37 (0.13, 1.05)0.07
Oral cancer2990370.48 (0.24, 0.93)0.030.51 (0.18, 1.41)0.190.44 (0.15, 1.29)0.140.30 (0.09, 1.03)0.06
 Oropharyngeal cancer2641370.66 (0.31, 1.40)0.280.49 (0.15, 1.57)0.230.57 (0.18, 1.85)0.350.43 (0.11, 1.68)0.23
 FGOral and oropharyngeal cancer combinedLagou et al.
GWAS/
GAME-ON (Lesseur et al., 2016)		6034281.06 (0.68, 1.66)0.791.20 (0.62, 2.30)0.591.13 (0.60, 2.12)0.711.11 (0.48, 2.56)0.80
Oral cancer2990281.05 (0.58, 1.92)0.871.15 (0.48, 2.72)0.750.99 (0.44, 2.23)0.991.25 (0.39, 4.01)0.70
Oropharyngeal cancer2641281.39 (0.77, 2.51)0.281.24 (0.51, 3.03)0.631.36 (0.59, 3.18)0.481.38 (0.45, 4.18)0.58
 FIOral and oropharyngeal cancer combinedLagou et al.
GWAS/GAME-ON		6034170.81 (0.23, 2.89)0.750.75 (0.20, 2.87)0.680.60 (0.03, 10.79)0.740.11 (0.001, 22.47)0.43
Oral cancer2990170.96 (0.22, 4.16)0.960.46 (0.08, 2.47)0.370.45 (0.01, 19.02)0.680.21 (0.0004, 107.21)0.63
Oropharyngeal cancer2641170.68 (0.16, 2.87)0.590.66 (0.12, 3.67)0.630.48 (0.05, 4.99)0.550.09 (0.0002, 40.04)0.45
 SBPOral and oropharyngeal cancer combinedEvangelou et al., 2018 GWAS/GAME-ON (Lesseur et al., 2016)6034831.00 (0.97, 1.03)0.890.99 (0.94, 1.03)0.550.98 (0.88, 1.09)0.661.06 (0.92, 1.23)0.39
Oral cancer2990831.01 (0.96, 1.06)0.740.99 (0.93, 1.04)0.650.95 (0.84, 1.08)0.481.09 (0.90, 1.33)0.37
Oropharyngeal cancer2641830.99 (0.95, 1.03)0.650.99 (0.94, 1.05)0.771.00 (0.88, 1.13)0.941.03 (0.87, 1.23)0.71
 DBPOral and oropharyngeal cancer combinedEvangelou et al., 2018 GWAS/GAME-ON6034640.93 (0.87, 1.00)0.050.94 (0.86, 1.04)0.221.10 (0.88, 1.38)0.420.99 (0.80, 1.24)0.95
Oral cancer2990640.95 (0.87, 1.04)0.260.96 (0.86, 1.07)0.451.17 (0.88, 1.56)0.280.97 (0.74, 1.27)0.81
Oropharyngeal cancer2641640.92 (0.84, 1.00)0.050.94 (0.84, 1.05)0.291.10 (0.86, 1.41)0.451.00 (0.75, 1.30)0.93

  1. OR are expressed per 1 standard deviation (SD) increase in genetically predicted BMI (4.81 kg/m2), WC (0.09 unit), WHR (0.10 unit), T2D (1-log unit higher odds of T2D), FG (1-log unit increase in mmol/L fasting glucose), FI (1-log unit increase in mmol/L fasting insulin), HbA1c (1-log unit % higher glycated haemoglobin), SBP (1 unit mmHg increase), and DBP (1 unit mmHg increase).

  2. IVW, inverse variance weighted; OR, odds ratio; CI, confidence intervals; p, p-value; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; FG, fasting glucose; FI, fasting insulin; HbA1c, glycated haemoglobin; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Additional files

Supplementary file 1

Genetic variants associated with metabolic traits of interest.

https://cdn.elifesciences.org/articles/82674/elife-82674-supp1-v2.xlsx
Supplementary file 2

Showing tables 2A through 2L.

Supplementary file 2, Table 2A. Assessing weak instrument bias (F-statistic) and proportion of variance in the phenotype (R2) explained by metabolic phenotype instruments. Abbreviations: BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; FG, fasting glucose; FI, fasting insulin; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Supplementary file 2, Table 2B. Assessing heterogeneity of single nucleotide polymorphism effect estimates in inverse variance weighted (IVW) and MR-Egger regression for metabolic disorder analysis. Abbreviations: Q, Q-statistic; df, degrees of freedom; p, p-value; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; FG, fasting glucose; FI, fasting insulin; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Supplementary file 2, Table 2C. Assessing directional pleiotropy through MR-Egger intercept for metabolic disorder analysis. Abbreviations: SE, standard error; p, p-value; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; FG, fasting glucose; FI, fasting insulin; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Supplementary file 2, Table 2D. MR-PRESSO outliers detected results in the analysis of metabolic disorders on combined oral and oropharyngeal cancer risk. Abbreviations: Q-stat, Cochran’s Q-statistic; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; FG, fasting glucose; FI, fasting insulin; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Supplementary file 2, Table 2E. MR-PRESSO results for metabolic disorders on combined oral and oropharyngeal cancer. Abbreviations: SE, standard error; p, p-value; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; FG, fasting glucose; FI, fasting insulin; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Supplementary file 2, Table 2F. Outlier corrected results in the analysis of metabolic disorders on combined oral and oropharyngeal cancer risk. Abbreviations: SE, standard error; OR, odds ratio; CI, confidence intervals; IVW, inverse variance weighted; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; FG, fasting glucose; FI, fasting insulin; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Supplementary file 2, Table 2G. Assessing violation of the NO measurement error (NOME) assumption for instruments used in MR-Egger regression. Abbreviations: I2, I-squared statistic; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; FG, fasting glucose; FI, fasting insulin; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Supplementary file 2, Table 2H. SIMEX correction MR-Egger regression results for where NO measurement error (NOME) assumption may have been violated (I2 < 0.90). Abbreviations: OR, odds ratio; CI, confidence intervals; WC, waist circumference; FI, fasting insulin; SBP, systolic blood pressure; DBP, diastolic blood pressure.

Supplementary file 2, Table 2I. Mendelian randomization results evaluating instrument-risk factor effects. Abbreviations: IVW, inverse variance weighted; OR, odds ratio; CI, confidence intervals; p, p-value; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; DBP, diastolic blood pressure. OR are expressed per 1 standard deviation (SD) increase in genetically predicted BMI (4.81 kg/m2), WC (0.09 unit), WHR (0.10 unit), T2D (1-log unit higher odds of T2D), HbA1c (1-log unit % higher glycated haemoglobin), and DBP (1 unit mmHg increase). Outcome beta estimates reflect the standard deviation of the phenotype.

Supplementary file 2, Table 2J. Assessing heterogeneity in Mendelian randomization results evaluating instrument-risk factor effects. Abbreviations: Q, Q-statistic; df, degrees of freedom; p, p-value; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; DBP, diastolic blood pressure.

Supplementary file 2, Table 2K. Assessing directional pleiotropy in Mendelian randomization results evaluating instrument-risk factor effects. Abbreviations: SE, standard error; p, p-value; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; DBP, diastolic blood pressure.

Supplementary file 2, Table 2L. Outlier corrected Mendelian randomization results evaluating instrument-risk factor effects. Abbreviations: IVW, inverse variance weighted; OR, odds ratio; CI, confidence intervals; p, p-value; BMI, body mass index; WC, waist circumference; WHR, waist–hip ratio; T2D, type 2 diabetes mellitus; HbA1c, glycated haemoglobin; DBP, diastolic blood pressure. OR are expressed per 1 standard deviation (SD) increase in genetically predicted BMI (4.81 kg/m2), WC (0.09 unit), WHR (0.10 unit), T2D (1-log unit higher odds of T2D), HbA1c (1-log unit % higher glycated haemoglobin), and DBP (1 unit mmHg increase). Outcome beta estimates reflect the standard deviation of the phenotype.

https://cdn.elifesciences.org/articles/82674/elife-82674-supp2-v2.docx
Transparent reporting form
https://cdn.elifesciences.org/articles/82674/elife-82674-transrepform1-v2.pdf

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Open citations (links to open the citations from this article in various online reference manager services)

Cite this article (links to download the citations from this article in formats compatible with various reference manager tools)

  1. Mark Gormley
  2. Tom Dudding
  3. Steven J Thomas
  4. Jessica Tyrrell
  5. Andrew R Ness
  6. Miranda Pring
  7. Danny Legge
  8. George Davey Smith
  9. Rebecca C Richmond
  10. Emma E Vincent
  11. Caroline Bull
(2023)
Evaluating the effect of metabolic traits on oral and oropharyngeal cancer risk using Mendelian randomization
eLife 12:e82674.
https://doi.org/10.7554/eLife.82674