Metabolic traits and the risk of head and neck cancer: a systematic review and meta-analysis
============================================================================================

* Alexander J Gormley
* Charlotte Richards
* Francesca Spiga
* Alex Whitmarsh
* Emily Gray
* Jo Hooper
* Barry G Main
* Emma E Vincent
* Rebecca C Richmond
* Julian PT Higgins
* Mark Gormley

## Abstract

**Introduction** The overall incidence of head and neck cancer (HNC) continues to rise, despite a decline in smoking rates, particularly within developed countries. Obesity and related metabolic traits have been attributed to a growth in cancer rate, so further exploration of these risk factors is warranted in HNC. A comprehensive systematic review and meta-analysis was conducted in order to obtain the most precise observational estimates between metabolic trait exposures and risk of HNC.

**Methods** A search strategy was developed with an information and content specialists. Multiple databases including Cochrane Library, OVID SP versions of Medline, EMBASE, pre-prints and the grey literature were searched. The primary outcome for included studies was incident HNC and exposures included obesity defined by body mass index (BMI), type 2 diabetes, dyslipidaemia, and hypertension, using pre-specified definitions. A combined risk effect across studies was calculated using both fixed and random-effects meta-analysis. Heterogeneity was assessed between studies using the Cochrans Q and I2 statistical tests. The ROBINS-E preliminary tool was used to assess the bias in each included result.

**Results** The search generated 7,316 abstracts, of these 197 full text articles were assessed for eligibility and 36 were included for full qualitative and quantitative synthesis. In the analysis of 5 studies investigating the association between obesity and incidence of HNC, there was an overall RR of 1.06 (95%CI (0.76, 1.49), *P* heterogeneity <0.024, I2= 73.2%) using a random-effects model. 6 studies reported on the association between type 2 diabetes and incidence of HNC, with an overall RR of 1.13 (95%CI (0.95, 1.34), *P* heterogeneity= <0.0001, I2= 80.0%) using a random-effects model. An increased risk of hypertension was consistent across HNC subsites, with the strongest association found in the larynx (RR= 1.17, 95%CI (1.10, 1.25), *P* heterogeneity= 0.186, I2= 37.7%). For dyslipidaemia, only 2 studies were available for meta-analysis in the laryngeal subsite, with some evidence of an increased risk association of both low high-density lipoprotein (RR= 1.12, 95%CI (1.07, 1.18), *P* heterogeneity= 0.103, I2= 62.5%) and high triglyceride levels (RR= 1.10, 95%CI (1.05, 1.15), *P* heterogeneity= 0.319, I2= 0.0%)) using random-effects models. Over 80% of studies were judged to be at Very High or High risk of bias using the ROBINS-E tool.

**Conclusion** Despite individual studies suggesting an association between BMI and HNC, limited effect was demonstrated in this meta-analysis. There was evidence of an association between hypertension and dyslipidaemia on incident HNC, however caution is required due to the high levels of heterogeneity recorded in these studies. Observational associations are susceptible to confounding, bias and reverse causality so these results must be interpreted with caution. Future work should include meta-analysing studies separately by geographic region, as this appears to be a clear source of heterogeneity.

## Introduction

Despite the significant reduction in smoking 1, particularly in developed countries, the overall incidence of head and neck cancer (HNC) continues to rise 2, 3. Given this potentially changing aetiology 4, further exploration of other risk factors is warranted to enhance prevention strategies. According to the World Health Organization (WHO), at least 40% of all cancer cases may be preventable, with smoking, alcohol consumption and obesity identified as three of the most important modifiable lifestyle factors 5. The growth in cancer attributable to metabolic risk is increasing, with a global attributable fraction due to high body mass index (BMI) at around 3.6%, equating to almost half a million new cases each year 6,7. Public health strategies have been unsuccessful in addressing the current obesity epidemic at the population level, which could also result in more overall cancer cases in the years to come 8. It has been projected that obesity could even supersede smoking as the primary driver of cancer in the coming decades 5.

People living with obesity and associated metabolic traits have an increased risk of developing cancer 5, but the epidemiological evidence surrounding HNC is not conclusive 9,10. In the largest pooled analysis, obesity defined by higher BMI was associated with a protective effect for HNC in current smokers (hazard ratio (HR)= 0.76, 95%CI (0.71, 0.82), *p* <0.0001, per 5 kg/m2) and conversely, a higher risk in never smokers (HR= 1.15, 95%CI (1.06, 1.24) per 5 kg/m2, *p* <0.001) 11. More recent cohort studies have failed to show a clear association between BMI and HNC 12–16. Similarly, a meta-analysis of observational studies investigating type 2 diabetes with oral and oropharyngeal subsites found an increased risk ratio (RR) of 1.15, 95%CI (1.02, 1.29) 17. This finding has been supported by more recent independent cohorts 12,18–20. Lastly, hypertension (raised systolic blood pressure (SBP) or diastolic blood pressure (DBP)), has been correlated with head and neck cancer risk across multiple studies 19–23. There are a number of proposed mechanisms linking these metabolic traits and cancer, with none specifically linked to HNC. In short, hyperglycaemia may provide a substrate for tumour growth or lead to direct DNA mutation. Insulin itself is a potent growth factor, thought to promote proliferation and carcinogenesis both directly and indirectly through insulin like growth factor-1 (IGF-1) 24–26. Insulin resistance can lead to a chronic proinflammatory environment and the formation of macrophages which produce mutagenic agents and reactive oxygen species leading to DNA damage 27. Any underlying mechanism between hypertension and cancer risk remains unknown 23. Lipids are an essential component of the cell wall and so increased cellular uptake of cholesterol has been seen in many cancers, while other malignancies have demonstrated lipid metabolism reprogramming which can help promote a cancer cell phenotype including proliferation, differentiation and metastasis 28.

We aimed to conduct the largest, most comprehensive systematic review and meta-analysis to date, in order to obtain the most precise observational association between metabolic traits and risk of HNC. If available, results were stratified by subsite (i.e., oral cavity, oropharynx, hypopharynx, and larynx), with pre-defined subgroup analyses performed to enhance interpretation, and additional sensitivity analysis to improve certainty in the results.

## Methods

### Research Question

‘Do metabolic traits affect the risk of developing head and neck cancer?’

### Eligibility criteria

**Population-** participants over 18 years old, of either sex, from any country or ethnic background.

**Exposures**-1) obesity, 2) type 2 diabetes, 3) hypertension and 4) dyslipidaemia as defined in the study protocol below. These will be collectively described as metabolic traits.

**Comparison-** participants who have not been diagnosed with a metabolic trait(s).

**Outcome-** incident head and neck squamous cell carcinoma (HNC).

### Study Selection

We focused on prospective observational studies reporting the risk of metabolic traits on incident HNC. This was to ensure a direct temporal relationship between the exposure and onset of head and neck cancer. Eligibility criteria are detailed in **Table 1**.

View this table:
[Table 1.](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/T1)

Table 1. Study selection criteria

### Data sources and search strategy

A systematic search strategy (**Supplementary Table 1**) was formulated by clinicians, scientific researchers and a specialist librarian for Head & Neck Surgery and Medicine. Medical Subject Headings (MeSH) and keywords were iteratively combined with the Boolean operators AND, OR and NOT. Multiple databases were searched from January 1966 – November 2021, including OVID SP versions of Medline and EMBASE, using the formulated search strategy (**Supplementary Table 1**). Pre-print servers *medRxiv* (since June 2019) and *BioRxiv* (since November 2013) were also searched. In addition, the following electronic bibliographic databases were comprehensively searched: EThOS, Google Scholar, Open Grey and ClinicalTrials.gov, to identify articles from the grey literature and conference proceedings using a modified search strategy (**Supplementary Table 2**). References extracted from the full-length articles were reviewed to identify other publications of interest. Duplicate articles were removed using Covidence (*Covidence systematic review software*, Veritas Health Innovation, Melbourne, Australia) 29.

View this table:
[Supplementary Table 1](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/T3)

Supplementary Table 1 Search strategy

View this table:
[Supplementary Table 2](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/T4)

Supplementary Table 2 Modified search strategy

To ensure transparent and reproducible methodology, this study has been reported in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement and was pre-registered on International Prospective Register of Systematic Reviews (PROSPERO) in 2021 (CRD42021250520). The protocol was also accepted for publication in *BMJ Open* 30.

### Data screening and extraction

All titles and abstracts were initially screened by two authors (A.G. and C.R.) using Covidence 29. Any conflicts were resolved by a third independent reviewer (M.G). A PRISMA flowchart 31 representing the study selection process is shown in **Figure 1**. Full text of articles included after title and abstract screening were retrieved and screened by (A.G., C.R and M.G.). Data were then extracted from articles which met the inclusion criteria (**Table 1**).

![Figure 1.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F1.medium.gif)

[Figure 1.](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F1)

Figure 1. PRISMA flowchart representing the publication selection process

This was performed independently by three authors (E.G., C.R and M.G.) using a standardised data collection spreadsheet containing: study ID; the first author’s last name, year of publication, study design, sample size, age and sex of the subjects, country of origin, duration of follow-up in cohort studies, assessment of exposure and outcome, adjusted covariates, number of cases, the effect estimates with 95%CIs and follow-up duration. From each study, risk estimates were extracted which reflected the greatest degree of adjustments for potential confounders. If study data appeared in more than one article, the most recent publication or largest e.g., pooled result were extracted for the statistical analysis.

### Exposure definitions

There are differences in metabolic trait definitions used, depending on the country where guidelines were written, age, sex, and ethnicity 32–34. For this study, the following metabolic traits definitions were used: 

1.  The WHO definition of obesity as body mass index (BMI) of ≥30 kg/m2 must have been used 35. BMI was acceptable if calculated from self-reported height and weight or taken from medical records or measured at baseline.

2.  The WHO defines type 2 diabetes as a chronic disease that occurs when the body cannot effectively use the insulin it produces and is largely the result of excess body weight and physical inactivity. Diagnosis was defined as symptoms such as polyuria or polydipsia, plus36,37:

*   random blood plasma glucose concentration ≥ 11.1 mmol/l or;

*   fasting plasma glucose concentration ≥ 7.0 mmol/l (whole blood ≥ 6.1 mmol/l) or;

*   two-hour plasma glucose concentration ≥ 11.1 mmol/l two hours after 75g anhydrous glucose in an oral glucose tolerance test (OGTT) or;

*   haemoglobin (HbA1c) 6.5% or more (48 mmol/mol and above)

A diagnosis of type 2 diabetes via self-report, International Classification of Diseases (ICD) codes or medical records was also accepted.

1.  The WHO definition of hypertension as 38:

*   systolic blood pressure (SBP) reading of ≥130 mmHg; and/or

*   diastolic blood pressure (DBP) readings on both days of ≥85 mmHg.

ICD-codes and patients taking antihypertensive medication were also acceptable diagnoses.

1.  Dyslipidaemia classified as serum total cholesterol (TC), low density lipoprotein cholesterol (LDL-C), triglycerides, apolipoprotein B, or lipoprotein(a) concentrations above the 90th percentile, or high density lipoprotein cholesterol (HDL-C) or apolipoprotein concentrations below the 10th percentile for the general population in which the study was conducted 39. ICD-codes and patients taking cholesterol-lowering medication were also acceptable diagnoses.

### Outcome definitions

All diagnoses of incident head and neck squamous cell carcinoma should have been confirmed using histology by a trained pathologist and reported using appropriate ICD codes40. Cancer cases for the disease of interest have the following ICD-10 codes: oral cavity (C02.0 – C02.9, C03.0 – C03.9, C04.0 – C04.9, C05.0 – C06.9) oropharynx (C01.9, C02.4, C09.0 – C10.9), hypopharynx (C13.0 – C13.9), overlapping (C14 and combination of other sites) and 25 cases with unknown ICD code (other). Older versions of ICD code e.g., ICD-8 and 9 were also accepted.

### Quality assessment of studies

This assessment was used in evaluating the strength of evidence from the included studies. Three authors (A.G., C.R. and M.G.) independently assessed papers for risk of bias during data extraction using the Risk of Bias in Non-randomised Studies – of Exposures (ROBINS-E) tool (available at: [https://www.riskofbias.info/welcome/robins-e-tool](https://www.riskofbias.info/welcome/robins-e-tool))41. The ROBINS-E tool aims to assess the result of interest in an observational study for risk of bias, where that study is designed to assess the effect of exposure on outcome 42. Pre-defined confounding variables that were agreed, included: age, sex, smoking status, alcohol intake, ethnicity, and socio-economic status. Further confounding variables that were recognised as important, but possibly to a lesser extent, included: dietary intake, physical exercise, and other metabolic traits (e.g., where BMI was the exposure, adjusting for hypertension, dyslipidaemia, and type 2 diabetes). The ‘appropriateness’ of each study to address the review question was also considered.

Data visualisations of the ROBINS-E tool were created using the *robvis* tool (available at: [https://mcguinlu.shinyapps.io/robvis/](https://mcguinlu.shinyapps.io/robvis/)). An individual panel has been developed for each individual exposure. Individual results from each included study were assessed using the tool for each included study. There was no difference in risk of bias assessment within each study and therefore the overall risk of bias assessment is presented below.

### Data analysis and synthesis

The analysis and synthesis process were performed in two phases. The first phase consisted of estimating the effect of each metabolic trait (obesity, type 2 diabetes, dyslipidaemia, and hypertension) on incident HNC separately for each included study. All included studies were cohort studies, with most using non-exposed comparisons and reporting risk (RR) and hazard ratios (HR). Other studies using external general population comparisons reported standardised incidence ratios (SIR). Because the absolute risk of HNC is low, all the above measures yield similar estimates of RR and this approach has been taken previously in similar head and neck cancer meta-analyses 17. The second phase focussed on estimating the combined effect across studies using meta-analysis, using *metan* software in Stata (*Stata Statistical Software: Release 16*. College Station, TX: StataCorp LLC). Summary effects with their corresponding 95%CIs were derived using the method described by DerSimonian and Laird for both fixed effects and random effects models, which incorporates between-study variability 43. This was performed separately for each metabolic trait, with a preference for random-effects if there is no important difference between the two methods.

The analysis yielded information about the heterogeneity of effects across studies, using Cochran’s Q test and Higgins’ I2 statistic. To further explore the sources of heterogeneity, subgroup, and meta-regression analyses were performed according to age, geographic location, year of publication, methods of exposure assessment, length of follow-up, the number of cancer cases and adjustment for confounding factors including age, sex, smoking and alcohol use. Small study effects, which may reflect publication bias, study quality or other sources of heterogeneity were assessed using funnel plots, which show effect estimates from individual studies plotted against the standard error as a measure of each study’s size or precision. In the absence of bias and between study heterogeneity, the scatter will be due to sampling variation alone, with the plot resembling a symmetrical inverted funnel. Smaller studies will be scattered more widely at the bottom, with larger studies nearer the top and any gaps where smaller studies are missing will be evident 44.

Sensitivity analyses were performed by excluding two large pooled analyses from the meta-analyses, one for BMI 11 and the other for type 2 diabetes 17, and calculating an estimate for the remainder of the studies to evaluate whether the results were significantly affected by these studies.

## Results

The search generated 7,316 potential records of which 6,129 were screened following the removal of 1,187 duplicates. Of these, 197 full-text articles of interest were assessed for eligibility. One hundred and sixty one of these articles were subsequently excluded: 105 due to study designs that could not infer a temporal relationship between the exposure and outcome, 29 where the exposure was not relevant or as defined above, 16 in which the outcome was not relevant or as defined above, 1 with an ineligible comparison group (diabetics treated versus not treated), 7 with no original data or reported effect statistic (e.g., short abstracts), 2 studies had continuous data only which was not comparable and 1 article that was not written in English. The remaining 36 eligible articles were included for full qualitative and quantitative synthesis to investigate the association between metabolic traits and HNC risk (Figure 1). One large pooled analysis for BMI 11 included the results from 4 of the studies 45–48 and similarly, a pooled analysis for type 2 diabetes17 included the results from 6 of the studies 9,49–53, so only the two pooled analyses were represented in the final **Table 2** and subsequent analyses.

View this table:
[Table 2.](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/T2)

Table 2. Descriptive characteristics of studies of metabolic traits and risk of head and neck c3a2n8cer

Eight studies reported the association between BMI (n= 2,859 cases) and risk of HNC 11,12,14–16,54–56, with 15 studies reporting the association with type 2 diabetes (n= 17,582 cases) 12,17–20,23,24,57–64, six studies hypertension (n= 12,151 cases) 12,19–23, and four studies dyslipidaemia (n= 6,410 cases) 12,19,20,65 and HNC risk (**Table 2**). Adjustments were made for potential confounders of two or more factors, including age, in all studies.

### Obesity and risk of HNC

In the analysis of five studies investigating the association between obesity and incidence of combined head and neck cancer sites, there was an overall RR of 1.06 (95%CI (0.76, 1.49)) using a random-effects model, little evidence of an effect (Figure 2a). The result was similar when using a fixed-effects model (RR= 1.10, 95%CI (0.97, 1.25)). There was however evidence for the presence of substantial heterogeneity between studies (*P* heterogeneity <0.024, I2= 73.2%) (Figure 2a). With respect to individual subsites, there was again little evidence of an association of obesity on oral cavity (RR= 0.92, 95%CI (0.81, 1.04)), oropharyngeal (RR= 1.05, 95%CI (0.69, 1.59)), or laryngeal (RR= 0.95, 95%CI (0.81, 1.11)) cancer using a random-effects model. There were similar results when using a fixed-effects model (Figure 2b – d).

![Figure 2.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F2.medium.gif)

[Figure 2.](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F2)

Figure 2. Relative risks for the association between obesity and incidence of head and neck cancer combined and by subsite

### Type 2 diabetes mellitus and risk of HNC

In the analysis of 6 studies on the association between type 2 diabetes and incidence of combined head and neck cancer, there was an overall RR of 1.13 (95%CI (0.95, 1.34)) using a random-effects model. Strong evidence of heterogeneity was present between studies (*P* heterogeneity= <0.0001, I2= 80.0%) (Figure 3a). There was a trend towards an increased risk association with oral cavity (RR= 1.13, 95%CI (0.97, 1.31)), oropharyngeal (RR= 1.16, 95%CI (0.73, 1.83)) and laryngeal (RR= 1.05, 95%CI (0.87, 1.26)) cancer using a random effects model, but again all results crossed the null, with strong evidence of heterogeneity across all subsites (Figure 3b – d). When using a fixed effects model, the result for combined head and neck cancer sites showed a stronger increased risk effect (RR= 1.22, 95%CI (1.16, 1.29)), but attenuated towards the null in the individual subsites.

![](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F3/graphic-14.medium.gif)

[](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F3/graphic-14)

![](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F3/graphic-15.medium.gif)

[](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F3/graphic-15)

Figure 3. Relative risks for the association between type 2 diabetes mellitus and incidence of head and neck cancer combined and by subsite

### Hypertension and risk of HNC

In the analysis of only 2 studies on the association between hypertension and incidence of combined head and neck cancer, there was limited evidence of an increased risk (RR= 1.11, 95%CI (0.92, 1.33)) using a random-effects model. There was evidence of substantial heterogeneity among studies (*P* heterogeneity= 0.134, I2= 55.6%) (Figure 4a). An increased risk of hypertension was consistent across HNC subsites, with a RR of 1.16 (95%CI (1.05, 1.28)) found for combined oral cavity and oropharyngeal cancer, and the strongest effect found in the larynx (RR= 1.17, 95%CI (1.10, 1.25)) (Figure 4b – c). The amount of heterogeneity was similar in both oral cavity and oropharyngeal combined (*P* heterogeneity= 0.108, I2= 50.6%) and laryngeal cancer analyses (*P* heterogeneity= 0.186, I2= 37.7%). The results remained similar when using a fixed effects model across all sites.

![Figure 4.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F4.medium.gif)

[Figure 4.](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F4)

Figure 4. Relative risks for the association between hypertension and incidence of head and neck cancer combined and by subsite

### Dyslipidaemia and risk of HNC

There were only 2 studies available for meta-analysis, both investigating the effect of dyslipidaemia on incident laryngeal cancer risk. There was some evidence of an increased risk association of both low HDL (RR= 1.12, 95%CI (1.07, 1.18)) and high triglyceride levels (RR= 1.10, 95%CI (1.05, 1.15)) using a random-effects model (Error! Not a valid bookmark self-reference.**a – b**). In the low HDL analysis, there was evidence of moderate heterogeneity (*P* heterogeneity= 0.103, I2= 62.5%), which was not found in the high triglycerides analysis (*P* heterogeneity= 0.319, I2= 0.0%) (Error! Not a valid bookmark self-reference.**a – b**). The results remained similar when using a fixed effects model across all sites. These findings were not supported in the single study by Jiang *et al.*12 which investigated low HDL (RR= 1.01, 95%CI (0.84, 1.22)) and high triglycerides (RR= 0.95, 95%CI (0.81, 1.10)) on head and neck combined cancer (**Table 2**).

### Subgroup analyses

Pre-defined subgroup meta-analyses were conducted for all four metabolic traits according to age, geographic location, year of publication, methods of exposure assessment, risk of bias score, length of follow-up, the number of cancer cases and adjustment for confounding factors including age, sex, smoking and alcohol use (**Supplementary Table 3**).

View this table:
[Supplementary Table 3](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/T5)

Supplementary Table 3 Subgroup analyses for the association of metabolic disorders and risk of head and neck cancer

Stratifying by geographic location of the study resulted in a difference for both obesity (European (RR= 0.97, 95%CI (0.90, 1.04)) vs USA (RR= 1.40, 95%CI (1.08, 1.81), *P* difference= 0.007), type 2 diabetes (Asian (RR= 1.23, 95%CI (1.13, 1.33)) vs European (RR= 1.00, 95%CI (0.90, 1.12)) vs USA (RR= 0.81, 95%CI (0.72, 0.90), *P* difference <0.0001) and dyslipidaemia (Asian (RR= 1.12, 95%CI (1.08, 1.16) vs European (RR= 0.97, 95%CI (0.87, 1.10) vs USA (RR= 0.45, 95%CI (0.21, 0.98), *P* difference= 0.006). Various methods of exposure assessment did result in a differences for both obesity and type 2 diabetes. Within obesity, there were differences between the use of medical records (RR= 0.94, 95%CI (0.86, 1.04) vs self-report (RR= 0.63, 95%CI (0.38, 1.06)) vs measurements taken at baseline (RR= 1.02, 95%CI (0.91, 1.15)) vs a combination of the aforementioned methods (RR= 1.40, 95%CI (1.08, 1.81)) (*P* difference = 0.011).

When stratifying by year of publication (before or after 2010), by adjustment for confounding factors, length of follow-up and number of cancer cases, this resulted in differences only within dyslipidaemia. However, there was only one small study by Chyou *et al.* (1992)65 representing the before 2010, no adjustment, >10 year follow-up and <50 cancer cases categories which is of high risk of bias.

### Risk of bias assessment

Risk of bias assessment was undertaken using the Risk of Bias In Non-randomized Studies – of Exposure (ROBINS-E) tool 42. From all included studies, each result (n= 85) included in the meta-analysis was individually risk assessed across the seven domains of bias, accounting for confounding variables which were pre-defined as important (i.e., age, sex, smoking status, alcohol intake, ethnicity, and socio-economic status). There was no difference between individual results within each study, therefore the overall study ratings only are presented in **Supplementary Figures 1 – 4**. One study was rated as ‘Very high’ risk of bias (3.7%) 18, 21 studies were rated as ‘High’ (77.8%) 15–17,19–24,54–65, one study as ‘Some concerns’ (3.7%) 66 and the remaining studies as ‘Low’ (14.8%) 12,14. All ‘High’ and ‘Very high’ ratings were due to bias that arose due to inadequate adjustment for confounding and/or the handling of missing data.

Funnel plots for each metabolic trait were used to assess asymmetry which may be as a result of true heterogeneity, poor methodological quality in smaller studies, reporting or publication bias 44. For obesity, there appeared to be only limited asymmetry, with only one outlier by Gaudet *et al.*11 and potential smaller studies missing from the bottom right corner (Figure 6a). For type 2 diabetes mellitus, there were significantly more larger studies, many of which lay outside the 95%CI suggesting the presence of heterogeneity. There were eight results from several studies which appear to be outliers in the type 2 diabetes funnel plot, including results from Atchison *et al.* 58, Lo *et al.* 59, Tseng *et al.* 24, Lin *et al.* 62, and Kim *et al.* 19,20. Four out of five of these outlier studies were distinct in that they were conducted in Asian only populations (Figure 6b). There were too few (≤10) studies for hypertension and dyslipidaemia) to confidently distinguish true asymmetry (Figure 6c-d).

### Sensitivity analyses

No association was found when dropping the largest pooled analysis by Gaudet *et al.*11 for obesity in combined head and neck cancer (RR= 0.82, 95%CI (0.43, 1.56) with evidence again for heterogeneity (I2 = 65.9%, *P heterogeneity* = 0.087). This lack of effect for obesity was consistent across subsites, with results all crossing the null. Similarly, there was no difference in the effect of type 2 diabetes on oral cancer risk when leaving out the other large pooled analysis by Gong *et al.*17 (RR= 0.98, 95%CI (0.95, 1.01), with no change in evidence for substantial heterogeneity (I2 = 95.8%, *P heterogeneity =* <0.001).

## Discussion

This is the most comprehensive and up-to-date systematic review and meta-analysis on this topic, considering the association of metabolic traits on HNC across subsites at a global population level. Despite observations from individual studies suggesting an association between BMI and HNC, no effect was found within any subsite in this study, using random-effects meta-analyses (Figure 2). There was also limited evidence of an association for type 2 diabetes (Figure 3), but perhaps a trend towards increased risk for overall HNC, which is in keeping with previous pooled estimates 12,19,20,59. Conversely, there was evidence of an increased HNC risk for both hypertension and dyslipidaemia across subsites, but with fewer studies included in these meta-analyses (Figure 4 & Figure 5). Furthermore, over 80% of studies were judged to be at ‘Very High’ or ‘High’ risk of bias using the ROBINS-E tool (**Supplementary Figures 1 – 4**), so caution must be exercised when interpreting these results. Of those studies assessed as low risk of bias, their results suggest limited evidence for an effect of obesity, but an increased risk association of type 2 diabetes 12 on HNC risk. Conversely, this was not the case for hypertension and dyslipidaemia, where the one study assessed as low risk of bias for these traits did not provide evidence of an effect on HNC risk 12. Variability in the individual study results may exist due to differences in timing of exposure measurement, confounder adjustment, or from other sources of bias.

![Figure 5.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F5.medium.gif)

[Figure 5.](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F5)

Figure 5. Relative risks for the association between dyslipidaemia and incidence of laryngeal cancer

![](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F6/graphic-18.medium.gif)

[](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F6/graphic-18)

![](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F6/graphic-19.medium.gif)

[](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F6/graphic-19)

Figure 6. Funnel plots showing effect estimates for metabolic traits on overall head and neck cancer isk (all subsites)

### Dyslipidaemia and risk of HNC

There were only 2 studies available for meta-analysis, both investigating the effect of dyslipidaemia on incident laryngeal cancer risk. There was some evidence of an increased risk association of both low HDL (RR= 1.12, 95%CI (1.07, 1.18)) and high triglyceride levels (RR= 1.10, 95%CI (1.05, 1.15)) using a random-effects model (Error! Not a valid bookmark self-reference.**a – b**). In the low HDL analysis, there was evidence of moderate heterogeneity (*P* heterogeneity= 0.103, I2= 62.5%), which was not found in the high triglycerides analysis (*P* heterogeneity= 0.319, I2= 0.0%) (Error! Not a valid bookmark self-reference.**a – b**). The results remained similar when using a fixed effects model across all sites. These findings were not supported in the single study by Jiang *et al.*12 which investigated low HDL (RR= 1.01, 95%CI (0.84, 1.22)) and high triglycerides (RR= 0.95, 95%CI (0.81, 1.10)) on head and neck combined cancer (**Table 2**).

An apparent obesity paradox demonstrated in individual studies 55,67 might be explained by BMI as a collider acting as a form of selection bias 68. While tobacco smoking is correlated with obesity, this may be dose dependant and change following smoking cessation 69.

Furthermore, many HNC patients are more likely to be underweight at presentation, particularly those with alcohol use disorders, often correlated with heavy smoking behaviour. Variability in the results may exist due to differences in timing of the exposure measurement, whereby misclassification of participants into the wrong BMI category could have occurred. This is because around one-third to a half of HNC patients experience between 5 – 10% weight loss during the period prior to diagnosis 70. Obesity may also provide additional nutritional reserves giving patients a survival advantage, if for example there was latent tumour progression prior to the baseline diagnosis in these observational studies. There may be issues with measurement error again leading to misclassification in this study, given BMI is simply a function of mass and height and does not specifically indicate metabolically unhealthy adiposity or take into account lean muscle mass. In future, more accurate measures body composition such as bioelectrical impedance or dual-energy X-ray absorptiometry (DEXA) scanning could be used to better differentiate fat mass from lean tissue. This study also lacked detail on HPV status or tumour stage, with more advanced stage disease likely resulting in more significant weight loss 71. Subgroup differences were found across metabolic traits with respect to geographic location. Stronger effects were found for obesity in USA compared with the UK, and for both type 2 diabetes and dyslipidaemia in Asian populations compared with the other continents, which is in keeping with global epidemiology 72,73. Larger subgroup effects were also found when patient’s self-reported their diagnoses compared to researchers using medical records, suggesting recall bias may be present.

### Strengths and limitations of this study

This meta-analysis includes the largest available studies, resulting in a sample size of n= 39,002 HNC cases in a total population size of over 84 million. This suggests the study was well powered to detect any association between metabolic traits and HNC, however there were fewer studies for hypertension and dyslipidaemia compared to the other traits. As there were <10 studies in these analysis, a comprehensive assessment of publication or reporting bias could not be made. However, the lack of smaller positive or negative effect studies for type 2 diabetes suggests potential publication or reporting bias. Both the obesity and type 2 diabetes funnel plots revealed outlier results. Leaving out the two outlier studies in the obesity meta-analysis had little effect on the overall null finding. Evidence of substantial heterogeneity was present (I2 values >50%) between studies across all metabolic traits, which increased when combining HNC subsites as would be expected. In particular for type 2 diabetes whereby four out of five outlier studies were in Asian only populations. This could be due to the differences in HNC aetiology in Asia e.g., betel quid chewing. Future work should include meta-analysing studies separately by geographic region or restricting to particular populations, as this appears to be a clear source of heterogeneity. Other forms of bias such as collider, recall and information bias may have been present, as described in the previous section. Despite multivariable adjustment within studies, residual confounding may still exist, given that HNC risk factors, such as smoking, alcohol and socioeconomic status are difficult to fully account for. While we conducted a number of subgroup analyses, there was insufficient data in the included studies to examine the effect of sex. This may have been important as previous analyses have suggested that the effect of type 2 diabetes on HNC was mainly found in females 17, which may indicate poorer metabolic control, elevated exposure to insulin and subsequent oxidative DNA damage. There were limited data for smoking at specific subsites across the metabolic traits and therefore meta-analysis at this level was not possible.

## Conclusions

The meta-analysis performed in this study suggested there was limited evidence for an association between obesity or type 2 diabetes and HNC risk. However, there was some evidence for increased risk associations with hypertension and dyslipidaemia. These observational associations are susceptible to confounding, bias and reverse causality and results must be interpreted with caution given the high risk of bias detected.

## Authors’ contributions

AG and CR jointly contributed to the development of the protocol, and the drafting, writing and editing of this manuscript. FS contributed to the development of this study and editing of the manuscript. JH contributed to the development of the search strategy. BM, EG, EV, RR and JPTH contributed to the development of this study. MG was responsible for conceptualising this study, drafting and editing this manuscript. All authors contributed to the development of the search strategy. All authors have approved and contributed to the final written manuscript.

## Funding statement

A.G. is a National Institute for Health and Care Research (NIHR) doctoral research fellow (NIHR302605). C.R. is currently supported by a Wellcome Trust GW4-Clinical Academic Training PhD Fellowship. F.S. was supported by a Cancer Research UK (C18281/A29019) programme grant (the Integrative Cancer Epidemiology Programme). E.E.V is supported by Diabetes UK (17/0005587). E.E.V is also supported by the World Cancer Research Fund (WCRF UK), as part of the World Cancer Research Fund International grant programme (IIG_2019_2009). R.C.R. is a de Pass VC research fellow at the University of Bristol (no grant number). J.P.T.H supported by the NIHR Biomedical Research Centre at University Hospitals Bristol and Weston NHS Foundation Trust and the University of Bristol (no grant number), the NIHR Applied Research Collaboration West at University Hospitals Bristol and Weston NHS Foundation Trust (no grant number) and is a National Institute for Health Research (NIHR) Senior Investigator (no grant number). M.G. was supported by a Wellcome Trust GW4-Clinical Academic Training PhD Fellowship. This research was funded in part, by the Wellcome Trust [Grant number 220530/Z/20/Z]. For the purpose of Open Access, the author has applied a CC BY public copyright licence to any Author Accepted Manuscript version arising from this submission. The views expressed in this publication are those of the author(s) and not necessarily those of Wellcome, NIHR, the NHS or Department of Health.

## Competing interests statement

None declared

## Patient and Public Involvement

Patients and/or the public were not involved in the development, conduct or proposed dissemination of this systematic review protocol.

## Data Availability

All data produced in the present work are contained in the manuscript.

## Supplementary information

Supplementary Appendix 1 Systematic review protocol

Supplementary Appendix 2 PRISMA checklist

Supplementary Figures 1 – 4 Risk of bias assessments for each metabolic trait using the ROBINS-E tool

![Supplementary Figure 1](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F7.medium.gif)

[Supplementary Figure 1](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F7)

Supplementary Figure 1 
Obesity

![Supplementary Figure 2](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F8.medium.gif)

[Supplementary Figure 2](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F8)

Supplementary Figure 2 
Type 2 Diabetes

![Supplementary Figure 3](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F9.medium.gif)

[Supplementary Figure 3](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F9)

Supplementary Figure 3 
Hypertension

![Supplementary Figure 4](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2023/09/20/2023.09.19.23295793/F10.medium.gif)

[Supplementary Figure 4](http://medrxiv.org/content/early/2023/09/20/2023.09.19.23295793/F10)

Supplementary Figure 4 
Dyslipidaemia

## Footnotes

*   * Joint first authors

*   Received September 19, 2023.
*   Revision received September 19, 2023.
*   Accepted September 20, 2023.


*   © 2023, Posted by Cold Spring Harbor Laboratory

This pre-print is available under a Creative Commons License (Attribution-NonCommercial-NoDerivs 4.0 International), CC BY-NC-ND 4.0, as described at [http://creativecommons.org/licenses/by-nc-nd/4.0/](http://creativecommons.org/licenses/by-nc-nd/4.0/)

## References

1.  1.Office for National Statistics. Data for 1974 onwards: Office for National Statistics. Adult Smoking habits in Great Britain., <[https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/drugusealcoholandsmoking/datasets/adultsmokinghabitsingreatbritain](https://www.ons.gov.uk/peoplepopulationandcommunity/healthandsocialcare/drugusealcoholandsmoking/datasets/adultsmokinghabitsingreatbritain).> (2020).
    
    

2.  2.Sung, H. et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA: A Cancer Journal for Clinicians, doi:10.3322/caac.21660 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3322/caac.21660&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=33538338&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

3.  3.Johnson, D. E. et al. Head and neck squamous cell carcinoma. Nature Reviews Disease Primers 6, 92, doi:10.1038/s41572-020-00224-3 (2020).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41572-020-00224-3&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=33243986&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

4.  4.Thomas, S. J., Penfold, C. M., Waylen, A. & Ness, A. R. The changing aetiology of head and neck squamous cell cancer: A tale of three cancers? Clin Otolaryngol 43, 999–1003, doi:10.1111/coa.13144 (2018).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/coa.13144&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

5.  5.World Health Organization (WHO). WHO European Regional Obesity Report 2022. [Internet] [cited 2022 *Jun 06]*, <[https://apps.who.int/iris/bitstream/handle/10665/353747/9789289057738-eng.pdf](https://apps.who.int/iris/bitstream/handle/10665/353747/9789289057738-eng.pdf)> (2022).
    
    

6.  6.Arnold, M. et al. Global burden of cancer attributable to high body-mass index in 2012: a population-based study. Lancet Oncol 16, 36–46, doi:10.1016/s1470-2045(14)71123-4 (2015).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S1470-2045(14)71123-4&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25467404&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

7.  7.Tran, K. B. et al. The global burden of cancer attributable to risk factors, 2010-2019: a systematic analysis for the Global Burden of Disease Study 2019. The Lancet 400, 563–591, doi:10.1016/S0140-6736(22)01438-6 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S0140-6736(22)01438-6&link_type=DOI) 

8.  8.Davey, R. C. The obesity epidemic: too much food for thought? British Journal of Sports Medicine 38, 360–363, doi:10.1136/bjsm.2003.007443 (2004).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6ODoiYmpzcG9ydHMiO3M6NToicmVzaWQiO3M6ODoiMzgvMy8zNjAiO3M6NDoiYXRvbSI7czo1MDoiL21lZHJ4aXYvZWFybHkvMjAyMy8wOS8yMC8yMDIzLjA5LjE5LjIzMjk1NzkzLmF0b20iO31zOjg6ImZyYWdtZW50IjtzOjA6IiI7fQ==) 

9.  9.Wotton, C. J., Yeates, D. G. & Goldacre, M. J. Cancer in patients admitted to hospital with diabetes mellitus aged 30 years and over: record linkage studies. Diabetologia 54, 527–534, doi:10.1007/s00125-010-1987-2 (2011).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s00125-010-1987-2&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21116605&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

10. 10.Stott-Miller, M. et al. History of diabetes and risk of head and neck cancer: a pooled analysis from the international head and neck cancer epidemiology consortium. Cancer Epidemiol Biomarkers Prev 21, 294–304, doi:10.1158/1055-9965.EPI-11-0590 (2012).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiY2VicCI7czo1OiJyZXNpZCI7czo4OiIyMS8yLzI5NCI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDIzLzA5LzIwLzIwMjMuMDkuMTkuMjMyOTU3OTMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

11. 11.Gaudet, M. M. et al. Anthropometry and head and neck cancer:a pooled analysis of cohort data. Int J Epidemiol 44, 673–681, doi:10.1093/ije/dyv059 (2015).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/ije/dyv059&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=26050257&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

12. 12.Jiang, H. et al. The effect of metabolic syndrome on head and neck cancer incidence risk: a population-based prospective cohort study. Cancer & metabolism 9, 25, doi:10.1186/s40170-021-00261-w (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/s40170-021-00261-w&link_type=DOI) 

13. 13.Diaz, Y. et al. Body mass index and waist circumference in relation to the risk of 26 types of cancer: a prospective cohort study of 3.5 million adults in Spain. BMC Medicine 19, 10, doi:10.1186/s12916-020-01877-3 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/s12916-020-01877-3&link_type=DOI) 

14. 14.Cao, Z. et al. Association of obesity status and metabolic syndrome with site-specific cancers: a population-based cohort study. Brit J Cancer 123, 1336–1344, doi:10.1038/s41416-020-1012-6 (2020).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41416-020-1012-6&link_type=DOI) 

15. 15.Gribsholt, S. B. et al. Hospital-diagnosed overweight and obesity related to cancer risk: a 40-year Danish cohort study. Journal of Internal Medicine 287, 435–447, doi:10.1111/joim.13013 (2020).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/joim.13013&link_type=DOI) 

16. 16.Ward, H. A. et al. Measured Adiposity in Relation to Head and Neck Cancer Risk in the European Prospective Investigation into Cancer and Nutrition. Cancer epidemiology, biomarkers & prevention: a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology 26, 895–904, doi:10.1158/1055-9965.EPI-16-0886 (2017).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiY2VicCI7czo1OiJyZXNpZCI7czo4OiIyNi82Lzg5NSI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDIzLzA5LzIwLzIwMjMuMDkuMTkuMjMyOTU3OTMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

17. 17.Gong, Y., Wei, B., Yu, L. & Pan, W. Type 2 diabetes mellitus and risk of oral cancer and precancerous lesions: a meta-analysis of observational studies. Oral oncology 51, 332–340, doi:10.1016/j.oraloncology.2015.01.003 (2015).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.oraloncology.2015.01.003&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25650271&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

18. 18.Saarela, K. et al. Cancer incidence among Finnish people with type 2 diabetes during 1989-2014. Eur J Epidemiol 34, 259–265, doi:10.1007/s10654-018-0438-0 (2019).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s10654-018-0438-0&link_type=DOI) 

19. 19.Kim, H.-B., Kim, G.-J., Han, K.-d. & Joo, Y.-H. Changes in metabolic syndrome status and risk of laryngeal cancer: A nationwide cohort study. PLoS ONE 16, e0252872–e0252872, doi:10.1371/journal.pone.0252872 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pone.0252872&link_type=DOI) 

20. 20.Kim, S.-Y., Han, K.-d. & Joo, Y.-H. Metabolic Syndrome and Incidence of Laryngeal Cancer: A Nationwide Cohort Study. Scientific reports 9, 667, doi:10.1038/s41598-018-37061-0 (2019).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41598-018-37061-0&link_type=DOI) 

21. 21.Stocks, T. et al. Blood pressure and risk of cancer incidence and mortality in the Metabolic Syndrome and Cancer Project. Hypertension 59, 802–810, doi:10.1161/HYPERTENSIONAHA.111.189258 (2012).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1161/HYPERTENSIONAHA.111.189258&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=22353615&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

22. 22.Christakoudi, S. et al. Blood pressure and risk of cancer in the European Prospective Investigation into Cancer and Nutrition. Int J Cancer 146, 2680–2693, doi:10.1002/ijc.32576 (2020).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/ijc.32576&link_type=DOI) 

23. 23.Seo, J.-H., Kim, Y.-D., Park, C.-S., Han, K.-d. & Joo, Y.-H. Hypertension is associated with oral, laryngeal, and esophageal cancer: a nationwide population-based study. Scientific reports 10, 10291, doi:10.1038/s41598-020-67329-3 (2020).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41598-020-67329-3&link_type=DOI) 

24. 24.Tseng, K. S., Lin, C., Lin, Y. S. & Weng, S. F. Risk of head and neck cancer in patients with diabetes mellitus: a retrospective cohort study in Taiwan. JAMA Otolaryngol Head Neck Surg 140, 746–753, doi:10.1001/jamaoto.2014.1258 (2014).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1001/jamaoto.2014.1258&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25058016&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

25. 25.Gatenby, R. A. & Gillies, R. J. Why do cancers have high aerobic glycolysis? Nat Rev Cancer 4, 891–899, doi:10.1038/nrc1478 (2004).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/nrc1478&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=15516961&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000224815700015&link_type=ISI) 

26. 26.Warburg, O., Wind, F. & Negelein, E. The metabolism fo tumors in the body. J Gen Physiol 8, 519–530, doi:10.1085/jgp.8.6.519 (1927).
    
    [FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6MzoiUERGIjtzOjExOiJqb3VybmFsQ29kZSI7czozOiJqZ3AiO3M6NToicmVzaWQiO3M6NzoiOC82LzUxOSI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDIzLzA5LzIwLzIwMjMuMDkuMTkuMjMyOTU3OTMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

27. 27.Scully, T., Ettela, A., LeRoith, D. & Gallagher, E. J. Obesity, Type 2 Diabetes, and Cancer Risk. Frontiers in Oncology 10, doi:10.3389/fonc.2020.615375 (2021).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3389/fonc.2020.615375&link_type=DOI) 

28. 28.Wang, W., Bai, L., Li, W. & Cui, J. The Lipid Metabolic Landscape of Cancers and New Therapeutic Perspectives. Frontiers in Oncology 10, doi:10.3389/fonc.2020.605154 (2020).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3389/fonc.2020.605154&link_type=DOI) 

29. 29.Covidence. Covidence systematic review software, Veritas Health Innovation, Melbourne, Australia*. [Internet] [cited* 2022 *July 29]*, <[https://www.covidence.org/](https://www.covidence.org/)> (2022).
    
    

30. 30.Gormley, A. et al. Metabolic disorders and the risk of head and neck cancer: a protocol for a systematic review and meta-analysis. BMJ Open 12, e058392, doi:10.1136/bmjopen-2021-058392 (2022).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NzoiYm1qb3BlbiI7czo1OiJyZXNpZCI7czoxMjoiMTIvNS9lMDU4MzkyIjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjMvMDkvMjAvMjAyMy4wOS4xOS4yMzI5NTc5My5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

31. 31.Liberati, A. et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. BMJ 339, b2700, doi:10.1136/bmj.b2700 (2009).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6MzoiYm1qIjtzOjU6InJlc2lkIjtzOjE3OiIzMzkvanVsMjFfMS9iMjcwMCI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDIzLzA5LzIwLzIwMjMuMDkuMTkuMjMyOTU3OTMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

32. 32.Alberti, K. G. & Zimmet, P. Z. Definition, diagnosis and classification of diabetes mellitus and its complications. Part 1: diagnosis and classification of diabetes mellitus provisional report of a WHO consultation. Diabet Med 15, 539–553, doi:10.1002/(sici)1096-9136(199807)15:7<539::Aid-dia668>3.0.Co;2-s (1998).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/(SICI)1096-9136(199807)15:7<539::AID-DIA668>3.0.CO;2-S&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=9686693&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000074777700003&link_type=ISI) 

33. 33.Alberti, K. G., Zimmet, P. & Shaw, J. The metabolic syndrome--a new worldwide definition. Lancet 366, 1059–1062, doi:10.1016/s0140-6736(05)67402-8 (2005).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/S0140-6736(05)67402-8&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=16182882&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000232114000008&link_type=ISI) 

34. 34.National Cholesterol Education Program (NCEP) Expert Panel. Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation 106, 3143–3421 (2002).
    
    [FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiRlVMTCI7czoxMToiam91cm5hbENvZGUiO3M6MTQ6ImNpcmN1bGF0aW9uYWhhIjtzOjU6InJlc2lkIjtzOjExOiIxMDYvMjUvMzE0MyI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDIzLzA5LzIwLzIwMjMuMDkuMTkuMjMyOTU3OTMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

35. 35.35 World Health Organization (WHO). Obesity and overweight [Internet] [cited 2022 July 26], <[https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight](https://www.who.int/news-room/fact-sheets/detail/obesity-and-overweight)> (2021).
    
    

36. 36.36 World Health Organization (WHO). Definition and Diagnosis of Diabetes Mellitus and Intermediate Hyperglycaemia [Internet] [cited 2022 July 26], <[https://www.diabetes.org.uk/resources-s3/2017-09/Definition\_and_diagnosis1111.pdf](https://www.diabetes.org.uk/resources-s3/2017-09/Definition_and_diagnosis1111.pdf)> (2006).
    
    

37. 37.37 World Health Organization (WHO). Use of Glycated Haemoglobin (HbA1c) in the Diagnosis of Diabetes Mellitus [Internet] [cited 2022 July 26], <[https://www.diabetes.org.uk/resources-s3/2017-09/Definition\_and_diagnosis1111.pdf](https://www.diabetes.org.uk/resources-s3/2017-09/Definition_and_diagnosis1111.pdf)> (2011).
    
    

38. 38.38 World Health Organization (WHO). Hypertension [Internet] [cited 2022 July 26], <[https://www.who.int/news-room/fact-sheets/detail/hypertension](https://www.who.int/news-room/fact-sheets/detail/hypertension).> (2021).
    
    

39. 39.39 British Medical Journal. Hypercholesterolemia [Internet] [cited 2022 July 26], <[https://bestpractice.bmj.com/topics/en-us/170](https://bestpractice.bmj.com/topics/en-us/170)> (2019).
    
    

40. 40.40 World Health Organization (WHO). World Health Organization (WHO): International statistical classification of diseases and related health problems (10th ed.) [Internet] [cited 2022 July 26]. <[https://icd.who.int/browse10/2019/en#](https://icd.who.int/browse10/2019/en#)> (2016).
    
    

41. 41.ROBINS-E Development Group (Higgins J, M. R., Rooney A, Taylor K, Thayer K, Silva R, Lemeris C, Akl A, Arroyave W, Bateson T, Berkman N, Demers P, Forastiere F, Glenn B, Hróbjartsson A, Kirrane E, LaKind J, Luben T, Lunn R, McAleenan A, McGuinness L, Meerpohl J, Mehta S, Nachman R, Obbagy J, O’Connor A, Radke E, Savović J, Schubauer-Berigan M, Schwingl P, Schunemann H, Shea B, Steenland K, Stewart T, Straif K, Tilling K, Verbeek V, Vermeulen R, Viswanathan M, Zahm S, Sterne J). Risk Of Bias In Non-randomized Studies - of Exposure (ROBINS-E). Launch version, 1 June 2022., <[https://www.riskofbias.info/welcome/robins-e-tool](https://www.riskofbias.info/welcome/robins-e-tool)]> (2022).
    
    

42. 42.ROBINS-E Development Group, H. J., Morgan R, Rooney A, Taylor K, Thayer K, Silva R, Lemeris C, Akl A, Arroyave W, Bateson T, Berkman N, Demers P, Forastiere F, Glenn B, Hróbjartsson A, Kirrane E, LaKind J, Luben T, Lunn R, McAleenan A, McGuinness L, Meerpohl J, Mehta S, Nachman R, Obbagy J, O’Connor A, Radke E, Savović J, Schubauer-Berigan M, Schwingl P, Schunemann H, Shea B, Steenland K, Stewart T, Straif K, Tilling K, Verbeek V, Vermeulen R, Viswanathan M, Zahm S, Sterne J. Risk Of Bias In Non-randomized Studies - of Exposure (ROBINS-E) [Internet] *[cited* 2022 *July 29]*, <[https://www.riskofbias.info/welcome/robins-e-tool](https://www.riskofbias.info/welcome/robins-e-tool).> (2022).
    
    

43. 43.DerSimonian, R. & Laird, N. Meta-analysis in clinical trials. Control Clin Trials 7, 177–188, doi:10.1016/0197-2456(86)90046-2 (1986).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/0197-2456(86)90046-2&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=3802833&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=A1986F013900001&link_type=ISI) 

44. 44.Sterne, J. A. C. et al. Recommendations for examining and interpreting funnel plot asymmetry in meta-analyses of randomised controlled trials. BMJ 343, d4002, doi:10.1136/bmj.d4002 (2011).
    
    [FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiRlVMTCI7czoxMToiam91cm5hbENvZGUiO3M6MzoiYm1qIjtzOjU6InJlc2lkIjtzOjE3OiIzNDMvanVsMjJfMS9kNDAwMiI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDIzLzA5LzIwLzIwMjMuMDkuMTkuMjMyOTU3OTMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

45. 45.Andreotti, G. et al. Body mass index, agricultural pesticide use, and cancer incidence in the Agricultural Health Study cohort. Cancer Causes Control 21, 1759–1775, doi:10.1007/s10552-010-9603-9 (2010).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s10552-010-9603-9&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=20730623&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

46. 46.Gaudet, M. M. et al. Prospective studies of body mass index with head and neck cancer incidence and mortality. Cancer epidemiology, biomarkers & prevention 21, 497–503, doi:10.1158/1055-9965.EPI-11-0935 (2012).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiY2VicCI7czo1OiJyZXNpZCI7czo4OiIyMS8zLzQ5NyI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDIzLzA5LzIwLzIwMjMuMDkuMTkuMjMyOTU3OTMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

47. 47.Hashibe, M. et al. Tobacco, alcohol, body mass index, physical activity, and the risk of head and neck cancer in the prostate, lung, colorectal, and ovarian (PLCO) cohort. Head & neck 35, 914–922, doi:10.1002/hed.23052 (2013).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/hed.23052&link_type=DOI) 

48. 48.Etemadi, A. et al. A prospective cohort study of body size and risk of head and neck cancers in the NIH-AARP diet and health study. Cancer epidemiology, biomarkers & prevention 23, 2422–2429, doi:10.1158/1055-9965.EPI-14-0709-T (2014).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiY2VicCI7czo1OiJyZXNpZCI7czoxMDoiMjMvMTEvMjQyMiI7czo0OiJhdG9tIjtzOjUwOiIvbWVkcnhpdi9lYXJseS8yMDIzLzA5LzIwLzIwMjMuMDkuMTkuMjMyOTU3OTMuYXRvbSI7fXM6ODoiZnJhZ21lbnQiO3M6MDoiIjt9) 

49. 49.Hjalgrim, H. et al. Cancer and diabetes--a follow-up study of two population-based cohorts of diabetic patients. J Intern Med 241, 471–475, doi:10.1111/j.1365-2796.1997.tb00004.x (1997).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/j.1365-2796.1997.tb00004.x&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=10497622&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=A1997XL81900004&link_type=ISI) 

50. 50.Wideroff, L. et al. Cancer incidence in a population-based cohort of patients hospitalized with diabetes mellitus in Denmark. J Natl Cancer Inst 89, 1360–1365, doi:10.1093/jnci/89.18.1360 (1997).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/jnci/89.18.1360&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=9308706&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=A1997XW47700010&link_type=ISI) 

51. 51.Stocks, T. et al. Blood glucose and risk of incident and fatal cancer in the metabolic syndrome and cancer project (me-can): analysis of six prospective cohorts. PLoS Med 6, e1000201, doi:10.1371/journal.pmed.1000201 (2009).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pmed.1000201&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=20027213&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

52. 52.Tseng, C.-H. Oral cancer in Taiwan: is diabetes a risk factor? Clinical oral investigations 17, 1357–1364, doi:10.1007/s00784-012-0820-3 (2013).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s00784-012-0820-3&link_type=DOI) 

53. 53.Lai, G. Y., Park, Y., Hartge, P., Hollenbeck, A. R. & Freedman, N. D. The association between self-reported diabetes and cancer incidence in the NIH-AARP Diet and Health Study. J Clin Endocrinol Metab 98, E497–502, doi:10.1210/jc.2012-3335 (2013).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1210/jc.2012-3335&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=23408570&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000316417200011&link_type=ISI) 

54. 54.Samanic, C., Chow, W. H., Gridley, G., Jarvholm, B. & Fraumeni, J. F., Jr.. Relation of body mass index to cancer risk in 362,552 Swedish men. Cancer Causes Control 17, 901–909, doi:10.1007/s10552-006-0023-9 (2006).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s10552-006-0023-9&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=16841257&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000239040100004&link_type=ISI) 

55. 55.Bhaskaran, K., Douglas, I., Forbes, H. & Smeeth, L. Associations between body mass index and a range of cancers: A population-based study within the UK Clinical Practice Research Datalink. Eur J Epidemiol 28, S97, doi:10.1007/s10654-013-9820-0 (2013).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s10654-013-9820-0&link_type=DOI) 

56. 56.Maasland, D. H., van den Brandt, P. A., Kremer, B. & Schouten, L. J. Body mass index and risk of subtypes of head-neck cancer: the Netherlands Cohort Study. Sci Rep 5, 17744, doi:10.1038/srep17744 (2015).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/srep17744&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

57. 57.Chodick, G. et al. Diabetes and risk of incident cancer: A large population-based cohort study in Israel. Cancer Causes and Control 21, 879–887, doi:10.1007/s10552-010-9515-8 (2010).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s10552-010-9515-8&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=20148361&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000277709800010&link_type=ISI) 

58. 58.Atchison, E. A., Gridley, G., Carreon, J. D., Leitzmann, M. F. & McGlynn, K. A. Risk of cancer in a large cohort of U.S. veterans with diabetes. Int J Cancer 128, 635–643, doi:10.1002/ijc.25362 (2011).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/ijc.25362&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=20473855&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000285264000014&link_type=ISI) 

59. 59.Lo, S. F. et al. Modest increase in risk of specific types of cancer types in type 2 diabetes mellitus patients. Int J Cancer 132, 182–188, doi:10.1002/ijc.27597 (2013).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/ijc.27597&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=22510866&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

60. 60.Walker, J. J. et al. Type 2 diabetes, socioeconomic status and risk of cancer in Scotland 2001-2007. Diabetologia 56, 1712–1715, doi:10.1007/s00125-013-2937-6 (2013).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s00125-013-2937-6&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=23661106&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

61. 61.Nakamura, K. et al. Diabetes mellitus and risk of cancer in Takayama: a population-based prospective cohort study in Japan. Cancer Sci 104, 1362–1367, doi:10.1111/cas.12235 (2013).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/cas.12235&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=23859808&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

62. 62.Lin, C. C. et al. Cancer risks among patients with type 2 diabetes: A 10-year follow-up study of a nationwide population-based cohort in Taiwan. BMC Cancer 14, 381, doi:10.1186/1471-2407-14-381 (2014).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1186/1471-2407-14-381&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=24884617&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

63. 63.He, X., Shi, L. & Wu, J. Retrospective database analysis of cancer risk in patients with type 2 diabetes mellitus in China. Current Medical Research and Opinion 34, 1089–1098, doi:10.1080/03007995.2017.1421527 (2018).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1080/03007995.2017.1421527&link_type=DOI) 

64. 64.Qi, J. et al. Cancer risk among patients with type 2 diabetes: A real-world study in Shanghai, China. J Diabetes 11, 878–883, doi:10.1111/1753-0407.12926 (2019).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/1753-0407.12926&link_type=DOI) 

65. 65.Chyou, P. H., Nomura, A. M., Stemmermann, G. N. & Kato, I. Prospective study of serum cholesterol and site-specific cancers. Journal of clinical epidemiology 45, 287–292 (1992).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/0895-4356(92)90089-6&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=1569426&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=A1992HR88500011&link_type=ISI) 

66. 66.Gaudet, M. M. et al. Anthropometry and head and neck cancer:a pooled analysis of cohort data. Int J Epidemiol 44, 673–681, doi:10.1093/ije/dyv059 (2015).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1093/ije/dyv059&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=26050257&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

67. 67.Chen, Y. et al. Body mass index and the risk of head and neck cancer in the Chinese population. Cancer Epidemiol 60, 208–215, doi:10.1016/j.canep.2019.04.008 (2019).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.canep.2019.04.008&link_type=DOI) 

68. 68.Holmberg, M. J. & Andersen, L. W. Collider Bias. JAMA 327, 1282–1283, doi:10.1001/jama.2022.1820 (2022).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1001/jama.2022.1820&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=35285854&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

69. 69.Dare, S., Mackay, D. F. & Pell, J. P. Relationship between smoking and obesity: a cross-sectional study of 499,504 middle-aged adults in the UK general population. PLoS One 10, e0123579, doi:10.1371/journal.pone.0123579 (2015).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1371/journal.pone.0123579&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=25886648&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

70. 70.Jager-Wittenaar, H. et al. Critical weight loss in head and neck cancer--prevalence and risk factors at diagnosis: an explorative study. Support Care Cancer 15, 1045–1050, doi:10.1007/s00520-006-0212-9 (2007).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s00520-006-0212-9&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=17277925&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 
    
    [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000249122500005&link_type=ISI) 

71. 71.Strulov Shachar S. & Williams, G. R. The Obesity Paradox in Cancer-Moving Beyond BMI. Cancer Epidemiol Biomarkers Prev 26, 13–16, doi:10.1158/1055-9965.Epi-16-0439 (2017).
    
    [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6NDoiY2VicCI7czo1OiJyZXNpZCI7czo3OiIyNi8xLzEzIjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjMvMDkvMjAvMjAyMy4wOS4xOS4yMzI5NTc5My5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 

72. 72.Saeedi, 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, 9(th) edition. Diabetes Res Clin Pract 157, 107843, doi:10.1016/j.diabres.2019.107843 (2019).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.diabres.2019.107843&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom) 

73. 73.Blüher, M. Obesity: global epidemiology and pathogenesis. Nature Reviews Endocrinology 15, 288–298, doi:10.1038/s41574-019-0176-8 (2019).
    
    [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1038/s41574-019-0176-8&link_type=DOI) 
    
    [PubMed](http://medrxiv.org/lookup/external-ref?access_num=30814686&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2023%2F09%2F20%2F2023.09.19.23295793.atom)