Oesophageal cancer risk factors

89% of oesophageal cancer cases each year in the UK are linked to major lifestyle and other risk factors.[1]

Oesophageal cancer is associated with a number of risk factors.[2-4]

Oesophageal Cancer Risk Factors

Oesophageal cancer risk is strongly related to age and sex.

Tobacco smoking is classified by the International Agency for Research on Cancer (IARC) as a cause of oesophageal cancer.[1] An estimated 66% (63% in males and 71% in females) of oesophageal cancers in the UK are linked to tobacco smoking.[2]

Oesophageal adenocarcinoma (AC) risk is 85-96% higher in ever-smokers compared with never-smokers, meta- and pooled analyses have shown.[1,2]

Oesophageal AC risk is 2.7-2.8 times higher in smokers with 45-60+ pack-years , compared with never-smokers, meta- and pooled analyses have shown; risk increases with number of pack-years.[2,3] Oesophageal AC risk is 2.3 times higher in people with 40+ years of cigarette smoking, compared with never-smokers, a meta-analysis showed.[1]

Oesophageal AC risk is 2.5 times higher in people who smoke 20+ cigarettes per day, compared with never-smokers, one meta-analysis showed;[1] however no association was found in a pooled analysis.[3]

Oesophageal AC risk is 29% lower in ex-smokers who quit 10+ years previously compared with continuing smokers, a pooled analysis showed; however those ex-smokers are still at 72% higher risk compared with never-smokers.[2]

Oesophageal cancer (overall) risk is associated with waterpipe tobacco smoking (smoking flavoured tobacco through a long pipe attached to a water bowl), a meta-analysis from mainly African and Asian studies showed; however confounding by cigarette smoking is possible.[4]

Oesophageal squamous cell carcinoma (SCC)  risk is 4.2 times higher in current smokers in Europe compared with never-smokers, a meta-analysis showed.[1]

Oesophageal SCC risk is 5.6 times higher in smokers with 60+ pack-years , compared with never-smokers, a pooled analysis showed; risk increases with number of pack-years.[2]

Oesophageal SCC risk among smokers is not associated with number of cigarettes smoked per day, a pooled analysis showed.[2]

Smoking and drinking alcohol have a synergistic effect on oesophageal SCC risk: their effect in combination is almost double the sum of their effects individually, a meta-analysis showed.[3]

Smokeless tobacco and betel quid (with and without tobacco), are classified by the International Agency for Research on Cancer (IARC) as causes of oesophageal cancer.[1]

Oesophageal squamous cell carcinoma (SCC)  risk is around three times higher in people who chew areca nut (often included in betel quid), compared with non-users, a meta-analysis showed.[2] Oesophageal SCC risk is around 6.8 times higher in people who chew areca nut and smoke, compared with people who do neither, a meta-analysis showed.[2]

Oesophageal cancer risk is 60% higher in ever-users of smokeless tobacco (including snuff and snus), compared with never-users, a meta-analysis showed.[3]

Body fatness is classified by the International Agency for Research on Cancer (IARC) and World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) as a cause of oesophageal adenocarcinoma (AC) (not squamous cell carcinoma (SCC)).[1,2] An estimated 22% (27% in males and 11% in females) of oesophageal cancers in the UK are linked to overweight and obesity.[3]

Oesophageal AC risk is 13% higher per 5-unit body mass index (BMI) increase, a meta-analysis of case-control and cohort studies showed;[4] meta-analyses of case-control studies only show a larger effect.[5,6]

Oesophageal AC risk is 2.5 times higher in people with the highest abdominal fatness (measured by waist circumference or waist-to-hip ratio), compared with healthy abdominal fatness level, a meta-analysis showed.[7]

Oesophageal  squamous cell carcinoma (SCC)  risk may be decreased with higher body mass index (BMI), meta-analyses have shown; however this may reflect confounding by smoking.[1,2]

Consumption of alcoholic beverages (and acetaldehyde associated with this) is classified by the International Agency for Research on Cancer (IARC) and World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) as a cause of oesophageal cancer (WCRF/AICR classify an association with oesophageal squamous cell carcinoma [SCC] only).[1,2] An estimated 21% (25% in males and 11% in females) of oesophageal cancers in the UK are linked to drinking alcohol.[3]

Oesophageal adenocarcinoma (AC)  risk is not associated with alcohol drinking overall (versus not drinking), drink-years (one drink year is e.g. one alcoholic drink per day for one year, or 2 drinks per day for half a year) or number of alcoholic drinks consumed per day, meta-analyses have shown.[1,2]

Oesophageal squamous cell carcinoma (SCC) risk is 34% higher in people who consume up to 12.5g (1.5 units) of alcohol per day, 2.6 times higher in those who consume around 12.5-50g (1.5-6 units) of alcohol per day, and 5.5 times higher in those who consume 50g+ (6+ units) of alcohol per day, compared with non- or occasional drinkers, a meta-analysis showed.[1] The risk increase is independent of, but compounded by, smoking.[2]

Oesophageal SCC risk is 3.8 times higher in smokers with 200+ drink-years, compared with never-drinkers, a meta-analysis showed; risk increases with number of drink-years.[2]

Oesophageal cancer risk is no higher in ex-drinkers who quit 16.5 years previously compared with never-drinkers, a pooled analysis showed.[3]

Smoking and alcohol have a synergistic effect on oesophageal SCC risk.

Oesophageal squamous cell carcinoma (SCC) risk is 57% higher in people with the highest red meat intake, compared with those with the lowest, a meta-analysis showed.[1]

Consumption of processed meat is classified by the World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) as a possible cause of oesophageal squamous cell carcinoma (SCC), based on limited-suggestive evidence.[1] 

Oesophageal SCC risk is 55% higher in people with the highest processed meat intake, compared with those with the lowest, a meta-analysis showed.[2]

Consumption of hot maté (a south American drink) is classified by the International Agency for Research on Cancer (IARC) and World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) as a probable cause of oesophageal cancer, based on limited evidence.[1]

Oesophageal squamous cell carcinoma (SCC) risk is 2.6 times higher in maté ever-drinkers versus never-drinkers, a meta-analysis showed.[2] However, a pooled analysis of case-control studies found the risk increase was only 60%.[3]

Oesophageal SCC risk is higher in those with higher maté intake, and those who drink maté at higher temperatures.[2-4]

Consumption of high-temperature drinks is classified by the World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) as a possible cause of oesophageal cancer, based on limited evidence.[1] 

Oesophageal cancer (overall) risk is higher in people who consume tea, coffee, other drinks or food at higher temperatures, a systematic review showed.[2]

Consumption of pickled vegetables (traditional Asian style) is classified by the International Agency for Research on Cancer (IARC) as a probable cause of oesophageal cancer, based on limited evidence.[1,2]

Oesophageal cancer (overall) risk is around doubled in people who consume pickled vegetables, compared with those who do not, a meta-analysis showed; most evidence is for squamous cell carcinoma (SCC).[3]

Oesophageal adenocarcinoma (AC)  risk is more than 11 times higher in people with Barrett’s oesophagus (BO)  versus the general population, a cohort study showed.[1] Oesophageal AC develops in around 1 per 1,000 BO patients per year, cohort studies have shown.[1,2] on the basis of these risk levels 3-10% of people with Barrett’s oesophagus in the UK will develop oesophageal AC in their lifetime.[3] However, previous studies estimate a much higher risk of around 1 oesophageal AC case per 160-190 BO patients per year.[4,5] on the basis of these risk levels 7-13% of people with Barrett’s oesophagus in the UK will develop oesophageal AC in their lifetime.[3]

Oesophageal AC risk among BO patients increases with BO extent (higher in long-segment than short-segment) and severity (progressively higher through non-dysplastic[4,6] low-grade dysplastic or high-grade dysplastic).Oesophageal AC risk among BO patients may be higher in males than females, and in smokers than non-smokers.[7]

Risk of oesophageal AC or BO with high-grade dysplasia, among BO patients, is 71% lower in those using proton pump inhibitors (PPIs), and 36% lower in those using cyclooxygenase (COX) inhibitors, versus non-users, meta-analyses have shown.[8,9] The risk reduction with PPI use may be greater in, or limited to, longer-term use.[8] Oesophageal AC risk among BO patients is 36% lower in non-steroidal anti-inflammatory drugs (NSAIDs) users versus non-users, and 41-47% lower in statin users, both compared with non-users, meta-analyses have shown.[10-13] The risk reduction with statins may be limited to those with high-grade dysplasia and may be confounded by NSAIDs use.[14]

Up to 1.6% of the adult population may have BO, small cross-sectional studies suggest.[1] Because BO prevalence appears to be low, population screening is not recommended.[1] However regular endoscopic surveillance for BO patients is recommended.[1]

Oesophageal squamous cell carcinoma (SCC) risk is not associated with BO.

Barrett’s oesophagus (BO) risk is higher in males, older people, white people, and those with a history of reflux.[1,2] BO risk is also higher in ever-smokers,[3] people with higher waist circumference,[4,5] people with hiatal hernia,[6] and people with low fibre intake.[7] BO risk may be lower in people with H. Pylori infection, but evidence is mixed.[8,9]

Oesophageal adenocarcinoma (AC)  risk is 4.9 times higher in people who have gastro-oesophageal reflux disease (GORD, or GERD in American English spelling) symptoms at least weekly, versus people who have GORD symptoms less frequently or never, a meta-analysis showed.[1] Oesophageal AC risk is 7.4 times higher in people who have GORD symptoms daily, versus people who have GORD symptoms less frequently or never, a meta-analysis showed.[1]

Oesophageal AC risk is 2.8 times higher in people who have had GORD symptoms for under 10 years, and 6.2 times higher in those who have had GORD symptoms for 20 years or more, both versus people who have never had GORD symptoms, a pooled analysis showed.[2]

Barrett’s oesophagus (BO) develops in 5-13% of people with GORD.[3] Long-segment BO risk is 3-4.9 times higher in people with GORD symptoms, versus people without GORD symptoms, a meta-analysis showed.[4] Short-segment BO risk is probably not associated with GORD, but results are mixed.[4]

Around 15% of people in Europe have GORD, a systematic review showed.[5]

Oesophageal squamous cell carcinoma (SCC) risk is 94% higher in people with gastric atrophy  versus those without, a meta-analysis showed. Oesophageal adenocarcinoma (AC) risk is not associated with gastric atrophy.[1]

Oesophageal cancer (overall) risk is 28% higher in diabetic men compared with non-diabetic men, a meta-analysis showed.[1] Oesophageal cancer risk in women is not associated with diabetes, a meta-analysis showed.[1] The association with diabetes may be stronger for oesophageal adenocarcinoma (AC).[1]

Oesophageal cancer (overall) risk in diabetics is probably not associated with diabetes treatment type, a case-control study showed.[2]

Oesophageal adenocarcinoma (AC) risk may be higher in people hospitalised with asthma, cohort studies suggest.[1] Risk is 55% higher in theophylline (an asthma medication) users compared with non-users, a meta-analysis showed.[2] Asthma is associated with gastro-oesophageal reflux disease (GORD),[3] this may be because theophylline relaxes the oesophageal sphincter so can cause or exacerbate reflux.[2]

Oesophageal cancer risk is higher in people with achalasia,[1] coeliac disease [2], or Plummer-Vinson syndrome.[3]

Oesophageal cancer (overall) risk is 74% higher in bisphosphonate users compared with non-users, one meta-analysis showed;[1] however previous meta-analyses showed no association.[2,3] The risk increase may be higher in/limited to long-term use, and Etidronate (not Aledronate).[1]

Oesophageal adenocarcinoma (AC) risk is 66% higher in anticholinergics users compared with non-users, a meta-analysis showed.[1] Anticholinergics relax the oesophageal sphincter so can cause or exacerbate reflux.

Oesophageal  squamous cell carcinoma (SCC)  risk is not associated with anticholinergics use.[1]

X-radiation and Gamma radiation are classified by the International Agency for Research on Cancer (IARC) as causes of oesophageal cancer.[1] An estimated 3% (2% in males and 4% in females) of oesophageal cancers in the UK are linked to ionising radiation, mainly radiotherapy for previous cancer.[2]

Oesophageal cancer (overall) associated with radiotherapy for previous cancer occurs in survivors of breast (though evidence is mixed[3]), lung, oropharyngeal and laryngeal cancers.[4]

Oesophageal cancer accounts for an estimated 46% of radiotherapy-attributable second cancers in breast cancer survivors in the UK.[4]

Oesophageal cancer (overall) risk is higher in survivors of Hodgkin lymphoma (5.2 times higher), larynx cancer (3.2 times higher), oral cavity and pharynx cancer (2.2 times higher), female breast cancer (2.1 times higher), lung cancer (80% higher), non-Hodgkin lymphoma (98% higher), testicular cancer (95% higher), or cervical cancer (37% higher), 10 or more years after their primary cancer, a pooled analysis showed.[1]

Oesophageal cancer secondary to another cancer is typically squamous cell carcinoma (SCC), though the SCC proportion varies by site of the primary cancer.[1]

This may reflect the effect of treatment for the primary cancer, or shared lifestyle, environmental or genetic risk factors.

Working in dry-cleaning (entailing exposure to tetrachloroethylene [PCE] and trichloroethylene [TCE]), or the rubber production industry, are classified by the International Agency for Research on Cancer (IARC) as probable causes of oesophageal cancer, based on limited evidence.[1,2] An estimated 3% (3% in males and 1% in females) of oesophageal cancers in the UK are linked to occupational exposures to soots or tetrachloroethylene.[3,4]

Oesophageal adenocarcinoma (AC) risk is not associated with occupational TCE exposure, a pooled analysis showed.[5]

Oesophageal adenocarcinoma (AC) risk is 3.5 times higher in people with a parent diagnosed with oesophageal cancer (overall), a cohort study showed.[1] The risk may be higher in people with an affected mother, and in people with a parent with squamous cell carcinoma (SCC).[1]

Oesophageal SCC risk is not associated with oesophageal cancer in a parent.[1]

Around 7% of people with Barrett’s oesophagus (BO) have a first-degree relative (parent, sibling, or child) with BO or oesophageal AC, a cohort study showed.[2]

Oesophageal cancer risk is higher in people whose spouse has/had upper aerodigestive tract cancer, perhaps indicating a role for shared behavioural risk factors such as smoking in familial clustering of oesophageal cancer.[3]

Oesophageal squamous cell carcinoma (SCC) risk is higher in people with tylosis.[1]

Oesophageal adenocarcinoma (AC) risk is 25-32% lower in hormone replacement therapy (HRT)  ever-users compared with never-users, meta-analyses have shown.[1,2]

Oesophageal cancer (overall) risk is not associated with HRT use, a pooled analysis of case-control studies showed.[3]

Oesophageal cancer (overall) risk is 58% lower in mothers who breastfed for at least 12 months, compared with mothers who did not breastfeed at all, a pooled analysis showed.[1]

Oesophageal cancer (overall) risk is 69% higher in people with AIDS , compared with uninfected people or the general population, a pooled analysis showed.[1]

The risk increase is greater for oesophageal adenocarcinoma (AC) (91% higher risk) than for oesophageal squamous cell carcinoma (SCC) (47% higher risk), pooled analyses have shown.[1]

Oesophageal cancer risk overall is not associated with HIV infection in the absence of AIDS, or HIV and AIDS combined, meta- and pooled analyses have shown.[1,2]

Oesophageal squamous cell carcinoma (SCC) risk is 2.4-3.6 times higher in people with HPV-16 infection, compared with uninfected people, meta-analyses have shown.[1,2]

Oesophageal SCC risk is not associated with HPV-18 infection.[1]

Oesophageal cancer (overall) risk may be higher in recipients of solid organ transplant or stem cell transplant, compared with the general population;[1-3] however the size of the risk increase varies widely, and some studies show no association.[4]

Oesophageal cancer risk is 86% higher in people with systemic lupus erythematosus, compared with the general population, a meta-analysis showed.[1]

Non-starchy vegetables, fruits (oesophageal squamous cell carcinoma only), and physical activity are classified by the World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) as possibly protective against oesophageal cancer, based on limited-suggestive evidence.[1] An estimated 46% of oesophageal cancers in the UK are linked to eating fewer than five portions of fruit and vegetables per day.[2]

Oesophageal cancer risk is lower in people with the highest intake of the following foods, versus those with the lowest intake, meta- and pooled analyses or systematic reviews have shown:

• Vegetables (oesophageal adenocarcinoma (AC)) – 24% lower risk.[3]
• Vegetables (oesophageal squamous cell carcinoma (SCC)) – 44% lower risk.[4]
• Fruit (oesophageal AC) – 27% lower risk.[3]
• Fruit (oesophageal SCC) – 47% lower risk.[4]
• Citrus fruit (oesophageal SCC) – 37% lower risk. [5]
• Dietary folate (oesophageal AC) – 43% lower risk.[6]
• Dietary/blood folate (oesophageal SCC) – 37% lower risk.[6]
• Dietary fibre (oesophageal AC) – 50% lower risk.[7]
• Dietary fibre (oesophageal SCC) – 47% lower risk.[7]
• Dietary carotenoids – 19-42% lower risk (some evidence of variation by oesophageal cancer type and population).[8]

Oesophageal cancer risk is lower in people who use the following medications or have the following medical conditions, compared with people who do not, meta- and pooled analyses, systematic reviews or large cohort studies have shown:

• Aspirin (oesophageal adenocarcinoma (AC)) – 25% lower risk in twice-weekly users versus non-users (risk reduction plateaus at use 4.5 times per week).[1]
• Aspirin (oesophageal AC) – 47% lower risk in users of 3 years versus non-users (risk reduction plateaus at 6 years use).[1]
• Aspirin (oesophageal squamous cell carcinoma (SCC)) – 39% lower in regular users versus never-users.[2]
• Non-aspirin non-steroidal anti-inflammatory drugs (NA-NSAIDs) (oesophageal AC) – 32-35% lower risk in ever-users versus non-users.[3,4]
• NA-NSAIDs (oesophageal SCC) – risk around halved in ever-users versus non-users.[5]
• Statins (oesophageal cancer overall) – 14-28% lower in users versus non-users.[6-8]
• Nitrates (oesophageal SCC) – risk lower in users versus non-users.[9]
• Angiotensin-converting-enzyme (ACE) inhibitors and angiotensin-receptor blockers (oesophageal cancer overall) – risk lower in users versus non-users.[10]
• Parkinson’s disease – 15% lower risk in people with the condition versus those without (probably linked with lower smoking rates in people with Parkinson’s disease).[11]

Oesophageal adenocarcinoma (AC) risk is 21-32% lower in the most physically active people compared with the least, meta-analyses have shown.[1,2] Oesophageal squamous cell carcinoma (SCC) risk is 34% lower in the most physically active people compared with the least, one meta-analysis showed;[1] however another meta-analysis showed no association.[2]

Oesophageal adenocarcinoma (AC) risk is 41-43% lower in people with H. Pylori infection, compared with uninfected people, meta-analyses have shown.[1,2] This may be linked with lower gastric acid levels in H. pylori infection.[2]

Oesophageal squamous cell carcinoma (SCC) risk is overall not associated with H. Pylori infection, meta-analyses have shown, though this varies by H. Pylori type and population.[1,2]

The World Cancer Research Fund/American Institute for Cancer Research (WCRF/AICR) make no judgement on the association between oesophageal adenocarcinoma (AC) or oesophageal squamous cell carcinoma (SCC) risk and intake of dietary fibre; fruit (oesophageal AC only); red meat; processed meat (oesophageal AC only); total meat; poultry; fish; coffee; high-temperature drinks; mate (oesophageal AC only); alcohol (oesophageal AC only); pyridoxine; vitamin C; vitamin E; folate; beta-carotene; body fatness (oesophageal SCC only); adult attained height; patterns of diet; cereals (grains) and their products; starchy roots, tubers, and plantains; pulses (legumes); soya and soya products; herbs, spices, and condiments; poultry; fish; eggs; milk and dairy products; total fat; saturated fatty acids; monounsaturated fatty acids; polyunsaturated fatty acids; sugary foods and drinks; salt; salting; fermenting; pickling; smoked and cured foods; nitrates and nitrites; frying; grilling (broiling) and barbecuing (charbroiling); protein; vitamin A; retinol; thiamin; riboflavin; calcium; iron; zinc; pro-vitamin A carotenoids; beta-cryptoxanthin; or energy intake, due to limited evidence.[1]

Oesophageal cancer risk is not associated with the following factors, meta- and pooled analyses or systematic reviews have shown:

• Zinc intake (though some evidence of risk decrease limited to Asian populations).[2]
• Dietary fibre (oesophageal squamous cell carcinoma (SCC)).[3]
• Green tea (though some evidence of risk decrease limited to Asian populations, case-control studies, females).[4,5]
• Black tea.[4]
• Coffee.[4,6] (though some evidence of risk decrease)
• Fish (though some evidence of SCC risk decrease limited to hospital-based case-control studies).[7]
• Red and processed meat (oesophageal AC) (though some evidence of risk increase limited to case-control studies).[9] Note red and processed meat intake is associated with oesophageal SCC risk.
• Parity (oesophageal cancer overall).[9]
• Age at menarche (oesophageal cancer overall).[9]
• Age at menopause (oesophageal cancer overall).[9]
• Oral contraceptives (AC alone,[10] and oesophageal cancer overall).[9]
• Gallstones.[11]
• Dietary acrylamide (though risk increase for obese people).[12]
• Vitamin D supplements.[13]
• Allium vegetables.[14]
• Nitrate and nitrite.[15]