Advertisement

Oral Cancer

, Volume 2, Issue 1–2, pp 57–65 | Cite as

Tea, coffee, and head and neck cancer risk in a multicenter study in east Asia

  • Shuang Li
  • Yuan-Chin Amy Lee
  • Qian Li
  • Chien-Jen Chen
  • Wan-Lun Hsu
  • Pen-Jen Lou
  • Cairong Zhu
  • Jian Pan
  • Hongbing Shen
  • Hongxia Ma
  • Lin Cai
  • Baochang He
  • Yu Wang
  • Xiaoyan Zhou
  • Qinghai Ji
  • Baosen Zhou
  • Wei Wu
  • Jie Ma
  • Paolo Boffetta
  • Zuo-Feng Zhang
  • Min Dai
  • Mia Hashibe
Original Article
  • 216 Downloads
Part of the following topical collections:
  1. Epidemiology and Etiology

Abstract

Purpose

The role of tea or coffee on head and neck cancer (HNC) risk has not been consistent, especially in Asian populations. The aim of the study was to investigate whether tea or coffee is associated with HNC risks in a Chinese population.

Methods

Our case–control study included 921 head and neck cancer cases and 806 controls. We obtained information on tea and coffee consumption frequency, duration, age at start, strength, and types. Unconditional logistic regression was used to estimate the odds ratios (ORs) and 95% confidence intervals (95% CIs).

Results

No association between tea drinking and HNC risk was observed after adjusting for potential confounders. Coffee drinking was inversely associated with oral cavity cancer risk with dose–response trends for both frequency (p for trend = 0.0149) and the duration (p for trend = 0.0167) of intake. When stratified by age, sex, and cigarette smoking, coffee drinking was not clearly associated with head and neck cancer risk in terms of frequency or duration.

Conclusion

Coffee consumption was associated with a decreased risk of oral cavity cancer. Further investigations with more specific tea consumption-related factors are warranted to examine the relationship between tea drinking and HNC risks.

Keywords

Head and neck cancer risk Coffee consumption Tea consumption 

Introduction

Tobacco smoking and alcohol consumption are established major risk factors for head and neck cancers (HNC), and have more than multiplicative joint effects [1]. Tobacco smoking and alcohol consumption are prevalent in China, but the HNC incidence rates are very low. The estimated HNC age-standardized incidence rate in China was 2.7 per 100,000, compared with the world rate of 8.0 per 100,000 in 2012 [2]. The incidence rate for oral and pharyngeal cancers in Taiwan in 2011 was 22 per 100,000, which is higher than the incidence in mainland China [3]. Tea and coffee are the most popular hot beverages in the world [4]. Few studies have been conducted to study the relationship between HNC risk and tea or coffee consumption in Chinese population. A study with 396 HNC cases and 413 controls reported an inverse association between tea consumption and HNC risk in Taiwan (OR for every cup per day = 0.96, 95% CI 0.93–0.99) [5]. Another study with 723 oral cavity cancer cases and 857 controls reported that green tea may reduce oral cancer risk among Chinese males (≥ 8 g/day compared with < 4 g/day OR = 0.72, 95% CI 0.54–0.93) [6]. Other case–control studies, however, did not find statistically significant associations between tea and HNC risk [7, 8, 9, 10, 11, 12].

The results between coffee drinking and HNC risk have been inconsistent among studies. Some studies found no overall association between coffee drinking and HNC risk [7, 8, 10, 11]. Franco et al. reported that coffee was associated with HNC risk, but not associated after adjusting for potential confounders [8]. In a large pooled data of case–control studies including 5139 cases and 9028 controls, coffee consumption was protective against HNC [9]. A large study in Japan of 961 cases and 2883 controls also reported a decreased risk of upper aerodigestive tract cancers for coffee drinkers [13].

In terms of cohort studies, a study in the US including 145 head and neck cancers reported no association with coffee or tea drinking and the risk of head and neck cancer [1]. Another cohort study in Japan including 37 oral cavity cancer patients did not report any significant associations between tea consumption and oral cancer risk, but the risk estimates were suggestive of reduced risks in women [14]. A cohort study in Japan of 157 upper aerodigestive tract cancers reported a 50% decreased risk due to coffee intake [15].

Given the high prevalence of tea and/or coffee consumption, as well as low incidence and mortality among the Chinese population, it is of interest to investigate the relationship between tea or coffee consumption and the risk of HNC. The aim of the study is to investigate whether tea or coffee is associated with head and neck cancer risks in this multicenter case–control study of Chinese population.

Materials and methods

Study design and population

We conducted a multicenter case–control study in eight centers in East Asia. Starting from December 2010, the eight centers (Beijing, Fujian, Henan, Jiangsu, Liaoning, Shanghai, Sichuan, and Taiwan) recruited 921 incident HNC cases, including oral cavity, oropharynx, hypopharynx, and larynx, and 806 controls as of February 2015. The face-to-face interviews of both cases and controls were structured to obtain information on the current and previous alcohol consumption (ever status, duration, and frequency), dietary habits (12 food items), tobacco consumption (ever status, duration, and frequency) including cigarette, pipe, chewing tobacco, betel quid, family history of cancer, occupation, medical history, oral hygiene, and vitamin supplementation. Written consents for participation were obtained from all study participants. Ethical approval for human subject research was obtained at the University of Utah (University of Utah IRB no. 00041033 11/10/2010), Fujian (Fujian IRB 3/10/2011), Henan (Henan Cancer Hospital 2011), Shanghai (Fudan University IRB no. 101293-15 12/27/2010), Sichuan (Sichuan University IRB 9/29/2010), Taiwan (National Taiwan University Hospital no. 201006077R 8/10/2010), and Beijing (Chinese Academy of Medical Sciences 2012).

The inclusion criteria for cases were: 1) age 18–80 years, 2) incident cases of HNC [tumors were assigned to one of the five categories as follows (ICD-0-2): (1) oral cavity (includes lip, tongue [except for base of tongue], gum, floor of mouth, hard palate, and cheek mucosa): codes C00.3–C00.9, C02.0–C02.3, C03.0, C03.1, C03.9, C04.0, C04.1, C04.8, C04.9, C05.0, C06.0–C06.2, C06.8, and C06.9; (2) oropharynx (includes base of tongue, lingual tonsil, soft palate, uvula, palatine tonsil, and oropharynx): codes C01.9, C02.4, C05.1, C05.2, C09.0, C09.1, C09.8, C09.9, C10.0, C10.2–C10.4, C10.8, and C10.9; (3) hypopharynx (includes pyriform sinus and hypopharynx): codes C12.9, C13.0–C13.2, C13.8, and C13.9; (4) oral cavity, pharynx unspecified, or overlapping: codes C02.8, C02.9, C05.8, C05.9, C14.0, C14.2, and C14.8; and (5) larynx (includes glottis, supraglottis, and subglottis): codes C10.1, C32.0–C32.3, and C32.8–C32.9], 3) final diagnosis based on histological or cytological confirmation, and 4) interviews performed within 6 months of cancer diagnosis. Controls were frequency-matched by sex, 5-year age group, ethnicity, and residence area from hospitals at each of the centers. Controls were selected from a defined list of diseases unrelated to alcohol, tobacco, or dietary practices. The proportion of hospital controls within a particular diagnostic group did not exceed 33%. These groups were 1) benign tumor disorders [D10–D36], 2) endocrine and metabolic disorders [E00–E07, E24–E27], 3) skin, subcutaneous tissue, and musculoskeletal disorders [M05–M06, M15–M19, M50, M51, M53, M54, N60, L04.0, L05, L21, L60.0, L72.0, R22], 4) trauma [S00-T98], 5) circulatory disorders [I83], 6) ear, eye, and mastoid disorders [H00–H24, H27, H29–H95, except H36.0], 7) diseases of upper-respiratory tract [J30–J39], 8) diseases of the oral cavity, salivary gland, and jaws [K01–K02], 9) gastro-intestinal disorders [K35–K38, K40–K46, K50–K52, K55–K63, K80, K81, K83.0], 10) disorders of the nervous system [G00–G19, G23–G99], 11) other diseases [A49.9, I10, J00, J02.9, Q87.0, Q89.2], and 12) no diagnosis (healthy population). Hospital controls were randomly chosen from subjects admitted as in-patients or out-patients in the same hospital as the cases, and they were in the hospital for less than 1 month when recruited.

We excluded individuals who had missing education status (n = 3), and restricted to squamous cell carcinomas. In the final analysis data set, there were 921 cases (424 oral cavity, 106 oropharynx, 81 hypopharynx, 85 larynx, and 225 unspecified or overlapping) and 806 controls.

Ever tea or coffee drinker was defined as having at least one cup of tea or coffee per week for 6 months or more. During the interview, both cases and controls were asked whether they were tea or coffee drinkers, and then asked about the frequency, duration, age at start, strength of the drink, and types of coffee/tea. In addition, temperature of tea and additives to coffee were also assessed. In the questionnaire, the number of cups consumed per week and the average volume of the cup used to drink the beverages were assessed; total volume consumed per week was calculated using these variables and converted to standardized cups (236.59 ml = 1 cup).

Statistical analysis

The associations between HNC risk and tea and coffee drinking were investigated using unconditional logistic regression. We used SAS version 9.4 to calculate the crude and adjusted odds ratios (ORs) and 95% confidence intervals (CIs) adjusting on: (1) ethnicity (Han and other), age group (as shown in Table 1), sex, education (categorical as shown in Table 1), and center; (2) ethnicity, age group, sex, education, center, cigarette years smoked (categorical: 0,1–19, 20–39, or ≥ 40 years), cigarette average per day (categorical: 0, 1–19, 20–39, or ≥ 40 cigarettes per day), alcohol duration (categorical: 0, 1–19, 20–39, or ≥ 40 years), and alcohol drinks per day (categorical: 0, 1, or ≥ 2 drinks per day). Both former smokers and current smokers were included in the models through the duration and frequency of cigarette smoking. We then stratified by age (18–54 or 55–85), sex, cigarette smoking (ever cigarette smoker or never cigarette smoker), cancer subsite, and center (China mainland or Taiwan). Although betel quid chewing is a strong risk factor for HNC, mainly in the Taiwan center, it was not associated with coffee or tea intake, and thus, we did not adjust on betel quid chewing.
Table 1

Characteristics of head and neck cancer cases and controls

Characteristic

Cases

Mean tea intake

(cups/week)

Mean coffee intake

(cups/week)

Controls

Mean tea intake

(cups/week)

Mean coffee intake

(cups/week)

N

%

N

%

Total

921

 

9.8

1.1

806

 

7.6

1.6

Age

 18–44

146

15.9

7.0

1.6

257

31.8

4.8

2.1

 45–54

273

29.7

9.0

1.5

215

26.7

10.9

1.6

 55–64

297

32.2

12.3

0.9

222

27.7

9.0

1.6

 65–85

205

22.2

9.2

0.7

112

14.0

5.1

0.7

Education

 Illiterate

59

6.4

5.7

0.1

24

3.0

1.7

0.0

 Primary school

228

24.8

9.8

0.4

129

16.0

5.2

0.5

 Junior/middle school

261

28.3

9.8

0.8

150

18.6

8.1

0.7

 Senior/high school

244

26.5

10.0

1.3

170

21.1

7.8

1.2

 College/university and above

129

14.0

11.1

2.9

333

41.3

8.6

2.8

Sex

 Male

726

78.8

11.1

1.1

556

69.0

9.4

1.9

 Female

195

21.2

5.2

1.0

250

31.0

3.7

1.0

Center

 Beijing

54

5.9

31.0

0.4

52

6.5

20.2

0.2

 Jiangsu

63

6.8

8.9

0.0

77

9.6

4.0

0.1

 Shanghai

55

6.0

3.2

0.0

56

7.0

0.0

0.1

 Henan

26

2.8

8.0

0.2

44

5.5

2.2

0.0

 Fujian

60

6.5

7.1

0.5

50

6.2

10.2

0.0

 Liaoning

57

6.2

4.0

0.0

75

9.3

3.2

0.0

 Sichuan

124

13.5

13.6

0.3

51

6.3

6.9

0.2

 Taiwan

482

52.3

8.4

1.9

401

49.8

9.0

3.2

Ethnicity

 Han

556

60.4

10.2

0.7

407

50.5

6.3

0.2

 Other

365

39.6

9.0

1.8

399

49.5

9.0

3.1

Cigarette smoking

 Never smoker

319

34.7

7.8

0.9

462

57.4

4.9

1.4

 Ever smoker

600

65.3

11.0

1.2

343

42.6

11.4

1.9

Subsite

 Oral cavity

424

46.0

9.1

0.8

    

 Oropharynx

106

11.5

11.9

0.9

    

 Hypopharynx

81

8.8

8.1

1.7

    

 Larynx

85

9.2

13.6

0.7

    

 Unspecified or overlapping

225

24.4

9.1

1.8

    

*The unit of mean tea and coffee intake is cups per week

Results

Most participants were between 45 and 65 years old, and 52.3% were recruited from Taiwan (Table 1). The cases were more likely to be male, less educated, and a higher proportion was of Han ethnicity (p values < 0.0001). The mean tea intake was 9.8 cups per week [standard deviation (SD) = 18.7] among cases, and 7.6 cups per week (SD = 17.6) among controls. The mean coffee consumption was 1.1 cups per week (SD = 3.4) among cases and 1.6 cups per week (SD = 3.7) among controls. Differences in coffee and tea consumption between cases and controls were detected (t test, p values < 0.0001). When stratified by age group, sex, and ethnicity, differences in mean tea and coffee consumption were also observed between cases and controls. Both coffee and tea intake were higher among ever smokers than among never smokers.

When adjusted for ethnicity, age group, sex, education, and center, tea intake was not clearly associated with head and neck cancer risk (Table 2). When stratified by cancer subsites, an association with overall tea consumption was detected for laryngeal cancer (OR 2.28, 95% CI 1.10–4.75), and specifically for hot tea (Supplemental Table 1). However, dose–response relations with the frequency and duration of tea intake were not observed for laryngeal cancer risk. Compared to never tea drinkers, drinking other types of tea including black tea, flower tea, or combination of tea types were associated with an increased risk of laryngeal cancer (OR 2.47, 95% CI 1.05, 5.85), but an association was not evident for overall HNC (OR 0.89, 95% CI 0.65, 1.21).
Table 2

Tea intake and the risk of head and neck cancer

 

Case

Control

Adjusted ORa

95% CI

Adjusted ORb

95% CI

Drink tea

 Never drinker

496

493

1.00

 

1.00

 

 Ever drinker

419

313

1.19

(0.95, 1.49)

1.05

(0.82, 1.34)

Tea frequency (cups per week)

 Never drinker

496

493

1.00

 

1.00

 

 1–5

47

58

0.81

(0.51, 1.28)

0.72

(0.43, 1.19)

 6–10

64

49

1.26

(0.52, 1.92)

1.23

(0.77, 1.95)

 11–15

92

75

1.04

(0.72, 1.51)

1.02

(0.68, 1.53)

 16–25

53

38

1.36

(0.84, 2.19)

1.03

(0.61, 1.72)

 25 +

117

78

1.30

(0.91, 1.85)

1.06

(0.72, 1.54)

 p trend

   

0.0785

 

0.6649

Tea years

 Never drinker

496

493

1.00

 

1.00

 

 1–14

120

121

1.05

(0.77, 1.45)

0.98

(0.70, 1.38)

 15–29

145

113

1.22

(0.90, 1.66)

1.14

(0.81, 1.59)

 30–44

102

61

1.10

(0.75, 1.61)

0.82

(0.54, 1.26)

 45 +

43

17

1.98

(1.03, 3.77)

1.72

(0.87, 3.40)

 p trend

   

0.0705

 

0.5762

Tea strength

 Never drinker

496

493

1.00

 

1.00

 

 Strong tea

126

77

1.37

(0.97,1.93)

1.03

(0.70, 1.52)

 Moderate tea

198

156

1.12

(0.85,1.47)

0.99

(0.73, 1.34)

 Light tea

90

80

1.08

(0.74,1.56)

1.14

(0.77, 1.70)

Tea temperature

 Never drinker

496

493

1.00

 

1.00

 

 Cold tea

18

14

1.06

(0.48, 2.35)

0.94

(0.41, 2.18)

 Warm tea

138

85

1.38

(0.99, 1.92)

1.25

(0.87, 1.80)

 Hot tea

229

184

1.16

(0.89, 1.51)

1.01

(0.76, 1.36)

 Very hot tea

20

20

0.72

(0.36, 1.43)

0.60

(0.28, 1.29)

Tea types

 Never drinker

496

493

1.00

 

1.00

 

 Green tea only

80

67

1.05

(0.70, 1.55)

0.85

(0.55, 1.30)

 Oo-long only

133

81

1.48

(1.05, 2.09)

1.42

(0.98, 2.07)

 Other

191

63

1.03

(0.77, 1.36)

0.89

(0.65, 1.21)

aAdjusted for ethnicity, age group, sex, education, and center

bAdjusted for ethnicity, age, sex, education, center, cigarette years smoked, cigarette average per day, alcohol duration, and alcohol drinks per day

Approximately 14.7% of cases and 23.2% of controls were ever coffee drinkers. The adjusted OR showed an inverse relationship between HNC risk and ever coffee drinking (OR 0.71, 95% CI 0.53–0.96; Table 3). However, when adjusted on cigarette smoking and alcohol drinking, the association was no longer observed. Having no coffee additives (no sugar/milk) and caffeinated coffee were significantly inversely associated with HNC risk. When stratified by cancer subsite (Table 4), coffee drinking was observed to be inversely associated with oral cavity cancer risk with dose–response trends for both frequency (p for trend = 0.0149) and the duration (p for trend = 0.0167) of intake. Other types of coffee such as decaffeinated coffee, espresso appeared to be positively associated with cancer of the larynx, but dose–response relations were not observed.
Table 3

Coffee intake and the risk of head and neck cancer

 

Case

Control

Adjusted ORa

95% CI 

Adjusted ORb

95% CI

Drink coffee

 Never drinker

780

617

1.00

 

1.00

 

 Ever drinker

134

187

0.71

(0.53, 0.96)

0.79

(0.57, 1.09)

Coffee frequency per week (cups)

 Never drinker

780

617

1.00

 

1.00

 

 1–5

43

65

0.71

(0.46, 1.12)

0.83

(0.51, 1.35)

 6–10

57

74

0.72

(0.48, 1.09)

0.83

(0.53, 1.29)

 10 +

32

47

0.67

(0.40, 1.13)

0.68

(0.39, 1.19)

 p trend

   

0.0349

 

0.1314

Coffee years

 Never drinker

780

617

1.00

 

1.00

 

 1–14

83

130

0.64

(0.45, 0.90)

0.74

(0.51, 1.07)

 15–29

32

42

0.75

(0.44, 1.27)

0.83

(0.48, 1.46)

 30 +

15

13

1.15

(0.51, 2.61)

1.04

(0.44, 2.46)

 p trend

   

0.1658

 

0.3738

Coffee—age at start

 Never drinker

780

617

1.00

 

1.00

 

 1–24

22

43

0.63

(0.35, 1.14)

0.63

(0.33, 1.20)

 25–49

80

113

0.78

(0.55, 1.12)

0.87

(0.60, 1.28)

 50 +

30

30

0.57

(0.32, 1.02)

0.72

(0.39, 1.34)

 p trend

   

0.0252

 

0.2141

Coffee strength

 Never drinker

780

617

1.00

 

1.00

 

 Strong coffee

25

28

0.85

(0.45, 1.58)

0.81

(0.42, 1.60)

 Moderate coffee

85

129

0.67

(0.47, 0.94)

0.80

(0.55, 1.17)

 Light coffee

15

24

0.62

(0.30, 1.27)

0.64

(0.30, 1.37)

 Not sure

8

5

1.52

(0.44, 5.17)

1.14

(0.30, 4.28)

Coffee additives

 Never drinker

780

617

1.00

 

1.00

 

 Sugar only

7

8

0.75

(0.25, 2.28)

0.86

(0.25, 2.95)

 Milk/cream/soy milk only

19

40

0.44

(0.24, 0.84)

0.53

(0.27, 1.02)

 Both

91

84

1.01

(0.70, 1.45)

1.12

(0.75, 1.65)

 No sugar/milk

13

52

0.30

(0.16, 0.58)

0.32

(0.16, 0.66)

Coffee types

 Never drinker

780

617

1.00

 

1.00

 

 Caffeinated coffee/espresso only

95

162

0.61

(0.44, 0.85)

0.69

(0.49, 0.97)

 Other

37

23

1.39

(0.76, 2.56)

1.42

(0.75, 2.70)

aAdjusted for ethnicity, age group, sex, education, and center

bAdjusted for ethnicity, age, sex, education, center, cigarette years smoked, cigarette average per day, alcohol duration, and alcohol drinks per day

Table 4

Coffee intake and the risk of head and neck cancer subsites

 

Oral cavitya

Oropharynxa

Hypopharynxa

Larynxa

Drink coffee

 Never drinker

1.00

 

1.00

 

1.00

 

1.00

 

 Ever drinker

0.66

(0.42, 1.03)

0.99

(0.48, 2.07)

0.79

(0.34, 1.85)

1.84

(0.71, 4.78)

Coffee frequency (cups per week)

 Never drinker

1.00

 

1.00

 

1.00

 

1.00

 

 1–5

0.88

(0.45, 1.73)

1.49

(0.54, 4.14)

0.39

(0.07, 2.05)

2.39

(0.72, 7.95)

 6–10

0.76

(0.42, 1.38)

0.79

(0.26, 2.43)

0.93

(0.26, 3.31)

1.34

(0.27, 6.72)

 10+

0.25

(0.09, 0.72)

0.79

(0.23, 2.71)

1.06

(0.32, 3.50)

1.56

(0.30, 8.02)

 p trend

 

0.0149

 

0.6824

 

0.9199

 

0.4073

Coffee years

 Never drinker

1.00

 

1.00

 

1.00

 

1.00

 

 1–14

0.77

(0.47, 1.26)

0.83

(0.34, 2.01)

0.62

(0.21, 1.83)

1.47

(0.44, 4.97)

 15–29

0.57

(0.23, 1.43)

1.36

(0.42, 4.42)

0.57

(0.12, 2.75)

1.19

(0.21, 6.84)

 30+

0.06

(0.00, 0.95)

1.28

(0.24, 6.87)

2.98

(0.53, 16.60)

3.61

(0.62, 21.12)

 p trend

 

0.0167

 

0.7405

 

0.8837

 

0.2186

Coffee strength

 Never drinker

1.00

 

1.00

 

1.00

 

1.00

 

 Strong coffee

0.37

(0.12, 1.13)

1.53

(0.45, 5.19)

0.57

(0.11, 3.03)

1.57

(0.30, 8.36)

 Moderate coffee

0.84

(0.51, 1.40)

0.83

(0.33, 2.10)

0.73

(0.25, 2.11)

2.25

(0.74, 6.85)

 Light coffee

0.29

(0.09, 1.00)

1.12

(0.26, 4.76)

1.05

(0.18, 6.12)

---b

(---b, ---b)

 Not sure

0.73

(0.11, 4.65)

---b

(---b, ---b)

2.45

(0.16, 36.66)

10.21

(0.86, 120.96)

Coffee additives

 Never drinker

1.00

 

1.00

 

1.00

 

1.00

 

 Sugar only

1.00

(0.19, 5.35)

1.96

(0.20, 18.87)

1.63

(0.15, 18.16)

1.31

(0.08, 21.95)

 Milk/cream/soy milk only

0.54

(0.21, 1.34)

---b

(---b, ---b)

0.18

(0.01, 2.64)

3.17

(0.73, 13.86)

 Both

0.96

(0.56, 1.65)

1.53

(0.66, 3.55)

1.10

(0.37, 3.25)

2.07

(0.60, 7.15)

 No sugar/milk

0.13

(0.03, 0.57)

0.55

(0.13, 2.32)

0.42

(0.08, 2.21)

0.84

(0.10, 7.18)

Coffee types

 Never drinker

1.00

 

1.00

 

1.00

 

1.00

 

 Caffeinated coffee/espresso only

0.61

(0.37, 0.99)

0.92

(0.41, 2.07)

0.68

(0.27, 1.76)

1.38

(0.47, 4.05)

 Other

0.92

(0.38, 2.27)

1.30

(0.33, 5.07)

0.90

(0.16, 4.97)

4.66

(1.05, 20.63)

aAdjusted for ethnicity, age, sex, education, center, cigarette years smoked, cigarette average per day, alcohol duration, and alcohol drinks per day

bValue not available

Among never cigarette smokers, positive associations were found with having more than 25 cups of tea per week, and with drinking for more than 45 years, but dose–response trends were not identified (Supplemental Table 2). In mainland China, the overall tea consumption was associated with an increased HNC risk (OR 1.51, 95% CI 1.04–2.18), with a dose–response relationship for the frequency of tea. When stratified by age, sex, cigarette smoking, and mainland China vs. Taiwan, coffee drinking was not clearly associated with HNC in terms of frequency or duration (Supplemental Table 3).

Discussion

To the best of our knowledge, this is one of the largest case–control studies on the tea and coffee consumption and the risks for HNC in a Chinese population. Our observation of no association between tea and coffee intake and HNC risk overall was consistent with some of the previous studies [7, 8, 9, 10, 11, 12]. However, coffee consumption was associated with a decreased risk for one of the HNC subsites, oral cavity cancer with a dose–response relation for frequency and duration. Although two recent studies from China reported decreased risks of HNC due to tea intake, we did not observe such a protective effect. For specific HNC subsites, associations between tea consumption and laryngeal cancer risk were observed but without a dose–response relationship.

Tea drinking was common as expected in this population, but coffee drinking was fairly low with an average of 1.1 cups per day among cases and 1.6 cups per day among controls. In fact, coffee drinking was largely observed in the cases and controls from the Taiwan population. In our stratified analysis, we did not observe a clear association for coffee and HNC risk in the Taiwan center, but, perhaps, the sample size was too limited.

We observed differences in the amount of coffee and tea intake among never smokers and ever smokers. This is a concern, since residual confounding due to cigarette smoking may affect our results, although we carefully adjusted for cigarette smoking frequency and duration. In the never smoker subgroup, we did not observe clear associations for tea or coffee. For tea intake, we observed some increased risks, which may be due to residual confounding. For coffee intake, we observed decreased risks for oral cavity cancer, which is the opposite direction of estimates expected if residual confounding was a problem, since cigarette smokers tended to drink more coffee.

A few studies in Asian populations reported that tea was associated with reduced risks of HNC or oral cavity cancer [5, 6] as mentioned earlier, while several studies reported no statistical significant associations between tea consumption and HNC risk [7, 8, 9, 10, 11, 12]. The increased risk due to tea intake for laryngeal cancer was unexpected, although strong dose–response trends with frequency or duration were not observed. Although there were no differences in the increased laryngeal cancer risks observed for tea strength, we did observe an increased risk for hot tea, suggesting that the increase in risk might be due to thermal injury. A recent study in Japan also reported an increased risk of upper aerodigestive tract cancers due to green tea and suggested that thermal injury may be the explanation for the observed association [13].

There are several potential limitations to our study. A large proportion of the cases were unspecified or overlapping cancer subsites. Although this reduced the statistical powder for the subsite investigation, the inclusion of this group enhances the statistical power for the assessment for HNC overall. In a case–control study, the interview after cancer diagnosis might result in recall bias for exposures of interest, because cancer cases and controls might recall differently based on their cancer status. However, the associations between tea and coffee consumption and HNC development have not been well established, and thus, we would expect that recall bias is minimal with regard to coffee or tea intake. In addition, controls were selected from outpatient clinics or were hospitalized for less than a month. This could result in selection bias, since the consumption of tea and coffee might be different compared with the general public. In this study, people with low tea consumption might be misclassified as never drinkers, and the results might be biased toward the null. Furthermore, although we asked about drinking different types of tea, it was difficult to sort out the impact of one single type of tea, because individuals from mainland China may drink a mix of different variety of tea. Finally, the specific preparation of coffee drinking regarding whether the coffee was freeze dried, percolated, or expressed by steam was not assessed in this study. This factor may modify the associations with coffee drinking. Thus, it should be considered in future investigations on coffee drinking.

Strengths of our study include the large sample size and detailed information on potential confounders such as tobacco smoking and alcohol drinking which allowed for adjustments in the statistical models. We collected information on the frequency and duration of coffee drinking, so that dose–response relationship could be examined for HNC risk and by subsite. Collection of additional information on tea temperature, tea strength, and tea types was a strength in investigating these characteristics of tea intake.

In conclusion, our study showed an inverse association between coffee intake and the risk of oral cavity cancers. Larger studies may be necessary to detect any associations with tea intake.

Notes

Acknowledgements

This investigation was supported by the Department of Family and Preventive Medicine, and the University of Utah Study Design and Biostatistics Center, with funding in part from the National Cancer Institute through Cancer Center Support P30 CA042014 awarded to Huntsman Cancer Institute, and the National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant 8UL1TR000105 (formerly UL1RR025764).

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Supplementary material

41548_2018_9_MOESM1_ESM.docx (47 kb)
Supplementary material 1 (DOCX 46 kb)
41548_2018_9_MOESM2_ESM.xlsx (36 kb)
Supplementary material 2 (XLSX 35 kb)

References

  1. 1.
    Hashibe M, Galeone C, Buys SS, Gren L, Boffetta P, Zhang ZF, La Vecchia C (2015) Coffee, tea, caffeine intake, and the risk of cancer in the PLCO cohort. Br J Cancer 113(5):809–816.  https://doi.org/10.1038/bjc.2015.276 Epub 2015 Aug 20. PubMed PMID: 26291054; PubMed Central PMCID: PMC4559834 CrossRefPubMedPubMedCentralGoogle Scholar
  2. 2.
    Ferlay J, Soerjomataram I, Dikshit R, Eser S, Mathers C, Rebelo M et al (2015) Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer 136(5):E359–E386.  https://doi.org/10.1002/ijc.29210 Epub 2014 Oct 9 PubMed PMID: 25220842 CrossRefPubMedGoogle Scholar
  3. 3.
    Hwang TZ, Hsiao JR, Tsai CR, Chang JS (2015) Incidence trends of human papillomavirus-related head and neck cancer in Taiwan, 1995–2009. Int J Cancer 137(2):395–408.  https://doi.org/10.1002/ijc.29330 Epub 2014 Nov 24 PubMed PMID: 25395239 CrossRefPubMedGoogle Scholar
  4. 4.
    IARC Working Group on the Evaluation of Carcinogenic Risks to Humans (1991) Coffee, tea, mate, methylxanthines and methylglyoxal. Lyon, 27 February to 6 March 1990. IARC Monogr Eval Carcinog Risks Hum 51:1–513Google Scholar
  5. 5.
    Huang CC, Lee WT, Tsai ST, Ou CY, Lo HI, Wong TY et al (2014) Tea consumption and risk of head and neck cancer. PLoS ONE 9(5):e96507.  https://doi.org/10.1371/journal.pone.0096507.eCollection2014 PubMed PMID: 24796481; PubMed Central PMCID: PMC4010481 CrossRefPubMedPubMedCentralGoogle Scholar
  6. 6.
    Fu JY, Gao J, Zhang ZY, Zheng JW, Luo JF, Zhong LP, Xiang YB (2013) Tea consumption and the risk of oral cancer incidence: a case-control study from China. Oral Oncol 49(9):918–922.  https://doi.org/10.1016/j.oraloncology.2013.05.002 Epub 2013 May 31 PubMed PMID: 23731795 CrossRefPubMedGoogle Scholar
  7. 7.
    Bundgaard T, Wildt J, Frydenberg M, Elbrønd O, Nielsen JE (1995) Case-control study of squamous cell cancer of the oral cavity in Denmark. Cancer Causes Control 6(1):57–67 PubMed PMID: 7718736 CrossRefPubMedGoogle Scholar
  8. 8.
    Franco EL, Kowalski LP, Oliveira BV, Curado MP, Pereira RN, Silva ME et al (1989) Risk factors for oral cancer in Brazil: a case-control study. Int J Cancer 43(6):992–1000 PubMed PMID: 2732011 CrossRefPubMedGoogle Scholar
  9. 9.
    Galeone C, Tavani A, Pelucchi C, Turati F, Winn DM, Levi F et al (2010) Coffee and tea intake and risk of head and neck cancer: pooled analysis in the international head and neck cancer epidemiology consortium. Cancer Epidemiol Biomark Prev 19(7):1723–1736.  https://doi.org/10.1158/1055-9965.epi-10-0191 Epub 2010 Jun 22. PubMed PMID: 20570908; PubMed Central PMCID: PMC3047460 CrossRefGoogle Scholar
  10. 10.
    Ouyang Z, Wang Z, Jin J (2014) Association between tea and coffee consumption and risk of laryngeal cancer: a meta-analysis. Int J Clin Exp Med 7(12):5192–5200 eCollection 2014. PubMed PMID: 25664021; PubMed Central PMCID: PMC4307468 PubMedPubMedCentralGoogle Scholar
  11. 11.
    Takezaki T, Hirose K, Inoue M, Hamajima N, Kuroishi T, Nakamura S et al (1996) Tobacco, alcohol and dietary factors associated with the risk of oral cancer among Japanese. Jpn J Cancer Res 87(6):555–562 PubMed PMID: 8766517 CrossRefPubMedPubMedCentralGoogle Scholar
  12. 12.
    Tavani A, Bertuzzi M, Talamini R, Gallus S, Parpinel M, Franceschi S, La Vecchia C (2003) Coffee and tea intake and risk of oral, pharyngeal and esophageal cancer. Oral Oncol 39(7):695–700 PubMed PMID: 12907209 CrossRefPubMedGoogle Scholar
  13. 13.
    Shinoda M, Tajima K, Tanaka H (2014) Coffee and green tea consumption is associated with upper aerodigestive tract cancer in Japan. Int J Cancer 135(2):391–400.  https://doi.org/10.1002/ijc.28653 Epub 2013 Dec 17 PubMed PMID: 24310779 CrossRefPubMedGoogle Scholar
  14. 14.
    Ide R, Fujino Y, Hoshiyama Y, Mizoue T, Kubo T, Pham TM, JACC Study Group et al (2007) A prospective study of green tea consumption and oral cancer incidence in Japan. Ann Epidemiol 17(10):821–826 Epub 2007 Jul 2. PubMed PMID: 17606381 CrossRefPubMedGoogle Scholar
  15. 15.
    Naganuma T, Kuriyama S, Kakizaki M, Sone T, Nakaya N, Ohmori-Matsuda K et al (2008) Coffee consumption and the risk of oral, pharyngeal, and esophageal cancers in Japan: the Miyagi cohort study. Am J Epidemiol 168(12):1425–1432.  https://doi.org/10.1093/aje/kwn282 Epub 2008 Oct 29 PubMed PMID: 18974083 CrossRefPubMedGoogle Scholar

Copyright information

© Springer International Publishing AG, part of Springer Nature 2018

Authors and Affiliations

  • Shuang Li
    • 1
  • Yuan-Chin Amy Lee
    • 1
  • Qian Li
    • 2
  • Chien-Jen Chen
    • 3
    • 4
  • Wan-Lun Hsu
    • 3
  • Pen-Jen Lou
    • 5
  • Cairong Zhu
    • 6
  • Jian Pan
    • 7
  • Hongbing Shen
    • 8
  • Hongxia Ma
    • 8
  • Lin Cai
    • 9
  • Baochang He
    • 9
  • Yu Wang
    • 10
    • 11
  • Xiaoyan Zhou
    • 11
    • 12
  • Qinghai Ji
    • 10
    • 11
  • Baosen Zhou
    • 13
  • Wei Wu
    • 13
  • Jie Ma
    • 14
  • Paolo Boffetta
    • 15
  • Zuo-Feng Zhang
    • 16
  • Min Dai
    • 17
  • Mia Hashibe
    • 1
  1. 1.Division of Public Health, Department of Family and Preventive Medicine, and Huntsman Cancer InstituteUniversity of Utah School of MedicineSalt Lake CityUSA
  2. 2.Departments of Preventive MedicineIcahn School of Medicine at Mount SinaiNew YorkUSA
  3. 3.Genomics Research Center, Academia SinicaTaipeiTaiwan
  4. 4.Graduate Institute of Epidemiology and Preventive MedicineNational Taiwan UniversityTaipeiTaiwan
  5. 5.Department of OtolaryngologyNational Taiwan University HospitalTaipeiTaiwan
  6. 6.Department of Epidemiology and Biostatistics, West China School of Public HealthSichuan UniversityChengduChina
  7. 7.Department of Oral Surgery, West China Hospital of StomatologySichuan UniversitySichuanChina
  8. 8.Department of Epidemiology and Biostatistics, Jiangsu Key Lab of Cancer Biomarkers, Prevention and Treatment, Collaborative Innovation Center for Cancer Personalized Medicine, School of Public HealthNanjing Medical UniversityNanjingChina
  9. 9.Department of Epidemiology and Biostatistics, School of Public HealthFujian Medical UniversityFujianChina
  10. 10.Department of Head and Neck SurgeryFudan University Shanghai Cancer CenterShanghaiChina
  11. 11.Department of Oncology, Shanghai Medical CollegeFudan UniversityShanghaiChina
  12. 12.Department of PathologyFudan University Shanghai Cancer CenterShanghaiChina
  13. 13.Department of Epidemiology, School of Public HealthChina Medical UniversityLiaoningChina
  14. 14.Department of Head and Neck OncologyHenan Cancer HospitalHenanChina
  15. 15.Tisch Cancer Institute, Icahn School of Medicine Mount SinaiNew York CityUSA
  16. 16.Department of Epidemiology and Center for Environmental GenomicsUCLA Fielding School of Public HealthLos AngelesUSA
  17. 17.National Office of Cancer Prevention and Control Cancer Institute and Hospital, and Chinese Academy of Medical SciencesBeijingChina

Personalised recommendations