The effect of Greek herbal tea consumption on thyroid cancer: a case-control study

Abstract

Background: Although in the last decade several studies have addressed the protective role of black and green tea on several diseases, including cancer, there are only few and controversial studies on the effect of tea on benign and malignant thyroid diseases. Methods: An age and gender group matched case-control study conducted in Athens, Greece, was designed. 113 Greek patients with histologically confirmed thyroid cancer and 286 patients with benign thyroid diseases along with 138 healthy controls were interviewed with a pre-structured questionnaire in person by trained interviewers. Results: An inverse association between chamomile tea consumption and benign/malignant thyroid diseases was found ( P < 0.001). The odds of chamomile tea consumption, two to six times a week, after controlling for age, gender and BMI, were 0.30 (95% CI: 0.10–0.89) and 0.26 (95% CI: 0.12–0.5) for developing thyroid cancer and benign thyroid diseases, respectively when compared with not consumption. The duration of consumption was also inversely associated with the diseases. Thirty years of consumption significantly reduced the risk of thyroid cancer and benign thyroid diseases development by almost 80%. Similar, although weaker protective association, was found for sage and mountain tea. Adjustment for smoking, alcohol and coffee consumption did not alter the results. Conclusions: Our findings suggest for the first time that drinking herbal teas, especially chamomile, protects from thyroid cancer as well as other benign thyroid diseases.

Introduction

Tea is one of the most commonly consumed beverages in the world, second only to water. ¹

Most studies have examined the effect of tea on cancer looking at black and green tea separately, probably because the 80% of the tea consumed in Europe and the USA is black, whereas green tea is the most widely consumed beverage in China, Japan and other Asian countries. ²

In Greece, herbal tea has been widely consumed for centuries. In particular, chamomile (genus Marticaria Chamomila , family Composita), sage (family Salvia fruticosa L.) and mountain tea (genus Sideritis , family Lamiaceae) grow extensively in the Greek countryside and mountains. They are locally collected and home-prepared for direct consumption without any special manufacturing process, like the black tea.

In the last decade several studies described the beneficial properties derived by tea consumption: antioxidative, ³ anti-inflammatory, ⁴ antibacterial, ⁵ antiarthritic, ⁶ antiviral, ⁷ antiangiogenic, ⁸ neuroprotective. ⁹

Furthermore, it is now well known that the active components of tea can also play a role in lowering cholesterol levels ¹⁰ and in preventing cancer ¹¹ and cardiovascular diseases. ¹²

In the endocrine field, tea is appreciated as a precious ally on reaching metabolic control and preventing diabetes, ^13,14 although in high doses it can exert anti-thyroid and goitrogenic effects. ¹⁵

Additionally, though there are plenty of data describing a protective effect of black and green tea on colon rectum, breast, lung, prostate, stomach and ovarian carcinoma, ^16–21 there are only few findings regarding endocrine tumors.

This study examines the effect of the consumption of three different types of herbal tea on thyroid cancer risk [malignant (MC) and benign cases (BC)] from a population-based case-control study in Athens, Greece.

Methods

Participants

Between June 1990 and 1993, 113 Greek patients with histologically confirmed thyroid cancer were admitted to the departments of endocrinology or endocrine surgery in two major hospitals in the Athens metropolitan area. The observed proportion for the histological types of cancer follows the international literature: Papillary 78% of cases, Follicular 8%, Medullary 10% and Others 4%. 138 controls, group matched with the thyroid cancer cases by age (±5 years; 65+ as one group), gender and health unit, were selected from the same two hospitals. The controls were either healthy subjects or patients of other departments with a variety of diseases; persons with diseases related to thyroid cancer, other endocrine abnormalities or diet-related conditions were excluded. The above data have been used to analyze the effects of dietary patterns to thyroid cancer in a previous study. ²² In the same period, and using the same matching criteria, 286 patients with benign thyroid diseases were enrolled for this study.

Patients and controls were interviewed in person, after signing an informed consent, by trained interviewers using a pre-structured questionnaire covering a wide range of demographic and socio-economic factors, medical history and lifestyle characteristics, such as smoking, alcohol, coffee and tea consumption as well as dietary practices. The interviewers were blind with regard to the disease status and unaware of the specific hypothesis to be tested. Information on tea consumption included type of tea consumed, frequency and duration of tea drinking. Tea consumption was categorized in four categories ‘None’, ‘at most once a week’, ‘two to six times a week’ and ‘at least once a day’; the duration of tea drinking was categorized as ‘0 years’, ‘≤15 years’, ‘15–30 years’ and ‘>30 years’; alcohol and coffee consumption in ‘No’ or ‘Yes’. For the various types of tea, we had no information on the frequency or the duration of consumption for 0.7 to 7% of the controls, 0.8 to 6% of the MC and 0.3 to 7% of the BC. These subjects were excluded from our analyses.

Statistical analysis

Association between demographic and lifestyle characteristics for MC, BC and healthy controls were examined by using analysis of variance, and where appropriate the non-parametric Kruskal-Wallis test, and chi-square tests. Univariate and multivariate odds ratios, along with their 95% CIs were derived by using logistic regression before and after adjusting for demographic and lifestyle characteristics. Finally to explore trends with increased frequency and number of years of tea drinking, non-parametric tests for trend were used. All analyses were carried out in Stata (STATA corp., College Station, TX).

Results

Table 1 shows the association between demographic and lifestyle characteristics for malignant and benign cases and healthy controls. Cigarette smoking is defined as the number of cigarettes smoked per day multiplied by the years the person smoked.

Table 1

Demographics and risk factors for cases and controls

	Controls (N = 138)	BC ( N = 286)	MC ( N = 113)	Statistic
Age				P = 0.003
Mean (SD)	38.2 (13.1)	42.1 (14.9)	44.5 (15.7)
Gender				P = 0.38
Female	95 (69%)	218 (76%)	81 (72%)
Male	43 (31%)	68 (24%)	32 (28%)
BMI				P = 0.07
Mean (SD)	24.3 (3.7)	24.5 (3.6)	25.4 (4.5)
Smoking(cigarette-years)				P = 0.004
Mean (SD)	237.5 a (325.8)	174.5 a (334.4)	219.3 (385.7)
Alcohol				P < 0.001
Y	106 (77%)	162 (57%)	58 (51%)
N	32 (23%)	124 (43%)	55 (49%)
Coffee				P = 0.35
Y	127 (92%)	259 (91%)	98 (87%)
N	11 (8%)	27 (9%)	15(13%)

a: statistically significant at 1% level of significance.

A significant difference in the distribution of alcohol drinkers among cases and controls was noticed [57% of subjects in BC and 51% in MC group, respectively consume alcohol vs. 77% of controls (C); P < 0.001]; along with a small difference in age (42.1 ± 14.9 years for BC vs. 44.5 ± 15.7 MC vs. 38.2 ± 13.1 C, P = 0.003) and in cigarette smoking (174.5 ± 334.4 number of cigarettes for BC vs. 219.3 ± 385.7 MC vs. 237.5 ± 325.8 C, P = 0.004).

There were no differences in body mass index (BMI), gender and coffee consumption. The number of obese patients in the three groups is equally distributed (15/113 among MC, 25/286 BC and 14/138 C, P = 0.398).

The odds of thyroid disease cases for each type of tea consumption and years of drinking are shown in table 2 . There was a significant reduction in the odds of malignant, benign and any type of thyroid disease with an increased frequency of chamomile consumption ( table 2 , Panel A), when adjusting for age, gender and BMI ( P for trend <0.01). Those drinking chamomile from two to six times a week, were significantly less likely of developing cancer of the thyroid [adjusted odds ratio (AOR) = 0.30, 95% CI: 0.10–0.89], having a benign thyroid disease (AOR = 0.26, 95% CI: 0.12–0.58) and developing any type of thyroid disease (AOR = 0.27, 95% CI: 0.13–0.57) as compared with non-drinkers. Further adjustment for smoking, alcohol and coffee consumption do not change significantly the above results, although a non-significant association with an increased frequency of consumption was found for MC ( P for trend = 0.13). A similar inverse association was noticed with an increased number of years of chamomile consumption ( P for trend = 0.01 for malignant and <0.001 benign and any type of thyroid disease). Those consuming chamomile for over 30 years had significantly reduced odds of 0.20 (95% CI: 0.06–0.67) for thyroid cancer risk, 0.19 (95% CI: 0.07–0.51) for benign and 0.20 (95% CI: 0.08–0.49) for any type of thyroid disease than those who do not drink at all. The above results are not much altered for BC and any type of thyroid disease, when smoking, alcohol and coffee consumption were taken into account, while little association was found for MC ( P for trend 0.07).

Table 2

Odds of thyroid case for Chamomile tea, Sage tea and Mountain tea

	Controls (N)		MC ( N )				BC ( N )				All Cases combined ( N )
		N	OR (95% CI)	AOR 1 (95% CI)	AOR 2 (95% CI)	N	OR (95% CI)	AOR 1 (95% CI)	AOR 2 (95% CI)	N	OR (95% CI)	AOR 1 (95% CI)	AOR 2 (95% CI)
Panel A
Chamomile tea
None	85	80	1.0	1.0	1.0	237	1.0	1.0	1.0	317	1.0	1.0	1.0
At most once a week	32	19	0.63 (0.33–1.20)	0.52 (0.26–1.02)	0.58 (0.28–1.19)	30	0.33 (0.19–0.58)	0.28 (0.16–0.52)	0.31 (0.17–0.57)	49	0.41 (0.24–0.68)	0.35 (0.20–0.59)	0.38 (0.22–0.67)
Two to six times a week	16	5	0.33 (0.11–0.94)	0.30 (0.10–0.89)	0.31 (0.09–0.95)	14	0.31 (0.14–0.67)	0.26 (0.12–0.58)	0.27 (0.12–0.61)	19	0.32 (0.15–0.64)	0.27 (0.13–0.57)	0.28 (0.13–0.61)
At least once a day	4	8	2.12 (0.61–7.33)	1.36 (0.37–4.95)	1.64 (0.42–6.36)	2	0.18 (0.03–0.99)	0.12 (0.02–0.70)	0.12 (0.02–0.75)	10	0.67 (0.20–2.19)	0.45 (0.13–1.54)	0.53 (0.15–1.88)
P for trend			0.17	<0.001	0.13		<0.001	<0.001	<0.001		<0.001	<0.001	<0.001
Years of consumption
0	85	80	1.0	1.0	1.0	237	1.0	1.0	1.0	317	1.0	1.0	1.0
≤15	15	8	0.56 (0.22–1.41)	0.59 (0.23–1.53)	0.63 (0.23–1.69)	11	0.26 (0.12–0.59)	0.25 (0.11–0.58)	0.24 (0.10–0.58)	19	0.34 (0.16–0.69)	0.33 (0.16–0.70)	0.33 (0.15–0.72)
15–30	18	12	0.71 (0.32–1.56)	0.56 (0.24–1.26)	0.54 (0.23–1.28)	9	0.18 (0.07–0.41)	0.16 (0.06–0.37)	0.15 (0.06–0.37)	21	0.31 (0.16–0.61)	0.26 (0.13–0.51)	0.27 (0.13–0.55)
>30	12	6	0.53 (0.19–1.48)	0.20 (0.06–0.67)	0.23 (0.06–0.78)	14	0.42 (0.18–0.94)	0.19 (0.07–0.51)	0.25 (0.09–0.67)	20	0.45 (0.21–0.95)	0.20 (0.08–0.49)	0.25 (0.10–0.62)
P for trend			0.10	0.01	0.07		<0.001	<0.001	<0.001		<0.001	<0.001	<0.001
Panel B
Sage tea
None	109	94	1.0	1.0	1.0	258	1.0	1.0	1.0	352	1.0	1.0	1.0
At most once a week	18	9	0.57 (0.24–1.35)	0.54 (0.22–1.31)	0.58 (0.23–1.47)	15	0.35 (0.17–0.72)	0.32 (0.15–0.68)	0.33 (0.55–0.73)	24	0.41 (0.21–0.78)	0.37 (0.19–0.74)	0.39 (0.19–0.79)
Two to six times a week	9	5	0.64 (0.21–1.98)	0.54 (0.16–1.74)	0.44 (0.13–1.51)	7	0.33 (0.12–0.90)	0.26 (0.09–0.75)	0.29 (0.09–0.85)	12	0.41 (0.17–1.0)	0.35 (0.14–0.88)	0.35 (0.13–0.92)
At least once a day	1	4	4.6 (0.51–42.2)	3.1 (0.32–30.0)	3.3 (0.31–34.9)	3	1.26 (0.13–12.3)	0.82 (0.08–8.41)	0.94 (0.09–9.56)	7	2.16 (0.26–17.8)	1.54 (0.18–13.0)	1.76 (0.20–15.5)
P for trend			0.46	0.10	0.37		<0.001	0.01	0.05		0.01	0.01	0.08
Years of consumption
0	109	94	1.0	1.0	1.0	258	1.0	1.0	1.0	352	1.0	1.0	1.0
≤15	6	6	1.16 (0.36–3.71)	1.07 (0.31–3.68)	0.92 (0.25–3.42)	8	0.56 (0.19–1.66)	0.58 (0.19–1.75)	0.58 (0.19–1.80)	14	0.72 (0.27–1.92)	0.73 (0.27–1.98)	0.70 (0.25–1.97)
15–30	12	9	0.87 (0.35–2.15)	0.77 (0.30–1.96)	0.72 (0.27–1.92)	3	0.10 (0.03–0.38)	0.10 (0.02–0.38)	0.11 (0.03–0.40)	12	0.31 (0.13–0.71)	0.28 (0.12–0.66)	0.29 (0.12–0.69)
>30	5	2	0.46 (0.08–2.44)	0.20 (0.03–1.16)	0.20 (0.03–1.21)	7	0.59 (0.18–1.90)	0.30 (0.08–1.11)	0.39 (0.10–1.54)	9	0.55 (0.18–1.69)	0.28 (0.08–0.96)	0.34 (0.09–1.21)
P for trend			0.60	0.13	0.41		<0.001	<0.001	0.06		0.01	0.01	0.08
Panel C
Mountain tea
None	72	68	1.0	1.0	1.0	229	1.0	1.0	1.0	297	1.0	1.0	1.0
At most once a week	33	20	0.64 (0.34–1.22)	0.60 (0.31–1.17)	0.63 (0.31–1.27)	19	0.27 (0.15–0.48)	0.26 (0.14–0.47)	0.29 (0.16–0.53)	49	0.36 (0.21–0.60)	0.33 (0.19–0.56)	0.36 (0.20–0.62)
Two to six times a week	17	11	0.68 (0.29–1.56)	0.66 (0.27–1.59)	0.77 (0.31–1.93)	19	0.35 (0.17–0.71)	0.33 (0.16–0.69)	0.36 (0.17–0.78)	30	0.43 (0.22–0.82)	0.39 (0.20–0.78)	0.43 (0.21–0.87)
At least once a day	14	13	0.98 (0.43–2.24)	0.76 (0.32–1.82)	1.02 (0.42–2.50)	5	0.11 (0.04–0.32)	0.09 (0.03–0.29)	0.10 (0.03–0.30)	18	0.31 (0.15–0.65)	0.25 (0.11–0.54)	0.28 (0.13–0.63)
P for trend			0.35	0.28	0.84		<0.001	<0.001	<0.001		<0.001	<0.001	0.02
Years of consumption
0	72	68	1.0	1.0	1.0	229	1.0	1.0	1.0	297	1.0	1.0	1.0
≤15	24	10	0.44 (0.19–0.99)	0.56 (0.24–1.30)	0.50 (0.21–1.20)	15	0.19 (0.09–0.39)	0.23 (0.11–0.47)	0.22 (0.10–0.46)	25	0.25 (0.13–0.46)	0.30 (0.16–0.58)	0.28 (0.15–0.55)
15–30	23	17	0.78 (0.38–1.59)	0.70 (0.34–1.46)	0.72 (0.33–1.55)	11	0.15 (0.07–0.32)	0.14 (0.06–0.31)	0.14 (0.06–0.31)	28	0.29 (0.16–0.54)	0.25 (0.13–0.48)	0.27 (0.14–0.51)
>30	9	12	1.41 (0.55–3.56)	0.80 (0.28–2.25)	1.06 (0.36–3.13)	13	0.45 (0.18–1.10)	0.28 (0.10–0.76)	0.43 (0.15–1.22)	25	0.67 (0.30–1.50)	0.40 (0.16–0.98)	0.55 (0.21–1.41)
P for trend			0.58	0.45	0.89		<0.001	<0.001	<0.001		<0.001	<0.001	0.01

¹ Adjusted for age, gender and BMI; ² Adjusted for age, gender, BMI, smoking, alcohol and coffee consumption.

For sage tea drinkers ( table 2 , Panel B), a significant inverse association with an increased frequency and duration of consumption was noticed for benign and any type of thyroid disease ( P for trend = 0.01), but not for MC ( P for trend = 0.10 for frequency and 0.13 for duration of consumption). Those consuming sage tea from two to six times a week, were 46% less likely to develop cancer of the thyroid (AOR = 0.54, 95% CI: 0.16–1.74) and had a significantly reduced odds of 0.26 (95% CI: 0.09–0.75) of benign disease and 0.35 (95% CI: 0.14–0.88) of developing any type of thyroid disease. Also, sage tea consumption for over 30 years will confer a reduced odds of 0.20 (95% CI: 0.03–1.16) for thyroid cancer risk, 0.30 (95% CI: 0.08–1.11) for benign and 0.28 (95% CI: 0.08–0.96) for any type of thyroid disease. Further adjustment for smoking, alcohol and coffee consumption do not alter much the above odds, although little association was found with increased frequency and years of sage tea consumption for benign ( P for trend = 0.05 for frequency and 0.06 for years of consumption) and any type of thyroid disease ( P for trend = 0.08).

A similar inverse association was noticed with an increased frequency and duration of mountain tea drinking ( table 2 , Panel C) for benign ( P for trend < 0.001) and any type of thyroid disease ( P for trend < 0.001). Further adjustment for smoking, alcohol and coffee consumption did not significantly change the above pattern. Those consuming mountain tea from two to six times a week are 34% less likely to develop cancer of the thyroid (AOR = 0.66, 95% CI: 0.27–1.59) and significantly less likely for benign (AOR = 0.33, 95% CI: 0.16–0.69) and any type of thyroid disease (AOR = 0.39, 95% CI: 0.20–0.78). Further adjustment for smoking, alcohol and coffee consumption did not significantly change the above odds, while little association was found for over 30 years of mountain tea consumption (AOR = 1.06, 95% CI: 0.36–3.13 for malignant; 0.43, 95% CI: 0.15–1.22 for benign and 0.55, 95% CI: 0.21–1.41 for any type of thyroid disease).

Discussion

Although many studies have addressed the protective role of black and green tea on several diseases, ²³ including cancer, ¹¹ there are only few and controversial studies on the effect of tea on benign and malignant thyroid diseases. The first study is a female population-based case control study of thyroid cancer in Los Angeles County that found a significant reduction of cancer risk with tea consumption. ²⁴ A pooled, though, analysis of 14 case control studies conducted in the USA, Europe and Asia on thyroid cancer showed no association of thyroid cancer with tea consumption, even if the reported studies displayed a high degree of heterogeneity in their findings. ²⁵

More recently, Michikawa et al.²⁶ reported that green tea consumption is inversely associated with postmenopausal thyroid cancer risk but positively associated with premenopausal risk. This observation may suggest an etiological difference between pre- and postmenopausal thyroid cancer, reconsidering the etiological role of estrogen in thyroid tumorigenesis in light of the studies that described the estrogenic activity of green tea catechins. ²⁷

Herbal teas are widely consumed in Greece and their role in influencing thyroid homeostasis is widely underestimated, especially taking into account a recent study that revealed the spreading use of complementary and alternative medicine among patients with thyroid cancer. ²⁸ This is the first time to our knowledge that the effects of different herbal teas on thyroid diseases, including thyroid cancer, is reported.

The generic mechanisms behind herbal tea consumption that protect from benign and malignant diseases can be attributed to anti-oxidant, anti-inflammatory and anti-cancer activity, mainly due to the polyphenols, flavonoids and catechins that these teas contain. ^23,29

Given its antioxidant potential, tea may have a preventive effect on thyroid carcinogenesis. Indeed, it is well known that the thyroid gland requires hydrogen peroxide, a reactive oxygen species, for thyroid hormone biosynthesis, and recent studies demonstrate that the development of thyroid cancer might be partly explained by abnormalities in the hydrogen peroxide generating systems. ³⁰

Furthermore, some basic studies have been published in the last years about the properties of the Epigallocatechin-3-gallate (EGCG), a major catechin in green tea, and its role in human thyroid carcinoma cell lines. It seems that EGCG has an inhibitory role on thyroid cancer cell proliferation and motility with concomitant loss of epithelial-to-mesenchymal cell transition markers ³¹ and that it induces growth inhibition and apoptosis of human anaplastic thyroid carcinoma cells. ³² Moreover, it was recently described that theaflavins, the bioactive components of black tea, can have a direct effect on the successful induction of apoptosis in human medullary thyroid carcinoma cell line. ³³

Among the three types of herbal tea we studied, chamomile was found to be the most protective in reducing thyroid cancer risk, as well as any type of thyroid diseases, when consumed from two to six times a week for over a period of 30 years. Sage tea, another herbal tea was also found be protective against any type of thyroid disease, though it was not associated with a significant reduction of the risk of thyroid cancer. When it is consumed for at least 30 years; however, sage tea reduces the odds of developing cancer of the thyroid by 80%. The fact that these odds did not reach statistical significance, although they are of the same magnitude as those of chamomile tea, can be explained by the small number of people who consumed sage tea for such a long period of time.

Greek mountain tea was also found to have a protective effect on the development of thyroid cancer or any type of thyroid disease, but its effect was not as strong as that of chamomile or sage tea. Further adjustment for lifestyle characteristics such as smoking, alcohol and coffee consumption did not seem to substantially change the above results.

Even if a significant age difference ( P = 0.003) is present among patients and controls, it is very narrow and linked to the matching criteria (±5 years; 65+ as one group). In any case we controlled for age in the estimation of the AORs. BMI and the prevalence of obese patients did not differ statistically significantly among the three groups ( P = 0.07 and 0.398, respectively).

Among the advantages of this study, we can consider the relative large number of MC, as well as the structure of the questionnaire, which apart from the demographic and life-style characteristics also included detailed information on the drinking habits of tea, i.e. frequency and duration of consumption.

However, as in all case-control studies, our paper in influenced by some factors that deserve to be considered. First, tea in Greece is not consumed as widely as in Asia. The relatively small number of tea consumers in this study (especially on more regular basis) leads to somehow unstable results in categories with few cases. Second, other factors such as the Mediterranean diet may or may not have an added protective role as well as explained by our group in a previous study. ²² Furthermore, people consuming tea may have other habits beneficial for thyroid disease which may not be captured in our data. Third, although patients with diseases related to thyroid cancer, other endocrine abnormalities or diet-related conditions were excluded from the control group, the information about the medical anamnesis or family history for thyroid diseases is lacking. Last, the data used for this analysis are fairly old (gathered in the period 1990–1993). There was no specific reason why this analysis was not conducted earlier, apart that, recent studies showed for the first time an in vitro activity of the tea components in preventing and contrasting cancer. ^31–33 These articles have led us to re-analyze our data and supported us in publishing our findings. Of course dietary habits, disease prevalence and general living conditions of the population may have changed considerably, but we think that the variables considered in the analysis are still valid and not impaired by the years passed.

After decades, the number of studies focused on the preventive role of tea in thyroid disorders is finally increasing. Our results demonstrate that drinking herbal teas, especially chamomile, could protect from thyroid cancer as well as other benign thyroid diseases. Further clinical and basic studies are needed to confirm the epidemiological findings and shed light on the protective effects derived by tea consumption.

Funding

This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.

Conflicts of interest : None declared.

Key points

• An age and gender group matched case-control study was designed.

• Inverse association between chamomile tea consumption/thyroid diseases was found.

• The duration of consumption was also inversely associated with the diseases.

• Similar although weaker protective association was found for sage and mountain tea.