Journal of Clinical Nutrition and Food Science (JCNFS)
ISSN: 2641-2292 | Vol. 4, Issue 1, 2021, pp. 16-31

Coffee Consumption and Risk of Pancreatic Cancer: An Updated Meta-analysis of Cohort Studies

Mingzhu Zhou1, Na Zhang1, Jianfen Zhang1, Man Zhang1, Hairong He1, Xinyu Yan1, Yibin Li1, Wen Guo1, Yue Zhang1, Guansheng Ma1

1Department of Nutrition and Food Hygiene, School of Public Health, Peking University, 38 Xue Yuan Road, Haidian District, Beijing 100191, China; Laboratory of Toxicological Research and Risk Assessment for Food Safety, Peking University, 38 Xue Yuan Road, Haidian District, Beijing 100191, China
DOI: | Published: 2021-03-25 | Article page and PDF

Abstract

Objective: To review the association between coffee consumption and risk of pancreatic cancer in a meta-analysis of cohort

studies and evaluate their dose–response relationship.Methods: We searched PubMed and EMBASE for cohort studies of coffee consumption and pancreatic cancer risk up to 5 January 2020. A fixed-effects model was used for pooling overall relative risk (RR) and corresponding 95% confidence interval (CI) for the highest versus lowest coffee consumption category, while the analysis was also stratified by sex, region and coffee type. Egger’s and Begg’s test was adopted to exam the publication bias. The dose-response analysis was performed as well. Results: A total of 21 high-quality studies from 18 publications were selected, including 2,859,065 participants and 11,023 pancreatic cancer cases. The pooled relative risk of 21 studies for highest versus lowest category of coffee consumption was 0.95 (95% CI 0.87-1.02, I2 = 43.4%, P for heterogeneity = 0.018). Subgroup analysis showed that coffee consumption was associated with a reduced pancreatic cancer risk in males (RR 0.72, 95%CI 0.53-0.91) but not in females (RR 0.89, 95% CI0.73-1.05). For dose-response analysis, the overall RR for increment of one cup/day of coffee consumption was 1.00 (95%CI, 0.99-1.02) of 15 studies, without statistically significant.Conclusions: This meta-analysis indicates non-significant association between coffee consumption and pancreatic cancer risk. No evidence of publication bias was found.

Keywords: Coffee; Pancreatic cancer; Meta-analysis

Introduction

Pancreatic cancer is a highly malignant tumor of the digestive system. In recent years, the prognosis of patients is improving but is still poor with an overall five-year survival rate of 8% [1]. According to the most recent data from the American Cancer Society, pancreatic cancer is the fourth leading cause of cancer deaths, and estimates that 23,800 and 21,950 deaths (in men and women, respectively) due to pancreatic cancer are projected to occur in the United States in 2019 [2]. In China, though pancreas cancer is not among the top five causes of cancer-related deaths, its mortality rate has increased by 9% in the past 10 years [3]. Therefore, pancreatic cancer has become a major public health problem threatening human life and health and has attracted more and more attention [1].

Coffee is one of the most commonly consumed beverages all over the world [4], and has been consumed for more than 1000 years. It is a complex mixture of numerous chemical compounds, which may have either beneficial or detrimental effects on human health [5]. Therefore, the relationship between coffee consumption and health has been studied widely, including in a recent umbrella analysis [6] that indicated consistent benefits of coffee consumption for all-cause mortality, liver disease, type-2 diabetes, and some cancers, a finding which was also replicated elsewhere [7]. However, there is conflicting evidence about the effect of coffee consumption pancreatic cancer, with one meta-analysis concluding that coffee consumption was related with the increasing risk of pancreatic cancer [8], and others showing a beneficial effect on pancreatic cancer risk [9,10]. Finally, another meta-analysis indicated no clear trend between coffee consumption and risk of pancreatic cancer [11]. The most recent meta-analysis [12] of 13 cohort studies (up to 1 February 2018), including 959,992 study participants and 3,831 pancreatic cancer cases, included a limited number of studies, which could weaken the power of review. In addition, in the past two years, several new large cohort studies referring to coffee intake and pancreatic cancer [13-15] have emerged, covering more 1.4 million participants and 7,000 cases of pancreatic cancer. Thus, we sought to perform an updated meta-analysis including a large number of high-quality studies to explore the association between coffee consumption and risk of pancreatic cancer.

Methods

Literature search

On 5 January 2020, we searched two electronic databases (PubMed and EMBASE), with the following search terms: (coffee OR caffeine OR intake OR consumption OR diet OR dietary OR beverages OR drinks OR drinking) AND (pancreas OR pancreatic) AND (neoplasia OR neoplasm OR cancer OR tumor OR adenocarcinomas OR malignancy OR malignancies). To obtain more data ensuring the power of evaluation, we did not have a limit on starting date. Furthermore, we limited our search to human studies which were written in English. We followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses statement) guidelines to report this study.

Inclusion and exclusion criteria

Two authors (Mingzhu Zhou and Na Zhang) selected studies independently, studies were included if they met the following criteria: (1) cohort study; (2) the exposure was coffee consumption, including total coffee, caffeinated coffee or decaffeinated coffee, filtered coffee or boiled coffee; (3) the outcome was pancreatic cancer; and (4) provided the relative risk (RR) or hazard ratio (HR) adjusted for smoking and their 95% confidence intervals (CI) of pancreatic cancer for every category of coffee consumption. Studies were excluded if they were as follows: (1) information was insufficient, including the number of participants and cases, RR or HR, and the dose of coffee consumption in each exposure category; (2) duplicate reports.

Data extraction and quality assessment

Date extraction and quality assessment of included publications were conducted independently by two researchers (Mingzhu Zhou and Na Zhang) and disagreements were resolved by discussion.

Figure 1: Flow chart of study selection (search until 5 January 2020).
Figure 1: Flow chart of study selection (search until 5 January 2020).

For each study, the following information was extracted: the first author’s name and year of publication, country, period of followup, age at baseline, number of cases, number of subjects, categories of coffee consumption, RR or HR of pancreatic cancer and corresponding 95% CIs for every category of coffee consumption, and variates adjusted in the statistical analysis. The quality of each publication was evaluated with Newcastle-Ottawa Quality Assessment Scale (NOS), the scale ranges from 0 to 9 points, if scale ≥7 points, the study was considered high quality.

Figure 2: Forest plot of overall RRs (fixed effects model) of coffee consumption (highest versus lowest category) and pancreatic cancer risk.
Figure 2: Forest plot of overall RRs (fixed effects model) of coffee consumption (highest versus lowest category) and pancreatic cancer risk.

The pooled relative risk indicates non-significant association between coffee consumption and pancreatic cancer risk.

Statistical analysis

If a study reported RR for both sexes (males and females) separately, but did not have an overall result, the results for each sex were treated as a separate study. For all the included cohort studies, we computed overall RRs with 95% CIs for the highest versus lowest category of coffee consumption. The heterogeneity 2 2 was measured by I statistics and the chi-square test, if I >50% or P<0.10 was observed, results were defined as heterogeneous, and then the random effect model was used as the pooling method. If not, the fixed effect model was adopted. Besides, Egger’s test, Begg’s test and a funnel plot were used to evaluate a potential publication bias and P<0.05 suggested the presence of publication bias. A sensitivity analysis was conducted by dropping each study for each time to see whether the significant heterogeneity was driven by one specific study. Subgroup analysis was performed for sex, region and coffee type. For the dose–response analysis, we used the method proposed by Greenland, et al. [16] to evaluate linear trends from the correlated natural logs of the RRs through categories of coffee consumption and only included studies that reported at least three exposure categories. For every study, the midpoint of the range was assigned as the average exposure

Figure 3: Begg’s funnel plot for publication bias. No evidence of publication bias was found.
Figure 3: Begg’s funnel plot for publication bias. No evidence of publication bias was found.

value in each category, and the open-ended upper category was assumed to have the same amplitude as the adjacent category. All statistical analyses were conducted with Stata/SE12.0 software (Stata Corp, College Station, TX).

Results

Search results and the quality of included publications

A total of 4,748 records were found from PubMed and EMBASE, while 61 publications were selected for further assessment after removing duplicates and screening titles or abstracts. Finally, 18 cohort studies [13-15, 17-31] were identified based on full test, including 2,859,065 participants and 11,023 cases of pancreatic cancer. Among these 18 cohorts, three of them reported 2 separate outcomes (males and females) [22,24,26], therefore, a total of 21 studies were included in this meta-analysis. The process of selecting studies is summarized in Figure 1. The main characteristics of included studies are shown in Table 1. Of the 18 cohorts, 7 were located in Europe (1 in UK, 1 in Norway and Sweden, 2 in Finland, 2 in Sweden and 1 in multiple European countries), 3 in Asia (Japan) and 8 in North America (United States). All of the included studies had a relatively high quality (NOS score ≥7) and the assessment is reported in supplemental material (Supplementary data).

Coffee consumption and risk of pancreatic cancer

The overall RR of pancreatic cancer risk for the highest versus the lowest category of coffee consumption are shown in Figure 2. The summary RR for the 18 cohorts (21 studies) included in the meta-analysis was 0.95 (95% CI 0.87-1.02, I = 43.4%, P for heterogeneity =0.018). Neither Egger’s test (P for bias =0.450) nor Begg’ test (P for bias = 0.634) indicated a significant publication bias. The funnel plot is shown in Figure 3. And sensitivity analysis confirmed the stability of our results.

Subgroup analysis

Subgroup analysis was performed according to sex, region and coffee type. For sex, nine studies were conducted in females, and analysis of these studies showed that there was no association between coffee consumption and pancreatic cancer risk (RR 0.89, 95% CI 0.73-1.05, I = 0.0%, P for heterogeneity =0.521). On the contrary, when nine studies conducted in males were pooled, coffee consumption was associated with a reduced pancreatic cancer risk (RR 0.72, 95%CI 0.53-0.91, I = 28.0%, P for heterogeneity =0.196). We also divided selected studies into

Table 1: Characteristics of included studies.
Table 1: Characteristics of included studies.

Coffee NOS

Subjects RR/HR(95%CI) Adjustments consumption score

smoking, dietary 0 1.00 (reference) 8 energy intake, type of meat consumed, height, BMI, 309,797 social deprivation, women alcohol intake, educational qualifications, 1-2 cups/day 1.02 (0.83, 1.26) region 193,439 ≤1 cup/day 1.00 (reference) smoking status, BMI, sex, selfreported history of diabetes age, sex, smoking variables, race, marital status, education, alcohol Never 1.00 (reference) consumption, body mass index, 922,896 physical activity, 8 family history of cancer, red and processed meat/ vegetable intake, and current tea drinking ≤1 cup/day 1.01 (0.92, 1.10) Guertin, K. A., US 1995-2006 62(50-71) 1541 et al. (2015)[17] Bidel, S., et al.

Finland 1972-2006 26–74 235 Per 2 cups/day 1.00 (0.97, 1.02) Age, sex, smoking, diabetes, race/ ethnicity, BMI, None 1.00 (reference) highest level of education, alcohol consumption, health status, use of nutritional supplements, physical activity, history of cardiovascular disease, family history of cancer, total energy intake, the nutrient density-adjusted <1 cup/day 1.05 (0.85, 1.30) intakes age, study year, education, cigarette smoking, alcohol consumption, 0 1.00 (reference) leisure time 60,041 8 physical activity, history of diabetes, tea consumption, and body mass index Bhoo-Pathy, N., et al. (2013) Europe 1992-2000 51.2 865 [19] Nilsson, L. M., et al. (2010) Sweden 1992– 2007 50 74 [20] Luo, J., et al.

Japan 1990-2003 40-69 233 sex, center, and age at diagnosis in No intake 1.09 (0.80, 1.50) 1-year categories, and adjusted for height, weight, smoking status, history of diabetes, highest attained education, physical activity, energy intake, as well as red meat, processed meat, alcohol, tea, soft drink, fruit, and Low 1.00 (reference) vegetable intake Moderately low 1.11 (0.92, 1.34) Moderately high 0.99 (0.81, 1.21) High intake 1.07 (0.86, 1.33) Per 100 mL 1.00 (0.97, 1.02) <1 cup/day 1.00 (reference) age, sex, BMI, smoking, 64,603 education, and 7 recreational physical activity 4 cups/day 1.50 (0.57, 3.92) body mass index, leisure-time physical activity in terms of frequency of sports, smoking 102,137 Rarely 1.00 (reference) 7 status, alcohol intake, history of diabetes , history of cholelithiasis, study area and age 13 women 1,634 women 12 men 1,524 men Khan, M. M., et Japan 1984-2002 58 StolzenbergSolomon, R. 27,111 male Finland 1985-1997 58(50–69) 163 Z., et al. (2002) [23] 90 M 44,646 men 62 F and 54,881 Lin, Y., et al.

Japan 1988–1990 57.3(40–79) 1–2 cups/week 1.0 (0.7, 1.4) 3–4 cups/week 1.1 (0.7, 1.7) 1–2 cups/day 0.9 (0.6, 1.3) 3 or more cups/ 0.8 (0.4, 1.3) day 3,158 Men nondrinkers 1.0 (reference) age and smoking Men drinkers 0.6 (0.2, 2.2) age, health status, Women health education, 1.0 (reference) nondrinkers health screening & smoking Women drinkers 0.2 (0.01, 1.8) ≤321.4 g/day 1.00 (reference) 321.4-450.0 g/ 1.48 (0.89, 2.46) day 450.0-624.9 g/ age and years of 1.12 (0.61, 2.03) 7 smokers day smoking 624.9-878.6 g/ 1.72 (1.01, 2.86) day >878.6 g/day 0.95 (0.54, 1.68) 99,527 Males women age and cigarette 1–2 cups/month 0.74 (0.37, 1.49) smoking in pack- 8 years 1–4 cups/week 0.58 (0.32, 1.08) 0.59 (0.26, 1 cup/day 1.33) Isaksson, B., et Sweden 1969-1997 56 131 1986-1998 40-75 (HPFS) 288 Michaud, D.

S., et al. (2001) United States 47,794 men; [26] 1980-1996 30–55(NHS) NHS 88,799 2–3 cups/day 0.75 (0.36, 1.59) Females 1–2 cups/month 1.27 (0.64, 2.54) 1–4 cups/week 0.74 (0.36, 1.50) 2–3 cups/day 0.31 (0.07, 1.33) 1.80 (0.24, ≥4 cups/day 13.66) 0–2 cups/day 1.00 (reference) sex, age and 21,884 3–6 cups/day 0.91 (0.60, 1.38) 7 cigarette smoking 136,593 Men (HPSF) None 1.00 (reference) age in 5-year <1 cup /day 1.04 (0.67, 1.61) categories, pack-years of smoking, BMI, 1 cup /day 0.48 (0.24, 0.95) (HPFS history of diabetes mellitus, history of 2–3 cups /day 0.89 (0.56, 1.40) cholecysectomy, women) energy intake, >3 cups /day 0.37 (0.16, 0.88) period Women (NHS) None 1.00 (reference) <1 cup /day 0.72 (0.36, 1.44) 1 cup /day 0.71 (0.38, 1.30) 2–3 cups /day 0.88 (0.58, 1.34) >3 cups /day 0.88 (0.56, 1.38) Harnack, L. J., et al. (1997) Iowa 1986-1994 ≥55 66 [27] Shibata, A., et United States 1981-1990 65-85 65 Zheng, W., et 17,633 White United States 1966-1986 ≥35 56 Mills, P. K., et United States 1976-1983 ≥25 40 Hiatt, R. A., et United States 1978-1984 40.8 49 three group according to region (Europe, Asia, America). Seven studies conducted in Europe and no differences were identified in the association between coffee intake and pancreatic cancer risk (RR 0.92, 95%CI 0.78-1.05, I = 49.8%, P for heterogeneity =0.063), and similar result was found in the five studies in Asia (RR 0.70, 95%CI 0.33-1.07, I = 0.0%, P for heterogeneity =0.493) and nine studies in America (RR 0.93, 95%CI 0.68-1.19, I = 54.8%, P for heterogeneity =0.024). After we tested for coffee type, the pooled RR was 0.75 (95% CI 0.57, 0.93), 0.90 (95% CI 0.60, 1.20), 1.07 (95% CI 0.87, 1.27), 1.01 (95% CI 0.76, 1.25) for filtered coffee, boiled coffee, caffeinated coffee and decaffeinated coffee, ≤7 cups /week 1.00 (reference) 8-17.5 cups / age, smoking 33,976 1.91 (0.92, 4.00) week status, and pack- 7 women years of smoking ＞17.5 cups / 2.15 (1.08, 4.30) week < 1 cup/day 1.00 (reference) sex, age and 13,976 7 cigarette smoking < 3 cups/day 1.0 (reference) age, smoking index, and alcohol 7 men 5-6 cups/day 0.7 (0.4, 1.6) index ＞7 cups/day 0.9 (0.3, 2.4) Never 1.00 (reference) 34,000 <Daily 1.41 (0.56, 3.58) age, sex, cigarettes 7 >Daily 2.00 (0.91, 4.38) age, sex, ethnic origin, blood glucose level, 122,894 treatment for 7 < 1cup/day 0.8 (0.3, 2.6) diabetes, smoking, and consumption of alcohol >4 cups/day 0.7 (0.2, 1.9) respectively. Results of subgroup analysis are shown in Table 2.

Dose–response analysis

A total of 15 outcomes from 13 publications [13- 15,17,18,20,21,24,26-30] were included in the dose-response analysis for the relationship between coffee consumption and pancreatic cancer risk. The pooled RR for an increment of one cup/day of coffee consumption was 1.00 (95%CI, 0.99-1.02), and was not statistically significant. The Goodness-of-fit indicated no significant heterogeneity among these studies (Q = 19.32, P = 0.153).

Table 2: Results of subgroup analysis.

Subgroup Number of studies RR (95% CI)

Sex

Male 9 0.72 (0.53, 0.91) Female 9 0.89 (0.73, 1.05)

Region

Europe 7 0.92 (0.78, 1.05) Asia 5 0.70 (0.33, 1.07) America 9 0.93 (0.68, 1.19)

Coffee type

Filtered 2 0.75 (0.57, 0.93) Boiled 2 0.90 (0.60, 1.20) Caffeinated 2 1.07 (0.87, 1.27) Decaffeinated 2 1.01 (0.76, 1.25)

Discussion

Coffee may impact the etiology of cancer of various sites along multiple pathways, such as regulation of DNA repair [32] and induction of apoptosis [33]. There exists a possibility that caffeine could reduce insulin sensitivity [34] and induce pancreatic cancer.

In the past few decades, many investigations have been carried out on coffee intake and pancreatic cancer but with inconsistent results [8-12]. Thus, this meta-analysis was conducted to examine the association between coffee consumption and pancreatic cancer risk with 21 high-quality studies from 18 cohorts and to evaluate dose-response trends with 15 studies described in 13 publications.

Findings suggested that there was no association between coffee consumption and pancreatic cancer risk (RR 0.95, 95% CI 0.87- 1.02), which was similar to the results of a meta-analysis based on 37 case-control and 17 cohort studies (RR 1.13, 95% CI 0.99- 1.29) [12]. Subgroup analysis showed significant differences when stratified by sex. The results suggested that there was no association between coffee consumption and pancreatic cancer risk (RR 0.89, 95% CI 0.73-1.05) when studies conducted in females were pooled but coffee consumption was associated with a reduced pancreatic cancer risk (RR 0.72, 95%CI 0.53-0.91) among studies with males, which were similar to the result of a meta-analysis with fourteen cohort studies [35]. Given the impact of cultural, dietary and lifestyle factors, the biological make-up of the subjects in different region [36,37], we tested for region, no significant association I P 28.00% 0.196 0.00% 0.521 49.80% 0.063 0.00% 0.493 54.80% 0.024 0.00% 0.69 53.30% 0.143 0.00% 1 0.00% 0.825 between coffee consumption and the risk of pancreatic cancer was observed in Europe (RR 0.92, 95%CI 0.78-1.05), Asia (RR 0.70, 95%CI 0.33-1.07) and America (RR 0.93, 95%CI 0.68-1.19), respectively. The coffee type [38] and brewing method [39,40] can vary from country to country, which may have an effect on the results of study. Just as studies showed that the compounds retained in coffee was dependent on the preparation method [41]. So we tested for coffee type, we noted that the filtered coffee might play a positive role on pancreatic cancer (RR 0.75, 95% CI 0.57- 0.93), but not for boiled coffee, caffeinated coffee and decaffeinated coffee, it is possible that because a small number of studies which described coffee type were included.

Certain strengths of this meta-analysis should be acknowledged. Firstly, this meta-analysis included cohort studies which were considered high quality, with a larger sample size than metaanalysis performed ever before, which could provide enough power to find association. Additionally, recall bias was greatly minimized with prospective design. Since there are no human experimental studies, these cohort studies could be the best epidemiological evidence available [42]. Secondly, study selection, data extraction and quality assessment of selected studies were performed by two authors independently and discrepancies were resolved by discussion. Furthermore, subgroup analysis was performed to explore if there were any differences among different gender, region and coffee type. Finally, sensitivity analysis confirmed that our result was robust.

There were also several limitations to this study. Firstly, the data of coffee consumption in most studies was obtained by dietary frequency questionnaire, which is often based on the participants’ self-reporting. And it’s possible that some participates have changed their coffee consumption habits after the baseline assessment. But evidence has emerged about reproducibility and validity of information on coffee consumption [43]. Secondly, the assessment of exposure was defined by cups in the selected studies, however, no study told us the information about cup size. The sizes of cups can be different between different countries, for example, standard coffee cups are larger in the United States than in Europe or Japan [44]. Finally, not all included studies adjusted potentially confounding factors, like race, sex, and alcohol intake.

As described in introduction, previous meta-analyses have shown different results between coffee consumption and pancreatic cancer. Li T, et al. [12] showed that the risk of pancreatic cancer was increased with the increment of coffee intake (RR 1.06, 95% CI 1.05-1.07). While Ran H, et al. [9] showed the opposite (RR 0.75, 95% CI 0.63-0.86). And Turati F, et al. [11] concluded no relationship between them (RR 1.13, 95% CI 0.99-1.29). Some researchers [8] put forward that it was possible that any relationship between coffee and pancreatic cancer wasn’t causal but explainable through something between them such as biological, social and environmental factors. For instance, some epidemiological studies [45,46] have shown that the risk of pancreatic cancer is related to blood type. Person whose blood type were A, AB and B have a significantly higher risk of pancreatic cancer than those with type O. In addition, the onset of pancreatic cancer was associated with smoking [47], alcohol [48], obesity [49], dietary [50], age [51], ethnicity, family history and genetic history, infection [52] and so on.

All in all, the results of this study indicated no significant association between coffee consumption and the risk of pancreatic cancer. Of course, there must be many factors affect the relationship between coffee consumption and the risk of pancreatic cancer, and more exploration is needed in future research.

Acknowledgments

This project has been funded by Danone Institute. Danone Institute was not involved in the design and implementation of the study, data collection, analysis and interpretation, writing and editing of the manuscript, and decision to submit the manuscript for publication.

The authors thank Danone Institute for providing the financial support for this manuscript. We also thank Erica T. Perrier for providing suggestions for revising this manuscript.

References

  1. Liu, M., Ji, S., Xu, X., Yu, X. (2020) Advances in basic research, clinical diagnosis and treatment of pancreatic cancer in 2019. China Oncology, (01):1-10.
  2. Siegel, RL. Miller, KD., Jemal, A. (2019) Cancer statistics, 2019. CA Cancer J Clin, 69(1): 7-34.
  3. Feng, R,., Zong, YN., Cao, SM., Xu, RH. (2019) Current cancer situation in China: good or bad news from the 2018 Global Cancer Statistics? Cancer Commun (Lond), 39(1): 22.
  4. Cornelis, MC. (2015) Toward systems epidemiology of coffee and health. Curr Opin Lipidol, 26(1): 20-29.
  5. John Rozar Raj, B., Vishnu Priya, V., Gayathri, R. (2016) Relationship between coffee consumption and cancer- A review. Journal of pharmaceutical sciences and research,
  6. Poole, R., Kennedy, OJ., Roderick, P., Fallowfield, JA., Hayes, PC., Parkes, J. (2017) Coffee consumption and health: umbrella review of meta-analyses of multiple health outcomes. BMJ, 359.
  7. La Vecchia, C., Tavani, A. (2007) Coffee and cancer risk: an update. Eur J Cancer Prev, 16(5): 385-389.
  8. Nie, K., Xing, Z., Huang, W., Wang, W., Liu, W. (2016) Coffee intake and risk of pancreatic cancer: an updated metaanalysis of prospective studies. Minerva Med, 107(4): 270-278.
  9. Ran, HQ., Wang, JZ., Sun, CQ. (2016) Coffee Consumption and Pancreatic Cancer Risk: An Update Meta-analysis of Cohort Studies. Pak J Med Sci, 32(1): 253-259.
  10. Yu, X., Bao, Z., Zou, J., Dong, J. (2011) Coffee consumption and risk of cancers: a meta-analysis of cohort studies. BMC Cancer, 11: 96.
  11. Turati, F., Galeone, C., Edefonti, V., Ferraroni, M., Lagiou, P., La Vecchia, C., et al. (2012) A meta-analysis of coffee consumption and pancreatic cancer. Ann Oncol, 23(2): 311-318.
  12. Li, TD., Yang, HW., Wang, P., Song, CH., Wang, KJ., Dai, LP., et al. (2019) Coffee consumption and risk of pancreatic cancer: a systematic review and dose-response metaanalysis. Int J Food Sci Nutr, 70(5): 519-529.
  13. Zhou, CD., Kuan, AS., Reeves, GK., Green, J., Floud, S., Beral, V., et al. (2019) Coffee and pancreatic cancer risk among never-smokers in the UK prospective Million Women Study. Int J Cancer, 145(6): 1484-1492.
  14. Lukic, M., Nilsson, LM., Skeie, G., Lindahl, B., Braaten, T. (2018) Coffee consumption and risk of rare cancers in Scandinavian countries. Eur J Epidemiol, 33(3): 287-302.
  15. Gapstur, SM., Anderson, RL., Campbell, PT., Jacobs, EJ., Hartman, TJ., Hildebrand, JS., et al. (2017) Associations of coffee drinking and cancer mortality in the cancer prevention study-II. Cancer Epidemiology Biomarkers and Prevention, 26(10): 1477-1486.
  16. Greenland S, Longnecker MP. (1992) Methods for trend estimation from summarized dose-response data, with applications to meta-analysis. Am J Epidemiol, 135(11):1301–1309.
  17. Guertin, KA., Freedman, ND., Loftfield, E., StolzenbergSolomon, RZ., Graubard, BI., Sinha, R. (2015) A prospective study of coffee intake and pancreatic cancer: Results from the NIH-AARP Diet and Health Study. Br J Cancer, 113(7):
  18. Bidel, S., Hu, G., Jousilahti, P., Pukkala, E., Hakulinen, T., Tuomilehto, J. (2013) Coffee consumption and risk of gastric and pancreatic cancer - A prospective cohort study. Int J Cancer, 132(7): 1651-1659.
  19. Bhoo-Pathy, N., Uiterwaal, CSPM., Dik, VK., Jeurnink, SM., Bech, BH., Overvad, K., et al. (2013) Intake of coffee, decaffeinated coffee, or tea does not affect risk for pancreatic cancer: Results from the european prospective investigation into nutrition and cancer study. Clinical Gastroenterology and Hepatology, 11(11): 1486-1492.
  20. Nilsson, LM., Johansson, I., Lenner, P., Lindahl, B., Van Guelpen, B. (2010) Consumption of filtered and boiled coffee and the risk of incident cancer: a prospective cohort study. Cancer Causes Control, 21(10): 1533-1544.
  21. Luo, J., Inoue, M., Iwasaki, M., Sasazuki, S., Otani, T., Ye, W., et al. (2007) Green tea and coffee intake and risk of pancreatic cancer in a large-scale, population-based cohort study in Japan (JPHC study). Eur J Cancer Prev,
  22. Khan, MM., Goto, R., Kobayashi, K., Suzumura, S., Nagata, Y., Sonoda, T.,et al. (2004) Dietary habits and cancer mortality among middle aged and older Japanese living in hokkaido, Japan by cancer site and sex. Asian Pac J Cancer Prev, 5(1): 58-65.
  23. Stolzenberg-Solomon, RZ., Pietinen, P., Taylor, PR., Virtamo, J., Albanes, D. (2002) Prospective study of diet and pancreatic cancer in male smokers. Am J Epidemiol,
  24. Lin, Y., Tamakoshi, A., Kawamura, T., Inaba, Y., Kikuchi, S., Motohashi, Y., et al. (2002) Risk of pancreatic cancer in relation to alcohol drinking, coffee consumption and medical history: findings from the Japan collaborative cohort study for evaluation of cancer risk. Int J Cancer,
  25. Isaksson, B., Jonsson, F., Pedersen, NL., Larsson, J., Feychtin, M., Permert, J. (2002) Lifestyle factors and pancreatic cancer risk: a cohort study from the Swedish Twin Registry. Int J Cancer, 98(3): 480-482.
  26. Michaud, DS., Giovannucci, E., Willett, WC., Colditz , GA., Fuchs, CS. (2001) Coffee and alcohol consumption and the risk of pancreatic cancer in two prospective United States cohorts. Cancer Epidemiol Biomarkers Prev, 10(5): 429- 437.
  27. Harnack, LJ., Anderson, KE., Zheng, W., Folsom, AR., Sellers, TA., Kushi, LH. (1997) Smoking, alcohol, coffee, and tea intake and incidence of cancer of the exocrine pancreas: the Iowa Women's Health Study. Cancer Epidemiol Biomarkers Prev, 6(12): 1081-1086.
  28. Shibata, A., Mack, TM., Paganini-Hill, A., Ross, RK., Henderson, BE. (1994) A prospective study of pancreatic cancer in the elderly. Int J Cancer, 58(1): 46-49.
  29. Zheng, W., McLaughlin, JK., Gridley, G., Bjelke, E., Schuman, LM., Silverman, DT., et al. (1993) A cohort study of smoking, alcohol consumption, and dietary factors for pancreatic cancer (United States). Cancer Causes Control,
  30. Mills, PK., Beeson, WL., Abbey, DE., Fraser, GE., Phillips, RL. (1988) Dietary habits and past medical history as related to fatal pancreas cancer risk among Adventists. Cancer,
  31. Hiatt, RA., Klatsky, AL., Armstrong, MA. (1988) Pancreatic cancer, blood glucose and beverage consumption. Int J Cancer, 41(6): 79.
  32. Bøhn, SK., Blomhoff, R., Paur, I. (2014) Coffee and cancer risk, epidemiological evidence, and molecular mechanisms. Mol Nutr Food Res, 58(5): 915-930.
  33. Alferink, LJM., Kiefte-de Jong, JC., Darwish Murad, S (2018) Potential Mechanisms Underlying the Role of Coffee in Liver Health. Semin Liver Dis, 38(3): 193-214.
  34. Keijzers, GB., De Galan, BE., Tack, CJ., Smits, P. (2002) Caffeine can decrease insulin sensitivity in humans. Diabetes Care, 25(2): 364-369.
  35. Dong, J., Zou, J., Yu, XF. (2011) Coffee drinking and pancreatic cancer risk: a meta-analysis of cohort studies. World J Gastroenterol, 17(9): 1204-1210.
  36. Platt, DE., Ghassibe-Sabbagh, M., Salameh, P., Salloum, AK., Haber, M., Mouzaya, F., et al. (2016) Caffeine Impact on Metabolic Syndrome Components Is Modulated by a CYP1A2 Variant. Ann Nutr Metab, 68(1): 1-11.
  37. Sartini, M., Bragazzi, NL., Spagnolo, AM., Schinca, E., Ottria, G., Dupont, C., et al. (2019) Coffee Consumption and Risk of Colorectal Cancer: A Systematic Review and MetaAnalysis of Prospective Studies. Nutrients, 11(3): 694.
  38. Cavin, C., Marin-Kuan, M., S. Langouët, C. Bezençon, G. Guignard, C. Verguet, D. Piguet, D. Holzhäuser, R. Cornaz and B. Schilter. (2008). Induction of Nrf2-mediated cellular defenses and alteration of phase I activities as mechanisms of chemoprotective effects of coffee in the liver. Food Chem Toxicol, 46(4), 1239-1248.
  39. Kalschne, DL., Biasuz, T., De Conti, AJ., Viegas, MC., Corso, MP., Benassi, MT. (2019) Sensory characterization and acceptance of coffee brews of C. arabica and C. canephora blended with steamed defective coffee. Food Res Int, 124: 234-238.
  40. Rendón, MY., Dos Santos Scholz, MB., Bragagnolo, N. (2018) Physical characteristics of the paper filter and low cafestol content filter coffee brews. Food Res Int, 108: 280-285.
  41. Rothwell, JA., Loftfield, E., Wedekind, R., Freedman, N., Kambanis, C., Scalbert, A., et al. (2019) A Metabolomic Study of the Variability of the Chemical Composition of Commonly Consumed Coffee Brews. Metabolites, 9(1): 7.
  42. Zhou, Y., Tian, C., Jia, C. (2012) A dose-response metaanalysis of coffee consumption and bladder cancer. Prev Med, 55(1): 14-22.
  43. Ferraroni, M., Tavani, A., Decarli, A., Franceschi, S., Parpinel, M., Negri, E., et al. (2004) Reproducibility and validity of coffee and tea consumption in Italy. Eur J Clin Nutr, 58(4): 674-680.
  44. Bracken, MB., Triche, E., Grosso, L., Hellenbrand, K., Belanger, K., Leaderer, BP. (2002) Heterogeneity in assessing self-reports of caffeine exposure: implications for studies of health effects. Epidemiology, 13(2): 165- 171.
  45. Wolpin, BM., Chan, AT., Hartge, P., Chanock, SJ., Kraft, P., Hunter, DJ., et al. (2009) ABO blood group and the risk of pancreatic cancer. J Natl Cancer Inst, 101(6): 424-431.
  46. Wolpin, BM., Kraft, P., Gross, M., Helzlsouer, K., Buenode-Mesquita, HB., Steplowski, E., et al. (2010). Pancreatic cancer risk and ABO blood group alleles: results from the pancreatic cancer cohort consortium. Cancer Res, 70(3):
  47. Bosetti, C., Lucenteforte, E., Silverman, DT., Petersen, G., Bracci, PM., Ji, BT., et al. (2012) Cigarette smoking and pancreatic cancer: an analysis from the International Pancreatic Cancer Case-Control Consortium (Panc4). Ann Oncol, 23(7): 1880-1888. Allen NE, Beral V, Casabonne D, et al. Moderate alcohol intake and cancer incidence in women (2009) J Natl Cancer Inst 101(5):296–305.
  48. Allen NE, Beral V, Casabonne D, et al. (2009) Moderate alcohol intake and cancer incidence in women. J Natl Cancer Inst, 101(5):296–305.
  49. Davoodi, SH., Malek-Shahabi, T., Malekshahi-Moghadam, A., Shahbazi R., Esmaeili, S. (2013) Obesity as an important risk factor for certain types of cancer. Iran J Cancer Prev,
  50. Maisonneuve P., Lowenfels, AB. (2015) Risk factors for pancreatic cancer: a summary review of meta-analytical studies. Int J Epidemiol, 44(1):186-198.
  51. Midha, S., Chawla, S., Garg, PK. (2016) Modifiable and nonmodifiable risk factors for pancreatic cancer: A review. Cancer Lett, 381(1): 269-277. Stevens RJ, Roddam AW, Beral V. (2007) Pancreatic cancer in type 1 and youngonset diabetes: systematic review and meta-analysis. Br J Cancer, 96(3):507–509.
  52. Stevens RJ, Roddam AW, Beral V. (2007) Pancreatic cancer in type 1 and young-onset diabetes: systematic review and meta-analysis. Br J Cancer, 96(3):507–509.