How can green tea polyphenols affect drug metabolism and should we be concerned?

KEYWORDS:

• Camellia sinensis
• cancer chemotherapeutics
• Cardiovascular disease (CVD) drugs
• drug metabolism
• green tea
• green tea polyphenols
• polyphenol-drug interactions
• polyphenol metabolism

1. Introduction

The cornerstones of any drug treatment are safety and efficacy considerations, both features are commonly assessed using randomized controlled trials (RCTs). However, confounding variables related to the concurrent use of other exogenous compounds have rarely been evaluated, with the earlier exemption of alcohol-drug interactions [1] as outlined also recently [2], and caffeine-drug interactions [3]. Currently, some interest is directed to polyphenols (PP) of green tea (GT) or green tea extracts (GTE) in connection with PP-drug interactions, an issue to be addressed in this report.

2. Putative molecular events of PP-drug interactions

Based on endoscopic studies in humans, the intestinal tract with preference of the colon contains multiple isoforms of cytochrome P450 (CYP or CYP450), corroborated for the small intestine using animal models [4]. Studies in human and animal tissues also revealed that GT PP like epigallocatechin-3-gallate (EGCG) epicatechin-3-gallate (ECG), epicatechin (EC), epigallocatechin (EGC), and epigallocatechin-3-catechins (EGCG) modify but are not metabolized by intestinal and hepatic CYP isoforms including CYP2C8, CYP2B6, CYP3A4, CYP2D6, and CYP2C19, as these PP upregulate or downregulated specific CYP activities [5,6]. Putative mechanisms of PP metabolism may also include intestinal and hepatic efflux transporter systems such as p-glycoprotein, OATP1A2 or OATP2B1, whereby OATP is derived from Organic-Anion-Transporting Polypeptide [6]. In humans, PP are also metabolized by phase II enzymes such as UGTs (Uridine-5´-diphosphate glucuronosyltransferase) and SULTs (Sulfotransferase) [7]. Evidently, possible PP-drug interactions are under the control of various complex and complicated mechanisms.

3. Clinical, pharmacodynamic, pharmacokinetic, and bioavailability studies

Human health conditions can be impaired by a few hundred different diseases, on the top are cardiovascular diseases (CVD) [6,8–13] and cancer [14]. Both disease categories are, as suspected by others, related to the generation of reactive oxygen species (ROS) [15,16], causing respective patients to believe that GT polyphenols could be helpful fighting against these diseases. Indeed, many patients use GT polyphenols in addition to conventional drugs for CVD or to cancer chemotherapeutics [6,8–14]. Because the metabolism of the polyphenols may interfere with drug disposition leading to unwanted adverse effects, a closer look on possible clinical relevant interactions of polyphenols and drugs is warranted with a focus on CVD and cancer.

3.1. Cardiovascular diseases

Pharmacokinetic studies in humans have shown that GT polyphenol use influences differently bioavailability parameters in individuals undergoing therapy with CVD drugs [6,8–13]. The number of respective studies in humans is small, a few relevant ones will be discussed as examples. Among 15 healthy volunteers and compared to digoxin alone, the concomitant administration of digoxin and GT PP significantly reduced the systemic exposure of digoxin: the peak plasma concentration (C_max) and the area under the concentration-time curve (AUC) from time 0 to the last measurable time (AUC_last) [8]. A pharmacokinetic study in volunteers with sildenafil showed a significant decrease of AUC_0–∞ by 50%, interpreted as a consequence following inhibition of intestinal CYP 3A [6,9]. Other pharmacokinetic studies on simvastatin for AUC_0-6 presented either small but not significant increases or significant increases by 22%, which were interpreted as p-glycoprotein-mediated efflux and inhibition of intestinal CYP 3A4 causing a reduced first-pass metabolism [6]. An additional pharmacokinetic study on rosuvastatin provided for AUC_last a small decline by 19%, which was explained as inhibition of intestinal OATP1A2 or OATP2B1 and resulting reduced uptake [6,10]. Results from a combined pharmacokinetic and pharmacodynamic study on nadolol showed for AUC_0-48 a significant reduction by 85%, caused by an inhibition of intestinal OATP1A2-mediated uptake; in addition, there was also a significant suppression of the systolic blood pressure-lowering effect [6,11]. Interaction of GT with warfarin was of clinical relevance as evidenced by an abnormal INR (International Normalized Ratio), requiring substitution therapy with vitamin K; additional pharmacokinetic data on warfarin were not provided, not allowing a statement on the putative mechanism [6,12]. Finally and of potential interest for patients with a heart transplantation, studies with tacrolimus presented a two-fold increase versus previous level, explained as intestinal p-glycoprotein-mediated efflux and inhibition of intestinal CYP 3A4-mediated first-pass metabolism [6,13].

3.2. Cancer chemotherapeutics

Broadly used in patients with cancer and undergoing treatment with chemotherapeutics, PP and GT received special attention, with most results derived from in vitro studies or cell preparations [14]: (1) in rats, GTE increased the bioavailability of 5-FU, C_max and AUC, associated with a prolongation of the half-life of 5-FU in plasma; (2) in rats, co-administration of EGCG and irinotecan altered the pharmacokinetics of irinotecan because pretreatment with EGCG increased irinotecan AUC in plasma by 57.7% associated with a substantial prolongation of the half-life, and it has been argued that the increased plasma concentrations of irinotecan may be associated with enhanced pharmacological effects or toxicity; (3) in rats, plasma C_max and AUC_0–∞ of sunitinib were significantly decreased by 47.7% and 51.5%, indicating a reduced bioavailability of sunitinib and suggesting that patients receiving sunitinib therapy should abstain from using GT or PP [14]. Of note, bortezomib-EGCG chemical interaction has been reported [17] as another example of GT vs. anti-cancer drug interaction.

The above results derived from experimental studies merit additional comment, in particular, whether the translation of these experimental data to a clinical setting is warranted and of potential benefit of patients. Based on present experimental knowledge, promotion of concurrent use of PP with any chemotherapeutic drug in cancer patients must be discouraged. Instead, a call for caution seems more appropriate: first, using in experimental 5-FU studies GTE rather than small amounts of PP as mentioned above and published [14] is outside the clinical scope, because GTE are potentially toxic to the liver and not recommended for human use [5], an important difference in potential toxicity between the extract and the usual beverage. The second point requiring clarification refers to the statement in the abstract of the review article [14] mentioning previous studies, which would have demonstrated that the combination of chemotherapeutic drugs and green tea extract or tea polyphenols could synergistically enhance treatment efficacy and reduce the adverse side effects of anticancer drugs in cancer patients. Again, and as outlined in a recent analysis [5], the use of GTE with its high liver injury risk cannot be recommended due to a negative benefit: risk constellation, an unacceptable doubled hepatotoxic risk for patients with cancer and under a potentially hepatotoxic cytostatic therapy are facing if GTE and the cytostatic drugs are used concomitantly. The third point is directed to the published belief [14] that the combination of multidrug cancer treatment with GTE or GT PP would improve treatment efficacy and diminish negative side effects. This is not substantiated due to lack of clinical or experimental evidence base, rather more appropriate is a general cautionary in view of their referenced experimental studies on increased bioavailability of cancer chemotherapeutic drugs caused by PP as outlined above, similar to the warning above specified for sunitinib. Finally, in reference to previous studies it has been stated that survival rates of cancer patients could be improved by using chemotherapeutic drugs together with green tea; this combination was also claimed as reducing the cancer risk, a vague statement because the patients of the addressed cohort had already cancer, and there is no base to reduce cancer risk [14].

In clinical trial setting, interactions between GT PP and chemotherapeutic drugs have not been reported for patients with cancer [14]. However, such a combined therapy approach is certainly not a good clinical option and likely not of benefit for the cancer patients. The overall promotion to use PP, GT, or GTE to neutralize ROS is based on the assumption that cancer among many other human diseases is causally related to the generation of ROS [15,16], a basically interesting and stimulating experimental theory but not justified, difficult to evaluate in human organs, and not yet proven in human diseases with the required degree of certainty.

4. Expert opinion

Green tea (GT) with its polyphenols (PP) is heavily promoted and largely consumed under the assumption it could help improve health conditions, but consumers often use in addition conventional drugs that potentially lead to serious PP-drug interactions. GT with their PP as ingredients is a popular and commonly well-tolerated beverage appreciated in a large number of countries throughout the world. The authors conclude that little attention was paid so far to PP-drug interactions in clinical studies or randomized controlled trials (RCTs). They also noticed that concomitant use of GT or PP with conventional drugs was frequently observed in patients with cardiovascular disease (CVD) and cancer that could lead to drug toxicity or lack of drug efficacy. Therefore, these two cohorts were specifically analyzed.

A few examples of the CVD cohort may illustrate the variable results: no clinically relevant PP-drug interactions were reported for digoxin, sildenafil, simvastatin, and rosuvastatin, but most of these drugs experienced some impairment of bioavailability due to PP use; conversely, adverse effects caused by PP were published for nadolol as evidenced by a significant suppression of the systolic blood pressure, and for warfarin as evidenced by problems with a blood clotting parameter requiring substitution therapy by vitamin K. Caution is required for PP use in patients treated for heart transplantation using tacrolimus with its narrow therapeutic index because PP may cause a two-fold increase of tacrolimus bioavailability. The authors are of the opinion that these risky clinical conditions are not acceptable for reasons of safety.

The authors viewed conditions of the cancer cohort as more complex because clinical data are limited and published conclusions were mostly derived from experimental studies with disputable translation to humans. Despite these limitations, the authors accept that experimental studies call for caution using PP or GT in combination with 5-FU, irinotecan, or sunitinib, because all these cancer therapeuticals are conflicted either with a low or high bioavailability due to PP or GT, a clinical problem of high risk due to the narrow therapeutic index of these drugs and the related potential of drug underdosing or overdosing.

The authors strongly recommend patients undergoing therapy with drugs known for their narrow therapeutic index to abstain from concomitant use of GT and PP, which can severely modify drug bioavailability with the associated risk of drug underdose or overdose.

Authors’ contributions

R Teschke wrote the first draft, which was carefully edited by TD Xuan, and the final version was prepared. Both authors agreed on the final manuscript to be submitted.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.