Herb-Drug Interactions: Isoflavones

Isoflavonoids

This is a large group of related compounds with similar structures and biological properties in common, which are widely available as additives in dietary supplements as well as the herbs or foods that they were originally derived from. Isoflavones are the subject of intensive investigations and new information is constantly being published.

You may have come to this monograph via a herb that contains isoflavones. The information in this monograph relates to the individual isoflavones, and the reader is referred back to the herb (and vice versa) where appropriate. It is very difficult to confidently predict whether a herb that contains one of the isoflavones mentioned will interact in the same way. The levels of the isoflavone in the particular herb can vary a great deal between specimens, related species, extracts and brands, and it is important to take this into account when viewing the interactions described below.

Types, sources and related compounds

Isoflavones are plant-derived polyphenolic compounds that are a distinct group of flavonoids. They can exert oestrogen-like effects, and therefore belong to the family of ‘phytoestrogens’. Most occur as simple isoflavones, but there are other derivatives such as the coumestans, the pterocarpans and the rotenoids, some of which also have oestrogenic properties.

The isoflavones are found in small amounts in many legumes, seeds, grains and vegetables, but soya, is by far the most concentrated dietary source; it contains principally genistein and daidzein. There are various other isoflavone-rich supplements, including those derived from alfalfa and red clover (both of which are rich in biochanin A and formononetin), and kudzu, which contains puerarin. In addition, some popular herbal medicines, such as astragalus and shatavari contain isoflavones as well as other types of active constituents.

In plants, isoflavones are usually found in the glycoside form, i.e. bound to a sugar molecule, but digestion results in the release of the sugar molecule leaving the aglycone. The most important isoflavones are: genistein and daidzein, which are hydrolysed from their glycosides genistin, daidzin and puerarin (daidzein 8-C-glucoside); glycetein and its glycoside glycitin; formononetin, biochanin A, isoformono-netin, prunetin, calycosin, ononin, orobol; and others.

Use and indications

Epidemiological studies show that a high dietary intake of isoflavones from foods such as soya might be protective against certain cancers (breast, endometrium, prostate) and degenerative diseases, in the same way as the flavonoids. Although isoflavone supplements are used for these possible benefits, it remains to be seen whether they are effective. Many of their biological effects, as with the flavonoids, appear to be related to their ability to modulate cell signalling pathways, and genistein in particular has been widely investigated for its tyrosine kinase-inhibiting properties, and it is now also considered by some to be a SERM (selective oestrogen receptor modulator). Some biologically active constituents of genistein have given cause for concern, as it can be genotoxic and cause cell damage, and it is a topoisomerase II inhibitor.

Isoflavones have weak oestrogenic effects, but under certain conditions (for example, in premenopausal women) they can also act as oestrogen antagonists by preventing the more potent natural compounds, such as estriol, from binding to receptor sites. In some cases the activities are tissue selective. Isoflavones also inhibit the synthesis and activity of enzymes involved in oestrogen and testosterone metabolism, such as aromatase.

Because of their oestrogenic effects, isoflavone supplements have been investigated for treating menopausal symptoms such as hot flushes (hot flashes) and for prevention of menopausal osteoporosis, with generally modest to no benefits when compared with placebo in randomised controlled studies. Isoflavones have also been extensively studied for lipid lowering, and there are a few studies on other cardiovascular benefits, and effects on cognitive function. In a 2006 analysis, the American Heart Advisory Committee concluded that the efficacy and safety of soya isoflavones were not established for any indication and, for this reason, they recommended against the use of isoflavone supplements in food or pills.

Pharmacokinetics

The uptake, metabolism and disposition of the isoflavones are highly complex and have not yet been fully elucidated. Isoflavone glycosides are probably hydrolysed in the gut wall by intestinal beta-glucosidases to release the aglycones (genistein, daidzein, etc.), which can then be absorbed. Intestinal bacteria may also hydrolyse the glycosides, and, in some people, they metabolise daidzein to the more active oestrogen equol. After absorption, the aglycones are conjugated, predominantly to glucuronic acid. Gut bacteria also extensively metabolise isoflavones: for example, daidzein may be metabolised to equol, a metabolite with greater oestrogenic activity than daidzein, but also to other compounds that are less oestrogenic. Because of differences in gut flora, there are individual differences in the metabolism of isoflavones, which might have important implications for their effects: for example, studies measuring urinary equol excretion after soya consumption indicate that only about one-third of Western individuals metabolise daidzein to equol.

In a study in 9 healthy subjects the isoflavone puerarin, given orally, was rapidly absorbed and reached peak levels at 2 hours, and had a half-life of about 4.5 hours. The elimination half-life was not significantly altered after repeated administration. The authors concluded that three times a day dosing is recommended, as accumulation will not occur, and plasma levels remain at levels that are biologically active, even 8 hours after the last steady-state dose. For mention that colonic bacteria hydrolyse puerarin to the more active aglycone daidzein, see Isoflavones + Antibacterials.

In an in vitro study in human liver microsomes, fluvoxamine was a potent inhibitor of genistein and tangeretin metabolism. This finding suggests that these isoflavones are principally metabolised by CYP1A2, of which fluvoxamine is a potent inhibitor. The relevance of this to the activity of these isoflavones is unknown, since the relative activity of the metabolites to the parent isoflavone is unknown.

The isoflavones genistein and equol were found to inhibit the cytochrome P450 isoenzymes CYP2E1 and CYP1A2, see also theophylline, but note also that infant formulas, including soya-based formulas, appear to induce CYP1A2, see Soya + Caffeine.

In vitro, the soya isoflavones daidzein and genistein and a hydrolysed soya extract inhibited CYP3A4. and CYP2C9. However, in one of these studies, St John’s wort also inhibited CYP3A4, but clinically this herb is known to be an inducer of CYP3A4. This highlights the problems of extrapolating the findings of in vitro studies to the clinical situation.

Genistein and biochanin A inhibit P-glycoprotein-mediated drug transport, for example, see paclitaxel and digoxin.

Interactions overview

The interactions covered in the following sections relate to individual isoflavones. Some of these may be directly applicable to isoflavone supplements; however, caution must be taken when extrapolating these interactions to herbs or foods known to contain the isoflavone in question. This is because the amount of the isoflavone found in the herb or food can be highly variable, and might not be known, and other constituents present in the herb or food might affect the bioavailability or activity of the isoflavone. Therefore, although data on isolated isoflavones are useful, it is no substitute for direct studies of the herb or food in question.

Isoflavones

Isoflavones + Antibacterials

The interaction between isoflavones and antibacterials is based on a prediction only.

Clinical evidence

No interactions found.

Experimental evidence

It is well known that colonic bacteria are involved in the metabolism of isoflavones. For example, it has been shown that equol is exclusively formed from daidzin by colonic bacteria, but that only about one-third of people are equol producers. In another study, the isoflavone glycosides puerarin and daidzin were incubated with human intestinal bacteria. All bacteria hydrolysed daidzin to the aglycone daidzein, and a few bacteria also transformed puerarin to daidzein. Human faecal specimens hydrolysed puerarin and daidzin to daidzein, but their hydrolysing activities varied between individual specimens. When the oestrogenic effects of the glycosides puerarin and daidzin were compared with those of the aglycone daidzein, the aglycone metabolite was more potent. Intestinal bacteria have also been reported to metabolise daidzein to the more active oestrogen equol.

However, it is also now established that beta-glucosidases in the intestinal wall are also important for the hydrolysis of glycosides to form the aglycones (see Pharmacokinetics).

Mechanism

Colonic bacteria appear to play an important role in the metabolism of soya isoflavones; therefore, it is possible that antibacterials that decimate colonic bacteria could alter isoflavone metabolism and biological activity.

Importance and management

Evidence is limited to experimental studies that were not designed to study drug interactions; however, what is known suggests that the concurrent use of antibacterials active against gut flora might theoretically alter or reduce the efficacy of some isoflavones. However, there is no clinical evidence to support this supposition and, in any case, the effect is likely to be temporary. No action is therefore needed.

Isoflavones + Antidiabetics

The interaction between isoflavones and antidiabetics is based on experimental evidence only.

Evidence, mechanism, importance and management

In various studies in animal models of diabetes, a couple of which are cited for information, puerarin, an isoflavone found in kudzu, has demonstrated blood glucose-lowering effects.

Some have interpreted these studies to indicate that kudzu might have additive effects with antidiabetic drugs, and that the dose of antidiabetic medications might need to be adjusted. Given the nature of the evidence, and the fact that it relates to isolated isoflavone constituents, this appears to be a very cautious approach.

Isoflavones + Benzodiazepines

The interaction between isoflavones and benzodiazepines is based on experimental evidence only.

Evidence, mechanism, importance and management

In two experimental studies, the isoflavone puerarin has been shown to be a weak benzodiazepine antagonist. It is therefore theoretically possible that puerarin might reduce the effects of benzodiazepines if given concurrently. However, there is no clinical evidence to support this supposition. The fact that the information relates to an isolated isoflavone, and the effect was only weak, suggests that a clinically important interaction between isoflavones and benzodiazepines is unlikely.

Isoflavones + Cardiovascular drugs; Miscellaneous

The interaction between isoflavones and miscellaneous cardiovascular drugs is based on experimental evidence only.

Evidence, mechanism, importance and management

Some experimental studies have shown that isoflavones from kudzu, may inhibit of platelet aggregation. In one small study in patients with angina, treatment with the isoflavone puerarin reduced the activation of platelet surface activity protein.

Some have interpreted these studies to indicate that, theoretically, kudzu might increase the risk of bleeding when used with antiplatelet drugs or anticoagulants, and that caution is warranted on concurrent use. Given the nature of the evidence, and the fact that it relates to isolated isoflavone constituents of kudzu, this appears to be a very cautious approach.

Note that puerarin injection is used in China to treat angina and cardiovascular disease. Clinical studies comparing standard Western treatment (nitrates, beta blockers, calcium-channel blockers, aspirin, anticoagulants, etc.) with or without puerarin injection have been reviewed. It was concluded that, although adverse events were inadequately reported, treatment including the injection tended to result in more adverse effects.

Isoflavones + Digoxin

The interaction between isoflavones and digoxin is based on experimental evidence only.

Clinical evidence

No interactions found.

Experimental evidence

In a study in rats, biochanin A 100 mg/kg increased the AUC and maximum serum levels of an oral 20-mg/kg dose of digoxin by 75% and 71%, respectively. No significant changes in mean residence time and terminal half-life of digoxin were observed, suggesting a negligible effect of biochanin A on renal elimination.

Mechanism

Digoxin is a substrate for P-glycoprotein. Biochanin A may modestly inhibit P-glycoprotein, resulting in a moderate increase in oral bioavailability of digoxin.

Importance and management

There appears to be no clinical data regarding an interaction between biochanin A and digoxin, and the clinical relevance of the experimental data needs to be determined. However, until more is known, because of the narrow therapeutic index of digoxin, it may be prudent to be cautious if patients taking digoxin also wish to take supplements containing high doses of biochanin A. Patients should be alert for any evidence of adverse effects, such as bradycardia, and if these occur it may be prudent to monitor digoxin levels.

Isoflavones + Fexofenadine

The interaction between isoflavones and fexofenadine is based on experimental evidence only.

Clinical evidence

No interactions found.

Experimental evidence

The effects of biochanin A on the pharmacokinetics of fexofenadine was investigated in rats. Biochanin A 100 mg/kg decreased the oral bioavailability and peak plasma concentration of fexofenadine 20mg/kg by about 30% and 57%, respectively. No significant changes in mean residence time and terminal half-life were observed, suggesting a negligible effect of biochanin A on fexofenadine hepatic or renal elimination.

Mechanism

Fexofenadine is a substrate for P-glycoprotein and OATP, both of which affect fexofenadine uptake. Biochanin A appears to preferentially inhibit OATP over P-glycoprotein in the intestine, leading to the decreased oral absorption of fexofenadine.

Importance and management

There appear to be no clinical data regarding an interaction between biochanin A and fexofenadine, and therefore the clinical relevance of the experimental data needs to be determined. However, the modest reduction in bioavailability of fexofenadine suggests that a clinically important interaction is unlikely. No action is therefore required.

Isoflavones + Food

No interactions found

Isoflavones + Herbal medicines

No interactions found. Isoflavones are regularly ingested as part of the diet.

Isoflavones + Nicotine

Soya isoflavones slightly decrease the metabolism of nicotine.

Clinical evidence

The effects of soya isoflavones on nicotine metabolism were investigated in a study in 7 healthy Japanese subjects who were non-smokers. Taking an isoflavone tablet (containing daidzein 4.2mg, genistein 5.2 mg and glycitein 600 micrograms) six times daily for 5 days, in addition to their usual diet, slightly reduced nicotine metabolism by about 24% (measured 2 hours after chewing a piece of nicotine gum). This was when compared with nicotine metabolism after abstaining from soya foods for one week.

Experimental evidence

In vitro, these isoflavones decreased nicotine metabolism, by inhibiting the cytochrome P450 isoenzyme CYP2A6.

Mechanism

Isoflavones slightly decrease the metabolism of nicotine by the inhibition of CYP2A6.

Importance and management

Although evidence is limited to one study, it is a well-designed clinical study. The minor change in nicotine metabolism when the subjects were taking isoflavones suggests that isoflavone supplements are unlikely to have a clinically relevant effect on the efficacy of nicotine replacement therapy. No action is therefore needed.

Isoflavones + Paclitaxel

The interaction between isoflavones and paclitaxel is based on experimental evidence only.

Clinical evidence

No interactions found.

Experimental evidence

In a study in rats, genistein 10 mg/kg given orally 30minutes before a single dose of oral or intravenous paclitaxel modestly increased the AUC of paclitaxel by 55% and 43%, respectively. The increase in AUC with a lower dose of genistein 3.3 mg/kg was not significant, although it did increase the peak concentration of paclitaxel by 67%.

In another similar study, biochanin A 100 mg/kg caused a marked 3.8-fold increase in the oral bioavailability of paclitaxel 20 mg/kg.

Mechanism

It seems that these isoflavones increase the systemic exposure of oral paclitaxel by inhibiting P-glycoprotein. In addition, isoflavones might reduce paclitaxel drug resistance via their effects on P-glycoprotein.

Importance and management

The available evidence for an interaction between isoflavones and paclitaxel is from experimental studies, the clinical relevance of which needs to be determined. Furthermore, paclitaxel is given intravenously, and the effect of biochanin A has only been assessed with oral paclitaxel. However, genistein modestly increased the AUC of intravenous paclitaxel, and therefore, until more is known, some caution might be appropriate with high doses of these individual isoflavones, in view of the possibility of increased exposure and increased toxicity of paclitaxel.

Isoflavones + Tamoxifen

The available evidence on the effect of isoflavone supplements on the efficacy of tamoxifen in breast cancer is inconclusive, and the effect of isoflavones on breast tissue appears to be complex. It is possible that whether the effect is beneficial or antagonistic might be related to the dose of isoflavones used, and also the oestrogen status of the patient (pre- or postmenopausal).

Evidence and mechanism

(a) Breast cancer

In various animal studies, soya isoflavones have either inhibited or enhanced the preventative effect of tamoxifen on the development of breast cancer. Note that the body of evidence is vast, and only a selection of representative papers has therefore been cited. For example, in a study in rats given tamoxifen, a diet supplemented with daidzein increased protection against chemically induced breast cancer, whereas a diet supplemented with genistein reduced protection, when compared with tamoxifen alone. In another study, a ‘low-dose’ isoflavone-enriched diet (genistein plus daidzein) halved the protective effect of tamoxifen against the development of breast tumours, whereas a soy meal or ‘high-dose’ isoflavones did not have any effect. In yet another study, genistein and tamoxifen had a synergistic effect on delaying the growth of oestrogen-dependent breast tumours in mice, especially at lower levels of tamoxifen.

Note that disparate findings (both prevention and stimulation) have been found for genistein alone on induction of mammary tumours in animals. It has been suggested that the effect might depend on age, with a preventative effect seen at a young age, and a stimulatory effect seen when oestrogen levels are low, as occurs postmenopausally. Note also that there is a large body of epidemio-logical data on the effect of dietary soya products on the risk of breast cancer, which suggest a possible reduction in risk.

Some animal studies have clearly shown that genistein can antagonise the inhibitory effect of tamoxifen on growth of oestrogen-dependent human breast cancer. In an in vitro study, this effect was shown to be biphasic, with low levels of genistein simulating cancer cell growth, and high levels of genistein inhibiting cancer cell growth. Similarly, in vitro studies have shown that genistein has a synergistic or additive inhibitory effect on the growth of breast cancer cells exposed to tamoxifen, or antagonises the response of breast cancer cells to tamoxifen.

Note that, in one study in 17 women with biopsy-confirmed breast cancer, supplementation with soya isoflavones 200 mg daily for 2 weeks did not increase tumour growth over the 2 to 6 weeks before surgery. There was a trend towards cancer growth inhibition in the isoflavone treatment group, manifested as an increase in the apoptosis/mitosis ratio, when compared with those from a historical control group, although this was not statistically significant.However, in another study in women requiring surgery for a benign or malignant breast tumour, supplementation with dietary soy, containing isoflavones 45 mg daily for 2 weeks, increased proliferation markers in a healthy zone of the breast. Similarly, another study of dietary supplementation with soya protein (providing 37.4 mg of genistein daily) for 6 months found an increase in breast secretion (an assessment of breast gland function) in premenopausal women, but a small or lack of an increase, in postmenopausal women and epithelial hyperplasia in about one-third of the women, which was suggestive of oestrogenic breast tissue stimulation in response to genistein.

(b) Menopausal symptoms

In a placebo-controlled crossover study in 149 women with a history of breast cancer, about two-thirds of whom were taking tamoxifen, soya isoflavones (genistein, daidzein and glycitein) 50 mg three times daily for 4 weeks had no effect on the incidence of hot flushes. Another similar study also reported a lack of efficacy for hot flushes in 157 women breast cancer survivors given a soya beverage (90 mg isoflavones daily) or placebo beverage for 12 weeks, of whom about one-third were taking tamoxifen. Vaginal spotting was reported by 4 women who drank the soya beverage and one woman who drank the placebo beverage, but this was not thought to be due to the soya. In a third study, in 72 women with breast cancer, 78% of whom were taking tamoxifen, a soya supplement (35 mg of isoflavones twice daily; Phytosoya) for 12 weeks had no effect on menopausal symptoms when compared with placebo.

These studies are probably too short, and too small, to detect any possible effect of the isoflavones on the efficacy of tamoxifen. Nevertheless, they show that isoflavones are probably no more effective than placebo for one of the most common reasons for which they are used in this patient group.

(c) Tamoxifen metabolism

In a cross-sectional study in 380 Asian-American women (including Chinese and Japanese Americans) with breast cancer the serum levels of tamoxifen and its major metabolites were unrelated to serum levels of isoflavones (genistein, daidzein, equol) or reported dietary soya intake.

In an in vitro study using female rat liver micro somes, genistein inhibited alpha-hydroxylation of tamoxifen (a minor metabolic route), but did not affect 4-hydroxylation, N-demethylation or W-oxidation (major metabolic routes). A combination of three to five isoflavones (genistein, daidzein and glycitein, or these three isoflavones plus biochanin A and formononetin) inhibited tamoxifen alpha-hydroxylation to a greater extent, but did not decrease the formation of other metabolites. Studies using selective chemical inhibitors showed that tamoxifen alpha-hydroxylation was mainly mediated by CYP1A2 and CYP3A1/2 in rats. Although alpha-hydroxytamoxifen is a minor metabolite of tamoxifen, it is thought to be responsible for DNA adduct formation and increased risk of endometrial cancer with tamoxifen. The authors concluded that using genistein and its isoflavone analogues with tamoxifen might potentially be beneficial because of the inhibition of the formation of alpha-hydroxytamoxifen.However, this requires confirmation in humans. Also, note that isoflavones themselves may not be free of endometrial adverse effects, for example, in one study, long-term clinical use of isoflavones (genistein, daidzein, glycitein) induced endometrial hyperplasia in some women.

Importance and management

The available evidence on the effect of isoflavone supplements on the efficacy of tamoxifen in breast cancer is inconclusive, and the effect of isoflavones on breast tissue appears to be complex. It is possible that whether the effect is beneficial or antagonistic might be related to the dose of isoflavones used, and also the oestrogen status of the patient (pre- or postmenopausal). Because of differences in gut flora, there are individual differences in the metabolism of isoflavones, which might have important implications for their effects: for example, studies measuring urinary equol (which has more potent oestrogenic effects than daidzein) excretion after soya consumption indicate that only about one-third of Western individuals metabolise daidzein to equol.

Most authorities recommend that patients taking oestrogen antagonists (that is, drugs such as tamoxifen and the aromatase inhibitors) for breast cancer should avoid isoflavone supplements. Given the available evidence, this seems a sensible precaution, particularly because there is no clear clinical evidence that isoflavones are beneficial for menopausal symptoms in these women. The advice to avoid isoflavone supplements is not usually extended to soya foods, although some have argued that available data do not appear to warrant making this distinction. Further study is needed.

Isoflavones + Theophylline

High doses of isoflavones might modestly increase theophylline levels.

Clinical evidence

In a placebo-controlled study in 20 healthy non-smoking subjects, pre-treatment with daidzein 200 mg twice daily for 10 days increased the AUC and maximum level of a single 100-mg dose of theophylline by about 34% and 24%, respectively, and increased the elimination half-life from about 9 hours to about 12 hours.

Experimental evidence

The isoflavones genistein and equol were found to inhibit the cytochrome P450 isoenzyme CYP1A2. Conversely, note that soya-based infant formula induced CYP1A2 in vitro, see Soya + Caffeine.

Mechanism

Daidzein, and some other isoflavones, appear to be moderate inhibitors of cytochrome P450 isoenzyme CYP1A2, of which theophylline is a substrate.

Importance and management

The dose of daidzein used in this study was higher than that usually taken in isoflavone supplements, or as part of the diet, and the effects on theophylline pharmacokinetics were modest. Nevertheless, bear in mind that high doses of isoflavones might modestly increase theophylline levels and that this could be clinically important in patients with theophylline levels already at the higher end of the therapeutic range. Note that an increase in theophylline levels has been seen in a patient given the synthetic isoflavone, ipriflavone.Aminophylline would be expected to interact similarly.

Note also that, conversely, there is evidence that infants receiving formula feeds, (which may include soya-based formula) require higher doses of caffeine (which, like theophylline, is a substrate of CYP1A2) than those that are breastfed, see Soya + Caffeine.