Dietary Supplements in Patients With Cancer: Risks and Key Concepts, Part 2

Laura Boehnke Michaud, Julie Phillips Karpinski, Kellie L. Jones and Janet Espirito


Am J Health Syst Pharm. 2007;64(5):467-480. 

In This Article

Agents with Hormonal Properties

Another pharmacologic category of dietary supplements that may pose a risk to patients with cancer includes agents with hormonal properties ( Table 5 ). Much like the antioxidants, this diverse group of products has very complex mechanisms of action, many of which have yet to be fully elucidated. The potential risks associated with these agents may be obvious or subtle, and their level of risk depends on the type of cancer in question.

Certain types of cancer appear to rely heavily on hormonal influences. These include breast, prostate, endometrial, and ovarian cancers. Breast, endometrial, and ovarian cancers are under the control of estrogen and, to a lesser degree, progesterone. However, the relationship between these two hormones in these diseases is not well understood. Prostate cancer relies on testosterone as a driving force in its growth and development. However, not all breast, endometrial, ovarian, and prostate cancers are under hormonal control. Hormone sensitivity is determined in all breast tumors at diagnosis. This entails the identification and quantification of the estrogen receptor (ER) and progesterone receptor (PR). Up to two thirds of breast cancer patients will have ER- or PR-positive disease or both, and the incidence increases with age.[114] This represents a large population of patients who will receive some type of hormonal therapy at some point during cancer treatment. The lack of these receptors indicates a low likelihood of responding favorably to endocrine therapy for breast cancer.[115] This fact is often stated when debating the use of hormonal supplements in patients who have hormone-receptor negative breast cancer. However, there are some data to indicate that sampling errors for determination of ER and PR status may occur; however, breast cancers, regardless of hormone receptor status, are generally considered to be under hormonal control. This topic is controversial and encompasses postmenopausal hormone-replacement therapy (HRT) as well as hormonal dietary supplements.

While hormone receptors are not routinely measured in endometrial and ovarian cancer tissue, hormone therapy remains a therapeutic option for these women. The mainstay of therapy for endometrial cancer is surgery, but a multidisciplinary approach combining radiotherapy with chemotherapy or hormone therapy is often used. Treatment of ovarian cancer primarily involves surgery or chemotherapy or both; however, hormone therapy is given to patients with recurrent disease.

Many conventional cancer therapies used to treat hormonally responsive malignancies alter hormone levels (e.g., aromatase inhibitors, leutenizing hormone-releasing hormone [LHRH] agonists) or have hormonal effects (e.g., SERMs). Table 6 and Table 7 list the mechanisms of conventional hormonal therapy for breast and prostate cancers, respectively. Many herbs and other natural products possess intrinsic hormonal properties, functioning similarly to mammalian hormones and conventional cancer therapies. Supplements may alter normal physiological levels of hormones (e.g., estrogen, progesterone, testosterone, leutenizing hormone [LH], follicle-stimulating hormone [FSH]), potentially altering the activities of these hormones or interfering with cancer therapies. Supplements may also affect hormone receptors or other biological targets, which may be detrimental to patients with cancer or counter-act cancer therapies. The complex nature of interactions between these conventional therapies and supplements with hormonal properties makes predicting clinical outcomes rather difficult when relying solely on available endocrine data. The mechanisms of pharmacologic activity of dietary supplements are often demonstrated through animal or cell-line experiments, making it difficult to extrapolate the results to humans. If clinical data are available, they are usually from healthy subjects, making extrapolation to cancer patients difficult. Anecdotal information may offer clinical clues about a product’s effects. For instance, anecdotal reports may indicate that a supplement affects menses or menopausal symptoms in women or increases sexual desire in men. If these claims are true, this agent may possess some hormonal properties. However, it is important to note that a placebo effect or other mechanisms may be responsible for these clinical outcomes, and more stringent scientific evidence is required to prove causality. Anecdotal reports are often the only information available about a dietary supplement’s potential hormonal effects, and patients should be told that such supplements are "potentially harmful" if they have a hormonally sensitive malignancy.

Many patients with breast, endometrial, or ovarian cancer turn to these types of supplements as options for "natural" hormone replacement therapy (HRT), but they may also use these therapies to help with menopausal symptoms associated with their cancer therapy. The potential for drug interactions between "natural" hormones to treat these symptoms and the hormone therapies to treat their cancer may pose serious risks to patients, potentially negating the effects of their cancer therapy. An extensive review of all dietary supplements with hormonal effects is beyond the scope of this article; however, the major issues related to phytoestrogens with breast, endometrial, ovarian, and prostate cancers are reviewed. The goal of this discussion is to introduce topics related to hormonal supplements to provide a template with which to evaluate other products of interest not mentioned here.

For cancer patients and clinicians, the use of phytoestrogens is perhaps one of the most controversial and fiercely debated topics related to dietary supplements. Phytoestrogens are diphenol compounds that comprise one of the largest groups of "natural" hormonal supplements consumed today.[116] These agents are categorized according to their chemical structures, with isoflavones, lignans, coumestans, and resorcyclic acid lactones comprising the four major classes. Isoflavones are the most abundant compounds in commonly consumed foods, with 230 types identified. Three common isoflavone aglycones—genistein, daidzein, and glycitein—are present in most dietary sources. Coumestrol, another isoflavone, is present in soybeans but in much lower concentrations than genistein, daidzein, and glycitein. Isoflavones also exist as glucosides, acetyl glucosides, and malonyl glucosides. These differing forms of isoflavones determine absorption, with aglycone forms more readily absorbed than the corresponding glucoside conjugates. Glucoside derivatives of isoflavones undergo hydrolization in the large intestine to their corresponding aglycone, which is readily absorbed. These aglycones are then metabolized by the liver to glucorinic acid or sulfate conjugates and enterohepatically circulated through the gut, with potential for further metabolism and reabsorption. The glucuronide fraction of these isoflavone aglycones represents up to 90% of circulating isoflavones in both rats and humans and is considered biologically inactive. Free and sulfated forms, while in lower concentrations, are considered biologically active. Variances in the pharmacokinetics of isoflavones are apparent and based on age, sex, and ethnicity. Details surrounding these differences are beyond the scope of this article but are important indicators of the complex nature of phytoestrogens and represent how individual differences in liver and gut metabolism may play vital roles in determining the activity of these compounds and the potential risks or benefits.

The most significant dietary sources of phytoestrogens include soybeans, clover, alfalfa sprouts, and oilseeds (e.g., flaxseed).[117] Components of these plants vary with geographic location, soil type, year, and environmental conditions of growth.[116] Processing also influences the amount and form of isoflavones in soy products. Processing soybeans for soy-containing food products increases hydrolization to the aglycone forms. Fermenting soybeans to produce products such as tempeh and natto partially hydrolyzes isoflavone glucosides to the aglycone form.[116] The U.S. Department of Agriculture and Iowa State University developed an isoflavone database in 1999 to use as a reference for determining the isoflavone values for many foods. This database expresses all values as aglycones and provides an accurate and comparable estimate of isoflavone content for many different foods.[118] It is important to recognize that eliminating all phytoestrogens from the diet is impossible, but limiting soy-containing foods and supplements is feasible. The effects of supplementation with synthetic sources of phytoestrogens are just beginning to be studied. All of these factors may account for varying results in observational and clinical trials and should be considered when evaluating studies of phytoestrogenic supplements.

The estrogenic activity of phytoestrogens was first observed in Australian ewes. It was found that some animals suffered reproductive disorders, resulting in permanent infertility after eating red clover. Genistein and daidzein, constituents of red clover, were later identified as the agents most likely to have caused this effect in the animals.[117,119] Estrogenic activity has been demonstrated by many soy-containing foods, as well as the individual components discussed above. In vitro and in vivo studies have demonstrated hormonal activity through ER-binding assays. The binding affinity of phytoestrogens to the estrogen receptor has been compared with 17ß-estradiol, illustrating their weaker estrogenic potential (potential in descending order: 17ß-estradiol, coumestrol, genistein, equol, daidzein, and biochanin A).[120] These experiments did not distinguish between agonistic and antagonistic activity after receptor binding or binding differences between the α and ß isoforms of ER. Therefore, corresponding clinical outcomes related to these biochemical activities are uncertain.

Many other mechanisms of biological activity have been demonstrated with various phytoestrogens. Genestein, for example, has demonstrated both proliferative and anti-proliferative effects, depending on the concentrations studied. At low concentrations, genestein stimulates the growth of breast cancer cells; at high concentrations, growth inhibition is seen.[121] Antiestrogenic effects have also been observed with increased concentrations of other phytoestrogens.[122] Mechanisms other than ER binding may also elicit hormonal effects. Daidzein has been shown to inhibit aromatase and 5-α-reductase, while other phytoestrogens have been proposed to decrease estrogen levels via stimulation of sex hormone binding globulin.[117,123] Other hypothesized mechanisms of biological activity include inactivation of tyrosine kinase, inhibition of epidermal growth factor, and inhibition of type II topoisomerase.[119] Corresponding biological outcomes related to these activities have yet to be determined. However, these effects would seem to be beneficial for breast, endometrial, ovarian and prostate cancers, potentially inhibiting cancer cell growth through similar mechanisms as prescription drugs administered as cancer therapy. Evidence supporting or refuting these biological effects in cancer patients has yet to be reported.

In humans, phytoestrogens have been shown to alter endocrine function in premenopausal, perimenopausal, and postmenopausal women. Studies investigating the effects of different forms of isoflavone supplementation have found conflicting results regarding changes in menstrual cycle, hormone levels, and other measures of endocrine function (e.g., changes in vaginal epithelium). Many studies investigating the endocrine effects of phytoestrogens in healthy women have been of short duration and are not designed to determine the presence (or absence) of clinical benefit based on their findings. This information could be extrapolated to individual patient scenarios, but caution should be used when making these assumptions, as these are complex interactions.

Breast cancer. The effects of phytoestrogens on breast tissue are not well understood. Early epidemiologic studies appeared to indicate an inverse association between soy consumption and breast cancer incidence;[124,125,126,127,128,129] however, other studies have failed to confirm these results.[130,131,132,133] This result may be due to study design, a true lack of effect, or variability in other factors related to phytoestrogen metabolism or processing. Studies investigating the timing of phytoestrogen exposure appear to indicate that consumption early in life, especially during breast development, may indeed be protective against breast cancer. This may be due to enhanced mammary gland development, leading to fewer terminal end buds, which are the most vulnerable to carcinogenesis. These inferences are based on animal studies and have yet to be substantiated in humans. However, a small retrospective study conducted in China to evaluate soy food intake during childhood and adolescence found an inverse association with adult breast cancer risk.[129] These results appeared to be significant, regardless of menopausal status. It is, therefore, postulated that early exposure to a diet high in phytoestrogens may be beneficial, but supplementation or increases in dietary phytoestrogens later in life may offer no established benefit on breast tissue.

Effects of phytoestrogens on breast tumors have been studied in cell lines and animal models. Conflicting reports exist, demonstrating both protective and procarcinogenic effects with phytoestrogens.[134,135,136,137,138] A study investigating the effects of dietary genistein in combination with tamoxifen in an athymic nude mouse model demonstrated a decrease in tumor growth inhibition with tamoxifen when given in combination with the phytoestrogen.[139] At least two human studies have demonstrated a stimulatory effect on breast tissue with soy supplementation.[140,141] All of these data raise concerns about phytoestrogen supplementation in women with a history of breast cancer, regardless of whether they are currently receiving active hormonal therapy.

Endometrial and ovarian cancers. The effects of phytoestrogens on the endometrium are even less well understood. Unopposed estrogen replacement therapy has long been known to increase the risk of endometrial hyperplasia and cancer. This effect is not evident when either continual or cyclic progesterone therapy is added to the replacement regimen. Epidemiologic evidence seems to support the notion that increased consumption of phytoestrogens decreases the risk of endometrial cancer.[142,143] However, in vitro data have shown that high concentrations of phytoestrogens induce stromal cell proliferation to nearly the same degree as estradiol. In the presence of estradiol, phytoestrogens antagonize the proliferative effects of estradiol by 10–20%. This demonstrates the SERM-like qualities of phytoestrogens on the endometrium.[144] A recent randomized, double-blind, placebo-controlled clinical trial investigating the use of a soy tablet (150 mg of isoflavone daily) in 376 postmenopausal women found 6 patients (4%) with endometrial hyperplasia, compared with none of the patients receiving placebo.[145] The phytoestrogen supplement was administered for five years, and the cases of hyperplasia were diagnosed at five years through a scheduled endometrial biopsy that was part of the study. Another study investigating the rates of endometrial hyperplasia with conventional HRT also reported rates as high as 3% in the placebo group of the trial.[146] Therefore, the overall risk of endometrial cancer with phytoestrogen supplementation seems rather low, but one case of endometrial cancer has been reported in a woman who was ingesting many herbs and vitamins with known phytoestrogenic components.[146] This information would lead to the conclusion that phytoestrogens should be avoided in patients with endometrial cancer. No data regarding the effects of phytoestrogens on ovaries or ovarian cancer could be found, making it difficult to ascertain the risks associated with phytoestrogen supplementation in ovarian cancer.

Other potential uses. In terms of efficacy, there is an abundance of information related to the efficacy of phytoestrogens for a number of physiological or pathophysiological states. Physiological changes in women related to phytoestrogen consumption have been reported relative to their use as "natural" HRT (in postmenopausal women) or as a cancer prevention strategy (for all women), primarily focusing on breast cancer prevention.

The effects on menopausal symptoms, primarily vasomotor symptoms (hot flashes), have been a major focus of research in this arena as well. It is prudent to recognize the importance of a placebo effect (nearly 20–30% reduction in vasomotor symptoms) in these studies. Soy supplementation through foods,[147] soy protein isolate,[148,149] or soy extracts[150,151,152] appears to add an additional 10–20% improvement above that seen with placebo. However, conflicting evidence from other studies demonstrate no benefit over placebo with a soy protein isolate,[153,154] a soy beverage supplement,[155] or a soy extract.[156] Also, two studies of red clover extracts have been published and failed to find a benefit over placebo for the management of hot flashes.[157,158] The soy products used in these studies had differing amounts of isoflavone content, which may have led to discrepancies in the outcomes. Again, these varying results could be due to individual differences in metabolism or the source of phytoestrogens used. As little as 30 mg daily of soy isoflavones (intact with soy protein or as a semipurified extract) may reduce the frequency of hot flashes. However, the placebo effect should be taken into account, and each individual woman must evaluate the relatively small additional benefit of soy phytoestrogen used to treat hot flashes.[159] Of greater concern, no safety information exists regarding this type of supplementation in women with a history of breast, endometrial, or ovarian cancer. One of the above-mentioned studies included breast cancer patients, but neither cancer status nor disease recurrence rates were mentioned.[156] Most of these studies addressing hot flashes were of short duration, and the overall effect of continued supplementation on cancer recurrence is unknown.

Other proposed benefits of HRT include positive effects on cardiovascular disease and osteoporosis. While even large, prospective, randomized, placebo-controlled trials investigating prescription HRT have failed to substantiate epidemiologic and case–control data relating to these positive effects, investigations have commenced with "natural" HRT. In 1999, the Food and Drug Administration (FDA) approved the health claim "Diets low in saturated fat and cholesterol that include 25 g of soy protein/day may reduce the risk of heart disease."[160] This approval was based on data from numerous studies indicating that an average of 47 g daily of soy protein lowers total and low-density-lipoprotein cholesterol and produces a trend toward increased high-density-lipoprotein cholesterol.[161] Since this approval, studies have questioned whether isoflavones are responsible for the lipid-lowering effects of soy. Isoflavones have been shown to participate in this effect but must be consumed with other components of soy protein to see a benefit.[152,162,163,164,165,166,167,168,169,170,171] Isoflavone extracts may have other effects on the cardiovascular system that may be beneficial (e.g., improved arterial compliance), but these are just beginning to be investigated.[171,172]

The effects of phytoestrogens on bone health are not as strongly supported by human evidence. Animal studies investigating effects on bone mineral density and bone turnover have been conducted, but little human evidence is available to substantiate these claims.[173,174,175] Reports from clinical trials appear mixed, demonstrating mild benefit or no effect.[144] Studies suggest a short-term benefit with soy protein supplementation, but clinical outcomes (e.g., fracture rates) related to these benefits have not been demonstrated. Further studies are required to establish whether these effects are sustained over a prolonged period and whether they translate into decreased fracture rates and an overall improvement in postmenopausal health. Also, the question of whether bone loss can be prevented with soy supplementation has not been answered. Again, women included in these studies did not have a history of breast, endometrial, or ovarian cancer; hence, the effects on cancer recurrence with soy protein supplementation are not known. Therefore, risk–benefit analyses have not adequately been accomplished and safety concerns still exist.

Based on this information, recommendations for individual patients are very complex. For patients faced with a myriad of menopause- or cancer-treatment-related symptoms, the question of whether to use a "natural" hormonal preparation is fraught with anxiety and confusion. Communication surrounding this issue should be sensitive to these anxieties and attempt to fully explain the controversial nature of the evidence available to date. Discussion of the conflicting evidence presented previously and the fact that there are still many unanswered questions is paramount to ensure that patients make informed decisions. Generally, recommendations include an emphasis on moderation and stress the lack of information with synthetic or concentrated forms of phytoestrogens. As stated previously, lower concentrations of phytoestrogens appear to be estrogenic and stimulate cancer cell growth in vitro. However, it is important to reiterate that it is impossible to eliminate phytoestrogens from the diet. For food sources with high phytoestrogenic concentrations or high isoflavone content (e.g., soy milk, tofu), no more than one serving of these products per day is recommended. This recommendation is based on other published recommendations that will be discussed later.

The normal growth and differentiation of the prostate require the presence of androgens, specifically testosterone.[176] The most physiologically active form of testosterone is dihydrotestosterone (DHT). Conversion of testosterone to the more active moiety occurs through a reaction catalyzed by the enzyme 5-α-reductase. Testosterone production is regulated through the release of LHRH, which interacts with the pituitary gland to release FSH and LH, ultimately regulating the level of testosterone through a negative feedback mechanism. LHRH agonists have been developed to target this pathway, activating the negative feedback mechanism and decreasing serum testosterone levels. Similar outcomes can be achieved through surgery (orchiectomy) or with other types of drugs, such as ketoconazole ( Table 7 ). Directly targeting prostate cancer cells may also be accomplished with antiandrogens (e.g., flutamide, bicalutamide) that bind to the androgen receptor and prevent DHT from acting and stimulating growth. Another mechanism to target prostate cancer is to block the conversion to the active form of testosterone (DHT) through 5-α-reductase inhibition. This mechanism has not proven effective as prostate cancer treatment but may play a role in prostate cancer prevention.

Men in the general population commonly take supplements for symptoms of prostatitis, treatment of benign prostatic hypertrophy, or symptoms of sexual dysfunction. Others use these products for the treatment and prevention of prostate cancer. Some dietary supplements may effectively treat these problems but may also pose a risk for prostate cancer patients, rendering their cancer treatment ineffective or stimulating cancer cell growth.

Phytoestrogen supplementation is also extremely controversial for the treatment of prostate cancer yet may prove to be more beneficial than for patients with breast, endometrial, or ovarian cancer. Epidemiologic and laboratory evidence seems to support the idea that phytoestrogens are protective against prostate cancer. Asian men have been shown to have a decreased incidence of prostate cancer, which is believed to be a result of a diet high in phytoestrogens.[120] The full mechanism of protection has not been clearly elucidated, but most studies suggest a relationship to the inhibition of 5-α-reductase in genital skin fibroblasts and prostate tissue.[177] ER-ß receptors have been identified in the prostate, but expression of this protein is lost in malignant prostate cells and prostate cancer tissues. Some phytoestrogens may alter expression of ER-receptor proteins over a man’s life span, but the clinical outcomes of these biological events are unknown.[178] A study using a mouse xenograft model for prostate cancer incorporated increasing concentrations of soy protein into the animals’ diet to determine the effects on prostate tumor growth.[179] Diets with the highest concentrations of phytoestrogens were associated with tumor growth inhibition, while diets containing lower levels of phytoestrogens were associated with reduced tumor growth compared with diets without supplementation. Tumor growth inhibition in this study was associated with apoptotic mechanisms and decreased angiogenesis. Whether the mechanisms are active and these concentrations of phytoestrogens achievable in humans has yet to be determined. Urban and colleagues[180] failed to find any effect on prostate specific antigen (PSA) in men with baseline elevations in PSA who were undergoing supplementation with a soy protein beverage. This was a small study and does not eliminate the potential for beneficial effects on the prostate by similar phytoestrogenic compounds. Until more evidence is available, phytoestrogenic supplements should be used with caution in men with a history of prostate cancer and their use should be accurately recorded for any patient, particularly patients who are participating in clinical trials. Dietary sources of phytoestrogens may also pose some risk, but eliminating all dietary sources of phytoestrogens is impossible. Therefore, moderation would still be a prudent course of action based on other published recommendations.

There are no easy answers regarding the use of dietary supplements with hormonal properties in people with a history of cancer. Hopefully, the concerns outlined above are adequate for clinicians to establish a template with which to evaluate these types of compounds. Due to the lack of conclusive evidence regarding both safety and efficacy, patients should be counseled on the theoretical risks and potential benefits and need to recognize that our understanding of this subject is lacking. Overall, it is important to review each patient’s disease and the goals of cancer therapy and supplement therapy while addressing these concerns. Moderation would appear to be a prudent course of action related to phytoestrogens, limiting the amount of soy and avoiding the use of synthetic or concentrated forms of phytoestrogens. Some countries have established maximum daily allowances for phytoestrogens based on the available scientific evidence. Much of the widespread concern surrounds thyroid alterations found in adolescents who received soy-based infant formulas[181] and decreases in the number of proliferating cells in the intestine of piglets fed soy-based formulas compared with cow’s-milk formulas.[182] The French have issued a public statement recommending a maximum daily intake of 1 mg/kg of phytoestrogens.[144] According to Sirtori and colleagues,[144] the Italian Health Authority released a public statement recommending a daily intake of phytoestrogens as dietary supplements below 80 mg, which represents approximately 1 mg of phytoestrogens per kilogram of body weight. Utilizing the isoflavone database mentioned previously,[118] foods and supplements can be categorized based on their isoflavone components, and patients can be counseled on the appropriate amounts of foods and supplements to intake to stay within these limits. Working closely with a dietitian who specializes in this type of education and counseling is very helpful. As a more generalized recommendation for cancer patients, clinicians may recommend no more than one serving of soy per day with a well-balanced diet that includes fruits, vegetables, and grains. After reviewing all of the important information, patients must decide whether to take such supplements.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.