Review Article

Prostate cancer: Altering the natural history by dietary changes [PDF]

ASEEM R. SRIVASTAVA, D. DALELA

ABSTRACT
The importance of diet on the development and progression of prostate cancer was initially suggested by epidemiological studies. Since then, there has been a vast amount of research in this field. Compelling evidence now provides hope that evidence-based dietary alterations may markedly alter the natural history of this disease. Is there enough evidence for clinicians to be able to advise dietary modifications? The preliminary results no doubt are encouraging, but at present there seems to be no evidence to justify the widespread use of these proposed dietary interventions. However, as public awareness increases, all physicians involved with the care of patients with cancer of the prostate will need to be better armed with the current updates and advice on this issue.

Natl Med J India 2004;17:248–53

INTRODUCTION
The incidence and mortality rates of prostate cancer differ greatly among countries and ethnic groups. At any given age, men all over the world share the same risk for latent cancer, whereas the incidence of clinically manifest prostate cancer differs dramatically,1 with mortality rates as low as 4 in 100 000 in Japan to about 20 per 100 000 in the USA. So, if American and Japanese men start with the same risk for latent prostate cancer, what accounts for the difference in the rates of clinical cancer? Why do these latent cancers progress to an aggressive form in so many American men and not in Japanese men? What are the factors that transform a latent tumour into a life-threatening one? Is it genetic susceptibility? Or is it because of environmental influences?
    Shimizu et al.2 have shown that the incidence of prostate cancer in Japanese immigrants to the USA increases in one or two generations to about half that of the indigenous North American population. Migrants have the same genetic make-up as their families in their native countries, and thus this study suggests that there must be an environmental influence that causes the change in risk, possibly diet.

Prostate cancer prevention and tumour biology
Malignancies most likely to benefit from preventive measures must have an early event and a slow progression to malignant transformation. In these types of tumours, interactions between environmental and genetic influences are the greatest and, therefore, are potentially benefited by a preventive strategy. The development of invasive prostate cancer is a culmination of a complex series of initiating and promotional events, under the influence of many genetic, hormonal, dietary and environmental factors over a period of several decades. Prostate cancer may thus be an ideal solid tumour to evaluate preventive strategy.3 The PTEN or MMAC1 gene located on chromosome 10 regulates cellular motility and matrix interactions and is associated with advanced prostate cancer.4 Other tumour suppressor genes implicated include p53 (present in 25%–75% of patients with prostate cancer)5 the retinoblastoma gene on chromosome 13, and KA 11 on chromosome 11.
    In untreated primary prostate cancer, expression of another tumour suppressor gene p16 is reduced.6 Oncogenes that may be activated in prostate cancer include c-myc, bcl-2, c-met and ras. Because many risk factors such as age, race and family history are beyond an individual’s control, it is plausible that altering the diet (if it has a major influence) may minimize the risk of developing prostate cancer. One such plausible link (between diet and prostate cancer) seems to be oxidative stress,7 another could be the insulin-like growth factor (IGF-1) system. The ability of IGF-1 to activate androgen receptors even in the absence of androgens suggests its possible role in the progression of prostate cancer.8 IGF-1 also has mitogenic and anti-apoptotic effects on prostate epithelial cells.8,9 While high levels of IGF-1 are associated with prostate cancer, low levels are found in patients whose diets have been supplemented (see lycopene).
    A typical Asian diet is protective against prostate cancer (with India likely to be no exception). However, with a western diet gaining increasing acceptance, the incidence of clinical prostate cancer may begin to rise. Most studies in this field are preliminary and often conflicting, but since there is no curative therapy for advanced prostate cancer, identification of dietary risk factors may help in the development of possible preventive strategies in the future.

Role of serum biomarkers
In view of the long natural history of prostate cancer, if the incidence of cancer acts as the only end-point measured in trials, prolonged periods of observation will be necessary to obtain results. Biomarkers of carcinogenesis, tailored to the agent under investigation, are therefore essential. Examples of such biomarkers are prostate specific antigen (PSA) (serum levels of which often correlate with tumour volume), DNA adduct levels such as 8-oxo-dG or M1G, which can be measured in prostate tissue or in white blood cells as a surrogate for the prostate, to detect antioxidant changes elicited by dietary supplementation. Similarly, the soluble p105 component of p185erbB-2 proto-oncogene may also be monitored as a serum biomarker.

METHODS
Data were gathered from a wide range of sources, including a variety of electronic databases (Pubmed, Medscape), peer-reviewed publications and technical reports. Original publications were reviewed wherever possible to properly interpret the data and associated experimental methods. This was especially important where there was an apparent conflict in the published literature.

FATS
Studies demonstrate a direct relation between a country’s prostate cancer-specific mortality rate and average total calories from fat consumed by the country’s population,10,11 possibly resulting from prolonged exposure to androgens under these circumstances. In the USA, where there is a high intake of fat, the mortality rate from prostate cancer is high, whereas in Japan, which has one of the lowest fat consumptions in the world, mortality from the disease is low.12
    Men with a high fat consumption are not only more likely to develop prostate cancer but are also more likely to develop a more aggressive form of the disease.13 Also the risk of its progressing to an advanced stage is higher among men with a high fat intake.14
    Mice model studies show that a low fat diet is associated with slower tumour growth and this effect is probably mediated through the modulation of the IGF axis.15 Omega-6 fatty acids (derived from linolenic acid) stimulate the growth of prostate cancer cell lines, whereas omega-3 fatty acids such as docosahexaenoic acid and eicosapentaenoic acid, inhibit the growth of these cells.16 A reduced risk of developing prostate cancer has also been seen among men with higher eicosapentaenoic acid levels in their erythrocytes,17 which is likely to be due to its effect on prostaglandin synthesis. Similarly, in a follow up of men in the Health Professionals Follow-up Study,18 a notable finding was that eating fish more than three times a week is associated with a decreased risk of prostate cancer, with the strongest association for metastatic cancer. Intake of marine fatty acids from other sources showed a similar but weaker association.
    However, not all investigators have uniformly reached these conclusions. A rat model study19 of sex hormone-induced cancers, to examine the effects of high-fat diet on the incidence/latency of prostate cancer, revealed no difference in the pattern of carcinogenesis and weight of the prostate between rats on a high fat diet and controls on a standard, low fat diet. This study does not support the role of dietary fat in promoting sex hormone-induced prostate cancer. Similarly, in a randomized controlled trial,20 the intervention group received intensive counselling to consume a diet low in fat and high in fibre, fruits and vegetables, and the control group received a standard brochure on a healthy diet. There was no difference in the PSA slopes and in the frequency of elevated PSA values for those with an elevated PSA at baseline. The incidence of prostate cancer was also similar among the two groups. This study offers no evidence that dietary modifications over a 4-year period with reduced fat and increased fruits, vegetables and fibre affect the incidence of prostate cancer and that diet has an impact on serum PSA levels in men.
    There are suggestions that decreasing the saturated fat content in the diet of men with prostate cancer may reduce the chances of actually dying from the disease and potentially prolong life.21 However, a review of published studies on the relationship between dietary fat and prostate cancer risk reveals that approximately half these studies found an increased risk with increased dietary fat while the remaining half found no association.22

VITAMIN D
Calcitriol is a steroid hormone obtained by dietary means or synthesized in the body on exposure to UV light. In 1990, Schwartz and Hulka23 proposed that a low level of circulating calcitriol is a risk factor for prostate cancer. Since then, many investigators have attempted to explain the regional/racial risk factors of prostate cancer by this hypothesis. A higher mortality from prostate cancer in the northern latitudes of the USA has been linked with a lower exposure to UV rays and hence lower calcitriol levels.24 Similarly, racial pigmentation has been incriminated in the higher incidence of prostate cancer among people of African descent.25
    The normal prostate gland may be a vitamin D target organ;26 receptors for vitamin D are widely expressed in the normal prostate tissue and more so in the peripheral zone.27 Calcitriol inhibits the growth of normal prostate epithelium28 and has an antiproliferative effect on human prostate cancer,29 initially presumed to be due to vitamin D-induced cell cycle arrest. However, studies now propose the induction of apoptosis as an additional mechanism.30
    To date, there have been two clinical trials on calcitriol in the treatment of human prostate cancer. In one, 7 patients of early recurrent prostate cancer were treated with calcitriol. The results showed a significant slowing of the rate of PSA increase in 6 patients.31 Unfortunately, results from the other trial, in which 14 patients with hormone-refractory metastatic disease were treated, failed to demonstrate any response.32 More importantly, in both these studies, the dose of calcitriol was limited by hypercalcaemia and hypercalciuria.

SOY
The protective role of soy in prostate cancer was postulated from studies that indicated that men in Southeast Asia, who consume 20–50 times more soy daily than American men, have a 10-fold lower incidence of clinically significant prostate cancer and prostate cancer-related deaths.33,34 Soy contains large amounts of isoflavones, genistein, daidzein and their glycosides, and has been implicated in the prevention of prostate cancer, possibly via a mild oestrogenic effect.
    The initial evidence that genistein inhibits human prostate growth was provided by studies on the histoculture of minced, surgically resected prostate tissue from men with benign prostatic hypertrophy (BPH).35 In these studies, it caused inhibition of DNA synthesis. Genistein inhibits the growth of rat prostate adenocarcinoma cells in vitro as well as in Lobund–Wistar rats when injected subcutaneously.36 In these susceptible rats, soy proteins decrease the number of prostatic tumours induced by testosterone37 and protect against the development of chemically induced prostate cancer.38
    In contrast to genistein, daidzein (another major soy isoflavone) is only a weak inhibitor of tumour cell growth in vitro and has no inhibitory effect on the growth of DU-145 or LNCaP cells.39 However, recent reports suggest that an active metabolite of daidzein (equol) may have potent antiproliferative effects on benign and malignant prostatic epithelial cells.40
    Although most studies suggest a protective role of soy proteins against prostate cancer, some show no effect and one suggests a paradoxically increased growth. Urban et al.41 in a double-blind study found that short term exposure of elderly men with elevated serum PSA values to soy protein containing isoflavones decreases serum cholesterol but not the serum biomarkers PSA and p105erbB-2.
    Results from another study show that neither an isoflavone-rich soy protein isolate (SPI), nor conjugated linoleic acid (CLA) inhibit the in vivo growth and development of prostate tumour cells when administered in the diet of male Copenhagen rats inoculated with androgen independent R–3327–AT–1 rat prostate tunel cells. Moreover, at the highest concentrations of SPI and CLA there was a significant increase in tumour volume compared with controls. Administration of 10% SPI in the diet also enhanced tumour growth, whereas 5% SPI exerted no measurable effect.42 Therefore, the authors have cautioned that isoflavone-rich soy protein isolates should not be used in human studies involving advanced prostate cancer until further data on their safety are available.

SELENIUM
There is increasing evidence that selenium, an essential trace element,43,44 may have protective properties against prostate cancer. Selenium is a constituent of the enzyme glutathione peroxidase, one of the body’s antioxidants and, since oxidant damage has been linked to many cancers, investigators suggest that the anticancer benefits of selenium may arise from its antioxidant function. In cell culture, it reduces the effect of a number of described mutagens45–47 and may alter the metabolism of other carcinogens.48–50 Other potential actions suggested include effects on the immune and endocrine systems, production of cytotoxic selenium metabolites, initiation of apoptosis, inhibition of protein synthesis, as well as inhibition of specific enzymes.51–54 There are sufficient data to suggest that supplementation with selenium decreases the risk of many chemically induced cancers,55–61 spontaneous animal tumours,62 and transplanted animal tumour lines.63 Studies of geographical areas with varying dietary selenium content demonstrate an inverse relationship between selenium intake and cancer risk.64,65 Epidemiological studies have shown mixed results, with significant and inverse relationships encountered in some studies,66–77 while others have not encountered a higher risk in patients with low selenium levels or a low selenium intake.78–83 Salonen et al. studied serum samples collected from 111 subjects, obtained from the Hypertension Detection Follow up Programme, who developed cancer during the subsequent 5 years and compared them with serum samples from 210 cancer-free subjects matched for age, race, sex and smoking history. The mean serum selenium level was lower in cancer cases than in controls and an association between low selenium level and cancer was strongest for gastrointestinal and prostate cancer.66
    Similarly, a multi-institutional study designed to prevent skin cancer randomized a group of 1312 patients with a history of basal cell or squamous cell carcinoma of the skin to either 200 mg selenium supplementation per day or placebo.84 The baseline serum PSA levels in both arms were also measured. After an average follow up of 6.4 years, selenium-treated patients developed only about one-third as many prostate tumours as patients receiving placebo. Importantly, no patient experienced toxicity due to selenosis, a side-effect that has been reported in association with chronic intake of selenium above 5 ppm.85 Although previous studies suggest that selenium and vitamin E (alone or in combination) may reduce the risk of developing prostate cancer, only a large clinical trial can confirm these initial findings.

LYCOPENE
Carotenoids are encountered in a number of vegetables, notably tomatoes, and are best absorbed if cooked with fats or oils. The most abundant carotenoid in tomato is lycopene, followed by phytoene, phytofluene, zeta-carotene, gamma-carotene, beta-carotene, neurosporene and lutein. These have been postulated to have anticancer properties, attributable to their antioxidant effect.
    Earlier studies evaluating the links between lycopene and prostate cancer risk generally did not find an association, with only one study showing a reduced risk.86–89
    In 1995, an analysis of the Physicians’ Health Study90 revealed a one-third reduction in prostate cancer risk in men with the highest consumption of tomato products, attributed to the lycopene in these vegetables. Several in vitro and in vivo studies since then suggest the protective effects of lycopene on specific cancers, including prostate cancer.91,92
    These reports prompted further studies, the results of which have been mixed.93,94 A prospective study examining the relationship between the plasma concentration of several antioxidants and the risk for prostate cancer reveals that lycopene is the only antioxidant found in significantly lower levels in patients with prostate cancer than in matched controls.95 A similar study found that lycopene levels in prostatic tissue are significantly lower in patients with cancer than in controls. However, serum and tissue levels of beta-carotene and other major carotenoids were not different between the two groups.96 This is surprising, as dietary studies on rat models show a significant decrease in the risk of developing prostate cancer in rats fed on tomato powder compared with lycopene alone.97 Thus it is possible that the influence of tomato on the development of prostate cancer may relate not only to lycopene but to other phytochemicals also. However, in epidemiological studies, beta-carotene has not been found to be associated with prostate cancer risk, or is even associated with an increased risk.98
    From the available literature, it appears that the IGF-1 (a potent mitogen on prostate epithelial cells) axis may play a central role in the progression of prostate cancer. While high levels of IGF-1 have been associated with prostate cancer, low levels are found in those with increased tomato intake.99 Furthermore, lycopene has been shown to inhibit IGF-1 induced proliferation of several tumour lines. Kucuk et al. randomized 33 men with prostate cancer and assigned them to consume lycopene or no supplement for 30 days before prostatectomy; postoperatively, in the lycopene-supplemented group, PSA levels fell by 20% and cancer had spread in only 33%. However, in 75% of the control group, the cancer had spread and their PSA levels remained unchanged.100 In a more recent study,101 32 patients with localized prostate cancer consumed tomato sauce-based pasta dishes before radical prostatectomy. The mean serum PSA concentrations and leucocyte 80HdG decreased after tomato sauce consumption, and the resected tissues had a lower prostate 80HdG and a higher apoptotic index compared with controls.
    A review of epidemiological studies of tomatoes, lycopene and prostate cancer risk102 shows that of the 15 studies only 5 support a 30%–40% reduction in risk associated with high tomato or lycopene consumption, whereas 3 suggest a 30% reduction in risk, but the results were not statistically significant. Seven studies do not support any association. The largest relevant dietary study found that consumption of 2–4 servings of tomato sauce per week was associated with a 35% reduction in the risk of prostate cancer and a 50% reduction of advanced prostate cancer. In the largest serum sample-based study, a similar reduction in risk was observed for all and advanced prostate cancer for a high versus low concentration of lycopene.
    Most of these reports do suggest an association between a lower prostate cancer risk and high lycopene intake. However, this evidence is weak because previous studies were not controlled for total vegetable intake (i.e. separating the effect of tomatoes from vegetables in general), dietary intake instruments are poorly able to quantify lycopene intake, and other potential biases.103
    More important, follow up for a sufficient period of time is required so that other end-points are also captured, as supplementation of the diet with higher than physiological doses of micro-nutrients has at times caused unexpected and unwanted results, e.g. an 18% increase in lung cancers observed in the beta-carotene arm of the ATBC trial.54
    The final recommendations therefore await further research.

FLAXSEED
Flaxseed is the richest plant source of omega-3 fatty acids and is high in dietary fibre. As previously mentioned, omega-3 fatty acids and dietary fibre may protect against cancer. A diet supplemented with flaxseed has been shown to inhibit the growth and development of prostate cancer in a transgenic adenocarcinoma mouse prostate (TRAMP) model.104 However, the quantity of flaxseed given to each mouse was 5% of its total food intake, which would probably be too high to be realistic for a human diet.

WINE
The finding that people in Mediterranean countries, with a high intake of legumes, olive oil, wine and vegetables, have a much lower incidence of prostate cancer and that polyphenols in green tea may induce apoptosis in human prostate cancer cell lines led to the hypothesis that polyphenols in red wine may have a similar effect. Romero et al.105 studied 5 polyphenol constituents of red wine and found that they all inhibited the growth of LNCaP cells at different concentrations and induced apoptosis.

HERBAL THERAPIES
PC-SPES (PC stands for prostate cancer, SPES is Latin for hope), a mixture of 7 medicinal herbs and saw palmetto, was introduced in the USA as a dietary supplement in 1996. Each capsule contains 320 mg of the herbal combination powder with an unknown ratio of each herbal extract.106Early studies suggested that PC-SPES was effective in reducing the levels of PSA in both hormone-responsive and -resistant prostate cancer patients. The clinical activity and adverse effects of PC-SPES are strikingly similar to high-dose oestrogen (diethyl stilboestrol)107,108 and it is not known if PC-SPES provides a significant additive benefit.
    A chemical analysis of PC-SPES has shown its contamination with warfarin, diethyl stilboestrol and indomethacin. Diethyl stilboestrol and indomethacin are known to have anticancer properties. Later preparations of PC-SPES, which contained less diethyl stilboestrol and indomethacin, showed a 6-fold lower anticancer potency.109

SUMMARY
Although there are too many hypotheses on this subject, there are little hard and conclusive data. From the published literature, it is clear that diet does have a role to play in the development and progression of prostate cancer. However, clear and convincing recommendations will have to await the results of long term prospective clinical trials.110 Presently, there seems to be little or no evidence to support an advise beyond a healthy diet that is low in fat, and rich in fruits, vegetables and fibre.
    However, continuing research in this field may lead to an improved understanding of the mechanism by which diet influences prostate glad oncogenesis and may provide means to lower the risk of the disease.