Cancer

Lycopene and Prostate Cancer: No Magic Bullet

By Howell Sasser, PhD

Research Review Coordinator, Manila Consulting, McLean, VA; Adjunct Lecturer, Department of Epidemiology and Community Health, New York Medical College, Valhalla, NY

Dr. Sasser reports no financial relationships relevant to this field of study.

Prostate cancer is the most commonly diagnosed non-dermatologic cancer and the second leading cause of cancer-related death in U.S. men.1 Its etiology and predisposing factors are inadequately characterized, but probably include some combination of older age, north American and northern European residence (perhaps a diet-related issue), African American race, and genetic similarities within families. Oxidative stress is also a common etiologic element suggested for many kinds of cancer. Antioxidant-rich foods and supplements are marketed widely as having protective properties with respect to cancer, but few have rigorous scientific evidence to back up such claims.

Lycopene (one of the carotene family of antioxidants) came to be of interest early in the study of diet and cancer. It is present in meaningful quantities in tomatoes and tomato products, watermelon, pink grapefruit, and red bell peppers, and is known to be taken up by human prostate tissue.2 However, its precise mechanism of anticancer action is still mostly speculative. It might function by scavenging intracellular free radicals, by mediating programmed cell death (apoptosis), by inhibiting specific growth factors, or by some combination of these.2 This uncertainty, combined with the attractive prospect of a chemoprotective agent with virtually no side effects and the general difficulty of assessing the impact of any single factor in so multifactorial a process as cancer, has kept lycopene on the research agenda.

The available published research is divided into observational and experimental studies. The former are usually larger and include most of the evidence for lycopene as contained in whole foods. The latter are smaller, but often have more precisely controlled estimates of exposure (through the use of supplements with known contents). Some studies avoid the exposure measurement issue altogether by using serum lycopene levels. This produces greater efficiency for the researchers, but at the expense of providing less useful information for clinicians.

Observational Evidence

Much of the early evidence for lycopene was in the form of observational studies that looked first at lycopene-rich whole foods and later at supplements containing lycopene.3 A comprehensive meta-analysis of this body of research included both prospective and retrospective studies, and covered a wide range of definitions of lycopene intake. These included raw and cooked tomato products, and dietary and serum lycopene (measured in mg). There was general agreement among the studies included that a serving of raw tomatoes was about 200 g, but there was no similar agreement about cooked tomatoes. Because of this and other definitional issues, intake of both kinds — food and supplements — was divided arbitrarily into quintiles, and then further consolidated into high (the fifth [highest] quintile), moderate (the second and third; or second, third, and fourth quintiles, depending on the study), and low. The common outcome derived from all studies was the “risk” of prostate cancer. This could be estimated directly in the prospective studies because all participants were prostate cancer-free when those studies started. In the retrospective studies, risk was inferred indirectly from the difference in prior experience (of lycopene exposure) between those with and without prostate cancer.

Summing across all studies, there was a small but not statistically significant reduction in the risk of prostate cancer with higher consumption of raw tomatoes (relative risk [RR] = 0.94; 95% confidence interval [CI], 0.88-1.01). When the analysis was limited to the higher-quality evidence of the prospective studies, the effect was more pronounced (RR = 0.78; 95% CI, 0.66-0.92). There appeared to be a small and significant increase in risk with increasing intake of cooked tomatoes (RR = 1.07; 95% CI, 1.06-1.08), but this must be seen in light of the disagreement among studies about units of consumption mentioned above. Dietary lycopene levels showed a similar pattern to raw tomatoes — a nonsignificant reduction in risk when all studies were included, and a more dramatic reduction (RR = 0.38; 95% CI, 0.34-0.42, for a change in lycopene consumption of 12.7 mg/day) in an analysis restricted to prospective studies. Findings were also positive and statistically significant when comparing the highest quintile of intake with the four lower quintiles of consumption.

Since the publication of the first meta-analysis, a few other researchers have reported observational results. For example, one study reported finding no significant association between serum lycopene levels and prostate cancer risk (odds ratio [OR] = 0.98; 95% CI, 0.68-1.43).4 This study, conducted in Australia, had a relatively strong design — a case-control study nested in a larger prospective study — and avoided the complexities of intake quantification by using serum levels, but was small in size. It falls well within the range of prior results and does little to support any therapeutic claim for lycopene.

Kristal et al also conducted a case-control study nested in a clinical trial of a prescription drug for the prevention of prostate cancer.5 This virtually guaranteed them a population free from prostate cancer until a known and recent date, as well as clear evidence of prediagnosis levels of lycopene in the form of frozen blood samples collected just before and just after random assignment in the trial. Those diagnosed as cases during the trial were compared with a sample of trial participants who did not develop prostate cancer. The results showed no association of lycopene with reduced cancer risk (OR for a 10 mg/dL increase in lycopene were 0.99, 1.01, and 1.02 for rising grades of prostate tumors). This study also combines powerful methods with minimal findings.

Agalliu and colleagues examined dietary factors in a subgroup of men in a large Canadian study of cancer.6 A sample of men stratified by age reported what they ate and what supplements they took at baseline using a food-frequency questionnaire. From this, their intakes of a variety of nutrients were calculated. The main outcome was time to a diagnosis of prostate cancer or the end of study follow-up (about 7 years), whichever came sooner. The study’s main objective, to evaluate the joint effect of a group of pro- and anti-oxidant factors, proved to be unsupported statistically. However, there was a statistically significant reduction in the risk of prostate cancer with lycopene consumption under certain circumstances. Specifically, an effect was shown when the cases included in the analysis were restricted to those diagnosed 2 or more years after the start of follow-up. Men consuming higher levels of lycopene (second through fifth quintiles as compared with the first) had hazards (relative risks) of prostate cancer between 0.64 and 0.75. It should be noted, however, that there was no evidence of a trend to greater protection with greater consumption, and, in fact, the reduction in risk appeared to attenuate with rising lycopene levels. This study was large and had the virtue of measuring actual dietary patterns, but produced results that at best require further evidence to confirm.

Finally, Beydoun and colleagues took a slightly different approach, looking for associations between antioxidant intake and values for prostate specific antigen (PSA) commonly used as markers of prostate cancer.7 Dietary and PSA data were abstracted from the records of 3800 men in three waves of the National Health and Nutrition Evaluation Survey (NHANES). Lycopene was divided into quartiles of consumption, and PSA was reported as total PSA > 10 ng/mL, > 4 ng/mL, and > 2.5 ng/mL, as well as free PSA as a proportion of total PSA (< 15% and < 25%). It was found in unadjusted analyses that rising levels of lycopene consumption were associated with a lower chance of being above the > 4 and > 2.5 ng/mL PSA threshold levels (P = 0.007 and 0.033, respectively). However, after adjusting for likely confounding factors such as age, race, education, and body mass index, these associations were no longer statistically significant. This study is interesting because it uses a common screening marker, still in use despite recent controversy, rather than the disease outcome itself, moving the association of interest further “upstream.” However, the balance of observational research still does not give a clear answer about the role of lycopene in prostate cancer prevention.

Experimental Evidence

Numerous experimental studies have focused on the effects of single or repeated doses of lycopene-rich foods (usually tomato products) in preventing DNA damage and other processes known or believed to play a role in the genesis of prostate cancer. Ellinger and colleagues reviewed this literature in 2006 and concluded that there is some evidence to support such a role.8 They found that the bulk of the available evidence relates to lycopene consumed in the form of whole foods rather than in supplements, and characterized the strength of the collected findings (reviewed systematically but not meta-analytically) as, “the regular ingestion of tomatoes or tomato products might prevent” prostate cancer (emphasis added).

Another group reasoned that the antioxidant effects of lycopene might slow the progression of prostate disease that is already established.9 The authors conducted a “Phase II” study in a group of 71 men with histologically proven prostate cancer who had rising PSA levels and were on no other therapies. Participants were randomly assigned to receive 15 mg of lycopene alone (in the form of a “tomato extract” capsule) or lycopene plus a soy isoflavone mixture, twice a day for up to 6 months. All participants showed continued increases in their PSA values, but a larger proportion of those in the lycopene-alone group than in the lycopene plus soy group showed enough slowing in the rate of rise to be classified as having stable disease (95% vs 67%, P = 0.003, though the test result was not reported in the paper).

In another study related to disease progression, Bunker and colleagues moved one step earlier in the disease process, assembling a group of Afro-Caribbean men considered to be at high risk for prostate cancer — those with high-grade prostatic intraepithelial neoplasia (HGPIN) or other cellular changes that often precede the development of cancer.10 The men were randomly assigned to receive 30 mg/day of lycopene plus a multivitamin or the multivitamin only for 4 months. PSA was measured at baseline, and again 1 and 4 months later. The investigators found that PSA declined in both treatment groups between baseline and month 1, but then returned to prestudy levels by month 4. There was no significant difference in PSA between the treatment groups at any measurement.

Mohanty et al and Schwarz et al also considered lycopene’s role in preventing or delaying progression to prostate cancer.11,12 Mohanty began with men with HGPIN. The study was small (n = 40), and the lycopene dose was lower than in most other trials (4 mg/day), but the 1-year duration of follow-up was longer than that of the Bunker study. The results showed a reduction in average serum PSA in the lycopene group and an increase in average PSA in the comparison group (which received no active therapy). Curiously, no statistical tests were reported. Schwarz recruited 40 men with proven benign prostatic hyperplasia (BPH), but who were free from prostate cancer, and randomly assigned them to receive 15 mg/day of lycopene or placebo for 6 months. PSA level, measured at 6 months and 1 year, was the main outcome. The lycopene group showed a significant (P < 0.05) reduction in serum PSA, while the placebo group did not. However, the magnitude of the difference between the two groups over time was not statistically significant.

Dose

It is worth summarizing briefly the forms and doses that have been tested to date. Supplement doses as low as 4 mg/day and as high as 30 mg/day have been tested. The most common dose used for research appears to be 15 mg/day. None of the experimental studies reviewed used whole foods as part of the dosing regimen. An analogous statement about the studies that used whole foods is difficult to make. It can be said that one cup of a concentrated tomato product (like soup or sauce) contains about 25 mg of lycopene, and a serving of fresh tomato contains about 4 mg.13

Conclusion

On balance, there seems to be little in vivo evidence that lycopene produces a clinically meaningful amount of protection against prostate cancer. This appears to be the case regardless of the form in which the lycopene is consumed and regardless of the presence or absence of early cancer precursors. None of the studies identified for this review presented consistent, statistically significant results that withstand adjustment for known or suspected confounding factors.

Caution is required when speculating about why this may be the case, but a few possibilities may be noted. First, it may genuinely be the case that lycopene does not play a role in preventing prostate cancer. With lingering uncertainty as to its mechanism of action, it is difficult to know exactly what to test and how. A related possibility is that lycopene’s main role in disease prevention takes place earlier in the life course. All of the studies reviewed here focused on men in the age range in which prostate cancer typically presents. If lycopene’s function has more to do with inhibiting initiation than promotion of disease, it may be necessary to look for exposure to it among younger men (or further in the past for those currently in the prostate cancer window). This possibility cannot be ruled in or out on the basis of the available literature, though a clear majority of the studies reviewed are based on the assumption that lycopene can and does have value later in the natural history of prostate cancer.

As with many phytochemicals, it is also important to consider dosing and the vehicle of delivery. It is a common observation that constituents of whole foods that appear to have desirable properties often show disappointing results when they are extracted and delivered in isolation. The limited positive evidence for lycopene comes mostly from studies measuring its consumption in whole foods. All of the experimental studies included here used supplements of some kind, although with some variation in the dose delivered. It may be that careful research with prolonged exposure to a controlled dietary exposure (as difficult as that may be) is needed to settle the question definitively.

Recommendation

There appears to be little support for clinicians to recommend or endorse their patients’ use of lycopene as part of a prostate cancer prevention strategy. Although there appears to be virtually no risk of toxicity with lycopene use at conventional levels, and the possibility of benefit cannot be ruled out definitively, it is not possible to say how much benefit there might be or for whom. It is important that the availability and promotion of this and other supplements not distract patients from other strategies, such as PSA screening and digital rectal exams, which can aid in the detection and early treatment of prostate cancer. These are “secondary” forms of prevention (that is, means to detect disease early rather than prevent it from starting), but they are of proven value. Clinicians should also emphasize the value of consuming a wide variety of nutrients in whole foods, since the research seems to show that if there is any benefit to be had from phytochemicals, it is most likely to come in that form.

References

1. American Cancer Society. 2009. Cancer Facts & Figures 2009. Atlanta, GA: American Cancer Society.

2. Hsueh-Li T, et al. Tomato-based food products for prostate cancer prevention: What have we learned? Cancer Metastasis Rev 2010;29:553-568.

3. Etminan M, et al. The role of tomato products and lycopene in the prevention of prostate cancer: A meta-analysis of observational studies. Cancer Epidemiol Biomarkers Prev 2004;13:340-345.

4. Beilby J, et al. Serum levels of folate, lycopene, β-carotene, retinol, and vitamin E and prostate cancer risk. Eur J Clin Nutr 2010;64:1235-1238.

5. Kristal AR, et al. Serum lycopene concentration and prostate cancer risk: Results from the Prostate Cancer Prevention Trial. Cancer Epidemiol Biomarkers Prev 2011;20:638-646.

6. Agalliu I, et al. Oxidative balance score and risk of prostate cancer: Results from a case-cohort study. Cancer Epid 2011;35:353-361.

7. Beydoun HA, et al. Association of serum vitamin A and carotenoid levels with markers of prostate cancer detection among US men. Cancer Causes Control 2011;22:1483-1495.

8. Ellinger S, et al. Tomatoes, tomato products and lycopene in the prevention and treatment of prostate cancer: Do we have the evidence from intervention studies? Curr Opin Clin Nutr Metab Care 2006;9:722-727.

9. Vaishampayan U, et al. Lycopene and soy isoflavones in the treatment of prostate cancer. Nutr Cancer 2007;59:1-7.

10. Bunker CH, et al. A randomized trial of lycopene supplementation in Tobago men with high prostate cancer risk. Nutr Cancer 2007;57:130-137.

11. Mohanty NK, et al. Lycopene as a chemopreventive agent in the treatment of high-grade prostate intraepithelial neoplasia. Urol Onc 2005;23:383-385.

12. Schwarz S, et al. Lycopene inhibits disease progression in patients with benign prostate hyperplasia. J Nutr 2008;138:49-53.

13. Cancer dietitician. Available at: www.cancerdietitian.com. Accessed August 20, 2012.