This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Kristal, A. R. Search for Related Content PubMed PubMed Citation Articles by Kristal, A. R.

Diet and Trend in Prostate-Specific Antigen: Inferences for Prostate Cancer Risk

Fred Hutchinson Cancer Research Center, Seattle, WA

THE LIST OF PURPORTED dietary risk factors for prostate cancer has become quite long. Only a decade ago, fat and red meat were considered the principal dietary factors affecting prostate cancer risk, but now these factors include (but are not limited to) total energy, calcium, lycopene, selenium, soy, vitamin E, brassica vegetables and -linolenic acid. Many scientists doubt that more of the same types of observational studies used to identify these risk factors, especially when based on recall of usual diet, will provide the clarity necessary to understand whether and how dietary factors affect prostate cancer risk. Instead, advances in our understanding of diet and prostate carcinogenesis are more likely to come from studies based on biologically-based measures of dietary exposures and, when sufficiently well motivated, by randomized clinical trials. In this light, the very interesting and provocative report by Shike et al,¹ based on a secondary analysis from the Polyp Prevention Trial, is a unique contribution to a complex and frustratingly inconsistent literature.

In this report, Shike et al¹ conclude that adopting a diet low in fat and high in fiber, fruits, and vegetables has no effect on 4-year trends in serum prostate-specific antigen (PSA) concentration. To my knowledge, this is the first published manuscript to examine associations of diet with trends in PSA over time, and certainly this is the only large study to examine the question of diet and PSA trends using an experimental design. The results of this study are robust, given the carefully collected data and the use of a randomized design rather than self-report as the means of assessing dietary exposures. However, it is important to carefully consider the interpretation of this study’s results, focusing on three questions: (1) What diet-related hypotheses are being tested by the low-fat, high-fiber intervention? (2) What is the interpretation of modest changes in PSA over a 4-year period? and (3) Are the results of this study inconsistent with the literature on diet and prostate cancer risk?

This study’s simple experimental design, a contrast between the low-fat and high-fiber intervention and the control arms, leads to a very complex interpretation when based on nutrient exposures and their metabolic consequences. For example, two consequences of markedly reduced fat consumption are decreased intake of vitamin E and lower absorption of lycopene, and both of these antioxidants have been associated with reduced prostate cancer risk.^2,3 And, though there were modest changes in weight attributable to the intervention, we do not know how or whether the dietary intervention affected serum androgens or insulin-like growth factors, which may be the mediators of at least some of the effects of diet on prostate carcinogenesis.^4,5 It may well be that dietary prevention of prostate cancer is not addressed by a low-fat and high-fiber food pattern, whose positive and negative consequences offset each other and provide no overall benefit.

The role of PSA as a marker of prostate cancer is not entirely straightforward. PSA is produced by both benign and malignant prostate tissues. Although prostate cancer tissue produces more PSA than benign tissue, benign prostatic hyperplasia (a very common disease of older men) can contribute between 10 and 20 g of additional tissue to the prostate.⁶ Very little is known about diet and benign prostate growth, but it is possible that the dietary factors that influence benign and malignant growth differ. Thus, we should be duly cautious when interpreting trends in PSA without long-term follow-up for both benign and malignant disease.

Lastly, I do not believe that the findings of this study are inconsistent with the scientific literature on diet and prostate cancer risk. One generalization from the epidemiologic literature is that associations of diet with prostate cancer risk tend to be stronger for regional and distant disease (stages C and D) compared with local disease (stages A and B).^7,8 Indeed, some studies that have stratified analyses of diet and prostate cancer risk by stage of disease have found associations only for late-stage disease.⁹ My interpretation of this generalization is that many dietary factors, including fat and calcium, seem to promote the development of invasive disease while having little or no effect on earlier stages of carcinogenesis. This is consistent with international studies, which report that rates of autopsy-detected, latent prostate cancer are similar in countries with vastly different rates of prostate cancer mortality. If we hypothesize that dietary factors are not associated with risk of early-stage disease, then it is not altogether surprising that a low-fat intervention had no effect on trends in PSA in a healthy population.

Despite the complexity in interpreting the results of this study, I believe that the conclusion that a low-fat and high-fiber dietary pattern does not affect prostate cancer risk over a 4-year period is sound. However, we should remember that this study does not address whether a low-fat and high-fiber dietary pattern affects very early processes in carcinogenesis, before a detectable increase in PSA would occur, nor does it address whether dietary change affects the development of invasive disease. I suspect that further analyses of data from this study could address many outstanding questions about diet and the etiology of prostate cancer. For example, hypotheses that relate trends in PSA to weight change, serum fatty acids, and serum lycopene could be readily addressed. And, as mentioned earlier, it would be interesting to examine whether baseline levels or changes in serum steroid hormones and insulin-related growth factors affect PSA trends. This report demonstrates the value of randomized clinical trials to test whether dietary change can reduce cancer risk. These trials are difficult and expensive, and their results are often not consistent with hypotheses that were based on observational studies. Nevertheless, the unanticipated results of clinical trials have been scientifically challenging and have often led to a new understanding of diet and cancer risk.

REFERENCES

1. Shike M, Latkany L, Riedel E, et al: Lack of effect of a low-fat, high-fruit, -vegetable, and -fiber diet on serum prostate-specific antigen of men without prostate cancer: Results from a randomized trial. J Clin Oncol 20: 3592-3598, 2002[Abstract/Free Full Text]

2. Heinonen OP, Albanes D, Virtamo J, et al: Prostate cancer and supplementation with -tocopherol and ß-carotene: Incidence and mortality in a controlled trial. J Natl Cancer Inst 90: 440-446, 1998[Abstract/Free Full Text]

3. Giovannucci E, Clinton S: Tomatoes, lycopene, and prostate cancer. Proc Soc Exp Biol Med 218: 129-139, 1998[Medline]

4. Gann PH, Hennekens CH, Ma J, et al: Prospective study of sex hormone levels and risk of prostate cancer. J Natl Cancer Inst 88: 1118-1126, 1996[Abstract/Free Full Text]

5. Chan J, Stampfer M, Giovannucci E, et al: Plasma insulin-like growth factor 1 and prostate cancer risk: A prospective study. Science 279: 563-566, 1998[Abstract/Free Full Text]

6. Ellis W, Brawer M: The role of tumor markers in the diagnosis and treatment of prostate cancer, in Lepor H, Lawson R (eds): Prostate Diseases. Philadelphia, PA, W.B. Saunders, 1993, pp 276-292

7. Giovannucci E, Rimm EB, Colditz GA, et al: A prospective study of dietary fat and risk of prostate cancer. J Natl Cancer Inst 85: 1571-1579, 1993[Abstract/Free Full Text]

8. Kolonel LN, Hankin JH, Whittemore AS, et al: Vegetables, fruits, legumes and prostate cancer: A multiethnic case-control study. Cancer Epidemiol Biomarkers Prev 9: 795-804, 2000[Abstract/Free Full Text]

9. Kristal AR, Cohen J, Qu P, et al: Associations of energy, fat, calcium and vitamin D with prostate cancer risk. Cancer Epidemiol Biomarkers Prev (in press)

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Kristal, A. R. Search for Related Content PubMed PubMed Citation Articles by Kristal, A. R.