Obesity and prostate cancer: a growing problem.

Obesity is an epidemic in American society with approximately one in three adults in the United States being considered obese (1). This compares to ‘‘only’’ one in six adults being considered obese 20 years ago (Fig. 1; ref. 1). The exact reasons for this dramatic increase in obesity prevalence in the last 20 years likely include alterations in diet and amount of physical activity. Obesity is associated with numerous chronic medical problems including coronary artery disease, hypertension, and diabetes (2). In a landmark study, Calle et al. from the American Cancer Society enrolled over 900,000 adults who were free of cancer at the time of study enrollment in 1982 (3). The study participants were then followed for 16 years to assess for risk of death from cancer. The authors found that higher body mass index (BMI) was positively associated with risk of death from 12 different types of cancer among men, including prostate cancer. Category I obese men (BMI, 30.0-34.9 kg/m) were 20% more likely to die from prostate cancer than normal weight men (BMI, 18.524.9 kg/m), whereas men who were category II obese (BMI, 35.0-39.9 kg/m) were 34% more likely to die from prostate cancer. Similar observations have been made in multiple other prospective cohort studies (4, 5), including a very similar study reported by Rodriguez et al. also from the American Cancer Society, which enrolled over 400,000 men in 1959 and followed them through to 1972 (4). Collectively, these studies have enrolled over 1,000,000 men in prospective cohort studies and found relatively strong evidence to suggest that obese men are at increased risk for prostate cancer death. In terms of early-stage disease, several reports, including the study by Strom et al. in this issue, have consistently suggested that men with a higher BMI are at increased risk for biochemical progression following radical prostatectomy (6–9), although the association did not reach statistical significance in all studies (8, 9). Why is obesity associated with more aggressive prostate cancer (e.g., greater risk of prostate cancer death and greater risk of progression after surgery)? Undoubtedly, the reasons are complex and multifactorial. Therefore, rather than trying to cover every possible explanation for this apparent association, I have selected just a few leading hypotheses, keeping in mind that these hypotheses are not mutually exclusive. One hypothesis is that the way we screen for and treat prostate cancer is biased against obese men. Viewed alternatively, it is possible that it may be harder to find a cancer in an obese man and even when we find the cancer the adiposity may make treatment of the cancer more difficult. First, anecdotally it is harder to perform a proper digital rectal examination in an obese man and thus it may be possible to miss a cancer. Second, obese men have larger prostate sizes making it harder to feel a cancer and harder to find the cancer at the time of biopsy (10). The analogy that comes to mind is, if doing a prostate biopsy looking for cancer is like searching for a needle in a haystack, then the larger the haystack, the harder to find the needle. Third, despite larger prostate sizes, obese men may have lower serum prostatespecific antigen (PSA) values, presumably due to lower testosterone and higher estradiol levels among obese men (11, 12). Lower PSA concentrations would make obese men less likely to have an abnormal PSA test and undergo biopsy resulting in fewer cancers detected, such that the cancers that were detected would be more advanced. The combination of these three factors may result in obese men presenting at a later stage of disease than nonobese men. In addition, there is the added bias that surgery is more difficult to do among obese men resulting in a greater risk of a positive surgical margins (6) and capsular incision (inadvertent incision into the prostate at the time of radical prostatectomy; ref. 13), suggesting that surgery may be less efficacious among obese men. Ultimately, the net result of these biases would be for obese men to present with more advanced disease and even adjusting for this to have a worse outcome after radical prostatectomy. It is important to note that this ‘‘nonbiological’’ detection bias hypothesis is in part based upon the association between obesity and larger prostate size and lower serum PSA values, both of which are biological phenomena. Moreover, the positive association between obesity and risk of death from prostate cancer, as noted by Rodriguez et al., was observed in men from the 1950s and 1960s, long before PSA screening (4). In addition, during that time, not only were patients rarely diagnosed early enough to justify surgery, but surgery itself was rarely done due to excess morbidity. Thus, detection bias issues related to PSA or less efficacious surgery cannot solely explain the association between obesity and risk of prostate cancer death. Instead, some truly ‘‘biological’’ explanation is warranted to completely understand the association between obesity and aggressive prostate cancer. Various ‘‘biological’’ explanations have been given including