|Home | About | Journals | Submit | Contact Us | Français|
Across multiple studies, obesity is associated with increased risk of higher grade disease and PSA recurrence after radical prostatectomy (RP). Whether these associations vary by race is unknown. We examined the association between obesity and outcome after RP stratified by race.
We performed a retrospective analysis of 1,415 men treated with RP between 1989 and 2008 in the SEARCH database. We examined the association between increased body mass index (BMI) and adverse pathology and biochemical recurrence using multivariate logistic regression and Cox models, respectively. Data were examined stratified by race.
After adjusting for preoperative clinical characteristics, higher BMI was associated higher tumor grade (p=0.008) and positive surgical margins (p<0.001) in white men, with similar but statistically non-significant trends in black men. We noted no significant interaction between race and BMI for associations with adverse pathology (p-interaction≥0.12). After adjusting for preoperative clinical characteristics, higher BMI was associated with increased recurrence risk in both white (p=0.001) and black men (p=0.03). After further adjusting for pathological variables, higher BMI was associated with significantly increased recurrence risk in white (p=0.002) and black men (p=0.01). There was no significant interaction between race and BMI for predicting biochemical progression either adjusting for pre-operative factors (p-interaction=0.35) or pre-op and pathological features (p-interaction=0.47).
Obesity was associated with greater recurrence risk among both black and white men. Obesity does not appear to be more or less influential in one race than another, but rather is a risk factor for aggressive cancer regardless of race.
Obesity is a major public health issue, which is associated with increased risk of death from several cancers including the most common non-cutaneous cancer in men – prostate cancer.1 Even though the specific biologic processes linking obesity to prostate cancer remain to be completely elucidated, previous studies have shown that obesity is associated with higher grade tumors and poorer outcomes after surgery.2–4
Similar to obese men, black men are also at increased risk for prostate cancer death. Specifically, there are marked racial differences in both prostate cancer incidence and mortality with black men having the highest risk in the world.5 They also tend to present with more advanced stages than white men.5 The degree to which this is attributable to socioeconomic reasons like impaired access to health care or due to worse tumor biology is debatable.6 Given that there is a higher prevalence of obesity among black men with prostate cancer,3 the degree to which excess obesity may explain or contribute to these poorer outcomes among black men is unknown. Moreover, whether obesity is correlated with poor prostate cancer outcomes equally for black and white men is likewise unknown.
We propose that obesity holds the same negative consequences regardless of race. To test this hypothesis, we examined the association between body mass index (BMI) and adverse clinical and pathological characteristics and the risk of biochemical progression after radical prostatectomy (RP) in both black and white men. In order to ensure the validity and generalizability of our findings, we used the multi-center multi-ethnic Shared Equal Access Regional Cancer Hospital (SEARCH) Database.7 Of note, we have previously examined the association between obesity and outcome within the SEARCH database and found that obesity was associated with increased progression risk.2 However, these analyses were not stratified by race. Given a recent report which suggested obesity may be a stronger predictor of outcome among black men,8 we re-examined an updated cohort from the SEARCH database to assess this issue.
After obtaining Institutional Review Board approval from each institution to abstract and combine data, we combined data from patients undergoing RP at the Veterans Affairs Medical Centers in West Los Angeles and Palo Alto, California and Augusta, Georgia, and Durham, North Carolina, and Birmingham, Alabama into the SEARCH database.7 This database includes information on patient age at surgery, race, height, weight, clinical stage, grade of cancer on diagnostic biopsies, preoperative PSA, surgical specimen pathology (specimen weight, tumor grade, stage, and surgical margin status), and follow-up PSA data. BMI was calculated as weight in kilograms divided by height in meters squared (kg/m2). Patients treated with preoperative androgen deprivation or radiation therapy were excluded. Of the 2,320 men within the SEARCH Database, we excluded 54 diagnosed from a transurethral resection as this affects PSA, 98 with missing PSA values, 191 with missing biopsy Gleason scores, 306 with missing clinical stage data, 155 with missing BMI data and 101 men who were neither black nor white. This resulted in a study population of 1,415. Biochemical progression was defined as a single PSA > 0.2 ng/ml, two concentrations at 0.2 ng/ml, or secondary treatment for an elevated postoperative PSA. Men who received adjuvant treatment for an undetectable PSA were censored as having not recurred at the time of treatment. The prostatectomy specimens were sectioned per each institution’s protocol.7
We explored differences in the distribution of clinicopathological characteristics across the BMI (kg/m2) groups of normal weight (<25), overweight (25–29.9), obese (30–34.9), and moderately and severely obese (≥35) using analysis of variance for continuous variables or the chi-squared test for categorical variables. The odds ratio (OR) of the following binary pathological outcomes was estimated for BMI categories using logistic regression: high-grade disease (Gleason score ≥7), positive surgical margins, extraprostatic extension, and seminal vesicle invasion. Few men had lymph node metastases (n =18). BMI was entered into all multivariable models as a series of indicator variables for each BMI category. We tested for trend by entering the median BMI of each BMI category as a continuous term into the model and evaluating the coefficient by the Wald test. We adjusted for pre-operative PSA (continuous variable), age at RP (continuous), year of surgery (continuous variable), clinical stage (T2/T3 versus T1c), center (categorical) and biopsy Gleason score (3+4, ≥4+3 versus 2–6). Because the data for pre-operative PSA were not normally distributed, we examined the data after logarithmic transformation. Data was analyzed as a whole as well as separately by race. To test whether the association of obesity with adverse outcome varied by race, we also tested for the interaction between race and BMI strata by including a cross product term in the multivariate analysis.
Time to biochemical progression was compared among the BMI categories using Kaplan-Meier plots and the log-rank test. To estimate the relative risk (RR) of progression associated with obesity, we used a Cox proportional hazards regression model. We ran two separate multivariable Cox models to examine whether obesity was associated with biochemical progression and whether this association was independent of any association between obesity and pathological findings. The first model adjusted for only pre-operative clinical characteristics (age, year of surgery, clinical stage, biopsy Gleason score, center, and pre-operative PSA), while the second model adjusted for pre-operative clinical characteristics (except biopsy Gleason score) and pathological variables (logarithmic transformation of prostate weight, pathological Gleason score, positive surgical margins, extraprostatic extension, seminal vesicle invasion, and lymph node metastasis). Data were examined as a whole and stratified by race. We tested for an interaction between race and BMI strata as described above.
The point estimates for all crude and age-adjusted models for predicting both the binary pathological outcomes and time to progression were similar to the multivariable adjusted point estimates and therefore only results for the multivariable models are shown. The distribution of all clinicopathological variables was similar among the SEARCH sites. Therefore, data from all centers were combined for analyses. All statistical analyses were performed using STATA 10.1 (Stata Corp., College Station, TX).
Black men constituted 47% (n=662) of the patient population. In univariate analysis, there was no association between BMI and race (chi-squared, p=0.49). When stratified by race, higher BMI was associated with more recent year of surgery and younger age in both races, though these associations did not always reach statistical significance (table 1 and and2).2). In white men, higher BMI was associated with lower PSA (p=0.05) and larger prostate size (p=0.001). Though the direction of the association between BMI and PSA and prostate size was in general similar in black men, the association was weaker and not statistically significant. In white men (p=0.02), but not black men (p=0.40), higher BMI was associated with increased prevalence of positive surgical margins. In both white and black men, BMI was not significantly related to tumor grade, clinical stage, extraprostatic extension, seminal vesicle invasion or lymph node involvement.
After adjusting for multiple preoperative clinical characteristics, higher BMI was significantly associated higher pathological tumor grade (p=0.008) and positive surgical margins (p<0.001) in white men (Table 3). Though a similar trend was noted among black men these associations did not achieve statistical significance. Higher BMI was not significantly associated with extraprostatic extension or seminal vesicle invasion in either racial group. In addition, the interaction between race and BMI was not statistically significant for any of these adverse pathological features (p-interaction≥0.12). Since associations between obesity and prostate cancer outcomes may theoretically differ based upon type of surgery performed (i.e. retropubic vs. laparoscopic), we performed a subanalysis in which type of surgery was included in the multivariate analysis. When this was done, none of the results were materially changed.
Among white men, mean and median follow up was 4.0 and 3.3 years during which, 222 men (30%) developed a recurrence. Among black men, mean and median follow up were 3.6 and 2.8 years, during which, 230 men (35%) recurred. Among all men, black race was associated with a 27% increased risk of PSA progression (HR 1.27, 95% CI 1.06–1.53, p=0.01). After adjusting for preoperative clinical characteristics, higher BMI was associated with increased risk of biochemical recurrence in both white (p=0.001, figure 1) and black men (p=0.03, figure 2) (Table 4). After further adjusting for multiple pathological variables, obesity was associated with a significantly increased risk of recurrence in both white (p=0.002) and black men (p=0.01). There was no significant interaction between race and BMI for predicting biochemical progression whether adjusting for only pre-operative factors (p-interaction=0.35) or pre-operative and pathological features (p-interaction=0.47).
Prior studies have shown that obesity is associated with increased risk of adverse pathological features as well as biochemical progression after RP.2–4 Black men are more likely to be morbidly obese compared to white men9 and they present with more aggressive tumors and in general at a more advanced stage.5 However, whether obesity and race interact such that the combination of black race and obesity creates a particularly aggressive cancer is unknown. We examined whether the association between obesity and adverse outcome after RP varied as a function of race. Using a large multi-center equal access population, we found that regardless of race, obesity was associated increased risk of biochemical progression after RP. Thus, obesity appears to be related to aggressive disease, regardless of race.
The prevalence of obesity has steadily been increasing over the last 20 years and currently 30% of US men are obese.10 Its impact on overall health in general is now accepted as a major public health issue. However, the role of obesity in modulating prostate cancer risk and aggressiveness remains hotly debated. Even though recent evidence suggests obesity may be associated with a lower risk of prostate cancer incidence11–13 there is a consensus that obesity is associated with higher risk of prostate cancer mortality.1, 12 Moreover, multiple studies, including prior studies from the SEARCH database have found that higher BMI was associated with increased risk of prostate cancer recurrence after RP.2–4
Black men the United States have the highest incidence and mortality from prostate cancer in the world.5 However, data regarding outcomes after RP are conflicting. Several large multi-institutional cohorts have shown black race to be associated with increased risk of recurrence compared to white men.3, 14 However, data from the SEARCH cohort found that after controlling for baseline differences between the races, including PSA, race was not an independent predictor of recurrence, though black men did have a non-significantly increased recurrence risk.7 Of note, though the current study found no association between race and BMI, many other studies have found that black men undergoing prostatectomy were more likely to be obese.3, 15 Also, black men in the general population are more likely to be morbidly obese.9 Therefore, it remains plausible that obesity may contribute to the more aggressive tumor biology in black men.
While obesity may contribute to poorer outcomes among black men in general, the as yet unanswered question is whether obesity is more influential as a risk factor for aggressive prostate cancer in black men compared to white. A recent study from Spangler et al. found that among black men, obesity was associated with a significant five-fold higher risk of recurrence after RP, while among white men, obesity was associated with only a 44% non-significant increased risk of recurrence.8 However, this study was limited by the inclusion of only 140 black men, of which only 51 were obese. In the current much larger study of 662 black men of which 209 were obese, we did not find obesity was more predictive of recurrence after RP among black men than among white men. Rather, in the current study, obesity portended a poor prognosis, regardless of race. Moreover, the associations between obesity and pathological findings were more or less similar in both races. We did note that among white men, obesity was associated with a significantly higher incidence of high grade tumors and positive surgical margins, whereas within black men, no significant associations between tumor grade and margin status BMI were found. However, the general trends were for more positive margins and more high-grade disease among obese black men, though these did not reach statistical significance. Furthermore, there were no significant interactions between race and BMI predicting adverse pathology suggesting there was no significant difference in the influence of obesity predicting adverse pathological features in black and white men. As such, we are unable to explain the differences between our findings and those of Spangler et al. Whether these stem from differences in patient populations (equal access vs. tertiary care), the overall association between race and obesity (not associated in our study, black men more obese in Spangler et al.), power issues (limited number of obese black men in Spangler et al.) or true differences is unclear and requires further study.
Given the lack of significant interaction between race and obesity for predicting either adverse pathological features or biochemical progression, it is important to understand the definitive mechanisms through which obesity results in aggressive cancer and poor outcome after surgery, regardless of race. Though these mechanisms are yet to be fully elucidated, one potential mechanism is the alteration of sex hormone levels. Obesity is associated with higher levels of estrogen and lower levels of testosterone. Even though the evidence is equivocal on overall risk of development of prostate cancer, lower levels of testosterone may result in the promotion of aggressive cancers.16, 17 Another possible mechanism may be through the insulin-like growth factor (IGF) axis. Obesity is associated with higher levels of free IGF-1,18 and higher IGF-1 levels are positively associated with increased risk of several cancers including prostate cancer.19 Leptin and adiponectin – two cytokines released by the adipose tissue may form yet another postulated biologic link between obesity and prostate cancer. Leptin is increased in obese men and may be associated with increased risk of high grade tumors.20 Conversely, adiponectin levels are inversely related to BMI and may have protective effects.21 However, the evidence on the association between adipokines and prostate cancer risk and aggressiveness is conflicting and further studies are required before their exact role is determined. Ultimately, the exact molecular mechanism(s) linking obesity and aggressive prostate cancer remain poorly characterized and are a fertile area of research.
Our study has several limitations. Even though there were significant trends of obesity being associated with poor outcome regardless of race, the risks did not steadily increase as BMI increased. In fact, among black men, the moderately and severely obese group seemed to have a lower risk of recurrence than the mildly obese group. This may in part be due to the modest numbers in these categories. However, the true importance of this finding is unknown and we await further studies to explore this issue. The median follow up period in this study is relatively short and longer follow up will be needed to confirm these findings. Additionally, we assessed biochemical progression as our primary end point. However, since early biochemical progression is associated with more definite end points like metastatic disease and prostate cancer specific death22 and a principal instrument for post-surgical decision making, it is a reasonable intermediate end-point. This is a retrospective analysis of a RP cohort and hence there is a selection bias for early stage disease and therefore the findings may not be generalizable to more advanced stages or the general population. Finally the pathological evaluation was not centralized and therefore may lead to individual variations in pathological grade and stage reporting, though this should not have been deferential by race or BMI and therefore its influence on the current findings is unclear.
Obesity was associated with greater risk of recurrence after RP among both white and black men. Obesity does not appear to be more or less important in one race than another, but rather is a risk factor for aggressive cancer regardless of race.
Supported by the Department of Veterans Affairs, National Institute of Health R01CA100938 (WJA), NIH Specialized Programs of Research Excellence Grant P50 CA92131-01A1 (WJA), the Georgia Cancer Coalition (MKT), the Department of Defense, Prostate Cancer Research Program, (SJF), and the American Urological Association Foundation/Astellas Rising Star in Urology Award (SJF). Views and opinions of, and endorsements by the author(s) do not reflect those of the US Army or the Department of Defense.