Baseline prostate inflammation is associated with a reduced risk of prostate cancer in men undergoing repeat prostate biopsy: Results from the REDUCE study
ClinicalTrials.gov Identifier: NCT00056407.
The current study was performed to evaluate whether baseline acute and chronic prostate inflammation among men with an initial negative biopsy for prostate cancer (PCa) increased the risk of subsequent PCa detection in a clinical trial with systematic biopsies.
A retrospective analysis was performed of 6238 men aged 50 years to 75 years with prostate-specific antigen levels between 2.5 ng/mL and 10 ng/mL and a prior negative biopsy in the REduction by DUtasteride of PCa Events study who completed a 2-year biopsy. PCa, acute prostate inflammation, and chronic prostate inflammation were assessed by central review. The association between inflammation in baseline prostate biopsies and positive 2-year and 4-year repeat biopsies was evaluated with the chi-square test and logistic regression analysis adjusting for baseline covariates.
Acute and chronic inflammation and both were detected in 46 baseline biopsies (1%), 3931 baseline biopsies (63%), and 892 baseline biopsies (14%), respectively. Acute and chronic inflammation were found to be significantly associated with each other (P < .001). Acute inflammation at baseline biopsy was associated with younger age, lower prostate-specific antigen levels, and a smaller prostate (all P < .01), whereas chronic inflammation was associated with older age and larger prostate glands (all P < 0.01). At the 2-year biopsy, the prevalence of PCa was 14% (N = 900 patients). On univariable and multivariable analysis, both acute and chronic inflammation were found to be significantly associated with a lower PCa risk (acute univariable: odds ratio [OR], 0.65 [P < .001] and multivariable: OR, 0.75 [P = .012] and chronic univariable: OR, 0.61 [P < .001] and multivariable: OR, 0.65 [P < .001]). At the time of 4-year biopsy, only acute inflammation was found to be associated with a lower PCa risk.
Baseline acute and chronic inflammation were both found to be independently associated with a lower PCa risk. From a clinical standpoint, inflammation in negative biopsies for PCa may lower the risk of subsequent PCa detection. Cancer 2014;120:190–196. © 2013 American Cancer Society.
Inflammation is known to be associated with several cancers and in some cases it is directly linked to carcinogenesis. Since 1863, when Rudolf Virchow noted leukocytes in neoplastic tissues and made a connection between inflammation and cancer, several inflammation-related tumors have been described. For example, inflammatory bowel disease has been correlated with colorectal carcinogenesis. In addition, certain infections accompanied by a chronic inflammatory reaction such as viral hepatitis and schistosomiasis are associated with an increased risk of liver and bladder cancers, respectively. However, the association between inflammation and prostate cancer (PCa) is controversial. There is evidence suggesting that antiinflammatory agents such as aspirin may reduce the incidence of PCa.[2, 3] In addition, multiple studies have shown an association between clinical prostatitis and risk of PCa.[4-7] For example, a meta-analysis of 11 studies published through 2000 revealed a 60% increased risk of PCa in men with a history of prostatitis. Furthermore, a questionnaire-based study of 68,675 men demonstrated an increased incidence of PCa in subjects with a self-reported history of prostatitis and/or sexually transmitted diseases. However, other studies failed to demonstrate such an association, which suggests the possibility of recall and/or detection bias given that men with urinary symptoms are more likely to be screened for PCa.[8, 9]
Notably, histological inflammation is frequently present in prostate biopsies, radical prostatectomy specimens, and tissue resected for the treatment of benign prostatic hyperplasia. However, to the best of our knowledge, the clinical significance of asymptomatic prostatic inflammation in negative biopsies for PCa is not completely known. Some small studies have suggested that histological prostate inflammation is associated with a lower risk of cancer after repeat prostate biopsies.[11, 12] However, because prostate inflammation can elevate prostate-specific antigen (PSA) levels and men are typically selected for repeat prostate biopsy based on their PSA level, it is possible that men with inflammation are selected for repeat biopsy more often than those without it, thus having more opportunities for cancer detection. Alternatively, these men with prostate inflammation and an elevated PSA level are likely to have a lower risk of cancer when compared with men with similar PSA levels and no inflammation given that the reason for the elevated PSA is inflammation as opposed to cancer. Consequently, the ideal design with which to evaluate the association between prostate inflammation and cancer involves an approach in which men are selected for rebiopsy regardless of their PSA levels. The REduction by DUtasteride of PCa Events (REDUCE) trial is a 4-year, placebo-controlled study evaluating the use of daily dutasteride to reduce the risk of biopsy-detectable PCa among men with a negative baseline prostate biopsy. Repeat prostate biopsies were performed at 2 years and 4 years regardless of PSA values. Thus, REDUCE uniquely allows for the examination of the relation between inflammatory changes on baseline-negative biopsies and the subsequent risk of PCa detection independent of the influence of changes in PSA levels. Therefore, we evaluated whether baseline acute and chronic prostate inflammation as well as inflammation severity among men with an initial negative biopsy for PCa were associated with the risk of subsequent PCa detection among men in the REDUCE trial.
MATERIALS AND METHODS
The design of the REDUCE study was published previously. In brief, eligible men were those aged 50 years to 75 years who had a serum PSA level ≥ 2.5 ng/mL or 3.0 ng/mL according to age (50 years-60 years and 60 years-75 years, respectively) but ≤10 ng/mL, and a single negative prostate biopsy (6 to 12 core needle biopsy specimens) within 6 months of enrollment. Men were excluded if they had history of PCa, high-grade intraepithelial neoplasia, atypical small acinar proliferation, a prostate volume (PV) > 80 mL, had undergone previous prostate surgery, or had an International Prostate Symptom Score ≥ 25 or ≥ 20 while receiving α-blockers. Medical history was collected at baseline. All men were randomized in a double-blind fashion to receive either dutasteride at a dose of 0.5 mg or placebo orally daily and be followed every 6 months for 4 years. To maintain the blinded nature of the study, PSA levels in the men treated with dutasteride were doubled (given that dutasteride reduces PSA levels by approximately 50%) and randomly adjusted by 0.1 ng/mL so that the final reported values were equally even and odd. PV was measured by ultrasonography at the time of randomization and 2 years and 4 years later. Ten-core, transrectal, ultrasound-guided biopsies were performed as part of the protocol at 2 years and 4 years regardless of changes in PSA. Baseline biopsies had been performed before the start of the study (and independently of the study) and were reread centrally. Biopsies that were performed as part of the study were also read centrally. The central pathology laboratory had no access to the randomization codes. PCa and acute and chronic prostate inflammation were coded as present or absent. Inflammation was also coded based on severity as mild, moderate, and marked based on the average cell density and the extent of tissue involvement in each biopsy core specimen. Chronic inflammation consisted mainly of lymphocytes and a variable number of plasma cells and macrophages. Acute inflammation consisted of a neutrophilic infiltrate. Mild inflammation was defined as small scattered or patchy aggregates and the presence of nests of inflammatory cells, usually no more than 10 to 15 cells per nest. Moderate inflammation was characterized by larger aggregates and nests, usually > 15 cells, and was invariably multifocal. Moderate inflammation was noticeable at low magnification. Sheets of inflammatory cells or extensive multifocal confluent masses of cells were noted in severe inflammation, which was obvious and noticeable at any magnification. Additional criterion for severe acute inflammation was tissue destruction of any size. The protocol was approved by the Institutional Review Board at each research site, and all participants provided written informed consent. Of the 8231 men enrolled in the study, 6316 had a per-protocol 2-year prostate biopsy. The details of the men who underwent at least 1 on-study biopsy have been published previously. Of these men, 78 (1%) were excluded because of missing data regarding 1 of the covariates analyzed. This resulted in a final study sample of 6238 men. Of these men, a total of 4574 (73.3%) also had undergone a per-protocol 4-year prostate biopsy and were included in the 4-year biopsy analysis.
Univariable comparisons of baseline characteristics between men with baseline prostate inflammation were performed using the chi-square test for categorical data and the Student t test for continuous variables. The association between acute and chronic inflammation in baseline prostate biopsies with positive 2-year and 4-year repeat biopsy was evaluated using the chi-square test in univariable analysis. To test the multivariable association between baseline inflammation and positive 2-year and 4-year repeat biopsy results, we used logistic regression controlling for pre-repeat biopsy data including age (continuous, in years), race (white, black, Asian, American Hispanic, or other), body mass index (BMI; continuous and log-transformed, in kg/m2), digital rectal examination (DRE; coded as normal or abnormal), PV (continuous and log-transformed, in cm3), PSA (continuous, in ng/mL), and treatment arm (dutasteride or placebo). All covariates were determined before the repeat biopsy was performed. We also performed 2 sensitivity analyses: the first stratified our sample according to baseline PSA (2.5-4.0 ng/mL, 4.1-7.0 ng/mL, and 7.0-10.0 ng/mL) and the second was restricted to the placebo arm only. The assumptions of linearity of the relationship between the log of the dependent variable and covariates, independence of the errors, homoscedasticity of the errors, and normality of the error distribution were verified for all models. All statistical analyses were performed using Stata 10.1 statistical software (StataCorp, College Station, Tex). A P value < .05 was considered to indicate statistical significance.
The mean baseline age, BMI, PSA, and PV of the study cohort were 62 years, 27.5 kg/m2, 4.3 ng/dL, and 45.6 cm,3 respectively. A total of 5721 men (91.7%) were white. An abnormal DRE was identified in 236 men (3.8%). At baseline biopsy, acute and chronic inflammation were detected in 938 cases (15%) and 4823 cases (77%), respectively. A total of 892 baseline biopsies (14%) showed both acute and chronic inflammation. The presence of baseline acute and chronic inflammation was found to be significantly associated with each other (P < .001). Men with acute inflammation at the time of baseline biopsy were significantly younger, had lower PSA levels, and had smaller prostates (all P < .01), whereas men with chronic inflammation were significantly older and had larger prostate glands (all P < .01). Both types of inflammation were found to be unrelated to race, BMI, DRE, and treatment arm (Table 1).
Table 1. Baseline Patient Characteristics by Chronic Prostate Inflammation
|No. (%)||938 (15.0)||5300 (85.0)||–||4823 (77.3)||1415 (22.7)||—|
|Mean age (SD), y||62.2 (6.0)||62.8 (6.0)||.001||62.9 (6.0)||62.3 (6.1)||.005|
|Race, no. (%)|| || ||.162|| || ||.162|
|White||857 (91.4)||4864 (91.8)|| ||4404 (91.3)||1317 (93.1)|| |
|Black||13 (1.4)||91 (1.7)|| ||85 (1.8)||19 (1.4)|| |
|Asian||36 (3.8)||214 (4.0)|| ||200 (4.1)||50 (3.5)|| |
|American Hispanic||24 (2.6)||78 (1.5)|| ||82 (1.7)||20 (1.4)|| |
|Other||8 (0.8)||53 (1.0)|| ||52 (1.1)||9 (0.6)|| |
|Mean BMI (SD), kg/m2||27.5 (4.1)||27.5 (3.9)||.595||27.5 (3.9)||27.4 (3.8)||.499|
|DRE, no. (%)|| || ||.517|| || ||.576|
|Normal||906 (96.6)||5096 (96.2)|| ||4637 (96.1)||1365 (96.5)|| |
|Abnormal||32 (3.4)||204 (3.8)|| ||186 (3.9)||50 (3.5)|| |
|Mean PSA (SD), ng/mL||3.97 (2.7)||4.33 (2.9)||<.001||4.25 (2.8)||4.37 (3.1)||.193|
|Mean PV (SD), cm3||43.5 (21.7)||46.0 (21.5)||.001||46.5 (22.1)||42.6 (19.2)||<.001|
|Randomization arm, no. (%)|| || ||.410|| || ||.331|
|Placebo||485 (51.7)||2663 (50.2)|| ||2450 (50.8)||698 (49.3)|| |
|Dutasteride||453 (48.3)||2637 (49.8)|| ||2373 (49.2)||717 (50.7)|| |
The number of positive-per-protocol, 2-year repeat prostate biopsies was 900 (14%). Of the 4574 men (73.3%) who underwent a per-protocol 4-year repeat prostate biopsy, 440 (10%) of these biopsies were positive for PCa. At the 2-year biopsy, both acute and chronic inflammation were found to be significantly associated with a lower risk of PCa on univariable analysis (odds ratio [OR], 0.65 [P < .001] and OR, 0.61 [P < .001], respectively). Baseline acute prostate inflammation was associated with an absolute decrease in the risk of cancer detection at the 2-year repeat biopsy of nearly 5% (from 15.3% for those without acute inflammation to 10.4% for those with acute inflammation). Similarly, baseline chronic inflammation was associated with an absolute decrease in the risk of cancer detection at the 2-year repeat biopsy of 6.7% (from 19.7% for those without chronic inflammation to 13.0% for those with acute inflammation). On multivariable analysis, both acute (OR, 0.75; P = .012) and chronic (OR, 0.65; P < .001) inflammation were found to be independently associated with a lower risk of PCa (Table 2). However, at the 4-year biopsy, only acute inflammation was found to be significantly associated with a lower risk of PCa on both univariable (OR, 0.69; P = .017) and multivariable (OR, 0.71; P = .031) analyses. Chronic inflammation was not found to be associated with positive biopsy on either univariable (OR, 0.90; P = .392) or multivariable (OR, 0.98; P = .870) analysis at the 4-year repeat biopsy.
Table 2. Association Between Baseline Acute and Chronic Prostate Inflammation With 2-Year and 4-Year Repeat Prostate Biopsy Positivity
|Acute|| || || || || || || || || || |
|Absent||15.3% (812/5326)||Reference||–||Reference||–||10.1% (389/3868)||Reference||–||Reference||–|
|Present||10.4% (98/943)||0.654 (0.524–0.817)||<.001||0.745 (0.592–0.938)||.012||7.3% (54/741)||0.694 (0.515–0.936)||.017||0.713 (0.525–0.969)||.031|
|Chronic|| || || || || || || || || || |
|Absent||19.7% (280/1422)||Reference||–||Reference||–||10.5% (102/972)||Reference||–||Reference||–|
|Present||13.0% (630/4847)||0.606 (0.518–0.708)||<.001||0.651 (0.553–0.766)||<.001||9.4% (341/3637)||0.902 (0.713–1.142)||.392||0.980 (0.769–1.250)||.870|
The analysis of inflammation severity was not possible for moderate acute inflammation or marked acute and chronic inflammation due to low numbers. Nevertheless, mild baseline acute inflammation was associated with a lower risk of positive 2-year and 4-year repeat prostate biopsy (all P < .05 except for multivariable analysis at the 4-year biopsy, at which P = .055). Mild and moderate baseline chronic inflammation were associated with a lower risk of a positive 2-year repeat biopsy (all P < .05) and were found to be unrelated to the risk of cancer detection in the 4-year prostate biopsy (Table 3).
Table 3. Association Between Baseline Inflammation Severity and 2-Year and 4-Year Repeat Prostate Biopsy Positivity
|Acute|| || || || || || || || || |
|Mild||920 (14.7)||0.645 (0.515–0.809)||<.001||0.720 (0.566–0.915)||.007||0.709 (0.526–0.956)||.024||0.735 (0.537–1.007)||.055|
|Chronic|| || || || || || || || || |
|Mild||4297 (68.9)||0.608 (0.519–0.712)||<.001||0.645 (0.547–0.760)||<.001||0.918 (0.724–1.165)||.483||0.982 (0.769–1.254)||.883|
|Moderate||516 (8.3)||0.589 (0.441–0.789)||<.001||0.724 (0.529–0.989)||.042||0.790 (0.521–1.199)||.269||0.976 (0.629–1.515)||.914|
Because some studies have suggested that histological prostate inflammation is associated with elevated PSA levels and men are selected for prostate biopsy based on elevated PSA levels, it is conceivable that these men with inflammation potentially had a lower risk of PCa because the elevation of their baseline PSA was due to inflammation rather than cancer. Therefore, we performed a sensitivity analysis stratifying our sample according to baseline PSA (2.5-4.0 ng/mL, 4.1-7.0 ng/mL, and 7.0-10.0 ng/mL). In this analysis, both acute and chronic inflammation were found to be significantly associated with a lower risk of PCa on univariable analysis (data not shown) at 2-year biopsy. Similarly, when only patients randomized to the placebo arm (N = 3148) were included in univariable analysis, chronic inflammation was found to be significantly associated with a lower risk of prostate cancer at the 2-year biopsy (OR, 0.61; P < .001) and there was a trend toward a lower risk of cancer detection among those with acute inflammation (OR, 0.77; P = .067).
Although the causes of prostate inflammation are to the best of our knowledge poorly understood, acute and chronic inflammatory infiltrates are frequently observed in prostate biopsies. Previous studies have indicated that between 35% to 100% of prostate biopsies performed for concern about PCa had some histological evidence of inflammation.[18-21] The variability in prevalence is likely due to the fact that inflammation is not routinely evaluated or reported by pathologists. In the current study, in which inflammation was systematically evaluated, approximately 15% of baseline negative biopsies for PCa demonstrated evidence of acute inflammation, whereas 77% had histological findings compatible with chronic inflammation. These results are in agreement with studies indicating that inflammation is an extremely common histological finding in prostate biopsies, even when there are no clinical signs of prostatitis. Moreover, the results of the current study suggest that inflammation should be routinely evaluated and reported in prostate biopsies.
The clinical significance of histological prostate inflammation is controversial. There is some evidence suggesting a relationship between prostate inflammation and the degree of lower urinary tract symptoms. However, the association between prostate inflammation with the development, diagnosis, and even management of PCa remains the subject of much debate. For example, several authors have demonstrated a positive correlation between urogenital infection, prostatitis, and sexually transmitted disease with PCa.[7, 8, 23] Conversely, more recent data have suggested that histological inflammation in the biopsy specimen may actually be associated with a decreased risk of PCa detection in subsequent biopsies.[12, 24, 25] In our cohort of 6238 men undergoing repeat prostate biopsy after negative baseline biopsy, baseline acute and chronic inflammation were both found to be independently associated with a lower risk of PCa. At the 2-year repeat biopsy, both baseline acute and chronic inflammation were significantly associated with a nearly 35% to 40% reduction in the diagnosis of PCa. At the 4-year repeat biopsy, acute baseline inflammation was associated with a significant reduction in the risk of PCa by approximately 30%. Thus, if these findings are confirmed in future studies, a more selective rebiopsy strategy would be reasonable in men who had histological inflammation in a prostate biopsy that was negative for cancer.
Despite the evidence linking inflammation to specific types of cancer, the traditional teaching indicates that the inflammatory and immune systems can inhibit carcinogenesis by reacting against tumor-specific and tumor-associated antigens. This process is called immunosurveillance, in which the host's immune system recognizes the transformed malignant cells as foreign agents and eliminates them before they can become an established tumor. Thus, it is conceivable that inflammation may correlate with immunosurveillance, leading to a favorable immunomodulation that could potentially explain why inflammation was associated with a lower risk of PCa in repeat prostate biopsies in the current study. If this hypothesis is true, our results support inflammation, immunosurveillance, and immunomodulation as potential areas in which therapies for preventing and treating PCa could be developed.
The presence of confounding factors could possibly be the reason why inflammation and lower PCa risk are related. For example, PSA selection for prostate biopsy may pose a potential bias. Given that some studies have suggested histological prostate inflammation to be associated with elevated PSA[19, 27] and that men are traditionally selected for prostate biopsy based on their PSA levels, men with inflammation would therefore appear to have a lower risk of PCa (ie, their elevated PSA was due to inflammation rather than cancer). To minimize this potential bias, we adjusted our analysis to PSA values. Moreover, men with inflammation had a lower PSA than those without it and men were rebiopsied regardless of PSA values. Finally, we performed a sensitivity analysis stratifying patients based on baseline PSA values and the results were virtually unchanged. Thus, although the biological link between inflammation and reduced PCa risk is not elucidated by the present study, from a clinical standpoint, inflammation in a negative biopsy for PCa may lower the risk of subsequent PCa detection on repeat biopsies. Furthermore, the association between inflammation and lower risk of PCa appears to be independent of PSA levels.
The REDUCE study has many strengths, including its large international and multicentric population, per-protocol biopsies regardless of PSA levels, prospective data acquisition, and central pathology review. The main limitation of the current study is that only men with PSA values between 2.5 ng/mL and 10 ng/mL were selected. In addition, men with high-grade intraepithelial neoplasia, atypical small acinar proliferation, or a PV > 80 mL or those who had undergone previous prostate surgery or had an International Prostate Symptom Score ≥ 25 or ≥ 20 while receiving α-blockers were excluded. Although these exclusions increase the homogeneity of the sample, they limit the generalizability of our results. We did not consider off-protocol biopsies, including transrectal/transperineal needle biopsies, transurethral resections, and prostatectomies of any kind. In addition, men were followed for 4 years and therefore we were unable to study the long-term consequences of prostate inflammation on PCa risk. In addition, we were unable to determine why chronic inflammation did not remain significantly inversely associated with cancer at the time of 4-year biopsy. We did not evaluate tumor grade in the current study. Lastly, the limited number of nonwhite men prevented us from exploring whether race influenced the association between inflammation and PCa risk.
Among men undergoing repeat prostate biopsy after an initial negative baseline biopsy, baseline acute and chronic inflammation were both found to be independently associated with a lower risk of PCa at the 2-year biopsy. At the 4-year biopsy, acute baseline inflammation was associated with a lower PCa risk. Whether these associations are driven by a false PSA increase suggesting a higher PCa risk than actually exists or a direct biological link between inflammation and reduced PCa carcinogenesis is unclear. From a clinical standpoint, inflammation in a negative biopsy for PCa may lower the risk of subsequent PCa detection on repeat biopsy.
No specific funding was disclosed.
CONFLICT OF INTEREST DISCLOSURES
Dr. Nickel has acted as a consultant for GlaxoSmithKline, Pfizer, Astellas, and Lilly. He has provided expert testimony for GlaxoSmithKline and has received a grant from the company for the BPH-Prostatitis Investigator-Initiated Study. Dr. Andriole has received a grant from GlaxoSmithKline as well as a consulting fee for acting as chair of the REDUCE Steering Committee. He has also acted as a consultant for Bayer, Genomic Health, GlaxoSmithKline, and Myriad Genetics. Dr. Castro-Santamaria is employed by GlaxoSmithKline. Dr. Freedland has received a grant from GlaxoSmithKline.