Abstract
The role of the insulin-like growth factor (IGF) axis and whether IGFs interact with androgen-suppressing agents in relation to prostate carcinogenesis is unclear. This nested case-control study (n=1652 cases/1543 controls) examined whether serum IGF1, IGF2, IGFBP2, IGFBP3 and the IGF1:IGFBP3 ratio were associated with prostate cancer in the Prostate Cancer Prevention Trial, a randomized, placebo-controlled trial of finasteride for prostate cancer prevention. Presence or absence of cancer was determined by prostate biopsy. Baseline serum was assayed for IGF-axis analytes using ELISA. Logistic regression estimated odds ratios and 95% confidence intervals for risk of total, low-grade (Gleason 2–6) and high-grade (Gleason 7–10) cancers. Results were stratified by intervention assignment. In both the placebo and finasteride arms, serum IGF1, IGF2, IGFBP3 and the IGF1:IGFBP3 ratio were not associated with prostate cancer. However men in the highest vs. lowest quartile of serum IGFBP2 had a 48% (P-trend =0.02) and 55% (P-trend=0.01) increased risk for total and low-grade cancers respectively. These IGFBP2 associations were attenuated and no longer statistically significant in the finasteride arm. Our results suggest that in general, serum IGF-axis analytes were not associated with prostate cancer risk in the PCPT where presence or absence of all cancers was biopsy-determined. The exception was the finding that high serum IGFBP2 is a risk factor for low-grade disease, which was attenuated for men on finasteride. Further research is needed to understand better the risk incurred by high IGFBP2 and whether androgen-suppressing agents such as finasteride influence aspects of IGFBP2 physiology relevant to prostate carcinogenesis.
Keywords: prostate cancer, insulin-like growth factors, randomized trials, 5-α reductase inhibitors
INTRODUCTION
The insulin-like growth factors (IGFs) are potent mitogens and anti-apoptotic factors (1). Unlike other regulatory peptides, they have characteristics of both tissue growth factors and circulating growth hormones. Thus, while they are expressed in many tissues where they have local actions, they are also present in the circulation, where levels are physiologically regulated and vary with both genetic and lifestyle factors (2). The bioactivity of IGFs is modulated by a family of high-affinity binding proteins (IGFBPs), which are also expressed in most tissues and are present in the circulation (3).
Relationships between circulating concentrations of IGFs and IGFBPs with cancer risk in general and prostate cancer risk in particular have been investigated for more than a decade. The first prospective study related to prostate cancer, based on the Physicians’ Health Study cohort, showed an approximate 4-fold increase in risk from the lowest to highest quartile of serum IGF1 concentrations, and decreased risk with increasing serum IGFBP3 concentrations (4), but did not investigate IGFBP2 or the IGF1:IGFBP3 molar ratio. Many (5–11), but not all (12, 13), subsequent studies confirmed increased risk of prostate cancer with increasing serum IGF1 concentrations, although the effect size was considerably lower than that observed in the original PHS study. The European Prospective Investigation into Cancer and Nutrition (EPIC) recently reported a prostate cancer odds ratio of 1.69 for men in the highest vs. the lowest quartile of serum IGF1, but no other IGF-axis analytes were reported (14). The Endogenous Hormones and Prostate Cancer Collaborative Group pooled data from 12 prospective cohort studies to examine associations of serum IGF1, IGF2, IGFBP2 and IGFBP3 with prostate cancer risk (15). Men in the top quintile of serum IGF1 had a modest, but significant, increased risk of prostate cancer compared to the lowest quintile (OR=1.38, 95% CI 1.19–1.60, P-trend <0.001) (15). The authors noted marked heterogeneity across the studies for associations of serum IGFBP3 with prostate cancer risk, consistent with previous findings showing considerable variation across studies with respect to the direction and magnitude of association of IGFBPs with prostate cancer risk (5, 7, 9). Still, the pooled odds ratio for IGFBP3 reported by the Collaborative group was modestly elevated (OR=1.23, 95% CI 1.06–1.43) (15). The Collaborative group found no association of serum IGF2 or IGFBP2 with prostate cancer risk, but fewer data were available on these analytes thereby limiting power to detect associations (15).
Here we examine associations of serum concentrations of IGF1, IGF2, IGFBP2, IGFBP3 and the IGF1:IGFBP3 ratio with prostate cancer risk using a nested case-control study in the Prostate Cancer Prevention Trial (PCPT) in both the intervention and placebo arms of the trial (16). Despite the strong biological plausibility and the many studies that have previously examined associations of IGF-axis analytes and prostate cancer risk, several aspects of the PCPT are unique rendering it an optimal setting in which to examine these and other questions concerning prostate cancer risk. Specifically, willing and eligible men (n=9060 of the 18,882 PCPT participants) had a prostate biopsy either during or at the conclusion of the study to detect the presence or absence of prostate cancer. The remaining participants did not have a biopsy due to the early cessation of the trial or because they elected not to undergo a biopsy (16). For those with biopsies, centralized and uniform pathological grading was used to categorize prostate cancer endpoints. While almost all prostate cancer cases in PCPT were diagnosed as local stage, detection bias was minimized and pathological grading of cases was rigorous and standardized. In addition, we have the important opportunity within this randomized controlled trial to test whether associations of IGF analytes with prostate cancer risk varies by randomization to the PCPT intervention agent, finasteride (a 5-α-reductase inhibitor), or placebo.
MATERIALS AND METHODS
Study Design and Study Population
The Prostate Cancer Prevention Trial (PCPT) was a randomized, placebo-controlled trial testing whether the 5α-reductase inhibitor, finasteride, could reduce the 7-year period prevalence of prostate cancer. Details regarding study design and participant characteristics have been described previously (16). Briefly, at 221 clinical centers across the United States 18,880 men aged 55 years and older with a normal digital rectal exam (DRE) and prostate-specific antigen (PSA) level ≤ 3.0 ng/ml, as well as no history of prostate cancer, severe benign prostatic hyperplasia, or clinically significant co-morbid conditions that would have precluded successful completion of the study protocol, were randomized to receive finasteride (5 mg/day) or placebo. During the course of the PCPT, men underwent annual DRE and PSA measures and a prostate biopsy was recommended for all men with an abnormal DRE or a finasteride-adjusted PSA of ≥ 4.0 ng/ml (17). At the final study visit, all men without a previous diagnosis of prostate cancer were offered an end-of-study biopsy. Biopsies were collected under transrectal ultrasonographic guidance and a minimum of six biopsy specimens (cores) were collected from each participant. All biopsies were reviewed both by a local study pathologist and a central study pathologist (18, 19). Discordant pathology interpretations were arbitrated by a referee pathologist and concordance was achieved in all cases (16, 18, 19). Pathologists were blinded to the randomization arm of all participants. Tumors were graded with the Gleason system by central pathology review at the Prostate Diagnostic Laboratory (Denver, CO). Study procedures were approved by Institutional Review Boards at each of the participating clinical centers, the Southwest Oncology Group (SWOG, San Antonio, TX) and the SWOG Statistical Center (Fred Hutchinson Cancer Research Center, Seattle, WA). All men signed informed consent. An independent data safety and monitoring committee met every six months throughout the course of the trial to review data on safety, adherence and diagnosis of prostate cancer (16).
This report presents data from a nested case-control study in the PCPT. Cases were men with biopsy-determined prostate cancer identified either during a for-cause interim biopsy prompted by abnormal DRE or elevated PSA or an end-of-study biopsy (for-cause and not for-cause) and who had baseline serum available for analysis (n=1,803). Tumors were classified as low-grade = Gleason 2–6; high-grade = Gleason 7–10 as was done in the original trial report (16). Controls were selected from men who completed the end-of-study biopsy procedure, had no evidence of prostate cancer and had available baseline serum samples (n=1,797). Controls were frequency matched to cases by age (in 5-year age groups), PCPT treatment arm (finasteride vs. placebo) and positive family history for first degree relative with prostate cancer. Controls were oversampled to include all eligible non-white men. Men with self-reported diabetes, reported at any time before cancer diagnosis or negative biopsy, were further excluded from these analyses due to dysregulation of the insulin and IGF axis among diabetics (20, 21) leaving n=1,652 cases and n=1,543 controls for analysis.
Data Collection
Blood Collection and Processing
Nonfasting blood specimens were collected at screening (approximately three months prior to randomization) and yearly thereafter. Venous blood was drawn into collection tubes without anticoagulant, refrigerated and shipped via overnight courier to the PCPT specimen repository where they were centrifuged, aliquotted and stored at −70°C until analysis (22).
Laboratory Analysis
Concentrations of IGF1, IGF2, IGFBP2 and IGFBP3 were assayed in the baseline serum samples with a standard ELISA using a single production lot of reagents (Diagnostic Systems Limited, Webster, TX). All assays were conducted in duplicate and the mean of the duplicate measures are reported. Two sets of QC samples (from pooled specimens) were included for quality control and the coefficients of variation (CVs) from these QC pools were as follows: IGF1 (7.1% and 5.3%), IGF2 (5.0% and 4.2%), IGFBP2 (5.5% and 8.9%) and IGFBP3 (4.2% and 4.8%). Laboratory technicians were blinded to both the randomization assignment and case-control status of all participants. The primary analyses in this report are from baseline measures of the entire case-control sample. Year 2 serum samples from a randomly selected subset of n=244 participants (121 cases and 123 controls) were used to assess potential finasteride associated change in IGF axis analytes.
Other Data
Demographic characteristics, personal medical history, family history of prostate cancer and lifestyle habits, such as smoking, usual diet, alcohol and physical activity habits were collected by self-report at baseline. The measurement characteristics of many of these self-assessment tools are published (23–25). Height and weight were assessed at the baseline clinic visit using a standard protocol (26) and weight was assessed annually thereafter. Body mass index (BMI) was computed as [weight(kg)/height(m2)] and standard cutpoints categorized BMI as normal = BMI < 25.0 kg/m2; overweight = BMI 25.0 to < 30.0 kg/m2; and obese = BMI ≥ 30.0 kg/m2 (27). Circumferences of the abdomen, waist, hip and thigh were measured at 1-year post randomization (28). As the body circumference measurements were voluntary, some clinical centers did not participate, resulting in missing data for 10% (n=319) of the participants.
Statistical Analysis
We compared baseline demographic and lifestyle characteristics of prostate cancer cases and controls by t-tests for continuous variables and Chi-squared tests for categorical variables. We compared baseline and year 2 serum concentrations of IGF-axis analytes using a paired t-test in the subset of participants who had values at both time points. We used logistic and polytomous logistic regression models to estimate associations of serum IGF1, IGF2, IGFBP2, IGFBP3 and the IGF1:IGFBP3 molar ratio with risks of total, low-grade and high-grade prostate cancer. Results are given separately for the finasteride and the placebo arms because we hypothesized a priori that finasteride treatment could modify associations between the IGF axis and prostate cancer risk. Models were adjusted for the matching factors [age, family history of first-degree relative with prostate cancer], the oversampling of non-white men, and other covariates selected based on a priori information (age, race, family history) and evidence for potential confounding in this cohort based on our data diagnostics procedures (protein intake, smoking, BMI) (29–31). The final covariates were age, race (white/non-white), family history of prostate cancer, protein intake (grams/day, continuous), BMI (continuous) and cigarette smoking (pack-years of smoking). Other variables examined, but determined non-influential on the results and therefore not included, were physical activity, education, waist circumference and waist:hip ratio. Serum concentrations of IGFs and IGFBPs were categorized into quartiles based on the distribution in the controls. Tests for linear trend across the quartiles were based on an ordinal variable taking values of 1,2,3 and 4 corresponding to rank from lowest to highest category (32). Exploratory analyses used the Wald test to investigate multiplicative interactions by entering cross-product terms of IGF axis analytes with treatment arm. These subgroup analyses examined whether risk estimates differed between for-cause and not-for-cause cancers, when stratified by BMI (< 25.0 kg/m2, 25.0–29.9 kg/m2, ≥ 30.0 kg/m2) and when stratified by baseline serum PSA (for IGFBP3 only since PSA cleaves IGFBP3). All statistical tests were two-sided with p<0.05 considered statistically significant. SAS (version 9.2, Cary, NC) was used for all statistical analyses.
RESULTS
Table 1 gives demographic and health-related characteristics of the study population by case-control status. Due to the sampling design for this nested case-control study, there were more non-white controls compared to cases and no differences between cases and controls with respect to the matching factors of age, family history of prostate cancer and intervention arm. Cases and controls did not differ by measures of adiposity (BMI, body circumferences), dietary intake of dairy and protein or alcohol use and smoking history, but a greater proportion of cases had advanced college degrees compared to controls. Two-thirds (68.9%) of prostate cancer cases were low-grade (Gleason < 7).
Table 1.
Demographic, health and lifestyle characteristics of Prostate Cancer Prevention Trial prostate cancer cases and controls (n=3195)
Characteristic1 | Cases (n=1,652) |
Controls (n=1,543) |
---|---|---|
Mean (SD) | ||
Age (y) | 63.6 (5.6) | 63.6 (5.6) |
Waist circumference (cm) | 101.2 (9.8) | 101.4 (10.3) |
Height (inches) | 70 (2.9) | 69.8 (2.8) |
Waist:hip ratio | 1.0 (0.1) | 1.0 (0.1) |
Smoking (pack-years) | 13.8 (16.2) | 14.9 (16.8) |
Alcohol intake (g/d) | 10.1 (15.5) | 9.3 (13.8) |
Protein intake (g/d) | 92.8 (37.6) | 92.7 (37.9) |
Dairy intake (svg/wk) | 10.3 (8.8) | 9.7 (8.2) |
n (%) | ||
Race/ethnicity | ||
White | 1541 (93.3) | 1268 (82.2) |
Non-White | 111 (6.7) | 275 (17.8) |
Family history of prostate cancer | 358 (21.7) | 338 (21.9) |
BMI (kg/m2) | ||
Normal (<25.0) | 485 (29.6) | 413 (27.0) |
Overweight (25.0–29.9) | 848 (51.8) | 821 (53.8) |
Obese (≥30.0) | 304 (18.6) | 293 (19.2) |
Education | ||
High school or less | 271 (16.4) | 290 (18.8) |
Some college | 444 (26.9) | 453 (29.4) |
Graduate/professional school | 927 (56.7) | 799 (51.8) |
Alcohol Intake | ||
Non-drinker | 344 (20.8) | 336 (21.8) |
< 30 grams/d | 1151 (69.7) | 1072 (69.5) |
≥ 30 grams/d | 157 (9.5) | 135 (8.7) |
Prostate Cancer Characteristics | ||
Low grade (Gleason 2–6) | 1138 (68.9) | N/A |
High-grade (Gleason 7–10) | 445 (26.9) | N/A |
All characteristics were assessed at baseline, excluding dietary intake (protein, dairy, alcohol), waist circumference and hip circumference, which were assessed at year 1.
Table 2 compares finasteride vs. placebo baseline to year 2 values for serum IGF1, IGF2, IGFBP2 and IGFBP3 (means and 95% CI). Serum concentrations of IGF1 decreased from baseline to year 2 significantly more among controls on finasteride than on placebo (P=0.03). Decreases in serum IGF1 were of suggestively greater magnitude for all prostate cancer cases on finasteride compared to placebo (P=0.16), but there were no differences for high-grade disease. There were no other differences in baseline to follow-up measures of IGF axis anaytes by study arm in either cases or controls.
Table 2.
Baseline and year 2 comparisons of serum IGF-axis analytes in cases and controls by PCPT intervention arm
Finasteride | Placebo | ||||||
---|---|---|---|---|---|---|---|
Analyte | n | Baseline | Follow-Up | n | Baseline | Follow-Up | P-value2 |
IGF1 (ng/mL) | |||||||
Controls | 57 | 223.3 (203.9, 242.7) | 204.9 (188.9, 221.0) | 66 | 192.3 (177.5, 207.1) | 189 (175.2, 202.7) | 0.03 |
All Prostate Cancer | 56 | 212.7 (195.0, 230.4) | 199.2 (185.4, 212.9) | 65 | 202.1 (188.4, 215.7) | 198.6 (183.3, 213.8) | 0.16 |
High-Grade Prostate Cancers | 24 | 207.8 (182.9, 232.6) | 197.7 (172, 223.4) | 14 | 208.5 (181.2, 235.8) | 195.8 (162.5, 229.1) | 0.79 |
IGF2 (ng/mL) | |||||||
Controls | 57 | 1779.0(1666.5,1891.5) | 1696.3(1585.1,1807.4) | 66 | 1679.2(1554.9,1803.4) | 1584.6(1455.9,1713.4) | 0.79 |
All Prostate Cancer | 56 | 1657.8(1551.4,1764.2) | 1601.8(1501.5,1702.2) | 65 | 1743.3(1644.8,1841.8) | 1671.1 (1567.1, 1775) | 0.68 |
High-Grade Prostate Cancer | 24 | 1695.6(1536.3,1854.9) | 1587.9(1434.7,1741.1) | 14 | 1934.7(1755.4,2114.1) | 1856.1 (1663.1, 2049) | 0.70 |
IGFBP2 (ng/mL) | |||||||
Controls | 57 | 557.2 (473.0, 641.4) | 592.5 (486.8, 698.1) | 66 | 586.2 (510.7, 661.8) | 660.3 (576.3, 744.3) | 0.31 |
All Prostate Cancer | 56 | 608.9 (510.7, 707.1) | 662.9 (553.2, 772.6) | 65 | 536.5 (458, 614.9) | 601.7 (515.4, 688.0) | 0.75 |
High-Grade Prostate Cancer | 24 | 594.1 (416.6, 771.5) | 616.5 (452.0, 781.0) | 14 | 515.0 (388.1, 642.0) | 524.4 (415.0, 633.8) | 0.78 |
IGFBP3 (ng/mL) | |||||||
Controls | 57 | 4221.1(3953.8,4488.5) | 3934.4 (3677, 4191.8) | 66 | 3882.1(3607.2,4156.9) | 3694.1(3414.4,3973.8) | 0.28 |
All Prostate Cancer | 56 | 3962.9(3706.4,4219.4) | 3802.8(3561.9,4043.8) | 65 | 4029.0(3800.7 4257.3) | 3890.5(3639.8,4141.2) | 0.79 |
High-Grade Prostate Cancer | 24 | 4000.7(3582.2,4419.1) | 3848.7(3459.3,4238.1) | 14 | 4486.3(4107.9,4864.6) | 4191.8(3700.9,4682.8) | 0.43 |
n=244 randomly selected participants had follow-up measures using bloods drawn at year 2. Values are means (95% CI).
P-values are from t-tests comparing change from baseline to follow-up values between finasteride and treatment arms.
Table 3 gives associations of serum concentrations of IGF1, IGF2, IGFBP3, IGFBP2 and the IGF1:IGFBP3 molar ratio with total, low-grade and high-grade prostate cancer risk stratified by PCPT treatment arm (placebo or finasteride). In both the placebo and finasteride arms we found no associations between serum IGF1, IGF2, IGFBP3 and the IGF1:IGFBP3 with prostate cancer risk. However, higher vs. lower serum IGFBP2 was associated with a 48% increased risk (P, trend = 0.02) of total prostate cancer and a 55% increased risk (P, trend = 0.01) of low-grade prostate cancer for men randomized to placebo. These associations were attenuated and no longer statistically significant for men using finasteride. Despite these differences in risk estimates by PCPT treatment arm, none of the P-values was statistically significant from the models testing the interaction of IGF analytes with treatment in relation to prostate cancer risk. Additional subgroup analyses revealed neither differences between cases diagnosed for-cause and not-for cause nor any differences by BMI or baseline PSA in either the placebo or finasteride arms (data not shown).
Table 3.
Associations between serum IGF-1, IGF-2, IGFBP3, IGFBP3 and IGF1:IGFBP3 with risk of total and high grade prostate cancer by treatment arm in the Prostate Cancer Prevention Trial
Odds Ratios (95% CI) |
|||||
---|---|---|---|---|---|
Serum IGF-1 (ng/mL)1 |
|||||
Placebo | Q1 <167.1 ng/mL |
Q2 167.1 to <203.7 ng/mL |
Q3 203.7 to <250.2 ng/mL |
Q4 ≥250.2 ng/mL |
P-trend |
All cases | 1.0 (ref) | 0.89(0.67–1.20) | 1.10(0.83–1.45) | 1.06(0.79–1.42) | 0.39 |
No. of cases | 184 | 179 | 240 | 210 | |
Gleason 2–6 | 1.0 (ref) | 0.83(0.60–1.14) | 1.07(0.79–1.45) | 1.11(0.81–1.52) | 0.24 |
No. of cases | 138 | 129 | 181 | 172 | |
Gleason 7–10 | 1.0 (ref) | 1.14(0.70–1.86) | 1.28(0.80–2.05) | 0.90(0.53–1.53) | 0.90 |
No. of cases | 37 | 42 | 52 | 31 | |
Finasteride | |||||
All cases | 1.0 (ref) | 0.92(0.66–1.30) | 1.24(0.88–1.76) | 1.02(0.73–1.43) | 0.55 |
No. of cases | 134 | 142 | 157 | 160 | |
Gleason 2–6 | 1.0 (ref) | 0.92(0.62–1.37) | 1.31(0.88–1.95) | 1.05(0.71–1.55) | 0.45 |
No. of cases | 76 | 82 | 97 | 98 | |
Gleason 7–10 | 1.0 (ref) | 0.95(0.61–1.49) | 1.15(0.72–1.82) | 0.96(0.61–1.51) | 0.96 |
No. of cases | 53 | 56 | 54 | 55 | |
P-values for interaction tests (treatment * IGF1) were 0.87 (all cancers), 0.90 (Gleason 2–6) and 0.87 (Gleason 7–10) | |||||
Serum IGF-2 (ng/mL)1 |
|||||
Q1 <1448.3 ng/mL |
Q2 1448.3 to <1722.3 ng/mL |
Q3 1722.3 to <1999.7 ng/mL |
Q4 ≥ 1999.7 ng/mL |
||
Placebo | |||||
All cases | 1.0 (ref) | 1.23(0.92–1.63) | 1.22(0.92–1.63) | 1.13(0.84–1.52) | 0.46 |
No. of cases | 167 | 214 | 230 | 202 | |
Gleason 2–6 | 1.0 (ref) | 1.19(0.87–1.62) | 1.21(0.89–1.65) | 1.13(0.82–1.56) | 0.48 |
No. of cases | 125 | 160 | 177 | 158 | |
Gleason 7–10 | 1.0 (ref) | 1.27(0.78–2.07) | 1.18(0.72–1.92) | 1.08(0.65–1.81) | 0.85 |
No. of cases | 37 | 44 | 44 | 37 | |
Finasteride | |||||
All cases | 1.0 (ref) | 0.99 (0.71–1.40) | 1.14(0.81–1.59) | 1.02(0.73–1.44) | 0.70 |
No. of cases | 132 | 144 | 164 | 153 | |
Gleason 2–6 | 1.0 (ref) | 1.19(0.80–1.77) | 1.29(0.87–1.91) | 1.13(0.76–1.69) | 0.52 |
No. of cases | 70 | 92 | 99 | 92 | |
Gleason 7–10 | 1.0 (ref) | 0.80(0.51–1.27) | 1.01(0.65–1.57) | 0.93(0.59–1.47) | 0.99 |
No. of cases | 55 | 48 | 60 | 55 | |
P-values for interaction tests (treatment * IGF2) were 0.87 (all cancers), 0.90 (Gleason 2–6) and 0.95 (Gleason 7–10) | |||||
Serum IGFBP3 (ng/mL)2 |
|||||
Q1 <3418.2 ng/mL |
Q2 3418.2 to < 3997.7 ng/mL |
Q3 3997.7 to < 4644.2 ng/mL |
Q4 ≥ 4644 ng/mL |
||
Placebo | |||||
All cases | 1.0 (ref) | 1.03(0.76–1.38) | 1.12(0.82–1.54) | 1.02(0.71–1.47) | 0.77 |
No. of cases | 174 | 197 | 226 | 216 | |
Gleason 2–6 | 1.0 (ref) | 1.01(0.73–1.39) | 1.08(0.77–1.51) | 0.99(0.67–1.46) | 0.95 |
No. cases | 129 | 150 | 172 | 169 | |
Gleason 7–10 | 1.0 (ref) | 1.01(0.60–1.68) | 1.30(0.77–2.22) | 1.10(0.59–2.05) | 0.56 |
No. of cases | 39 | 37 | 47 | 39 | |
Finasteride | |||||
All cases | 1.0 (ref) | 0.86(0.60–1.23) | 1.22(0.83–1.78) | 1.23(0.79–1.91) | 0.16 |
No. of cases | 134 | 130 | 170 | 159 | |
Gleason 2–6 | 1.0 (ref) | 0.91(0.60–1.38) | 1.34(0.87–2.08) | 1.30(0.78–2.17) | 0.14 |
No. of cases | 75 | 78 | 105 | 95 | |
Gleason 7–10 | 1.0 (ref) | 0.82(0.50–1.32) | 1.08(0.65–1.80) | 1.30(0.72–2.34) | 0.25 |
No. of cases | 54 | 47 | 57 | 60 | |
P-values for interaction tests (treatment * IGFBP3) were 0.98 (all cancers), 0.89 (Gleason 2–6) and 0.92 (Gleason 7–10) | |||||
Serum IGFBP2 (ng/mL)
3 |
|||||
Q1 <318.2 ng/mL |
Q2 318.2 to < 466.9 ng/mL |
Q3 466.9 to < 675.9 ng/mL |
Q4 ≥ 675.9 ng/mL |
||
Placebo | |||||
All cases | 1.0 (ref) | 1.20(0.90–1.61) | 1.33(0.98–1.81) | 1.48(1.06–2.06) | 0.02 |
No. of cases | 169 | 212 | 205 | 227 | |
Gleason 2–6 | 1.0 (ref) | 1.21(0.88–1.66) | 1.40(1.00–1.96) | 1.55(1.08–2.22) | 0.01 |
No. of cases | 123 | 158 | 162 | 177 | |
Gleason 7–10 | 1.0 (ref) | 1.42(0.88–2.31) | 1.19(0.70–2.04) | 1.40(0.79–2.49) | 0.40 |
No. of cases | 37 | 50 | 35 | 40 | |
Finasteride | |||||
All cases | 1.0 (ref) | 1.16(0.81–1.67) | 0.89(0.61–1.30) | 1.16(0.77–1.73) | 0.77 |
No. of cases | 124 | 149 | 147 | 173 | |
Gleason 2–6 | 1.0 (ref) | 1.09(0.73–1.65) | 0.76(0.49–1.18) | 1.16(0.73–1.84) | 0.81 |
No. of cases | 77 | 88 | 79 | 109 | |
Gleason 7–10 | 1.0 (ref) | 1.35(0.82–2.20) | 1.19(0.71–1.98) | 1.25(0.72–2.17) | 0.60 |
No. of cases | 41 | 56 | 62 | 59 | |
P-values for interaction tests (treatment * IGFBP2) were 0.64 (all cancers), 0.46 (Gleason 2–6) and 0.47 (Gleason 7–10) | |||||
IGF1:IGFBP31 |
|||||
Q1 <0.045 |
Q2 0.045 to <0.052 |
Q3 0.052 to < 0.059 |
Q4 ≥ 0.059 |
||
Placebo | |||||
All cases | 1.0 (ref) | 0.92(0.70–1.22) | 0.89(0.67–1.17) | 1.07(0.80–1.42) | 0.77 |
No. of cases | 202 | 206 | 204 | 201 | |
Gleason 2–6 | 1.0 (ref) | 0.81(0.60–1.10) | 0.85(0.63–1.15) | 1.11(0.82–1.50) | 0.48 |
No. of cases | 157 | 143 | 156 | 164 | |
Gleason 7–10 | 1.0 (ref) | 1.53(0.96–2.43) | 1.08(0.66–1.78) | 0.93(0.54–1.59) | 0.49 |
No. of cases | 35 | 57 | 41 | 29 | |
Finasteride | |||||
All cases | 1.0 (ref) | 0.98 (0.70–1.38) | 1.10 (0.79–1.55) | 0.87 (0.62–1.20) | 0.51 |
No. of cases | 149 | 140 | 151 | 153 | |
Gleason 2–6 | 1.0 (ref) | 0.95(0.64–1.41) | 1.18(0.81–1.73) | 0.83(0.57–1.21) | 0.53 |
No. of cases | 87 | 81 | 97 | 88 | |
Gleason 7–10 | 1.0 (ref) | 0.95(0.61–1.49) | 0.96(0.61–1.51) | 0.81(0.52–1.26) | 0.37 |
No of cases | 60 | 53 | 50 | 55 | |
P-values for interaction tests (treatment * IGF1:IGFBP3) were 0.46 (all cancers), 0.36 (Gleason 2–6) and 0.96 (Gleason 7–10) |
Adjusted for age, race (white vs. non-white), family history of first degree relative with prostate cancer, body mass index, dietary protein intake and pack-years of smoking.
Also adjusted for serum IGF1
Also adjusted for serum IGF2
DISCUSSION
In the Prostate Cancer Prevention Trial, the majority of prostate cancer cases were low-grade and asymptomatic and the presence or absence of all cancers was determined by prostate biopsy. Neither serum IGF1, IGF2, IGFBP3 nor the IGF1:IGFBP3 ratio were associated with prostate cancer risk. The null findings were consistent for total cancer, low-grade and high-grade prostate cancers and across both PCPT study arms. Only serum IGFBP2 was associated with a modest, but significant, increased risk of total (OR=1.48) and low-grade (OR=1.55) prostate cancer among placebo-randomized men, but not finasteride-randomized men.
To our knowledge this is the first report implicating serum IGFBP2 in prostate cancer risk. The pooled analysis of 12 cohort studies found no association of serum IGFBP2 with prostate cancer risk, but power was limited to detect associations since only four of the 12 studies had data on IGFBP2 (15). The PCPT finding that high vs. low IGFBP2 is associated with increased prostate cancer risk is somewhat novel and is supported by data from in vitro and animal model studies. Meherian-Shai et al used expression profiling of prostate cancer xenografts to demonstrate that serum IGFBP2 may be a serum biomarker of PTEN status and activation of the PI3/Akt pathway in prostate cancer (33). In their experiments, these investigators found that elevated IGFBP2 expression was common in PTEN-mutant tumors (33). Since PTEN is a well-known tumor suppressor gene, the finding is potentially important in terms of identifying both the etiology of some prostate cancers as well as confirming a role for molecules in the IGF family and their relationship to activation of the PI3/AKT pathway. IGFBP2 has also been suggested to be a growth factor for DU145 human prostate cells and it may be involved in growth regulation of both normal and neoplastic prostate cells (34). Thus, these PCPT results have may have important biological relevance.
The lack of an association between circulating IGFI concentrations and prostate cancer risk in the PCPT was unanticipated, given the results of many prior studies as summarized in the 2008 meta-analysis and more recently the results from the large EPIC cohort (14, 15). Assay inaccuracy is an unlikely explanation for the lack of an association in the PCPT, as internal controls were satisfactory and the expected relationships between IGFI concentrations and age and between IGF1 and IGFBP3 concentrations were observed (data not shown). Prior studies of the Physicians’ Health Study cohort noted that the IGFI related risk was greater in the pre-PSA screening era than after PSA screening became common (4, 9). This observation suggested that IGFI related risk may not operate early in carcinogenesis, but rather that high serum IGFI influenced rate of progression from subclinical to symptomatic disease. Thus, when cases were assessed earlier in the natural history due to PSA screening, the impact of IGFI as a risk factor becomes reduced. Notably, routine PSA screening is not performed in the countries participating in EPIC where higher vs. lower serum IGF1 was associated with a 69% increased risk of prostate cancer (14). We speculate that in the closely-monitored PCPT population, IGFI and other IGF analytes did not emerge as prostate cancer risk factors for this reason. Interestingly, however, this does not preclude utility of serum IGFI concentration as a predictor of risk of clinically significant disease, an issue not investigated in PCPT.
We had hypothesized that associations of serum IGF analytes with prostate cancer risk might vary by PCPT treatment arm. Finasteride blocks the conversion of testosterone to the more potent dihydrotestosterone by inhibiting 5-α-reductase. While androgens are the primary target of finasteride, evidence suggests cross-talk exists between androgens and IGFs or their signaling pathways (1, 35–37). For example, one recent report demonstrated an increase in steroid hormone synthesis following insulin treatment of prostate cancer cell lines (36). Other data suggest direct interaction may exist between the androgen receptor (AR) and the IGF receptor (IGF-IR) (37) offering biological plausibility to support these findings from the PCPT. Despite the biological rationale and the modest attenuation of the IGFBP2 odds ratios in the finasteride arm, we observed no conclusive evidence for an interaction of PCPT treatment arm with IGF analytes in relation to prostate cancer risk.
This study has several strengths. The PCPT was a large placebo-controlled randomized trial. Part of the trial design specified that prostate cancer outcomes would be based on for-cause or end-of-study biopsy results. As such, the control group used in these analyses all had negative prostate biopsies, thus reducing the possibility that controls may have had undiagnosed or undetected disease. Other strengths include the carefully collected data throughout the course of the trial, the central pathology laboratory for uniform adjudication of all cases (including adjudication of Gleason). Limitations should also be noted, including the fact that the PCPT included few minorities. While we oversampled non-white controls to increase power for analyses by race, the power for any race-specific substrata was limited and thus not performed for this report. In addition, most of the cases were low-grade so power was limited to detect differential associations by tumor grade. Finally, few deaths from prostate cancer have occurred in the PCPT so we are unable to conduct analyses to examine mortality as an endpoint.
In conclusion, in this nested case-control study from the Prostate Cancer Prevention Trial, we found that higher vs. lower serum IGFBP2 was associated with a 55% increased risk of low-grade prostate cancer cancers. Unlike several previous studies, though, we found no association of any of the other IGF axis analytes with prostate cancer risk and no effect modification by finasteride.
Acknowledgements
The authors gratefully acknowledge the Southwest Oncology Group (SWOG) and the SWOG PCPT Biorespository at the University of Colorado, Denver.
Funding: This work was supported by U01 CA37429 (Prostate Cancer Prevention Trial), P01 CA108964 (Biology of the Prostate Cancer Prevention Trial), National Cancer Institute, National Institutes of Health, United States Department of Health and Human Services
Footnotes
No author has declared a conflict of interest
References
- 1.Pollak M. Insulin and Insulin-like growth factor signalling in neoplasia. Nature Reviews - Cancer. 2008;8(12):915–928. doi: 10.1038/nrc2536. [DOI] [PubMed] [Google Scholar]
- 2.Harrela M, Koistinen H, Kaprio J, Lehtovirta M, Tuomilehto J, Eriksson J, et al. Genetic and environmental components of interindividual variation in circulating levels of IGF-I, IGF-II, IGFBP-1, and IGFBP-3. Journal of Clinical Investigation. 1996;98(11):2612. doi: 10.1172/JCI119081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Firth SM, Baxter RC. Cellular actions of the insulin-like growth factor binding proteins. Endocrine Reviews. 2002;23(6):824–854. doi: 10.1210/er.2001-0033. [DOI] [PubMed] [Google Scholar]
- 4.Chan J, Stampfer M, Giovannucci E, Gann P, Ma J, Wilkinson P, et al. Plasma insulin-like growth factor 1 and prostate cancer risk: a prospective study. Science. 1998;279:563–566. doi: 10.1126/science.279.5350.563. [DOI] [PubMed] [Google Scholar]
- 5.Signorello LB, Brismar K, Bergstrom R, Andersson SW, Wolk A, Trichopoulos D, et al. Insulin-like growth factor-binding protein-1 and prostate cancer. Journal of the National Cancer Institute. 1999;91(22):1965–1967. doi: 10.1093/jnci/91.22.1965. [DOI] [PubMed] [Google Scholar]
- 6.Stattin P, Bylund A, Rinaldi S, Biessy C, Déchaud H, Stenman U, et al. Plasma insulin-like growth factor-I, insulin-like growth factor-binding proteins, and prostate cancer risk: a prospective study. Journal of the National Cancer Institute. 2000;92:1910–1917. doi: 10.1093/jnci/92.23.1910. [DOI] [PubMed] [Google Scholar]
- 7.Harman S, Metter E, Blackman M, Landis P, Carter H. Serum levels of insulin-like growth factor I (IGF-I), IGF-II, IGF-binding protein-3, and prostate-specific antigen as predictors of clinical prostate cancer. Journal of Clinical Endocrinology & Metabolism. 2000;85:4258–4265. doi: 10.1210/jcem.85.11.6990. [DOI] [PubMed] [Google Scholar]
- 8.Chokkalingam AP, McGlynn KA, Gao YT, Pollak M, Deng J, Sesterhann IA, et al. Vitamin D receptor gene polymorphisms, insulin-like growth factors, and prostate cancer risk: a population-based case-control study in China. Cancer Research. 2001;61:4333–4336. [PubMed] [Google Scholar]
- 9.Chan JM, Stampfer MJ, Ma J, Gann P, Gaziano JM, Pollak M, et al. Insulin-like growth factor-I (IGF-I) and IGF binding protein-3 as predictors of advanced-stage prostate cancer. Journal of the National Cancer Institute. 2002;94:1099–1109. doi: 10.1093/jnci/94.14.1099. [DOI] [PubMed] [Google Scholar]
- 10.Woodson K, Tangrea JA, Pollak M, Copeland TD, Taylor PR, Virtamo J, et al. Serum insulin-like growth factor I: Tumor marker or etiologic factor? A prospective study of prostate cancer among Finnish men. Cancer Research. 2003;63(14):3991–3994. [PubMed] [Google Scholar]
- 11.Hsing AW, Chua S, T GY, Gentzschein E, Change L, Deng J, et al. Prostate cancer risk and serum levels of insulin and leptin: a population-based study. Journal of the National Cancer Institute. 2001;93:783–789. doi: 10.1093/jnci/93.10.783. [DOI] [PubMed] [Google Scholar]
- 12.Allen NE, Key TJ, Appleby PN, Travis RC, Roddam AW, Rinaldi S, et al. Serum insulin-like growth factor (IGF)-I and IGF-binding protein-3 concentrations and prostate cancer risk: results from the European Prospective Investigation into Cancer and Nutrition. Cancer Epidemiology Biomarkers & Prevention. 2007;16(6):1121–1127. doi: 10.1158/1055-9965.EPI-06-1062. [DOI] [PubMed] [Google Scholar]
- 13.Borugian MJ, Spinelli JJ, Sun Z, Kolonel LN, Oakley-Girvan I, Pollak MD, et al. Prostate cancer risk in relation to insulin-like growth factor (IGF)-I and IGF-binding protein-3: a prospective multiethnic study. Cancer Epidemiology Biomarkers & Prevention. 2008;17(1):252–254. doi: 10.1158/1055-9965.EPI-07-2694. [DOI] [PubMed] [Google Scholar]
- 14.Price AJ, Allen NE, Appleby PN, Crowe FL, Travis RC, Tipper SJ, et al. Insulin-like Growth Factor-I Concentration and Risk of Prostate Cancer: Results from the European Prospective Investigation into Cancer and Nutrition. Cancer Epidemiol Biomarkers Prev. 2012;21(9):1531–1541. doi: 10.1158/1055-9965.EPI-12-0481-T. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Roddam AW, Allen NE, Appleby P, Key TJ, Ferrucci L, Carter HB, et al. Insulin-like growth factors, their binding proteins, and prostate cancer risk: analysis of individual patient data from 12 prospective studies. Annals of Internal Medicine. 2008;149(7):461–471. doi: 10.7326/0003-4819-149-7-200810070-00006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Thompson IM, Goodman PJ, Tangen CM, Lucia MS, Miller GJ, Ford LG, et al. The influence of finasteride on the development of prostate cancer. New England Journal of Medicine. 2003;349(3):215–224. doi: 10.1056/NEJMoa030660. [DOI] [PubMed] [Google Scholar]
- 17.Thompson IM, Chi C, Ankerst DP, Goodman PJ, Tangen CM, Lippman SM, et al. Effect of finasteride on the sensitivity of PSA for detecting prostate cancer. Journal of the National Cancer Institute. 2006;98(16):1128–1133. doi: 10.1093/jnci/djj307. [DOI] [PubMed] [Google Scholar]
- 18.Lucia MS, Darke AK, Goodman PJ, La Rosa FG, Parnes HL, Ford LG, et al. Pathologic characteristics of cancers detected in the Prostate Cancer Prevention Trial: Implications for prostate cancer detection and chemoprevention. Cancer Prevention. 2008:1–7. doi: 10.1158/1940-6207.CAPR-08-0078. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lucia MS, Epstein JI, Goodman PJ, Darke AK, Reuter VE, Civantos F, et al. Finasteride and high-grade prostate cancer in the Prostate Cancer Prevention Trial. Journal of the National Cancer Institute Monographs. 2007;99(18):1375–1383. doi: 10.1093/jnci/djm117. [DOI] [PubMed] [Google Scholar]
- 20.Kaaks R, Lukanova A, Rinaldi S, Biessy C, Soderberg S, Olsson T, et al. Interrelationships between plasma testosterone, SHBG, IGF-1, insulin and leptin in prostate cancer cases and controls. European Journal of Cancer Prevention. 2003;13:309–315. doi: 10.1097/00008469-200308000-00011. [DOI] [PubMed] [Google Scholar]
- 21.Nyomba B, Berard L, Murphy L. Free insulin-like growth factor I (IGF-I) in healthy subjects: relationships with IGF binding proteins and insulin sensitivity. Journal of Clinical Endocrinology & Metabolism. 1997;82:2177–2181. doi: 10.1210/jcem.82.7.4070. [DOI] [PubMed] [Google Scholar]
- 22.Kristal AR, King IB, Albanes D, Pollak MN, Stanzyk FZ, Santella RM, et al. Centralized blood processing for the selenium and vitamin E cancer prevention trial: effects of delayed processing on carotenoids, tocopherols, insulin-like growth factor-I, insulin-like growth factor binding protein 3, steroid hormones, and lymphocyte viability. Cancer Epidemiology, Biomarkers & Prevention. 2005;14(3):727–730. doi: 10.1158/1055-9965.EPI-04-0596. [DOI] [PubMed] [Google Scholar]
- 23.Neuhouser ML, Kristal AR, McLerran D, Patterson RE, Atkinson J. Validity of short food frequency questionnaires used in cancer chemoprevention trials: Results from the Prostate Cancer Prevention Trial. Cancer Epidemiology Biomarkers & Prevention. 1999;8:721–725. [PubMed] [Google Scholar]
- 24.Neuhouser ML, Kristal AR, Patterson RE, Goodman PJ, Thompson IM. Dietary supplement use in the Prostate Cancer Prevention Trial: implications for prevention trials. Nutrition & Cancer. 2001;39(1):12–18. doi: 10.1207/S15327914nc391_2. [DOI] [PubMed] [Google Scholar]
- 25.Kristal AR, Arnold KB, Neuhouser ML, Goodman PJ, Platz EA, Albanes D, et al. Diet, supplement use, and prostate cancer risk: results from the Prostate Cancer Prevention Trial. American Journal of Epidemiology. 2010;172(5):566–577. doi: 10.1093/aje/kwq148. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Lohman T, Roche A, Martorell M. Anthropometric standardization reference manual. Champaign, IL: Human Kinetics Books; 1988. [Google Scholar]
- 27.Expert Panel on the Identification Evaluation and Treatment of Overweight and Obesity in Adults. Clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults. American Journal of Clinical Nutrition. 1998;68:899–917. doi: 10.1093/ajcn/68.4.899. [DOI] [PubMed] [Google Scholar]
- 28.Satia-Abouta J, Patterson RE, Schiller RN, Kristal AR. Energy from fat is associated with obesity in U.S. men: Results from the Prostate Cancer Prevention Trial. Preventive Medicine. 2002;34(5):493–501. doi: 10.1006/pmed.2002.1018. [DOI] [PubMed] [Google Scholar]
- 29.Giovannucci E, Pollak M, Liu Y, Platz EA, Majeed N, Rimm EB, et al. Nutritional predictors of insulin-like growth factor I and their relationships to cancer in men. Cancer Epidemiology Biomarkers & Prevention. 2003;12(2):84–89. [PubMed] [Google Scholar]
- 30.Yu H, Rohan T. Role of insulin-like growth factor family in cancer development and progression. Journal of the National Cancer Institute. 2000;92:1472–1489. doi: 10.1093/jnci/92.18.1472. [DOI] [PubMed] [Google Scholar]
- 31.Allen NE, Key TJ, Appleby PN, Travis RC, Roddam AW, Tjønneland A, et al. Animal foods, protein, calcium and prostate cancer risk: the European Prospective Investigation into Cancer and Nutrition. British Journal of Cancer. 2008;98(9):1574–1581. doi: 10.1038/sj.bjc.6604331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Breslow NE, Day NE. Statistical methods in cancer research. Vol 1 - The analysis of case-control studies. Lyon: Intl Agency for Research on Cancer; 1980. [PubMed] [Google Scholar]
- 33.Mehrian-Shai R, Chen CD, Shi T, Horvath S, Nelson SF, Reichardt JKV, et al. Insulin growth factor-binding protein 2 is a candidate biomarker for PTEN status and PI3K/Akt pathway activation in glioblastoma and prostate cancer. Proceedings of the National Academy of Sciences. 2007;104(13):5563–5568. doi: 10.1073/pnas.0609139104. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Miyako K, Cobb LJ, Francis M, Huang A, Peng B, Pintar JE, et al. PAPA-1 is a nuclear binding partner of IGFBP-2 and modulates its growth-promoting actions. Molecular Endocrinology. 2009;23(2):169–175. doi: 10.1210/me.2008-0168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Pollak M. The insulin receptor/insulin-like growth factor receptor family as a therapeutic target in oncology. Clinical Cancer Research. 2012;18(1):40–40. doi: 10.1158/1078-0432.CCR-11-0998. [DOI] [PubMed] [Google Scholar]
- 36.Lubik AA, Gunter JH, Hendy SC, Locke JA, Adomat HH, Thompson V, et al. Insulin increases de novo steroidogenesis in prostate cancer cells. Cancer Research. 2011;71(17):5754–5764. doi: 10.1158/0008-5472.CAN-10-2470. [DOI] [PubMed] [Google Scholar]
- 37.Wu J, Haugk K, Woodke L, Nelson PS, Coleman I, Plymate SR. Interaction of IGF signaling and the androgen receptor in prostate cancer progression. Journal of Cellular Biochemistry. 2006;99(2):392–401. doi: 10.1002/jcb.20929. [DOI] [PubMed] [Google Scholar]