Abstract
Endostatin inhibits endothelial cell proliferation and migration, prerequisites of angiogenesis. A functional missense mutation (D104N) in endostatin was associated with an increased prostate cancer risk in a small study. We undertook a larger, prospective study within the Physicians’ Health Study to examine D104N and prostate cancer risk and progression among 544 incident prostate cancer cases (1982-1995) and 678 matched controls. The association between endostatin genotype and cancer risk was estimated using logistic regression models. Among cases, Cox models were used to assess D104N and lethal prostate cancer. Given the role of endostatin in neovascularization of adipose tissue, we cross-classified individuals on D104N genotype and body mass index (BMI). The genotype frequency was 1.3% homozygous (NN), 14.5% heterozygous (DN), and 84.2% wildtype homozygous (DD). There was no overall association between carriage of the N allele and prostate cancer risk (RR=1.2, 95% CI: 0.9-1.6) or cancer-specific mortality (HR=1.2, 0.7-1.8). Cases with the polymorphic allele were less likely to be overweight (BMI 25 kg/m2 or greater, 26%) compared to men wildtype homozygous (48%), p<0.0001. Being overweight was associated with a 60% greater prostate cancer risk among those who were wildtype homozygous. In contrast, being overweight was associated a 50% lower risk of cancer among those with the N allele. We did not confirm earlier observation between the D104N polymorphism and prostate cancer. However, our data indicate that prostate cancer cases who carry the variant N allele are more likely to be overweight, and may be more susceptible to the angiogenic influences of obesity in prostate cancer pathogenesis.
Keywords: Endostatin, angiogenesis, obesity, prostate cancer
Introduction
Angiogenesis is a discrete event in carcinogenesis that is related to the aggressive potential of a tumor.(1,2) Through the building of new vasculature, angiogenesis provides an independent blood supply essential for tumor growth and expansion.(3) Prostate cancer has the ability to produce angiogenenic factors (4), and there is accumulating evidence that increased angiogenenic properties within the tumor are associated with poor prostate cancer outcomes. (5,6)
Endostatin, a cleavage product of the C-terminal of collagen, is a potent inhibitor of endothelial cell proliferation and migration(4), requisites of angiogenesis. Endostatin acts exclusively on the endothelium, and induces endothelial cell apoptosis in vitro.(7) In animal models, systemic administration of endostatin increases apoptotic activity in tumor cells, decreases microvessel density, and regresses primary tumors.(8)
A missense mutation in the coding region for endostatin involves an asp1437-to-asn change in the COL18A1 gene, a variant located at residue 104 in endostatin (D104N, rs1248337). The polymorphism does not appear to affect expression, since serum endostatin levels were similar among carriers and noncarriers of the variant allele.(9) Structural modeling analysis suggests that the polymorphism may affect protein conformation, which may subsequently impair the ability of endostatin to bind to other molecules.(9) While in cellular assays the variant allele did not influence binding to heparin, nor affect the ability of endostatin to inhibit endothelial cell migration, it did impair binding of endostatin to laminin.(10)
In a hospital-based, case-control study, Iughetti et al first examined endostatin genotype and prostate cancer risk, reporting a more than 2-fold increased risk among men who carried the variant allele.(9) In the present investigation, we present data from a larger, prospective study within the Physicians’ Health Study examining the D104N endostatin polymorphism in relation to prostate cancer risk and survival. In addition to its role as an angiogenic inhibitor, endostatin influences neovascularization of nonneoplastic tissue such as adipose.(12) Accumulating data suggests a relation between body mass index (BMI) and prostate cancer progression.(13,14) Thus, we examined potential effect modification of BMI on prostate cancer risk and survival by the endostatin polymorphism.
Material and Methods
Study population
Participants in the present investigation come from the Physicians’ Health Study, a randomized trial of aspirin and beta-carotene supplementation initiated in 1982 among 22,071 US male physicians ages 40-84. All were initially free of cancer at diagnosis. During 1982-1984, physicians were sent blood collection kits to obtain baseline samples prior to randomization. Prospectively collected samples are available from 14,916 (68%) of physicians, representing the study base for the nested case-control study.
Disease ascertainment
Participants are followed through annual questionnaires to collect data on lifestyle and medical factors, and biennally through postcards to ascertain compliance and health endpoints. Follow-up of the participants for morbidity and mortality is 99% complete and ongoing (11).
Diagnosis of incident prostate cancer was self-reported by the participants and confirmed by study investigators through review of medical records and pathology reports. Information on tumor stage and grade was abstracted through these records. Prostate cancer death is confirmed through death certificate review, and information on post-diagnostic metastases collected through annual follow-up questionnaires of the men with prostate cancer.
We sought to select controls randomly from participants who had not had a partial or total prostatectomy or prostate cancer by the date of the case’s diagnosis, and whom had sufficient plasma for biochemical analyses. Controls were matched, 1:1 or 1:2, to cases on age (±1 year, ±5 years for elderly participants) and smoking status (current, former, never). Given the low proportion (7%) of non-Caucasian males in the study, we limited this analysis to Caucasians in order to reduce possible bias from population stratification.
Included in this analysis were 544 incident prostate cancer cases diagnosed through 1995 and 678 controls with available genotyping data. Of the 544 cases, 147 (28.7%) were advanced stage tumors at diagnosis (Stage T3/T4/N1/M1), 223 (48.8%) had a high-grade tumor (Gleason 7 or higher, or poorly differentiated), and 129 (23.7%) of these cases developed lethal prostate cancer, defined as distant metastases or died of prostate cancer during follow-up. The mean age of cases and controls at baseline was 61 years (Table 1).
Table 1.
Baseline characteristics of prostate cancer cases and controls in the Physicians’ Health Study 1982-1995
| Characteristic | Prostate Cancer Cases |
Controls |
|---|---|---|
| N | 544 | 678 |
| Mean age at baseline, years (±SD) | 61.2 (±7.5) | 61.4 (±7.6) |
| Mean age at diagnosis, years (±SD) | 68.9 (±6.7) | N/A |
| Mean post-diagnostic follow-up, years (±SD) | 12.2 (±5.4) | N/A |
| Smoking Status in 1982, N (%) | ||
| Never | 250 (46.0) | 304 (44.8) |
| Former | 252 (46.3) | 318 (46.9) |
| Current | 42 (7.7) | 56 (8.3) |
| Body mass index in 1982, N (%) | ||
| <25 kg/m2 | 305 (56.1) | 423 (62.4) |
| 25-30 kg/m2 | 220 (40.4) | 237 (35.0) |
| 30+ kg/m2 | 19 (3.5) | 18 (2.6) |
| Endostatin Genotype, N (%) | ||
| DD | 448 (82.4) | 571 (84.2) |
| DN | 90 (16.5) | 98 (14.5) |
| NN | 6 (1.1) | 9 (1.3) |
| Tumor Stage, N (%) | ||
| T1/T2 | 366 (71.3) | N/A |
| T3/T4N1/M1 | 147 (28.7) | N/A |
| Gleason Score, N (%) | ||
| 2-6 | 234 (51.2) | N/A |
| 7 | 148 (32.4) | N/A |
| 8-10 | 75 (16.4) | N/A |
N/A = Information is not applicable for controls
Genotyping of endostatin D104N polymorphism
Genomic DNA was prepared from frozen peripheral blood using “QIAamp DNA Blood mini kit” (QIAGEN Inc, Valencia, CA), quantitated by OD260, and stored at 4°C. Genotypes were analyzed following to a previously established procedure(9) with modifications. Genomic DNA (200 ng) was amplified using 40 pmoles each of primers (Forward: 5-CACGGTTTCTCTTCCAGGAC-3′ and Reverse: 5′CTCTCAGAGCTGCTCACACG-3′) in 45 ul 1× Pfx amplification buffer (Invitrogen Corp., Carlsbad, CA) containing 0.56 mM MgSO4, 0.33 mM dNTP, and 1 unit of Platinum Pfx DNA Polymerase (Invitrogen). Thirtytwo cycles of primary PCR amplification [denaturing (94°C, 40 sec), annealing (62°C, 30 sec), and extension (72°C, 30sec)] were performed. One ul product from a primary PCR was subjected to a secondary PCR using exactly the same condition as for the primary PCR. The reaction was maintained at 4°C after cycling. These PCRs yielded a 168 bp amplicon. Ten ul of the final PCR product was digested by incubating with 4 units of MseI (New England BioLabs, MA) at 37°C for 1 hr. Digested products were visualized on a 2% agarose gel stained with ethidium bromide. Fragment patterns specific for three genotypes were: D(Asp)/D(Asp) (GAC/GAC; 168 bp), D(Asp)/N(Asn) (GAC/AAC; 168, 101, 67 bp), and N(Asn)/N(Asn) (AAC/AAC; 101, 67 bp). To confirm genotypes, eighty-eight PCR products were randomly picked up and subjected to DNA sequencing using the same forward/reverse primers described above. Results showed identical genotypes for all these samples.
Statistical analysis
Unconditional logistic regression models were employed to examine the relation between endostatin and prostate cancer risk, controlling for matching factors. Given the low prevalence of men homozygous for the variant (1%), we combined men who carried at least one of the N alleles. Odds ratios, as an estimate of the relative risk, and 95% confidence intervals (95 % CI) were calculated, with men homozygous wildtype (DD) as the reference category. Given the role of angiogenesis in tumor aggressiveness and growth, we examined the association of the D104N polymorphism by prostate cancer stage and tumor grade. Moreover, in a survival analysis, among the prostate cancer cases, we explored whether men who carried the N allele had a higher risk of prostate cancer death or developing metastases using proportional hazard regression models to account for potential confounding by age at diagnosis, tumor grade, and stage. Follow-up time was calculated from date of cancer diagnosis to development of distant metastases, prostate cancer death or censored at time of death from other causes or end of follow-up (March 1, 2008).
Results
Among controls, 84.2% men were wildtype homozygous (DD), 14.5% were heterozygous (DN), and 1.3% were homozygous carriers of the rare allele (NN) (Table 1) (permutation test of HWE: p = 0.05). The frequency of the N allele among the controls was similar to that previously reported.(15)
Carriage of the polymorphic N allele was not associated with risk of total prostate cancer risk (RR, 95% CI; 1.2, 0.9-1.6) (Table 2). The relative risks were similar for localized (1.3, 0.9-1.87) and advanced stage tumors (0.8, 0.5-1.4) at diagnosis, and did not differ according to tumor grade. Cancers diagnosed at an earlier age may indicate a familial component. In these data, however, there was no evidence that the endostatin D104N polymorphism was associated with prostate cancers diagnosed at a younger or older age (data available on request).
Table 2.
Endostatin D104N polymorphism in relation to prostate cancer risk and survival, the Physicians’ Health Study, 1982-2007
| NN/DN | DD | NN/DN vs. DD (REF) | |
|---|---|---|---|
| Genotype, N (%) | Relative risk (95% CI)1 | ||
| D104N polymorphism and prostate cancer risk | |||
| Controls | 107 (15.8) | 571 (84.2) | -- |
| Prostate Cancer Cases | 96 (17.6) | 448 (82.4) | 1.2 (0.9-1.6) |
| Stage at diagnosis | |||
| T3/N1/M1 | 20 (13.6) | 127 (86.4) | 0.8 (0.5-1.4) |
| T1/T2 | 69 (18.8) | 297 (81.2) | 1.3 (0.9-1.8) |
| Grade at diagnosis | |||
| High grade tumor | 34 (15.2) | 189 (84.8) | 1.0 (0.7-1.5) |
| Low grade tumor | 44 (18.8) | 190 (81.2) | 1.3 (0.9-2.0) |
| D104N polymorphism and prostate cancer survival | |||
|---|---|---|---|
| Prostate cancer deaths or mets, N (%) |
Hazard ratio (95% CI)2 | ||
| All prostate cancer cases | 24 (25.0) | 105 (23.4) | 1.2 (0.7-1.8) |
| Stage at diagnosis | |||
| T3/N1/M1 | 9 (45.0) | 59 (46.5) | 1.0 (0.5-2.1) |
| T1/T2 | 13 (18.8) | 45 (15.2) | 1.1 (0.6-2.1) |
Relative risk from unconditional logistic regression models, adjusted for matching factors age, smoking status, and follow-up time
Hazard ratios (95% Confidence Intervals) from Cox proportional hazard models among cases only (N=544), adjusted for age at diagnosis, tumor grade and stage
We undertook a survival analysis to explore the influence of the D104N polymorphism on risk of prostate cancer death or development of distant metastases over time (Table 2). Among the 544 incident cases, 129 men died of their cancer or developed metastases during follow-up through March 1, 2008. Controlling for clinical parameters, we found no association between carriage of the N allele and development of lethal prostate cancer (HR: 1.2, 95% CI: 0.8-1.8).
Interestingly prostate cancer cases with the polymorphic allele were less likely to be overweight or obese (26.0%) compared to men with the DD genotype (47.8%, p<0.0001). The mean BMI among cases was 24.9 kg/m2 among those who carried an N allele versus 24.1 kg/m2 for those who did not. There was no significant difference in the prevalence of overweight/obesity according to D104N genotype among the controls (p~0.41). Our group has previously shown within the PHS cohort that obese men have worse cancer survival compared to men with normal weight at baseline.(16) We therefore cross-classified men on BMI and D104N genotype for cancer risk and survival (Table 3). Among men homozygous wildtype for endostatin, we confirm the relation, demonstrating a 2-fold increase risk of prostate cancer among obese men (case control analysis), and a 2-fold increased risk of prostate cancer death or metastases over time among cases who were obese compared to those of normal weight (survival analysis). In contrast, among men who carried the polymorphic N allele, the relation with body mass index was inverse, such that the men who were healthy weight had a higher prostate cancer risk and a worse cancer prognosis.
Table 3.
Effect of body mass index on prostate cancer risk and survival, according to endostatin D104N genotype, Physicians’ Health Study 1982-2007
| Relative risk (95% CI) | |||
|---|---|---|---|
| NN/DN genotype |
DD genotype | P for interaction |
|
| Prostate Cancer Risk 1 | |||
| Overweight/Obese | 0.9 (0.5-1.5) | 1.6 (1.2-2.1) | <0.001 |
| Healthy weight | 1.8 (1.2-2.6) | REF | |
|
| |||
| Prostate Cancer Survival 2 | |||
| Overweight/Obese | 1.2 (0.5-3.0) | 1.7 (1.2-2.5) | 0.08 |
| Healthy weight | 1.7 (1.0-2.9) | REF | |
Relative risk from unconditional logistic regression models, adjusted for matching factors age, smoking status, and follow-up time
Hazard ratios (95% Confidence Intervals) from Cox proportional hazard models among cases only, adjusted for age at diagnosis, tumor grade and stage
Discussion
In this large, prospective study nested within the Physicians’ Health Study, we observed no overall association between the D104N endostatin polymorphism and prostate cancer risk or prognosis. Our data show a consistently null association of the genotype for high grade or advanced stage tumors. Given its characterization as an angiogenesis inhibitor, one would hypothesis a stronger effect of the D104N endostatin polymorphism on advanced stage tumors or cancer progression if there were truly a relation.
Neovascularization of tumors requires the coordination of multiple pathways involving endothelial cells, extracellular matrices and soluble factors.(17,18) Each step is regulated by angiogenesis inhibitors and inducers, and endostatin is one such inhibitor. Endostatin acts directly on endothelial cells to inhibit response to angiogenic signals, and accumulating evidence suggests a role for endostatin in inhibiting and reversing angiogenesis in prostate tumors. Transgenic mice predisposed to develop prostate cancer experienced greater survival when treated with endostatin compared to control. However, in a Col18A1 knockout model, there was no evidence of increased angiogenesis or tumor growth among mice compared to wild-type animals.(19)
The D104N polymorphism was identified by sequencing of the COL18A1 gene among patients with Knobloch syndrome, a condition characterized by vitreoretinal degeneration caused by mutation in the gene. The same investigators examined its association with prostate cancer in a hospital-based, case-control study, reporting that carriage of the polymorphic N allele was associated with a 2.5-fold increased risk of prostate cancer compared to those who were homozygous wild-type. They found a similar effect for high grade tumors and did not differ as a function of age at diagnosis.(9) More recent studies, however, have shown no association between the polymorphism in relation to prostate cancer risk overall or for aggressive prostate cancer.(15,20,21)
Although the main effects of the D104N genotype were null, we found that men with prostate cancer who carried the variant allele were less likely to be overweight or obese. Moreover, we observed a significant effect modification by body mass index, such that the effect of obesity/overweight on prostate cancer risk and progression depended on the D104N genotype. Adipose tissue produces pro- and anti-angiogenic factors, including endostatin. Indeed, individuals who are overweight or obese express higher levels of endostatin compared to healthy weight men,(22) while in an animal model, adipose tissue mass responds and regresses to exogenous angiogenesis inhibitors such as endostatin(12). Thus, the D104N polymorphism may impact prostate cancer progression via its modulation of adipose tissue.
One weaknesses of our study is the lack of comprehensive assessment of the endostatin gene, and thus, we cannot exclude that genetic variation across endostatin may be related to prostate cancer risk or progression. Moreover, our study did not focus on polymorphisms in other genes that modulate endostatin secretion or angiogenesis that could still impact overall prostate cancer incidence or progression. Still, the D104N is a missense mutation which impacts protein confirmation, and thus our study was hypothesis driven. Our study was focused on Caucasian men, and thus it is unclear whether our findings can be generalized to other populations such as African-American men who have a substantially higher risk of prostate cancer. In addition, we cannot confirm or refute a potential role of population stratification on our study findings.
The present study has a number of important strengths. The large size of our study, including a significant number of advanced stage or high grade cancers, permitted us to examine associations within important subgroups of clinical and tumor characteristics. Moreover, we were able to relate the D104N genotype to both prostate cancer incidence and mortality to look at the full spectrum from disease initiation to progression. Our study was the first to evaluate potential effect of the D104N genotype according to body mass index. Additional studies are needed to confirm or refute the apparent synergistic associations between the D104N genotype and body mass index on prostate cancer development and progression, as well as clarify the potential mechanisms by which this polymorphism could differentially influence prostate cancer outcomes according to degree of adiposity.
ACKNOWLEDGEMENTS
We are grateful to the ongoing participation of the men in the Physicians’ Health Study. The authors wish to acknowledge Joanne Smith for her work as a research coordinator and Haiyan Zhang for database management on the project. The Physicians’ Health Study is supported by grants CA-34944, CA40360 and CA-097193 from the National Cancer Institute. This work was supported by grants from the National Cancer Institute (5R01CA058684-13 and 5R01CA042182-20) (PI Stampfer).
References
- 1.Folkman J. Tumor angiogenesis: therapeutic implications. N Engl J Med. 1971;285(21):1182–1186. doi: 10.1056/NEJM197111182852108. [DOI] [PubMed] [Google Scholar]
- 2.Hanahan D, Folkman J. Patterns and emerging mechanisms of the angiogenic switch during tumorigenesis. Cell. 1996;86(3):353–364. doi: 10.1016/s0092-8674(00)80108-7. [DOI] [PubMed] [Google Scholar]
- 3.Folkman J. In: Angiogenesis. DeVita V, Hellman S, Rosenberg S, editors. Lippincott; Philadelphia: 1997. pp. 3075–3085. [Google Scholar]
- 4.van Moorselaar RJ, Voest EE. Angiogenesis in prostate cancer: its role in disease progression and possible therapeutic approaches. Mol Cell Endocrinol. 2002;197(1-2):239–250. doi: 10.1016/s0303-7207(02)00262-9. [DOI] [PubMed] [Google Scholar]
- 5.Borre M, Offersen BV, Nerstrom B, Overgaard J. Microvessel density predicts survival in prostate cancer patients subjected to watchful waiting. Br J Cancer. 1998;78(7):940–944. doi: 10.1038/bjc.1998.605. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Silberman MA, Partin AW, Veltri RW, Epstein JI. Tumor angiogenesis correlates with progression after radical prostatectomy but not with pathologic stage in Gleason sum 5 to 7 adenocarcinoma of the prostate. Cancer. 1997;79(4):772–779. doi: 10.1002/(sici)1097-0142(19970215)79:4<772::aid-cncr14>3.0.co;2-x. [DOI] [PubMed] [Google Scholar]
- 7.Dhanabal M, Ramchandran R, Waterman MJ, Lu H, Knebelmann B, Segal M, Sukhatme VP. Endostatin induces endothelial cell apoptosis. J Biol Chem. 1999;274(17):11721–11726. doi: 10.1074/jbc.274.17.11721. [DOI] [PubMed] [Google Scholar]
- 8.O’Reilly MS, Boehm T, Shing Y, Fukai N, Vasios G, Lane WS, Flynn E, Birkhead JR, Olsen BR, Folkman J. Endostatin: an endogenous inhibitor of angiogenesis and tumor growth. Cell. 1997;88(2):277–285. doi: 10.1016/s0092-8674(00)81848-6. [DOI] [PubMed] [Google Scholar]
- 9.Iughetti P, Suzuki O, Godoi PH, Alves VA, Sertie AL, Zorick T, Soares F, Camargo A, Moreira ES, di Loreto C, Moreira-Filho CA, Simpson A, Oliva G, Passos-Bueno MR. A polymorphism in endostatin, an angiogenesis inhibitor, predisposes for the development of prostatic adenocarcinoma. Cancer Res. 2001;61(20):7375–7378. [PubMed] [Google Scholar]
- 10.Menzel O, Bekkeheien RC, Reymond A, Fukai N, Boye E, Kosztolanyi G, Aftimos S, Deutsch S, Scott HS, Olsen BR, Antonarakis SE, Guipponi M. Knobloch syndrome: novel mutations in COL18A1, evidence for genetic heterogeneity, and a functionally impaired polymorphism in endostatin. Hum Mutat. 2004;23(1):77–84. doi: 10.1002/humu.10284. [DOI] [PubMed] [Google Scholar]
- 11.Henneckens CH, Buring JE, Manson JE, Stampfer MJ, Rosner B, Cook NR. Lack of effect of long-term supplementation with beta carotene on the incidence of malignant neoplasms and cardiovascular disease. New England Journal of Medicine. 1996;334:1145–1149. doi: 10.1056/NEJM199605023341801. al e. [DOI] [PubMed] [Google Scholar]
- 12.Rupnick MA, Panigrahy D, Zhang CY, Dallabrida SM, Lowell BB, Langer R, Folkman MJ. Adipose tissue mass can be regulated through the vasculature. Proc Natl Acad Sci U S A. 2002;99(16):10730–10735. doi: 10.1073/pnas.162349799. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Giovannucci E, Rimm EB, Liu Y, Leitzmann M, Wu K, Stampfer MJ, Willett WC. Body mass index and risk of prostate cancer in U.S. health professionals. J Natl Cancer Inst. 2003;95(16):1240–1244. doi: 10.1093/jnci/djg009. [DOI] [PubMed] [Google Scholar]
- 14.Rodriguez C, Patel AV, Calle EE, Jacobs EJ, Chao A, Thun MJ. Body mass index, height, and prostate cancer mortality in two large cohorts of adult men in the United States. Cancer Epidemiol Biomarkers Prev. 2001;10(4):345–353. [PubMed] [Google Scholar]
- 15.Macpherson GR, Singh AS, Bennett CL, Venzon DJ, Liewehr DJ, Franks ME, Dahut WL, Kantoff PW, Price DK, Figg WD. Genotyping and functional analysis of the D104N variant of human endostatin. Cancer Biol Ther. 2004;3(12):1298–1303. doi: 10.4161/cbt.3.12.1453. [DOI] [PubMed] [Google Scholar]
- 16.Ma J, Li H, Giovannucci E, Mucci L, Qiu W, Gaziano JM, Pollak M, Stampfer MJ. Prediagnostic body-mass index, plasma C-peptide concentration, and prostate cancer-specific mortality in men with prostate cancer: a long-term survival analysis. Lancet Oncol. 2008;9(11):1039–1047. doi: 10.1016/S1470-2045(08)70235-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Folkman J. Role of angiogenesis in tumor growth and metastasis. Semin Oncol. 2002;29(6 Suppl 16):15–18. doi: 10.1053/sonc.2002.37263. [DOI] [PubMed] [Google Scholar]
- 18.Folkman J. Fundamental concepts of the angiogenic process. Curr Mol Med. 2003;3(7):643–651. doi: 10.2174/1566524033479465. [DOI] [PubMed] [Google Scholar]
- 19.Fukai N, Eklund L, Marneros AG, Oh SP, Keene DR, Tamarkin L, Niemela M, Ilves M, Li E, Pihlajaniemi T, Olsen BR. Lack of collagen XVIII/endostatin results in eye abnormalities. Embo J. 2002;21(7):1535–1544. doi: 10.1093/emboj/21.7.1535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Li HC, Cai QY, Shinohara ET, Cai H, Cao C, Wang ZF, Teng M, Zheng W, Lu B. Endostatin polymorphism 4349G/A(D104N) is not associated with aggressiveness of disease in prostate [corrected] cancer. Dis Markers. 2005;21(1):37–41. doi: 10.1155/2005/650109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Nam RK, Zhang WW, Trachtenberg J, Jewett MA, Emami M, Vesprini D, Chu W, Ho M, Sweet J, Evans A, Toi A, Pollak M, Narod SA. Comprehensive assessment of candidate genes and serological markers for the detection of prostate cancer. Cancer Epidemiol Biomarkers Prev. 2003;12(12):1429–1437. [PubMed] [Google Scholar]
- 22.Silha JV, Krsek M, Sucharda P, Murphy LJ. Angiogenic factors are elevated in overweight and obese individuals. Int J Obes (Lond) 2005;29(11):1308–1314. doi: 10.1038/sj.ijo.0802987. [DOI] [PubMed] [Google Scholar]