Open Journal of Urology, 2013, 3, 287-292
Published Online November 2013 (
Open Access OJU
RNase L Variants Do Not Appear to Impact on Clinical
Features of Sporadic Prostate Cancer Patients*
Frank T. D’Arcy1,2, Ruth Foley1, Thomas H. Lynch2
1Institute of Molecular Medicine, Trinity College, Dublin, Ireland
2Department of Urology, St. James Hospital, Dublin, Ireland
Received November 4, 2013; revised November 23, 2013; accepted November 28, 2013
Copyright © 2013 Frank T. D’Arcy et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Introduction: Prostate cancer is the most common non-cutaneous male cancers, contributing to significant mortality
rates globally. Mutations of RNase L, an enzyme involved in inflammatory and immunological pathways, have been
speculated to predispose to cancer. This study assesses three different mutations of the RNase L gene in Irish prostate
cancer patients, including one linked with general cancer susceptibility never investigated before in prostate cancer
(rs3738579), and reports on links with aggressive cancer. Methods: 134 patients had their RNase L mutation status
determined by polymerase chain reaction (PCR) of serum DNA. Complementary clinical details for each patient are
statistically analysed. Results: No link to age of diagnosis, high grade disease or prostate specific antigen (PSA) level at
diagnosis was demonstrated with any of the studied single nucleotide polymorphisms (SNP). The SNP variation was
consistent with that of published international series. Conclusion: SNP genotypic frequencies in Ireland are consistent
with international findings. The studied RNase L mutations including rs3738579 do not appear to have a significant
impact on our patient population.
Keywords: RNase L; R462Q; D541E; Prostate Cancer
1. Introduction
RNase L is a gene found on the hereditary prostate can-
cer (HPC) locus of chromosome 1 that codes for a latent
endoribonuclease. This enzyme participates in an interferon
inducible RNA decay pathway that has a role in inflam-
mation and cellular immunity against viral infection. Sus-
tained activation will lead to apoptosis. The gene is 741
amino acids long with 8 exons and is roughly 13 kilo-
bases. It is converted from an inactive monomeric form
to a potent dimeric structure by the action of a series of
2’ to 5’ linked oligodenylates, commonly known as 2 -
5 A.
RNase L has been a candidate gene for prostate cancer
researchers for a number of years and many variants in-
cluding R462Q [1], E265X [1], M1I [2] and 471ΔAAAG [3]
have been described.
The R462Q RNase L mis-sense mutation (rs486927)
arises when a substitution of the “G” to “A” base along
exon 1 at mRNA position 1552 gives an amino acid
change of argenine to glutamine at amino acid position
462. The consequence of this is a variant of the gene that
produces an enzyme functioning at a third of the wild-
type enzyme’s efficiency [4]. It has been of interest to
researchers since it was linked to prostate cancer in 2002
[2]. Subsequent studies have been mixed with some show-
ing a link with this variant and familial prostate cancer
[5,6], whilst others have not [7]. A meta-analysis of 7
papers has failed to show a significant link between this
mutation and prostate cancer regardless of ethnicity or
family history [8].
D541E (rs627928) comes about following a substitution
of the “T” base at mRNA position 1790 with “G”, caus-
ing a change in the protein at amino acid position 541
from aspartate to glutamic acid. This area of the gene
codes for a region within the protein kinase domain of
the enzyme, an area with an important role in dimeriza-
tion of the protein into its active state. Despite the muta-
tion, this enzyme has comparable catalytic activity with
the wildtype variant [4]. Its significance is the source of
some debate within the literature. Studies have linked
*All funding for described work came via Trinity College Dublin and
The Irish Cancer Society.
this change to an increased risk of familial prostate can-
cer [7] and sporadic, metastatic disease [9], but in a Japa-
nese study [10] the wildtype variant was actually linked
to an increased risk of familial prostate cancer. Other
investigators have found no significant associations with
it [1,5,11].
An undescribed mutation (rs3738579) located in the 5’
untranscribed region (5’-UTR) has been identified as a
general indicator of increased cancer susceptibility [12].
The exact role of this variant on expression is unclear but
mutations are seen disproportionally in cervical as well
as head and neck squamous cell cancers.
Work in the fields of cervical cancer (human papillo-
mavirus) [13], gastric cancer (helicobacter pylori) [14],
B-cell lymphomas (Epstein-Barr virus) [15] and osteo-
genic sarcomas (chronic osteomyelitis) [16] have all found
definite links between chronic infection and neoplasia.
Whilst no such links have yet been established in prostate
cancer, there are inconsistent reports in the literature of
positive associations of the disease with sexually trans-
mitted infections (STIs) [17-19], and one study finding
men with 25 or more sexual partners being 2.8 times
more likely to be diagnosed with prostate cancer than a
man with 5 or less partners [20]. If inflammation or viral
infection plays a role in prostate cancer development, it
is speculated that polymorphisms in genes involved in
the inflammatory and infectious disease pathways like
RNase L could be important.
This study aims to sequence RNase L single nucleo-
tide polymorphisms (SNPs) in men with known cases of
sporadic prostate cancer and correlate to the clinical as-
pects of the cancer.
2. Methods and Materials
Patient samples were obtained from the Prostate Cancer
Research Consortium (PCRC) bio-repository. This is a
collaboration of three University institutions affiliated
with five major hospitals. Complementary clinical details
relating to each patient are also logged into a password
protected, secure database. All volunteers were pre-op-
erative radical prostatectomy patients with localised pros-
tate cancer, enrolled with full consent in line with ethical
advice. All cases of prostate cancer analysed were spo-
Using the database of known SNPs published online at
the National Centre for Biotechnology Information (NCBI),
the sequence of code surrounding the SNPs of interest
were obtained.
Primers were designed using Primer Express (Applied
Biosystems, USA). Details of the actual sequence of in-
terest and the primer set are given in Table 1. Primers
were supplied by MWG and were purchased at 0.01
mol scale, HPS purification.
DNA was extracted from peripheral blood DNA by the
Autopure automated system (Qiagen, USA) which uses
puregene chemistry. The product was purified (Qiaquick
PCR purification kit from Qiagen, USA) and sequencing
reactions with a fluorescent dye terminator (BigDye v3.1,
ABI) were performed. Results were interpreted with Se-
quencing Analysis v5.1 from Applied Biosystems.
Polymerase chain reaction (PCR) was carried out with
the product initially being exposed to 94˚C for thirty
seconds, followed by 35 cycles where the temperature
alternated from 94˚C for 30 more seconds, then [53.1˚C
(R462Q)/53˚C (D541E)/55˚C (5’-UTR)] for 30 seconds
followed by a minute at 72˚C. The reaction culminated in
30 minutes exposure to 72˚C.
3. Results
Table 2 illustrates the frequencies of each RNase L SNP
studied in the prostate cancer population. These figures
are in keeping with those published on the NCBI database.
9% of patients had the homozygous AA genotype
(R462Q), which corresponds to the less efficient variant of
the RNase L enzyme, while 52.2% were heterozygous
for this SNP. Table 3 outlines R462Q status versus histo-
logical disease aggression. On initial inspection it would
appear that a disproportionate amount of men with the
AA (i.e. the R462Q) genotype have tumours that are of
Gleason score 7 or greater (7 out of 11). This association,
whilst a trend did not demonstrate statistical significance
(p = 0.258). No association was seen between this SNP
and either percentage gland involvement by tumour (p =
0.57) or number of affected first-degree relatives (p =
0.69). Furthermore, anova 1-way p values for age and
prostate specific antigen (PSA) levels at diagnosis also
failed to demonstrate any statistical significance (p =
Table 1. Description of primers selected for each SNP studied.
DbSNP ID Base Amino acid Primer Sequence
rs486907 G/A R462Q Reverse 5’-TGCAGATCCTGGTGGGTGTA -3’
rs627928 T/G D541E Reverse 5’-TGAGGTCCTTAGTTTCCTCATCT-3’
rs3738579 C/T - Reverse 5’-AATGCCACCTGCTACCACTT-3’
Open Access OJU
F. T. D’ARCY ET AL. 289
Table 2. Summary of sequencing results for each SNP in
men with prostate cancer.
SNP Genotype Frequency (%)
AA (Mutation) 12/134 (9%)
AG 70/134 (52.2%)
GG (Wildtype) 52/134 (38.8%)
GG (Mutation) 35/91 (38.5%)
GT 38/91 (41.7%) D541E
TT (Wildtype) 18/91 (19.8%)
CC (Mutation) 18/136 (13.2%)
CT 55/136 (40.4%) 5’-UTR (rs3738579)
TT (Wildtype) 63/136 (46.4%)
Table 3. Gleason score and SNP status at the exon 1/R462Q
Gleason Score AA AG GGTotal numbers P value
4 - 6 4 39 23 66
7 - 9 7 26 23 56
Total 11 65 46 122 p = 0.258
0.60 and 0.44 respectively).
38.5% of prostate cancer patients in the study were
shown to carry the D541E mutation. Analysis of comple-
mentary clinical data revealed no association with either
percentage gland involvement (p = 0.58), number of
first-degree relatives affected (p = 0.058) or age at diag-
nosis (p = 0.68). Analysis of PSA at diagnosis (on Table
4) revealed that men with the mutation had a much
higher PSA at diagnosis however this fell just short of
statistical significance (p = 0.06).
Analysis of Gleason score of tumour and SNP status
failed to show any statistically significant association (p
= 0.68).
Statistical analysis of Gleason score (p = 0.24), per-
centage gland involvement (p = 0.5), family history (p =
0.49), age (p = 0.29) or PSA at diagnosis (p = 0.77)
failed to reveal any association with this SNP (see Table
4. Discussion and Review of the Literature
To our knowledge this is the first study of the RNase L
SNP rs3738579 in the context of prostate cancer and also
adds to the debate about the significance of D541E and
There are a number of studies in the literature exam-
ining prostate cancer and the R462Q mutation. The results
of these are summarised in Table 6 and a comparison of
the frequency that the mutation is detected across these
series are displayed in the Forest plot given in Figure 1.
This shows our findings to be consistent with interna-
tionally published series.
Table 4. Breakdown of PSA values (given as ng/ml) versus
genotype at D541E SNP.
Genotype Mean ± SDMedian Anova 1-way p value
GG (D541E)8.42 ± 3.138.25
GT 7.08 ± 3.266.25
TT 6.84 ± 2.076.6
Table 5. Analysis of low and high grade tumours versus
5’-UTR locus SNP status.
Gleason score CCCTTT Total numbersp value
4 - 6 8 29 29 66
7 - 9 8 22 26 56
Total 16 51 55 122 0.24
0.000.05 0.100.15 0.20 0.25
Wa ng
of AA
on Exon 1
AA Proportion
Figure 1. Expression of mutated form of Exon 1, AA (R 462Q),
as a proportion of total numbers examined for all major
studies, expressed on a Forest plot. Y-axis has study’s au-
thor’s name whilst X-axis contains proportion of AA geno-
type given with 95% confidence limits.
D541E mutation has been the focus of multiple studies.
We fail to demonstrate a significant clinical association
with this mutation. Results summary and comparison of
frequency of mutation are given in Table 7 and Figure 2
Rokman’s study focused on 492 patients with prostate
cancer, 47 of whom reported a positive family history.
They were consecutive cases diagnosed at the University
Hospital, Tampere, Finland, over a three-year period.
This population would have differed from the one we are
reporting on by including men with both organ confined
and metastatic disease. Control samples came from 566
healthy male blood donors. This study failed to show any
significant difference in D541E status between prostate
cancer, BPH and normal control groups and concluded
that this mutation doesn’t have an important role in pros-
tate cancer within their population. It did however dem-
onstrate an association between R462Q status and heredi-
tary prostate cancer.
Wang’s study came from Minnesota in the United
States and included a total of 825 patients with prostate
Open Access OJU
Table 6. Comparison of results of analysis of R462Q mutation (“AA” genotype) throughout different studies within the litera-
Study Author Population Patient group AA (Mut) GenotypeAG Genotype GG Genotype Total
Prostate cancer 39 (16.7%) 106 (45.5%) 88 (37.8%) 233
Rokman [1] Finnish Control 23 (13.1%) 84 (47.7%) 69 (39.2%) 176
Prostate cancer 102 (11.1%) 427 (46.5%) 389 (42.4%) 918
Wang [5] Hispanic and White. USA Control 67 (13.5%) 233 (47.3%) 193 (39.1%) 493
Prostate cancer 247 (15.2%) 778 (48.0%) 597 (36.8%) 1622
Wiklund [7] Swedish Control 115 (14.4%) 384 (48.2%) 297 (37.4%) 796
Prostate cancer 59 (16.3%) 171 (47.1%) 133 (36.6%) 363
Maier [11] German Control 37 (17.9%) 97 (46.9%) 73 (35.2%) 207
D’Arcy Irish Prostate cancer 12 (9.0%) 70 (52.2%) 52 (38.8%) 134
Table 7. Comparison of results of analysis of D541E mutation (“GG” genotype) throughout different studies within the litera-
Study Author Population Patient group GG (Mut) Genotype GT Genotype TT Genotype Total
Prostate cancer 78 (33.5%) 126 (54.1%) 29 (12.4%) 233
Rokman [1] Finnish Control 56 (31.8%) 91 (51.7%) 29 (16.5%) 176
Prostate cancer 181 (19.5%) 476 (51.2%) 272 (29.3%) 929
Wang [5] Hispanic and White. USA Control 107 (21.1%) 228 (44.9%) 173 (34.0%) 508
Prostate cancer 462 (33.9%) 668 (49.0%) 233 (17.1%) 1363
Wiklund [7] Swedish Control 257 (32.5%) 372 (47.0%) 162 (20.5%) 791
Prostate cancer 125 (34.4%) 176 (48.5%) 62 (17.1%) 363
Maier [11] German Control 69 (33.3%) 97 (46.9%) 41 (19.8%) 207
Prostate cancer 55 (36.7%) 73 (48.7%) 22 (14.6%) 150
Noonan-Wheeler [9] European-Americans Control 44 (25.7%) 94 (55.0%) 33 (19.3%) 171
Prostate cancer 51 (50.5%) 32 (31.7%) 18 (17.8%) 101
Nakazato [10] Japanese Control 59 (56.2%) 43 (40.9%) 3 (2.9%) 105
D’Arcy Irish Prostate cancer 35 (38.5%) 38 (41.7%) 18 (19.8%) 91
0.000.20 0.400.60 0.80
Wan g
Na ka za to
Rokma n
of GG
representa tion
onExon 3
GG Proportion
Figure 2. Expression of mutate d form of Exo n 3, GG ( D 541E),
as a proportion of total numbers examined for all major
studies, expressed on a Forest plot. Y-axis has study’s au-
thor’s name whilst X-axis contains proportion of GG geno-
type given with 95% confidence limits.
cancer, made up of 326 familial prostate cancer patients
and 499 cases of sporadic prostate cancer, all of the latter
group comprising of organ confined disease. Control group
came from men of a similar age from the local popula-
tion (Rochester Epidemiology Project), who were invited
to enrol and subsequently had a careful history, physical
examination and if necessary a TRUS biopsy. Once pros-
tate cancer had been ruled out, they were incorporated as
a control group. In total there were 510 such men. This
study also linked R462Q status with hereditary prostate
cancer and in such men disease was associated with an
earlier age of onset. Like Rokman’s work, this study also
failed to demonstrate a correlation between the D541E
mutation and prostate cancer risk.
Wiklund’s series came from patients registered on
Cancer Prostate, a Swedish nationwide database of pros-
tate cancer patients younger than 79 diagnosed between
July 1st 2001 and 30th September 2002. Patients were
invited to take part in the study and a blood sample for
analysis was taken. In total, 1636 prostate cancer patients
with a full spectrum of disease from early to advanced
were included. Control samples were matched to similar
people from the Swedish Population registry, who were
invited to take part. This involved filling a questionnaire
and giving a blood sample. In total 801 controls were
analysed. No significant association between R462Q status
and prostate cancer risk was demonstrated in this study.
It did reveal a significant link (p = 0.03) between familial
Open Access OJU
F. T. D’ARCY ET AL. 291
prostate cancer risk and D541E mutation, however no link
between this variant and either age at diagnosis or tu-
mour aggressiveness was found.
Maier’s patients were recruited from the Prostate Can-
cer Genetics Project. This is a database made up of men
from predominately from the South of Germany, the ma-
jority of whom would have undergone radical prostatec-
tomies. Patients are encouraged to enrol in this study by
their urologist and there are no selection criteria. At risk
families are identified by interview of the patient. There
were a total of 303 of such patients. 227 sporadic cases
were also sequenced as well as 207 control samples.
These samples came from healthy, elderly men with no
history of prostate cancer and negative DRE and/or nor-
mal PSA levels. D541E or R462Q failed to demonstrate any
significant association with prostate cancer in this study.
Noonan-Wheeler’s study originated in Missouri, USA.
It examined RNase L in men with aggressive, metastatic
cancer and healthy controls. Patients were recruited from
the outpatient department and were required to have a
PSA over 50 ng/ml or radiological/pathological evidence
of metastatic disease. The control group consisted of men
older than 75 with normal PSA levels and rectal exami-
nation with no background of prostate cancer. There
were a total of 150 patients and 171 controls examined. It
was shown that D541E was over-represented (p = 0.045)
in patients with metastatic disease, and concluded that
such patients were at an increased risk for sporadic, me-
tastatic disease (OR = 1.68).
Nakazato’s study came from Japan and examined
RNase L in familial prostate cancer cases and healthy
controls. It comprised 101 patients with a positive family
history of cancer (29 of whom had 3 or more affected
family members) and 105 controls. Prostate cancer pa-
tients ranged in age from 40 to 88 years, encompassed
organ confined and metastatic disease and had 76 men
with a Gleason score of 7 or greater and the remaining 26
with a Gleason score of 6 or less. Control cases were
recruited from the outpatients department and were of
similar age. Patients with an elevated PSA or abnormal
DRE were excluded from this group. Interestingly this
study demonstrated a significant link between the wild-
type DD variant and familial prostate cancer (p = 0.0004,
OR = 7.37), possibly an observation unique to the Japa-
nese population.
5. Conclusion
In conclusion, no statistically significant correlation was
proven in the Irish population between Gleason score,
percentage gland involvement, patient age, PSA or fam-
ily history with any of the studied SNPs. In particular the
SNP rs3738579 failed to highlight men with significant
clinical cancer.
[1] A. Rökman, T. Ikonen, E. H. Seppälä, N. Nupponen, V.
Autio, N. Mononen, J. Bailey-Wilson, J. Trent, J. Carpten,
M. P. Matikainen, P. A. Koivisto, T. L. Tammela, O. P.
Kallioniemi and J. Schleutker, “Germline Alterations of
the RNASEL Gene, a Candidate HPC1 Gene at 1q25, in
Patients and Families with Prostate Cancer,” The Ameri-
can Journal of Human Genetics, Vol. 70, No. 5, 2002, pp.
[2] J. Carpten, N. Nupponen, S. Isaacs, R. Sood, C. Robbins,
J. Xu, M. Faruque, T. Moses, C. Ewing, E. Gillanders, P.
Hu, P. Bujnovszky, I. Makalowska, A. Baffoe-Bonnie, D.
Faith, J. Smith, D. Stephan, K. Wiley, M. Brownstein, D.
Gildea, B. Kelly, R. Jenkins, G. Hostetter, M. Matikainen,
J. Schleutker, K. Klinger, T. Connors, Y. Xiang, Z. Wang,
A. De Marzo, N. Papadopoulos, O. P. Kallioniemi, R.
Burk, D. Meyers, H. Gronberg, P. Meltzer, R. Silverman,
J. Bailey-Wilson, P. Walsh, W. Isaacs and J. Trent, “Germ-
line Mutations in the Ribonuclease L Gene in Families
Showing Linkage with HPC1,” Nature Genetics, Vol. 30,
No. 2, 2002, pp. 181-184.
[3] A. Orr-Urtreger, A. Bar-Shira, D. Bercovich, N. Mata-
rasso, U. Rozovsky, S. Rosner, S. Soloviov, G. Rennert,
L. Kadouri, A. Hubert, H. Rennert and H. Matzkin,
“RNASEL Mutation Screening and Association Study in
Ashkenazi and Non-Ashkenazi Prostate Cancer Patients,”
Cancer Epidemiology, Biomarkers & Prevention, Vol. 15,
No. 3, 2006, pp. 474-479.
[4] Y. Xiang, Z. Wang, J. Murakami, S. Plummer, E. A. Klein,
J. D. Carpten, J. M. Trent, W. B. Isaacs, G. Casey and R.
H. Silverman, “Effects of RNase L Mutations Associated
with Prostate Cancer on Apoptosis Induced by 2’,5’-Oli-
goadenylates,” Cancer Research, Vol. 63, No. 20, 2003,
pp. 6795-6801.
[5] L. Wang, S. K. McDonnell, D. A. Elkins, S. L. Slager, E.
Christensen, A. F. Marks, J. M. Cunningham, B. J. Peter-
son, S. J. Jacobsen, J. R. Cerhan, M. L. Blute, D. J.
Schaid and S. N. Thibodeau, “No Association of Germ-
line Alteration of MSR1 with Prostate Cancer Risk,” Na-
ture Genetics, Vol. 35, 2003, pp. 128-129.
[6] G. Casey, P. J. Neville, S. J. Plummer, Y. Xiang, L. M.
Krumroy, E. A. Klein, W. J. Catalona, N. Nupponen, J. D.
Carpten, J. M. Trent, R. H. Silverman and J. S. Witte,
“RNASEL Arg462Gln Variant Is Implicated in up to 13%
of Prostate Cancer Cases,” Nature Genetics, Vol. 32, No.
4, 2002, pp. 581-583.
[7] F. Wiklund, B. A. Jonsson, A. J. Brookes, L. Stromqvist,
J. Adolfsson, M. Emanuelsson, H. O. Adami, K. Augusts-
son-Balter and H. Gronberg, “Genetic Analysis of the
RNASEL Gene in Hereditary, Familial, and Sporadic Pros-
tate Cancer,” Clinical Cancer Research, Vol. 10, No. 21,
2004, pp. 7150-7156.
[8] H. Li and B. C. Tai, “RNASEL Gene Polymorphisms and
the Risk of Prostate Cancer: A Meta-Analysis,” Clinical
Cancer Research, Vol. 12, 2006, pp. 5713-5719.
Open Access OJU
Open Access OJU
[9] F. C. Noonan-Wheeler, W. Wu, K. A. Roehl, A. Klim, J.
Haugen, B. K. Suarez and A. S. Kibel, “Association of
Hereditary Prostate Cancer Gene Polymorphic Variants
with Sporadic Aggressive Prostate Carcinoma,” Prostate,
Vol. 66, No. 1, 2006, pp. 49-56.
[10] H. Nakazato, K. Suzuki, H. Matsui, N. Ohtake, S. Nakata
and H. Yamanaka, “Role of Genetic Polymorphisms of
the RNASEL Gene on Familial Prostate Cancer Risk in a
Japanese Population,” British Journal of Cancer, Vol. 89,
No. 4, 2003, pp. 691-696.
[11] C. Maier, J. Haeusler, K. Herkommer, Z. Vesovic, J. Hoe-
gel, W. Vogel and T. Paiss, “Mutation Screening and
Association Study of RNASEL as a Prostate Cancer Sus-
ceptibility Gene,” British Journal of Cancer, Vol. 92,
2005, pp. 1159-1164.
[12] B. E. Madsen, E. M. Ramos, M. Boulard, K. Duda, J.
Overgaard, M. Nordsmark, C. Wiuf and L. L. Hansen,
“Germline Mutation in RNASEL Predicts Increased Risk
of Head and Neck, Uterine Cervix and Breast Cancer,”
PLoS One, Vol. 3, No. 6, 2008, Article ID: e2492.
[13] S. Nair and M. R. Pillai, “Human Papillomavirus and
Disease Mechanisms: Relevance to Oral and Cervical Can-
cers,” Oral Diseases, Vol. 11, 2005, pp. 350-359.
[14] A. Leodolter, M. Naumann and P. Malfertheiner, “Pre-
vention of Gastric Cancer by Helicobacter pylori Eradi-
cation,” Digestive Diseases, Vol. 22, 2004, pp. 313-319.
[15] M. A. Epstein, B. G. Achong and Y. M. Barr, “Virus Par-
ticles in Cultured Lymphoblasts from Burkitt’s Lym-
phoma,” Lancet, Vol. 15, 1964, pp. 702-703.
[16] J. E. McGrory, D. J. Pritchard, K. K. Unni, D. Ilstrup and
C. M. Rowland, “Malignant Lesions Arising in Chronic
Osteomyelitis,” Clinical Orthopaedics and Related Re-
search, Vol. 362, 1999, pp. 181-189.
[17] L. K. Dennis and D. V. Dawson, “Meta-Analysis of Mea-
sures of Sexual Activity and Prostate Cancer,” Epidemi-
ology, Vol. 13, 2002, pp. 72-79.
[18] R. B. Hayes, L. M. Pottern, H. Strickler, C. Rabkin, V.
Pope, G. M. Swanson, R. S. Greenberg, J. B. Schoenberg,
J. Liff, A. G. Schwartz, R. N. Hoover and J. F. Fraumeni
Jr., “Sexual Behaviour, STDs and Risks for Prostate Can-
cer,” British Journal of Cancer, Vol. 82, 2000, pp. 718-
[19] M. L. Taylor, A. G. Mainous 3rd and B. J. Wells, “Pros-
tate Cancer and Sexually Transmitted Diseases: A Meta-
Analysis,” Family Medicine Journal, Vol. 37, 2005, pp.
[20] A. V. Sarma, J. C. McLaughlin, L. P. Wallner, R. L. Dunn,
K. A. Cooney, D. Schottenfeld, J. E. Montie and J. T.
Wei, “Sexual Behavior, Sexually Transmitted Diseases
and Prostatitis: The Risk of Prostate Cancer in Black
Men,” Journal of Urology, Vol. 176, No. 3, 2006, pp.