DNA methylation in repeat negative prostate biopsies as a marker of missed prostate cancer.
Aged
Biomarkers, Tumor
/ genetics
Biopsy
Case-Control Studies
Cross-Sectional Studies
DNA Methylation
Epigenesis, Genetic
Genetic Predisposition to Disease
Glutathione S-Transferase pi
/ genetics
Humans
Italy
Logistic Models
Longitudinal Studies
Male
Middle Aged
Prostatic Neoplasms
/ diagnosis
Sensitivity and Specificity
DNA methylation
Negative prostate biopsies
Prostate cancer
Prostate cancer diagnosis
Journal
Clinical epigenetics
ISSN: 1868-7083
Titre abrégé: Clin Epigenetics
Pays: Germany
ID NLM: 101516977
Informations de publication
Date de publication:
30 10 2019
30 10 2019
Historique:
received:
29
04
2019
accepted:
22
09
2019
entrez:
1
11
2019
pubmed:
2
11
2019
medline:
28
7
2020
Statut:
epublish
Résumé
Men often undergo repeat prostate biopsies because of suspicion of missed cancer. We assessed if (i) methylation of selected genes in prostate tissue vary with aging and (ii) methylation alterations in repeat biopsies predict missed prostate cancer. We conducted a case-control study among men who underwent at least two negative prostate biopsies followed by a sampling either positive (cases n = 111) or negative (controls n = 129) for prostate cancer between 1995 and 2014 at the University Hospital (Turin, Italy). Two pathology wards were included for replication purposes. We analyzed methylation of GSTP1, APC, PITX2, C1orf114, GABRE, and LINE-1 in the first two negative biopsies. Conditional logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) of the association between genes methylation and prostate cancer. Age at biopsy and time interval between the two negative biopsies were not associated with methylation levels of the selected genes in neither cases nor controls. GSTP1 methylation in the first and in the second negative biopsy was associated with prostate cancer detection [OR per 1% increase: 1.14 (95% CI 1.01-1.29) for the second biopsy and 1.21 (95% CI 1.07-1.37) for the highest methylation level (first or second biopsy)]. A threshold > 10% for GSTP1 methylation corresponded to a specificity of 0.98 (positive likelihood ratio 7.87). No clear association was found for the other genes. Results were consistent between wards. Our results suggest that GSTP1 methylation in negative prostate biopsies is stable over time and can predict missed cancer with high specificity.
Sections du résumé
BACKGROUND
Men often undergo repeat prostate biopsies because of suspicion of missed cancer. We assessed if (i) methylation of selected genes in prostate tissue vary with aging and (ii) methylation alterations in repeat biopsies predict missed prostate cancer.
METHODS
We conducted a case-control study among men who underwent at least two negative prostate biopsies followed by a sampling either positive (cases n = 111) or negative (controls n = 129) for prostate cancer between 1995 and 2014 at the University Hospital (Turin, Italy). Two pathology wards were included for replication purposes. We analyzed methylation of GSTP1, APC, PITX2, C1orf114, GABRE, and LINE-1 in the first two negative biopsies. Conditional logistic regression was used to estimate odds ratios (ORs) and 95% confidence intervals (CIs) of the association between genes methylation and prostate cancer.
RESULTS
Age at biopsy and time interval between the two negative biopsies were not associated with methylation levels of the selected genes in neither cases nor controls. GSTP1 methylation in the first and in the second negative biopsy was associated with prostate cancer detection [OR per 1% increase: 1.14 (95% CI 1.01-1.29) for the second biopsy and 1.21 (95% CI 1.07-1.37) for the highest methylation level (first or second biopsy)]. A threshold > 10% for GSTP1 methylation corresponded to a specificity of 0.98 (positive likelihood ratio 7.87). No clear association was found for the other genes. Results were consistent between wards.
CONCLUSIONS
Our results suggest that GSTP1 methylation in negative prostate biopsies is stable over time and can predict missed cancer with high specificity.
Identifiants
pubmed: 31666119
doi: 10.1186/s13148-019-0746-6
pii: 10.1186/s13148-019-0746-6
pmc: PMC6820908
doi:
Substances chimiques
Biomarkers, Tumor
0
GSTP1 protein, human
EC 2.5.1.18
Glutathione S-Transferase pi
EC 2.5.1.18
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
152Références
J Urol. 2013 Mar;189(3):1110-6
pubmed: 22999998
Lancet Oncol. 2017 Feb;18(2):221-229
pubmed: 28094199
Cancer Epidemiol Biomarkers Prev. 2009 Oct;18(10):2717-22
pubmed: 19755651
Lancet. 2017 Feb 25;389(10071):815-822
pubmed: 28110982
Epigenetics. 2017 Jan 2;12(1):11-18
pubmed: 27892790
Eur Urol. 2015 Dec;68(6):1045-53
pubmed: 25656808
J Urol. 2014 Oct;192(4):1081-7
pubmed: 24747657
Curr Opin Urol. 2014 Mar;24(2):155-61
pubmed: 24451092
BJU Int. 2012 Jul;110(1):56-62
pubmed: 22077694
Curr Opin Urol. 2015 Jan;25(1):83-8
pubmed: 25405932
Eur Urol. 2017 Mar;71(3):353-365
pubmed: 27543165
PLoS One. 2013 Jul 09;8(7):e68162
pubmed: 23874531
Int J Mol Sci. 2014 Sep 18;15(9):16544-76
pubmed: 25238417
Eur Urol. 2013 Dec;64(6):876-92
pubmed: 23787356
N Engl J Med. 2018 May 10;378(19):1767-1777
pubmed: 29552975
Clin Cancer Res. 2007 Jul 1;13(13):3796-802
pubmed: 17606710
Eur Urol. 2017 Apr;71(4):618-629
pubmed: 27568654
JAMA. 2015 Jan 27;313(4):390-7
pubmed: 25626035
J Urol. 2013 Jun;189(6):2335-41
pubmed: 23159584
J Clin Oncol. 2013 Sep 10;31(26):3250-8
pubmed: 23918943
Clin Cancer Res. 2016 Feb 15;22(4):984-92
pubmed: 26475336