Citations

Abstract Background: Variation in mitochondrial genes in cancer cells alters the mitochondrial bioenergetic and biosynthetic state, which mediates cross-talk between mitochondria and the nucleus to modulate many cellular processes. The objective of this study was to comprehensively evaluate mitochondria-nuclear cross-talk in terms of SNP-SNP interactions associated with invasive epithelial ovarian cancer (EOC) risk, which may play an important role in unveiling the underlying mechanism of this complex lethal disease. Methods: Our analyses were based on 1,938 invasive EOC cases and 2,009 controls with greater than 80% European ancestry from five case-control studies genotyped using the Illumina 610quad chip. We evaluated 2,488 SNPs in over 1,500 genes in eight mitochondria-related pathways (apoptosis, oxidation-reduction, epithelial-mesenchymal transition, immunomoduation, transmembrane transport, small GTPases, BRCA1-interactors and circadian rhythm). SNPs with strong linkage disequilibrium of r2 > 0.8 were excluded. We examined 2-way SNP-SNP interactions associated with invasive EOC risk using the two-stage Random Forests plus Multivariate Adaptive Regression Splines (TRM) approach. Bootstrapping was further performed to select important variables and reduce false positive findings. Based on the scree plot using 1,000 bootstrap samples, the factors with more than 4% of bootstrap frequency were defined as significant factors. We then included each identified factors separately in model adjusting for study site and first principal component representing European ancestry. Results: Five significant SNP-SNP interactions were observed: (rs2270799 in IQSEC3 (12p13.33) + rs933305 in CUX2 (12q24.3), (p= 6.3x10−5); rs6063251 in PREX1 (20q13.3) + rs9705 in SLC39A8 (4q22-q24), (p= 3.5x10−5); rs933305 in CUX2 (12q24.3) + rs1712957 in ARHGAP15 (2q22.2-q22.3), (p=4.2x10−7), rs2861828 in SLIT1 (10q23.3-q24) + rs6063251 in PREX1 (20q13.3), (p=2.5x10−7); rs10954593 in SEMA3C (7q21-q31) + rs933305 in CUX2 (12q24.3), (p=2.8x10−3). Four of the five interactions involved SNPs from genes in the small GTPases pathway (IQSEC3; PREX1 (twice) and ARHGAP15). In addition, three of the five interaction pairs involved the CUX2 gene. This gene encodes a protein which contains three CUT domains and a homeodomain, which are DNA-binding motifs. The main mitochondrial SNP-SNP interaction involved the nuclear mitochondria gene SLC39A8. Conclusions: Our results suggest that genetic variation in mitochondrial-related pathways may be associated with invasive EOC risk. These findings highlight important genes/pathways that may interact as a network to influence invasive EOC risk and may generate novel therapeutic targets. External validation will be done using other independent studies for future analyses. Citation Format: Hui-Yi Lin, Ya-Yu Tsai, Y. Ann Chen, Zhihua Chen, Xiaotao Qu, Ellen L. Goode, Joellen Schildkraut, Edwin Iversen, Steven A. Narod, Thomas A. Sellers, Catherine M. Phelan. SNP-SNP interactions in mitochondria-related pathways are associated with invasive Epithelial Ovarian Cancer (EOC) risk . [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4839. doi:10.1158/1538-7445.AM2013-4839

Abstract Since most GWAS hits fall in non-coding regions of the genome and long non-coding RNAs (lncRNAs) are emerging as drivers of carcinogenesis, we hypothesized that SNPs in lncRNAs influence epithelial ovarian cancer (EOC) risk. A comprehensive lncRNA database (lncRNA db) that contained 104 human lncRNAs when downloaded in November 2011 was used to identify 1,737 variants in 63 unique lncRNAs that were represented in three genome-wide association studies from North America, the United Kingdom, and Poland (3,995 EOC cases and 3,277 controls of European background). SNPs with MAF <5% were excluded, and missing genotypes were inferred with Mach using the HapMap CEU population. Unconditional logistic regression treating alternate alleles as an ordinal variable was used to evaluate individual SNP-EOC risk associations. Subgroup analysis was conducted for serous adenocarcinomas (N= 2,066), the main histologic subtype of EOC. Models were adjusted for study and European ancestry using the first two principal components (PCs). We reduced correlated signals (r2 ≥ 0.8) within each lncRNA and kept the most statistically significant variants based on Wald tests, resulting in 651 SNPs for further analyses. Fisher's product method was used to aggregate evidence of multiple variants within each lncRNA using 10,000 permutations. To adjust for multiple comparisons, a false discovery rate (FDR) of 10% was used. A replication phase was carried out for top-ranked SNPs from an additional 9,854 EOC cases (5,802 serous) and 17,633 controls genotyped through an international effort known as the Collaborative Oncological Gene-Environment study (COGS). Four lncRNAs, WT1-AS (p = 0.005, n = 2 SNPs), Hoxa11as (p = 0.013, n = 2), AK082072 (p = 0.016, n = 4) and H19 (p = 0.020, n = 2), were significantly associated with EOC risk among all histologies (FDR ≤ 10%). The SNP in WT1-AS most significantly associated with EOC risk was rs3809061 (T>C) (minor allele frequency= 36%, odds ratio (OR) = 0.91, CI: 0.84-0.97, p = 0.008). Among serous cases, rs3809061 was also protective (OR = 0.92, p = 0.04). These findings were replicated by a correlated SNP rs2301250 (r2 = 0.84) evaluated in COGS among serous cases (OR = 0.94, CI: 0.90-0.99, p = 0.009) and less significant among all histologies (OR = 0.96, CI: 0.93-1.00, p = 0.054). We further interrogated its functional role by performing an expression quantitative trait locus analysis on WTI-AS mRNA expression using the Cancer Genome Atlas (TCGA) data (N = 462 serous cases). WT1-AS expression was significantly higher among carriers of the variant allele (compared to TT homozygotes) using the Mann-Whitney U test (p = 0.001). WT1-AS is the antisense and possible regulator of WT1, a tumor suppressor gene reported to be prognostic in advanced EOC. These findings implicate germline lncRNA variants associated with EOC risk that merit further investigation. Citation Format: Yian Ann Chen, Zhihua Chen, Jennifer Permuth-Wey, Ya-Yu Tsai, Hui-Yi Lin, Xiaotao Qu, Kate Lawrenson, David Fenstermacher, Catherine M. Phelan, Alvaro Monteiro, Simon A. Gayther, Steven A. Narod, Rebecca Sutphen, Michael J. Birrer, Nicolas Wentzensen, Joellen M. Schildkraut, Ellen L. Goode, Paul Pharoah, Thomas Sellers. Variants in long non-coding RNAs are associated with epithelial ovarian cancer risk in a pooled analysis of three genome-wide association studies. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4579. doi:10.1158/1538-7445.AM2013-4579

Abstract Background. Transmembrane transport (GO:0055085) is the process whereby a solute/hormone/ion/iron is transported from one side of a membrane to the other. Disruption of these processes leads to defects in homeostasis which contributes to cancer risk. We hypothesized that germline single nucleotide polymorphisms (SNPs) in the cellular transport genes are associated with EOC risk and histopathology subtype. Methods. We genotyped 305 SNPs from 120 cellular transport related genes in 14,525 cases and 23,448 controls of European ancestry, 387 African, and 2,388 Asian ancestries from 43 studies in the Ovarian Cancer Association Consortium (OCAC). A custom Illumina iSelect designed for the Collaborative Oncological Gene-environment Study (COGS) was used. For women of European ancestry, both invasive cancers combined and the four main histological subtypes (serous (n=8,369), mucinous (n=943), endometroid (n=2,067) and clear cell (n=1,024) carcinoma) were analyzed, while for women of African and Asian ancestry only the serous subtype was analyzed adjusting for study site and first 5 principal components. Odds ratios and 95% confidence intervals were calculated using unconditional logistic regression. We fitted a log-additive genetic model. Results. In women of European ancestry, the strongest evidence of an association for all invasive EOCs was the iron transport gene, HEPH rs17216603 (OR=0.85, 95%CI=0.78-0.93, P=2.8x10−4), which was also the most significant association for serous histological subtype (OR=0.81, 95%CI=0.73-0.91, P=2.0x10−4). For endometrioid EOC, the strongest association was observed in the glucuronidation gene, UGT1A5 rs11563251 (OR=0.82, 95%CI=0.73-0.92, P=6.6x10−4). For clear cell and mucinous EOC, the thyroid hormone transport protein, SERPINA7 rs1804495 showed the strongest associations (OR=1.06, 95%CI=1.06-1.38, P=0.0041) and (OR=0.85, 95%CI=0.73-0.98, P=0.0407), respectively. For women of African ancestry, the strongest association for serous EOC was the glutathione conjugation gene, MGST rs6488840 (OR=0.55, 95%CI=0.37-0.82, P=0.0035). This SNP was not significant in women of European or Asian ancestry. In women of Asian ancestry, the HEPH rs17216603 (OR=1.45, 95%CI=1.15-1.83, P=0.0019) demonstrated the strongest association, although the directionality of risk was reversed compared to European women. Conclusions. Our results suggest that variation in transmembrane transport genes may influence EOC risk and histopathologic subtype in women of European, African and Asian ancestries, highlighting the differences in the disease etiology across populations. Additional epidemiological and functional studies are warranted to elucidate translational and clinical utility of these associations. Citation Format: Ganna Chornokur, Jonathan Tyrer, Hui-Yi Lin, Gang Han, Xiaotao Qu, Chen Zhihua, Ya-Yu Tsai, Ellen L. Goode, Julie M. Cunningham, Edwin Iversen, Susan J. Ramus, Andrew Berchuk, Joellen M. Schildkraut, Alvaro Monteiro, Simon A. Gayther, Steven A. Narod, Paul Pharoah, Thomas A. Sellers, Catherine M. Phelan. Variation in transmembrane transport genes influence epithelial ovarian cancer risk and histopathologic subtype. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 3644. doi:10.1158/1538-7445.AM2013-3644

Abstract Introduction The epithelial-mesenchymal transition (EMT) pathway contributes to epithelial ovarian cancer (EOC) progression. Previously we identified associations of single nucleotide polymorphisms (SNPs) in EMT-related genes and EOC risk (data not shown). To further investigate the role of EMT-related gene variants in EOC and to derive more reliable risk estimates, we evaluated associations for 795 SNPs from 278 EMT-related genes in a replication study of 43 studies within the Ovarian Cancer Association Consortium (OCAC). Methods The study population included 14,736 cases and 23,448 controls of European ancestry, 89 cases and 200 controls of African ancestry, and 249 cases and 1574 controls of Asian ancestry. The 795 SNPs were genotyped using an Illumina Infinium iSelect BeadChip as part of the Collaborative Oncological Gene-environment Study (COGS). For women of European ancestry, both invasive cancers combined and the four main histological subtypes (serous [n=8,372], endometroid [n=2,068], clear cell [n=1,025] and mucinous [n=943] were analyzed, while for women of African and Asian ancestries only the serous subtype was analyzed. SNP analyses were conducted using unconditional logistic regression under a log-additive model separately for women of each ancestry. All analyses were adjusted for study site and population substructure within each ancestry. Results For women of European ancestry, the strongest evidence of an association for invasive cancers combined was observed for IGF1R rs10794486 (OR=1.05, 95%CI=1.02-1.09, P=0.002), which was also associated with serous (P=0.019), but not the other histological subtypes. In the histological subtype-specific analysis, the most significant association for serous was observed at NRP2 rs3771044 (OR=1.09, 95%CI=1.03-1.15, P=0.00014), for endometroid at rs1770247 (13q31.3, OR=1.16, 95%CI=1.07-1.25, P=0.00028), for clear cell at rs4848300 (2q13, OR=0.84, 95%CI=0.76-0.93, P=0.00099) and for mucinous at SEMA4B rs8030039 (OR=1.09, 95%CI=1.01-1.17, P=0.00019). For women of African ancestry, the strongest association for serous histological subtype was observed at F9 rs6048 (OR=1.71, 95%CI=1.17-2.50, P=0.006). This SNP was not significant in women of European or Asian ancestry. The strongest association for women of Asian ancestry was IGF1R rs10794486 (OR=1.68, 95%CI=1.16-2.45, P=0.007), which was also significant in women of European, but not African, ancestry. Conclusion Findings from this large study provide additional evidence that variants in EMT-related genes may be associated with ovarian cancer risk and that EMT susceptibility loci may differ by ovarian cancer histological subtypes. Furthermore, the findings suggest that susceptibility locus for the serous subtype may be different for women of African ancestry. Future studies are warranted to confirm these findings. Citation Format: Ernest K. Amankwah, Jonathan Tyrer, Hui-Yi Lin, Ya-Yu Tsai, Zhihua Chen, Gang Han, Xiaotao Qu, Ellen Goode, Julie Cunninghan, Edward Iverson, Susan Ramus, Andrew Berchuck, Joellen Schildkraut, Alvaro Monteiro, Simon Gayther, Steven Narod, Paul Pharoah, Thomas A. Sellers, Catherine Phelan. Epithelial-mesenchymal transition (EMT) gene variants influence epithelial ovarian cancer risk in women of European, African and Asian ancestry. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 1352. doi:10.1158/1538-7445.AM2013-1352

Abstract Background: Circadian rhythms of biological processes are regulated by endogenous clock genes and clock-controlled genes. Aberrant expression of circadian clock genes may have important consequences on the transactivation of downstream targets that control the cell cycle and cellular proliferation potentially promoting carcinogenesis. Animal models indicate that several circadian rhythm genes are expressed in the ovaries, where they influence and are influenced by estrous cycles. The goal of the current study was to examine the association of circadian gene variants and epithelial ovarian cancer (EOC) risk. Methods: Thirty-one SNPs from five circadian genes (i.e., ARNTL, CRY2, KLF10, NPAS2, PER3, TIMELESS) were genotyped in 14,736 cases and 23,448 control women of European ancestry from 43 studies in the Ovarian Cancer Association Consortium (OCAC), on a custom Illumina iSelect designed for the Collaborative Oncological Gene-Environment Study (COGS). Both invasive cancers combined and the four main histological subtypes (serous [n=8,372], endometroid [n=2,068], clear cell [n=1,025] and mucinous [n=943] were analyzed, SNP analyses were conducted using unconditional logistic regression under a log-additive model. All analyses were adjusted for study site and population substructure. Results: Eleven SNPs were found to be associated with ovarian cancer risk. The SNPs most associated with serous cancer risk were KLF10 rs2513928 (OR=.95 p=6.1x10−3), rs2513927 (OR=1.05 p=6.1x10−3), rs2511703 (OR=1.05 p=9.8x10−3), rs3191333 (OR=1.05, p=1.5x10−2) and NPAS2 rs13012930 (OR=.95, p=3.5x10−2). Endometroid cancer risk was significantly associated with ARNTL SNPs rs10732458 (OR=1.3, p=1.0x10−2) and rs7117836 (OR=1.2, p=4.6x10−2), while clear cell cancer risk was associated with ARNTL rs1562438 (OR=.88, p=1.5x10−2), rs1026071 (OR=.89, p=1.8x10−2), and rs3816360 (OR=.91, p=4.8x10−2) and KLF10 rs2388232 (OR=1.1, p=3.3x10−2). No variants were significantly associated with mucinous cancer risk. Four SNPs in KLF10 were associated with cancer invasiveness; they were rs2513928 (OR=.95, p=1.8x10−3), rs3191333 (OR=1.04, p=1.4x10−2), rs2513927 (OR=1.04, p=1.9x10−2), and rs2511703 (OR=1.03, p=2.8x10−2). Conclusions: Data from the current study suggest that polymorphisms in circadian genes ARNTL, KLF10, and NPAS2 are significantly associated with ovarian cancer histopathologic subtypes and invasiveness. These findings merit further investigation and replication. Funding: R01 CA149429 Citation Format: Heather Jim, Jonathan Tyrer, Hui-Yi Lin, Gang Han, Xiaotao Qu, Ellen L. Goode, Zhihua Chen, Ya-Yu Tsai, Julie M. Cunningham, Edward Iversen, Susan Ramus, Andrew Berchuck, Joellen Schildkraut, Alvaro Monteiro, Simon Gayther, Steven A. Narod, Thomas A. Sellers, Paul Pharoah, Catherine M. Phelan. Variation in circadian rhythm genes influence epithelial ovarian cancer risk and invasiveness. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4850. doi:10.1158/1538-7445.AM2013-4850 Note: This abstract was not presented at the AACR Annual Meeting 2013 because the presenter was unable to attend.

Abstract Background: ASP is hypothesized to decrease the risk of breast cancer (BRCA) and BRCA recurrences. Thus we performed an exploratory analysis of ASP and CC use within the NCIC CTG MA.27 adjuvant trial comparing exemestane (E;N=3789) to anastrozole (A; N=3787) with a factorial second randomization to CC or placebo (P). Neither A or E was superior in breast cancer outcomes. Baseline low-dose ASP use is an accepted surrogate of cardiovascular risk factors and was used as a stratification factor across all 4 arms. Randomization to CC-P was discontinued after 18 months (n = 1622) due to concerns of cardiac toxicity. Methods: Patients taking >81mg of ASP daily at baseline were ineligible for randomization and use of >81mg of ASP daily was not allowed. Women enrolled during CC randomization were included in the comparison of E and A, stratified by whether they had been randomized to CC [yes, no;N=1622] and concomitant low-dose ASP [≤81 mg/day (yes, no); N=2209]. Other stratification factors included: lymph-nodes (negative, positive, or unknown); prior adjuvant chemotherapy (yes, no). The primary endpoint, event-free-survival (EFS), was defined as time from randomization to time of locoregional or distant disease recurrence, new primary breast cancer, or death from any cause. Secondary endpoints included overall survival (OS) defined as time from randomization to time of death from any cause and distant disease-free-survival (DDFS), defined as time from randomization to time of distant disease recurrence. Univariate (uni) assessment of CC and ASP use was assessed with stratified log-rank test, adjusting for lymph-node status and adjuvant chemotherapy and applied by intention-to-treat. Exploratory multivariate (multi) analyses (N = 1622) had forced inclusion of treatment and used step-wise forward stratified Cox modeling to examine the effects of CC, ASP use and baseline patient characteristics on outcomes; a factor was added with Wald test statistic p ≤ 0.05. Results: At median follow-up of 4.1 years, 186/1622 (11%) patients had an EFS event; 125 (8%) had died from any cause, and 80 (5%) had distant BRCA relapse. CC did not have significant uni association with outcomes: EFS p-value=0.92; OS p-value p = 0.56; DDFS p-value=0.55. ASP use was associated with worse EFS [p = 0.006, HR 1.48 (95% CI 1.12–1.96)], worse OS [p = 0.0002, HR 1.87 (95% CI (1.35–2.61)], and non-significant difference in DDFS (p = 0.72). CC had no multi association with EFS, OS or DDFS. ASP use had no multi association with EFS and DDFS (p > 0.05). However, ASP use had multi prognostic association with worse OS [p = 0.01; HR 1.67 (95% CI 1.13–2.49)]. Conclusions: Users of either CC or low dose ASP had similar DDFS to non-users in MA.27. As expected, low-dose ASP users (with presumptive cardiovascular risk factors) had worse OS than non-users. Inadequate numbers of patients randomized to CC and lack of randomization to aspirin leaves inadequate evidence from MA.27 to determine whether anti-inflammatories influence breast cancer outcomes. Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P2-13-02.

Abstract Background: Epidermal Growth Factor Receptor (EGFR, HER1) is known to heterodimerize with HER2 and may modulate the effectiveness of trastuzumab or lapatinib. Historically, it has been difficult to measure, but recently we reported a new, standardized, quantitative immunofluorescence assay and a novel EGFR antibody which showed that high levels of EGFR were associated with decreased benefit in the concurrent arm of the North Central Cancer Treatment Group (NCCTG/Alliance) N9831 trial (Rimm et al, SABCS 2012). Here we assess its value in the neo-adjuvant setting in the NeoALTTO trial. Methods: NeoALTTO randomized 455 breast cancer patients to neoadjuvant trastuzumab, lapatinib or both and showed benefit in both single anti-HER2 therapy arms, and the combination therapy showing about twice as much benefit as either single drug arm. We used the previously described AQUA method of quantitative immunofluorescence to measure EGFR levels using the D38B1 antibody on between 4 and 140 fields of view per slide on 353 specimens. The averaged EGFR score was standardized to the absolute concentration using cell lines and western blots and the 13 ng/ug total protein established in the N9831 trial was tested to determine predictive value for pathological complete response (pCR) defined as ypT0/is. Results: In NeoALTTO, 19% of the patients had EGFR levels above 13 ng/ug, compared to 16% in NCCTG N9831. The pCR rate was 35.3% in the patients with low EGFR, compared with 29.8% for high EGFR but this difference was not statistically significant. However, when adjusted for treatment and hormone receptor (HR) status, high EGFR is statistically significantly associated with lower response rate(p = 0.038). Continuous analysis measures were not significant. The effect on overall response at 6 weeks was also weakly significant when adjusting for treatment and HR status (p = 0.06). Surprisingly, for HR negative patients in the lapatinib only arm, the pCR rate in the low EGFR group was 47.2% compared with 16.0% in the high group. While this difference is statistically significant (p = 0.014), it should be interpreted cautiously because of the small numbers of patients and the multiple tests performed. Conclusions: High expression of EGFR by the AQUA platform appears to be associated with decreased pCR rate from HER2 targeted therapy in the neoadjuvant setting. This observation is consistent with NCCTG N9831. Although underpowered, high EGFR levels predicted decreased pCR rate from lapatinib, the opposite of that anticipated for this small molecule dual EGFR/HER2 inhibitor. Further studies are required to determine the clinical utility of this assay and the biological mechanisms underlying these observations. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr P1-08-09.

ScopeWe reevaluated previously reported associations between variants in pathways of one‐carbon (1‐C) (folate) transfer genes and ovarian carcinoma (OC) risk, and in related pathways of purine and pyrimidine metabolism, and assessed interactions with folate intake.Methods and resultsOdds ratios (OR) for 446 genetic variants were estimated among 13 410 OC cases and 22 635 controls, and among 2281 cases and 3444 controls with folate information. Following multiple testing correction, the most significant main effect associations were for dihydropyrimidine dehydrogenase (DPYD) variants rs11587873 (OR = 0.92; p = 6 × 10−5) and rs828054 (OR = 1.06; p = 1 × 10−4). Thirteen variants in the pyrimidine metabolism genes, DPYD, DPYS, PPAT, and TYMS, also interacted significantly with folate in a multivariant analysis (corrected p = 9.9 × 10−6) but collectively explained only 0.2% of OC risk. Although no other associations were significant after multiple testing correction, variants in SHMT1 in 1‐C transfer, previously reported with OC, suggested lower risk at higher folate (pinteraction = 0.03–0.006).ConclusionVariation in pyrimidine metabolism genes, particularly DPYD, which was previously reported to be associated with OC, may influence risk; however, stratification by folate intake is unlikely to modify disease risk appreciably in these women. SHMT1 SNP‐by‐folate interactions are plausible but require further validation. Polymorphisms in selected genes in purine metabolism were not associated with OC.

Abstract Background: The Prostate Cancer Prevention Trial (PCPT) was a 7-year randomized, double-blind, placebo-controlled trial of the efficacy of finasteride for the prevention of prostate cancer with a primary outcome of histologically determined prevalence of prostate cancer at the end of 7 years. Methods: A systematic modeling process using logistic regression identified factors available at year 6 that are associated with end-of-study (EOS) biopsy adherence at year 7, stratified by whether participants were ever prompted for a prostate biopsy by year 6. Final models were evaluated for discrimination. At year 6, 13,590 men were available for analysis. Results: Participants were more likely to have the EOS biopsy if they were adherent to study visit schedules and procedures and/or were in good health (P < 0.01). Participants at larger sites and/or sites that received retention and adherence grants were also more likely to have the EOS biopsy (P < 0.05). Conclusions: Our results show good adherence to study requirements 1 year before the EOS biopsy was associated with greater odds that a participant would comply with the invasive EOS requirement. Impact: Monitoring adherence behaviors may identify participants at risk of nonadherence to more demanding study end points. Such information could help frame adherence intervention strategies in future trials. Cancer Epidemiol Biomarkers Prev; 23(8); 1638–48. ©2014 AACR.