Constitutional BRCA1 Promoter Methylation in Patients With Ovarian Cancer: Results of the Observational AGO-TR1 Study Authors: Mohamad Kayali, Eric Hahnen, Alexander Burges, Alexander Reuss, Julia Caro-Valenzuela, Nikolaus de Gregorio, Florian Heitz, Sandra Schmidt, Frederik Marmé, Felix Hilpert, Pauline Wimberger, Stefan Kommoss, Katharina Prieske, Christoph Engel, Paul Buderath, Dimo Dietrich, Rita Schmutzler, Jalid Sehouli, Kerstin Rhiem, Philipp Harter, Jan Hauke Published: 2026 ABSTRACT PURPOSEConstitutional epimutations arise early in development and are present across normal tissues, including peripheral blood. Constitutional BRCA1 promoter methylation has emerged as a risk factor for BRCA1-associated cancers, such as ovarian cancer (OC), and may serve as a biomarker for OC risk. This study retrospectively evaluated the clinical relevance of constitutional BRCA1 promoter methylation in 473 patients with OC enrolled in the observational AGO-TR1 study (ClinicalTrials.gov identifier: NCT02222883).MATERIALS AND METHODSBRCA1 promoter methylation was quantified by the methylation-specific real-time polymerase chain reaction using whole blood-derived DNA from 476 female controls and 473 patients with OC along with 473 corresponding tumor-derived DNA samples. Methylation levels ≥1.0% were considered methylation-positive.RESULTSBRCA1 promoter methylation in blood-derived DNA was detected in 42 of 473 patients with OC and in 26 of 476 controls (8.9% v 5.5%; odds ratio [OR], 1.69 [95% CI, 1.02 to 2.80], P = .0432), with the strongest association observed with methylation levels ≥10% (OR, 6.17 [95% CI, 1.37 to 27.72], P = .018). Patients with BRCA1 promoter methylation in blood-derived DNA were diagnosed at a younger median age than those without (54.0 v 60.0 years, P = .018). Constitutional BRCA1 promoter methylation was less frequent in patients carrying pathogenic germline variants in OC predisposition genes than in noncarriers (4.1% v 10.5%; OR, 0.37 [95% CI, 0.14 to 0.96], P = .04) and showed no association with a family history of cancer or platinum-based chemotherapy before blood draw. BRCA1 promoter methylation in blood-derived DNA was correlated with tumor BRCA1 promoter methylation (P < .001). Tumor BRCA1 promoter methylation was observed in 64 of 473 samples (13.5%), half (32 of 64) of which were attributable to constitutional BRCA1 promoter methylation also detectable in the blood.CONCLUSIONConstitutional BRCA1 promoter methylation accounts for a substantial proportion of OCs and represents a robust biomarker for individual OC risk. INTRODUCTION Women carrying pathogenic germline variants (PGVs) in BRCA1, BRCA2, BRIP1, PALB2, RAD51C, and RAD51D are at an increased risk for ovarian cancer (OC) compared with the general female population.1-3 In addition, OC is part of the tumor spectrum of the Lynch syndrome, and women with PGVs in MLH1, MSH2, and MSH6 are at an elevated OC risk.4 Up to 20% of patients with OC carry heterozygous PGVs in BRCA1/2, and up to 5% carry PGVs in further OC predisposition genes.5,6 A proportion of OCs may additionally be explained by unfavorable OC-specific polygenic risk scores (PRSs).3,7 However, the majority of unselected OCs are not caused by known heritable risk factors such as PGVs in established OC predisposition genes or PRSs. Pathogenic somatic variants (PSVs) in OC predisposition genes and BRCA1 promoter methylation observed in the tumor tissue are usually attributed to sporadic OC, not associated with a positive family history of cancer as shown for PGVs.8 PGVs and PSVs in BRCA1/2 and BRCA1 promoter methylation observed in the tumor tissue are associated with a homologous recombination–deficient tumor phenotype, predicting favorable response to treatment with DNA-damaging agents and DNA repair inhibition, such as platinum or poly(ADP-ribose) polymerase (PARP) inhibitors.9-11 Notably, PGVs and PSVs in BRCA1/2 and BRCA1 promoter methylation observed in the tumor tissue are mutually exclusive, with only a few exceptions described.8 CONTEXT Key Objective Does constitutional BRCA1 promoter methylation detectable in blood-derived DNA contribute to ovarian cancer (OC) risk, and how does it relate to tumor methylation, germline pathogenic variants, and other clinical variables? Knowledge Generated BRCA1 methylation in blood-derived DNA was detected in 8.9% of patients with OC and 5.5% of controls (odds ratio [OR] 1.69), with the strongest association at methylation levels ≥10% (OR, 6.17). Patients with constitutional BRCA1 methylation were diagnosed at a younger median age. Constitutional BRCA1 methylation was less frequent in carriers of pathogenic germline variants (4.1% v 10.5%) and not associated with cancer family history or chemotherapy before blood draw. BRCA1 methylation in blood-derived DNA correlated with tumor BRCA1 methylation. Tumor BRCA1 methylation was observed in 64 of 473 samples (13.5%), half of which were attributable to constitutional BRCA1 methylation. Relevance Assessment of constitutional BRCA1 methylation could support OC risk stratification. Prospective studies are required to quantify age-dependent OC risks. Key Objective Key Objective Does constitutional BRCA1 promoter methylation detectable in blood-derived DNA contribute to ovarian cancer (OC) risk, and how does it relate to tumor methylation, germline pathogenic variants, and other clinical variables? Does constitutional BRCA1 promoter methylation detectable in blood-derived DNA contribute to ovarian cancer (OC) risk, and how does it relate to tumor methylation, germline pathogenic variants, and other clinical variables? Knowledge Generated Knowledge Generated BRCA1 methylation in blood-derived DNA was detected in 8.9% of patients with OC and 5.5% of controls (odds ratio [OR] 1.69), with the strongest association at methylation levels ≥10% (OR, 6.17). Patients with constitutional BRCA1 methylation were diagnosed at a younger median age. Constitutional BRCA1 methylation was less frequent in carriers of pathogenic germline variants (4.1% v 10.5%) and not associated with cancer family history or chemotherapy before blood draw. BRCA1 methylation in blood-derived DNA correlated with tumor BRCA1 methylation. Tumor BRCA1 methylation was observed in 64 of 473 samples (13.5%), half of which were attributable to constitutional BRCA1 methylation. BRCA1 methylation in blood-derived DNA was detected in 8.9% of patients with OC and 5.5% of controls (odds ratio [OR] 1.69), with the strongest association at methylation levels ≥10% (OR, 6.17). Patients with constitutional BRCA1 methylation were diagnosed at a younger median age. Constitutional BRCA1 methylation was less frequent in carriers of pathogenic germline variants (4.1% v 10.5%) and not associated with cancer family history or chemotherapy before blood draw. BRCA1 methylation in blood-derived DNA correlated with tumor BRCA1 methylation. Tumor BRCA1 methylation was observed in 64 of 473 samples (13.5%), half of which were attributable to constitutional BRCA1 methylation. Relevance Relevance Assessment of constitutional BRCA1 methylation could support OC risk stratification. Prospective studies are required to quantify age-dependent OC risks. Assessment of constitutional BRCA1 methylation could support OC risk stratification. Prospective studies are required to quantify age-dependent OC risks. In addition, constitutional BRCA1 promoter methylation has emerged as a risk factor for BRCA1-associated cancers such as OC and breast cancer (BC).12 Constitutional epimutations such as constitutional BRCA1 promoter methylation occur prenatally, affecting all three germ layers, and are thus detectable in normal tissues such as blood. Constitutional BRCA1 promoter methylation has been detected in 4%-10% of female newborns, suggesting a common and early event during life.13,14 However, it remains unclear whether constitutional BRCA1 promoter methylation observed in blood-derived DNA is affected by factors such as individual disease status, age at blood draw, and/or chemotherapy before blood draw, among others. In a nested case-control study, Lønning et al analyzed blood-derived DNA samples of 3,493 women who were cancer-free at blood draw, including 511 women who developed incident high-grade serous OC, a BRCA1-associated histologic tumor phenotype. In this study, BRCA1 promoter methylation in blood-derived DNA was significantly associated with the risk of incident high-grade serous OC (hazard ratio [HR], 1.93 [95% CI, 1.36 to 2.73]; P < .001).12 Thus, constitutional BRCA1 promoter methylation detectable in blood-derived DNA may be a valuable biomarker to identify individuals prone to developing BRCA1-associated tumors, such as OC, and to improve the clinical management of diseased individuals. These data prompted us to retrospectively assess the role of constitutional BRCA1 promoter methylation in 473 patients with OC enrolled in the observational AGO-TR1 study (ClinicalTrials.gov identifier: NCT02222883). MATERIALS AND METHODS AGO-TR1 Study Patients Between March 2015 and July 2015, the AGO-TR1 study recruited 523 consecutive patients with invasive epithelial OC at 20 German study centers of the German Gynecological Oncology Study Group (AGO Study Group); 281 patients were included with a primary OC diagnosis, and 242 patients with platinum-sensitive recurrence of OC. All participants were older than 18 years at study inclusion and provided written informed consent before trial enrollment. The study protocol was approved by the ethics committee of the Landesärztekammer Nordrhein (Nr. 2014340) and registered (ClinicalTrials.gov identifier: NCT02222883). Demographic data, disease characteristics, family history, and medical history were documented as previously described.6 Venous blood samples were available from all 523 patients, and corresponding formalin-fixed paraffin-embedded (FFPE) tumor samples were available from 473 patients.6,8 A detailed description of the 473 study patients with available blood and tumor samples is provided in Table 1. TABLE 1.Patient CharacteristicsVariableAll (n = 473)Age at diagnosis, years, median (range)59 (16-93)Age at blood draw, years, median (range)61 (18-93)Time point of study inclusion, % Primary diagnosis of OC257 (54.3) Platinum-sensitive recurrence of OC216 (45.7)Time point of blood draw, % Before chemotherapy start, enrolled at primary diagnosis110 (23.3) After chemotherapy start, enrolled at primary diagnosis (median 112 days, range 1-983)132 (27.9) After chemotherapy start, enrolled at recurrence (median 1,098 days, range, 321-5,869)212 (44.8) Unknown/missing19 (4.0)Family history of cancer, %a Positive198 (41.9) Negative266 (56.2) Unknown/missing9 (1.9)Histologic subtype of OC, % High-grade serous373 (78.9) High-grade endometrial22 (4.7) Serous/papillary, grade unknown19 (4.0) Low-grade serous16 (3.4) Low-grade endometrial7 (1.5) Clear cell6 (1.3) Mucinous6 (1.3) Others/unspecified18 (3.8) Unknown/missing6 (1.3)Germline mutation status, %b With PGVs in BRCA1/297 (20.5) With PGVs in non-BRCA1/2 OC predisposition genes24 (5.1) With PGVs in any OC predisposition gene121 (25.6) Without PGVs in any OC predisposition gene352 (74.4)Abbreviations: BC, breast cancer; OC, ovarian cancer. a A positive family history was defined as at least one relative with BC or OC or BC in personal history. b BRCA1, BRCA2, BRIP1, PALB2, RAD51C, and RAD51D and the Lynch syndrome genes in MLH1, MSH2, and MSH6 were considered established OC predisposition genes. Patient Characteristics Abbreviations: BC, breast cancer; OC, ovarian cancer. A positive family history was defined as at least one relative with BC or OC or BC in personal history. BRCA1, BRCA2, BRIP1, PALB2, RAD51C, and RAD51D and the Lynch syndrome genes in MLH1, MSH2, and MSH6 were considered established OC predisposition genes. Control Individuals Female control individuals were recruited from the LIFE-Adult-Study, a population-based cohort study focusing on the prevalence, early onset markers, genetic predispositions, and the role of lifestyle factors of major civilization diseases.15 We selected 476 cancer-free female control individuals age between 59 and 68 years at the time of blood draw, with a median age at blood draw of 64 years. DNA Isolation Genomic DNA was isolated from blood samples and FFPE tumor samples using standard methods, as described in detail previously.6,8 Briefly, for the isolation of DNA from FFPE tumor samples, hematoxylin and eosin–stained 3-µm tissue sections were centrally investigated (Institute of Pathology, University Hospital Bonn, Bonn, Germany), and tumor areas containing >80% tumor nuclei were chosen for DNA isolation. Quantitative BRCA1 Promoter Methylation Analysis Quantitative BRCA1 gene promoter methylation analysis was performed as described previously,8 and the genomic BRCA1 gene promoter target region (hg38, Chr17:43125354-43125465) was shown to be affected by DNA methylation.8,16 We used a quantitative methylation-specific polymerase chain reaction (PCR; qMSP) assay to amplify the bisulfite-converted methylated BRCA1 gene promoter (forward primer: gaa​gtt​tta​gcg​agt​tta​cgt​c, reverse primer: cta​aaa​aac​tac​tac​tta​acg​ata​ac, hydrolysis probe: 6-FAM-ttcgcgttttttcgttgttacgga-BBQ-650). A quantitative PCR assay targeting the ACTB gene locus (hg38, Chr7:5532118-5532223) was used to quantify the total amount of bisulfite-converted template DNA (forward primer: ccc​tta​aaa​att​aca​aaa​acc​aca​a, reverse primer: gga​gga​ggt​tta​gta​agt​ttt​ttg, hydrolysis probe: Atto-647N-accaccacccaacacacaataacaaacaca-BHQ-2). The ACTB target region does not contain any CpG sites; therefore, its amplification is not methylation-sensitive. Briefly, 2 µg of DNA was bisulfite-converted using an innuCONVERT Bisulfite Basic Kit (Analytik Jena, Jena, Germany). For qMSP, 25 ng of bisulfite-converted DNA per PCR replicate was used as the template. Three nanograms of bisulfite-converted template DNA prepared from artificially methylated DNA (CpGenome Universal Methylated DNA, Millipore, Billerica, MA), as previously described, was used as the reference standard (calibrator). PCR oligonucleotide concentrations were 0.1 μM (ACTB) and 0.3 μM (BRCA1), respectively. PCR was performed in 20-µl reaction volumes using an AB 7500 Fast Real-Time PCR System (Life Technologies Corporation, Carlsbad, CA) with an initial denaturation/heat activation step at 95°C for 20 minutes, followed by 50 cycles of 2 seconds at 62°C, 45 seconds at 56°C, and 15 seconds at 95°C. The BRCA1 qMSP assay was performed in duplicate using a quantitative ACTB assay. Measurements were performed in triplicate for each DNA sample, and the median ΔCT values were computed. The percent methylation levels relative to the calibrator were calculated using the ΔΔCT method.17 Samples with BRCA1 promoter methylation levels <1.0% were considered unmethylated. Statistical Analyses Odds ratios (ORs) were calculated using the OR calculator (MedCalc Software Ltd, Version 23.2.1).18 ORs, 95% CIs, and P values are given in brackets. Independent, nonparametric variables were compared using the Mann-Whitney U test, with exact P values given in brackets. Dependent nonparametric variables were compared using the Wilcoxon signed-rank test. Correlations were assessed using Spearman's correlation test. The latter three statistical tests were performed using the DATAtab Online Statistics Calculator (DATAtab e.U. Graz, Austria).19 Statistical significance was set at P values <.05. RESULTS BRCA1 promoter methylation in blood-derived DNA was observed in 42 of 473 patients with OC (8.9%), with a median BRCA1 promoter methylation level of 4.0% (range, 1.1%-56.1%) in BRCA1 methylation–positive samples (Fig 1A). Patients with OC and BRCA1 promoter methylation in blood-derived DNA had an earlier age at diagnosis (AAD) than patients without BRCA1 promoter methylation. The median AAD was 54.0 years in patients with BRCA1 promoter methylation versus 60.0 years in patients without BRCA1 promoter methylation (exact P = .018). In addition to 473 patients with OC, we analyzed 476 female control individuals who were cancer-free at the time of blood draw. BRCA1 promoter methylation in blood-derived DNA was observed in 26 of the 476 control individuals (5.5%), with a median BRCA1 promoter methylation level of 3.7% (range, 1.1%-16.6%) in BRCA1 methylation–positive samples (Fig 1A). The median age at blood draw was 64 years (range, 60-67) in control individuals with BRCA1 promoter methylation versus 65 years (range, 59-68) in patients without BRCA1 promoter methylation (exact P = .289). (A) BRCA1 promoter methylation levels in blood-derived DNA of 476 female control individuals, 473 patients with OC, and BRCA1 promoter methylation levels in tumor-derived DNA of 473 corresponding tumor samples; (B) BRCA1 methylation levels in 57 patients with higher BRCA1 promoter methylation levels observed in tumor- versus blood-derived DNA; (C) BRCA1 methylation levels in 17 patients with lower BRCA1 promoter methylation levels observed in tumor- versus blood-derived DNA; median BRCA1 promoter methylation levels are shown as dashed lines. OC, ovarian cancer. The overall prevalence of BRCA1 promoter methylation–positive samples was significantly higher in patients with OC than in control individuals (8.9% v 5.5%, OR, 1.69 [95% CI, 1.02 to 2.80], P = .0432). We explored whether higher BRCA1 promoter methylation levels are similarly associated with OC (Table 2). BRCA1 promoter methylation levels of 5.0% or above were also significantly associated with OC in our study sample (4.0% v 1.5%, OR, 2.80 [95% CI, 1.17 to 6.73], P = .021), with the highest association with OC observed in the subgroup of individuals with BRCA1 methylation levels of 10% or above (OR, 6.17 [95% CI, 1.37 to 27.72], P = .018). TABLE 2.Prevalence of Constitutional BRCA1 Promoter Methylation in Blood-Derived DNA in Patients With OC (n = 473) and in Female Control Individuals (n = 476) According to Minimal BRCA1 Promoter Methylation Cutoff LevelsMinimal BRCA1 Promoter Methylation Cutoff (%)No. of OC Patients With BRCA1 Promoter Methylation in Blood (%)No. of Controls With BRCA1 Promoter Methylation in Blood (%)OR (95% CI, P)1.042/473 (8.9)26/476 (5.5)1.69 (1.02 to 2.80, .043)2.532/473 (6.8)17/476 (3.6)1.96 (1.07 to 3.58, .029)5.019/473 (4.0)7/476 (1.5)2.80 (1.17 to 6.73, .021)10.012/473 (2.5)2/476 (0.4)6.17 (1.37 to 27.72, .018)Abbreviations: OC, ovarian cancer; OR, odds ratio. Prevalence of Constitutional BRCA1 Promoter Methylation in Blood-Derived DNA in Patients With OC (n = 473) and in Female Control Individuals (n = 476) According to Minimal BRCA1 Promoter Methylation Cutoff Levels Abbreviations: OC, ovarian cancer; OR, odds ratio. In tumor-derived DNA, BRCA1 promoter methylation was detected in 64 of 473 patients with OC (13.5%), with a median BRCA1 promoter methylation level of 64.0% (range, 1.2%-316%) in BRCA1 methylation–positive samples (Fig 1A). We observed a correlation between BRCA1 promoter methylation levels in tumor-derived DNA and AAD, as described previously.8 The median AAD in patients with BRCA1 promoter methylation in tumor-derived DNA was 56.0 years (range, 32-78) versus 60.0 years (range, 16-93) in patients without BRCA1 promoter methylation (exact P = .041). A positive correlation between BRCA1 DNA methylation levels in blood- and in tumor-derived DNA samples was observed in the overall study sample (r = 0.57, P < .001, Spearman correlation test). Of the 42 patients with BRCA1 promoter methylation in blood-derived DNA, 76.2% (32 of 42) also showed BRCA1 promoter methylation in tumor-derived DNA. Of the 431 patients without BRCA1 promoter methylation in blood-derived DNA, 7.4% (32 of 431) showed BRCA1 promoter methylation in tumor-derived DNA. Thus, patients with BRCA1 promoter methylation detected in blood-derived DNA had a significantly higher probability of BRCA1 promoter methylation in tumor-derived DNA than patients without BRCA1 promoter methylation in blood-derived DNA (OR, 39.9 [95% CI, 18.0 to 88.5], P < .001). A prevalence of BRCA1/2 PGVs of 20.5% (97 of 473) was detected in the AGO-TR1 study sample of 473 patients with OC, and another 5.1% (24 of 473) carried PGVs in additional OC predisposition genes, as previously described.6,20 Patients with PGVs in any established OC predisposition gene (121 of 473, BRCA1, BRCA2, BRIP1, MSH2, MSH6, PALB2, RAD51C, RAD51D) show a significantly younger median AAD (53.0 years, range, 28-77) than patients without PGVs (61.0 years, range, 16-93; exact P < .001). As expected, a positive PGV status was significantly associated with a positive family history of cancer (71 of 121 PGV carriers v 127 of 352 noncarriers with a positive family history of cancer; OR, 2.52 [95% CI, 1.65 to 3.84], P < .0001) and a high-grade serous tumor phenotype (106 of 121 PGV carriers v 267 of 352 noncarriers with a high-grade serous tumor phenotype; OR, 2.25 [95% CI, 1.24 to 4.08], P = .0074). BRCA1 promoter methylation in blood-derived DNA was significantly less frequent in patients with PGVs in any OC predisposition gene (5 of 121, 4.1%) than in patients without PGVs (37 of 352, 10.5% [OR, 0.37 [95% CI, 0.14 to 0.96], P = .04). In addition, BRCA1 promoter methylation in tumor-derived DNA was significantly less frequent in patients with PGVs in any OC predisposition gene (1 of 121, 0.8%) than in patients without PGVs (63 of 352, 17.9%; OR, 0.04 [95% CI, 0.005 to 0.28], P = .0013). Regarding BRCA1 promoter methylation levels, 74 patients showed BRCA1 promoter methylation in blood- or tumor-derived DNA or both. We observed generally higher methylation levels in tumor-derived DNA (Fig 1A, P < .001, Wilcoxon signed-rank test). Of the 74 patients, 57 showed higher BRCA1 promoter methylation levels in the tumor than in blood-derived DNA (including 32 patients without BRCA1 promoter methylation levels in blood-derived DNA, Fig 1B). The remaining 17 patients showed lower BRCA1 promoter methylation levels in tumor- than in blood-derived DNA (including 10 without BRCA1 promoter methylation detected in tumor-derived DNA, Fig 1C). Notably, no PGV carriers were observed in the group of 57 patients with increased BRCA1 promoter methylation in tumor-derived DNA, whereas 5 PGV carriers were observed in the group of 17 patients with decreased BRCA1 promoter methylation in tumor-derived DNA (Fig 1C, thicker lines). Regarding further clinical variables, we identified no association between BRCA1 promoter methylation and a family history of cancer. Of the 42 patients with BRCA1 promoter methylation in blood-derived DNA, 19 reported a positive family history of cancer (49.8%) versus 41.5% (179 of 431) of the methylation-negative patients (OR, 1.15 [95% CI, 0.62 to 2.20], P = .6423). Of the 64 patients with BRCA1 promoter methylation in tumor-derived DNA, 26 reported a positive family history of cancer (40.6%) versus 42.1% (172 of 409) of the methylation-negative patients (OR, 0.94 [95% CI, 0.55 to 1.61], P = .83). The same holds true when restricting a positive family history definition to patients with first-degree relatives with BC and/or OC only. Of the 42 patients with BRCA1 promoter methylation in blood-derived DNA, five reported a positive family history of cancer (11.9%) versus 13.5% (58 of 431) of the methylation-negative patients (OR, 0.87 [95% CI, 0.33 to 2.30], P = .7776). Of the 64 patients with BRCA1 promoter methylation in tumor-derived DNA, six reported a positive family history of cancer (9.4%) versus 13.9% (57 of 409) of the methylation-negative patients (OR, 0.64 [95% CI, 0.26 to 1.55], P = .32). In addition, no association was observed when restricting positive family history definition to relatives with OC only (blood: 5 of 42, 11.9%, v 54 of 431, 12.5%; OR, 0.94 [95% CI, 0.36 to 2.50], P = .91; tumor: 7 of 64, 10.9%, v 52 of 409, 12.7%; OR, 0.84 [95% CI, 0.37 to 1.95], P = .69). In addition, BRCA1 promoter methylation in blood-derived DNA was not associated with the high-grade serous tumor phenotype (33 of 42, 78.6% v 340 of 431 without promoter methylation, 78.9%; OR, 0.98 [95% CI, 0.45 to 2.12], P = .962). Patients with BRCA1 promoter methylation in tumor-derived DNA tended to have a high-grade serous tumor phenotype, not reaching significance in our study sample (57 of 64, with promoter methylation, 89.1% v 340 of 431 without promoter methylation, 78.9%; OR, 2.18 [95% CI, 0.96 to 4.94], P = .0621). A total of 242 patients were enrolled at primary diagnosis of OC, with the time point of blood draw known (Table 1). In 110 of these patients, blood was drawn before starting platinum-based chemotherapy, and in 132 patients, blood was drawn after chemotherapy start, the latter with a median time difference between the start of chemotherapy and blood draw of 112 days (Table 1, range 1-983 days). Of the first mentioned 110 patients, 10 (9.1%) showed BRCA1 promoter methylation in blood-derived DNA versus 14 of 132 (10.6%) in the latter group of patients (OR, 0.84 [95% CI, 0.36 to 1.98], P = .694). In methylation-positive samples, methylation levels did not differ between both groups (median 5.87%, range, 1.72-24.64 v median 5.94%, range, 1.05-51.13, exact P = .841). Another 212 patients were enrolled at platinum-sensitive recurrence of OC and, by definition, had undergone chemotherapy before blood draw (Table 1, median 1,098 days, range, 321-5,869 days). Of these 212 patients, 17 (8.02%) showed BRCA1 promoter methylation in blood-derived DNA (median methylation levels of 2.93%, range, 1.72-56.1). No significant differences were observed when comparing the number of methylation-positive patients enrolled at platinum-sensitive recurrence of OC (17 of 212) versus patients enrolled at primary diagnosis of OC (24 of 242, 9.92%, OR, 0.79 [95% CI, 0.41 to 1.52], P = .482). Moreover, promoter methylation levels did not differ between patients enrolled at platinum-sensitive recurrence of OC versus patients enrolled at primary diagnosis of OC (exact P = .323). Thus, exposure to platinum-based chemotherapy before blood draw did not significantly affect BRCA1 promoter methylation levels in blood-derived DNA in our analyses. DISCUSSION In this retrospective analysis of an observational AGO-TR1 study, we confirmed a positive association between constitutional BRCA1 promoter methylation and OC (Table 2). A case-control analysis (934 patients, 1,698 controls) by Lønning et al published in 2018 revealed a positive association between constitutional BRCA1 promoter methylation and overall OC (OR, 1.83 [95% CI, 1.27 to 2.63]), with an elevated OR observed for high-grade serous OC (2.91 [95% CI, 1.85 to 4.56]).21 A larger, nested case-control study by Lønning et al12 published in 2022 (3,493 women who were cancer-free at the time of blood draw, including 511 women who developed incident high-grade serous OC) confirmed an association between constitutional BRCA1 promoter methylation and the risk of incident high-grade serous OC (HR, 1.93 [95% CI, 1.36 to 2.73]). In our case-control analysis, we demonstrated that higher BRCA1 promoter methylation levels were associated with higher ORs of up to 6.17 (95% CI, 1.37 to 27.72, P = .018, Table 2), suggesting that methylation levels may be considered for OC risk stratification. However, this finding was less pronounced in the more complex next-generation sequencing–based BRCA1 methylation analysis performed by Lønning et al.12 In line with previous studies, we demonstrated (1) that constitutional BRCA1 methylation and PGVs in OC predisposition genes are mutually exclusive, with a few exceptions only, indicating that PGVs and epimutations are independent risk factors, (2) that constitutional BRCA1 methylation is not associated with the family history of cancer, and (3) that constitutional BRCA1 methylation observed in controls does not correlate with the age at blood draw.12,22 In contrast to previous studies, however, we observed a significantly younger median AAD in patients with constitutional BRCA1 methylation than in patients without constitutional BRCA1 methylation (mean 56.0 v 60.0 years, respectively). This difference becomes more apparent when also considering the PGV status. In patients with PGVs in OC predisposition genes, the median AAD was 53.0 years (range, 28-77 years) versus 61.0 years (range, 16-93 years) in patients without (P < .001). In patients without PGVs and with constitutional BRCA1 promoter methylation, the median AAD was 54.0 years (range, 25-78 years). In patients without both risk factors, the mean AAD was 62.0 years (range, 16-93 years). Thus, we suggest that constitutional BRCA1 promoter methylation is associated with the risk of OC at an earlier age. It is a matter of debate whether constitutional BRCA1 promoter methylation is affected by the individual disease status or chemotherapy before blood draw.22 In our analyses, we demonstrated that exposure to platinum-based chemotherapy before blood draw did not significantly affect BRCA1 promoter methylation levels in blood-derived DNA. BRCA1 promoter methylation in tumor-derived DNA is a frequent finding observed in 9% to 20% of all patients with OC.23-26 In the AGO-TR1 study, we observed BRCA1 promoter methylation in tumor-derived DNA in 13.5% (64 of 473) of all patients with OC, which is in line with previous studies. For OC, however, it is still unclear to what extent constitutional BRCA1 promoter methylation detected in blood-derived DNA correlates with BRCA1 promoter methylation in tumor tissue. We demonstrated that half (32 of 64) of all OCs with BRCA1 promoter methylation in the tumor tissue were due to constitutional BRCA1 promoter methylation, whereas the remaining half (32 of 64) appeared to be tumor-specific. The cause of constitutional BRCA1 promoter methylation remains unclear. In rare cases, constitutional BRCA1 promoter methylation may be triggered by germline variants (secondary constitutional epimutations [SCEs]).22 In two families with BC and OC, Evans et al identified the c.-107A>T variant in the BRCA1 5'UTR to cause monoallelic BRCA1 promoter hypermethylation.27 In our study sample of 473 patients with OC, the c.-107A>T variant was absent, as shown by Sanger sequencing (data not shown). However, we cannot rule out SCEs because of yet unknown BRCA1 germline variants that trigger SCEs affecting the BRCA1 promoter. With a median age at blood draw of 61.0 years in patients with OC and 64.0 years in female control individuals, patients and control individuals were not perfectly age-matched. However, we observed no significant correlation between BRCA1 promoter methylation and age at blood draw in control individuals, consistent with previous findings. The number of OC patients with histologic tumor phenotypes other than high-grade serous OC was low (n = 88, Table 1) and probably too low to identify a negative association between BRCA1 promoter methylation in blood-derived DNA and non–high-grade serous OC subtypes described by others. Several samples showed high BRCA1 promoter methylation levels in tumor-derived DNA of up to 316% (Fig 1A). This may be due to (or a combination of) tumor-specific amplification of the methylated BRCA1 promoter, loss of the reference ACTB allele, or incomplete methylation of the artificially methylated DNA used as a reference standard to calculate methylation levels. We refrained from arbitrarily setting these values to 100%. In conclusion, constitutional BRCA1 promoter methylation detectable in blood-derived DNA appears to be a reliable biomarker for OC risk stratification not significantly affected by age at blood draw, individual disease status, or platinum-based chemotherapy before blood draw. We suggest the further development of elevated BRCA1 promoter methylation as a biomarker for the assessment of individual risks of OC and also other BRCA1-associated tumors, especially triple-negative BC. Prospective studies are required to quantify age-dependent cancer risks associated with constitutional BRCA1 promoter methylation to offer risk-adjusted preventive measures.