Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Nov 1.
Published in final edited form as: Gynecol Oncol. 2024 Aug 10;190:18–27. doi: 10.1016/j.ygyno.2024.07.690

Increased TP53 somatic evolution in peritoneal washes of individuals with BRCA1 germline mutations

Xin Ray Tee 1, Emma Hazard 1, Elena Latorre-Esteves 1, Brendan F Kohrn 1, Talayeh S Ghezelayagh 1,2, Jeanne Uy Fredrickson 1, CoohleenAnn Coombes 1, Marc R Radke 2, Enna Manhardt 2, Ronit Katz 2, T Rinda Soong 3, Elizabeth M Swisher 2, Barbara M Norquist 2,4, Rosa Ana Risques 1,4,*
PMCID: PMC11560739  NIHMSID: NIHMS2016232  PMID: 39128337

Abstract

Background:

Individuals with germline BRCA1 and BRCA2 pathogenic variants (BRCA carriers) are at high risk of developing high grade serous ovarian carcinoma (HGSC). HGSC is predominantly driven by TP53 mutations, but mutations in this gene are also commonly found in non-cancerous tissue as a feature of normal human aging. We hypothesized that HGSC predisposition in BRCA carriers may be related to increased TP53 somatic evolution, which could be detectable by ultra-deep sequencing of TP53 mutations in gynecological liquid biopsies.

Methods:

Duplex sequencing was used to identify TP53 mutations with high sensitivity in peritoneal washes and cervical liquid-based cytology (LBC) collected at surgery from 60 individuals including BRCA1 and BRCA2 carriers, and non-carriers. TP53 mutation pathogenicity was compared across groups and with TP53 cancer mutations.

Results:

TP53 mutations were more abundant in cervical LBC than in peritoneal washes but increased with age in both sample types. In peritoneal washes, but not in cervical LBC, pathogenic TP53 mutation burden was increased in BRCA1 carriers compared to non-carriers, independently of age. Five individuals shared identical pathogenic TP53 mutations in peritoneal washes and cervical LBC, but not in blood.

Conclusions:

Ultra-deep sequencing of TP53 mutations in peritoneal washes collected at surgery reveals increased burden of pathogenic TP53 mutations in BRCA1 carriers. This excess of pathogenic TP53 mutations might be linked to the elevated risk of HGSC in these individuals. In some patients, concordant TP53 mutations were found in peritoneal washes and cervical LBCs, but the cell of origin remains unknown and deserves further investigation.

Keywords: Pap tests, BRCA mutations, ovarian cancer risk, TP53 mutations, clonal expansions, pathogenic mutations

INTRODUCTION

High grade serous carcinoma (HGSC), which encompasses carcinomas of the ovary, fallopian tube, and peritoneum, is the most common and deadliest subtype of ovarian cancer, accounting for 70-80% of all ovarian cancer deaths (1). Individuals with pathogenic germline mutations in BRCA1 and BRCA2 (BRCA carriers) are at disproportionately higher risk of developing HGSC than the general population, with a cumulative risk at age 80 of 44% for BRCA1 carriers and 17% for BRCA2 carriers (2) compared to ~1.4% risk for the general population (3). BRCA1 carriers develop HGSC around 10 years earlier than either BRCA2 carriers or those with sporadic HGSC (4). The current clinical preventative recommendation is to undergo risk-reducing salpingo-oophorectomy (RRSO) at age 35-40 for BRCA1 carriers and at age 40-45 years for BRCA2 carriers. RRSO is associated with up to 80% reduction in risk of ovarian cancer in BRCA carriers (5), but it leads to premature menopause and loss of reproductive ability. For these reasons, many individuals opt to postpone surgery, which further increases their cancer risk (6). Therefore, there is an urgent need to understand the biological basis of the increased risk of HGSC in BRCA carriers to enable more effective approaches for cancer prediction and prevention.

Most HGSCs are driven by TP53 mutations and are thought to originate in the distal fallopian tube from precursor lesions harboring TP53 mutations (called p53 signatures or foci) (7). These lesions may take decades to develop into serous tubal intraepithelial carcinomas (STICs), which subsequently may progress to HGSC in 6-7 years (8-10). It has been postulated that TP53 mutant cells from STICs or even early precursor lesions may disseminate from the fallopian tubes into the peritoneal cavity, before seeding onto the ovaries and peritoneum to further develop into HGSC years later (11). Supporting this hypothesis, a recent study demonstrated temporal concordance between TP53 mutations in STICs identified at risk reducing surgery of BRCA carriers and HGSC developed at least 2 years later (12). Thus, investigation of potentially exfoliated TP53 mutant cells into the gynecological tract and peritoneal cavity is key to understanding disease pathogenesis and might enable better approaches for early cancer detection and interception.

In recent years, our group leveraged the high sensitivity of duplex sequencing, an ultra-accurate sequencing method (13,14) to detect rare mutant cells in gynecological liquid biopsies including peritoneal wash (15), uterine lavage (16,17), and cervical liquid based cytology (LBC), most commonly known as Pap test (18,19). These studies revealed multiple TP53 mutations (distinct from the tumor mutation), which increased with age and were largely pathogenic. These mutations occurred in individuals with and without HGSC, but the frequency and pathogenicity tended to be higher in those with HGSC (15,17-19), suggesting an association between TP53 clonal expansions and the development of HGSC. Collectively, these results are consistent with a growing body of literature demonstrating cancer mutations in non-cancerous tissue, which is now understood as a manifestation of somatic evolution (20-24). TP53 mutations appear to be ubiquitous in clonal expansions of multiple non-cancerous human tissues, including blood, skin, esophagus, and endometrium (20,24,25).

While TP53 mutations increase with age, it is unknown whether the burden of mutation is higher in individuals at high-risk for TP53-driven cancer. In this exploratory study, we aim to answer that question in the context of BRCA1 and BRCA2 germline mutation carriers without HGSC. We hypothesize that the risk of developing HGSC is related to the burden of pathological TP53 mutations in the gynecological tract and thus, BRCA1 carriers will have the highest burden, followed by BRCA2 carriers and individuals without BRCA germline mutations. Based on our success in prior studies using high fidelity sequencing to measure very low frequency TP53 mutations in gynecological biopsies (15-19), we used duplex sequencing to quantify TP53 mutation burden in peritoneal washes and cervical LBCs collected from BRCA carriers and non-carriers without HGSC. While LBCs were expected to be less sensitive than peritoneal washes due to the location of sampling, the analysis of two different sample types enabled a more comprehensive characterization of TP53 mutations as well as the determination of shared clones, which is critical to our understanding of the transit of TP53 mutant clones in the gynecological tract.

METHODS

Patients and Samples

The study consisted of 60 individuals ages 35 to 63 who underwent gynecological surgery with benign findings at the University of Washington, including 21 BRCA1 carriers, 22 BRCA2 carriers, and 17 patients without germline BRCA1 or BRCA2 mutations (BRCAwt) (Figure 1A, Supplementary Table S1). Patients were enrolled prior to surgery under an IRB-approved protocol at the University of Washington for tissue collection including cervical LBC, peritoneal wash and blood. BRCA carriers underwent risk reducing surgery. BRCA non-carriers underwent surgery because of pelvic masses (7 patients) or risk reducing surgery due to family history of cancer (10 patients). BRCA germline mutation status was determined with commercial tests or the in-house BROCA sequencing panel (26). 6 patients with benign masses did not have BRCA testing but were expected to be negative due to the low prevalence of BRCA germline mutations in the general population. LBC DNA from a subset of patients had been previously analyzed for TP53 mutations as part of two prior studies dedicated to exploring differences between patients with and without HGSC (4 patients from ref. (18) and 23 patients from ref. (19)). Serum levels of CA-125, a well-established ovarian cancer biomarker, were measured prior to surgery. Relevant clinical information including smoking history and BMI was retrieved from the medical records. LBCs were collected preoperatively with an endocervical cytobrush (Thinprep, Hologic, MA, USA) according to manufacturer’s protocol. Peritoneal wash (saline wash of the peritoneum) was collected during surgery. For all patients, peritoneal wash samples were analyzed by expert pathologists and determined to be negative for cancer cells. LBCs and peritoneal washes were centrifuged at 2700 rpm for 15 minutes and cell pellets were stored in 1.7mL of media at −80°C at the University of Washington Gynecologic Oncology Tissue Bank. Genomic DNA was extracted from LBC and peritoneal wash cell pellets using Qiagen kits. Blood DNA was extracted from buffy coats with a salting out method (Supplementary Methods).

Figure 1. TP53 mutations are identified in cervical LBC and peritoneal washes from patients with and without BRCA germline mutations and are associated with age.

Figure 1.

Open in a new tab

A. Experimental design. This study included 60 patients for whom cervical LBC and peritoneal washes were collected at gynecological surgery (illustrations reproduced and edited with permission © 2022 Terese Winslow LLC). B. TP53 ultra-deep sequencing revealed multiple TP53 coding and non-coding mutations across all groups and in both sample types. C. Coding and non-coding mutation frequency plots for cervical LBCs and peritoneal washes. For each sample, coding and non-coding mutation frequencies were calculated as the number of unique mutations divided by the total number of duplex nucleotides sequenced in coding and non-coding regions, respectively. Overlying box plots display the quartiles with whiskers extending up to 1.5x the interquartile range. p-values correspond to Mann-Whitney U tests. D. Overall and group specific age associations for TP53 coding mutation frequency. p-values were calculated with Spearman’s rank correlation test.

Sectioning and extensively examining the fimbriated end (SEE-FIM) was performed for all individuals undergoing risk reducing surgery to determine the presence of ovarian cancer precursor lesions. Inclusion criteria in this study cohort included no invasive carcinoma, no prior tubal ligation, age between 35 and 65 years, identification of a pathogenic variant for individuals with germline BRCA1 and BRCA2 mutations, and availability of peritoneal wash and cervical LBC with sufficient DNA yield for analysis (200ng for LBC, 400ng for peritoneal washes). Exclusion criteria included prior cancer history other than breast cancer, endometriosis (including endometriotic cysts), and pelvic inflammatory disease. Clinico-pathological information is shown in Supplementary Table S1.

Duplex Sequencing Library Preparation

Duplex sequencing libraries were prepared using commercially available kits (TwinStrand Biosciences, Seattle, WA) with 200ng of DNA for LBC and 400ng of DNA for peritoneal washes. The coding region of TP53 was captured with 120bp biotinylated probes (TP53 human panel v1.0 from TwinStrand Biosciences, Seattle, WA) using two rounds of hybridization. For 8 individuals with overlapping mutations between LBC and peritoneal wash, blood DNA (200ng) was sequenced using the same method. Libraries were sequenced using 2 x 150bp paired-end kits on Illumina sequencers (Illumina, San Diego, CA) (Supplementary Methods).

Mutational Analysis

Sequencing reads were analyzed as previously described (27) using Duplex Seq Pipeline v2.1.4 from https://github.com/Kennedy-Lab-UW/Duplex-Seq-Pipeline. For each sample, mean duplex depth was calculated as the average duplex depth in all coding positions sequenced. One peritoneal wash sample with mean depth <5000x was excluded from the analysis. VCF files were converted to MAF files, which were post processed with R version 4.3.0 (28). Pathogenicity was determined with two different algorithms: AlphaMissense and Seshat (Supplementary Methods). Mutations were compared with an in silico “no selection” model including all possible substitutions in TP53 (n=3420) and with the Catalog Of Somatic Mutations in Cancer (COSMIC) genome wide screen TP53 mutational data. COSMIC data was also used to define hotspot codons as codons with at least 1% of reported substitutions in cancer (Supplementary Table S2, Supplementary Methods).

For each sample, Mutation Frequency (MF) was calculated independently for coding and non-coding regions as the number of unique TP53 mutations divided by the total number of duplex nucleotides sequenced in the corresponding regions. TP53 Pathogenic mutation burden was calculated as the number of DNA molecules carrying TP53 pathogenic mutations (each duplex read corresponds to a different DNA molecule) divided by the total number of duplex coding nucleotides sequenced. TP53 Pathogenic mutation burden was independently calculated using the AlphaMissense and Seshat algorithms (Supplementary Methods). TP53 Common-in-Cancer mutation burden was calculated considering only mutations frequent in cancer according to Seshat.

Statistical Analysis

Pair-wise comparisons of mutation burden across groups were performed by Mann-Whitney U test, and correlations were tested with Spearman’s rank test. Associations between categorical variables were tested with Fisher Exact Tests and Pearson Chi-Square. Linear regression analysis was employed to assess the association of mutation burden as outcome with BRCA genotype as the main predictor, while adjusting for potential confounders. Confounders included age, prior breast cancer, prior chemotherapy, benign neoplasia, surgery type, BMI, smoking, and CA-125. CA-125 was included as a covariate in separate models because 3 patients had missing data. All tests were two-sided at α level (type 1 error rate) of 0.05. Statistical analyses were performed with R version 4.3.0 (28) and SPSS (IBM Corp. Released 2021. Version 28.0. Armonk, NY).

Data Access

Sequencing data from this study have been submitted to the NCBI BioProject database (https://www.ncbi.nlm.nih.gov/bioproject) under BioProject number PRJNA1144429.

RESULTS

The study cohort comprised 60 patients ages 35 to 63 including 21 patients with pathogenic BRCA1 germline mutations, 22 patients with pathogenic BRCA2 germline mutations, and 17 patients without BRCA germline mutations or unknown BRCA status but not meeting clinical family or personal cancer history indications for genetic testing. All patients underwent gynecological surgery in which LBC and peritoneal washes were collected (Figure 1A) and absence of ovarian cancer was confirmed. Peritoneal wash mean sequencing depth (7828x) was about double that of LBC (3827x) because peritoneal wash libraries were built with 400ng of DNA vs 200ng for LBC. This was done to increase sensitivity of mutation detection in peritoneal washes because prior studies indicated that TP53 mutations may be less abundant in peritoneal washes than LBC samples (15,18). Despite less depth, LBCs harbored more TP53 mutations than peritoneal washes (Figure 1B). Duplex depth across all groups (Figure 1B) and between samples within each group (Supplementary Fig. S1) were comparable. In total, we identified 795 TP53 variants: 481 coding and 314 non-coding. All coding variants were present at very low frequency (all VAF< 0.02, 99.8% VAF<0.004). Coding variants for LBCs and peritoneal washes are listed in Supplementary Tables S3 and S4, respectively.

TP53 coding mutations are more abundant than non-coding mutations and increase with age

To correct for differences in sequencing depth and enable quantitative comparisons, coding and non-coding TP53 MF were calculated for each LBC and peritoneal wash sample (Supplementary Tables S5 and S6). Coding TP53 MF was significantly higher than non-coding TP53 MF in LBCs and peritoneal washes across all groups suggesting positive selection of protein impacting mutations (Figure 1C). In addition, TP53 coding MF in LBCs and peritoneal washes significantly increased with age (Figure 1D) but TP53 non-coding MF did not (Supplementary Figure 2).

Pathogenic TP53 mutations are most frequent in peritoneal wash from BRCA1 carriers, and LBCs from all groups

We next investigated the distribution of pathogenic mutations across all individuals and between groups. Mutations were classified as pathogenic, ambiguous, or benign based on their type and AlphaMissense algorithm (29) (Supplementary Methods). Most individuals carried at least one pathogenic mutation in LBC and/or peritoneal wash sample (Figure 2A). In LBCs, the enrichment for pathogenic mutations was significantly higher than the no selection value (expected to be 36%) for BRCA1 carriers, BRCA2 carriers, and non-carriers (Fisher exact test, p<0.0001) but the differences between groups were not significant (Figure 2B). In peritoneal washes, again all groups were enriched for pathogenic mutations compared to the no selection value (Fisher exact test, p<0.01), but BRCA1 carriers had a significantly higher percentage compared to the other two groups (Pearson Chi-square p=0.04, Figure 2B). Eight individuals had a total of 10 mutations shared between LBCs and peritoneal washes, most of which (9/10 mutations) were pathogenic (Figure 2A). 7 shared mutations originated from five BRCA1 carriers (24% of BRCA1 carriers), 2 shared mutations originated from two BRCA2 carriers (9% of BRCA2 carriers), and 1 shared mutation originated from one non-carrier (6% of non-carriers). While shared mutations were more frequent in BRCA1 carriers, these differences were not significant.

Figure 2. Distribution of pathogenic TP53 mutations in cervical LBC and peritoneal wash samples of individuals with and without germline BRCA mutations.

Figure 2.

Open in a new tab

A. TP53 mutations by patient, represented by columns. Patients are grouped by germline mutation and sorted by ascending age. Relevant clinical variables are shown. The number of unique coding TP53 mutations for cervical LBC and peritoneal wash samples are indicated on the Y-axis and the number of pathogenic mutations based on AlphaMissense is indicated on the X-axis. Mutations are represented by squares and color coded according to their pathogenicity group based on AlphaMissense. The percentage of pathogenic mutations for each group is indicated. Letter S indicates mutations shared between cervical LBC and peritoneal wash samples from a given patient. Letter X denotes a patient without peritoneal wash data due to insufficient sequencing depth. B. Comparison of the percentage of TP53 pathogenic mutations across groups and with the no selection model. The no selection value corresponds to pathogenic mutations identified for all possible nucleotide substitutions in TP53. Numbers in brackets indicate the number of mutations in each group. Comparisons across groups were performed with Pearson Chi-Square (blue line) and group comparisons with no selection were performed with Fisher Exact Tests (black lines). * p<0.05; ** p < 0.01; *** p < 0.001; ns: not significant.

BRCA1 genotype, age, and prior chemotherapy are associated with higher TP53 pathogenic mutation burden in peritoneal washes

To further investigate the association between TP53 pathogenic mutations and BRCA1 genotype, for each sample we calculated TP53 pathogenic mutation burden, which quantifies the number of DNA molecules with pathogenic TP53 mutations (based on AlphaMissense) corrected by the total amount of nucleotides sequenced (Methods, Supplementary Tables S5 and S6). Multivariate linear regression models demonstrated that TP53 pathogenic mutation burden in peritoneal wash but not in LBC was associated with BRCA1 genotype, prior chemotherapy, and higher BMI, independently of age (Table 1, Figure 3A). Prior breast cancer, benign neoplasia, surgery type, and smoking were not predictors of TP53 pathogenic mutation burden in LBC or peritoneal washes in this cohort. Serum CA-125 levels were not associated with TP53 pathogenic mutation burden in LBC or peritoneal washes in univariate analyses (Supplementary Figure S3) or multivariate analyses (Supplementary Table S7). In addition, the four significant predictors of TP53 mutation burden (BRCA1 genotype, age, prior chemotherapy, and BMI) remained significant after further adjustment by CA-125 (Supplementary Table S7).

Table 1.

Linear regression of TP53 Pathogenic Mutation Burden in cervical LBC and peritoneal washes by BRCA genotype and clinical covariates

Variables TP53 Pathogenic
Mutation Burden
(cervical LBC)		 TP53 Pathogenic
Mutation Burden
(Peritoneal Washes)
N β (95% CI) p-
value		 N β (95% CI) p-
value
BRCA genotype
BRCA1 21 −4.13E-7 (−2.70E-6, 1.84E-6) 0.705 21 4.50E-7 (1.97E-7, 7.03E-7) 0.001
BRCA2 22 −2.96E-7 (−2.47E-6, 1.93E-6) 0.807 21 1.54E-7 (−9.27E-8, 4.00E-7) 0.216
BRCAwt 17 0 (ref) 17 0 (ref)
Other covariates
Age (per 10yr increase) 60 1.20E-6 (2.08E-7, 2.20E-6) 0.019 59 1.81E-7 (7.04E-8, 2.92E-7) 0.002
Prior breast cancer (yes) 12 −3.09E-8 (−4.17E-6, 3.55E-6) 0.873 12 −2.75E-7 (−7.05E-7, 1.55E-7) 0.204
Prior chemotherapy (yes) 10 6.55E-8 (−4.01E-6, 4.14E-6) 0.974 10 6.12E-7 (1.58E-7, 1.07E-6) 0.009
Benign neoplasia (yes) 9 −3.92E-7 (−2.69E-6, 1.91E-6) 0.733 8 −2.22E-7 (−5.05E-7, 6.06E-8) 0.121
Surgery type
risk reducing 53 0 (ref) 52 0 (ref)
benign pelvic masses 7 −6.15E-7 (−3.83E-6, 2.60E-6) 0.703 7 3.10E-7 (−5.66E-8, 6.76E-7) 0.096
Ever smoked 14 4.74E-7 (−1.23E-6, 2.18E-6) 0.579 14 1.38E-7 (−5.24E-8, 3.29E-7) 0.151
BMI (per unit increase) 60 −4.55E-8 (−1.47E-7, 5.64E-8) 0.374 59 1.79E-8 (6.52E-9, 2.93E-8) 0.003
Open in a new tab

Abbreviations: LBC: liquid-based cytology. BMI: body mass index. CI: confidence interval.

Figure 3. BRCA1 carriers harbor higher burden of TP53 pathogenic mutations in peritoneal washes.

Figure 3.

Open in a new tab

A. Linear models of predictors of TP53 pathogenic mutation burden in cervical LBC and peritoneal wash. All the variables included in the model are listed on the Y-axis. Age is quantified in decades. BMI is measured as kg/m2. Effect sizes are indicated with solid circles. Bars represent 95% CIs. * p<0.05; ** p < 0.01; *** p < 0.001. B-D. Associations between BRCA genotype for patients older than age 45 and TP53 pathogenic mutation burden measured in peritoneal washes. Measurements include pathogenicity based on the AlphaMissense (AM) algorithm (Cheng et al. 2023, B), pathogenicity based on the Seshat algorithm (Tikkanen et al. 2018, C), and TP53 mutations common in cancer based on the Seshat algorithm (Tikkanen et al. 2018, D). Overlying box plots display the quartiles with whiskers extending up to 1.5x the interquartile range. p-values correspond to Mann-Whitney U tests for pair-wise comparisons.

Although the age range across all groups was similar, BRCA carriers included younger individuals. Given the strong effect of age on TP53 mutations we restricted the analyses to individuals older than age 45 to enable proper comparisons across groups. This is a meaningful threshold in the pathogenesis of the disease because around age 45 is when the risk for ovarian cancer starts to increase rapidly in BRCA1 carriers (2) and BRCA1 carriers who undergo RRSO after age 45 have a marked increase likelihood of identification of occult neoplasia (30). For individuals older than 45 years of age, TP53 pathogenic mutation burden in peritoneal wash was significantly higher in BRCA1 carriers than in BRCA2 carriers or non-carriers (Figure 3B). To validate these results, we performed an independent assessment of predicted pathogenicity based on the Seshat algorithm, which we had used for the characterization of TP53 mutations in prior studies (16,18). Pathogenic mutations according to Seshat were also significantly higher in BRCA1 carriers than in BRCA2 carriers (Figure 3C). In addition, Seshat provides information about the observed frequency of each TP53 mutation in cancer. TP53 mutations common in cancer were also significantly more frequent in the peritoneal wash of BRCA1 carriers compared to BRCA2 carriers and non-carriers (Figure 3D). For LBC samples, none of these variables were significantly different across BRCA genotypes (Supplementary Fig. S4). Overall, these results indicate that BRCA1 carriers have increased TP53 somatic evolution in peritoneal wash compared to BRCA2 carriers and non-carriers as reflected by a higher burden of pathogenic TP53 mutations. The association between BRCA1 genotype and TP53 pathogenic mutations is independent of prior chemotherapy treatment, which also significantly increases TP53 pathogenic mutation burden in peritoneal washes.

TP53 mutations in LBCs and peritoneal washes resemble cancer mutations

To gain further insight into the profiles of LBC and peritoneal wash TP53 mutations, we plotted the location of all the identified substitutions along the coding region of the gene and compared this with the distribution of TP53 mutations in COSMIC. As expected, cancer mutations clustered in hotspot codons in the DNA binding domain coding region (exons 5 to 8). LBC and peritoneal wash mutations also tended to cluster in the same hotspots (Figure 4A). In both sample types, for each group, we identified a sizable proportion of indels, splice and/or nonsense mutations, approaching the observed distribution of TP53 mutation types in cancer (Supplementary Figure S5A). The mutational spectrum was dominated by C>T transitions, which are common age-related mutations very frequent in human cancers (31) (Supplementary Figure S5B). For all groups, more than 19% of the TP53 mutations identified in LBC and peritoneal washes were located in hotspot codons, which is significantly higher than the 6.6% expected randomly (Fisher exact test p<0.001 for all groups) (Supplementary Figure S5C).

Figure 4. Comparison of TP53 mutations in cervical LBC and peritoneal washes with mutations found in cancer.

Figure 4.

Open in a new tab

A. TP53 mutation distribution maps for cervical LBC and peritoneal washes compared to TP53 mutations identified in human cancer (COSMIC). Upper panels show the codon location of TP53 substitutions identified in cervical LBC (left) and peritoneal washes (right), with the mutation count on the Y-axis. Bottom panels show the codon location of TP53 substitutions reported in COSMIC (n=9296). Indels are excluded from plots as they may span multiple codons. TP53 gene domains are color-coded. B. Cumulative pathogenicity score plots for TP53 mutations identified in cervical LBC and peritoneal wash of BRCA1 carriers, BRCA2 carriers, and BRCAwt individuals. Each dot represents a mutation, which are sorted in ascending order of pathogenicity score based on the AlphaMissense algorithm (AM) and normalized for each group to proportionally distribute all mutations between 0 and 1. Pathogenicity score plots for all TP53 mutations identified in COSMIC (n=11256) and all possible substitutions in TP53 (no selection model, n=3420) are shown for comparison. C. Cumulative pathogenicity score plots for TP53 mutations identified in peritoneal washes of BRCA1 carriers, BRCA2 carriers, and BRCAwt individuals younger and older than 45 years of age. Plots built as in B.

Next, we tested whether AlphaMissense scores could reveal differences in pathogenicity between genotypes. For all genotypes, we plotted the pathogenicity scores of all identified mutations in ascending order after normalizing by the number of mutations in each group, and we compared to TP53 mutations in cancer (COSMIC) and those predicted in the no selection model (Figure 4B). As expected, we observed that the pathogenicity scores of TP53 cancer mutations increased rapidly, with most mutations harboring high values, whereas the scores in the no-selection model increased gradually, with a smaller fraction of mutations harboring high values. Interestingly, for LBCs, TP53 mutations in all groups had overlapping pathogenicity score plots located halfway between the cancer and the no-selection model. However, for peritoneal wash, the plot for BRCA1 carriers was shifted to the left, indicating that mutations were more similar in pathogenicity to those observed in cancer. The higher pathogenicity of TP53 mutations in peritoneal wash of BRCA1 carriers was observed even when separating by ages younger and older than 45 (Figure 4C). TP53 mutations in younger patients were less pathogenic than in older patients but BRCA1 carriers had more pathogenic mutations in both age groups. These results demonstrate the effect of BRCA1 genotype in addition to aging on the overall pathogenicity of TP53 peritoneal wash mutations.

Most mutations in peritoneal washes and LBCs, including those shared between both sample types, are not of blood origin

As shown in Figure 2A, 8 individuals shared 10 mutations between LBC and peritoneal wash samples (two individuals carried two such mutations). In the 8 individuals with shared mutations, duplex sequencing of blood (Supplementary Table S8) revealed that 6 of those 10 mutations were not present in blood, all of which were pathogenic (Figure 5A). Thus, 5 individuals from this study (3 BRCA1 carriers, 1 BRCA2 carrier, and 1 non-carrier) had shared mutations between LBC and peritoneal wash that were not identifiable in hematopoietic clones (P14: p.I251F, p.R273H; P15: p.V122L; P21: p.R273L; P31: p.Y220C; P47: p.R181C). Remarkably, while these shared mutations were detected with only one mutant read in LBCs, for 3 BRCA carriers, shared mutations were found in multiple duplex reads in peritoneal washes, indicating larger clones (P15: p.V122L; P21: p.R273L; P31: p.Y220C). These identical mutations in washes and LBC could represent the same original mutant clone or, alternatively, could have arisen independently in multiple clones. Overall, individuals shared few mutations across sample types (Figure 5B, Supplementary Figure S6). Most mutations in LBCs, peritoneal washes, and blood were unique (80%, 69%, and 88%, respectively) with only 4 mutations in blood being observed in LBCs and peritoneal washes.

Figure 5. Shared TP53 mutations in cervical LBC, peritoneal wash, and blood.

Figure 5.

Open in a new tab

A. TP53 mutations found in cervical LBC and peritoneal wash from the same individuals are listed, color-coded by BRCA genotype group. For each mutation, the number of mutant duplex reads is presented on the x-axis and labeled within a bar. For 8 individuals with shared mutations in cervical LBC and peritoneal wash, blood DNA was also sequenced and mutations corresponding to those found in cervical LBC and peritoneal washes are shown. Empty rows indicate that the specific mutation was not found in the blood of that individual. Patient IDs are listed on the left. Pathogenic mutations based on AlphaMissense are bolded. B. The Venn diagram indicates the number of TP53 mutations shared in cervical LBC, peritoneal wash, and blood for all patients listed in A. Mutations are graphically represented by dots color coded according to their presence in only one sample (grey) or in multiple samples (other colors). Larger dots represent clones in which the mutation was seen in more than one duplex read. For each sample type,” n” indicates the total number of TP53 mutations identified in the 8 patients.

DISCUSSION

By using ultra-deep TP53 sequencing, we have demonstrated an excess of pathogenic TP53 mutations in peritoneal wash of BRCA1 carriers. Age-related TP53 mutations were prevalent in cervical LBCs and peritoneal washes. However, the burden of TP53 pathogenic mutations was higher in BRCA1 carriers, independently of age, prior chemotherapy, or high BMI. While our study cannot prove causality, the association between pathogenic somatic TP53 mutations and BRCA1 germline mutation provides a plausible link to understand the susceptibility to HGSC in this population at risk.

We and others have demonstrated that mutations in cancer genes, including TP53, are common in individuals without cancer as part of a normal process of somatic evolution with aging (16,20,24). However, little is known about how these mutations might impact cancer risk (32). Results from our prior studies have indicated that patients with HGSC carry a higher TP53 mutational burden in peritoneal wash (15), cervical LBC (18), and uterine lavage (17). Although the mutant cancer clone often contributed to the burden, those samples also contained TP53 mutations not found in concurrent HGSC, indicating the presence of multiple somatic pathogenic clones in these patients. We now demonstrate that, while TP53 pathogenic clones are common in the peritoneal wash of individuals without HGSC, they are more abundant and more pathogenic in BRCA1 mutation carriers. These results suggest that BRCA1 carriers might harbor accelerated TP53 somatic evolution leading to a higher frequency of TP53 mutant clones present in the peritoneal cavity, which might be related to their increased risk of HGSC.

While BRCA2 carriers also have an increased risk of HGSC, the risk is lower (17% in BRCA2 vs. 44% in BRCA1) and not evident until ages 50 or later (2), which might explain the lack of significance in our study. Alternatively, HGSC risk in BRCA2 carriers might not be related to increased TP53 somatic evolution. Of note, we detected an increase of TP53 pathogenic mutations in peritoneal wash due to chemotherapy, which is in agreement with prior findings of chemotherapy related clonal expansions in leukocytes, including expansions driven by TP53 (33,34). We also detected an increased in TP53 pathogenic mutations in peritoneal wash due to high BMI, which has been demonstrated to be a risk factor for ovarian cancer in BRCA carriers (35,36). Importantly, the effects of chemotherapy and high BMI were independent of the effect of age or BRCA1 genotype and could be readily detected in multivariate models even with our limited sample size. These results highlight the compounding effect of genetic risk, exposures, and age in enhancing expansions of pathogenic clones and suggest that the measurement of TP53 pathogenic mutations using a combination of ultra-sensitive sequencing and accurate pathogenicity prediction might harbor value to estimate cancer risk.

An important unanswered question from this study is the origin of mutant clones in both sample types. While our method cannot determine cell of origin, the comparison of mutations between cervical LBCs and peritoneal washes revealed that most mutations were unique (not shared). These mutations might represent local clones arising from endometrium in the case of LBCs or from the peritoneal lining in the case of peritoneal washes. The normal endometrium is known to carry extensive clonal expansions in cancer genes, including TP53 (37-39), which could explain the higher TP53 MF in LBCs than in peritoneal washes. From the 10 concordant mutations between cervical LBCs and peritoneal washes from the same individual, 4 were also concordant with blood, likely corresponding to very low frequency TP53 mutant hematopoietic clones (40,41). The origin of the remaining six concordant mutations is unknown. One possibility is convergent evolution, given the fact that some of these mutations are present in common hotspot codons, and clones from different origins might be similarly selected. Another possibility is that these mutations correspond to exfoliated TP53 mutant clones, which, according to the ‘precursor escape’ hypothesis (11), could migrate from fallopian tube epithelium into the peritoneal cavity and later seed onto the ovaries and peritoneum to develop into HGSC. p53 foci are common in fallopian tubes of BRCA carriers, even in individuals without HGSC (42-44), and it is estimated that they take decades to develop into HGSC (9). In addition, a recent study demonstrated that STICs identified at risk reducing surgery of BRCA carriers had the same mutation than HGSC that developed several years after the surgery indicating very early dissemination of TP53 mutant cells years prior cancer development (12). Whether TP53 mutant cells may disseminate from earlier tubal precursor lesions or clones from other tissues is unknown and warrants further investigation.

One limitation of our study is that the BRCA non-carrier group included patients who underwent prophylactic surgery because of familial risk as well as patients who underwent gynecological surgery for pelvic masses but were cancer free. We demonstrated, however, that surgery type had no effect on TP53 mutation frequency in multivariate models. Another limitation is that the analysis of blood samples was only performed for individuals with overlapping mutations in cervical LBC and peritoneal wash because the analysis of blood for the full cohort was outside of the scope of this exploratory study. While most mutations in cervical LBC and peritoneal wash were not found in blood, a more detailed analysis of TP53 mutations in blood would be helpful to fully define the contribution of leukocyte clones to these gynecological samples and the role of clonal hematopoiesis in this high-risk population.

In conclusion, ultra-deep sequencing has revealed increased pathogenic TP53 mutations in peritoneal washes of BRCA1 germline mutation carriers. These findings suggest that, while TP53 somatic evolution is prevalent during normal aging, it might be accelerated or enhanced in individuals with specific genetic predispositions to cancer, such as pathogenic BRCA1 germline mutations. Further studies of TP53 mutations in the fallopian tubes of BRCA carriers are warranted to explore the origin of TP53 pathogenic mutations observed in peritoneal washes and to determine the role of these mutations in HGSC predisposition.

Supplementary Material

2
3
Supplementary Figures

HIGHLIGHTS.

  • TP53 mutations are prevalent in cervical LBCs and peritoneal washes and their frequency increases with age

  • Pathogenic TP53 mutation burden in peritoneal wash is associated with BRCA1 germline mutation, chemotherapy, BMI, and age

  • Most mutations in cervical LBCs and peritoneal washes were not identified in blood DNA

  • Identical mutations were identified in cervical LBCs and peritoneal washes of some patients, but cell of origin is unknown

Acknowledgments

We thank all the members of the Risques and Kennedy laboratories for their support; Dr. Thierry Soussi for facilitating the use of Seshat; and all the patients who donated samples, without which this research would not have been possible. This work was supported in part by R01 CA259384 (RAR), R21 CA240885 (RAR), the Tom and Nancy Jensen Gynecologic Cancer Research Fund (BMN), and the NIH/NCI Cancer Center Support Grant P30 CA015704.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Conflict of Interest Disclosure Statement:

RAR is a consultant and equity holder at TwinStrand Biosciences Inc. and an equity holder at NanoString Technologies Inc. RAR is named inventor on patents owned by the University of Washington and licensed to TwinStrand Biosciences Inc. RAR received research funding from a joint research grant with TwinStrand Biosciences Inc. and Ovartec GmbH.

REFERENCES

  • 1.Bowtell DD, Bohm S, Ahmed AA, Aspuria PJ, Bast RC, Beral V, et al. Rethinking ovarian cancer II: reducing mortality from high-grade serous ovarian cancer. Nat Rev Cancer. 2015. Nov;15(11):668–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Kuchenbaecker KB, Hopper JL, Barnes DR, Phillips KA, Mooij TM, Roos-Blom MJ, et al. Risks of Breast, Ovarian, and Contralateral Breast Cancer for BRCA1 and BRCA2 Mutation Carriers. JAMA. 2017. Jun 20;317(23):2402–16. [DOI] [PubMed] [Google Scholar]
  • 3.Pearce CL, Stram DO, Ness RB, Stram DA, Roman LD, Templeman C, et al. Population distribution of lifetime risk of ovarian cancer in the United States. Cancer Epidemiol Biomark Prev. 2015. Apr;24(4):671–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Norquist BM, Harrell MI, Brady MF, Walsh T, Lee MK, Gulsuner S, et al. Inherited Mutations in Women With Ovarian Carcinoma. JAMA Oncol. 2016. Apr;2(4):482–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Finch AP, Lubinski J, Moller P, Singer CF, Karlan B, Senter L, et al. Impact of oophorectomy on cancer incidence and mortality in women with a BRCA1 or BRCA2 mutation. J Clin Oncol. 2014. May 20;32(15):1547–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Rush SK, Swisher EM, Garcia RL, Pennington KP, Agnew KJ, Kilgore MR, et al. Pathologic findings and clinical outcomes in women undergoing risk-reducing surgery to prevent ovarian and fallopian tube carcinoma: A large prospective single institution experience. Gynecol Oncol. 2020. May;157(2):514–520. [DOI] [PubMed] [Google Scholar]
  • 7.Shih IM, Wang Y, Wang TL. The Origin of Ovarian Cancer Species and Precancerous Landscape. Am J Pathol. 2021. Jan;191(1):26–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Labidi-Galy SI, Papp E, Hallberg D, Niknafs N, Adleff V, Noe M, et al. High grade serous ovarian carcinomas originate in the fallopian tube. Nat Commun. 2017. Oct 23;8(1):1093. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Wu R, Wang P, Lin S, Zhang M, Song Q, Chu T, et al. Genomic landscape and evolutionary trajectories of ovarian cancer precursor lesions. J Pathol. 2019. May;248(1):41–50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Gerstung M, Jolly C, Leshchiner I, Dentro SC, Gonzalez S, Rosebrock D, et al. The evolutionary history of 2,658 cancers. Nature. 2020. Feb;578(7793):122–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Soong TR, Howitt BE, Horowitz N, Nucci MR, Crum CP. The fallopian tube, “precursor escape” and narrowing the knowledge gap to the origins of high-grade serous carcinoma. Gynecol Oncol. 2019. Feb;152(2):426–433. [DOI] [PubMed] [Google Scholar]
  • 12.Berg CB, van den, Dasgupta S, Ewing-Graham PC, Bart J, Bulten J, Gaarenstroom KN, et al. Does serous tubal intraepithelial carcinoma (STIC) metastasize? The clonal relationship between STIC and subsequent high-grade serous carcinoma in BRCA1/2 mutation carriers several years after risk-reducing salpingo-oophorectomy. Gynecol Oncol. 2024. Aug 1;187:113–9. [DOI] [PubMed] [Google Scholar]
  • 13.Salk JJ, Schmitt MW, Loeb LA. Enhancing the accuracy of next-generation sequencing for detecting rare and subclonal mutations. Nat Rev Genet. 2018. May;19(5):269–85. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Schmitt MW, Kennedy SR, Salk JJ, Fox EJ, Hiatt JB, Loeb LA. Detection of ultra-rare mutations by next-generation sequencing. Proc Natl Acad Sci U S A. 2012. Sep 4;109(36):14508–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Krimmel JD, Schmitt MW, Harrell MI, Agnew KJ, Kennedy SR, Emond MJ, et al. Ultra-deep sequencing detects ovarian cancer cells in peritoneal fluid and reveals somatic TP53 mutations in noncancerous tissues. Proc Natl Acad Sci U A. 2016. May 24;113(21):6005–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Salk JJ, Loubet-Senear K, Maritschnegg E, Valentine CC, Williams LN, Higgins JE, et al. Ultra-Sensitive TP53 Sequencing for Cancer Detection Reveals Progressive Clonal Selection in Normal Tissue over a Century of Human Lifespan. Cell Rep. 2019. Jul 2;28(1):132–144 e3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Ghezelayagh TS, Kohrn BF, Fredrickson J, Manhardt E, Radke MR, Katz R, et al. Uterine lavage identifies cancer mutations and increased TP53 somatic mutation burden in individuals with ovarian cancer. Cancer Res Commun. 2022. Oct;2(10):1282–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Krimmel-Morrison JD, Ghezelayagh TS, Lian S, Zhang Y, Fredrickson J, Nachmanson D, et al. Characterization of TP53 mutations in Pap test DNA of women with and without serous ovarian carcinoma. Gynecol Oncol. 2020. Feb;156(2):407–414. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Ghezelayagh TS, Kohrn BF, Fredrickson J, Krimmel-Morrison JD, Latorre-Esteves E, Tee XR, et al. TP53 somatic evolution in cervical liquid-based cytology and blood from individuals with and without ovarian cancer and BRCA1 or BRCA2 germline mutations. Oncogene. 2024. Jul;43(31):2421–2430. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Kakiuchi N, Ogawa S. Clonal expansion in non-cancer tissues. Nat Rev Cancer. 2021;21(4):239–56. [DOI] [PubMed] [Google Scholar]
  • 21.Martincorena I. Somatic mutation and clonal expansions in human tissues. Genome Med. 2019. May 28;11(1):35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Risques RA, Kennedy SR. Aging and the rise of somatic cancer-associated mutations in normal tissues. PLoS Genet. 2018. Jan;14(1):e1007108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Rane JK, Frankell AM, Weeden CE, Swanton C. Clonal Evolution in Healthy and Premalignant Tissues: Implications for Early Cancer Interception Strategies. Cancer Prev Res Phila Pa. 2023. Jul 5;16(7):369–78. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Evans MA, Walsh K. Clonal hematopoiesis, somatic mosaicism, and age-associated disease. Physiol Rev. 2023. Jan 1;103(1):649–716. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Yizhak K, Aguet F, Kim J, Hess JM, Kubler K, Grimsby J, et al. RNA sequence analysis reveals macroscopic somatic clonal expansion across normal tissues. Science. 2019. Jun 7;364(6444). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Walsh T, Casadei S, Lee MK, Pennil CC, Nord AS, Thornton AM, et al. Mutations in 12 genes for inherited ovarian, fallopian tube, and peritoneal carcinoma identified by massively parallel sequencing. Proc Natl Acad Sci U A. 2011. Nov 1;108(44):18032–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Kennedy SR, Schmitt MW, Fox EJ, Kohrn BF, Salk JJ, Ahn EH, et al. Detecting ultralow-frequency mutations by Duplex Sequencing. Nat Protoc. 2014. Nov;9(11):2586–606. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.R Core Team. R: A Language and Environment for Statistical Computing [Internet]. Vienna, Austria: R Foundation for Statistical Computing; 2022. Available from: https://www.R-project.org/ [Google Scholar]
  • 29.Cheng J, Novati G, Pan J, Bycroft C, Žemgulytė A, Applebaum T, et al. Accurate proteome-wide missense variant effect prediction with AlphaMissense. Science. 2023. Sep 19;381(6664):eadg7492. [DOI] [PubMed] [Google Scholar]
  • 30.Lamb JD, Garcia RL, Goff BA, Paley PJ, Swisher EM. Predictors of occult neoplasia in women undergoing risk-reducing salpingo-oophorectomy. Am J Obstet Gynecol. 2006. Jun;194(6):1702–9. [DOI] [PubMed] [Google Scholar]
  • 31.Alexandrov LB, Jones PH, Wedge DC, Sale JE, Campbell PJ, Nik-Zainal S, et al. Clock-like mutational processes in human somatic cells. Nat Genet. 2015. Dec;47(12):1402–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Rahal Z, Scheet P, Kadara H. Somatic Mutations in Normal Tissues: Calm before the Storm. Cancer Discov. 2024. Apr 4;14(4):605–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Coombs CC, Zehir A, Devlin SM, Kishtagari A, Syed A, Jonsson P, et al. Therapy-Related Clonal Hematopoiesis in Patients with Non-hematologic Cancers Is Common and Associated with Adverse Clinical Outcomes. Cell Stem Cell. 2017. Sep 7;21(3):374–382.e4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Bolton KL, Ptashkin RN, Gao T, Braunstein L, Devlin SM, Kelly D, et al. Cancer therapy shapes the fitness landscape of clonal hematopoiesis. Nat Genet. 2020. Nov;52(11):1219–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Qian F, Rookus MA, Leslie G, Risch HA, Greene MH, Aalfs CM, et al. Mendelian randomisation study of height and body mass index as modifiers of ovarian cancer risk in 22,588 BRCA1 and BRCA2 mutation carriers. Br J Cancer. 2019. Jul;121(2):180–92. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Kim SJ, Lubinski J, Huzarski T, Møller P, Armel S, Karlan BY, et al. Weight Gain and the Risk of Ovarian Cancer in BRCA1 and BRCA2 Mutation Carriers. Cancer Epidemiol Biomarkers Prev. 2021. Nov 1;30(11):2038–43. [DOI] [PubMed] [Google Scholar]
  • 37.Suda K, Nakaoka H, Yoshihara K, Ishiguro T, Tamura R, Mori Y, et al. Clonal Expansion and Diversification of Cancer-Associated Mutations in Endometriosis and Normal Endometrium. Cell Rep. 2018. Aug 14;24(7):1777–89. [DOI] [PubMed] [Google Scholar]
  • 38.Moore L, Leongamornlert D, Coorens THH, Sanders MA, Ellis P, Dentro SC, et al. The mutational landscape of normal human endometrial epithelium. Nature. 2020. Apr;580(7805):640–6. [DOI] [PubMed] [Google Scholar]
  • 39.Yamaguchi M, Nakaoka H, Suda K, Yoshihara K, Ishiguro T, Yachida N, et al. Spatiotemporal dynamics of clonal selection and diversification in normal endometrial epithelium. Nat Commun. 2022. Feb 17;13(1):943. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Jaiswal S, Ebert BL. Clonal hematopoiesis in human aging and disease. Science. 2019. Nov 1;366(6465). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Mitchell SR, Gopakumar J, Jaiswal S. Insights into clonal hematopoiesis and its relation to cancer risk. Curr Opin Genet Dev. 2021. Feb 1;66:63–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Visvanathan K, Shaw P, May BJ, Bahadirli-Talbott A, Kaushiva A, Risch H, et al. Fallopian Tube Lesions in Women at High Risk for Ovarian Cancer: A Multicenter Study. Cancer Prev Res Phila. 2018. Nov;11(11):697–706. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Norquist BM, Garcia RL, Allison KH, Jokinen CH, Kernochan LE, Pizzi CC, et al. The molecular pathogenesis of hereditary ovarian carcinoma: alterations in the tubal epithelium of women with BRCA1 and BRCA2 mutations. Cancer. 2010. Nov 15;116(22):5261–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Folkins AK, Jarboe EA, Saleemuddin A, Lee Y, Callahan MJ, Drapkin R, et al. A candidate precursor to pelvic serous cancer (p53 signature) and its prevalence in ovaries and fallopian tubes from women with BRCA mutations. Gynecol Oncol. 2008. May;109(2):168–73. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

2
NIHMS2016232-supplement-2.pdf (43.7KB, pdf)
3
NIHMS2016232-supplement-3.xlsx (107.9KB, xlsx)
Supplementary Figures
NIHMS2016232-supplement-Supplementary_Figures.pdf (431.5KB, pdf)

RESOURCES