US11286531B2

US11286531B2 - Assaying ovarian cyst fluid

Info

Publication number: US11286531B2
Application number: US15/749,887
Authority: US
Inventors: Yuxuan Wang; Bert Vogelstein; Kenneth W. Kinzler; Luis Diaz; Nickolas Papadopoulos; Karin Sundfeldt; Bjorg Kristjansdottir
Original assignee: Johns Hopkins University
Current assignee: Johns Hopkins University
Priority date: 2015-08-11
Filing date: 2016-08-11
Publication date: 2022-03-29
Also published as: US20180258490A1; WO2017027653A1; US20220205048A1

Abstract

A diagnostic test for ovarian cysts is based on the detection of mutations characteristic of the most common neoplasms giving rise to these lesions. With this test, tumor-specific mutations were detected in the cyst fluids of 19 of 24 (79%) borderline tumors and 28 of 31 (90%) malignant ovarian cancers. In contrast, we detected no mutations in the cyst fluids from 10 non-neoplastic cysts and 12 benign tumors. When categorized by the need for exploratory surgery (i.e., presence of a borderline tumor or malignant cancer), the sensitivity of this test was 85% and the specificity was 100%. These tests could inform the diagnosis of ovarian cysts and improve the clinical management of the large number of women with these lesions.

Description

This application is a National Stage application under 35 U.S.C. § 371 of International Application No. PCT/US2016/046453, having an International Filing Date of Aug. 11, 2016, which claims the benefit of U.S. Provisional Application Nos. 62/203,573, filed Aug. 11, 2015, each of which is incorporated herein by reference in its entirety.

This invention was made with government support under CA 43460, 57345, and 62924 awarded by National Institutes of Health. The government has certain rights in the invention.

TECHNICAL FIELD OF THE INVENTION

This invention is related to the area of DNA analysis. In particular, it relates to analysis of genes in clinical samples.

BACKGROUND OF THE INVENTION

Ovarian cancer is the most lethal gynecologic malignancy, with 21,980 estimated new cases and 14,270 estimated deaths in the United States in 2014. Approximately 1.3% of women will be diagnosed with ovarian cancer during their lifetime (1). These cancers commonly present as an adnexal mass with cystic components, but are not associated with specific symptoms. As a result, two-thirds of ovarian cancers are diagnosed at late stage (Stage III and IV), when the 5-year survival is less than 30% (1).

Complicating the diagnosis of ovarian cancer is the fact that ovarian cysts are common in women of all ages, with a prevalence of 35% and 17% in pre- and post-menopausal women, respectively (2). These cysts are frequently benign and found incidentally on routine imaging (2). Though malignancy is an unusual cause of the cysts, 30% of the cysts exhibit radiographic features suspicious for malignancy, such as solid areas or mass (2). In addition to the anxiety that such findings provoke, many women undergo unnecessary surgery for cysts that are not malignant and may not be responsible for the symptoms they have. For example, only 5% of 570 women in a large ovarian cancer screening randomized trial who underwent surgical evaluation actually had a malignancy (3). Compounding this issue is the fact that surgery for ovarian cysts requires general anesthesia and is associated with significant morbidity, causing serious complications in 15% of women. These complications include damage to nerves and ureters, bleeding, infection, perforation of adjacent viscera, as well as hormonal and fertility loss (in the case of bilateral oophorectomy) (4). Even minimal procedures such as ovarian cystectomy can affect fertility in premenopausal women by decreasing follicular response and oocyte number (5, 6). If a preoperative test could be performed that indicated whether the cystic lesion was benign or malignant, unnecessary surgery and its associated complications could be avoided in a large number of patients, particularly women of reproductive age who wish to preserve their fertility, as well as women whose medical comorbidities or functional status makes anesthesia and surgery hazardous.

Ovarian cysts and tumors are classified as non-neoplastic, benign, borderline, or malignant based on microscopic examination after surgical removal (FIG. 1). Non-neoplastic cysts are by far the most common class of ovarian cyst. They are frequently functional in pre-menopausal women, arising when an egg is not released properly from either the follicle or corpus luteum and usually resolve spontaneously within several months (7). Benign cystic tumors, such as cystadenomas and cystadenofibromas, rarely progress to malignancy (8, 9). No genetic alterations have yet been identified in either non-neoplastic cysts or in benign cystic tumors (9). Neither of these cyst types requires surgery unless they are symptomatic or have undergone torsion (8). These cysts can be easily sampled with ultrasound-guided fine-needle aspiration in an outpatient setting without the need for anesthesia (10).

At the other end of the spectrum are epithelial ovarian cancers, which are potentially lethal and unequivocally require surgery. A dualistic model has been proposed to classify these neoplasms (11). Type I tumors are composed of low-grade serous, low-grade endometrioid, clear cell, and mucinous carcinomas. They are clinically indolent, frequently diagnosed at early stage (Stage I or II), and develop from well-established precursor lesions (“borderline” or “atypical proliferative” tumors, as described below) (12). Type I cancers commonly exhibit mutations in KRAS, BRAF, CTNNB1, PIK3CA, PTEN, ARID1A, or PPP2R1A (11). In contrast, type II tumors are generally high-grade serous carcinomas. They are highly aggressive, most often diagnosed in late stage (Stage III or IV), and have suggested origins from the distal fallopian tube (13). Type II cancers almost always harbor TP53 mutations (14). Also unlike type I cancers, which are relatively chemo-resistant and more often treated only with surgical excision, type II cancers respond to conventional chemotherapy, particularly after maximal debulking to reduce tumor burden (15, 16).

“Borderline” or “atypical proliferative” tumors lie in the middle of this spectrum, between the malignant cancers and the relatively harmless (non-neoplastic or benign) lesions. They are distinguished from carcinomas by the absence of stromal invasion and are precursors of type I cancers. In light of their potential for malignancy, the standard of care for borderline tumors is surgical excision. Following surgery, the prognosis is excellent compared to ovarian cancers, with 5-year survival rates over 85% (17). A minor but significant portion of borderline tumors recur after surgery, however, and a subset of the recurrences are found to have advanced to type I cancers (18). This progression is consistent with molecular findings: serous borderline tumors typically exhibit mutations in BRAF or KRAS, like their malignant counterparts (low-grade serous carcinoma) (19, 20). The presence of a BRAF mutation in a borderline tumor is associated with better prognosis and a low probability of progression to carcinoma (21). In contrast, KRAS mutations are associated with the progression to type I cancers (22).

The examination of fluids from pancreatic, renal, and thyroid cysts is routinely used in clinical management (23-25). The fluids have historically been studied by cytology to identify malignant cysts. Ovarian cysts share many features with these other types of cysts, in that they are common, often diagnosed incidentally, and are nearly always benign. However, aspiration of ovarian cyst fluid for cytology is not standard-of-care. From a historical perspective, the difference in diagnostic management probably lies in the fact that cytology has not proven to be very informative for ovarian cysts, particularly for distinguishing benign vs. borderline tumors (26, 27).

More recently, genetic analysis of specific types of cyst fluids has been considered as an aid to cytology, given that conventional cytology often has limited sensitivity and specificity (23). Based on the emerging success of the molecular genetic evaluation of other types of cysts, we reasoned that a similar approach could be applied to ovarian cysts. Evaluation of DNA from cells and cell fragments shed into the cyst fluid would presumably allow the identification of tumor-specific mutations. Unlike other, conventional markers of neoplasia such as CA-125, cancer gene mutations are exquisitely specific indicators of a neoplastic lesion (29). Moreover, the type of mutation can in some cases indicate the type of neoplastic lesion present (30). Yamada et al. have demonstrated that mutations can be detected in the cystic fluid of ovarian tumors by querying exons 4 to 9 of TP53, achieving sensitivities of 12.5% and 10%, for borderline and malignant tumors, respectively (31). Extremely sensitive methods for mutation detection, capable of identifying one mutant template allele among thousands of normal templates in a panel of genes, have recently been developed (32-34). In this study, we here applied one of these technologies to determine whether mutations could be identified in ovarian cyst fluids, and if so, whether they provided information that could in principle be used in diagnosis and management.

Because there is currently no reliable way to determine whether an ovarian cyst is malignant prior to surgical excision, many women undergo unnecessary, invasive surgeries for non-malignant lesions. There is a need in the art for techniques to determine whether surgery is required or unnecessary.

SUMMARY OF THE INVENTION

According to one aspect of the invention a method is provided in which ovarian cyst fluid is tested for mutations in a panel of genes frequently mutated in ovarian neoplasms, wherein the panel comprises BRAF, KRAS, and TP53.

According to another aspect of the invention a method is provided in which ovarian cyst fluid is tested for mutations in a panel of genes frequently mutated in ovarian neoplasms, wherein the panel comprises BRAF, KRAS, TP53, and one or more of AKT1, APC, BRCA1, BRCA2, CDKN2A, EGFR, FBXW7, FGFR2, MAPK1, NRAS, PIK3R1, and POLE.

According to another aspect of the invention a method is provided in which ovarian cyst fluid is tested for mutations in a panel of genes frequently mutated in ovarian neoplasms, wherein the panel comprises BRAF, KRAS, TP53, and one or more of CTNNB1, PIK3CA, PTEN, ARID1A, and PPP2R1A.

According to an additional aspect of the invention a method is provided in which ovarian cyst fluid is tested for mutations in a panel of genes frequently mutated in ovarian neoplasms, wherein the panel comprises BRAF, KRAS, TP53, AKT1, APC, BRCA1, BRCA2, CDKN2A, EGFR, FBXW7, FGFR2, MAPK1, NRAS, PIK3R1, and POLE.

According to an additional aspect of the invention a method is provided in which ovarian cyst fluid is tested for mutations in a panel of genes frequently mutated in ovarian neoplasms, wherein the panel comprises BRAF, KRAS, TP53, CTNNB1, PIK3CA, PTEN, ARID1A, and PPP2R1A.

According to an additional aspect of the invention a method is provided in which ovarian cyst fluid is tested for mutations in a panel of genes frequently mutated in ovarian neoplasms, wherein the panel comprises BRAF, KRAS, TP53, AKT1, APC, BRCA1, BRCA2, CDKN2A, EGFR, FBXW7, FGFR2, MAPK1, NRAS, PIK3R1, POLE, CTNNB1, PIK3CA, PTEN, ARID1A, and PPP2R1A.

These and other embodiments which will be apparent to those of skill in the art upon reading the specification provide the art with powerful methods for assessing ovarian cysts without recourse to unnecessary surgeries.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1. Schematic showing classes of ovarian cysts and the diagnostic potential of the cyst fluid. Ovarian cysts and tumors are currently classified according to microscopic evaluation after surgical removal. The majority of ovarian cysts are non-neoplastic (often “functional” in premenopausal women). Ovarian tumors with combined cystic and solid components are either benign tumors, borderline tumors, or malignant cancers (type I or II). Only cysts associated with borderline tumors and cancers require surgical excision. We show here that the DNA purified from cyst fluid can be analyzed for somatic mutations commonly found in their associated tumors. The type of mutation detected not only indicates the type of tumor present but also could inform management.

FIGS. 2A-2B. Mutant allele fractions. (FIG. 2A) Classification by tumor type. No mutations were found in the DNA of non-neoplastic or benign cysts (red). Of the cysts that required surgery (blue), the median mutant allele fraction was higher in the cyst fluids associated with type II cancer (60.3%) than type I (7.8%) or borderline tumors (2.4%). (FIG. 2B) Classification by tumor stage. The DNA from cyst fluids of late-stage cancers had a higher median mutant allele fraction (51.2%) than those of early-stage cancers (7.4%) or borderline tumors (2.4%). Horizontal bars depict median and IQR.

FIGS. 3A-3D (Table 1.) Patient demographics. The clinical characteristics of patients in this study and their tumor characteristics are indicated.

FIGS. 4A-4E. (Table 2.) Mutations identified in tumors and cyst fluids. The mutations, mutant allele fractions, and amount of DNA recovered from cyst fluids are indicated.

FIG. 5 (Table 3.) Detection of tumor-specific mutations in cyst fluid. The fraction of samples detected and the median fraction of mutant alleles are indicated, grouped by cyst type, cancer stage, and the need for surgery.

FIG. 6 (Table 4.) Multivariate analysis for markers associated with need for surgery. The presence of a mutation, cyst DNA amount, and common serum biomarkers for ovarian cancer were analyzed for association with cysts that require surgical removal (Firth's penalized likelihood logistic regression).

FIG. 7A-7C (FIG. S1.) Markers associated with the need for surgery. Cyst DNA amount and levels of commonly used ovarian cancer serum biomarkers are plotted according to the cyst type and need for surgery. (FIG. 7A) The amounts of DNA in cyst fluids was generally higher in cysts requiring surgery (blue) than those that do not (red), but no significant correlation was found (p=0.69). (FIG. 7B) CA-125 levels were significantly higher in cysts that required surgery than those that do not (p=0.01). (FIG. 7C) Serum HE4 levels was not correlated with the need for surgery (p=0.92). P-values were calculated using Firth's penalized likelihood logistic regression in a multivariate model (See Example 1).

FIGS. 8A-811 (Table S1) Primer sequences used in multiplex assay; Forward primers (SEQ ID NO: 1-133); Reverse primers (SEQ ID NO: 134-266).

FIG. 9A-9B Mutated genes found in the cyst fluid samples. FIG. 9A shows non-neoplastic, benign, and borderline. FIG. 9B shows malignant Type I and malignant Type II. Yellow boxes represent mutations with mutant allele frequency (MAF) between 0.1% and 1%; orange boxes represent mutations with MAF between 1 and 10%; red boxes represent mutations with MAF greater than 10% (* indicates patients with insufficient DNA for analysis; ** indicates patients with two detected mutations).

DETAILED DESCRIPTION OF THE INVENTION

The inventors have developed an assay for testing cyst fluids. Cyst fluids are typically aspirated by a needle, preferably a fine needle. The aspiration can be performed under the guidance of a radiological technique, such as ultrasound. Other guidance techniques can be used as convenient. Cyst fluids can typically be collected from any type of ovarian cyst or cystic neoplasm, and the term “cyst” is used here to refer to all types of ovarian growths with a cystic component.

Non-neoplastic ovarian cysts typically do not require surgical removal and do not display mutations. In contrast, ovarian cysts that are associated with malignancy do require surgical removal and frequently display mutations; these mutations can further indicate the type and severity of the disease. Testing for a panel that includes markers for a broad range of ovarian cysts permits the identification of cyst type and prognosis. It also permits a clinical decision to surgically remove or not.

Other markers and clinical indication can be used in combination with the ovarian cyst fluid assay results. Plasma markers such as CA-125 and HE4 can be assessed in patient plasma. Other protein or genetic markers can be used in conjunction with the ovarian cyst fluid assay. Other clinical indicators, including radiological findings and physical findings may be used in conjunction with the ovarian cyst fluid assay.

Testing may be performed using any technique that is targeted for particular genes. These are not techniques that screen for any and all gene mutations. Rather, they are designed to detect mutations in certain predetermined genes. In some cases they are designed to detect certain mutations or mutations in certain codons. Any analytic technique can be used for detecting mutations as is convenient, efficient, and sufficiently sensitive to detect mutations in ovarian cyst fluid. The assays may be hybridization based, such as using specific probes or specific primers. The assays may employ labeled probes or primers. The assays may employ labeled secondary reagents that permit the primary reagents to be detected. Such labels include radiolabels, fluorescent labels, enzymatic labels, chromophores, and the like.

A variety of different mutation types can be detected and may be useful in providing prognosis or management decisions. Such mutations include LOH, point mutations, rearrangements, frameshifts, point mutations, and copy number variations. Specific detection techniques for these mutation types or generic detection techniques may be used. It may be desirable to use control samples from other parts of the patient's body, such as a body fluid, like plasma, saliva, urine, feces, and the like. Alternatively other control samples may include tissues such as normal tissue from a non-ovary, or cells or tissues from the ovarian cyst wall.

Cyst fluid may be obtained by any technique known in the art, including but not limited to needle aspiration. The aspiration may optionally be guided by a radiological technique such as ultrasound. Cyst fluid may be aspirated before or after initial surgical removal or subsequent surgical removal.

In some embodiments, primers will incorporate unique identification DNA sequence (UID) which are molecular barcodes. These can be randomly generated and attached to templates as a means to reduce errors arising from amplification and sequencing. Probes, primers, and UIDs can incorporate non-naturally occurring modifications to DNA sequences, by internucleotide linkage modifications, by sugar modifications, and by nucleobase modifications. For example, phosphorothioate (PS) linkages can be used in which sulfur substitutes for one nonbridging phosphate oxygen. This imparts resistance to nuclease degradation. Other modifications which can be used include N3′ phosplioramidate (NP) linkages, Boranophospliate internucleotide linkages, Phosphonoacetate (PACE) linkages, Morpholino phosphoramidates, Peptide nucleic acid (PNA), 2′-O-Me nucleoside analog, 2′F-RN A modification, 2′-deoxy-2′-fluoro-β-D-arabino nucleic acid (2′F-ANA) modification and Locked nucleic acid (LNA).

Other techniques which are unbiased toward particular genes can be used as well for assessing genes of interest in cyst fluid. Such techniques include whole-genome or whole exome techniques. These may include assessments by nucleotide sequencing. The nucleotide sequencing may be redundant nucleotide sequencing. Targeted sequencing methods can be used as well.

The methods described here achieve high degrees of sensitivity and specificity. Sensitivity may be at least 15%, at least 20%, at least 25%, at least 50%, at least 60%, at least 70%, at least 80%, at least 85%, at least 90%, at least 95% for borderline and malignant tumors. Specificity may be at least 15%, at least 20%, at least 25%, at least 50%, at least 60 at least 70%, at least 80%, at least 85%, at least 90%, at least 95% for borderline and malignant tumors.

Removal of ovarian cyst fluid assay from the body can be accomplished before any surgery occurs. Thus the results of the assay can help guide the decision to perform surgery. If surgery has occurred to remove the ovarian cyst, and if it returns, a sample of ovarian cyst fluid may be obtained from the body at that time. The assays will typically be performed in a clinical laboratory on samples that have been removed by a skilled clinician, such as an interventional radiologist or a surgeon. The samples may be assayed immediately or they may suitable stored and or shipped for testing. It is possible that DNA will be extracted from the sample prior to shipping it to a laboratory for testing. Results will generally be communicated back from the assaying laboratory to the clinician for communication to a patient. Results may be recorded in paper or electronic medical records.

Ovarian cancer is the most lethal gynecologic cancer in women. However screening is not recommended by the U.S. Preventive Services Task Force using current diagnostic approaches, which too frequently lead to “important harms, including major surgical interventions in women who do not have cancer” (Moyer and Force, 2012). We have demonstrated here that driver mutations in ovarian tumors are also present in their associated cyst fluids. Moreover, the mutant allele frequencies in the cyst fluids are relatively high (median 12.6%, IQR of 2.7% to 40.2%), facilitating their detection. There were no mutations detected in the cyst fluids that were not also present in the tumors, and vice versa. Also importantly, no mutation was identified in non-neoplastic cysts or cysts associated with benign tumors. Overall, mutations were detected in a major fraction (87%) of cysts requiring surgery but not in any cyst that did not require surgery.

Our results demonstrate that mutations present in ovarian tumors are also present in their associated cyst fluids. Moreover, the mutant allele frequencies in the cyst fluids are relatively high (median 12.6%, IQR of 2.7% to 40.2%), facilitating their detection. There were no mutations detected in the cyst fluids that were not also present in the tumors, and vice versa. And most importantly, mutations were detected in a major fraction (85%) of cysts requiring surgery but not in any cyst that did not require surgery (Tables 2 and 3).

Although most (85%) of the 55 cysts requiring surgery had detectable mutations in their fluidic compartment, eight did not. All of these eight cysts occurred in borderline tumors or type I cancers, while mutations were always (100%) detectable in type II cancers (Tables 2 and 3). There are two potential explanations for our failure to detect mutations in these eight cysts. First, it is possible that the mutant DNA concentration in these cysts was below the level of technical sensitivity of our assay (˜0.1% mutant allele fraction). We excluded this possibility by evaluating the tumors themselves: no mutations were detected in any of the tumors from these 8 patients. The second, and therefore more likely explanation, is that our panel of 133 amplicons, containing regions of 17 genes, was not adequate to capture the mutations that were present. Unlike type II cancers, which nearly always contain TP53 mutations (94% of the type II cancers we studied, for example), the genomic landscapes of type I cancers and borderline tumors are more heterogeneous and not as well studied (II). Further genetic evaluation of these tumors should facilitate the incorporation of additional amplicons in the panel to reach higher sensitivities. Nevertheless, the 100% sensitivity for type II cancers in our study is highly encouraging, given that these cancers account for over 90% of ovarian cancer deaths.

One limitation of our study is the number of patients evaluated. Though excision of ovarian cysts is one of the most commonly performed surgical procedures, banking of cyst fluids is not common, even in academic centers. Thus, we only had relatively small numbers (n=22) of non-neoplastic cysts and benign tumors available for study. Even so, the differences in genetic alterations among the various cyst types were striking (Tables 2 and 3). Our study will hopefully stimulate collection and analyses of ovarian cyst fluids that will be able to establish smaller confidence limits around the sensitivities and specificities reported in the current study.

A potential clinical limitation of our approach is the concern by gynecologists that needle puncture of a malignant ovarian cyst leads to seeding of the peritoneum. This concern is based on inconclusive evidence about the dangers of cyst rupture during surgery and is, at best, controversial (40). Moreover, leakage is expected to be much less likely when a tiny needle is inserted into the cyst under ultrasound-guidance than when cysts are manipulated during surgery. The idea that malignant cysts might shed cancer cells if needle-punctured also seems incongruent with the widespread practice of laparoscopic removal of ovarian cysts (41). Laparoscopic removal of a cyst carries a risk of cyst rupture, perhaps higher than needling (42). Finally, malignant pancreatic cysts are at least as dangerous as malignant ovarian cysts, yet the standard-of-care for pancreatic cysts involves repeated sampling of cyst fluid through endoscopic ultrasound over many years (43, 44). Though pancreatic cysts and ovarian cysts lie in different anatomical compartments, it is encouraging that aspiration of pancreatic cysts is not associated with an increased risk of mortality in patients with pancreatic cancer (45). Finally, recent advancements in methods to plug biopsy tracts, using materials such as absorbable gelatin slurry and torpedo, can further decrease the risk of tumor spillage associated with fine-needle aspirations (46, 47). On the basis of these observations and recent developments, we believe that ultrasound-guided aspiration of ovarian cyst fluids would likely be a safe and well-tolerated procedure.

As noted in the background of the invention section above, seven to ten patients with benign ovarian cyst lesions undergo surgery for each case of ovarian cancer found (48). In addition to the psychological impact a potential diagnosis of cancer has on patients, surgery for benign lesions entails considerable cost and morbidity. OVA-1 is the only FDA-cleared test to date that aims to distinguish benign versus malignant adnexal mass. It measures levels of five serum markers (CA-125, β-2 microglobulin, apolipoprotein A1, prealbumin, and transferrin) and is used to stratify patients who should consult a gynecologic oncologist rather than a general gynecologist for surgery. However the test has a specificity of 43% for ovarian cancer, which is even lower than that of CA-125 alone (49). While the test might encourage patients with suspected ovarian cancer to seek specialized care, it would not decrease the number of unnecessary surgeries for women with benign adnexal masses.

This study was driven by the need for a biomarker that would help distinguish malignant ovarian tumors from benign lesions and thereby reduce the number of unnecessary surgeries. Such distinction is often difficult based on symptoms and conventional diagnostic criteria. For example, in a large study of 48,053 asymptomatic postmenopausal women who underwent ultrasound examination by skilled sonographers, 8 (17%) of the 47 ovarian cancers that were identified occurred in women with persistently normal sonographic findings (Sharma et al., 2012). All eight cases were type II cancers, highlighting the potential utility of an additional assay to detect this highly lethal and aggressive type of ovarian cancer. On the other hand, of the 4367 women with abnormal sono graphic findings, less than 1% of cases proved to have malignancy upon surgery. Furthermore, of the 32 women with borderline or Type I cancers diagnosed, 22 (69%) had a serum CA-125 level within the clinically accepted normal range (≤35 units/mL). In our study, 18 of 18 (100%) type II cancers were detectable by virtue of the mutations found in cyst fluid DNA while none of the 18 benign or non-neoplastic cyst fluid contained such mutations. It is also important to note that the readout of our assay is quantitative and not dependent on the skill level of the reader (in contrast to sonography). Finally, the procedure can be performed minimally invasively in an outpatient setting. The goal of our test is not to replace clinical, radiologic, or sonographic evaluation but to augment them with molecular genetic markers.

Our study, though only proof-of-principle, illustrates one route to improve management of patients with ovarian cysts. Genetic analysis is not the only such route; proteomics could also provide clues to the correct diagnosis (50, 51). One can easily imagine how such additional information could be used to inform clinical practice in conjunction with current diagnostic methods. For example, if a cyst contained low amounts of DNA, no detectable mutations, and if the patient had low CA-125 levels, our data suggest that it is very unlikely to be a borderline tumor or malignant lesion. Either no surgery, or laparoscopic rather than open surgery, could be recommended for that patient, even if there was some solid component upon imaging. The option to avoid surgery would be particularly valuable for pre-menopausal women who generally have a low risk of ovarian cancer and might wish to preserve their fertility, as well as patients who are poor surgical candidates. However, our assay in its current format cannot completely rule out malignancy because a fraction of early-stage cancer patients did not have detectable mutations in their cysts. Therefore, patients whose clinical and functional status allows them to undergo surgery and anesthesia might still choose to have a surgical procedure. On the other hand, a minimally invasive test that provides additional, orthogonal information to patients and surgeons could inform their decision about the advisability of surgery.

Our data suggest that a cyst without any solid component upon imaging, and thereby unlikely via conventional criteria to be malignant, should be removed promptly if the cyst fluid contained a TP53 mutation. Radical, rather than conservative, surgery might be appropriate due to the high likelihood of an aggressive type II cancer. In contrast, if a BRAF mutation was identified, the lesion is presumably a borderline or low-grade tumor; thus conservative rather than radical surgery might be sufficient. Lastly, given that certain types of ovarian cancers (type II) tend to respond well to chemotherapy while others (type I) are relatively chemo-resistant, knowing the type of cancer present prior to surgery based on the mutation profile could help guide decisions regarding the use of neoadjuvant chemotherapy. Validation of these data in a much larger, prospective trial will of course be required before incorporation of this approach into clinical practice.

The above disclosure generally describes the present invention. All references disclosed herein are expressly incorporated by reference. A more complete understanding can be obtained by reference to the following specific examples which are provided herein for purposes of illustration only, and are not intended to limit the scope of the invention.

Example 1—Materials and Methods

Patient Samples

Cyst fluids were collected prospectively from 77 women presenting with a suspected ovarian tumor. Patients were diagnosed by transvaginal sonography or computed tomography and admitted for surgical removal of the cyst by gynecologic oncology surgeons at Sahlgrenska University Hospital, Gothenburg, Sweden. The study was approved by the ethical board of Gothenburg University and patients provided written consent. According to the approved protocol, ovarian cyst fluids were collected after removal of the cyst from the abdomen. All samples were immediately put in 4° C. for 15-30 minutes, centrifuged for 10 minutes at 500 g, and aliquoted into Eppendorf tubes. The fluids were transferred to −80° C., within 30-60 minutes after collection. All histology was reviewed by board-certified pathologists (Table 1).

Plasma HE4 concentrations were determined using a commercial HE4 EIA assay (Fujirebio Diagnostics) and plasma CA-125 levels were measured using the Architect CA 125 II (Abbott Diagnostics, USA). DNA was purified from tumor tissue (either freshly-frozen, or formalin-fixed and paraffin-embedded) after microdis section to remove neoplastic components. DNA was purified from tumors and from cyst fluids using an AllPrep DNA kit (Qiagen) according to the manufacturer's instructions. Purified DNA from all samples was quantified as previously described (52).

Statistical Analysis

A Wilcoxon rank-sum test was used to compare the amount of DNA in the cancers and borderline tumors with the amount of DNA in the simple cysts and benign tumors. The fraction of samples detected by tumor-specific mutations in the cyst fluid, as well as their 95% confidence intervals, was calculated for each tumor type (Table 3). When the presence of a mutation in the cyst fluid was used to predict the need for surgery, the sensitivity and specificity of the test, as well as their 95% confidence intervals, were calculated. Firth's penalized likelihood logistic regression was used to quantify the association between molecular features of cyst fluids and the need for surgery (Table 4) in a multivariate model. The model predictors included the presence of mutation, log 10(ng) of cyst DNA and indicators for normal CA-125 and HE4 values. Normal CA-125 values were defined as <35 U/mL and normal HE4 values were defined as <92 pmol/L and <121 pmol/L for pre- and post-menopausal women, respectively. Statistical analyses were performed using the R statistical package (version 3.1.2). Unless noted otherwise, all patient-related values are reported as means±SD.

Mutation Detection and Analysis

DNA from either cyst fluids or tumors was used for multiplex PCR, as previously described (34). One-hundred-and-thirty-three primer pairs were designed to amplify 110 to 142 bp segments containing regions of interest from the following 17 genes: AKT1, APC, BRAF, CDKN2A, CTNNB1, EGFR, FBXW7, FGFR2, KRAS, MAPK1, NRAS, PIK3CA, PIK3R1, POLE, PPP2R1A, PTEN, and TP53. Primer sequences are listed in Table S1. These primers were used to amplify DNA in 25 μL reactions as previously described (34). For each sample, three multiplex reactions, each containing non-overlapping amplicons, were performed. Reactions were purified with AMPure XP beads (Beckman Coulter) and eluted in 100 μL of Buffer EB (Qiagen). A fraction (2.5 μL) of purified PCR products were then amplified in a second round of PCR, as described (34). The PCR products were purified with AMPure and sequenced on an 11lumina MiSeq instrument.

We used Safe-SeqS, an error-reduction technology for detection of low frequency mutations as described to distinguish better between genuine mutations in the samples and artifactual variants arising from sequencing and sample preparation steps, (34). High quality sequence reads were selected based on quality scores, which were generated by the sequencing instrument to indicate the probability a base was called in error. The template-specific portion of the reads was matched to reference sequences. Reads from a common template molecule were then grouped based on the unique identifier sequences (UIDs) that were incorporated as molecular barcodes. Artifactual mutations introduced during the sample preparation or sequencing steps were reduced by requiring a mutation to be present in >90% of reads in each UID family (i.e., to be scored as a “supermutant”). In addition, DNA from normal individuals was used as a control to identify potential false positive mutations (see main text). Only supermutants in samples with frequencies far exceeding their frequencies in control DNA samples (i.e., >mean+5 standard deviations) were scored as positive.

Example 2—Characteristics of the Tumors and Cyst Fluid Samples

DNA was isolated from surgically excised ovarian cysts of 77 women. Ten of them had non-neoplastic cysts, 12 had benign tumors, 24 had borderline tumors, and 31 had cancers (13 Type I and 18 Type II). Age, histopathologic diagnosis, stage, and other clinical information are provided in Table 1. The median amount of DNA recovered from the cysts was 222 ng (interquartile range (IQR) of 53 to 3120 ng) (Table 2). There was no significant difference in the amounts of DNA between borderline tumors and type I or type II cancers (Table 2). However, the borderline tumors and cancers contained significantly more DNA than the non-neoplastic cysts or benign tumors (4453±6428 ng vs. 62±64 ng; p<0.001, Wilcoxon rank-sum test).

Example 3—a Multiplex PCR-Based Test to Identify Tumor-Specific Mutations in Cyst Fluid Samples

We designed a multiplex PCR-based test that could simultaneously assess the regions of 17 genes frequently mutated in ovarian tumors. The amount of DNA shed from neoplastic cells was expected to be a minor fraction of the total DNA in the cyst fluid, with most DNA emanating from normal cells. We therefore used a sensitive detection method, called Safe-SeqS (Safe-Sequencing System), to identify mutations in cyst fluid samples (34). In brief, primers were designed to amplify 133 regions, covering 9054 distinct nucleotide positions within the 17 genes of interest (Table S1). Three multiplex PCR reactions, each containing non-overlapping amplicons, were then performed on each sample. One primer in each pair included a unique identifier (UID) for each template molecule, thereby drastically minimizing the error rates associated with PCR and sequencing, as described previously (34) (Table S1). Under the conditions used in the current experiments, mutations present in >0.1% of template molecules could generally be reliably determined. We could not perform sequencing on five cysts (two simple cysts, two cystadenomas, one borderline tumor) because there was insufficient DNA (<3 ng recovered), and these were scored in a conservative fashion, as “negative” for mutations. When this test was applied to the 22 cyst fluids obtained from patients with simple cysts (n=9) or benign tumors (n=13), no mutations were identified (Tables 2 and 3). This was in stark contrast to the fluids obtained from the 18 patients with type II cancers, all of which were found to contain a mutation (Tables 2 and 3). Ten (77%) of the 13 cyst fluids from patients with type I cancers and 19 (79%) of the 24 cyst fluids from patients with borderline tumors contained at least one detectable mutation. When categorized by the need for surgery (i.e., presence of a borderline tumor or a type I or type II cancer), the sensitivity of this test was 85% (47 of 55 cysts; 95% confidence interval of 73% to 94%) and the specificity was 100% (95% confidence interval of 78% to 100%; Table 3).

Ovarian cancers are generally detected only late in the course of disease, explaining the poor prognosis of patients. Accordingly, only 11 of the 31 cysts associated with cancers in our study had early (Stage I or II) disease (Table 1). As expected, most of these were type I carcinomas (n=8). Nevertheless, it was encouraging that mutant DNA could be detected in nine (82%) of these 11 patients (Table 3). Mutations could be detected in 95% of the 20 patients with Stage III or IV cancers (Table 3).

A variety of control experiments were performed to confirm the integrity of these results. One informative positive control was provided by the results of sequencing of DNA from the tumors, using the identical method used to analyze DNA from the cyst fluids. Fifty-three of the 55 borderline and malignant cases had tumor available for this purpose. Every mutation identified in a tumor was found in its cyst fluid, and vice versa. As expected, the mutant allele frequencies in the tumors were often, but not always, higher than in the cyst fluid (Table 2). As another positive control, we used an independent PCR and sequencing reaction to confirm each of the cyst fluid mutations listed in Table 2. This validated not only the presence of a mutation, but also confirmed its fractional representation. The median relative difference between the fractions of mutant alleles in replicate experiments was 7.0% (IQR of 3.5% to 8.9%). Finally, four patients were found to have two independent mutations (Table 2). For example, the cyst fluid of patient OVCYST 081, who had high-grade endometrioid carcinoma, had a missense mutation (R280K) in TP53 plus an in-frame deletion of PIK3R1 at codons 458 and 459 of PIK3R1. The TP53 mutation was found in 3.0% of alleles while the PIK3R1 mutation was found in 3.7% of the alleles analyzed. Similar mutant allele frequencies among completely different mutations in the cyst fluid of three other patients provided further indicators of reproducibility (Table 2). All genetic assays were performed in a blinded manner, with the operator unaware of the diagnoses of the patients from whom the cyst fluids were obtained.

In addition to DNA from normal individuals used as controls, additional negative controls were provided by the simple cysts and benign tumors. Using the identical assay, none of the DNA from their cyst fluids contained detectable mutations (Table 2). A final control was provided by the borderline and malignant tumors themselves. In general, only one or two of the 9054 base-pairs (bp) queried were mutated in any one tumor (Table 2). The other ˜9000 bp could then be independently queried in the corresponding cyst fluid, and none of these positions were found to be mutated.

Example 4—Relationship Between the Type of Tumor Present and the Type of Mutation Found in the Associated Cyst Fluid Sample

The mutant allele fractions in the cyst fluids tended to be higher in the type II cancers (median of 60.3%) than the type I cancers (median of 7.8%) or borderline tumors (median of 2.4%), though there was considerable overlap (Tables 2 and 3). On the other hand, the type of mutation varied considerably among these cysts. In type I tumors, the genes mutated were BRAF (n=1), KRAS (n=5), NRAS (n=1), PIK3R1 (n=1), PPP2R1A (n=1), PTEN (n=1), or TP53 (n=3). Two distinct mutations were found per sample in three type I cancers. One type I cancer had a BRAF mutation. This BRAF mutation (V600_S605>D) is unusual that it resulted from an in-frame deletion/insertion rather than the base substitution (V600E) characteristic of the vast majority of BRAF mutations reported in the literature. This mutation has been observed in a papillary thyroid cancer and a cutaneous melanoma (35, 36). The deletion results in loss of a phosphorylation site in the activation loop of BRAF, while the insertion of an aspartic acid has been suggested to increase BRAF kinase activity by mimicking an activating phosphorylation (37). In contrast, all but one type II cancers (94% of 18) had mutations in TP53; the only exception was OVCYST 081, a high-grade endometrioid carcinoma. The borderline tumors were distinguished by yet a different pattern from that of the either type I or type II cancers. Of the 19 mutations in borderline tumors, 12 (63%) were at BRAF V600E, never observed in type I or type II cancers, and the remainder were at KRAS 12 or 61 (Table 2).

Example 5—Markers Associated with the Need for Surgery

A multivariate analysis was used to identify the most informative molecular features of cyst fluids and to compare them to the commonly used serum biomarkers for ovarian cancer, HE4 (human epididymis protein 4) and CA-125 (38, 39) (Table 4). We defined “informative” as indicating a need for surgery (i.e., borderline tumors or type I or II cancers). The amount of DNA in cyst fluids was generally, but not significantly, higher in the cysts requiring surgery (p=0.69, Table 4), though there were many cysts not requiring surgery that had higher DNA levels than cysts requiring surgery (FIG. S1A). Similarly, the serum CA-125 levels were significantly higher in cysts requiring surgery (p=0.01, Table 4), but there were many cysts not requiring surgery that had higher levels than those requiring surgery (FIG. S1B). Serum HE4 levels were not correlated with cyst type (P=0.92, Table 4; FIG. S1C). On the other hand, the presence of a mutation was highly informative for the presence of a cyst requiring surgery in the multivariate analysis, as no mutations were found in cysts not requiring surgery (P<0.001, Table 4).

REFERENCES

The disclosure of each reference cited is expressly incorporated herein.

1. N. A. Howlader N, Krapcho M, Garshell J, Miller D, Altekruse S F, Kosary C L, Yu M, Ruhl J, Tatalovich Z, Mariotto A, Lewis D R, Chen H S, Feuer E J, Cronin K A. (Bethesda, Md., 2014).
2. E. J. Pavlik, F. R. Ueland, R. W. Miller, J. M. Ubellacker, C. P. DeSimone, J. Elder, J. Hoff, L. Baldwin, R. J. Kryscio, J. R. van Nagell, Jr., Frequency and disposition of ovarian abnormalities followed with serial transvaginal ultrasonography. Obstetrics and gynecology 122, 210-217 (2013).
3. S. S. Buys, E. Partridge, M. H. Greene, P. C. Prorok, D. Reding, T. L. Riley, P. Hartge, R. M. Fagerstrom, L. R. Ragard, D. Chia, G. Izmirlian, M. Fouad, C. C. Johnson, J. K. Gohagan, P. P. Team, Ovarian cancer screening in the Prostate, Lung, Colorectal and Ovarian (PLCO) cancer screening trial: findings from the initial screen of a randomized trial. American journal of obstetrics and gynecology 193, 1630-1639 (2005).
4. S. S. Buys, E. Partridge, A. Black, C. C. Johnson, L. Lamerato, C. Isaacs, D. J. Reding, R. T. Greenlee, L. A. Yokochi, B. Kessel, E. D. Crawford, T. R. Church, G. L. Andriole, J. L. Weissfeld, M. N. Fouad, D. Chia, B. O'Brien, L. R. Ragard, J. D. Clapp, J. M. Rathmell, T. L. Riley, P. Hartge, P. F. Pinsky, C. S. Zhu, G. Izmirlian, B. S. Kramer, A. B. Miller, J. L. Xu, P. C. Prorok, J. K. Gohagan, C. D. Berg, P. P. Team, Effect of screening on ovarian cancer mortality: the Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Randomized Controlled Trial. Jama 305, 2295-2303 (2011).
5. A. Demirol, S. Guven, C. Baykal, T. Gurgan, Effect of endometrioma cystectomy on IVF outcome: a prospective randomized study. Reproductive biomedicine online 12, 639-643 (2006).
6. F. H. Loh, A. T. Tan, J. Kumar, S. C. Ng, Ovarian response after laparoscopic ovarian cystectomy for endometriotic cysts in 132 monitored cycles. Fertility and sterility 72, 316-321 (1999).
7. J. T. Christensen, J. L. Boldsen, J. G. Westergaard, Functional ovarian cysts in premenopausal and gynecologically healthy women. Contraception 66, 153-157 (2002).
8. D. Levine, D. L. Brown, R. F. Andreotti, B. Benacerraf, C. B. Benson, W. R. Brewster, B. Coleman, P. Depriest, P. M. Doubilet, S. R. Goldstein, U. M. Hamper, J. L. Hecht, M. Horrow, H. C. Hur, M. Marnach, M. D. Patel, L. D. Platt, E. Puscheck, R. Smith-Bindman, Management of asymptomatic ovarian and other adnexal cysts imaged at US: Society of Radiologists in Ultrasound Consensus Conference Statement. Radiology 256, 943-954 (2010).
9. E. J. Cheng, R. J. Kurman, M. Wang, R. Oldt, B. G. Wang, D. M. Berman, M. Shih Ie, Molecular genetic analysis of ovarian serous cystadenomas. Laboratory investigation; a journal of technical methods and pathology 84, 778-784 (2004).
10. D. Duke, J. Colville, A. Keeling, D. Broe, T. Fotheringham, M. J. Lee, Transvaginal aspiration of ovarian cysts: long-term follow-up. Cardiovascular and interventional radiology 29, 401-405 (2006).
11. R. J. Kurman, M. Shih Ie, The origin and pathogenesis of epithelial ovarian cancer: a proposed unifying theory. The American journal of surgical pathology 34, 433-443 (2010).
12. M. L. C. R. J. Kurman, C. S. Herrington, R. H. Young WHO Classification of Tumours of Female Reproductive Organs. (International Agency for Research on Cancer Lyon, France., ed. 4th, 2014).
13. Y. Lee, A. Miron, R. Drapkin, M. R. Nucci, F. Medeiros, A. Saleemuddin, J. Garber, C. Birch, H. Mou, R. W. Gordon, D. W. Cramer, F. D. McKeon, C. P. Crum, A candidate precursor to serous carcinoma that originates in the distal fallopian tube. The Journal of pathology 211, 26-35 (2007).
14. N. Cancer Genome Atlas Research, Integrated genomic analyses of ovarian carcinoma. Nature 474, 609-615 (2011).
15. K. M. Schmeler, C. C. Sun, D. C. Bodurka, M. T. Deavers, A. Malpica, R. L. Coleman, P. T. Ramirez, D. M. Gershenson, Neoadjuvant chemotherapy for low-grade serous carcinoma of the ovary or peritoneum. Gynecologic oncology 108, 510-514 (2008).
16. R. E. Bristow, R. S. Tomacruz, D. K. Armstrong, E. L. Trimble, F. J. Montz, Survival effect of maximal cytoreductive surgery for advanced ovarian carcinoma during the platinum era: a meta-analysis. Journal of clinical oncology: official journal of the American Society of Clinical Oncology 20, 1248-1259 (2002).
17. M. E. Sherman, P. J. Mink, R. Curtis, T. R. Cote, S. Brooks, P. Hartge, S. Devesa, Survival among women with borderline ovarian tumors and ovarian carcinoma: a population-based analysis. Cancer 100, 1045-1052 (2004).
18. K. K. Shih, Q. Zhou, J. Huh, J. C. Morgan, A. lasonos, C. Aghajanian, D. S. Chi, R. R. Barakat, N. R. Abu-Rustum, Risk factors for recurrence of ovarian borderline tumors. Gynecologic oncology 120, 480-484 (2011).
19. D. Mayr, A. Hirschmann, U. Lohrs, J. Diebold, KRAS and BRAF mutations in ovarian tumors: a comprehensive study of invasive carcinomas, borderline tumors and extraovarian implants. Gynecologic oncology 103, 883-887 (2006).
20. S. Jones, T. L. Wang, R. J. Kurman, K. Nakayama, V. E. Velculescu, B. Vogelstein, K. W. Kinzler, N. Papadopoulos, M. Shih Ie, Low-grade serous carcinomas of the ovary contain very few point mutations. The Journal of pathology 226, 413-420 (2012).
21. R. N. Grisham, G. Iyer, K. Garg, D. DeLair, D. M. Hyman, Q. Zhou, A. lasonos, M. F. Berger, F. Dao, D. R. Spriggs, D. A. Levine, C. Aghajanian, D. B. Solit, BRAF mutation is associated with early stage disease and improved outcome in patients with low-grade serous ovarian cancer. Cancer 119, 548-554 (2013).
22. Y. T. Tsang, M. T. Deavers, C. C. Sun, S. Y. Kwan, E. Kuo, A. Malpica, S. C. Mok, D. M. Gershenson, K. K. Wong, KRAS (but not BRAF) mutations in ovarian serous borderline tumour are associated with recurrent low-grade serous carcinoma. The Journal of pathology 231, 449-456 (2013).
23. J. L. Frossard, P. Amouyal, G. Amouyal, L. Palazzo, J. Amaris, M. Soldan, E. Giostra, L. Spahr, A. Hadengue, M. Fabre, Performance of endosonography-guided fine needle aspiration and biopsy in the diagnosis of pancreatic cystic lesions. The American journal of gastroenterology 98, 1516-1524 (2003).
24. J. D. Lin, T. C. Chao, B. Y. Huang, S. T. Chen, H. Y. Chang, C. Hsueh, Thyroid cancer in the thyroid nodules evaluated by ultrasonography and fine-needle aspiration cytology. Thyroid: official journal of the American Thyroid Association 15, 708-717 (2005).
25. A. Volpe, J. R. Kachura, W. R. Geddie, A. J. Evans, A. Gharajeh, A. Saravanan, M. A. Jewett, Techniques, safety and accuracy of sampling of renal tumors by fine needle aspiration and core biopsy. The Journal of urology 178, 379-386 (2007).
26. P. Martinez-Onsurbe, A. Ruiz Villaespesa, J. M. Sanz Anquela, P. L. Valenzuela Ruiz, Aspiration cytology of 147 adnexal cysts with histologic correlation. Acta cytologica 45, 941-947 (2001).
27. O. Moran, J. Menczer, G. Ben-Baruch, S. Lipitz, E. Goor, Cytologic examination of ovarian cyst fluid for the distinction between benign and malignant tumors. Obstetrics and gynecology 82, 444-446 (1993).
28. J. Wu, H. Matthaei, A. Maitra, M. Dal Molin, L. D. Wood, J. R. Eshleman, M. Goggins, M. I. Canto, R. D. Schulick, B. H. Edil, C. L. Wolfgang, A. P. Klein, L. A. Diaz, Jr., P. J. Allen, C. M. Schmidt, K. W. Kinzler, N. Papadopoulos, R. H. Hruban, B. Vogelstein, Recurrent GNAS mutations define an unexpected pathway for pancreatic cyst development. Science translational medicine 3, 92ra66 (2011).
29. B. Vogelstein, N. Papadopoulos, V. E. Velculescu, S. Zhou, L. A. Diaz, Jr., K. W. Kinzler, Cancer genome landscapes. Science 339, 1546-1558 (2013).
30. J. Wu, Y. Jiao, M. Dal Molin, A. Maitra, R. F. de Wilde, L. D. Wood, J. R. Eshleman, M. G. Goggins, C. L. Wolfgang, M. I. Canto, R. D. Schulick, B. H. Edil, M. A. Choti, V. Adsay, D. S. Klimstra, G. J. Offerhaus, A. P. Klein, L. Kopelovich, H. Carter, R. Karchin, P. J. Allen, C. M. Schmidt, Y. Naito, L. A. Diaz, Jr., K. W. Kinzler, N. Papadopoulos, R. H. Hruban, B. Vogelstein, Whole-exome sequencing of neoplastic cysts of the pancreas reveals recurrent mutations in components of ubiquitin-dependent pathways. Proceedings of the National Academy of Sciences of the United States of America 108, 21188-21193 (2011).
31. R. Yamada, N. Maeda, H. Oguri, Y. Adachi, T. Takeuchi, M. Furihata, T. Fukaya, Is it possible to diagnose malignancy from fluid in cystic ovarian tumors? European journal of obstetrics, gynecology, and reproductive biology 171, 96-100 (2013).
32. A. M. Newman, S. V. Bratman, J. To, J. F. Wynne, N. C. Eclov, L. A. Modlin, C. L. Liu, J. W. Neal, H. A. Wakelee, R. E. Merritt, J. B. Shrager, B. W. Loo, Jr., A. A. Alizadeh, M. Diehn, An ultrasensitive method for quantitating circulating tumor DNA with broad patient coverage. Nature medicine 20, 548-554 (2014).
33. M. Murtaza, S. J. Dawson, D. W. Tsui, D. Gale, T. Forshew, A. M. Piskorz, C. Parkinson, S. F. Chin, Z. Kingsbury, A. S. Wong, F. Marass, S. Humphray, J. Hadfield, D. Bentley, T. M. Chin, J. D. Brenton, C. Caldas, N. Rosenfeld, Non-invasive analysis of acquired resistance to cancer therapy by sequencing of plasma DNA. Nature 497, 108-112 (2013).
34. I. Kinde, J. Wu, N. Papadopoulos, K. W. Kinzler, B. Vogelstein, Detection and quantification of rare mutations with massively parallel sequencing. Proceedings of the National Academy of Sciences of the United States of America 108, 9530-9535 (2011).
35. S. Barollo, R. Pezzani, A. Cristiani, M. Redaelli, L. Zambonin, B. Rubin, L. Bertazza, M. Zane, C. Mucignat-Caretta, A. Bulfone, G. Pennelli, E. Casal Ide, M. R. Pelizzo, F. Mantero, S. Moro, C. Mian, Prevalence, tumorigenic role, and biochemical implications of rare BRAF alterations. Thyroid: official journal of the American Thyroid Association 24, 809-819 (2014).
36. F. Cruz, 3rd, B. P. Rubin, D. Wilson, A. Town, A. Schroeder, A. Haley, T. Bainbridge, M. C. Heinrich, C. L. Corless, Absence of BRAF and NRAS mutations in uveal melanoma. Cancer research 63, 5761-5766 (2003).
37. H. Davies, G. R. Bignell, C. Cox, P. Stephens, S. Edkins, S. Clegg, J. Teague, H. Woffendin, M. J. Garnett, W. Bottomley, N. Davis, E. Dicks, R. Ewing, Y. Floyd, K. Gray, S. Hall, R. Hawes, J. Hughes, V. Kosmidou, A. Menzies, C. Mould, A. Parker, C. Stevens, S. Watt, S. Hooper, R. Wilson, H. Jayatilake, B. A. Gusterson, C. Cooper, J. Shipley, D. Hargrave, K. Pritchard-Jones, N. Maitland, G. Chenevix-Trench, G. J. Riggins, D. D. Bigner, G. Palmieri, A. Cossu, A. Flanagan, A. Nicholson, J. W. Ho, S. Y. Leung, S. T. Yuen, B. L. Weber, H. F. Seigler, T. L. Darrow, H. Paterson, R. Marais, C. J. Marshall, R. Wooster, M. R. Stratton, P. A. Futreal, Mutations of the BRAF gene in human cancer. Nature 417, 949-954 (2002).
38. R. C. Bast, Jr., T. L. Klug, E. St John, E. Jenison, J. M. Niloff, H. Lazarus, R. S. Berkowitz, T. Leavitt, C. T. Griffiths, L. Parker, V. R. Zurawski, Jr., R. C. Knapp, A radioimmunoassay using a monoclonal antibody to monitor the course of epithelial ovarian cancer. The New England journal of medicine 309, 883-887 (1983).
39. I. Hellstrom, J. Raycraft, M. Hayden-Ledbetter, J. A. Ledbetter, M. Schummer, M. McIntosh, C. Drescher, N. Urban, K. E. Hellstrom, The HE4 (WFDC2) protein is a biomarker for ovarian carcinoma. Cancer research 63, 3695-3700 (2003).
40. H. S. Kim, J. H. Ahn, H. H. Chung, J. W. Kim, N. H. Park, Y. S. Song, H. P. Lee, Y. B. Kim, Impact of intraoperative rupture of the ovarian capsule on prognosis in patients with early-stage epithelial ovarian cancer: a meta-analysis. European journal of surgical oncology: the journal of the European Society of Surgical Oncology and the British Association of Surgical Oncology 39, 279-289 (2013).
41. W. S. Hilger, J. F. Magrina, P. M. Magtibay, Laparoscopic management of the adnexal mass. Clinical obstetrics and gynecology 49, 535-548 (2006).
42. L. J. Havrilesky, B. L. Peterson, D. K. Dryden, J. T. Soper, D. L. Clarke-Pearson, A. Berchuck, Predictors of clinical outcomes in the laparoscopic management of adnexal masses. Obstetrics and gynecology 102, 243-251 (2003).
43. K. J. Chang, P. Nguyen, R. A. Erickson, T. E. Durbin, K. D. Katz, The clinical utility of endoscopic ultrasound-guided fine-needle aspiration in the diagnosis and staging of pancreatic carcinoma. Gastrointestinal endoscopy 45, 387-393 (1997).
44. M. A. Eloubeidi, V. K. Chen, I. A. Eltoum, D. Jhala, D. C. Chhieng, N. Jhala, S. M. Vickers, C. M. Wilcox, Endoscopic ultrasound-guided fine needle aspiration biopsy of patients with suspected pancreatic cancer: diagnostic accuracy and acute and 30-day complications. The American journal of gastroenterology 98, 2663-2668 (2003).
45. S. Ngamruengphong, K. M. Swanson, N. D. Shah, M. B. Wallace, Preoperative endoscopic ultrasound-guided fine needle aspiration does not impair survival of patients with resected pancreatic cancer. Gut, (2015).
46. A. A. Tran, S. B. Brown, J. Rosenberg, D. M. Hovsepian, Tract embolization with gelatin sponge slurry for prevention of pneumothorax after percutaneous computed tomography-guided lung biopsy. Cardiovascular and interventional radiology 37, 1546-1553 (2014).
47. W. K. Tsang, W. H. Luk, A. Lo, Ultrasound-guided plugged percutaneous biopsy of solid organs in patients with bleeding tendencies. Hong Kong medical journal=Xianggang yi xue za zhi/Hong Kong Academy of Medicine 20, 107-112 (2014).
48. J. R. van Nagell, Jr., P. D. DePriest, F. R. Ueland, C. P. DeSimone, A. L. Cooper, J. M. McDonald, E. J. Pavlik, R. J. Kryscio, Ovarian cancer screening with annual transvaginal sonography: findings of 25,000 women screened. Cancer 109, 1887-1896 (2007).
49. F. R. Ueland, C. P. Desimone, L. G. Seamon, R. A. Miller, S. Goodrich, I. Podzielinski, L. Sokoll, A. Smith, J. R. van Nagell, Jr., Z. Zhang, Effectiveness of a multivariate index assay in the preoperative assessment of ovarian tumors. Obstetrics and gynecology 117, 1289-1297 (2011).
50. E. Bandiera, L. Zanotti, A. S. Fabricio, E. Bucca, E. Squarcina, C. Romani, R. Tassi, E. Bignotti, P. Todeschini, G. Tognon, C. Romagnolo, M. Gion, E. Sartori, T. Maggino, S. Pecorelli, A. Ravaggi, Cancer antigen 125, human epididymis 4, kallikrein 6, osteopontin and soluble mesothelin-related peptide immunocomplexed with immunoglobulin M in epithelial ovarian cancer diagnosis. Clinical chemistry and laboratory medicine: CCLM/FESCC 51, 1815-1824 (2013).
51. B. Kristjansdottir, K. Levan, K. Partheen, E. Carlsohn, K. Sundfeldt, Potential tumor biomarkers identified in ovarian cyst fluid by quantitative proteomic analysis, iTRAQ. Clinical proteomics 10, 4 (2013).
52. C. Rago, D. L. Huso, F. Diehl, B. Karim, G. Liu, N. Papadopoulos, Y. Samuels, V. E. Velculescu, B. Vogelstein, K. W. Kinzler, L. A. Diaz, Jr., Serial assessment of human tumor burdens in mice by the analysis of circulating DNA. Cancer research 67, 9364-9370 (2007).

Claims

The invention claimed is:

1. A method, comprising:

testing ovarian cyst fluid for mutations in a panel of genes mutated in an ovarian neoplasm, wherein the panel comprises BRAF, KRAS, and TP53, wherein the step of testing does not employ a whole-genome or whole-exome technique, wherein the step of testing employs a step of adding a unique identifier (UID) to each template DNA in the ovarian cyst fluid, and wherein the method has a sensitivity level of at least 70%.

2. The method of claim 1 wherein the panel further comprises one or more genes selected from a first group consisting of CTNNB1, PIK3CA, PTEN, ARID1A, and PPP2R1A.

3. The method of claim 1 wherein the step of testing employs gene-specific reagents.

4. The method of claim 1 wherein the step of testing employs mutation-specific reagents.

5. The method of claim 1 wherein the testing is performed on the ovarian cyst fluid and on a sample selected from the group consisting of cyst wall and normal, non-ovarian tissue.

6. The method of claim 1 wherein BRAF600, KRAS12, KRAS13, KRAS61, or combinations thereof are tested.

7. The method of claim 2 wherein all genes of the first group are in the panel.

8. The method of claim 1 further comprising the step of assaying amount of DNA in the cyst fluid.

9. The method of claim 1 further comprising the step of assaying amount of CA-125 levels in plasma.

10. The method of claim 1 wherein the panel further comprises one or more genes selected from a second group consisting of AKT1, APC, BRCA1, BRCA2, CDKN2A, EGFR, FBXW7, FGFR2, MAPK1, NRAS, PIK3R1, and POLE.

11. The method of claim 7 wherein the panel further comprises one or more genes selected from a second group consisting of AKT1, APC, BRCA1, BRCA2, CDKN2A, EGFR, FBXW7, FGFR2, MAPK1, NRAS, PIK3R1, and POLE.

12. The method of claim 1 wherein the ovarian cyst fluid is obtained by needle aspiration of an ovarian cyst.

13. The method of claim 1 wherein the ovarian cyst fluid is obtained prior to any surgical removal of the ovarian cyst.

14. The method of claim 1 wherein the ovarian cyst fluid is obtained after surgical removal of the ovarian cyst and recurrence of the ovarian cyst.

15. The method of claim 1 wherein the ovarian cyst fluid is from a cyst selected from the group consisting of: mesothelial cyst, follicular cyst, corpus luteal cyst, mucinous cystadenoma, endometriotic cyst, serous cystadenoma, serous cystadenofibroma, atypical proliferative serous tumor, atypical proliferative endometrioid tumor, serous carcinoma, mixed epithelial tumor, endometrioid carcinoma, clear cell carcinoma, metastatic tumors to the ovary, and mucinous carcinoma.

16. The method of claim 1 wherein a copy number variation, a loss of heterozygosity, or both, is determined in at least one of the genes in the panel.

17. The method of claim 1 wherein a point mutation, a rearrangement, a frameshift, or combinations thereof, is determined in at least one gene of the panel.

18. The method of claim 1, wherein the method has a sensitivity level of at least 85%.

19. The method of claim 1, wherein the method has a specificity level of at least 90%.

20. The method of claim 1, wherein the method has a specificity level of at least 95%.

21. The method of claim 1, wherein the ovarian neoplasm needs surgery.

22. The method of claim 1, wherein the ovarian neoplasm is a borderline tumor or malignant cancer.

23. The method of claim 1, wherein the ovarian neoplasm is a type I cancer or a type II cancer.