Douglas A. Levine
All cancer is genetic, meaning that all cancers have a genetic basis and result from an accumulation of mutations or other genetic defects. Cancer can be caused by many different factors, but all cancers function through genetic mutation or other alterations. Most cancers occur when the normal functioning of a single cell in a tissue of origin goes awry. The old tenet of cancer being an imbalance of growth and death still holds true, but over the years, we have learned much about the causes and details of how this growth/death dichotomy becomes muddled. For some, cancer will have an inherited origin passed down through generations; for others it will be a newly developed, or de novo, mutation obtained by the tissue of origin to turn normal tissue into cancer. Many of the cancer-causing genes can be grouped into categories of tumor suppressors and oncogenes. There are a number of cancer syndromes that predispose to the development of gynecologic malignancies. In some cases, environmental factors may increase the risk of certain cancer types.
PRINCIPLES OF CANCER MOLECULAR GENETICS
Oncogenes
Oncogenes are cancer drivers that have the ability to initiate tumor formation when turned on, most commonly by mutations. Before a gene becomes an oncogene or develops the ability to transform normal cells to malignant cells, it is referred to as a proto-oncogene, or a gene with oncogenic potential. In addition to mutation, proto-oncogenes can transform into oncogenes through structural rearrangements such as translocations, duplications, or splice variants, as well as overexpression of the gene product. Genes can function as oncogenes through increasing protein activity or by losing the ability to suppress negative regulators of growth. The first oncogene, src, was discovered in chickens. RAS and MYC were other early oncogenes found to regulate transcription and affect cell proliferation. Since then, many other oncogenes, which are often activated by somatic mutations, have been discovered. An example of a more recently discovered oncogene is PIK3CA, which is activated by cell surface receptor tyrosine kinases and regulates AKT activation, cell growth, and survival (Figure 2-1). Through sequence analysis of various human tumors, PIK3CA mutations in multiple human tumors were identified.1 Remarkably, most of these mutations were clustered at a limited number of nucleotide positions, termed hotspots, making them useful for cancer diagnostics and therapeutics. Mutations in related genes such as PIK3R1 and PIK3R2, which encode the regulatory and structural subunits of the PI3K protein, have also been identified. Currently, there are many drugs designed to target various subunits of PI3K that have the potential for effectiveness in tumors with activating PI3K mutations and others. Other regulatory genes, such as microRNAs, can function as oncogenes by promoting cancer development and growth. MicroRNAs usually negatively regulate gene expression, but if they release their normal negative inhibition, unsuppressed growth can result in oncogenic activity. Thus a gene can have direct or indirect oncogenic potential (Figure 2-2).
FIGURE 2-1. Schematic diagram of PI3K signaling. PTEN is a tumor suppressor gene inhibiting the pathway. Classically, both copies of PTEN need to be lost to release this inhibition and allow for uncontrolled AKT activity, as is seen in many solid tumors. PTEN can be lost through a combination of mutation, methylation, and deletion. PIK3CA encodes a catalytic subunit of PI3K. A single activating mutation in 1 copy of PIK3CA is necessary to activate AKT and result in uncontrolled growth, as it is an oncogene. (Reprinted [modified] by permission from Macmillan Publishers Ltd: Oncogene [Yuan TL, Cantley LC. PI3K pathway alterations in cancer: variations on a theme. Oncogene. 2008;27:5497-5510], copyright 2008.)
FIGURE 2-2. MicroRNAs can function as oncogenes or tumor suppressor genes. Let-7 inhibits the RAS oncogene and therefore is functioning as a TSG. When lost, it would release its inhibition, and the RAS oncogene would be active, as happens in many malignancies. The miR17-92 cluster of microRNAs functions as an oncogene in that it inhibits the TSGs PTEN. PTEN normally inhibits AKT activity, but when inhibited itself, the PI3K/AKT pathway is activated/no longer inhibited, and cancer can develop or progress. (Reprinted from Hammond SM. MicroRNAs as oncogenes. Curr Opin Genet Dev.2006;16:4-9, with permission from Elsevier.)
Tumor Suppressor Genes
Tumor suppressor genes (TSGs) are genes that have normal inhibitory function that, when lost, permit cell transformation or tumor growth. Unlike oncogenes, which promote growth when activated, TSGs require complete inactivation in order to fully release inhibitory activity. Classically, this has been referred to as the “two-hit hypothesis,” in which both copies of a TSG need to be lost in order to release inhibitory function sufficiently to promote cancer development and growth. The loss of both copies, or alleles, of a TSG can occur by 2 mutations developing over time, such that the first mutation increases the likelihood of developing a second mutation.
Alternatively, first copy of a TSG can be inactivated by mutation, and the second copy can be inactivated by a separate mechanism such as methylation, structural rearrangement, or loss of heterozygosity. BRCA1 is a good example of a TSG in that both alleles are lost in ovarian tumors. Interestingly, this has not been uniformly shown to be the case for breast cancer.2 Commonly mutated TSGs include PTEN, which suppresses activation of the PI3K/AKT pathway; RB1, the retinoblastoma gene; APC in colon cancer; and the most commonly mutated TSG, TP53, which is mutated in nearly all serous ovarian cancers and approximately half of most other epithelial malignancies. When 1 copy of a TSG is lost, such as when a BRCA1 mutation is inherited from a parent, this is termed haploinsufficiency. The functional impact of haploinsufficiency is unclear and likely varies between tissue types and biologic circumstances, but in some cases it may predispose to loss of the second allele.
Mismatch Repair Proteins
Mismatch repair (MMR) proteins help to correct normal errors in DNA replication. Every time a cell divides, its DNA undergoes replication, which is not a perfect process. In fact, when DNA is replicating, errors occur 1 in every 106 to 108 nucleotides. Considering the size of the human genome, many errors occur. Thus mechanisms to repair DNA replication errors must be robust. Most errors in replication are repaired during the replication process itself through a mechanism called proofreading. However, some errors persist after replication and are fixed through MMR. MMR is one of several DNA repair mechanisms. In general, DNA repair mechanisms can be grouped into 2 classes: those that repair single-strand DNA breaks and those that repair double-strand DNA breaks.
Mechanisms that repair single-strand DNA damage include base excision repair, nucleotide excision repair, and MMR. MMR is the only mechanism of single-strand repair that repairs undamaged, misaligned DNA primarily due to errors of replication. The MMR proteins that are most commonly mutated in gynecologic cancer syndromes are MLH1 and MSH2. MMR mutation leads to greater errors of DNA replication in repetitive regions of the genome within tumors. This phenomenon is referred to as microsatellite instability (MI), in which short repeat regions undergo expansion or contraction in the number of repetitive elements. It is particularly common in endometrial and colorectal cancers. MI can be detected by sequencing DNA from normal tissues (such as blood) and comparing it with malignant tissues. When the malignant tissue has greater or fewer repetitive elements than the normal tissue, this is referred to as MI. In addition to MI developing as a consequence of MMR mutations, which can often be inherited, MI can also develop from somatic changes, such as methylation of the MLH1 promoter.
Inheritance Patterns
Inheritance refers to traits or genes obtained from parents and can be referred to as hereditary or inherited. Traditional inheritance patterns are autosomal dominant, autosomal recessive, and sex- (or X-) linked. Dominant inheritance occurs when the trait or gene is functional or active by inheriting a single copy from either parent. Dominantly inherited traits have a 50% chance of being passed down to each offspring. BRCA1/2 mutations are a good example of a dominantly inherited trait in that each offspring has a 50% chance of inheriting the mutation from a single parent who carries the mutation. Recessive patterns of inheritance are less common and require inheritance of a trait or gene from each parent, as 2 affected alleles are required for disease manifestation. X-linked traits can be inherited from either parent, but recessive traits are manifest in males, who only carry one X chromosome, whereas dominant traits would be manifest equally in males and females. Recessive X-linked traits require inheritance from each parent for manifestation. All inherited traits or mutations do not manifest themselves. The genetic composition of mutations or other variations are referred to as genotypes, and the manifestation of these mutations or traits are referred to as phenotypes. All mutations (or mutated genotypes) do not result in a disease phenotype; this phenomenon is referred to as incomplete penetrance, in which the diseased gene is only manifest in a subset of people. For most biologic situations, we do not understand the factors (or modifiers) that affect penetrance. Consider a BRCA1 mutation that confers a lifetime risk of 40% for ovarian cancer, meaning that not everyone with a BRCA1 inherited mutation develops ovarian cancer. The penetrance of a BRCA1 mutation for ovarian cancer is approximately 40%. The patterns of inheritance refer to the germline, or every cell in your body. When you inherit a mutation from a parent, it is present in every cell of your body. A disease-causing mutation may not cause disease in every cell of your body, but it is present in every cell of your body. This is in contrast to somatic mutations, which are not inherited but develop within the host and are commonly found in cancer. Many germline mutations will predispose to somatic mutations, which can function as the “second hit” to inactivate a TSG and initiate tumor formation. This distinction between germline events or mutations and somatic mutations is key to understanding what tumor-causing mechanisms are inherited and which develop de novo.
BIOLOGY OF GYNECOLOGIC CANCER
Endometrial Cancer
Endometrial cancers are broadly classified into 2 groups. Type 1 tumors are of endometrioid histology and are directly related to estrogen excess. Most commonly, the excess estrogen is derived from adipocytes and thus is more frequent in obese women. Many years ago, the most common source of exogenous estrogen was unopposed ingested estrogen, before combination hormone replacement therapy was prescribed. The initial report from the New England Journal of Medicine in 1975 identified a 4.5-fold increased risk of endometrial cancer in users of unopposed estrogens.3 Once the link between unopposed estrogen and endometrial cancer was established, hormone replacement therapy began including progestins for women who had their uterus in place. Other causes of unopposed estrogen include polycystic ovarian syndrome, in which anovulation leads to metabolic and hormonal derangements, which increase the risk of endometrial cancer in premenopausal women.4
Type 2 tumors are of more aggressive histologic subtypes, with the prototypic tumor being of serous histology. The etiology of type 2 tumors is less understood, but they have a higher propensity for recurrence and likely benefit from adjuvant treatment after surgical resection. These tumors are not associated with obesity and have no specific epidemiologic predilection, although they appear more common in nonobese women. Other aggressive type 2 histology subtypes include carcinosarcoma and clear cell (Table 2-1).
Table 2-1 Comparison Between Type 1 and Type 2 Endometrial Cancers
Pathology of Type 1 and Type 2 Endometrial Carcinoma
Type 1 endometrioid tumors follow a natural progression from normal endometrium to endometrial hyper-plasia (with or without atypia) followed by invasive cancer. A premalignant lesion within regions of endometrial hyperplasia has been recently identified and termed endometrial intraepithelial neoplasia (EIN). EIN is associated with a far greater increase in endometrial cancer risk than simple endometrial hyperplasia. The risk of progression from a precursor lesion to endometrial carcinoma increases with greater complexity and nuclear atypia. From the most recent comprehensive review, the risk of progression to invasive endometrial cancer was < 1% for simple hyperplasia without atypia, 7% to 9% for either simple hyperplasia with atypia or complex hyperplasia without atypia, and 20% for complex hyperplasia with atypia. These risks are all approximately doubled in the presence of EIN.5
Type 2 serous tumors do not have a logical and stepwise progression from a well-defined precursor lesion to invasive carcinoma. Often, uterine serous carcinoma can be confined to a small endometrial polyp, yet present with metastatic disease, at the time of comprehensive surgical staging. A precursor lesion has been identified and termed endometrial intraepithelial carcinoma (EIC). EIC has been specifically associated with uterine serous tumors and uterine carcinosarcomas that contain a serous epithelial component and can be found adjacent to most established uterine serous carcinomas.6 Epidemiologic studies suggest that nearly 50% of apparently uterine-confined disease have spread beyond the uterus at the time of diagnosis.7 Unlike type 1 tumors, which typically develop in a background of hyperplasia, type 2 tumors, particularly serous tumors, develop in a background of relative endometrial atrophy, with uterine polyps or other focal changes occurring at the site of tumorigenesis. These tumors also lack the presence of hormone receptors commonly seen in type 1 tumors.
Endometrial carcinosarcoma is an aggressive type 2 tumor that was initially thought to be 2 tumor types that collided at a well-defined interface. Subsequent molecular studies clearly proved that these tumors arise through clonal evolution where a given malignant epithelial cell undergoes differentiation into 2 histologically distinct lineages. Thus these tumors have traditionally been considered as a subset of uterine sarcomas, but in fact, they represent poorly differentiated and divergent epithelial tumors with sarcomatous differentiation.8 If any part of the sarcomatous component contains cell types that are not native to the uterine corpus, this is referred to as a heterologous differentiation. If all cell types are commonly seen within the uterus, this is referred to as a homologous differentiation. In several well-designed studies, it has been reported that patients with endometrial carcinosarcomas having heterologous differentiation have a worse outcome as compared with patients having homologous differentiation for both early- and late-stage disease.9
Genetics of Type 1 and Type 2 Endometrial Carcinoma
In addition to the well-defined histologic progression of type 1 endometrioid tumors, the genetics of type 1 tumors are also fairly well understood. It is thought that PTEN, a tumor suppressor gene that negatively regulates the PI3K/AKT pathway, is lost early in the neoplastic process. In fact, PTEN mutations have been found in the preneoplastic EIN lesions. PTEN loss can occur through a number of mechanisms, including somatic mutation, promoter methylation, or deletion. In the uncommon Cowden syndrome, which includes development of hamartomas and increased risks of endometrial, breast, renal, thyroid, and possibly colorectal malignancies, PTEN mutation is present in the germline (and thus is inherited). Subsequent to PTEN loss, additional mutations are accumulated that also frequently occur in the PI3K/AKT pathway. PIK3CA encodes the catalytic subunit of the PI3K enzyme and contains activating mutations in approximately 25% of endometrioid and serous tumors. Interestingly, mutations in this gene are often identified at the same genetic position, or at the same nucleotide, and are therefore termed hotspot mutations. These types of hotspot mutations in PIK3CA and other genes can be easily detected through modern laboratory assays such as mass spectrometry and are therefore good targets for both diagnostics and therapeutics. In fact, there are many PIK3CA inhibitors being developed and in early-phase clinical trials. The regulatory and structural units of the PI3K enzyme are encoded by the genes PIK3R1 and PIK3R2, both of which are also frequently mutated in type 1 cancers.
Mutations in FGFR2 were first reported in 2007 and are present in 10% to 15% of type 1 tumors.10 This finding has clinical relevance, as there are small-molecule inhibitors of FGFR2 in addition to other targets. High-grade type 1 endometrial tumors, like other high-grade solid tumors, have fairly frequent mutations in TP53, in the order of 50%. Other commonly mutated genes include CTNNB1, KRAS, and occasionally AKT1. Type 1 tumors also have a 20% to 25% frequency of MI, which can be easily detected through a consensus panel of 5 microsatellite markers and has been adopted as a reference set by the National Cancer Institute.11 The clinical significance of MI in endometrial cancer is unclear, with studies demonstrating both improved and worse outcomes in the setting of MI. For colorectal cancer, it is well established that MI tumors have an improved clinical course.12
The lack of well-defined progression from normal histology to cancer for type 2 tumors is also reflected in the genetics of these lesions. EIC is thought to be the precursor lesion, which overexpresses and contains mutations in TP53. TP53 mutation is also found in 80% to 90% of uterine serous carcinomas, in contrast to the low frequency in low-grade endometrioid type 1 tumors and modestly higher frequency in high-grade type 1 tumors. Type 2 tumors are also characterized by the lack of MI and a very low frequency of KRAS, PTEN, and CTNNB1 mutations. There is a similar frequency of mutations in PIK3CA, offering the possibility that PI3K-targeted therapy may be equally successful in type I and type 2 tumors. TP53 mutations are also commonly seen in uterine carcinosarcomas but occur at low frequency in clear cell tumors.13
Lynch Syndrome
First described by Henry T. Lynch, this syndrome commonly includes malignancies of the uterus, colon, stomach, and ovary.14 Less frequent malignancies that are part of this syndrome include small bowel, upper urinary tract, brain, and biliary tract. Lynch syndrome is primarily caused by an inherited germline mutation in one of several MMR genes: MLH1, MSH2, and less often MSH6, PMS1, and PMS2. The syndrome was initially divided into 2 subtypes—Lynch I and Lynch II—based on whether or not extracolonic tumors were included in a given pedigree. It was quickly realized that the genetic basis of these 2 subtypes was similar, and the syndrome was commonly referred to as hereditary nonpolyposis colorectal cancer syndrome (HNPCC). This was due mostly to the fact that the predominant malignancy was colon cancer. However, this name soon became a misnomer when the high frequency of endometrial cancer was identified in women. Furthermore, the colons within which the colorectal tumors developed as part of HNPCC did in fact contain multiple colonic polyps in many cases; however, the extent of the polyposis was far less than that seen in the related familial adenomatous polyposis (FAP) syndrome, in which a great many polyps are found throughout the colon due to a germline mutation in APC. One main difference in the risk of colorectal cancer between these 2 syndromes is that the risk of colorectal cancer with FAP is nearly 100%, but the risk of colorectal cancer in HNPCC is approximately 80% and occurs at a later onset than with FAP. For these reasons, HNPCC is no longer an accurate description of the clinical and genetic syndrome, and Lynch syndrome (without any further distinction between type I and type II) is now the preferred terminology.
Pathogenesis and Pathology
Lynch syndrome is due to inherited mutations in 1 of 5 MMR genes. MLH1 and MSH2 are the most commonly mutated genes in this syndrome. These defects lead to faulty DNA repair and increased risk of malignancy. In women, endometrial cancer and colon cancer have an equal likelihood of being the sentinel malignancy in this syndrome. In women with Lynch syndrome, the lifetime risk of colon cancer and endometrial cancer are both approximately 50%, in contrast to the risk of colon cancer in men, which is 50% greater.15,16 The mean age of endometrial cancer diagnosis in Lynch syndrome patients is approximately 45 years, as compared with a mean age of 62 years for endometrial cancer in general. In women under 50 years of age with endometrial cancer, the likelihood of having Lynch syndrome as determined by a germline mutation in one of the MMR genes is 9%.17 Young patients with Lynch syndrome tend to have a lower body mass index than young patients with sporadic endometrial cancer. Immunohistochemical markers are now robustly available for testing for the absence (abnormal) of expression in MLH1, MSH2, MSH6, and PMS2.
The absence of MMR protein expression is indicative of MI, but a fair portion of these cases may be due to promoter methylation of MLH1 and not an inherited germline mutation, which is the only circumstance diagnostic of Lynch syndrome. If immunohisto-chemistry (IHC) only is applied to endometrial cancer patients younger than 50 years of age, 25% to 35% of tumors have been found to lack normal IHC staining, which increases when considering only nonobese women. Nonetheless, approximately one-third to one-half of these cases may be due to MI without a true germline mutation secondary to epigenetic/methylation silencing.18,19 The lifetime risk of Lynch syndrome–associated ovarian cancer ranges from 8% to 12%. Ovarian cancers appear to be moderate to high grade, mostly epithelial in nature, and, unlike sporadic ovarian cancer, predominately stage I or II. Endometrial cancer is diagnosed synchronously in approximately 20% of Lynch syndrome patients with ovarian cancer. The mean age of ovarian cancer diagnosis in Lynch syndrome patients is approximately 43 years, as compared with a mean age of 63 years for ovarian cancer in general.20 Women with synchronous ovarian and endometrial cancer have a 7% risk of having Lynch syndrome21 (Figure 2-3).
FIGURE 2-3. Lifetime risk for colon, endometrial, and ovarian cancer in individuals with Lynch syndrome compared with the general population. (Reproduced, with permission, from Schmeler KM, Lu KH. Gynecologic cancers associated with Lynch syndrome/HNPCC. Clin Transl Oncol. 2008;10:313-317.)
Endometrial tumors associated with Lynch syndrome are more commonly found in the lower uterine segment and are more often poorly differentiated and deeply invasive, with a higher mitotic rate and more tumor-infiltrating lymphocytes (TILs).22 However, a comprehensive review comparing Lynch-associated endometrial cancer with sporadic endometrial cancer and MLH1 promoter methylated tumors found a similar frequency of non-endometrioid histologies between groups, but the MLH1 methylated tumors were more likely to be undifferentiated.23 This review also found similar frequencies of myometrial invasion, high stage, and TILs. Many centers have adopted pathologic screening approaches to identify patients who may have Lynch syndrome and should be referred for genetic counseling. These varied algorithms typically start with standard IHC directed toward the 4 common MMR proteins: MLH1, MSH2, MSH6, and PMS2 in women under the age of 50 years with endometrial cancer, those with a family history suggestive of Lynch syndrome, or in patients with characteristic tumor morphology such as lower uterine segment involvement and prominent TILs. If MLH1 immuno-staining is absent/lost, a follow-up assay is performed for MLH1 methylation, as this is a common cause of both MLH1 loss and MI, but is not inherited or associated with Lynch syndrome. If there is IHC loss in any of the tested proteins and no evidence of MLH1 promoter methylation, patients should then be referred for genetic counseling and consideration of germline testing for inherited mutations in these 4 genes. At any time, if there is high clinical suspicion, patients should be referred for genetic counseling, regardless of screening test results.
Screening and Prevention
Ovarian and endometrial screening is recommended for women with Lynch syndrome in addition to all screening recommended for both sexes. Physical examination, CA-125 measurements, and transvaginal sonography should be performed twice yearly and begin at age 30 to 35 years. This screening approach is the standard for women at high risk for ovarian cancer and will also identify gross abnormalities within the uterine cavity based on sonography. However, this screening approach has never been proven to reduce the risk of death from these diseases; nonetheless, it appears reasonable and awaits full validation and, in the absence of better screening approaches, is currently the best management scheme. Because this approach will not fully evaluate the uterine cavity for early lesions, annual endometrial sampling/biopsy is also recommended. For women who have completed childbearing, risk-reducing total hysterectomy with bilateral salpingo-oophorectomy is recommended. This approach reduces the risk of ovarian and endometrial cancer to approximately zero. In one study, the incidence of endometrial cancer in a control group was 33% compared with none in the risk-reducing hysterectomy group. The incidence of ovarian cancer in a control group was 5% compared with none in the risk-reducing salpingo-oophorectomy group.24 Colonoscopy should be performed every 1 to 2 years for at-risk patients.
Endometrial Cancer in Young Women
Endometrial cancer diagnosed in young women presents a difficult clinical dilemma. Because the mean age of endometrial cancer diagnosis is 63 years, “young women” can be characterized as those under the age of 50 or even 40 years. For early noninvasive endometrial cancer, conservative therapy may be appropriate for women who have not yet completed childbearing. The specific algorithms and outcomes are beyond the scope of this section; however, a portion of these patients will present with synchronous ovarian cancer or develop metachronous ovarian cancer. Clearly, young women with endometrial cancer are candidates for genetic counseling and Lynch syndrome testing. Recent data suggest that fewer than 10% of women with synchronous endometrial and ovarian cancer have Lynch syndrome, suggesting other pathobiology for women with these synchronous tumors, of which only half are under the age of 50 years at diagnosis.21 Young, normalweight women with endometrial cancer appear to have a relatively high incidence of infertility and/or irregular menstrual cycles, likely due to anovulation, which may also contribute to the increased risk of endometrial cancer.25
Approximately one-quarter of the normal weight, young women with endometrial cancer also had a synchronous ovarian cancer. Most of the synchronous ovarian tumors are of endometrioid histology, suggesting a possible field effect or link through endometriosis, as most sporadic ovarian cancer is of serous histology. Approximately 15% of the obese young women with endometrial cancer also had a synchronous ovarian cancer, suggesting greater frequency and possible different pathoetiology of synchronous ovarian cancer in young women with endometrial cancer based on body mass index and associated hormonal dysfunction. Additional studies have confirmed the approximate 25% incidence of synchronous ovarian cancer in young women with endometrial cancer.26
Synchronous endometrioid histologies have been associated with more favorable outcomes. These findings suggest that synchronous ovarian cancers develop in young women with endometrial cancer at a high rate for reasons that are not entirely clear, but this should be discussed with patients considering conservative management of early endometrial cancer. Risk-reducing salpingo-oophorectomy and hysterectomy should be considered for these patients once childbearing is complete.
Ovarian Cancer
The majority of ovarian cancers are epithelial in nature; a small percentage of them are sarcomas, and even fewer are germ cell or sex–cord stromal tumors. These uncommon tumors will be discussed elsewhere in the text. The most common histologic subtypes of epithelial ovarian cancer are serous, endometrioid, clear cell, and mucinous. Transitional cell tumors were described as an epithelial subtype, but more recently, they have been considered a simple epithelial variant and not a separate histologic subtype as originally thought.
Epithelial tumors of serous histology were thought to arise from the ovarian surface epithelium. Two common hypotheses of their origin are based on incessant or repeated ovulation and excessive hormonal stimulation. The first hypothesis implicates the ovarian surface epithelium and repeated cycles of ovulatory damage and repair. This hypothesis is supported by data that late menarche, early menopause, multiparity, and oral contraceptive use all decrease the risk of ovarian cancer. However, the reduction in the number of ovulatory cycles does not account for the associated magnitude of risk reduction. The evidence to support a hormonal basis of ovarian cancer arises from several areas. Lower gonadotropin levels, present during pregnancy and oral contraceptive use, reduce the risk of ovarian cancer. Women with polycystic ovarian disease, who have increased circulating androgens, are at an increased risk of developing ovarian cancer. Inclusions cysts, which have also been proposed to be the precursor cell to ovarian cancer, are found within the cortex of the ovary in close proximity to the vasculature, circulating hormones, and follicular cysts, which have high levels of androgen.27 Therefore, although ovulation and hormones have been directly linked to ovarian tumorigenesis, there are insufficiencies that fail to fully explain epidemiologic findings, suggesting a different set of biochemical, anatomic, and hormonal interactions, which will be discussed later.
Ovarian cancers have been broadly classified into 2 groups. Type 1 tumors are of endometrioid, mucinous, clear cell, and low-grade serous histology and thought to arise from ovarian cysts and secondary mullerian sites, such as endometriosis. These tumors appear to progress in a stepwise fashion from benign to borderline or atypical to malignant and invasive tumors. In this regard, various precursor lesions can be potentially identified and used for screening or prevention as appropriate. Type 2 tumors are more aggressive and contain high-grade serous tumors. These tumors most likely have a precursor lesion within the fallopian tube (discussed later) and metastasize early, present at an advanced stage, and account for most deaths from this disease. High-grade endometrioid tumors were thought to be a small subset of the more aggressive, advanced-stage ovarian epithelial tumors, but recent work has suggested that when IHC is incorporated into the diagnostic algorithm, many high-grade endometrioid tumors are morphologic variants of high-grade serous carcinoma, and only a few high-grade endometrioid tumors truly represent progression from a low-grade endometrioid tumor.28,29 Carcinosarcoma is another aggressive type 2 histology subtype, but is relatively uncommon (Table 2-2).
Table 2-2 Comparison Between Type 1 and Type 2 Ovarian Cancers
Pathology of Type 1 and Type 2 Ovarian Carcinoma
Type 1 ovarian tumors include endometrioid, mucinous, clear cell, and low-grade serous tumors. Most of these tumors develop in a stepwise fashion from a precursor borderline or low malignant potential (LMP) tumor. They are often confined to the ovary at the time of diagnosis and infrequently spread beyond the ovary, as compared with the more aggressive type 2 ovarian tumors. Type 1 serous tumors are uniformly low grade. Many will present in association with a borderline tumor. Borderline/LMP tumors contain branched glandular serous cells, and a subset of these tumors have a greater papillary architecture, referred to as micropapillary. Micropapillary tumors are thought to have a greater propensity for metastasis and association with frank invasion.
There is some controversy regarding whether to consider early invasive borderline tumors to be low-grade serous carcinoma or persistent borderline tumors, simply with a focus of microinvasion. Either naming scheme applied to these tumors describes the same histologic morphology and portends a higher likelihood of metastasis, recurrence, and progression to fully invasive low-grade serous carcinoma.30,31Microinvasion and micropapillary architecture are the 2 most reproducible findings associated with decreased long-term outcomes in borderline/LMP tumors.
Mucinous tumors of the ovary are extremely rare. In the past, many gastrointestinal tumors metastatic to the ovary had been misclassified as primary ovarian mucinous neoplasms.32 Although IHC can assist with the differential diagnosis of an ovarian versus a gastrointestinal primary, anatomic features may be most helpful. When IHC is performed, primary ovarian mucinous tumors preferentially express CK7 over CK20 and are negative for nuclear CTNNB1 (β-catenin). Features favoring metastatic disease include bilateral involvement, surface involvement, signet ring cells, and small size. Mucinous tumors that are bilateral and of any size or unilateral and smaller than 13 cm are mostly metastatic tumors to the ovary. Tumors that are unilateral and larger than or equal to 13 cm in size are mostly primary ovarian malignancies. This algorithm correctly classified 98% of the primary ovarian tumors and 82% of the metastases.33 These findings are further confirmed by a recent cooperative group study.34
Of all potential advanced-stage mucinous ovarian neoplasms, approximately one-third are found to be primarily from the ovary, and the remainder are tumors metastatic to the ovary. Advanced mucinous tumors of both types have a worse overall survival than ovarian serous tumors, likely due to relative chemoresistance to standard agents. Classic pseudomyxoma peritonei, traditionally thought to be associated with metastatic mucinous ovarian neoplasms, is in fact uniformly associated with appendiceal neoplasms, with the rare exception of an appendiceal neoplasm that arises in the setting of a mature cystic teratoma of the ovary. Primary ovarian mucinous tumors are uniformly low grade at presentation and can grow to be quite large before metastasizing.
Borderline endometrioid tumors are often found in association with invasive endometrioid ovarian carcinoma, which is typically low grade at diagnosis and confined to the ovary. As with mucinous ovarian tumors, endometrioid ovarian carcinomas can be large at presentation, yet without evidence of metastasis to other organs. These tumors are also frequently seen in association with endometriosis, suggesting a pathogenic link between the 2 processes. Morphologically, these tumors resemble similar low-grade tumors of the endometrium. Pure borderline endometrioid tumors without invasive components follow an entirely benign course. Clear cell ovarian tumors represent a unique subset of type 1 tumors. They are similar to other type 1 ovarian tumors in that they can be large and unilateral at presentation, are often associated with a borderline clear cell tumor, and are commonly confined to the ovary at diagnosis. They are different from the other type 1 tumors in that they are high-grade lesions. These tumors are typically composed of hobnail cells with clear cytoplasm.
Type 2 high-grade serous tumors do not have a logical and stepwise progression from a well-defined precursor lesion to invasive carcinoma. High-grade serous tumors represent the vast majority of all serous ovarian invasive carcinomas. These tumors are generally at an advanced stage at presentation, with 94% of these tumors presenting at stage III or IV. These tumors are morphologically heterogeneous with marked nuclear atypia, frequent papillae, and intermittent presence of psammoma bodies. They are often gland forming, but can also present in a solid, sheet-like arrangement.
Genetics of Type 1 and Type 2 Ovarian Carcinoma
Type 1 ovarian tumors have well-defined molecular abnormalities. Low-grade serous carcinomas have frequent and mutually exclusive KRAS and BRAF mutations and rare mutation in either ERBB2 or TP53(< 10%). Both KRASand BRAF mutations occur in approximately 30% of low-grade serous carcinomas and their precursor borderline serous tumor. Low-grade tumors have much less chromosomal instability than high-grade tumors.35 Primary ovarian mucinous tumors are rare, and anatomic findings, discussed previously, best differentiate ovarian from gastrointestinal mucinous tumors. KRAS mutations are common in mucinous ovarian tumors, present in up to 50% of cases.36Endometrioid ovarian carcinomas harbor various mutations in well-known oncogenes and tumor suppressors. Mutations in KRAS and BRAF are present in approximately 10% of endometrioid tumors, and PTEN mutations are present in approximately 20%. Mutations in CTNNB1 (β-catenin) have been reported in up to 60% of low-grade endometrioid tumors. Mutations in these genes have also been found in precursor lesions of LMP tumors. The striking overlap between the mutational spectrum of low-grade endometrioid ovarian tumors and endometrioid tumors of the endometrium suggests a shared etiology, which is likely mediated through the process of endometriosis. Recently, next-generation sequencing of RNA (RNA-seq) from ovarian tumors identified common mutations in 2 novel genes: ARID1A and PPP2R1A. ARID1A mutations were identified in 30% of endometrioid ovarian tumors and are thought to function as a tumor suppressor. PPP2R1A mutations have been identified in approximately 10% of ovarian endometrioid tumors and are thought to function as oncogenes.37,38 Endometrioid tumors fail to express WT-1 and often have MI, in contrast to the more common high-grade serous tumors. Clear cell ovarian carcinomas also have mutations in ARID1A and PPP2R1A. These mutations were first identified in clear cell tumors.39,40ARID1Amutations are present in approximately 50% of ovarian clear cell tumors, and PPP2R1A mutations are present in 5% to 10%.37,38 Clear cell carcinomas also have mutations in PTEN, KRAS, and TP53, but at low frequency. Interestingly, for a high-grade tumor, clear cell carcinomas have a relatively low frequency of chromosomal instability. Clear cell carcinomas frequently overexpress HNF-1B and, similar to endometrioid tumors, have no overexpression of WT-1 or TP53.
The discussion of type 2 tumors will be limited to high-grade serous carcinomas. Carcinosarcomas of the ovary are rare. They are part of the type 2 spectrum, have frequent TP53 mutations, and, like carcinosarcomas of the endometrium, are thought to be morphologic variants of serous tumors. It has been known for some time that high-grade serous carcinomas have frequent mutations in TP53. In fact, high-grade serous carcinomas may be the solid tumor with the highest frequency of TP53 mutations, besides certain inherited cancer syndromes. It is now well established that TP53 mutations can be identified in greater than 95% of high-grade serous carcinomas of the ovary.41Although the majority of type 2 serous tumors develop as de novo high-grade lesions with early loss of TP53, there are rare cases in which high-grade serous tumors can be molecularly characterized as arising from previously established low-grade serous tumors.42
Some of the earlier molecular genetic studies of high-grade serous carcinoma have been hampered by the inclusion of other subtypes of epithelial ovarian carcinoma. As evident from the previous sections, epithelial ovarian cancer is a heterogeneous group of diseases, a fact only fully recognized of late. Germ-line and somatic mutations in BRCA1 and BRCA2 are also relatively common in high-grade serous carcinomas. Germline mutations in each gene are present in 6% to 8% of high-grade serous carcinomas, and somatic mutations in each gene are present in approximately 3% of cases. Beyond these molecular findings, there are few recurrently mutated genes in high-grade serous carcinomas, a finding determined by The Cancer Genome Atlas (TCGA) pilot project in ovarian carcinoma, which was recently completed. TCGA comprehensively sequenced the whole exome of more than 300 high-grade serous carcinomas of the ovary and found few genes mutated in more than 5% of the samples other than TP53, BRCA1, and BRCA2 (tcga.cancer.gov). Copy number alterations are found in most high-grade serous carcinomas throughout the genome, making high-grade serous carcinoma one of the most genomically complex solid tumors. Recurrent amplifications are found in CCNE1, PIK3CA, KRAS, and MYC. Focal deletions have been identified in PTEN, RB1, NF1, and CDKN2A. High-grade serous carcinomas commonly overexpress both TP53 and WT-1.
Origins of Ovarian Carcinoma
It is well established and logical that type 1 tumors originate in a stepwise fashion from precursor lesions such as borderline tumors and endometriotic lesions. However, the origins of type 2 tumors remain elusive. Few precursor lesions have been identified on the ovarian surface epithelium, which has been the putative site of origin for high-grade serous carcinoma. In many other solid tumors, such as colorectal cancer, the normal tissue of origin histologically resembles the malignant counterpart and undergoes a stepwise progression from normal to cancer. In the ovary, which is lined by a single-cell layer of modified mesothelium, there is little resemblance to high-grade serous carcinoma, which frequently displays a great degree of papillary architecture prior to becoming diffusely solid or anaplastic. Furthermore, the ovarian surface is, embryologically, a modified mesothelial layer of peritoneum, and the ovary itself is a mesonephric structure, both unlike established high-grade serous carcinoma, which has a Mullerian phenotype. The uterus and fallopian tubes are true Mullerian structures developing, embryologically, from the Mullerian ducts in the absence of Mullerian inhibitory factor. When risk-reducing bilateral salpingo-oophorectomy became common for women with germline mutations in BRCA1 or BRCA2 (discussed later), occult “ovarian” cancers were identified in a small portion of cases. Interestingly, these occult tumors occurred predominantly in the distal fallopian tube and not on the surface of the ovary. To date, unifocal tubal carcinoma has not been reported in the proximal fallopian tube. These findings led investigators to examine the distal fallopian tube as a site of origin for epithelial ovarian cancer. To assist with these efforts, new processing methods for the fallopian tube were established to comprehensively section and evaluate the entire fimbria and perform better interrogation of the remainder of the fallopian tube. This technique is referred to as SEE-FIM43 (Figure 2-4).
FIGURE 2-4. Suggested procedure for processing risk-reducing salpingo-oophorectomy specimens.
As the distal fallopian tube was further interrogated, a putative precursor lesion, called serous tubal intraepithelial carcinoma (STIC), was identified. STIC lesions are characterized by nuclear atypia, epithelial layering, overexpression of TP53, increased proliferation as measured by Ki-67, and lack of stromal invasion. When these lesions are clearly invading underlying stroma, they can be classified as truly invasive malignancies. STIC lesions or invasive carcinoma has been found in 3% to 8% of all risk-reducing salpingo-oophorectomy specimens. Molecular evidence supports the relationship between STIC and invasive serous carcinoma. In 2 surveys of advanced serous carcinoma, STIC lesions were identified in the fallopian tubes of approximately 50% of cases44,45 (Figure 2-5). In addition, identical TP53 lesions have been identified in both the STIC lesions and established invasive carcinomas in the same patients.
FIGURE 2-5. Comparison of frequency distribution of high-grade serous carcinomas in pelvic (nonuterine) sites based on criteria for primary site origin. Dark blue: Conventional criteria, regardless of the presence of tubal intraepithelial carcinoma (TIC). Light blue: Conventional criteria ± TIC (cases with TIC were classified as being of primary fallopian tube origin and cases without TIC were classified as either an ovarian, peritoneal, or tubal primary tumor based on conventional criteria). (Reproduced, with permission, from Przybycin CG, Kurman RJ, Ronnett BM, et al. Are all pelvic [nonuterine] serous carcinomas of tubal origin? Am J Surg Pathol. 2010;34:1407-1416.)
It has been suggested that ovarian inclusion cysts may be sites of serous ovarian carcinogenesis. The data to support this theory are the identification of both intraepithelial carcinoma within ovarian inclusion cysts as well as regions of aneuploidy within pathologically normal tissue.27 The inclusion cyst hypothesis is consistent with the fallopian tube hypothesis in that fallopian tube epithelium can be shed onto the ovarian surface and later incorporated into the inclusion cyst during ovulation and surface repair. Additionally, immunomarkers, such as PAX8, found within fallopian tube epithelium, have also been found in the ovarian inclusion cysts, but not on the ovarian surface epithelium.46 Taken together, these data suggest that the distal fallopian tube is a likely site of origin for approximately 50% of all serous carcinomas, suggesting that pelvic serous carcinoma (PSC) may be a better term for previously presumed “ovarian” carcinoma (Figure 2-6). Where do the other half of PSCs originate? It is possible that there is an additional site of origin, or that all PSCs originate in the distal fallopian tube, yet we are only able to identify a precursor lesion in half of cases. The preneoplastic or early neoplastic cell could be shed onto the ovary or peritoneum early in the carcinogenic process, preventing the ability to identify a true precursor lesion within the fallopian tube. Further experimental evidence will be required to answer these important questions.
FIGURE 2-6. Proposed development of low-grade (LG) and high-grade (HG) serous carcinoma. A. One mechanism involves normal tubal epithelium that is shed from the fimbria, which implants on the ovary to form an inclusion cyst. Depending on whether there is a mutation of KRAS/BRAF/ERRB2 or TP53, an LG or HG serous carcinoma develops, respectively. LG serous carcinoma often develops from a serous borderline tumor, which, in turn, arises from a serous cyst-adenoma. Another mechanism involves exfoliation of malignant cells from an STIC that implants on the ovarian surface, resulting in the development of an HG serous carcinoma. B. A schematic representation of direct dissemination or shedding of STIC cells onto the ovarian surface on which the carcinoma cells ultimately establish a tumor mass that is presumably arising from the ovary. Of note, there may be stages of tumor progression that precede the formation of an STIC.
Hereditary Breast and Ovarian Cancer Syndrome
Mutations in BRCA1 and BRCA2 are known to increase the lifetime risks of breast and ovarian carcinoma. In the early 1990s, BRCA1 and BRCA2 were identified, cloned, and sequenced. Mutations in these genes were initially found in families with early-onset breast cancer. BRCA1 is the breast cancer 1 gene, and its naming has also been ascribed to the location where it was first identified: Berkeley, California. Since its identification, we have learned that BRCA1 increases the risk of both breast and ovarian cancer, whereas BRCA2 increases the risks of these 2 cancers among others. Both genes function as classic tumor suppressors, and in the case of hereditary breast and ovarian cancer syndrome, the first allele is mutated in the germline, and the second allele is typically inactivated through loss of heterozygosity, epigenetic silencing through methylation of the BRCA1 promoter, or rarely, a second somatic mutation. Both genes play important roles in maintaining genomic stability, and loss of either impairs the ability of DNA to repair damage. These genes play a critical role in homologous recombination, which is the most efficient mechanism to repair double-strand DNA damage that is often introduced through cytotoxic chemotherapy (Figure 2-7).
FIGURE 2-7. The BRCA1 network. BRCA1 is an important component of pathways that regulate DNA repair, cell-cycle progression, ubiquitylation, and transcriptional regulation. (Reprinted by permission from Macmillan Publishers Ltd: Nat Rev Cancer [Narod SA, Foulkes WD. BRCA1 and BRCA2: 1994 and beyond. Nat Rev Cancer. 2004;4:665-676], copyright 2004.)
When homologous recombination is deficient, the cell must use other less efficient mechanisms of repair, such as nonhomologous end joining (NHEJ) or microhomology-mediated end joining (MMEJ). Both NHEJ and MMEJ are more error-prone mechanisms of double-strand DNA repair than homologous recombination. Therefore, under the stress of chemotherapy, homologous recombination–deficient cells, such as those found in BRCA-associated tumors, are less likely to survive. Other genes that are thought to participate in homologous recombination include PTEN, ATM, ATR, PALB2, RAD51, and the Fanconi Anemia genes. Mutations and deletions in these genes are individually less common in high-grade serous carcinomas than BRCA1/2 mutations, but when all potential homologous recombination defects are considered together, nearly 50% of high-grade serous carcinomas have defective homologous recombination, a concept supported by experimental evidence.47
Penetrance
BRCA1 and BRCA2 germline mutations are uncommon in the general population, with an estimated frequency of approximately 0.1%. In certain ethic groups and geographic regions, however, germline mutations are more common. The most well-studied population are the Ashkenazi Jews, who carry 2 common founder mutations in BRCA1 (185delAG, 5382insC) and 1 in BRCA2 (6174delT). Founder mutations are mutations that originally arose in a single individual and then spread through a closely knit population. The Ashkenazi Jewish population is rife with these mutations due to repetitive occurrences of population contraction and expansion such as the Spanish Inquisition, the Holocaust, and the crusades and pogroms. The prevalence of BRCA1/2 founder mutations in the Ashkenazi Jewish population is approximately 2.5%. Other populations, such as French Canadians, Icelanders, Turks, Pakistanis, and certain ethnic groups in Africa, also have BRCA1/2 founder mutations. Icelanders, for example, are geographically isolated, allowing for the enrichment of founder mutations over time.
BRCA1 germline mutations confer a lifetime risk of approximately 40% for ovarian cancer and 60% to 80% for breast cancer. The general population risk for ovarian cancer is approximately 1.5%. BRCA1mutations also greatly increase the risk of fallopian tube and peritoneal cancer, but because these cancers are uncommon, the lifetime risks remain low. Due to the fallopian tube hypothesis discussed previously, it is likely that these lifetime risks will change due to changes in diagnostic criteria for ovarian and tubal carcinoma. BRCA2 germline mutations confer a lifetime risk of approximately 20% for ovarian cancer and 60% to 80% for breast cancer (Figure 2-8). These mutations also increase the risks of fallopian tube and peritoneal cancer as well as cancers of the prostate, pancreas, melanoma, stomach, and biliary tract.
FIGURE 2-8. Lifetime risks of breast (
) and ovarian cancer (
) of BRCA1 (A) and BRCA2 carriers (B). (Reprinted from Antoniou A, Pharoah PD, Narod S, et al. Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case Series unselected for family history: a combined analysis of 22 studies. Am J Hum Genet. 2003;72:1117-1130, with permission from Elsevier.)
Whether or not BRCA1/2 mutations increase the risk of endometrial cancer, endometrial serous cancer in particular, is debatable, with published studies finding evidence both for and against such an association. BRCA1/2 somatic mutations are each found in approximately 3% of high-grade serous carcinomas. In Ashkenazi Jewish women with high-grade serous carcinoma, the likelihood of identifying a germline BRCA1/2 mutation is 30% to 40%. In all women with high-grade serous carcinoma, the prevalence of germline BRCA1/2 mutations is approximately 15%, leading some academic centers to recommend genetic testing for all women with high-grade serous carcinoma. Members of families at high risk for breast or ovarian cancer should undergo genetic testing. These families generally include multiple members with breast and/or ovarian cancer, often at an early age of onset. Specific testing guidelines can be found from various professional organizations, the National Cancer Institute, and the National Comprehensive Cancer Network (nccn.org).
Pathology and Clinical Course
Ovarian tumors that develop in BRCA1/2 mutation carriers are uniformly high-grade serous carcinomas. There have been reports of transitional cell and endometrioid tumors in this population, but with the in clusion of IHC, these tumors are not considered morphologic variants of high-grade serous carcinoma. Greater than 90% of BRCA-associated ovarian tumors have been reported to be high-grade serous carcinomas, with 1% to 2% reported as endometrioid, transitional, or mucinous. The non-serous tumors are likely to be serous variants, as mentioned, or incidental metastases from non-ovarian primary sites such as gastrointestinal tumors. Most cases present at an advanced stage, as is seen with most high-grade serous carcinomas. The mean age at diagnosis for BRCA1- and BRCA2-associated cases is 54 and 62 years, respectively, compared with the mean age of 63 years for sporadic cases. Thus BRCA1-associated cases are diagnosed approximately 10 years earlier than sporadic cases, but BRCA2-associated cases are diagnosed at the same age as sporadic cases.
Although women with BRCA1/2 cases are more likely to develop ovarian cancer, as discussed previously, multiple studies have confirmed that, when diagnosed with high-grade serous carcinoma, these patients have a more favorable clinical course. It seems that the superior outcomes seen in BRCA-associated cases are due to an improved response to platinum-based chemotherapy. Because these tumors have defective homologous recombination repair, they are more likely to respond to the DNA damaging agents, such as platinum drugs and now possibly poly (ADP ribose) polymerase (PARP) inhibitors. Multiple studies have demonstrated a 50% to 100% longer overall survival in BRCA heterozygotes with high-grade serous carcinoma compared with sporadic high-grade serous carcinoma and an approximately 50% improvement in progression-free survival.
It is not clear whether tumors that obtain somatic BRCA1 or BRCA2 mutations will have a similarly improved survival as those with germline mutations, but this is a logical, yet unproven, conclusion. It is also not entirely clear if the improved outcomes seen in BRCA- associated tumors are completely independent of age at diagnosis, which is also a consistently identified factor associated with improved outcome. Some studies have found mutation status and age to be independent predictors of outcome, others have not, and others still have not assessed this relationship. The distinction between outcomes specifically related to BRCA1-associated versus BRCA2-associated tumors also remains murky because most studies have combined both mutation types to increase statistical power. However, considering the differences in clinical features, such as age at onset, and different biologic functions, the possibility remains that the clinical course could be different. As more consortia are created to study these issues, reliable answers should be forthcoming.
The role of BRCA promoter methylation, present in 10% of high-grade serous carcinomas, in clinical outcome and response to therapy remains unclear. The most recent and largest analysis performed by TCGA has confirmed that BRCA promoter methylation is not associated with an improved overall survival compared with sporadic cases, both of which are not as favorable as BRCA-associated cases. Unlike BRCA-associated ovarian cancer, BRCA-associated breast cancers are more likely to be triple negative, node positive, and aggressive, resulting in a general worse outcome for BRCA-associated breast cancer.
Prevention
Because germline testing can readily identify carriers of BRCA1 and BRCA2 mutations, ovarian cancer prevention can be considered in these women, and it is highly effective. Risk-reducing bilateral salpingo-oophorectomy is strongly recommended for ovarian cancer risk reduction in known carriers of germline BRCA1 or BRCA2 mutations. Multiple studies have demonstrated an 80% to 90% reduction in ovarian cancer risk in mutation carriers who had risk-reducing bilateral salpingo-oophorectomy as compared with those who had not. This procedure also reduces the risk of estrogen receptor–positive breast cancer. The earlier the procedure is performed, the greater the magnitude of risk reduction for both breast and ovarian cancer. This procedure also reduces the risk of fallopian tube cancer, as the fallopian tubes are (and should be) removed along with the ovaries.
All risk-reducing salpingo-oophorectomy procedures should be performed through minimally invasive approaches, unless there are extenuating circumstances. This procedure, when performed properly, will result in a small residual portion of the interstitial fallopian tube remaining in the uterine cornu. Although previously thought to potentially increase the risk of subsequent fallopian tube cancer, multiple studies have now confirmed that fallopian tube cancer develops in the distal fallopian tube or fimbria, and no cases of fallopian tube cancer developing in the interstitial portion of the tube have been reported. The association between BRCA germline mutations and the risk of uterine serous carcinoma is more controversial, with studies both supporting and refuting this association. No definitive data have yet been published to strongly support the role of hysterectomy at the time of risk-reducing salpingo-oophorectomy for the sole purpose of reducing the risk of potential BRCA-associated uterine serous carcinoma. There are often other medical, oncologic, and gynecologic reasons to consider hysterectomy at the time of risk-reducing salpingo-oophorectomy, and the risks and benefits of an associated hysterectomy must be evaluated individually. A small ongoing lifetime risk of primary peritoneal cancer persists after risk-reducing bilateral salpingo-oophorectomy (approximately 2%-3%).
Cervical Cancer
Cervical cancer is one of the greatest gynecologic cancer problems worldwide. Improvements have been made through screening and prevention with vaccination. Cervical cancer is a unique gynecologic malignancy in that more than 90% of cases can be etiologically linked to infection with the human papillomavirus (HPV). There are more than 100 types of papillomaviruses, which are double-stranded DNA viruses. Multiple types can infect the lower gynecologic tract, but only certain ones have been found to cause cervical cancer. The genome of HPV contains various open reading frames (ORFs). ORF E6 and E7 are oncoproteins that have the ability to integrate into host cells and are necessary for malignant transformation.48 HPV infection alone is not sufficient for malignant transformation, as many HPV infections are cleared spontaneously or progress into dysplasia but not cancer. The immune system mediates the host’s ability to clear HPV infections and regress dysplastic lesions. HPV also contains 2 capsid proteins that allow for type-specific vaccine production.
REFERENCES
1. Samuels Y, Wang Z, Bardelli A, et al. High frequency of mutations of the PIK3CA gene in human cancers. Science. 2004;304(5670):554.
2. King TA, Li W, Brogi E, et al. Heterogenic loss of the wild-type BRCA allele in human breast tumorigenesis. Ann Surg Oncol. 2007;14(9):2510-2518.
3. Smith DC, Prentice R, Thompson DJ, et al. Association of exogenous estrogen and endometrial carcinoma. N Engl J Med. 1975;293(23):1164-1167.
4. Pillay OC, Te Fong LF, Crow JC, et al. The association between polycystic ovaries and endometrial cancer. Hum Reprod. 2006; 21(4):924-929.
5. Baak JP, Mutter GL, Robboy S, et al. The molecular genetics and morphometry-based endometrial intraepithelial neoplasia classification system predicts disease progression in endometrial hyperplasia more accurately than the 1994 World Health Organization classification system. Cancer. 2005;103(11):2304-2312.
6. Sherman ME, Bitterman P, Rosenshein NB, et al. Uterine serous carcinoma. A morphologically diverse neoplasm with unifying clinicopathologic features. Am J Surg Pathol. 1992;16(6): 600-610.
7. Boruta DM 2nd, Gehrig PA, Fader AN, et al. Management of women with uterine papillary serous cancer: a Society of Gynecologic Oncology (SGO) review. Gynecol Oncol. 2009;115(1):142-153.
8. McCluggage WG. Uterine carcinosarcomas (malignant mixed Mullerian tumors) are metaplastic carcinomas. Int J Gynecol Cancer. 2002;12(6):687-690.
9. Ferguson SE, Tornos C, Hummer A, et al. Prognostic features of surgical stage I uterine carcinosarcoma. Am J Surg Pathol. 2007;31(11):1653-1661.
10. Pollock PM, Gartside MG, Dejeza LC, et al. Frequent activating FGFR2 mutations in endometrial carcinomas parallel germline mutations associated with craniosynostosis and skeletal dysplasia syndromes. Oncogene.2007;26(50):7158-7162.
11. Boland CR, Thibodeau SN, Hamilton SR, et al. A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: development of international criteria for the determination of microsatellite instability in colorectal cancer. Cancer Res. 1998;58(22):5 248-5257.
12. Vilar E, Gruber SB. Microsatellite instability in colorectal cancer—the stable evidence. Nat Rev Clin Oncol. 2010;7(3): 153-162.
13. An HJ, Logani S, Isacson C, et al. Molecular characterization of uterine clear cell carcinoma. Mod Pathol. 2004;17(5):530-537.
14. Lynch HT, Shaw MW, Magnuson CW, et al. Hereditary factors in cancer. Study of two large midwestern kindreds. Arch Intern Med. 1966;117(2):206-212.
15. Aarnio M, Sankila R, Pukkala E, et al. Cancer risk in mutation carriers of DNA-mismatch-repair genes. Int J Cancer. 1999;81(2):214-218.
16. Meyer LA, Broaddus RR, Lu KH. Endometrial cancer and Lynch syndrome: clinical and pathologic considerations. Cancer Control. 2009;16(1):14-22.
17. Lu KH, Schorge JO, Rodabaugh KJ, et al. Prospective determination of prevalence of lynch syndrome in young women with endometrial cancer. J Clin Oncol. 2007;25(33): 5158-5164.
18. Matthews KS, Estes JM, Conner MG, et al. Lynch syndrome in women less than 50 years of age with endometrial cancer. Obstet Gynecol. 2008;111(5):1161-1166.
19. Walsh MD, Cummings MC, Buchanan DD, et al. Molecular, pathologic, and clinical features of early-onset endometrial cancer: identifying presumptive Lynch syndrome patients. Clin Cancer Res.2008;14(6):1692-1700.
20. Watson P, Bützow R, Lynch HT, et al. The clinical features of ovarian cancer in hereditary nonpolyposis colorectal cancer. Gynecol Oncol. 2001;82(2):223-228.
21. Soliman PT, Broaddus RR, Schmeler KM, et al. Women with synchronous primary cancers of the endometrium and ovary: do they have Lynch syndrome? J Clin Oncol. 2005;23: 9344-9350.
22. Garg K, Soslow RA. Lynch syndrome (hereditary non-polyposis colorectal cancer) and endometrial carcinoma. J Clin Pathol. 2009;62(8):679-684.
23. Broaddus RR, Lynch HT, Chen LM, et al. Pathologic features of endometrial carcinoma associated with HNPCC: a comparison with sporadic endometrial carcinoma. Cancer. 2006;106(1): 87-94.
24. Schmeler KM, Lynch HT, Chen LM, et al. Prophylactic surgery to reduce the risk of gynecologic cancers in the Lynch syndrome. N Engl J Med. 2006;354(3):261-269.
25. Schmeler KM, Soliman PT, Sun CC, et al. Endometrial cancer in young, normal-weight women. Gynecol Oncol. 2005;99(2): 388-392.
26. Walsh C, Holschneider C, Hoang Y, et al. Coexisting ovarian malignancy in young women with endometrial cancer. Obstet Gynecol. 2005;106(4):693-699.
27. Pothuri B, Leitao MM, Levine DA, et al. Genetic analysis of the early natural history of epithelial ovarian carcinoma. PLoS One. 2010;5:e10358.
28. Gilks CB, Ionescu DN, Kalloger SE, et al. Tumor cell type can be reproducibly diagnosed and is of independent prognostic significance in patients with maximally debulked ovarian carcinoma. Hum Pathol. 2008;39(8):1239-1251.
29. Madore J, Ren F, Filali-Mouhim A, et al. Characterization of the molecular differences between ovarian endometrioid carcinoma and ovarian serous carcinoma. J Pathol. 2010;220(3): 392-400.
30. Kurman RJ, Shih IeM. Pathogenesis of ovarian cancer: lessons from morphology and molecular biology and their clinical implications. Int J Gynecol Pathol. 2008;27(2):151-160.
31. McKenney JK, Balzer BL, Longacre TA. Patterns of stromal invasion in ovarian serous tumors of low malignant potential (borderline tumors): a reevaluation of the concept of stromal microinvasion. Am J Surg Pathol.2006;30(10):1209-1221.
32. Soslow RA. Histologic subtypes of ovarian carcinoma: an overview. Int J Gynecol Pathol. 2008;27(2):161-174.
33. Yemelyanova AV, Vang R, Judson K, et al. Distinction of primary and metastatic mucinous tumors involving the ovary: analysis of size and laterality data by primary site with reevaluation of an algorithm for tumor classification. Am J Surg Pathol. 2008;32(1):128-138.
34. Zaino RJ, Brady MF, Lele SM, et al. Advanced stage mucinous adenocarcinoma of the ovary is both rare and highly lethal: a Gynecologic Oncology Group study. Cancer. 2011;117(3):554-562.
35. Kuo KT, Guan B, Feng Y, et al. Analysis of DNA copy number alterations in ovarian serous tumors identifies new molecular genetic changes in low-grade and high-grade carcinomas. Cancer Res.2009;69(9):4036-4042.
36. Gemignani ML, Schlaerth AC, Bogomolniy F, et al. Role of KRAS and BRAF gene mutations in mucinous ovarian carcinoma. Gynecol Oncol. 2003;90(2):378-381.
37. McConechy MK, Anglesio MS, Kalloger SE, et al. Subtype-specific mutation of PPP2R1A in endometrial and ovarian carcinomas. J Pathol. 2011;223(5):567-573.
38. Shih IeM, Panuganti PK, Kuo KT, et al. Somatic mutations of PPP2R1A in ovarian and uterine carcinomas. Am J Pathol. 2011;178(4):1442-1447.
39. Wiegand KC, Shah SP, Al-Agha OM, et al. ARID1A mutations in endometriosis-associated ovarian carcinomas. N Engl J Med. 2010;363(16):1532-1543.
40. Jones S, Wang TL, Shih IeM, et al. Frequent mutations of chromatin remodeling gene ARID1A in ovarian clear cell carcinoma. Science. 2010;330(6001):228-231.
41. Ahmed AA, Etemadmoghadam D, Temple J, et al. Driver mutations in TP53 are ubiquitous in high grade serous carcinoma of the ovary. J Pathol. 2010;221(1):49-56.
42. Dehari R, Kurman RJ, Logani S, et al. The development of high-grade serous carcinoma from atypical proliferative (borderline) serous tumors and low-grade micropapillary serous carcinoma: a morphologic and molecular genetic analysis. Am J Surg Pathol. 2007;31(7):1007-1012.
43. Medeiros F, Muto MG, Lee Y, et al. The tubal fimbria is a preferred site for early adenocarcinoma in women with familial ovarian cancer syndrome. Am J Surg Pathol. 2006;30(2): 230-236.
44. Kindelberger DW, Lee Y, Miron A, et al. Intraepithelial carcinoma of the fimbria and pelvic serous carcinoma: evidence for a causal relationship. Am J Surg Pathol. 2007;31(2): 161-169.
45. Przybycin CG, Kurman RJ, Ronnett BM, et al. Are all pelvic (nonuterine) serous carcinomas of tubal origin? Am J Surg Pathol. 2010;34(10):1407-1416.
46. Auersperg N. The origin of ovarian carcinomas: a unifying hypothesis. Int J Gynecol Pathol. 2011;30(1):12-21.
47. Mukhopadhyay A, Elattar A, Cerbinskaite A, et al. Development of a functional assay for homologous recombination status in primary cultures of epithelial ovarian tumor and correlation with sensitivity to poly(ADP-ribose) polymerase inhibitors. Clin Cancer Res. 2010;16(8):2344-2351.
48. zur Hausen H. Papillomaviruses causing cancer: evasion from host-cell control in early events in carcinogenesis. J Natl Cancer Inst. 2000;92(9):690-698.