Cancers of the reproductive organs (i.e., ovaries, uterus and testes), like other cancers, occur as a result of a multi-stage interaction of genetic and environmental factors. A small proportion of cancers of the reproductive organs occur as part of a recognised cancer syndrome, as a result of inheritance of mutations in highly penetrant cancer susceptibility genes (e.g., BRCA1 , BRCA2 , MLH1 or MSH2 ). Recognition of individuals and families with inherited cancer predisposition syndromes and individuals at high risk due to familial cancer clustering is fundamentally important for the management and treatment of the current cancer and for future prevention of further cancers for the individual and their extended family.
Introduction
Over the past two decades, significant advances have been made in our understanding of cancer genetics. Familial clustering of cancer has been recognised since Roman times; however, it is only now we are beginning to appreciate the true role of genetic factors in cancer susceptibility. Tumour suppressor genes acting in accordance with Knudson’s two-hit hypothesis were first described for retinoblastoma. The tumour suppressor genes, BRCA1 and BRCA2 , were cloned following genome-wide linkage analysis in the 1990s after the recognition of autosomal dominant transmission of breast and ovarian cancer in large families.
With increasing knowledge of molecular pathways, genes within an appropriate biological pathway can be directly interrogated within families with cancer clustering. This candidate gene approach, exemplified by the verification of germ line TP53 mutations in the Li–Fraumeni syndrome, has led to identification of further susceptibility genes. This is most notable in breast cancer susceptibility, with the identification of mutations in the DNA repair genes BRIP1 , CHEK2 and PALB2 , which confer a moderate increase in risk of breast cancer (Relative risk 2–4).
However, in practice, these low-prevalence, moderate to high-penetrance, cancer susceptibility genes account for only a small proportion of hereditary cancer. Emerging evidence suggests that the majority of familial clustering of some cancers may be explained by several lower-penetrance genetic variants and recent years have witnessed the emergence of genome-wide association studies. In these studies conducted for both cancer and other common complex diseases, the frequency of a specified variant is compared in cases and controls. With detailed maps of the human genome, including maps of single nucleotide polymorphisms (a single base change in DNA), extensive analysis of thousands of common variants within the genome, between cases and controls, can be performed. In large case-control studies, associations between common variants and disease are recognised, although in practice each variant is associated with only a small increase in relative risk of the disease. This approach has led to the discovery of common genetic variants, which are believed to act multiplicatively to increase cancer susceptibility. This has been demonstrated for breast and prostate cancer , but may prove difficult for unexplained familial ovarian and endometrial cancer, as they occur more rarely in the general population, making large-scale studies less feasible.
Consequently for cancers of the reproductive organs, current clinical practice should comprise identification and appropriate management of individuals at risk of highly penetrant cancer predisposition syndromes, through both gynaecology and clinical genetics. In future years, accurate characterisation of an individual’s risk for a particular cancer type, based on a panel of genetic polymorphisms may be within reach.
Recognition of cancer syndromes
In the recognition of cancer syndromes, two scenarios may be encountered: either the observation of clustering of a specific type of cancer within a family (e.g., breast or colon cancer) or the occurrence of a number of different cancers in family members, representing a defined cancer syndrome (e.g., sarcomas, breast cancer and brain tumours in the Li–Fraumeni syndrome).
The cardinal features of hereditary cancer syndromes include early age of cancer diagnosis, bilateral cancers, multiple primaries and multiple affected family members, spanning a number of generations.
Approximately 5% of all cases of endometrial cancer and 10% cases of ovarian cancer are due to inherited mutations in high-penetrance, low-prevalence cancer susceptibility genes. The majority of these cases are due to two recognised cancer syndromes: hereditary breast–ovarian cancer (HBOC) syndrome or hereditary ovarian cancer (HOC) syndrome, associated with germ-line mutations in BRCA1 and BRCA2 and Lynch syndrome, also known as hereditary non-polyposis colon cancer (HNPCC), associated with mutations in the DNA mismatch repair (MMR) genes ( Tables 1 and 2 ).
| Syndrome | Percentage of Hereditary Ovarian Cancer | Typical Histology | Gene | Lifetime Ovarian Cancer Risk (General population risk <2%) |
|---|---|---|---|---|
| Hereditary breast and ovarian cancer syndrome/Hereditary ovarian cancer syndrome | 90% | Epithelial – predominantly serous. | BRCA1 | 39–63% |
| BRCA2 | 11–27% | |||
| Lynch Syndrome | 5%–10% | Epithelial, predominantly serous, other epithelial histology also observed. | Mismatch Repair Genes ( MLH1, MSH2, MSH6, PMS2 ) | 12% |
| Gorlin Syndrome | <1% | Ovarian fibromas or (rare) fibrosarcomas. | PTCH | <2% |
| Peutz Jeghers Syndrome | <1% | Sex cord-stromal tumours particularly ovarian sex-cord tumour with annular tubules (SCTAT). | STK11 | 18% (all gynaecological malignancy) |
| Syndrome | Gene | Lifetime endometrial cancer risk |
|---|---|---|
| Lynch Syndrome | MMR Genes ( hMLH1, hMSH2, hMSH6 and PMS2 ) | 30–60% |
| Cowden Syndrome | PTEN | 5–10% |
In some hereditary cases, clinical genetic testing and identification of a causative pathogenic mutation are possible. However, in a subset of seemingly hereditary cases, no mutation is identified and the familial clustering is likely to be accounted for by a combination of environmental factors and low-risk alleles or genetic polymorphisms acting together to increase susceptibility to certain cancers, which has yet to be fully understood.
Recognition of cancer syndromes
In the recognition of cancer syndromes, two scenarios may be encountered: either the observation of clustering of a specific type of cancer within a family (e.g., breast or colon cancer) or the occurrence of a number of different cancers in family members, representing a defined cancer syndrome (e.g., sarcomas, breast cancer and brain tumours in the Li–Fraumeni syndrome).
The cardinal features of hereditary cancer syndromes include early age of cancer diagnosis, bilateral cancers, multiple primaries and multiple affected family members, spanning a number of generations.
Approximately 5% of all cases of endometrial cancer and 10% cases of ovarian cancer are due to inherited mutations in high-penetrance, low-prevalence cancer susceptibility genes. The majority of these cases are due to two recognised cancer syndromes: hereditary breast–ovarian cancer (HBOC) syndrome or hereditary ovarian cancer (HOC) syndrome, associated with germ-line mutations in BRCA1 and BRCA2 and Lynch syndrome, also known as hereditary non-polyposis colon cancer (HNPCC), associated with mutations in the DNA mismatch repair (MMR) genes ( Tables 1 and 2 ).
| Syndrome | Percentage of Hereditary Ovarian Cancer | Typical Histology | Gene | Lifetime Ovarian Cancer Risk (General population risk <2%) |
|---|---|---|---|---|
| Hereditary breast and ovarian cancer syndrome/Hereditary ovarian cancer syndrome | 90% | Epithelial – predominantly serous. | BRCA1 | 39–63% |
| BRCA2 | 11–27% | |||
| Lynch Syndrome | 5%–10% | Epithelial, predominantly serous, other epithelial histology also observed. | Mismatch Repair Genes ( MLH1, MSH2, MSH6, PMS2 ) | 12% |
| Gorlin Syndrome | <1% | Ovarian fibromas or (rare) fibrosarcomas. | PTCH | <2% |
| Peutz Jeghers Syndrome | <1% | Sex cord-stromal tumours particularly ovarian sex-cord tumour with annular tubules (SCTAT). | STK11 | 18% (all gynaecological malignancy) |
| Syndrome | Gene | Lifetime endometrial cancer risk |
|---|---|---|
| Lynch Syndrome | MMR Genes ( hMLH1, hMSH2, hMSH6 and PMS2 ) | 30–60% |
| Cowden Syndrome | PTEN | 5–10% |
In some hereditary cases, clinical genetic testing and identification of a causative pathogenic mutation are possible. However, in a subset of seemingly hereditary cases, no mutation is identified and the familial clustering is likely to be accounted for by a combination of environmental factors and low-risk alleles or genetic polymorphisms acting together to increase susceptibility to certain cancers, which has yet to be fully understood.
Hereditary Breast and Ovarian Cancer (HBOC) syndrome
HBOC syndrome is typified by four or more breast or ovarian cancers within a family, typically occurring at young ages or bilaterally in the case of breast cancer. Ovarian cancers are typically epithelial, with a younger age of onset in BRCA1 carriers (mean: 54 years), compared with sporadic ovarian cancers (mean: 63 years), which have a comparable age of onset to BRCA2 carriers (mean: 62 years). Breast cancer in these families is also typically seen at younger ages, particularly in BRCA1 carriers, compared with that in the general population ( Figs. 1 and 2 ).
Identification of BRCA1 and BRCA2
BRCA1 was cloned on chromosome17q21 in 1994, following a long search for the gene using linkage analysis. This was closely followed by discovery of BRCA2 on chromosome 13q12–13 in 1995. Both are classic tumour suppressor genes, requiring loss of both alleles in the susceptible tissue for progression of tumourigenesis. Both genes are involved in the maintenance of genomic stability by facilitating DNA repair, primarily executing DNA double-strand break repair by homologous recombination. This property is now a target for therapeutic exploitation in the treatment of breast and ovarian cancer in women with inherited BRCA1 and BRCA2 mutations, with both platinum-based chemotherapy that induces double-strand DNA breaks and PARP inhibitors, which exploit the role of BRCA1 and BRCA2 in DNA repair.
Despite BRCA1 and BRCA2 initially appearing to be genes with similar functions, it is now clear that the two genes are different in terms of their molecular biology, protein interactions and the cancer risks they confer.
Contribution of BRCA1 and BRCA2 mutations to familial ovarian cancer
Most studies looking at the contribution of germ line BRCA1 and BRCA2 mutations to familial ovarian cancer have ascertained families due to clustering of breast rather than ovarian cancer cases. One large study, which first reported in 1999 and was recently revisited, ascertained families due to ovarian cancer clustering. They found the proportion of families with a mutation varied according to family history. In families with at least three close relatives with epithelial ovarian cancer, 63% were found to have a mutation in BRCA1 or BRCA2 , with the majority of mutations occurring in BRCA1 . Mutation frequency increased in families who also had cases of breast cancer, diagnosed before age 60 years and increasing numbers of ovarian cancers. In families with only two cases of ovarian cancer and no breast cancer, only 27% were found to have a mutation in one of the two genes, whereas in families with at least two ovarian and two breast cancer cases, BRCA mutations were detected in 83% families. In families without a detected mutation, it is possible that some mutations were missed due to insensitivity of detection method and mutations in the MMR genes, will account for further families. However, a proportion of site-specific familial ovarian cancer remains unexplained. As yet no single gene that confers increased susceptibility to ovarian cancer alone has been identified. It is possible that families with only two cases may be explained either by chance or inheritance of several lower susceptibility genes.
Contribution of BRCA1 and BRCA2 mutations to isolated ovarian cancer
Women with ovarian cancer or their close relatives are often concerned about genetic risk, due to the aggressive nature of ovarian carcinomas and the difficulty with early detection. A number of studies have been performed to assess the prevalence of BRCA1 and BRCA2 mutations in population series of ovarian cancer cases, unselected for family history. However, there are flaws in study design, including small numbers, patient ascertainment, lack of detailed family history on maternal and paternal sides or inclusion of women with founder mutations. Consequently, estimates of mutation prevalence in ovarian cancer overall, vary between 2% and 9% in studies for BRCA1 and 1% and 6% for BRCA2 .
Currently, the National Institute for Health and Clinical Excellence (NICE) guidelines recommend BRCA testing for individuals with ovarian cancer if there is a further case of ovarian cancer or breast cancer diagnosed below 50 years within the close family, but not in an isolated case.
Penetrance of BRCA1 and BRCA2 mutations
Inherited mutations in BRCA1 and BRCA2 are highly penetrant. However, estimations of cancer risk in mutation carriers vary between studies, mainly due to ascertainment bias. Early studies calculated risk from small numbers of large cancer families, and suggested that women with mutations in BRCA1 have up to an 87% lifetime risk (defined as risk up to age 70) of breast cancer and up to a 63% lifetime risk for ovarian cancer. Whilst women with mutations in BRCA2 have a comparably elevated lifetime risk of breast cancer of up to 84%, the degree of ovarian cancer risk is lower with lifetime risks of up to 27%. Male breast cancer, pancreatic cancer, prostate cancer and melanoma are more frequently observed in BRCA2 mutation carriers than in the general population. In BRCA2 , a higher risk of ovarian cancer is associated with mutations occurring in the ‘ovarian cancer cluster region’ in exon 11.
Notably in population-based studies of BRCA1 and BRCA2 mutation carriers unselected for family history, the estimated cancer risks are significantly lower. In a combined analysis of 22 studies in which BRCA1 and BRCA2 carriers had been identified independently of their family history, ovarian cancer lifetime risk was estimated at 39% for BRCA1 carriers and 11% for BRCA2 carriers. The higher penetrance figures determined in earlier studies are partly explained by ascertainment bias from large families with multiple cancer cases.
The variance in penetrance estimates between studies are also partly explained by modifying environmental factors such as breast feeding, parity and hormonal factors, as well as low-penetrance genetic variants, which can cluster in families. Thus, families with large numbers of affected individuals may have clustering of other genetic variants, each conferring a small increase in risk, in addition to BRCA1 and BRCA2 , explaining the higher penetrance estimates from the original studies that included families with multiple cancer cases.
Genome-wide association studies have identified common alleles, which are associated with increased breast cancer risk in the general population , and further work has demonstrated that some of these act multiplicatively to alter breast cancer risk in BRCA2 carriers, but the evidence is not so strong for interaction with BRCA1 . Similar genetic interactions need further clarification for ovarian cancer risk.
BRCA1 and BRCA2 Founder mutations
The Ashkenazi Jewish population is host to a number of founder mutations for inherited conditions. Three founder mutations occur in the BRCA genes: 185delAG and 5382insC in BRCA1 and 6174delT in BRCA2 . Up to 60% of ovarian cancer and 30% of early-onset breast cancer in this population is due to one of these mutations. Approximately 1 in 40 Ashkenazi Jews are mutation carriers. Consequently, mutation analysis of these three founder mutations can be performed in unaffected women with a family history of breast or ovarian cancer and Ashkenazi heritage.
Founder mutations have also been described in Northern European, French-Canadian and other populations.
Histology of ovarian cancer in BRCA1 and BRCA2 carriers
The majority of ovarian cancers occurring in women carrying germ-line mutations in BRCA1 or BRCA2 are diagnosed at a younger age, are high grade and advanced-stage serous carcinomas. Epithelial serous ovarian carcinoma is the most common histological subtype and is found in up to 90% of all cases. Endometrioid carcinomas account for most of the remaining cases, with other epithelial tumours occurring occasionally. Papillary serous peritoneal tumours and fallopian tube cancers should be considered as part of the tumour spectrum in women carrying BRCA1 or BRCA2 mutations.
Mucinous tumours are generally not thought to be characteristic of ovarian cancers in BRCA1 or BRCA2 mutation carriers , but are associated with HNPCC/Lynch syndrome. Similarly, borderline tumours of the ovary are also not characteristic of cancers in BRCA1 or BRCA2 mutation carriers. Germ cell tumours and sex-cord stromal tumours are not characteristically seen in women with BRCA1 or BRCA2 mutations. If there is a family history of cancer, an ovarian tumour of this histology should prompt consideration of another cancer syndrome, such as Peutz–Jeghers syndrome.
Prognosis of ovarian cancer in BRCA1 or BRCA2 carriers
Ovarian cancers occurring in women with BRCA1 or BRCA2 mutations are generally of a higher stage and grade at diagnosis compared with sporadic cancers. Some studies suggest that hereditary ovarian cancers have been found to have a better clinical outcome with improved survival and recurrence-free interval after chemotherapy than sporadic cancer. However, not all studies agree with this, with some studies suggesting poorer survival. In addition, the studies suggesting better survival have been criticised for selection bias. Improved survival could be due to the increased sensitivity of BRCA-deficient tumour cells to DNA-damaging agents such as cisplatin, which induce double-strand DNA breaks.
Lynch syndrome
The clustering in families of colon cancer, often also with cases of ovarian, endometrial, gastric, transitional cell carcinomas of the urological tract and biliary tract cancers, typify Lynch syndrome (LS) or HNPCC ( Fig. 3 ). Colonic carcinomas in LS/HNPCC are typically diagnosed at an early age, and predominantly occur in the proximal colon. The risk for individuals with LS/HNPCC of developing colon cancer is approximately 80% by age 70. The modified Amsterdam criteria are used clinically to identify families suitable for molecular testing and are more widely used than the original Amsterdam criteria, which do not take into account the extra-colonic malignancies.
Genetics of Lynch Syndrome
Inherited mutations in a set of genes known as the mismatch repair genes are responsible for LS/HNPCC. These genes are responsible for repair of nucleotide mismatches and insertion–deletion loops during DNA replication.
Loss of function in the MMR pathway results in microsatellite instability in tumour cells, the hallmark of LS/HNPCC, identified by the presence of extra repeats at a DNA microsatellite in the tumour, when compared with normal DNA (microsatellites are stretches of DNA in which a series of nucleotides are repeated).
Mutations in MLH1 and MSH2 account for 70–90% of germ-line mutations in LS families, mutations in MSH6 account for approximately 10% but mutations in PMS2 and PMS1 are also occasionally found.
LS is most frequently diagnosed by gastroenterologists, due to the predominance of colon cancer in this condition. However, a retrospective multicentre study of women with LS and both gastrointestinal and gynaecological cancer found that, in 51% of cases, women presented with a gynaecological cancer and 49% presented with gastrointestinal cancer. Consequently, it is important for gynaecologists to recognise the cardinal features of this syndrome.
The Amsterdam criteria were originally used to define LS/HNPCC families for research purposes, and classify a family clinically. They serve as a good discriminator for germ-line MMR mutations, with mutations identified in approximately 50% families fulfilling these criteria , but are not very sensitive, so the criteria were later widened to include extracolonic cancers in the “modified Amsterdam criteria” ( Table 3 ). However, it is widely recognised that families not meeting even the modified criteria may have germ-line MMR mutations, so the Amsterdam criteria may not be sensitive or specific enough to use in isolation. Therefore, analysis for microsatellite instability (MSI) in tumour samples may be performed in some families first as a screening test, because although MSI may be seen in about 15% of sporadic colorectal cancers, the majority of colon cancers in individuals with LS/HNPCC are MSI-high. It is also important to be aware of ‘red flag’ features in one individual, such as synchronous or metachronous cancers, or young age at diagnosis.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
