Genetic Testing in Gastrointestinal Cancers: A Case-Based Approach
Genetic Testing in Gastrointestinal Cancers: A Case-Based Approach
ABSTRACT: High-risk genetic mutations that predispose individuals to various gastrointestinal (GI) cancers account for only about 5% of the population burden of these diseases. However, because early identification of at-risk individuals can so dramatically affect primary disease prevention, it is imperative that families who harbor susceptibility to these cancers be identified. The benefits of determining an underlying genetic susceptibility are important both for an individual patient’s ongoing management and for his or her family, where early identification of at-risk persons, along with the adoption of frequent cancer screenings and/or prophylactic risk-reduction surgeries, can have dramatic lifeprolonging benefits. In this article, we use a case-based approach to focus on the hereditary aspects of the most common GI cancers, including pancreatic, gastric, and colon cancer.
Be suspicious: knowing when to refer a patient for genetic testing
Indicators that identify individuals who may harbor germline mutations predisposing them to cancer include a family history of multiple cancers that reflect a pattern of disease, young age at onset, multiple primary tumors, tumors with particular histologic phenotypes, the presence of precursor lesions, and other related manifestations that cosegregate with cancer in the family. For individuals identified as being at risk for a cancer predisposition syndrome, referral to a genetic counselor for genetic risk assessment and testing is necessary and helps to ensure a comprehensive genetic evaluation. The investigation of a family for an underlying genetic cancer predisposition is guided by specific questions, including the type and age of cancer diagnoses in the family; whether there is access to pathology reports confirming the site of cancer and indicating histologic subtype; and in the scenario where the patient is unaffected, whether there is access to blood, tissue, or stored DNA from an affected family member on which genetic testing can be performed.
Typically, genetic testing should begin with the individual in the family most likely to be affected with the cancer predisposition syndrome, ie, with the family member who has the most severe disease, earliest onset, or rarest phenotype. Testing this individual decreases the likelihood of testing a family member with a phenocopy (eg, colon cancer that occurred sporadically) and is typically the most informative approach.
Which cancers have a hereditary component?
Familial clustering, evident in many cancer types, can in part be attributed to shared environmental exposures as well as to genetic susceptibility conferred by low-, medium-, or high-risk inherited factors. The heritable component of disease can be measured through studies comparing rates of disease in monozygotic vs dizygotic twins. In the case of cancer, an increased heritability has been shown for a variety of cancers, including prostate, colorectal, and breast cancers. Further evidence for increased heritability can be ascertained from epidemiologic studies using cancer data from homogenous populations that show greater than expected observed rates of certain types of cancer within families. In the case of families with multiple affected individuals in subsequent generations, there is heightened suspicion for rare mutations in genes that confer a high risk of cancer. In families where the pattern is not as striking, the cancer predisposition may still be genetic in origin but could be due to lower-penetrance risk variants. Genome-wide association studies have uncovered common risk variants in the genome that confer susceptibility to certain types of cancer; however, independently, each of these low-risk variants only marginally increases a person’s risk of cancer above that of the general population. Multiplicative models of common risk variants could potentially explain a proportion of familial clustering or be used to modify risk within moderate- to high-risk pedigrees, but as of yet, the clinical utility of redefining risk based on common variation has not been demonstrated.
To date, a large portion of the genetic etiology of hereditary cancer predisposition remains unexplained; even many seemingly high-risk families will not have an identifiable mutation in known cancer predisposition genes. However, with current technological advances in sequencing, rather than testing an individual for a specific gene or genes, clinical genetic testing may eventually evolve to incorporate whole-exome or whole-genome testing, in which all genes are assessed and the culprit genetic changes are identified after testing. Theoretically, this could reveal high-, medium-, and low-risk factors that might contribute to a familial cancer in an individual. In addition, factors such as genetic background and epigenetics (changes not encoded in the genome, resulting instead from environmental exposures that influence gene expression) may also further influence disease penetrance and affect an individual’s cancer risk.
To illustrate the benefits of genetic testing in individuals with a hereditary predisposition to GI cancers, we have provided four scenarios that are typical of what is encountered in the clinic. For each case, the general considerations outlined in Table 1 should be taken into account during the genetic risk assessment process.
Familial Pancreatic Cancer
Case: Mr. X is a 58-year-old man with a recent diagnosis of metastatic pancreatic cancer. On review of family history, you find a history of pancreatic, breast, and prostate cancers on the maternal side of the family (Figure 1). There is no personal or family history of melanoma, and on examination of his skin you do not find perioral freckling, melanomatous lesions, or evidence of old skin surgeries.
Based on the recently updated National Comprehensive Cancer Network (NCCN) guidelines, hereditary breast ovarian cancer (HBOC) syndrome should be considered in individuals with pancreatic adenocarcinoma diagnosed at any age who have two or more close relatives with breast and/or ovarian and/or pancreatic cancer diagnosed at any age. Thus, after a discussion with the patient regarding the potential usefulness to him and his family members of identifying an inherited mutation, you refer the patient to clinical genetics for further discussion of genetic testing.
Genetic evaluation: A total of 43,920 new diagnoses of pancreatic cancer in US men and women is projected for 2012. A quarter of these will prove to be secondary to environmental factors such as smoking; however, roughly 4,392 individuals (10%) will prove to have familial clustering of the disease, with 2% having disease resulting from a mutation in a high-risk Mendelian susceptibility gene (Figure 2).
In view of Mr. X’s family history of pancreatic, breast, and prostate cancers, HBOC syndrome is considered. The BRCA2 cancer susceptibility gene has been shown to be associated with sporadic and familial pancreatic cancer,[6,7] with the Breast Cancer Linkage Consortium study and other studies observing a 3.5- to 7-fold increased risk of pancreatic cancer in families carrying a BRCA2 mutation.[8,9] In patients with familial pancreatic cancer, a BRCA2 mutation is identified in 11% to 17% of families.[7,10] The association between BRCA1 mutations and pancreatic cancer is not as well defined, but the relative risk of pancreatic cancer in BRCA1 mutation carriers has been estimated to be two-fold higher than in the general population.
It is notable that mutations in the BRCA1/2 genes, while rare in the general population, are more prevalent in certain ethnic groups, such as Ashkenazi Jews. Since Mr. X is of Ashkenazi Jewish ancestry, testing begins with evaluation for the three Ashkenazi founder mutations, which account for ~90% to 95% of BRCA mutations in the Ashkenazim. Genetic testing reveals the presence of the BRCA2 6174delT germline mutation, the most common Ashkenazi founder mutation. In patients of Ashkenazi ancestry (unselected for family history) with resected pancreatic cancer, 5.5% were found to harbor a BRCA founder mutation, and in Ashkenazi breast-pancreas families, 14.2% were BRCA mutation–positive, with nearly equal distribution of BRCA1 and BRCA2 mutation carriers, suggesting that both of these genes may be involved with pancreatic cancer risk.[13,14]
During the genetic assessment of patients with pancreatic cancer, a number of other hereditary pancreatic cancer predispositions may also be considered (Figure 2). Peutz-Jeghers syndrome, which has a relative risk of 132 for pancreatic cancer, is usually associated with a history of intestinal polyposis and often presents with intussusception, gastrointestinal bleeding, and/or perioral freckling. Familial adenomatous polyposis, which has a relative risk of 4.5 for pancreatic cancer, is associated with a history of colorectal polyps or colorectal cancer; extraintestinal features such as congenital hypertrophy of the retinal pigment epithelium; or extracolonic tumors, such as pediatric hepatoblastomas, sebaceous adenomas, osteomas, desmoids, or medulloblastomas. Lynch syndrome has a ~9-fold increase in risk for pancreatic cancer above that of the general population; a history of colorectal, endometrial, and other cancers can also usually be found. Familial atypical multiple mole melanoma (FAMMM) syndrome, associated with germline mutations in CDKN2A, has a relative risk of 13 to 22 for pancreatic cancer, and is generally associated with a history of melanomas or dysplastic nevi. In addition, PALB2 mutations, thought to be associated with both breast and pancreatic cancer susceptibility, may also be considered.[20-22] Furthermore, other very rare conditions can be associated with an increased risk of pancreatic cancer, including hereditary pancreatitis, which has a relative risk of 53 to 87 for pancreatic cancer (Table).[23-25]
In terms of oncologic treatment, identification of a BRCA mutation in a patient with pancreatic cancer is becoming increasingly important, since new therapies, such as poly(ADP-ribose) polymerase (PARP) inhibitors, have shown significant activity in patients with advanced BRCA-associated breast and ovarian cancers and hold potential for the treatment of BRCA-associated pancreatic cancer as well. In fact, clinical trials exploring the efficacy of PARP inhibitors in BRCA-associated pancreatic cancer are currently underway.
With the knowledge that a BRCA mutation has been identified in this patient, cascade testing of family members can be performed so that at-risk family members can undertake the recommended cancer screening and prevention strategies outlined in Table 2. Although pancreatic cancer screening studies to date have not demonstrated a decrease in pancreatic cancer mortality, participation in clinical trials assessing the efficacy of pancreatic cancer screening in high-risk individuals may be a consideration for some BRCA-positive families with a history of pancreatic cancer.