Abstract

Introduction

In Puerto Rico, colorectal cancer (CRC) represents the second leading cause of cancer in men and women [17]. Familial CRC accounts for 10–15% of all CRCs [10, 12] and several studies suggest that inheritance has a significant impact in the pathogenesis of up to a third of all CRC cases [8,9]. One such familial syndrome accounting for about 1% of all CRC cases is familial adenomatous polyposis (FAP) [1]. However, little is known about the prevalence, clinical manifestations and genetic mutations in US Hispanic individuals with familial polyposis syndromes.

Familial adenomatous polyposis is an autosomal dominant inherited disease that usually manifests as tens to thousands of adenomas in the colon and rectum tin the second decade of life. These patients have a 100% risk of developing CRC, which usually occurs a decade after adenomas first present [1]. FAP occurs in 1 out of every 8,300 people (1). Patients often go without symptoms for years until adenomas become so numerous and large that they manifest as nonspecific symptoms such as anemia, rectal bleeding, abdominal pain, constipation or diarrhea, palpable abdominal masses and weight loss [1]. Aside from CRC, FAP usually manifests with benign or malignant extracolonic abnormalities including osteomas, dental abnormalities, congenital hypertrophy of the retinal pigment epithelium, desmoids tumors and extra-colonic malignancies in the thyroid, liver, bile ducts and central nervous system [1]. Additionally, FAP presents with extracolonic polyps in the duodenum and the stomach [14]. FAP results from a germline mutation in the Adenomatous Polyposis Coli (APC) gene. Most patients (70%) have a family history of CRC [1].

Other individuals with adenomatous polyposis present with an autosomal recessive disorder and have a MUTYH mutation that causes MUTYH-associated polyposis (MAP). The MUTYH gene codes for an enzyme involved in DNA that repairs oxidative damage [18]. The two most common variants in the MUTYH gene are Y165C and G382D missense mutations, both accounting for more than 80% of all MYH variants reported in Caucasian populations. When an individual inherits two copies of a defective MUTYH gene, they inherit the familial cancer syndrome, MAP. This autosomal recessive condition is characterized by an increased risk of developing CRC and adenomas in both the upper and lower gastrointestinal tract [1].

The aim of the present study is to characterize the phenotype (including colonic and extra-colonic manifestations) and genotype (APC/MUTY genes) of US Hispanics individuals with polyposis syndromes recruited to the Puerto Rico Familial Colorectal Cancer Registry (PURIFICAR).

Methods

Individuals with either personal or family history of possible familial colorectal cancer and/or polyposis are referred to the Puerto Rico Familial Colorectal Cancer Registry (PURIFICAR; http://purificar.rcm.upr.edu/) [13] by gastroenterologists, oncologists, and colorectal surgeons. The Puerto Rico Central Cancer Registry (PRCCR), the state funded national cancer registry in Puerto Rico, also refers incident colorectal cancer cases to PURIFICAR. The probands in this study were identified from PURIFICAR (http://purificar.rcm.upr.edu/)[13]. PURIFICAR was established in 2005 at the University of Puerto Rico Comprehensive Cancer Center and has received direct or in kind support from the National Institutes of Health, the Puerto Rico Gastroenterology Association and the Puerto Rico Colorrectal Cancer Coalition (see Acknowledgment Section). Subjects enrolled in PURIFICAR are US Hispanic individuals with a clinical and/or genetic diagnosis of FAP, attenuated FAP (AFAP), MAP, hamartomatous polyposis syndromes and Lynch Syndrome. Consented probands complete a comprehensive baseline questionnaire capturing medical, environmental exposures and cancer family history. Probands are referred to the Hereditary Cancer Prevention Clinic (led by Dr. Marcia Cruz-Correa) where they are offered a vast array of services including but not limited to genetic counseling and testing, diagnostic and surveillance endoscopic, radiographic and/or surgical procedures, referral to specialists, and communication with insurance companies to facilitate coverage/approval of services (Figure 1).

Pedigrees for each proband are completed to trace the number of affected relatives with polyposis and/or cancer using PROGENY [21]. Pathology reports and medical/surgical records are obtained to confirm cancer diagnosis reported by probands. Affected family members living in Puerto Rico and US are also invited to participate in the registry.

Our study population consists of US Hispanic men and women ages 10 and older recruited through PURIFICAR. For the current study, analysis was limited to cases with both the clinical and/or genetic diagnosis of FAP and cases with genetic diagnosis of MAP. Genetic diagnosis of FAP and MAP was established by identifying mutations by commercial sequence analyses (MYRIAD Genetic Laboratories, Inc.) of the APC and MUTYH genes, respectively. Clinical diagnosis of FAP was defined as individuals with more than 20 neoplastic lesions taking into account age at onset and inheritance pattern (adenomas). All patients were ruled out for APC mutations prior to undergoing MUTYH sequence analysis.

The statistical analyses consisted of frequency distribution, Wilcoxon rank-sum test, chi-square, Fisher’s exact test as appropriate using STATA 11.0 software (STATA Corporation, Texas) [22].

Results

We report for the first time observations on Hispanics with familial polyposis syndrome recruited through the Puerto Rico Familial Colorectal Cancer Registry (PURIFICAR)[13]. Data from familial cancer registries have been reported previously in other populations, however, no data have been reported from a predominantly Hispanic population.

Discussion

PURIFICAR is the first familial cancer registry established in Puerto Rico. We report, for the first time to our knowledge, clinical and genetic mutation data from a Hispanic population with familial polyposis (44 cases from 32 families). Clinical and genetic data from familial polyposis registries (ranging from 77–122 patients; representing 33–96 families) have been previously reported in most Caucasian cohorts [3–6]. Our Hispanic cases had similar prevalence of colorectal and gastrointestinal polyps as noted in other non-Hispanic cohorts [4]. Extracolonic malignancies were found in 9.1% of our cohort, similar to that observed in other reported familial polyposis registries [6,15]. However, our findings emphasize the need to further evaluate genotypic and phenotypic data in familial cancer syndromes in other ethnic groups in order to make the best screening and testing strategies for at-risk patient management.

A total of 31 FAP cases from 19 families recruited from PURIFICAR were evaluated in this study. In contrast to other studies, we found a lower mean age at diagnosis (27 vs. 32–41 years), which may be possibly due to our inclusion of younger patients [3–6]. Unique genetic mutations were detected in the 42.2% of FAP families for which insurance companies provided coverage for genetic testing. The remaining 57.8% families had a clinical diagnosis of FAP. The fact that we are not able to provide germline mutation analyses to all patients represents a limitation of this study as some FAP patients may actually have MAP. However, the higher polyp burden observed in most patients classified as FAP (with over two-thirds having more than 100 colonic adenomas), the earlier age of onset (mean age at diagnosis 27 years of age) and the autosomal dominant inheritance pattern observed in the affected families support the FAP diagnosis. Extracolonic malignancies are not uncommon in patients with FAP. Previously, published studies report a prevalence of 25%–66% for congenital hypertrophy of the retinal pigment epithelium (CHRPE) among FAP cases; however, there were no available data for this condition within our study [5]. Limited use of routine ophthalmic examination of FAP individuals and/or non-referral of CHRPE-positive patients for genetic evaluation may be contributing factors to our observations. Data regarding benign extracolonic manifestations was not available for all our cases; however, 41.9% patients had upper gastrointestinal, 9.7% had osteomas, 6.5% epidermal cysts, and none had thyroid nodules. Desmoid tumors were detected in three cases (9.7%) in two families (Family IDs: 8521 & 9132); no cases of duodenal or thyroid cancer were identified in our cohort. We observed that a third of our FAP cases had prophylactic IRA. In contrast, 90% of prophylactic surgeries performed during the last decade were IPAA, which may reflect surgical trends and available technical expertise. Regarding prophylactic colorectal surgeries, less restorative proctocolectomies were performed in our Hispanic FAP patients compared to other populations (57% vs. 73%)[3].

Limited studies exist evaluating MAP cohorts regarding mutation and clinical spectrum. One investigation included 26 patients from 9 families with a mean age of diagnosis of 47 years (range 30–70 years) for patients with biallelic MUTYH mutations, with 11.5% of these patients having more than 100 colonic polyps [20]. Our data are similar to most previous studies in Caucasian populations with common biallelic MUTYH mutations variants (Y165C and G382D variants also known as Y179C and G396D) accounting for more than 80% of all reported mutant alleles [18]. There is very limited information on particular MUTYH sequence variants in racial/ethnic groups other than Caucasians [18, 23]. In our Hispanic cohort, we identified MUTYH variants not routinely seen among the Caucasian cohorts, which would have resulted in the failure to identify MAP in several Hispanic patients if only the MUTYH variants Y165C and G382D were tested. Hispanics are an admixed population resulting from the mixture of three races (European, African and Indian), which may account for the unusual MUTYH variants found in our cohort. Our results support the need for a more comprehensive characterization of the mutations associated with MAP in other racial/ethnic groups and the use of full MUTYH sequencing in non-Caucasian individuals.

A striking finding in our cohort was that 30.8% of the MAP patients had monoallelic MUTYH mutations. Individuals who inherit both MUTYH mutations develop the clinical phenotype of MAP. However, it is unclear whether carriers of only one mutation (monoallelic variants) are at increased risk for CRC [18, 25–28]. This uncertainty is largely due to the low frequency of monoallelic carriers for MUTYH (1–2%) [1,18]. The frequency of biallelic mutation carriers, on the other hand, is between 1 per 10,000 and 1 per 40,000 births (compared to 1 per 8,300 for FAP) [1]. Individuals with biallelic MUTYH mutations usually develop between 10 and 100 adenomas in the colon and rectum, and approximately one third develop adenomas in the upper GI tract. Among MAP Hispanic patients with biallelic mutations, we observed a similar frequency of gastrointestinal adenomas as seen in Caucasian cohorts, with most patients (93%) having < 100 colorectal adenomas and about a third (31%) having upper GI tract polyps. Some studies support that the number of adenomas are dependent on the number of mutated alleles in the MUTYH gene [24]. In our cohort, the monoallelic MUTYH mutation carriers presented a less aggressive phenotype with less than 20 polyps (ranging from 5–20 polyps) with no extra-intestinal manifestations. However, three out of the four probands had first-degree family members with adenomatous polyps whereas the other proband had two third-degree relatives with CRC at a young age (50 and 51 years of age). Although three of the four monoallelic mutations detected are not in the mutation hotspots, they may have the potential to adversely affect protein function which cannot be established solely based on sequence analyses. As genetic data from various racial/ethnic populations is still limited, it is vital to request a full DNA sequence of the MUTYH gene when evaluating non-Caucasian patients with polyposis.

MAP is a relatively newly recognized familial polyposis syndrome, therefore there is very limited information regarding cancer risk in these patients [1,18,23–29]. Biallelic MUTYH mutations have been reported to be associated with a 93-fold excess risk of CRC, with an almost complete penetrance at 60 years of age [27]. In addition, heterozygous and homozygous MUTYH mutation carriers were found to have a high risk for synchronous cancers (24% in the colorectum and 16% in the upper gastrointestinal tract), but did not show an increased risk for extra intestinal tumors [25]. Colorectal cancer risk among monoallelic mutation carriers still remains controversial. One meta-analysis showed that MUTYH mutation carriers had a slightly increased chance of CRC of 15% [18], while a Swiss cohort reported a significantly increased risk only for biallelic mutation carriers (71.4% vs. 0%, p <0.007) [24]. Similarly, a large Spaniard population-based study reported no increased risk for CRC (odds ratio, 1.11; 95% confidence interval, 0.90–1.37) among monoallelic MUTYH carriers [29]. Only two individuals, biallelic MUTYH mutation carriers developed cancer during our study.

In Puerto Rico, diagnosis of familial polyposis syndromes is mainly based on clinical phenotype. In fact, genetic mutation analysis was only available for 42.2% of FAP families whereas the availability of genetic mutation results in other studies ranged from 24.0%–92.6% [3–6]. Barriers to genetic testing include lack of insurance coverage, limited awareness by physicians and patients, and cultural beliefs regarding colorectal cancer and genetic diseases. Mutation analysis of the MUTYH and APC genes is costly and not covered by most health insurance companies in Puerto Rico unless the patient has been diagnosed with CRC. All of the MAP patients were subjected to genetic testing due to the late-onset of the disease and the fact that Medicare covers these types of tests. The earlier onset of FAP and non-coverage of the genetic tests by their corresponding medical insurance companies resulted in relying on a clinical FAP diagnosis in 57.8% of the individuals in our study. Since identification of mutation-positive individuals results in early detection, improved cancer surveillance and decreased mortality, insurance coverage for genetic testing is imperative for the evaluation and management of individuals with GI polyposis and their at-risk relatives [2, 12]. More intensive efforts to educate healthcare providers, policy makers and the general public in the proper use of genetic testing for hereditary cancer syndromes are needed.

Acknowledgments

Footnotes

References