|Home | About | Journals | Submit | Contact Us | Français|
Mutations in the breast cancer 1, early onset (BRCA1) and breast cancer 2 (BRCA2) genes are responsible for the majority of hereditary breast cancers. Knowledge of the incidence and prevalence of BRCA mutations in a specific population or ethnic group is necessary to provide accurate genetic counseling for breast cancer patients and their families. However, these data have not been gathered in the population of Puerto Rico. We conducted a retrospective study of female breast cancer patients undergoing genetic testing for BRCA mutations in the highest volume breast surgery practices in San Juan, Puerto Rico. Data collection includes 3-generation family cancer history and results from complete BRCA sequencing. A total of 6 different deleterious mutations were observed, including 1 mutation in BRCA1 and 5 mutations in BRCA2. Three recurrent mutations (BRCA1 del exon1–2, BRCA2 4150G>T, and BRCA2 6027del4) account for over 70% of all the BRCA mutations observed in this study population. This study examines for the first time the characteristics of hereditary breast cancer in Puerto Rico, and assesses the accuracy of existing genetic risk assessment tools in that population. This data is expected to contribute to providing accurate and efficient tools for the clinical management of hereditary breast cancer in Puerto Rico.
Risk factors for breast cancer include age, geographic location, socioeconomic status, reproductive events, exogenous hormones, lifestyle factors, breast density and family history of breast and other cancers.(1–3) Approximately 5–10% of all breast cancers occur due to the inheritance of rare deleterious mutations in highly penetrant predisposition genes, such as BRCA1 and BRCA2. (4) Women carrying a deleterious mutation in the BRCA1 or BRCA2 genes are at a substantially higher lifetime risk of developing breast or ovarian cancer compared to the general population.(5–10) BRCA1 and BRCA2 carriers are also at reportedly at increased risk for developing other cancers, including prostate, colon, stomach, melanoma and pancreatic cancer.(5,8,10) It is estimated that mutations in BRCA1 and BRCA2 account for approximately 16% of the familial risk of breast cancer.(11)
In managing hereditary risk of breast cancer, increased surveillance, prophylactic surgery and chemopreventative interventions are among the options available to BRCA carriers.(12) Research has shown that those who are educated about their increased risk of breast cancer are more likely to engage in risk-reducing behaviors and early detection strategies such as monthly self-breast exam, physician visits, mammography and breast MRI screening.(13) Furthermore, the use of chemo- and or surgical- preventive measures, in addition to cancer surveillance, has a positive impact on survival in BRCA1/2 carriers.(14) In newly diagnosed cancer patients, BRCA testing was also shown to affect surgical decision-making.(15,16) The benefits of Cancer risk assessment and BRCA testing also extend to other family members. Cancer risk assessment can differentiate between those at high risk and those at average risk, thus relieving fears and anxieties about impending cancer, and sparing some individuals from unnecessary risk reducing options such as prophylactic surgeries.(15)
Cancer risk counseling and genetic testing serves to educate individuals about inherited predisposition to develop breast and other cancers. Throughout most of the United States, genetic counseling and testing is typically offered to breast cancer patients or their family members based on the risk of carrying a BRCA mutation estimated by personal and family history of cancer. This is considered the standard of care, according to national guidelines of multiple medical professional organizations, including the American Society of Clinical Oncology (17), the National Comprehensive Cancer Network (18), the U.S. Preventive Services Task Force (19), and the National Society of Genetic Counselors (20,21). Individuals at a high risk of carrying a BRCA mutations are recognized by a personal history of ovarian cancer, early onset (<50 years of age) breast cancer, and/or a family history of breast cancer, bilateral breast cancer, male breast cancer and/or ovarian cancer.(20) Ancestry has also been shown to modify the risk of carrying a BRCA mutation. For example, individuals of Ashkenazi Jewish ancestry are at significantly higher risk of carrying a BRCA mutation than those of European descent.(22) Several statistical models have been developed to predict the likelihood of identifying a deleterious BRCA mutation based on an individual's personal and family history. The BRCAPRO model uses a Bayesian approach, incorporating the autosomal dominant Mendelian characteristics of the genes, and the prevalence and penetrance of BRCA mutations based on published data.(23) The Myriad tables are derived from empiric data derived from individuals undergoing BRCA testing.(22) For a given family history, both models compute higher likelihood of identifying a BRCA mutation in individuals of Ashkenazi Jewish descent.(22,24) However, they fail to recognize that other ancestries may also be associated with differences in mutation frequencies.
According to the 2010 United States Census data, Hispanics make up 16.3% of the population, and accounted for 56% of the national growth between 2000 and 2010.(25) Breast cancer is the most commonly diagnosed cancer in Hispanic women (26), and it more likely to be diagnosed at a later stage in Hispanic women compared to non-Hispanic whites.(27–29) Even though part of these ethnic disparities have been attributed to socioeconomic factors (30), some differences in tumor stage and size at the time of diagnosis remain after accounting for these confounding factors.(30,31) Despite a lower incidence of breast cancer in US Hispanics, Hispanic women are more likely to die of breast cancer than non-Hispanic white women diagnosed at similar age and stage.(27) Puerto Ricans (8.1 million individuals living on the island and on the continent) make up 18% of the Hispanic population and comprise the second largest Hispanic subgroup in the country. Mitochondrial DNA analysis has demonstrated that the maternal ancestry of the population of Puerto Rico is mixed, with a contribution of 61.3% Native Americans, 27.2% Sub-Saharan African, and 11.5% West Eurasian.(32) This mixture makes this population unique from a genetic standpoint and possibly different from other Hispanic and non-Hispanic populations.
To date, there have been no published studies of the BRCA1 and BRCA2 genes in the Hispanic population of Puerto Rico. Genetic testing for hereditary cancer in Puerto Rico is limited to a few practices in the largest urban area of San Juan. The purpose of the present study was to investigate BRCA mutations in the population of Puerto Rico, and assess access to genetic testing services.
Participants were recruited from high-volume breast surgery clinics in San Juan, Puerto Rico, under Institutional Review Board. Eligible participants had undergone BRCA genetic testing through Myriad Genetics Laboratories (Salt Lake City, Utah). All participants were the first individuals in their family to undergo genetic testing, and all but two had a personal history of breast cancer (both participants with no personal history of cancer had a family history of early onset breast cancer with or without family history of ovarian cancer). Due to the limited availability of cancer genetic services in Puerto Rico, genetic testing was performed directly by the surgery clinic. Typically, a personal or family history of breast cancer before 50 years and/or a family history of ovarian cancer are the criteria used to identify families at risk of breast cancer.
Each participant met with study personnel to complete a family history of cancer (including cancer types and ages of diagnosis) for all first- and second- degree relatives. BRCA test results were obtained. Personal and family history information was used to estimate pre-test likelihood of mutation using the BRCAPRO (23) and Myriad II (22) models. The carrier risk was calculated independently for maternal and paternal sides of each family. Participants completed a questionnaire regarding medical insurance coverage and out-of-pocket expenses associated with the BRCA genetic analysis.
Table 1 presents the deleterious mutations and variants identified in this study population. Twenty-four unrelated women were screened as eligible based on having undergone BRCA genetic testing. Twenty-three of the 24 subjects who had undergone BRCA testing agreed to participate. Overall, 11 of the 23 (47.8%) participants were identified as carrying a BRCA1 or BRCA2 mutation. Two out of the 23 (8.7%) participants had been identified as carriers of a deleterious mutation in BRCA1, and 9 (39.1%) as carriers of a deleterious mutation in BRCA2. A total of 6 different deleterious mutations were identified, 1 in BRCA1 and 5 in BRCA2. Recurrent mutations included the BRCA1 del exon1–2 (n=2), BRCA2 4150G>T (n=4), and BRCA2 6027del4 (n=2). One participant had a variant classified as a polymorphism and 2 had variants of uncertain significance, both in individuals who were also found to carry a known deleterious mutation. Thus, mutations in BRCA2 represented 82% of the BRCA carriers, in contrast with 18% of the positive cases with mutations in BRCA1.
Both of the two breast cancers associated with BRCA1 mutations were diagnosed before 40 years of age in comparison to 22% and 25% in BRCA2 carriers and non-carriers, respectively. BRCA1 carriers were also more likely to develop a second primary cancer (50%) than BRCA2 carriers (14%) or non-carriers (0%). BRCA2 carriers were less likely to have a family history of ovarian cancer (11%) when compared to BRCA1 carriers (50%), but more likely to have a positive history of other cancers including prostate (33% of BRCA2 carrier families), pancreatic (11% of BRCA2 carrier families) and colorectal (11% of BRCA2 carrier families). The ages of diagnosis and family history characteristics of the study population are described in Table 2.
The median pre-test likelihood of carrying a BRCA mutation was estimated at 78.2% for BRCA1 carriers, 12.4% for BRCA2 carriers and 3.5% for non-carriers, under the BRCAPRO model (table 3). The Myriad II tables predicted median carrier risks of 23.5%, 10.5% and 6.8% for BRCA1, BRCA2 and non-carriers, respectively (table 3). The distribution and ranges of the pre-test likelihood of carrying a BRCA mutation as predicted by BRCAPRO and the Myriad tables are illustrated in Figure 1.
During the period of this retrospective study, over 50% of the genetic tests requested by healthcare practitioners were subsequently canceled due to a lack of insurance coverage. Although the majority of women who proceeded with testing paid less than $312 for a comprehensive BRCA1 and BRCA2 analysis, payment ranged up to the full retail price of the test by Myriad Laboratories ($3,120) (Data not shown).
A limited number of studies have looked at the BRCA mutation distribution, prevalence and penetrance in U.S. Hispanic populations.(33–35) To our knowledge, this is the first study describing BRCA testing and results in the Hispanic population of Puerto Rico. In the current study, the most frequent mutation encountered was the E1308X nonsense mutation in exon 11 of BRCA2, which was previously reported as a recurrent mutation in Latin America and the Caribbean.(36) This mutation has also been observed in Chilean families (37) and US Hispanics (33). We also noted recurrence of the BRCA1 exon 1–2 deletion in Puerto Rico, which has not previously been identified in Hispanics. The rarity of this mutation in the literature and the Breast Cancer Information Core (BIC) database may however be a technical bias, as this type of rearrangement is usually not detected by traditional sequencing or mutation screening approaches. The BRCA 1 H1421Y amino acid substitution of uncertain significance has previously been observed in Hispanics from the US.(33) None of the remaining deleterious mutations, polymorphisms or variants of uncertain significance we observed has previously been reported in other Hispanic populations.
A total of 6 different deleterious mutations were identified, 3 of them explaining more than three-fourths of the positive cases. The recurrent mutations were observed in apparently unrelated individuals, but a possible founder effect cannot be excluded. In Ashkenazi Jews, three recurrent founder mutations have been shown to be present at higher frequency and account for a significant proportion of the hereditary breast and ovarian cancer cases in this population.(38–41) Founder BRCA mutations have also been identified in families from Iceland (42–44), Poland (45, 46), Quebec (47, 48), Great-Britain (49), Netherlands (50, 51), Norway (52), Russia (53), and Sweden (54), among others. Based on those observations, a simplified screening panel has been developed for initial mutation analyses in some of those populations. From a cost-effective standpoint, the development of a similar test panel may benefit the population of Puerto Rico if these results can be replicated in a larger population. Interestingly, we did not find the BRCA1 187delAG mutation in our study population. This mutation is known to be one of the three most prevalent mutations in the Ashkenazi Jewish population (38, 40), but is also reportedly the most frequent mutation in several Hispanic populations including Hispanics from California (34) and Texas (33). It was also observed in Chilean (55, 56), but not in Colombian (57) or Mexican (58) BRCA carriers. Its origin in the Hispanic population may be traced back to Spain as it was also detected in non-Jewish families from Spain.(59–62) The presence of few recurrent BRCA mutations and the absence of mutations commonly observed in other Hispanic populations may be explained by the genetic bottleneck effect that are expected to have occurred as a result of the colonization of the New World by European and African ancestral populations. In the US, the term `Hispanic' refers mostly a heterogeneous group that shares a common culture and language. While most Hispanic populations are expected to be a mix of 3 ancestral populations (African, European and Native American), the relative proportion of each ancestral genetic background has been shown to vary within and across Hispanic populations.(63) Previous studies using Ancestry Informative Markers (AIMs) have shown that Puerto Ricans carry more European and African ancestry, but significantly less Native American ancestry when compared to Mexicans.(64)
There are limitations to the current study design and recruitment strategy: the small sample size, retrospective ascertainment and recruitment, and testing limited almost exclusively to women with a personal history of breast cancer. Previous admixture analysis has shown that ancestral genetic background is influenced by geography and socio-economic status in Puerto Rico.(65) The recruitment of patients that have had access to genetic testing is likely to have resulted in the selection of subjects of higher socioeconomic status, and consequently stronger European ancestry. In addition, recruitment was concentrated in a few offices in the San Juan metropolitan area, which is home to half of the population of the island. As a result, the ancestral background of the study sample may not be representative of that of the island. Additional samples from other areas of Puerto Rico will be necessary to generalize these findings. The small sample size of this study also limits the conclusion to be drawn. First, because of the limited number of participants, it is unlikely that we have detected all the mutations present in the population. More accurate estimates of the mutation frequencies, assessments of the performance of carrier risk prediction models, and statistical comparisons of the family history characteristics in BRCA1, BRCA2 and non-BRCA carriers will also require larger sample sizes. The BRCA carrier frequency was 47.8% in this sample. However, this number may reflect a tendency of the medical professionals to recommend testing in higher-risk women and/or an increased likelihood of insurance coverage in higher risk women. The full description of the mutation spectrum and accurate evaluation of BRCA carrier frequencies in Puerto Rico will require the analysis of a larger, unselected group of women.
Overall, the BRCA1 mutation carriers obtained higher carrier prediction scores when compared to the BRCA2 carriers. This might be a consequence of the higher ages at the time of breast cancer diagnosis observed in BRCA2 carriers. It is noteworthy that one of the BRCA2 cases produced a carrier risk of less than 5% for both prediction models (Myriad II and BRCAPRO). This same case was diagnosed at 54 years of age and would not meet the NCCN guidelines for genetic testing. Two additional BRCA2 carriers cases were predicted to have a risk below 10% using both models, but would have qualified for testing under the NCCN guidelines due to their young ages of diagnosis (42 and 33 years of age), even in absence of family history. Therefore, according to the currently available models and national guidelines for making genetic testing decision, 1 out of 9 BRCA2 carriers would have gone untested in this sample. However, our sample size is not sufficient to draw general conclusion about the specificity and sensitivity of the models and national guidelines in this population. The BRCAPRO and Myriad II models have been developed based mostly on empiric data from populations of Ashkenazi Jewish and European descent.(22–24) Assuming that mutation rates may vary across different populations, the accuracy of the prediction models would need to be validated in other populations. BRCAPRO was shown to perform equally well in African American families as in white and Jewish families.(66) In Asian Americans, Myriad II and BRCAPRO underestimated BRCA carriers by two-fold.(67) In Hispanic populations, the validation of carrier risk estimation models has produced mixed results. A study of Hispanics from Texas reported a good discrimination of BRCA carrier with BRCAPRO (33), whereas in the San Francisco Bay area this model showed the poorest performance in Hispanics (68). Although these inconsistencies may result from methodological differences between studies, they may also reflect variations in BRCA mutation frequencies and penetrance among the different Hispanic subgroups. Further studies will be required to fully validate the specificity and sensitivity of existing models for risk estimates in non-white women. (69)
For women receiving a positive BRCA test result, several options of breast cancer risk management are available, including increased surveillance, chemoprevention and prophylactic surgery.(12) The limited resources available in genetic counseling impact the current cancer management model in Puerto Rico, and the majority of BRCA carriers are identified through surgery or oncology practices. As a result, BRCA carriers are often identified after receiving a cancer diagnosis and the genetic test results may not always be available at the moment of making treatment decision. Previous studies have shown that hereditary cancer counseling and testing is likely to impact surgical treatment decisions.(15, 16) Furthermore, preventive strategies such as tamoxifen and prophylactic surgeries have been shown to have a greater impact on survival and quality-adjusted survival when initiated at earlier ages.(14) In Puerto Rico, it was found that having a health care provider recommendation was an important factor in the decision to undergo breast cancer genetic testing.(70) The establishment of cancer genetic education programs for community clinicians can impact knowledge and practice.(71) Such programs could allow identifying individuals at high risk of hereditary cancer earlier and more efficiently in Puerto Rico. The National Cancer Institute Cancer Genetics Services Directory lists only one cancer genetics professional for Puerto Rico. Even though medical geneticists that are not specialized in cancer may also facilitate genetic counseling and testing, these services are also limited in Puerto Rico. Given the high proportions of carriers identified in this sample, more cancer genetic specialists will be required to serve this potentially high-risk population.
In conclusion, we presented the first report identifying BRCA1 and BRCA2 mutations in Puerto Rico. Our data and that of others have shown that the BRCA mutation prevalence and spectrum may vary dramatically between different Hispanic and Caribbean populations. In addition, there appear to be significant barriers to BRCA genetic testing in this population. The majority of the tests sent to Myriad Genetics have been canceled due to lack of insurance coverage. In a study exploring women's interests and attitudes towards genetic counseling and testing for genetic testing, Puerto Rican women expressed a high overall interest for hereditary breast cancer genetic testing but perceived the associated cost as the main barrier.(70) If the recurrence of a subset of mutations is confirmed in Puerto Rico, a cost-effective screening approach prior to the comprehensive sequence analysis may be feasible in this population. This may contribute to improving accessibility to genetic testing. However, in order to meet the US standards of care in Puerto Rico, several issues will have to be addressed.
The authors are grateful to all the women who participated in this study. We also want to thank the clinic teams at Auxilio Mutuo and at the Advanced Surgical Institute in San Juan, Puerto Rico. This work was supported by a pilot project grant of the National Institute of Health, National Cancer Institute U56 Ponce School of Medicine/H. Lee Moffitt Cancer Center Partnership (1U56 CA 126379) to J. Dutil and R. Sutphen.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.