GENETIC TESTING AND MOLECULAR BIOMARKERS Volume 17, Number 10, 2013 ª Mary Ann Liebert, Inc. Pp. 763–767 DOI: 10.1089/gtmb.2013.0105
Detection of Carriers in the Ashkenazi Jewish Population: An Objective Comparison of High-Throughput Genotyping Versus Gene-by-Gene Testing Susan Klugman,1 Nicole Schreiber-Agus,2,3 Shivani Nazareth,4 and Eric A. Evans 4
Background: High-throughput genotyping allows rapid identification of targeted mutations at a fraction of the cost of current gene-by-gene testing methodologies. An objective comparison of the two methodologies allows providers to assess the clinical validity/utility of high-throughput carrier screening and establish a comfort level with new genomic technologies. Aim: To verify that high-throughput genotyping accurately determines patient carrier status, DNA samples from previously identified carriers (n = 31) of Ashkenazi Jewish genetic diseases were anonymized and submitted for retesting by high-throughput genotyping. Results: The results were 100% concordant (95% CI: 0.998–1), demonstrating that high-throughput genotyping assays accurately identify carriers of targeted mutations in the Ashkenazi Jewish population. In addition, carrier status for diseases and mutations not previously tested was uncovered using the high-throughput assay. Conclusions: High-throughput genotyping is a cost-effective and clinically valid approach to carrier screening. The use of a broader screen for Ashkenazi Jewish individuals increases the detection of carriers in this population.
Introduction
C
arrier screening for Tay-Sachs disease in the Ashkenazi Jewish population began in the 1970s and has expanded over the past few decades to include over a dozen autosomal recessive conditions. Cultural acceptance of screening for a large number of diseases, even when the carrier frequency is lower than the widely accepted 1/100 threshold, has made the Ashkenazi Jewish population a model community for implementation of carrier screening as a means of risk reduction and disease prevention (Scott et al., 2010). Traditionally, single-gene assays for targeted mutations have been performed at a rate of a few hundred dollars per test. This additive cost translates to thousands of healthcare dollars spent per Ashkenazi Jewish panel ordered, depending on the number of conditions included in the panel. Many insurance companies will not pay for these tests unless a woman is already pregnant, thereby decreasing the options for disease prevention. Screening programs for Tay-Sachs and the like have been successful, in part, because of philanthropic contributions from the Jewish community, as well as a common cultural goal to educate couples, encourage screening,
and ultimately prevent the transmission of disease to the next generation (Laberge et al., 2010). For example, the Program for Jewish Genetic Health of Yeshiva University/Einstein of New York and the Jewish Genetic Disease Consortium engage philanthropists to raise funds for community-based education and screening events (B. Lander and S. Ungerleider, pers. comm). Without such funding, cost is a barrier to test uptake and to expansion of the panel to include diseases that meet screening criteria in the Ashkenazi Jewish population (Little et al., 2010). Providers must be cognizant of the growing number of screening tests available and their potential relevance (Klugman and Gross, 2010; Srinivasan et al., 2010). Conditions appropriate for inclusion in a screening test should meet the standard criteria, such as significant morbidity/moribundity in homozygous affected offspring, availability of clinically and technically valid testing methodologies, cost-effective screening, availability of prenatal diagnosis and/or other reproductive options, process for consent and access to experienced professionals such as genetic counselors (National Human Genome Research Institute, 2008). Two factors suggest examination of a more sustainable approach to Ashkenazi Jewish screening: the decreasing reliability of racial and
1 Department of Obstetrics & Gynecology and Women’s Health, Albert Einstein College of Medicine of Yeshiva University, Program for Jewish Genetic Health, Montefiore Medical Center, Bronx, New York. 2 Human Genetics Lab, Jacobi Medical Center, Bronx, New York. 3 Department of Genetics, Albert Einstein College of Medicine; Program for Jewish Genetic Health, Bronx, New York. 4 Counsyl, Inc., South San Francisco, California.
763
764
KLUGMAN ET AL.
ethnic categorizations, and the recent ability to screen for a larger number of conditions with no concomitant increase in cost to patients (Lazarin et al., 2013). In this article, we argue that the availability of less expensive high-throughput multiplex testing methodologies should be evaluated as a costeffective replacement for the current gene-by-gene screening model. Adoption of analytically and clinically validated multiplex platforms would help the Ashkenazi Jewish community take advantage of screening for a larger number of serious genetic disorders at a reduced cost. Materials and Methods DNA samples from 31 patients who had previously undergone testing at the Human Genetics Laboratory at the Jacobi Medical Center (Luminex panel) between 2008 and 2011 were selected for retesting at Counsyl. A subset of the samples from 2011 had additional disease testing at the Mount Sinai Genetic Testing Laboratory. All 31 participants were self-identified as Ashkenazi Jewish. Test requisitions and patient consents for the Jacobi and Mount Sinai Laboratories were New York State Department of Health approved. Each individual undergoing testing was given the option to (1) let the laboratory retain the sample with the understanding that it may be anonymized and used for population research with all identifiers removed or (2)
alternatively request that the specimen be destroyed once testing is done or within 60 days. Only those samples for which the patient consented to the use of their DNA, in research, were retained. All patient identifiers were removed and no DNA sample could be linked back to any individual. Human subject research protocols were approved by the Einstein Institutional Review Board under number 2012–0877. Genomic DNA was originally isolated from peripheral blood using the Gentra Puregene Blood Kit and stored at - 20C for up to 4 years. Selected samples were numbered and aliquots containing 2000 ng of DNA in 50 mL of elution buffer were plated before shipment to Counsyl for testing. Test requisitions were generated for each sample, which were subsequently accessioned as standard clinical samples. All samples met automated DNA quality control acceptability criteria, and were introduced into the high-throughput genotyping workflow at Counsyl in parallel with ordinary patient samples. Each sample was tested for the mutations listed in Table 1. Testing for HEXA pseudodeficiency alleles was available upon request and was included in this analysis. Genotype calls were generated by 2D Gaussian mixture model classifiers trained for each variant and dynamically normalized to account for signal intensity variation between batches. Positive test results were confirmed by retesting before test result reporting, and all samples met quality control
Table 1. Diseases and Mutations Analyzed Using High-Throughput Genotyping Disease (gene) Bloom syndrome (BLM) Canavan disease (ASPA) Cystic fibrosis (CFTR)
Dihydrolipoamide dehydrogenase deficiency (DLD) Familial dysautonomia (IKBKAP) Familial hyperinsulinism (ABCC8) Fanconi anemia group C (FANCC) Gaucher disease type I (GBA) Glycogen storage disease type 1A (G6PC) Joubert syndrome (TMEM216) Maple syrup urine disease type 1B (BCKDHB) Mucolipidosis type IV (MCOLN1) Niemann-pick disease type A (SMPD1) Nemaline myopathy (NEB) Hex A deficiency (including Tay-Sachs disease) (HEXA) Usher syndrome type IF (PCDH15) Usher syndrome type III (CLRN1)
Variants tested 2281del6ins7 E285A, Y231X, A305E, IVS2-2A > G G85E, R117H, R334W, R347P, A455E, G542X, G551D, R553X, R560T, R1162X, W1282X, N1303K, F508del, I507del, 2184delA, 3659delC, 621 + 1G > T, 711 + 1G > T, 1717-1G > A, 1898 + 1G > A, 2789 + 5G > A, 3120 + 1G > A, 3849 + 10kbC > T, E60X, R75X, E92X, Y122X, G178R, R347H, Q493X, V520F, S549N, P574H, M1101K, D1152H, 2143delT, 394delTT, 444delA, 1078delT, 3876delA, 3905insT, 1812-1G > A, 3272-26A > G, 2183AA > G, S549R(A > C), R117C, L206W, G330X, T338I, R352Q, S364P, G480C, C524X, S549R(T > G), Q552X, A559T, G622D, R709X, K710X, R764X, Q890X, R1066C, W1089X, Y1092X, R1158X, S1196X, W1204X(c.3611G > A), Q1238X, S1251N, S1255X, 3199del6, 574delA, 663delT, 935delA, 936delTA, 1677delTA, 1949del84, 2043delG, 2055del9 > A, 2108delA, 3171delC, 3667del4, 3791delC, 1288insTA, 2184insA, 2307insA, 2869insG, 296 + 12T > C, 405 + 1G > A, 405 + 3A > C, 406-1G > A, 711 + 5G > A, 712-1G > T, 1898 + 1G > T, 1898 + 5G > T, 3120G > A, 457TAT > G, 3849 + 4A > G, Q359K/T360K 105insA, G229C IVS20 + 6T > C, R696P F1388del, V187D, 3992-9G > A IVS4 + 4A > T, 322delG, R548X N370S, L444P, 1035insG, IVS2 + 1G > A, V394L, R496H, D409H, D409V, R463C, R463H R83C, Q347X, Q27fsdelC, 459insTA, R83H, G188R, Q242X 35G > T R183P, G278S, E322X 511_6944del, IVS3-2A > G fsP330, L302P, R496L, p.R608del R2478_D2512del 1278insTATC, IVS12 + 1G > C, G269S, IVS9 + 1G > A, R178H, IVS7 + 1G > A, 7.6kb del, G250D, R170W R245X N48K
COMPARATIVE PLATFORM EVALUATION FOR CARRIER SCREENING acceptability criteria according to Counsyl standard operating procedures. Genotyping test results were transcribed and reported to the Jacobi Medical Center Human Genetics Laboratory for comparison to earlier test results. Results Thirty-one samples were analyzed using a high-throughput genotype assay for a total of 17 diseases and 153 associated mutations (Table 1). Samples were selected on the basis of mutational diversity and initially represented 38 known mutations. The samples were collectively positive for 35 independent mutations, including pseudodeficient Tay-Sachs alleles. All 38 carrier states were confirmed, and 3 additional carrier states were detected owing to a greater number of diseases and mutations included in the high-throughput platform (Table 2). The three additional carrier states were identified in samples 2, 3, and 18, for Maple Syrup Urine Disease Type 1B, Gaucher disease, and NEB-related nemaline myopathy, respectively. In total, 2207 genotypes were confirmed by high-throughput testing with an average accuracy of 100% (95% CI: 0.998–1). Sample 31 was negative for all mutations analyzed, both initially and with the high-throughput analysis. However, this patient was shown to be a carrier for Tay-Sachs using biochemical enzyme analysis. The initial DNA testing confirmed that the patient was negative for 7 common Tay-Sachs mutations tested by the Luminex panel: G269S, 1278insTATC, IVS12 + 1G > C, IVS9 + 1G > A, 7.6kb del, R247W, and R249W. The Counsyl assay determined that the patient was negative for nine common Tay-Sachs mutations (and also the two pseudodeficiency mutations): R178H, G269S, 1278insTATC, IVS12 + 1G > C, IVS7 + 1G > A, IVS9 + 1G > A, 7.6kb del, G250D, and R170W. These two DNA panels account for 11 independent Tay-Sachs mutations. Discussion An objective comparison of testing platforms is a useful tool for clinicians to assess the utility and reliability of such tests for adoption in their everyday practice. Our data indicate that high-throughput genotyping is an accurate and reliable method for determining patient carrier states in the Ashkenazi Jewish population. Beyond confirming all known carrier states, the high-throughput platform identified three additional carrier states in the patient cohort. The ease with which diseases/mutations could be added to or subtracted from the multiplex platform at no increased cost makes it an appealing solution for expanding disease panels. Professional guidelines for carrier screening in the Ashkenazi Jewish population recommend testing for fewer disorders than included in our study. This is partially cost driven. The American College of Medical Genetics (ACMG) recommends screening for nine disorders, including the four recommended by the American Congress of Obstetricians and Gynecologists (ACOG). In addition to compliance with the basic tenets of screening, the ACMG guidelines advise that the tested disease should have an allele frequency of greater than or equal to 1% or a detection rate of 90% or higher (Gross et al., 2008). The Ashkenazi Jewish community, on the other hand, has demonstrated an acceptance and desire for screening for at least 16 conditions (Scott et al., 2010). This includes carrier screening for the Usher Syndrome Type 1, despite the carrier frequency and detection rate falling well below the ACMG threshold.
765
Looking ahead, another reason to implement expanded carrier screening in the Ashkenazi Jewish population is the increasing ethnic admixture within the community as a result of interfaith marriages, adoption, and changes in patterns of self-identification (Bray et al., 2010). Genetic drift and migration, in combination with cultural patterns, have contributed to the current genetic makeup of the Ashkenazi population. It has become increasingly evident that self-reported ethnic identification is subject to the fallacies of social experience that shape patient preferences and is consequently a substandard measure of genetic risk (Lee et al., 2008; Ross, 2012). Along this line, of 23,453 patients who recently underwent expanded carrier testing, 26.3% of carriers (total n = 76) of the Ashkenazi Jewish founder mutation in familial dysautonomia had not reported any known Jewish ancestry (Lazarin et al., 2013). This statistic could be indicative of a pervasive lack of awareness about personal Jewish ancestry, and/or a higher prevalence of the founder mutation outside of the Ashkenazi Jewish population. Finally, it is known that targeted mutation analysis will always miss carriers, and residual risk must be discussed with patients (Shore et al., 1992). For diseases like Tay-Sachs or cystic fibrosis, hundreds of mutations in the respective HEXA or CFTR genes have been identified. Case in point, the patient represented by sample 31 in this study is likely a carrier of a rare Tay-Sachs mutation, similar to what has been described previously in the literature (Schneider et al., 2009; Nakagawa et al., 2012). Fortunately, this patient was picked up as a carrier by the enzyme assay, which remains a highly sensitive test and should be offered in conjunction with targeted DNA analysis for optimal Tay-Sachs disease carrier detection across ethnicities (Schneider et al., 2009). In part due to the dramatically declining rate of costs-per-megabase for DNA sequencing technologies (National Human Genome Research Institute, 2012), carrier screening by targeted mutation analysis may soon fall out of favor to be replaced by targeted sequencing panels, and eventually, even by whole exome/genome approaches. Even before large-scale population screening by sequencing is feasible, it may be worthwhile to consider sequencing as a reflex test for the noncarrier partner in carrier/ noncarrier couples to better assess residual risk. While sequencing based technologies will allow us to identify these additional carriers and further reduce the risk of the noncarriers by eliminating biases based on ethnicity and mutation frequency, these technologies are not without challenges and bottlenecks. The genetics community must use a discerning approach to contend with a potentially overwhelming amount of clinically vague information, along with the ethical issues related to communication of results to patients and their family members (Tarini and Goldenberg, 2012). That said, practitioners have historically dealt with technological advances and the subsequent disclosure of sensitive information within all fields of medicine; genetics is not necessarily unique in this regard (Evans, 2011). For now, as we transition toward widespread adoption of cost-effective genomic technologies, high-throughput genotyping for previously reported clinically relevant mutations is an acceptable and preferred alternative to carrier screening through current gene-by-gene methodologies. Acknowledgments We would like to thank Dr. Jose Carlos Ferreira and Jie Zhan for their contributions in the Jacobi genetic testing laboratory.
766
2009 2009 2009 2009
2009 2009
2009 2009 2009 2009
2009 2009
2010 2010 2010 2010 2010 2010 2010 2011 2011 2011
2011 2011 2009
7 8 9 10
11 12
13 14 15 16
17 18
19 20 21 22 23 24 25 26 27 28
29 30 31
SMPD1:p.Leu302Pro heterozygote HEXA:c.1421 + 1G > C heterozygote CFTR:p.Ala455Glu heterozygote CFTR:p.Asn1303Lys heterozygote HEXA:p.Gly269Ser heterozygote IKBKAP:p.Arg696Pro heterozygote CFTR:p.Gly542X heterozygote CFTR:c.262_263delTT heterozygote BCKDHB:p.Gly278Ser heterozygote CFTR:p.Trp1282X heterozygote FANCC:c.456 + 4A > T heterozygote IKBKAP:c.2204 + 6T > C heterozygote CFTR:c.1585-1G > A heterozygote DLD:p.Gly229Cys heterozygote No disease-causing mutations detected.a
SMPD1:p.Arg608del heterozygote HEXA:p.Arg247Trp FANCC:c.456 + 4A > T heterozygote HEXA:p.Arg249Trp CFTR:p.Arg117His heterozygote ASPA:p.Ala305Glu heterozygote HEXA:c.1274_1277dupTATC heterozygote SMPD1:p.Arg496Leu heterozygote GBA:p.Val433Leu heterozygote CFTR:c.3717 + 10kbC > T heterozygote GBA:c.115 + 1G > A heterozygote MCOLN1:c.406-2A > G heterozygote HEXA:c.1274_1277dupTATC heterozygote GBA:c.84dupG heterozygote GBA:p.Arg535His heterozygote ASPA:p.Glu285Ala heterozygote SMPD1{NM_000543.3}:c.996delC heterozygote HEXA:c.1073 + 1G > A heterozygote ASPA:p.Tyr231X heterozygote CFTR:p.Phe508del heterozygote BLM:c.2207_2212delATCTGAinsTAGATTC heterozygote MCOLN1:g.190619_197051del heterozygote GBA:p.Asn409Ser homozygote mutant
Known carrier states
Patient is a Tay-Sachs carrier via biochemical analysis.
2008 2008 2008
4 5 6
a
2008 2008 2008
1 2 3
Sample
Year collected
SMPD1:p.Leu302Pro heterozygote HEXA:c.1421 + 1G > C heterozygote CFTR:p.Ala455Glu heterozygote CFTR:p.Asn1303Lys heterozygote HEXA:p.Gly269Ser heterozygote IKBKAP:p.Arg696Pro heterozygote CFTR:p.Gly542X heterozygote CFTR:c.262_263delTT heterozygote BCKDHB:p.Gly278Ser heterozygote CFTR:p.Trp1282X heterozygote FANCC:c.456 + 4A > T heterozygote IKBKAP:c.2204 + 6T > C heterozygote CFTR:c.1585-1G > A heterozygote DLD:p.Gly229Cys heterozygote No disease-causing mutations detected.
SMPD1:p.Arg608del heterozygote HEXA:p.Arg247Trp FANCC:c.456 + 4A > T heterozygote HEXA:p.Arg249Trp CFTR:p.Arg117His heterozygote ASPA:p.Ala305Glu heterozygote HEXA:c.1274_1277dupTATC heterozygote SMPD1:p.Arg496Leu heterozygote GBA:p.Val433Leu heterozygote CFTR:c.3717 + 10kbC > T heterozygote GBA:c.115 + 1G > A heterozygote MCOLN1:c.406-2A > G heterozygote HEXA:c.1274_1277dupTATC heterozygote GBA:c.84dupG heterozygote GBA:p.Arg535His heterozygote ASPA:p.Glu285Ala heterozygote SMPD1{NM_000543.3}:c.996delC heterozygote HEXA:c.1073 + 1G > A heterozygote ASPA:p.Tyr231X heterozygote CFTR:p.Phe508del heterozygote BLM:c.2207_2212delATCTGAinsTAGATTC heterozygote MCOLN1:g.190619_197051del heterozygote GBA:p.Asn409Ser homozygote mutant
High-throughput results
Table 2. Data Comparison
83 83 69
69 69 69 69 69 69 69 81 83 83
69 69
69 69 69 69
69 69
69 69 69 69
69 69 69
69 69 69
Original mutations tested
1 1 0
1 1 1 1 1 1 1 1 1 3
1 2
1 1 1 2
1 2
1 1 1 2
1 1 2
1 1 2
Mutations confirmed
N/A N/A N/A
N/A NEB:p.R2478_D2512del heterozygote N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
N/A N/A N/A N/A
N/A N/A
N/A N/A N/A N/A
N/A BCKDHB:p.Arg183Pro GBA:p.Arg502Cys heterozygote N/A N/A N/A
Additional mutations identified
COMPARATIVE PLATFORM EVALUATION FOR CARRIER SCREENING Author Disclosure Statement E.E. and S.N. are employees of Counsyl. The samples were selected from patients participating in screening through the Program for Jewish Genetic Health at Einstein/Yeshiva University. Collaboration with S.K. and N.S.A. does not represent an endorsement by their respective institutions for any specific commercial laboratory, but instead for the 100% concordance of the newer methodology of high-throughput expanded carrier screening with the older method of gene-bygene testing. References Bray SM, Mulle JG, Dodd AF, et al. (2010) Signatures of founder effects, admixture, and selection in the Ashkenazi Jewish population. Proc Natl Acad Sci U S A 107:16222–16227. Evans JP (2011) Looking ahead, looking behind. Genet Med 13:177–178. Gross SJ, Pletcher BA, Monaghan KG (2008) Professional Practice and Guidelines Committee. Carrier screening in individuals of Ashkenazi Jewish descent. Genet Med 10:54–56. Klugman S, Gross SJ (2010) Ashkenazi Jewish screening in the twenty-first century. Obstet Gynecol Clin North Am 37:37–46. Laberge AM, Watts C, Porter K, Burke W (2010) Assessing the potential success of cystic fibrosis carrier screening: lessons learned from Tay-Sachs disease and beta-thalassemia. Public Health Genomics 13:310–319. Lazarin G, Haque I, Nazareth S, et al. (2013) An empirical estimate of carrier frequencies for 400 + causal Mendelian variants: results from an ethnically diverse clinical sample of 23,453 individuals. Genet Med 15:178–186. Lee SS, Mountain J, Koenig B, et al. (2008) The ethics of characterizing difference: guiding principles on using racial categories in human genetics. Genome Biol 9:404. Little SE, Janakiraman V, Kaimal A, et al. (2010) The cost-effectiveness of prenatal screening for spinal muscular atrophy. Am J Obstet Gynecol 202:253.e1–253.e17. Nakagawa S, Zhan J, Sun W, et al. (2012) Platelet hexosaminidase A enzyme assay effectively detects carriers missed by targeted DNA mutation analysis. JIMD Rep 6:1–6.
767
National Human Genome Research Institute (2008) Populationbased carrier screening for single gene disorders: LESSONS learned and new opportunities. Rockville, MD: February 6–7, 2008. Available at www.genome.gov/27026048 (Accessed August 3, 2013). National Human Genome Research Institute (2012) DNA sequencing costs: data from the NHGRI Large-Scale Genome Sequencing Program. Available at www.genome.gov/ sequencing costs (Accessed August 3, 2013). Ross LF (2012) A re-examination of the use of ethnicity in prenatal carrier screening. Am J Med Genet 158A:19–23. Schneider A, Nakagawa S, Keep R, et al. (2009) Populationbased Tay-Sachs screening among Ashkenazi Jewish young adults in the 21st century: hexosaminidase A enzyme assay is essential for accurate testing. Am J Med Genet Part A 149A:2444–2447. Scott SA, Edelmann L, Liu L, et al. (2010) Experience with carrier screening and prenatal diagnosis for 16 Ashkenazi Jewish genetic diseases. Hum Mutat 31:1240–1250. Shore S, Tomczak J, Grebner EE, Myerowitz R (1992) An unusual genotype in an Ashkenazi Jewish patient with Tay-Sachs disease. Hum Mutat 1:486–490. Srinivasan BS, Evans EA, Flannick J, et al. (2010) A universal carrier test for the long tail of Mendelian disease. Reprod Biomed Online 21:537–551. Tarini BA, Goldenberg AJ (2012) Ethical issues with newborn screening in the genomics era. Annu Rev Genom Hum Genet 13:381–393.
Address correspondence to: Susan Klugman, MD Division of Reproductive Genetics Department of Obstetrics & Gynecology and Women’s Health Albert Einstein College of Medicine of Yeshiva University Program for Jewish Genetic Health Montefiore Medical Center 1695 Eastchester Rd., Suite 301 Bronx, NY 10461 E-mail: [email protected]
Detection of Carriers in the Ashkenazi Jewish Population: An Objective Comparison of High-Throughput Genotyping Versus Gene-by-Gene Testing Susan Klugman,1 Nicole Schreiber-Agus,2,3 Shivani Nazareth,4 and Eric A. Evans 4
Background: High-throughput genotyping allows rapid identification of targeted mutations at a fraction of the cost of current gene-by-gene testing methodologies. An objective comparison of the two methodologies allows providers to assess the clinical validity/utility of high-throughput carrier screening and establish a comfort level with new genomic technologies. Aim: To verify that high-throughput genotyping accurately determines patient carrier status, DNA samples from previously identified carriers (n = 31) of Ashkenazi Jewish genetic diseases were anonymized and submitted for retesting by high-throughput genotyping. Results: The results were 100% concordant (95% CI: 0.998–1), demonstrating that high-throughput genotyping assays accurately identify carriers of targeted mutations in the Ashkenazi Jewish population. In addition, carrier status for diseases and mutations not previously tested was uncovered using the high-throughput assay. Conclusions: High-throughput genotyping is a cost-effective and clinically valid approach to carrier screening. The use of a broader screen for Ashkenazi Jewish individuals increases the detection of carriers in this population.
Introduction
C
arrier screening for Tay-Sachs disease in the Ashkenazi Jewish population began in the 1970s and has expanded over the past few decades to include over a dozen autosomal recessive conditions. Cultural acceptance of screening for a large number of diseases, even when the carrier frequency is lower than the widely accepted 1/100 threshold, has made the Ashkenazi Jewish population a model community for implementation of carrier screening as a means of risk reduction and disease prevention (Scott et al., 2010). Traditionally, single-gene assays for targeted mutations have been performed at a rate of a few hundred dollars per test. This additive cost translates to thousands of healthcare dollars spent per Ashkenazi Jewish panel ordered, depending on the number of conditions included in the panel. Many insurance companies will not pay for these tests unless a woman is already pregnant, thereby decreasing the options for disease prevention. Screening programs for Tay-Sachs and the like have been successful, in part, because of philanthropic contributions from the Jewish community, as well as a common cultural goal to educate couples, encourage screening,
and ultimately prevent the transmission of disease to the next generation (Laberge et al., 2010). For example, the Program for Jewish Genetic Health of Yeshiva University/Einstein of New York and the Jewish Genetic Disease Consortium engage philanthropists to raise funds for community-based education and screening events (B. Lander and S. Ungerleider, pers. comm). Without such funding, cost is a barrier to test uptake and to expansion of the panel to include diseases that meet screening criteria in the Ashkenazi Jewish population (Little et al., 2010). Providers must be cognizant of the growing number of screening tests available and their potential relevance (Klugman and Gross, 2010; Srinivasan et al., 2010). Conditions appropriate for inclusion in a screening test should meet the standard criteria, such as significant morbidity/moribundity in homozygous affected offspring, availability of clinically and technically valid testing methodologies, cost-effective screening, availability of prenatal diagnosis and/or other reproductive options, process for consent and access to experienced professionals such as genetic counselors (National Human Genome Research Institute, 2008). Two factors suggest examination of a more sustainable approach to Ashkenazi Jewish screening: the decreasing reliability of racial and
1 Department of Obstetrics & Gynecology and Women’s Health, Albert Einstein College of Medicine of Yeshiva University, Program for Jewish Genetic Health, Montefiore Medical Center, Bronx, New York. 2 Human Genetics Lab, Jacobi Medical Center, Bronx, New York. 3 Department of Genetics, Albert Einstein College of Medicine; Program for Jewish Genetic Health, Bronx, New York. 4 Counsyl, Inc., South San Francisco, California.
763
764
KLUGMAN ET AL.
ethnic categorizations, and the recent ability to screen for a larger number of conditions with no concomitant increase in cost to patients (Lazarin et al., 2013). In this article, we argue that the availability of less expensive high-throughput multiplex testing methodologies should be evaluated as a costeffective replacement for the current gene-by-gene screening model. Adoption of analytically and clinically validated multiplex platforms would help the Ashkenazi Jewish community take advantage of screening for a larger number of serious genetic disorders at a reduced cost. Materials and Methods DNA samples from 31 patients who had previously undergone testing at the Human Genetics Laboratory at the Jacobi Medical Center (Luminex panel) between 2008 and 2011 were selected for retesting at Counsyl. A subset of the samples from 2011 had additional disease testing at the Mount Sinai Genetic Testing Laboratory. All 31 participants were self-identified as Ashkenazi Jewish. Test requisitions and patient consents for the Jacobi and Mount Sinai Laboratories were New York State Department of Health approved. Each individual undergoing testing was given the option to (1) let the laboratory retain the sample with the understanding that it may be anonymized and used for population research with all identifiers removed or (2)
alternatively request that the specimen be destroyed once testing is done or within 60 days. Only those samples for which the patient consented to the use of their DNA, in research, were retained. All patient identifiers were removed and no DNA sample could be linked back to any individual. Human subject research protocols were approved by the Einstein Institutional Review Board under number 2012–0877. Genomic DNA was originally isolated from peripheral blood using the Gentra Puregene Blood Kit and stored at - 20C for up to 4 years. Selected samples were numbered and aliquots containing 2000 ng of DNA in 50 mL of elution buffer were plated before shipment to Counsyl for testing. Test requisitions were generated for each sample, which were subsequently accessioned as standard clinical samples. All samples met automated DNA quality control acceptability criteria, and were introduced into the high-throughput genotyping workflow at Counsyl in parallel with ordinary patient samples. Each sample was tested for the mutations listed in Table 1. Testing for HEXA pseudodeficiency alleles was available upon request and was included in this analysis. Genotype calls were generated by 2D Gaussian mixture model classifiers trained for each variant and dynamically normalized to account for signal intensity variation between batches. Positive test results were confirmed by retesting before test result reporting, and all samples met quality control
Table 1. Diseases and Mutations Analyzed Using High-Throughput Genotyping Disease (gene) Bloom syndrome (BLM) Canavan disease (ASPA) Cystic fibrosis (CFTR)
Dihydrolipoamide dehydrogenase deficiency (DLD) Familial dysautonomia (IKBKAP) Familial hyperinsulinism (ABCC8) Fanconi anemia group C (FANCC) Gaucher disease type I (GBA) Glycogen storage disease type 1A (G6PC) Joubert syndrome (TMEM216) Maple syrup urine disease type 1B (BCKDHB) Mucolipidosis type IV (MCOLN1) Niemann-pick disease type A (SMPD1) Nemaline myopathy (NEB) Hex A deficiency (including Tay-Sachs disease) (HEXA) Usher syndrome type IF (PCDH15) Usher syndrome type III (CLRN1)
Variants tested 2281del6ins7 E285A, Y231X, A305E, IVS2-2A > G G85E, R117H, R334W, R347P, A455E, G542X, G551D, R553X, R560T, R1162X, W1282X, N1303K, F508del, I507del, 2184delA, 3659delC, 621 + 1G > T, 711 + 1G > T, 1717-1G > A, 1898 + 1G > A, 2789 + 5G > A, 3120 + 1G > A, 3849 + 10kbC > T, E60X, R75X, E92X, Y122X, G178R, R347H, Q493X, V520F, S549N, P574H, M1101K, D1152H, 2143delT, 394delTT, 444delA, 1078delT, 3876delA, 3905insT, 1812-1G > A, 3272-26A > G, 2183AA > G, S549R(A > C), R117C, L206W, G330X, T338I, R352Q, S364P, G480C, C524X, S549R(T > G), Q552X, A559T, G622D, R709X, K710X, R764X, Q890X, R1066C, W1089X, Y1092X, R1158X, S1196X, W1204X(c.3611G > A), Q1238X, S1251N, S1255X, 3199del6, 574delA, 663delT, 935delA, 936delTA, 1677delTA, 1949del84, 2043delG, 2055del9 > A, 2108delA, 3171delC, 3667del4, 3791delC, 1288insTA, 2184insA, 2307insA, 2869insG, 296 + 12T > C, 405 + 1G > A, 405 + 3A > C, 406-1G > A, 711 + 5G > A, 712-1G > T, 1898 + 1G > T, 1898 + 5G > T, 3120G > A, 457TAT > G, 3849 + 4A > G, Q359K/T360K 105insA, G229C IVS20 + 6T > C, R696P F1388del, V187D, 3992-9G > A IVS4 + 4A > T, 322delG, R548X N370S, L444P, 1035insG, IVS2 + 1G > A, V394L, R496H, D409H, D409V, R463C, R463H R83C, Q347X, Q27fsdelC, 459insTA, R83H, G188R, Q242X 35G > T R183P, G278S, E322X 511_6944del, IVS3-2A > G fsP330, L302P, R496L, p.R608del R2478_D2512del 1278insTATC, IVS12 + 1G > C, G269S, IVS9 + 1G > A, R178H, IVS7 + 1G > A, 7.6kb del, G250D, R170W R245X N48K
COMPARATIVE PLATFORM EVALUATION FOR CARRIER SCREENING acceptability criteria according to Counsyl standard operating procedures. Genotyping test results were transcribed and reported to the Jacobi Medical Center Human Genetics Laboratory for comparison to earlier test results. Results Thirty-one samples were analyzed using a high-throughput genotype assay for a total of 17 diseases and 153 associated mutations (Table 1). Samples were selected on the basis of mutational diversity and initially represented 38 known mutations. The samples were collectively positive for 35 independent mutations, including pseudodeficient Tay-Sachs alleles. All 38 carrier states were confirmed, and 3 additional carrier states were detected owing to a greater number of diseases and mutations included in the high-throughput platform (Table 2). The three additional carrier states were identified in samples 2, 3, and 18, for Maple Syrup Urine Disease Type 1B, Gaucher disease, and NEB-related nemaline myopathy, respectively. In total, 2207 genotypes were confirmed by high-throughput testing with an average accuracy of 100% (95% CI: 0.998–1). Sample 31 was negative for all mutations analyzed, both initially and with the high-throughput analysis. However, this patient was shown to be a carrier for Tay-Sachs using biochemical enzyme analysis. The initial DNA testing confirmed that the patient was negative for 7 common Tay-Sachs mutations tested by the Luminex panel: G269S, 1278insTATC, IVS12 + 1G > C, IVS9 + 1G > A, 7.6kb del, R247W, and R249W. The Counsyl assay determined that the patient was negative for nine common Tay-Sachs mutations (and also the two pseudodeficiency mutations): R178H, G269S, 1278insTATC, IVS12 + 1G > C, IVS7 + 1G > A, IVS9 + 1G > A, 7.6kb del, G250D, and R170W. These two DNA panels account for 11 independent Tay-Sachs mutations. Discussion An objective comparison of testing platforms is a useful tool for clinicians to assess the utility and reliability of such tests for adoption in their everyday practice. Our data indicate that high-throughput genotyping is an accurate and reliable method for determining patient carrier states in the Ashkenazi Jewish population. Beyond confirming all known carrier states, the high-throughput platform identified three additional carrier states in the patient cohort. The ease with which diseases/mutations could be added to or subtracted from the multiplex platform at no increased cost makes it an appealing solution for expanding disease panels. Professional guidelines for carrier screening in the Ashkenazi Jewish population recommend testing for fewer disorders than included in our study. This is partially cost driven. The American College of Medical Genetics (ACMG) recommends screening for nine disorders, including the four recommended by the American Congress of Obstetricians and Gynecologists (ACOG). In addition to compliance with the basic tenets of screening, the ACMG guidelines advise that the tested disease should have an allele frequency of greater than or equal to 1% or a detection rate of 90% or higher (Gross et al., 2008). The Ashkenazi Jewish community, on the other hand, has demonstrated an acceptance and desire for screening for at least 16 conditions (Scott et al., 2010). This includes carrier screening for the Usher Syndrome Type 1, despite the carrier frequency and detection rate falling well below the ACMG threshold.
765
Looking ahead, another reason to implement expanded carrier screening in the Ashkenazi Jewish population is the increasing ethnic admixture within the community as a result of interfaith marriages, adoption, and changes in patterns of self-identification (Bray et al., 2010). Genetic drift and migration, in combination with cultural patterns, have contributed to the current genetic makeup of the Ashkenazi population. It has become increasingly evident that self-reported ethnic identification is subject to the fallacies of social experience that shape patient preferences and is consequently a substandard measure of genetic risk (Lee et al., 2008; Ross, 2012). Along this line, of 23,453 patients who recently underwent expanded carrier testing, 26.3% of carriers (total n = 76) of the Ashkenazi Jewish founder mutation in familial dysautonomia had not reported any known Jewish ancestry (Lazarin et al., 2013). This statistic could be indicative of a pervasive lack of awareness about personal Jewish ancestry, and/or a higher prevalence of the founder mutation outside of the Ashkenazi Jewish population. Finally, it is known that targeted mutation analysis will always miss carriers, and residual risk must be discussed with patients (Shore et al., 1992). For diseases like Tay-Sachs or cystic fibrosis, hundreds of mutations in the respective HEXA or CFTR genes have been identified. Case in point, the patient represented by sample 31 in this study is likely a carrier of a rare Tay-Sachs mutation, similar to what has been described previously in the literature (Schneider et al., 2009; Nakagawa et al., 2012). Fortunately, this patient was picked up as a carrier by the enzyme assay, which remains a highly sensitive test and should be offered in conjunction with targeted DNA analysis for optimal Tay-Sachs disease carrier detection across ethnicities (Schneider et al., 2009). In part due to the dramatically declining rate of costs-per-megabase for DNA sequencing technologies (National Human Genome Research Institute, 2012), carrier screening by targeted mutation analysis may soon fall out of favor to be replaced by targeted sequencing panels, and eventually, even by whole exome/genome approaches. Even before large-scale population screening by sequencing is feasible, it may be worthwhile to consider sequencing as a reflex test for the noncarrier partner in carrier/ noncarrier couples to better assess residual risk. While sequencing based technologies will allow us to identify these additional carriers and further reduce the risk of the noncarriers by eliminating biases based on ethnicity and mutation frequency, these technologies are not without challenges and bottlenecks. The genetics community must use a discerning approach to contend with a potentially overwhelming amount of clinically vague information, along with the ethical issues related to communication of results to patients and their family members (Tarini and Goldenberg, 2012). That said, practitioners have historically dealt with technological advances and the subsequent disclosure of sensitive information within all fields of medicine; genetics is not necessarily unique in this regard (Evans, 2011). For now, as we transition toward widespread adoption of cost-effective genomic technologies, high-throughput genotyping for previously reported clinically relevant mutations is an acceptable and preferred alternative to carrier screening through current gene-by-gene methodologies. Acknowledgments We would like to thank Dr. Jose Carlos Ferreira and Jie Zhan for their contributions in the Jacobi genetic testing laboratory.
766
2009 2009 2009 2009
2009 2009
2009 2009 2009 2009
2009 2009
2010 2010 2010 2010 2010 2010 2010 2011 2011 2011
2011 2011 2009
7 8 9 10
11 12
13 14 15 16
17 18
19 20 21 22 23 24 25 26 27 28
29 30 31
SMPD1:p.Leu302Pro heterozygote HEXA:c.1421 + 1G > C heterozygote CFTR:p.Ala455Glu heterozygote CFTR:p.Asn1303Lys heterozygote HEXA:p.Gly269Ser heterozygote IKBKAP:p.Arg696Pro heterozygote CFTR:p.Gly542X heterozygote CFTR:c.262_263delTT heterozygote BCKDHB:p.Gly278Ser heterozygote CFTR:p.Trp1282X heterozygote FANCC:c.456 + 4A > T heterozygote IKBKAP:c.2204 + 6T > C heterozygote CFTR:c.1585-1G > A heterozygote DLD:p.Gly229Cys heterozygote No disease-causing mutations detected.a
SMPD1:p.Arg608del heterozygote HEXA:p.Arg247Trp FANCC:c.456 + 4A > T heterozygote HEXA:p.Arg249Trp CFTR:p.Arg117His heterozygote ASPA:p.Ala305Glu heterozygote HEXA:c.1274_1277dupTATC heterozygote SMPD1:p.Arg496Leu heterozygote GBA:p.Val433Leu heterozygote CFTR:c.3717 + 10kbC > T heterozygote GBA:c.115 + 1G > A heterozygote MCOLN1:c.406-2A > G heterozygote HEXA:c.1274_1277dupTATC heterozygote GBA:c.84dupG heterozygote GBA:p.Arg535His heterozygote ASPA:p.Glu285Ala heterozygote SMPD1{NM_000543.3}:c.996delC heterozygote HEXA:c.1073 + 1G > A heterozygote ASPA:p.Tyr231X heterozygote CFTR:p.Phe508del heterozygote BLM:c.2207_2212delATCTGAinsTAGATTC heterozygote MCOLN1:g.190619_197051del heterozygote GBA:p.Asn409Ser homozygote mutant
Known carrier states
Patient is a Tay-Sachs carrier via biochemical analysis.
2008 2008 2008
4 5 6
a
2008 2008 2008
1 2 3
Sample
Year collected
SMPD1:p.Leu302Pro heterozygote HEXA:c.1421 + 1G > C heterozygote CFTR:p.Ala455Glu heterozygote CFTR:p.Asn1303Lys heterozygote HEXA:p.Gly269Ser heterozygote IKBKAP:p.Arg696Pro heterozygote CFTR:p.Gly542X heterozygote CFTR:c.262_263delTT heterozygote BCKDHB:p.Gly278Ser heterozygote CFTR:p.Trp1282X heterozygote FANCC:c.456 + 4A > T heterozygote IKBKAP:c.2204 + 6T > C heterozygote CFTR:c.1585-1G > A heterozygote DLD:p.Gly229Cys heterozygote No disease-causing mutations detected.
SMPD1:p.Arg608del heterozygote HEXA:p.Arg247Trp FANCC:c.456 + 4A > T heterozygote HEXA:p.Arg249Trp CFTR:p.Arg117His heterozygote ASPA:p.Ala305Glu heterozygote HEXA:c.1274_1277dupTATC heterozygote SMPD1:p.Arg496Leu heterozygote GBA:p.Val433Leu heterozygote CFTR:c.3717 + 10kbC > T heterozygote GBA:c.115 + 1G > A heterozygote MCOLN1:c.406-2A > G heterozygote HEXA:c.1274_1277dupTATC heterozygote GBA:c.84dupG heterozygote GBA:p.Arg535His heterozygote ASPA:p.Glu285Ala heterozygote SMPD1{NM_000543.3}:c.996delC heterozygote HEXA:c.1073 + 1G > A heterozygote ASPA:p.Tyr231X heterozygote CFTR:p.Phe508del heterozygote BLM:c.2207_2212delATCTGAinsTAGATTC heterozygote MCOLN1:g.190619_197051del heterozygote GBA:p.Asn409Ser homozygote mutant
High-throughput results
Table 2. Data Comparison
83 83 69
69 69 69 69 69 69 69 81 83 83
69 69
69 69 69 69
69 69
69 69 69 69
69 69 69
69 69 69
Original mutations tested
1 1 0
1 1 1 1 1 1 1 1 1 3
1 2
1 1 1 2
1 2
1 1 1 2
1 1 2
1 1 2
Mutations confirmed
N/A N/A N/A
N/A NEB:p.R2478_D2512del heterozygote N/A N/A N/A N/A N/A N/A N/A N/A N/A N/A
N/A N/A N/A N/A
N/A N/A
N/A N/A N/A N/A
N/A BCKDHB:p.Arg183Pro GBA:p.Arg502Cys heterozygote N/A N/A N/A
Additional mutations identified
COMPARATIVE PLATFORM EVALUATION FOR CARRIER SCREENING Author Disclosure Statement E.E. and S.N. are employees of Counsyl. The samples were selected from patients participating in screening through the Program for Jewish Genetic Health at Einstein/Yeshiva University. Collaboration with S.K. and N.S.A. does not represent an endorsement by their respective institutions for any specific commercial laboratory, but instead for the 100% concordance of the newer methodology of high-throughput expanded carrier screening with the older method of gene-bygene testing. References Bray SM, Mulle JG, Dodd AF, et al. (2010) Signatures of founder effects, admixture, and selection in the Ashkenazi Jewish population. Proc Natl Acad Sci U S A 107:16222–16227. Evans JP (2011) Looking ahead, looking behind. Genet Med 13:177–178. Gross SJ, Pletcher BA, Monaghan KG (2008) Professional Practice and Guidelines Committee. Carrier screening in individuals of Ashkenazi Jewish descent. Genet Med 10:54–56. Klugman S, Gross SJ (2010) Ashkenazi Jewish screening in the twenty-first century. Obstet Gynecol Clin North Am 37:37–46. Laberge AM, Watts C, Porter K, Burke W (2010) Assessing the potential success of cystic fibrosis carrier screening: lessons learned from Tay-Sachs disease and beta-thalassemia. Public Health Genomics 13:310–319. Lazarin G, Haque I, Nazareth S, et al. (2013) An empirical estimate of carrier frequencies for 400 + causal Mendelian variants: results from an ethnically diverse clinical sample of 23,453 individuals. Genet Med 15:178–186. Lee SS, Mountain J, Koenig B, et al. (2008) The ethics of characterizing difference: guiding principles on using racial categories in human genetics. Genome Biol 9:404. Little SE, Janakiraman V, Kaimal A, et al. (2010) The cost-effectiveness of prenatal screening for spinal muscular atrophy. Am J Obstet Gynecol 202:253.e1–253.e17. Nakagawa S, Zhan J, Sun W, et al. (2012) Platelet hexosaminidase A enzyme assay effectively detects carriers missed by targeted DNA mutation analysis. JIMD Rep 6:1–6.
767
National Human Genome Research Institute (2008) Populationbased carrier screening for single gene disorders: LESSONS learned and new opportunities. Rockville, MD: February 6–7, 2008. Available at www.genome.gov/27026048 (Accessed August 3, 2013). National Human Genome Research Institute (2012) DNA sequencing costs: data from the NHGRI Large-Scale Genome Sequencing Program. Available at www.genome.gov/ sequencing costs (Accessed August 3, 2013). Ross LF (2012) A re-examination of the use of ethnicity in prenatal carrier screening. Am J Med Genet 158A:19–23. Schneider A, Nakagawa S, Keep R, et al. (2009) Populationbased Tay-Sachs screening among Ashkenazi Jewish young adults in the 21st century: hexosaminidase A enzyme assay is essential for accurate testing. Am J Med Genet Part A 149A:2444–2447. Scott SA, Edelmann L, Liu L, et al. (2010) Experience with carrier screening and prenatal diagnosis for 16 Ashkenazi Jewish genetic diseases. Hum Mutat 31:1240–1250. Shore S, Tomczak J, Grebner EE, Myerowitz R (1992) An unusual genotype in an Ashkenazi Jewish patient with Tay-Sachs disease. Hum Mutat 1:486–490. Srinivasan BS, Evans EA, Flannick J, et al. (2010) A universal carrier test for the long tail of Mendelian disease. Reprod Biomed Online 21:537–551. Tarini BA, Goldenberg AJ (2012) Ethical issues with newborn screening in the genomics era. Annu Rev Genom Hum Genet 13:381–393.
Address correspondence to: Susan Klugman, MD Division of Reproductive Genetics Department of Obstetrics & Gynecology and Women’s Health Albert Einstein College of Medicine of Yeshiva University Program for Jewish Genetic Health Montefiore Medical Center 1695 Eastchester Rd., Suite 301 Bronx, NY 10461 E-mail: [email protected]
