- Email: [email protected]
Genetic testing for Ataxia in north america
Molecular Diagnosis Vol. 5 No. 2 2000
Original Research Genetic Testing for Ataxia in North America NICHOLAS T. POTTER, PhD,* MARTHA A. NANCE, MD,* FOR THE ATAXIA MOLECULAR DIAGNOSTIC TESTING GROUP* Knoawille, Tennessee; St Louis Park, Milmesota
Background: The Ataxia Molecular Diagnostics Testing Group was established to generate quantitative proficiency and outcomes data regarding molecular testing for the autosomal dominant cerebellar ataxias (spinocerebellar ataxia types 1 [SCA-1] through -3, -6, and -7, and dentatorubral-pallidoluysian atrophy) in North America. Methods and Results: Twenty-four North American laboratories that offered diagnostic testing for one or more ataxia genes were initially identified through GeneTests (Children's Health Care System, Seattle, WA). Eighteen laboratories agreed to participate in the study, which consisted of completing a technical survey, clinical survey, and molecular proficiency test. One laboratory returned the completed surveys but did not perform the proficiency testing. Ten of 18 laboratories (56%) provided data on test volumes, and these laboratories collectively performed 2,240 tests; approximately 5% of the tests yielded a positive result (i.e., identification of a pathological trinucleotide (CAG) repeat expansion). In proficiency testing, 100% of the laboratories correctly genotyped all samples, and 93% of the laboratories were within 1 SD of the mean for sizing normal alleles (one repeat unit or less). Ninety percent of the laboratories were within 1 SD for sizing expanded alleles. Conclusions: Proficiency testing showed little difference between laboratories with respect to allele sizing. However, additional phenotype/genotype correlations are necessary to define CAG repeat-length descriptors for SCA-1, SCA-2, and SCA-7 alleles of intermediate size. Key words: ataxia, CAG repeat, proficiency testing.
From the *NeurogeneticsLaboratoo,, Developmentaland Genetic Center, The UniversiO,of TennesseeMedical Center, Knoxville, TN; and the tDepartment of Neurosciences, Park Nicollet Clinic, St Louis Park, MN. Supported in part from a grant from the National Ataxia Foundation. *Ataxia Molecular Diagnostic Testing Group consists of: D.J. Allingham-Hawkins, PhD (North York, Canada), D. Bellissimo, PhD (Milwaukee, WI), P. Bridge, PhD, Lisa Graham (Calgary, Canada), C. Brown, PhD (Charlotte, NC), A. Garber, PhD (Milwaukee, WI), Debra Leonard, MD, PhD, Hanna Rennert, PhD (Philadelphia, PA), Russell Margolis, MD (Baltimore, MD), N. Mclntosh, MSc (Stratford, CT), K. Muralidharan, PhD (Atlanta, GA), B. Popovich, PhD, K. Anoe (Portland, OR), P. Ray, PhD (Toronto, Canada), C.S. Richards, PhD, P. Gunaratne, PhD (Houston, TX), F. Schaefer, PhD (Tulsa, OK), W. Seltzer, PhD (Worcester, MA), K. Sims, MD, W. Xin, PhD (Charlestown, MA), K. Snow, PhD (Rochester, MN), K. Stephens, PhD (Seattle, WA), M. Wick, PhD (Minneapolis, MN). Reprint requests: Nicholas T. Potter, PhD, Neurogenetics Laboratory, The University of Tennessee Medical Center, Physicians' Office Building III, Suite 435, 1930 Alcoa Hwy, Knoxville TN 37920. Email: [email protected] Copyright © 2000 by Churchill Livingstone® 1084-8592/00/0502-0002510.00/0 doi: 10. 1054/xd.2000.7180
92
Molecular Diagnosis Vol. 5 No. 2 June 2000
Issues related to the perfornaance and quality assurance of molecular diagnostic testing have recently been raised in the clinical literature [ 1]. With the rapid increase in availability of molecular diagnostic testing for the hereditary ataxias, it was believed that a mechanism for formally addressing issues of quality assurance and control needed to be developed. In the absence of an established proficiency testing p r o ~ a m designed to address the specific challenges of diagnostic testing for the ataxias, the Ataxia Molecular Diagnostic Testing Group (AMDTG) was established in 1998. The group's first goal was to generate quantitative proficiency and outcomes data regarding molecular testing for the dominantly inherited ataxias (spinocerebellar ataxia types 1 [SCA-1] through -3, -6, and -7) and dentatorubral-pallidoluysian atrophy (DRPLA). It is the intent of this group, which includes molecular diagnostic service laboratories in both the academic and private sectors, to provide an annual forum for sample exchange, clinical case presentation, technical discussion, and policy formulation regarding issues specifically related to ataxia gene testing [2,3].
Methods Laboratory Recruitment Laboratories listed in GeneTests (www.genetests. org) as offering clinical testing for at least one of the SCAs or DRPLA were identified for participation in this study. Of the 24 laboratories identified, 21 (88%) expressed a willingness to participate and were sent technical surveys and DNA proficiency samples. Eighteen of 21 laboratories (86%) returned the completed surveys. At the time of the completion of the survey (June 1998), six of 18 laboratories (33%) offered testing for all six diseases, nine of 18 (50%) for at least five diseases, 11 of 18 (61%) for at least four diseases, and 17 of 18 (94%) for at least three diseases. All laboratories offered testing for at least SCA-1 and SCA-3. The participants included 15 US and three Canadian laboratories. Fifteen laboratories were academic based and three laboratories were commercial facilities.
Proficiency Testing The Neurogenetics Laboratory at the University of Tennessee Medical Center validated each of the
six disease-specific challenges. Genomic DNA was amplified using standard published PCR conditions for each disorder. PCR products were visualized by autoradiography after separation on 6% 35- × 43cm denaturing acrylamide gels, and repeat size was determined by comparison to an M13 sequencing ladder.
Technical Survey The technical questionnaire contained questions regarding specific methods used, including descriptions of disease-specific primer pairs and amplification conditions, allele separation and sizing methods, (CAG)n reference ranges and turnaround times for reporting results, and test charges. Clinical Survey
The clinical questionnaire included questions pertaining to the general characteristics of the laboratory setting, including type of professionals in the laboratory, type of institution, type of testing offered (i.e., diagnostic, predictive, prenatal), and referral patterns. In addition, specific information about criteria and policies regarding sample acceptance, genetic counseling, informed consent, and DNA banking was requested. A copy of the technical and clinical survey questions can be obtained from the authors on request.
Results All 18 laboratories used PCR-based methods for the genotyping of normal and expanded alleles. No laboratory routinely used Southem blot analysis for detection of expanded alleles. The majority of laboratories (>75% for each disease) used literaturederived primer pairs, whereas the remaining laboratories used in-house primers. For analysis for SCA-1, only three of 18 laboratories (17%) reported routine use of SfaN I restriction enzyme digestion to analyze alleles containing repeats in the intermediate-size range (Table 1). For all ataxia gene testing, all laboratories reported the use of either denaturing polyacrylamide gel electrophoresis (89%) or capillary gel electrophoresis for sample analysis (11%). For sizing of the PCR products, 14 of 18 laboratories (78%) indicated the use of a sequencing ladder, 15 of 18 laboratories (83%) used previously genotyped patient samples (positive and negative con-
Ataxia Technical Survey and Proficiency Test
•
Potter and Nance
93
Table 1. (CAG) n Ranges Ascertained From Diagnostic Laboratory Specimens Genotyped by the AMDTG Disease
Normal Alleles*
Intermediate Alleles*
Expanded Alleles
DRPLA SCA- 1* SCA-2 SCA-3 SCA-6 SCA-7
7-29 14-39 (+44)* 15-34 12--41 6-17 7-15
NR NR NR NR NR NR
60-70 37-78 33-52 (+220 infant)II 55-82 21-28 39-57
NR, none reported. *CAG length descriptors are defined from literature-derived data for each disorder. *Intermediate alleles are defined as alleles carrying repeat lengths that lie on the boundaries between confirmed normal and pathological size ranges. Genotypic assignments are based on clinical correlations, molecular characteristics (i.e., the presence/absence of repeat interruptions), and familial segregation patterns of the allele (i.e., does it show meiotic stability or instability on transmission from one generation to the next). *Genotypic assignments based on determination of presence/absence of CAT sequences and clinical correlation. SCA-I alleles with 37 repeats have been identified with and without SfaN I sites, and alleles of 38, 39, and 44 have been reported with SfaN I sites. See reference [13]. IISee reference [ 19]. trois), and 12 of 18 laboratories (67%) used both. Two of 18 laboratories (11%) indicated the use of cloned sequences and patient samples. Other reference standards included patient-based contiguous allelic ladders and digested h DNA ladders. In proficiency testing, 100% of the laboratories correctly genotyped all samples, and 93% of the laboratories were within 1 SD of the mean for sizing normal alleles (one repeat unit or less). Ninety percent of the laboratories were within 1 SD for sizing expanded alleles. Only for SCA-3 and SCA-7 did the SD represent more than two repeats for expanded alleles. A summary of disease-specific proficiency testing and (CAG), ranges ascertained from diagnostic specimens derived from the collective laboratory data are shown in Table 1 and Fig. 1, respectively. Two thirds of the laboratories indicated a turnaround time for reporting results of 7 to 14 days, and 90% of all laboratories released results within 21 days. Ninety-three percent of the laboratories included CAG repeat number in their final reports. Fees for testing generally were within the range of US $200 to $300 for a single test (61% of laboratories), whereas 28% charged between $100 and $200 and 11% greater than $300. Results of the clinical survey indicated that all participating laboratories were directed by a boardcertified PhD (American Board of Medical Genet-
ics certification in clinical molecular genetics) or a board-certified MD, and 54% of the laboratories indicated the routine use of a genetic counselor. Eighty-five percent reported additional staff in the laboratories that included board-certified PhD and MD geneticists, MD pathologists, and PhD and MD fellows. All US laboratories were Clinical Laboratory Improvement Amendments (CLIA) certified, 46% of the laboratories were university affiliated, 15% were commercial facilities, and 15% were private/public hospital-based laboratories. The remaining specific institutional designations were unknown. All laboratories indicated they accepted samples for diagnostic testing, whereas 61% accepted samples for predictive testing and 39% accepted prenatal specimens. All laboratories required a referral from a physician or genetic counselor before accepting a specimen. No laboratory accepted a sample directly from a patient. All laboratories indicated adherence to established practice guidelines regarding specimen identification, handling, and follow-up [4]. Regarding issues of counseling and consent for testing, 46% of the laboratories required signed consent for diagnostic testing and 80% of the laboratories offering predictive testing required evidence of informed consent. No laboratory discarded DNA samples after analysis, and 77% of the laboratories indicated they banked DNA from diagnostic specimens indefinitely primarily
94
Molecular Diagnosis Vol. 5 No. 2 June 2000 DRPLA 99-
70$050CAG Rep*at Number 40-
30-
20-
!
i
$CA-1 60
50
,80
CAG Rep*al Number 3o
20
LiJ~,~,kmt
Ilnormal at/~ oa~,~c~oeaee~
29
29
29 51
29 s~
29
29 s*
aO
29 ~2
2Q 52
29 s*
29 ~t
~5
30 s~
30 s~
29
'~.'
SCA-2
so-
eat Numb*r
=
eAGR~
.
_[_l`=
L
,
,
Fig. 1. Interlabo~ato~ comparison of (CAG). repeat determination for normal and expanded aUeles at the DRPLA, SCA-1 through ,3, -6, and -7 loci. DRPLA: Number of laboratories reporting; n = 12. Normal allele: mean, 16; SD, 0;3, mode, 16; range, I5-16. Expanded allele: mean, 69. $D, 1,9; mode, 70; range, 65-72. SCA-I: n = 17. N o d a l alMe: mean, 29, SD, 0.4; mode, 29; range, 29-30. Expanded allele: meant 52; SD, 1.0; mode, 52; range, 51-54, SCA-2: n = 13. Normal allele: mean, 22; SD, 0.6; mode, 22; range, 20-22. Expanded allele: mean, 46; SD, !.4; mode; 47; range, 43-48.
Ataxia Technical Survey and Proficiency Test
•
Potter and N a n c e
95
5CA-3
80-
70SO-
SO-
CAO Re,is! Number 40-
3020-
10.
!
2
3
4
$
d
22 77
22 74
22 77
22 76
22 74
-~ t~
n
76
re
.
~a
7,
75 73 ;,6 7 ~ 1 ~ I
SCA4
CAn Repest Number
2
3
4
5
It
II
II
11
22
22
22
22
--iF 6
7
8
It
II
tO
l!
I!
21
2!
22
23
SCA-7
8O I
Ii
m
50 m
40
I
CAG Repeal Number 3O
2O
fO
LJb*,,I
O
I $
!
II
I l0
10
4
S
II
I0 63
I; $5
II, 5;
Fig. 1. SCA-3: n = 17. Normal allele: mean, 22; SD, 0.4; mode, 22; range, 21-23. Expanded allele: mean, 76; SD, 2.7; mode, 77; range, 73-84. SCA-6: n = 10. Normal allele: mean, 11; SD, 0.3; mode, I 1; range, 10-1 I. Expanded allele: mean, 22; SD, 0,6; mode, 22; range, 21-23. SCA-7: n = 9. Normal allele: mean, 10; SD, 0.7; mode, 10; range, 10-12. Expanded allele: mean; 58; SD, 2.5; mode, 57; range, 55-63.
96
Molecular Diagnosis Vol. 5 No. 2 June 2000
for use in future clinical test development and follow-up diagnostic testing as new autosomal dominant cerebellar ataxia (ADCA) loci are identified.
Discussion The autosomal dominant cerebellar ataxias (ADCAs) represent a clinically heterogeneous group of neurological diseases with at least 12 distinct genetic loci identified to date [5-8]. Because the collective incidence of these disorders is high in both unselected and selected ataxia cohorts [9,10] and the clinical distinctions between them is diagnostically unreliable, the role of molecular diagnostics is significant. At the time of this survey, diagnostic testing was available for six of these disorders (SCA-1 through -3, -6, -7, and DRPLA). However, with the exception of SCA-3 proficiency challenges included in the 1999 College of American Pathologists (CAP)/American College of Medical Genetics Molecular Genetics (ACMG) Survey, there remains no established mechanism to provide proficiency testing for these diseases. In addition, because the CAP/ACMG survey does not provide a mechanism to adch'ess test interpretation or the relative benefits of varying methods in the analysis of unusual cases, the primary initiative for the AMDTG was to develop a comprehensive and timely program for addressing all aspects of ataxia gene testing. In our exchange, proficiency testing showed little difference between laboratories with respect to allele sizing, supporting the contention that all the participating laboratories (71% of all ataxia testing laboratories listed in GeneTests) provided accurate results despite some differences in testing methods. One general comment voiced by the group was the need for a set of disease-specific thxeshold standards that could be used by the AMDTG to accurately genotype patients with large, normal, or small expanded alleles. Whereas the results of proficiency testing showed no major concerns, several diseasespecific issues were discussed at the 1998 meeting of the AMDTG held during the 48th Annual Meeting of the American Society of Human Genetics in Denver, CO.
SCA-1 The literature suggests an overlap in the distribution of normal and pathological alleles carrying be-
tween 39 and 44 repeats [I1-13]. Three samples reported by the AMDTG suggested that this intermediate range extends from 37 to 44 repeats (Table 1). In this series, as suggested in the literature [14], CAT interruption was associated with apparent nonpathogenicity for these alleles. Based on this, the AMDTG reached a consensus that all alleles containing 37 to 44 CAG repeats be analyzed for the presence or absence of the CAT interruption to assist with their genotypic classification. Furthermore, because data suggest age of onset is determined by the number of uninterrupted repeats [15], CAT analysis would be useful in patients with SCA-] with atypical later-onset disease.
SCA-2 Participating laboratories reported an overlap between nornlal and expanded alleles in the 33- to 34-repeat range (Table 1). The recent literatule also suggests an intermediate range between 33 and 35 CAG repeats in which the phenotypic significance of the results is uncertain [16]. Normal alleles containing 30 to 32 repeats have been reported [16], and an allele with 33 repeats has been identified in a population-based study [17]; however, it is currently not known whether this represents a pathological allele associated with reduced penetrance, later age of onset, or a large meiotically stable normal allele. Although the smallest pathological allele reported contains 34 repeats in an individual with late-onset disease [18], we are aware of a symptomatic patient also with late-onset disease who carties an allele with 33 CAG repeats (A.R. La Spada, personal communication, December 1999), The clinical significance of the CAA interruption within the SCA-2 gene is also uncertain [16], and there is presently no clinical assay to determine the presence or absence of the repeat. Allele contraction has also been observed in this disorder, with an allele with 32 repeats reported in an asymptomatic 19-year-old whose affected father carried an allele with 40 repeats [16], Thus, care should be exercised in interpreting results of cases with repeats in this range, Finally; members of the AMDTG have previously reported the identification of "extreme expansions" in patients with the infantile or juvenile forms of SCA-2 in which repeat length can exceed 200 triplets [19]. Although alleles in the repeat range of approximately 200 may be detectable by conventional PCR [19], alternative methods of al-
Ataxia Technical Survey and Proficiency Test
lele detection (i.e., Southern blot or long-range PCR) should be considered in these situations if warranted, particularly if the case is unusual [20].
SCA-3 No significant concerns were noted.
SCA -6 Although no concerns regarding the sizing of SCA-6 alleles were identified in this study, the literature reflects a bimodal distribution of normal and expanded alleles with an overlap at 20 CAG repeats [21,22]. This includes the identification of a patient with a repeat number of 20 and a clinical diagnosis of episodic ataxia type 2 [22]. As such, some caution should be used in interpreting the results in patients carrying alleles of this size.
SCA-7 A review of the literature suggests that the genotypic classification of SCA-7 alleles in the 28- to 35-CAG repeat range remains unclear. Although no participating laboratory reported a patient with an allele in this size range, the classification of these alleles was discussed in light of new data on the subject. The largest reported normal allele contains 35 repeats, and the smallest expanded allele carries 37 repeats [23]. However, the identification of an allele with 36 repeats associated with an at-risk haplotype suggests this is a pathological allele [24]. Because meiotically unstable intermediate alleles with 28 and 35 repeats have also been documented in SCA-7 pedigrees [25,26], an allele within this size range should be considered unstable. As with SCA-2, the recent identification of extreme repeat sizes (>300 triplets) in infantile-onset SCA-7 warrants the availability of molecular testing strategies that will ensure the detection of very large expansions [25-28].
DRPLA No significant concerns were noted. The literature suggests that the collective incidence of these disorders (ADCAs; SCA1, -2, -3, -6, -7) ranges from 28% to 61% in unselected and selected ataxia cohorts, respectively [9,10]. As such, the collective "hit rate" of 5% from the AMDTG is considerably less than that previously reported. This may be the result of several factors, including
•
Potter and Nance
97
the multiplexing of tests (i.e., one patient was simultaneously tested for all six disorders), the genotyping of patients with a molecularly uncharacterized ADCA, and the testing of patients who do not fulfill the clinical criteria for a hereditary ataxia. Although this study did not address which of the approximately 95% of negative tests were ordered appropriately, it is clear that the AMDTG data are more consistent with the incidence of CAG repeat expansion in individuals with no family history of ataxia (i.e., sporadic ataxia) [10]. Although the AMDTG data are based only on the responses of approximately half (10 of 18) of the participating laboratories, it suggests that a greater emphasis may need to be placed on the education of the referring health professional regarding the appropriate use of DNA diagnostic testing strategies in patients with ataxia. Issues related to consent and banking of tested samples remain unresolved and will require further discussion. Forty-six percent of the laboratories required signed consent for diagnostic testing, a statistic consistent With that reported by McGovern et al. [I] in their large quality assurance study of 245 molecular genetic testing laboratories. However, it is interesting that the AMDTG cohort was almost twice as likely to require signed consent for predictive testing than the larger cohort studied by MeGovern et al. [1] (80% vs 44%). This may reflect an increased awareness by the AMDTG of the issues involved with presymptomatic testing for neurologically relevant diseases because many of the laboratories were directly involved in the development of practice guidelines for predictive testing for Huntington disease. Although it is encouraging that the majority of AMDTG laboratories appear to be adhering to established guidelines for predictive testing, it is still somewhat disconcerting that compliance was not 100%. A second point of concern pertains to the banking of diagnostic specimens. No laboratory discarded specimens after testing, and greater than 75% indicated the use of banked samples for future test development and follow-up diagnostic testing. Although this practice was considered routine by the group, less than half the laboratories required signed consent for diagnostic testing, which raises questions regarding the interpretation of the Common Rule and institutional review board requirements for the use of these materials [29]. Although we did not specifically ask whether institutional r e -
98
Molecular Diagnosis, Vol. 5 No. 2 June 2000
view board approval was sought, recent proposals aimed at limiting the retention and use of archived specimens [30]: wan'ants a comprehensive effort toward establishing a consensus regarding the use of archived specimens for clinical test development and validation I29]. These unresolved issues will become focal points for future discussions by the ~'oup. In summary, the organization of the A M D T G consortium has enabled the group to initiate woficiency testing for the ADCAs in a timely manner. It also provides an annual forum to self-monitor laboratory practice and discuss difficult or unusual cases. It is the intent of the group to continue to meet annually to ensm'e the quality and integrity of specialized testing for this group of neurological disorders.
Received December 29, 1999. Received in revised form Janual3, 26, 2000. Accepted February 4, 2000.
8.
9.
10.
11.
12.
13.
R e f e r e n ces 14. 1. McGovern MM, Benach MO, Wallenstein S, Desnick R J, Keenlyside R: Quality assurance in molecular genetic testing laboratories. JAMA 1999; 281:835-840 2. Potter NT, Nance MA, for The Ataxia Molecular Diagnostic Testing Group: Genetic testing for ataxia in North America. Am J Hum Genet 1998;63:A239 (abstr) 3. Wick MJ, Matthias-Hagen VL, Allingham-Hawkins DJ, et al., for The Ataxia Molecular Diagnostic Testing Group: Genetic testing for Friedreich ataxia. Am J Hum Genet 1999;65:A412 (abstr) 4. Watson MS, Altmiller DH, de Martinville B, et al.: Standards mad guidelines for clinical genetics laboratories, 2nd ed. American College of Medical Genetics, Bethesda, 1999 5. Koide R, Ikeuchi T, Onodera O, et al.: Unstable expansion of CAG repeat in hereditary dentatorubral-pallidoluysian atrophy (DRPLA). Nat Genet 1994;6:9-13 6. Koob MD, Moseley ML, Schut LJ, et al.: An untranslated CTG expansion causes a novel form of spinocerebeHar ataxia ~SCAS). Nat Genet 1999;21: 379-384 7. Worth PF, Giunti P, Gardner-Thorpe C, Dixon PH, Davis MB, Wood NW: Autosomal dominant cerebN~ar ataxia type III: Linkage in a Iarge British
15.
16.
17.
18.
19.
20.
family to a 7.6-cm region on chromosome 15q1421.3. Am J Hum Genet 1999;65:420--426 Hohnes SE, O'Hearn E, McInnis MG, et al.: Exp.ansion of a novel CAG repeat in the 5 region of PPP2R2B is associated with SCAI2. Nat Goner 1999;23:391-392 Gunaratue PH, Richards CS: Estimated contribution of known ataxia genes in ataxia patients undergoing DNA testing. Genet Testing 1997/1998;1:275-278 Moseley ML, Benzow KA, Schut LJ, et al.: Incidence of dominant spinocerebellar and Friedreich ataxia triplet repeats among 361 ataxia families. Neurology 1998;51:1666-1671 Matitla T, Volpini V, Genis D, et al.: Presymptomatic analysis of spinocerebellar ataxia type 1 (SCA1) via the expansion o f the SCA1 CAG-repeat in a large pedigree displaying anticipation and parental male bias. Hum Mol Genet 1993;2:2123-2128 Ranum LPW, Chung M-Y, Banff S, et al.: Molecular and clinical cola'elations in spinocerebellar ataxia type 1 (SCA1): Evidence for familial effects on the age of onset. Am J Hum Genet 1994;55:244 252 Quan F, Janas J, Popovich BW: A novel CAG repeat configuration in tbe SCAi gene: Implications for the molecular diagnostics of spinocerebellar ataxia type i, Hum Mol Goner 1995;4:2411-2413 Chung M-Y, Ranum LPW, Duvick L, Servadio A, Zoghbi HY, Orr HT: Evidence for a mechanism predisposing to intergenerational CAG repeat instability in spinocerebellar ataxia type I. Nat Genet 1993; 5:254-258 Matsuyama Z, Izumi Y, Kameyama M, Kawakami H, Nakamura S: Tlie effect of CAT trinucleotide interruptions on the age at onset :of spinocerebellar ataxia type 1 (SCA1). J Meal Genet 1999;36:546548 Pulst S-M: Spinocerebellar ataxia type 2. In Wells RD, Warren ST~ Genetic instabilities and hereditary neurological diseases, 1st ed. Academic Press, San Diego; 1998 Leggo J, Dalton A, Morrison PJ, et al.: Analysis of spinocerebellar ataxia types 1, 2, 3, and 6, denta, torubral-pallidoluysian atrophy, and Friedreich's ataxia genes in spinocerebellar ataxia patients in tlae UK, J Med Genet 1997;34:982,985 Malandrini A, Galli Iz, Villanova M, et ai.: CAG repeat expansion in an Italian family with spinoCerebellar ataxia type 2 (SCA2): A clinical and genetic study. Eur Nem'oi 1998;40:164-168 Babovic-Vuksanovic D, Snow K, Patterson MC, Miclaels VV: SpinocerebeHar ataxia type 2 (SCA2) in an infant with extreme CAG repeat expansion. Am:J Med Genet 1998;79:383,387 Mao R, Snow K: Screening for large CAG-repeat expansions in the SCA2 and SCA7 genes in infantile
Ataxia Technical Survey and Proficiency Test
21.
22.
23.
24.
25.
and juvenile-onset spinocerebellar ataxia (SCA). J Mol Diagn 1999;1:43 (abstr) Ishikawa K, Tanaka H, Saito M, et al.: Japanese families with autosomal dominant pure cerebellar ataxia map to chromosome 19p 13.1-p 13.2 and are strongly associated with mild CAG expansions in the spinocerebellar ataxia type 6 gene in chromosome 19p13.1. Am J Hum Genet 1997;61:336-346 Jodice C, Mantuano E, Veneziano L, et al.: Episodic ataxia type 2 (EA2) and spinocerebellar ataxia type 6 (SCA6) due to CAG repeat expansion in the CACNA1A gene on chromosome 19p. Hum Mol Genet 1997;6:1973-1978 David G, Durr A, Stevanin G, et al.: Molecular and clinical correlations in autosomal dominant cerebellar ataxia with progressive macular dystrophy (SCA7). Hum Mol Genet 1998;7:165-170 Benton CS, de Silva R, Rutledge SL, Bohlega S, Ashizawa T, Zoghbi HY: Molecular and clinical studies in SCA-7 define a broad clinical spectrum and the infantile phenotype. Neurology 1998;51: 1081-1086 Stevanin G, Giunti P, Belal GDS, et al.: De novo
26.
27.
28.
29.
30.
•
Potter and Nance
99
expansion of intermediate alleles in spinocerebellar ataxia 7. Hum Mol Genet 1998;7:1809-1813 Giunti P, Stevanin G, Worth PF, David G, Brice A, Wood NW: Molecular and clinical study of 18 families with ADCA type II: Evidence for genetic heterogeneity and de novo mutation. Am J Hum Genet 1999;64:1594-1603 Gunaratne PH, Gannavarapu A, Palmer S, Richards CS: Improvement in clinical genetic testing for SCA7: Development of a Southern to detect extremely large expansions. Am J Hum Genet 1999;65:A299 (abstr) Matsuura T, Khajavi M, de Silva R, Ashizawa T: A very large SCA7 CAG expansion is compatible with cell viability in somatic mosaicism. Am J Hum Genet 1999;65:A460 (abstr) Merz JF, Leonard DGB, Miller ER: IRB review and consent in human tissue research. Science 1999;283: 1647-1648 Clayton EW, Steinberg KK, Khoury MJ, et al.: Informed consent for genetic research on stored tissue samples. JAMA 1995;274:1786-1792
Original Research Genetic Testing for Ataxia in North America NICHOLAS T. POTTER, PhD,* MARTHA A. NANCE, MD,* FOR THE ATAXIA MOLECULAR DIAGNOSTIC TESTING GROUP* Knoawille, Tennessee; St Louis Park, Milmesota
Background: The Ataxia Molecular Diagnostics Testing Group was established to generate quantitative proficiency and outcomes data regarding molecular testing for the autosomal dominant cerebellar ataxias (spinocerebellar ataxia types 1 [SCA-1] through -3, -6, and -7, and dentatorubral-pallidoluysian atrophy) in North America. Methods and Results: Twenty-four North American laboratories that offered diagnostic testing for one or more ataxia genes were initially identified through GeneTests (Children's Health Care System, Seattle, WA). Eighteen laboratories agreed to participate in the study, which consisted of completing a technical survey, clinical survey, and molecular proficiency test. One laboratory returned the completed surveys but did not perform the proficiency testing. Ten of 18 laboratories (56%) provided data on test volumes, and these laboratories collectively performed 2,240 tests; approximately 5% of the tests yielded a positive result (i.e., identification of a pathological trinucleotide (CAG) repeat expansion). In proficiency testing, 100% of the laboratories correctly genotyped all samples, and 93% of the laboratories were within 1 SD of the mean for sizing normal alleles (one repeat unit or less). Ninety percent of the laboratories were within 1 SD for sizing expanded alleles. Conclusions: Proficiency testing showed little difference between laboratories with respect to allele sizing. However, additional phenotype/genotype correlations are necessary to define CAG repeat-length descriptors for SCA-1, SCA-2, and SCA-7 alleles of intermediate size. Key words: ataxia, CAG repeat, proficiency testing.
From the *NeurogeneticsLaboratoo,, Developmentaland Genetic Center, The UniversiO,of TennesseeMedical Center, Knoxville, TN; and the tDepartment of Neurosciences, Park Nicollet Clinic, St Louis Park, MN. Supported in part from a grant from the National Ataxia Foundation. *Ataxia Molecular Diagnostic Testing Group consists of: D.J. Allingham-Hawkins, PhD (North York, Canada), D. Bellissimo, PhD (Milwaukee, WI), P. Bridge, PhD, Lisa Graham (Calgary, Canada), C. Brown, PhD (Charlotte, NC), A. Garber, PhD (Milwaukee, WI), Debra Leonard, MD, PhD, Hanna Rennert, PhD (Philadelphia, PA), Russell Margolis, MD (Baltimore, MD), N. Mclntosh, MSc (Stratford, CT), K. Muralidharan, PhD (Atlanta, GA), B. Popovich, PhD, K. Anoe (Portland, OR), P. Ray, PhD (Toronto, Canada), C.S. Richards, PhD, P. Gunaratne, PhD (Houston, TX), F. Schaefer, PhD (Tulsa, OK), W. Seltzer, PhD (Worcester, MA), K. Sims, MD, W. Xin, PhD (Charlestown, MA), K. Snow, PhD (Rochester, MN), K. Stephens, PhD (Seattle, WA), M. Wick, PhD (Minneapolis, MN). Reprint requests: Nicholas T. Potter, PhD, Neurogenetics Laboratory, The University of Tennessee Medical Center, Physicians' Office Building III, Suite 435, 1930 Alcoa Hwy, Knoxville TN 37920. Email: [email protected] Copyright © 2000 by Churchill Livingstone® 1084-8592/00/0502-0002510.00/0 doi: 10. 1054/xd.2000.7180
92
Molecular Diagnosis Vol. 5 No. 2 June 2000
Issues related to the perfornaance and quality assurance of molecular diagnostic testing have recently been raised in the clinical literature [ 1]. With the rapid increase in availability of molecular diagnostic testing for the hereditary ataxias, it was believed that a mechanism for formally addressing issues of quality assurance and control needed to be developed. In the absence of an established proficiency testing p r o ~ a m designed to address the specific challenges of diagnostic testing for the ataxias, the Ataxia Molecular Diagnostic Testing Group (AMDTG) was established in 1998. The group's first goal was to generate quantitative proficiency and outcomes data regarding molecular testing for the dominantly inherited ataxias (spinocerebellar ataxia types 1 [SCA-1] through -3, -6, and -7) and dentatorubral-pallidoluysian atrophy (DRPLA). It is the intent of this group, which includes molecular diagnostic service laboratories in both the academic and private sectors, to provide an annual forum for sample exchange, clinical case presentation, technical discussion, and policy formulation regarding issues specifically related to ataxia gene testing [2,3].
Methods Laboratory Recruitment Laboratories listed in GeneTests (www.genetests. org) as offering clinical testing for at least one of the SCAs or DRPLA were identified for participation in this study. Of the 24 laboratories identified, 21 (88%) expressed a willingness to participate and were sent technical surveys and DNA proficiency samples. Eighteen of 21 laboratories (86%) returned the completed surveys. At the time of the completion of the survey (June 1998), six of 18 laboratories (33%) offered testing for all six diseases, nine of 18 (50%) for at least five diseases, 11 of 18 (61%) for at least four diseases, and 17 of 18 (94%) for at least three diseases. All laboratories offered testing for at least SCA-1 and SCA-3. The participants included 15 US and three Canadian laboratories. Fifteen laboratories were academic based and three laboratories were commercial facilities.
Proficiency Testing The Neurogenetics Laboratory at the University of Tennessee Medical Center validated each of the
six disease-specific challenges. Genomic DNA was amplified using standard published PCR conditions for each disorder. PCR products were visualized by autoradiography after separation on 6% 35- × 43cm denaturing acrylamide gels, and repeat size was determined by comparison to an M13 sequencing ladder.
Technical Survey The technical questionnaire contained questions regarding specific methods used, including descriptions of disease-specific primer pairs and amplification conditions, allele separation and sizing methods, (CAG)n reference ranges and turnaround times for reporting results, and test charges. Clinical Survey
The clinical questionnaire included questions pertaining to the general characteristics of the laboratory setting, including type of professionals in the laboratory, type of institution, type of testing offered (i.e., diagnostic, predictive, prenatal), and referral patterns. In addition, specific information about criteria and policies regarding sample acceptance, genetic counseling, informed consent, and DNA banking was requested. A copy of the technical and clinical survey questions can be obtained from the authors on request.
Results All 18 laboratories used PCR-based methods for the genotyping of normal and expanded alleles. No laboratory routinely used Southem blot analysis for detection of expanded alleles. The majority of laboratories (>75% for each disease) used literaturederived primer pairs, whereas the remaining laboratories used in-house primers. For analysis for SCA-1, only three of 18 laboratories (17%) reported routine use of SfaN I restriction enzyme digestion to analyze alleles containing repeats in the intermediate-size range (Table 1). For all ataxia gene testing, all laboratories reported the use of either denaturing polyacrylamide gel electrophoresis (89%) or capillary gel electrophoresis for sample analysis (11%). For sizing of the PCR products, 14 of 18 laboratories (78%) indicated the use of a sequencing ladder, 15 of 18 laboratories (83%) used previously genotyped patient samples (positive and negative con-
Ataxia Technical Survey and Proficiency Test
•
Potter and Nance
93
Table 1. (CAG) n Ranges Ascertained From Diagnostic Laboratory Specimens Genotyped by the AMDTG Disease
Normal Alleles*
Intermediate Alleles*
Expanded Alleles
DRPLA SCA- 1* SCA-2 SCA-3 SCA-6 SCA-7
7-29 14-39 (+44)* 15-34 12--41 6-17 7-15
NR NR NR NR NR NR
60-70 37-78 33-52 (+220 infant)II 55-82 21-28 39-57
NR, none reported. *CAG length descriptors are defined from literature-derived data for each disorder. *Intermediate alleles are defined as alleles carrying repeat lengths that lie on the boundaries between confirmed normal and pathological size ranges. Genotypic assignments are based on clinical correlations, molecular characteristics (i.e., the presence/absence of repeat interruptions), and familial segregation patterns of the allele (i.e., does it show meiotic stability or instability on transmission from one generation to the next). *Genotypic assignments based on determination of presence/absence of CAT sequences and clinical correlation. SCA-I alleles with 37 repeats have been identified with and without SfaN I sites, and alleles of 38, 39, and 44 have been reported with SfaN I sites. See reference [13]. IISee reference [ 19]. trois), and 12 of 18 laboratories (67%) used both. Two of 18 laboratories (11%) indicated the use of cloned sequences and patient samples. Other reference standards included patient-based contiguous allelic ladders and digested h DNA ladders. In proficiency testing, 100% of the laboratories correctly genotyped all samples, and 93% of the laboratories were within 1 SD of the mean for sizing normal alleles (one repeat unit or less). Ninety percent of the laboratories were within 1 SD for sizing expanded alleles. Only for SCA-3 and SCA-7 did the SD represent more than two repeats for expanded alleles. A summary of disease-specific proficiency testing and (CAG), ranges ascertained from diagnostic specimens derived from the collective laboratory data are shown in Table 1 and Fig. 1, respectively. Two thirds of the laboratories indicated a turnaround time for reporting results of 7 to 14 days, and 90% of all laboratories released results within 21 days. Ninety-three percent of the laboratories included CAG repeat number in their final reports. Fees for testing generally were within the range of US $200 to $300 for a single test (61% of laboratories), whereas 28% charged between $100 and $200 and 11% greater than $300. Results of the clinical survey indicated that all participating laboratories were directed by a boardcertified PhD (American Board of Medical Genet-
ics certification in clinical molecular genetics) or a board-certified MD, and 54% of the laboratories indicated the routine use of a genetic counselor. Eighty-five percent reported additional staff in the laboratories that included board-certified PhD and MD geneticists, MD pathologists, and PhD and MD fellows. All US laboratories were Clinical Laboratory Improvement Amendments (CLIA) certified, 46% of the laboratories were university affiliated, 15% were commercial facilities, and 15% were private/public hospital-based laboratories. The remaining specific institutional designations were unknown. All laboratories indicated they accepted samples for diagnostic testing, whereas 61% accepted samples for predictive testing and 39% accepted prenatal specimens. All laboratories required a referral from a physician or genetic counselor before accepting a specimen. No laboratory accepted a sample directly from a patient. All laboratories indicated adherence to established practice guidelines regarding specimen identification, handling, and follow-up [4]. Regarding issues of counseling and consent for testing, 46% of the laboratories required signed consent for diagnostic testing and 80% of the laboratories offering predictive testing required evidence of informed consent. No laboratory discarded DNA samples after analysis, and 77% of the laboratories indicated they banked DNA from diagnostic specimens indefinitely primarily
94
Molecular Diagnosis Vol. 5 No. 2 June 2000 DRPLA 99-
70$050CAG Rep*at Number 40-
30-
20-
!
i
$CA-1 60
50
,80
CAG Rep*al Number 3o
20
LiJ~,~,kmt
Ilnormal at/~ oa~,~c~oeaee~
29
29
29 51
29 s~
29
29 s*
aO
29 ~2
2Q 52
29 s*
29 ~t
~5
30 s~
30 s~
29
'~.'
SCA-2
so-
eat Numb*r
=
eAGR~
.
_[_l`=
L
,
,
Fig. 1. Interlabo~ato~ comparison of (CAG). repeat determination for normal and expanded aUeles at the DRPLA, SCA-1 through ,3, -6, and -7 loci. DRPLA: Number of laboratories reporting; n = 12. Normal allele: mean, 16; SD, 0;3, mode, 16; range, I5-16. Expanded allele: mean, 69. $D, 1,9; mode, 70; range, 65-72. SCA-I: n = 17. N o d a l alMe: mean, 29, SD, 0.4; mode, 29; range, 29-30. Expanded allele: meant 52; SD, 1.0; mode, 52; range, 51-54, SCA-2: n = 13. Normal allele: mean, 22; SD, 0.6; mode, 22; range, 20-22. Expanded allele: mean, 46; SD, !.4; mode; 47; range, 43-48.
Ataxia Technical Survey and Proficiency Test
•
Potter and N a n c e
95
5CA-3
80-
70SO-
SO-
CAO Re,is! Number 40-
3020-
10.
!
2
3
4
$
d
22 77
22 74
22 77
22 76
22 74
-~ t~
n
76
re
.
~a
7,
75 73 ;,6 7 ~ 1 ~ I
SCA4
CAn Repest Number
2
3
4
5
It
II
II
11
22
22
22
22
--iF 6
7
8
It
II
tO
l!
I!
21
2!
22
23
SCA-7
8O I
Ii
m
50 m
40
I
CAG Repeal Number 3O
2O
fO
LJb*,,I
O
I $
!
II
I l0
10
4
S
II
I0 63
I; $5
II, 5;
Fig. 1. SCA-3: n = 17. Normal allele: mean, 22; SD, 0.4; mode, 22; range, 21-23. Expanded allele: mean, 76; SD, 2.7; mode, 77; range, 73-84. SCA-6: n = 10. Normal allele: mean, 11; SD, 0.3; mode, I 1; range, 10-1 I. Expanded allele: mean, 22; SD, 0,6; mode, 22; range, 21-23. SCA-7: n = 9. Normal allele: mean, 10; SD, 0.7; mode, 10; range, 10-12. Expanded allele: mean; 58; SD, 2.5; mode, 57; range, 55-63.
96
Molecular Diagnosis Vol. 5 No. 2 June 2000
for use in future clinical test development and follow-up diagnostic testing as new autosomal dominant cerebellar ataxia (ADCA) loci are identified.
Discussion The autosomal dominant cerebellar ataxias (ADCAs) represent a clinically heterogeneous group of neurological diseases with at least 12 distinct genetic loci identified to date [5-8]. Because the collective incidence of these disorders is high in both unselected and selected ataxia cohorts [9,10] and the clinical distinctions between them is diagnostically unreliable, the role of molecular diagnostics is significant. At the time of this survey, diagnostic testing was available for six of these disorders (SCA-1 through -3, -6, -7, and DRPLA). However, with the exception of SCA-3 proficiency challenges included in the 1999 College of American Pathologists (CAP)/American College of Medical Genetics Molecular Genetics (ACMG) Survey, there remains no established mechanism to provide proficiency testing for these diseases. In addition, because the CAP/ACMG survey does not provide a mechanism to adch'ess test interpretation or the relative benefits of varying methods in the analysis of unusual cases, the primary initiative for the AMDTG was to develop a comprehensive and timely program for addressing all aspects of ataxia gene testing. In our exchange, proficiency testing showed little difference between laboratories with respect to allele sizing, supporting the contention that all the participating laboratories (71% of all ataxia testing laboratories listed in GeneTests) provided accurate results despite some differences in testing methods. One general comment voiced by the group was the need for a set of disease-specific thxeshold standards that could be used by the AMDTG to accurately genotype patients with large, normal, or small expanded alleles. Whereas the results of proficiency testing showed no major concerns, several diseasespecific issues were discussed at the 1998 meeting of the AMDTG held during the 48th Annual Meeting of the American Society of Human Genetics in Denver, CO.
SCA-1 The literature suggests an overlap in the distribution of normal and pathological alleles carrying be-
tween 39 and 44 repeats [I1-13]. Three samples reported by the AMDTG suggested that this intermediate range extends from 37 to 44 repeats (Table 1). In this series, as suggested in the literature [14], CAT interruption was associated with apparent nonpathogenicity for these alleles. Based on this, the AMDTG reached a consensus that all alleles containing 37 to 44 CAG repeats be analyzed for the presence or absence of the CAT interruption to assist with their genotypic classification. Furthermore, because data suggest age of onset is determined by the number of uninterrupted repeats [15], CAT analysis would be useful in patients with SCA-] with atypical later-onset disease.
SCA-2 Participating laboratories reported an overlap between nornlal and expanded alleles in the 33- to 34-repeat range (Table 1). The recent literatule also suggests an intermediate range between 33 and 35 CAG repeats in which the phenotypic significance of the results is uncertain [16]. Normal alleles containing 30 to 32 repeats have been reported [16], and an allele with 33 repeats has been identified in a population-based study [17]; however, it is currently not known whether this represents a pathological allele associated with reduced penetrance, later age of onset, or a large meiotically stable normal allele. Although the smallest pathological allele reported contains 34 repeats in an individual with late-onset disease [18], we are aware of a symptomatic patient also with late-onset disease who carties an allele with 33 CAG repeats (A.R. La Spada, personal communication, December 1999), The clinical significance of the CAA interruption within the SCA-2 gene is also uncertain [16], and there is presently no clinical assay to determine the presence or absence of the repeat. Allele contraction has also been observed in this disorder, with an allele with 32 repeats reported in an asymptomatic 19-year-old whose affected father carried an allele with 40 repeats [16], Thus, care should be exercised in interpreting results of cases with repeats in this range, Finally; members of the AMDTG have previously reported the identification of "extreme expansions" in patients with the infantile or juvenile forms of SCA-2 in which repeat length can exceed 200 triplets [19]. Although alleles in the repeat range of approximately 200 may be detectable by conventional PCR [19], alternative methods of al-
Ataxia Technical Survey and Proficiency Test
lele detection (i.e., Southern blot or long-range PCR) should be considered in these situations if warranted, particularly if the case is unusual [20].
SCA-3 No significant concerns were noted.
SCA -6 Although no concerns regarding the sizing of SCA-6 alleles were identified in this study, the literature reflects a bimodal distribution of normal and expanded alleles with an overlap at 20 CAG repeats [21,22]. This includes the identification of a patient with a repeat number of 20 and a clinical diagnosis of episodic ataxia type 2 [22]. As such, some caution should be used in interpreting the results in patients carrying alleles of this size.
SCA-7 A review of the literature suggests that the genotypic classification of SCA-7 alleles in the 28- to 35-CAG repeat range remains unclear. Although no participating laboratory reported a patient with an allele in this size range, the classification of these alleles was discussed in light of new data on the subject. The largest reported normal allele contains 35 repeats, and the smallest expanded allele carries 37 repeats [23]. However, the identification of an allele with 36 repeats associated with an at-risk haplotype suggests this is a pathological allele [24]. Because meiotically unstable intermediate alleles with 28 and 35 repeats have also been documented in SCA-7 pedigrees [25,26], an allele within this size range should be considered unstable. As with SCA-2, the recent identification of extreme repeat sizes (>300 triplets) in infantile-onset SCA-7 warrants the availability of molecular testing strategies that will ensure the detection of very large expansions [25-28].
DRPLA No significant concerns were noted. The literature suggests that the collective incidence of these disorders (ADCAs; SCA1, -2, -3, -6, -7) ranges from 28% to 61% in unselected and selected ataxia cohorts, respectively [9,10]. As such, the collective "hit rate" of 5% from the AMDTG is considerably less than that previously reported. This may be the result of several factors, including
•
Potter and Nance
97
the multiplexing of tests (i.e., one patient was simultaneously tested for all six disorders), the genotyping of patients with a molecularly uncharacterized ADCA, and the testing of patients who do not fulfill the clinical criteria for a hereditary ataxia. Although this study did not address which of the approximately 95% of negative tests were ordered appropriately, it is clear that the AMDTG data are more consistent with the incidence of CAG repeat expansion in individuals with no family history of ataxia (i.e., sporadic ataxia) [10]. Although the AMDTG data are based only on the responses of approximately half (10 of 18) of the participating laboratories, it suggests that a greater emphasis may need to be placed on the education of the referring health professional regarding the appropriate use of DNA diagnostic testing strategies in patients with ataxia. Issues related to consent and banking of tested samples remain unresolved and will require further discussion. Forty-six percent of the laboratories required signed consent for diagnostic testing, a statistic consistent With that reported by McGovern et al. [I] in their large quality assurance study of 245 molecular genetic testing laboratories. However, it is interesting that the AMDTG cohort was almost twice as likely to require signed consent for predictive testing than the larger cohort studied by MeGovern et al. [1] (80% vs 44%). This may reflect an increased awareness by the AMDTG of the issues involved with presymptomatic testing for neurologically relevant diseases because many of the laboratories were directly involved in the development of practice guidelines for predictive testing for Huntington disease. Although it is encouraging that the majority of AMDTG laboratories appear to be adhering to established guidelines for predictive testing, it is still somewhat disconcerting that compliance was not 100%. A second point of concern pertains to the banking of diagnostic specimens. No laboratory discarded specimens after testing, and greater than 75% indicated the use of banked samples for future test development and follow-up diagnostic testing. Although this practice was considered routine by the group, less than half the laboratories required signed consent for diagnostic testing, which raises questions regarding the interpretation of the Common Rule and institutional review board requirements for the use of these materials [29]. Although we did not specifically ask whether institutional r e -
98
Molecular Diagnosis, Vol. 5 No. 2 June 2000
view board approval was sought, recent proposals aimed at limiting the retention and use of archived specimens [30]: wan'ants a comprehensive effort toward establishing a consensus regarding the use of archived specimens for clinical test development and validation I29]. These unresolved issues will become focal points for future discussions by the ~'oup. In summary, the organization of the A M D T G consortium has enabled the group to initiate woficiency testing for the ADCAs in a timely manner. It also provides an annual forum to self-monitor laboratory practice and discuss difficult or unusual cases. It is the intent of the group to continue to meet annually to ensm'e the quality and integrity of specialized testing for this group of neurological disorders.
Received December 29, 1999. Received in revised form Janual3, 26, 2000. Accepted February 4, 2000.
8.
9.
10.
11.
12.
13.
R e f e r e n ces 14. 1. McGovern MM, Benach MO, Wallenstein S, Desnick R J, Keenlyside R: Quality assurance in molecular genetic testing laboratories. JAMA 1999; 281:835-840 2. Potter NT, Nance MA, for The Ataxia Molecular Diagnostic Testing Group: Genetic testing for ataxia in North America. Am J Hum Genet 1998;63:A239 (abstr) 3. Wick MJ, Matthias-Hagen VL, Allingham-Hawkins DJ, et al., for The Ataxia Molecular Diagnostic Testing Group: Genetic testing for Friedreich ataxia. Am J Hum Genet 1999;65:A412 (abstr) 4. Watson MS, Altmiller DH, de Martinville B, et al.: Standards mad guidelines for clinical genetics laboratories, 2nd ed. American College of Medical Genetics, Bethesda, 1999 5. Koide R, Ikeuchi T, Onodera O, et al.: Unstable expansion of CAG repeat in hereditary dentatorubral-pallidoluysian atrophy (DRPLA). Nat Genet 1994;6:9-13 6. Koob MD, Moseley ML, Schut LJ, et al.: An untranslated CTG expansion causes a novel form of spinocerebeHar ataxia ~SCAS). Nat Genet 1999;21: 379-384 7. Worth PF, Giunti P, Gardner-Thorpe C, Dixon PH, Davis MB, Wood NW: Autosomal dominant cerebN~ar ataxia type III: Linkage in a Iarge British
15.
16.
17.
18.
19.
20.
family to a 7.6-cm region on chromosome 15q1421.3. Am J Hum Genet 1999;65:420--426 Hohnes SE, O'Hearn E, McInnis MG, et al.: Exp.ansion of a novel CAG repeat in the 5 region of PPP2R2B is associated with SCAI2. Nat Goner 1999;23:391-392 Gunaratue PH, Richards CS: Estimated contribution of known ataxia genes in ataxia patients undergoing DNA testing. Genet Testing 1997/1998;1:275-278 Moseley ML, Benzow KA, Schut LJ, et al.: Incidence of dominant spinocerebellar and Friedreich ataxia triplet repeats among 361 ataxia families. Neurology 1998;51:1666-1671 Matitla T, Volpini V, Genis D, et al.: Presymptomatic analysis of spinocerebellar ataxia type 1 (SCA1) via the expansion o f the SCA1 CAG-repeat in a large pedigree displaying anticipation and parental male bias. Hum Mol Genet 1993;2:2123-2128 Ranum LPW, Chung M-Y, Banff S, et al.: Molecular and clinical cola'elations in spinocerebellar ataxia type 1 (SCA1): Evidence for familial effects on the age of onset. Am J Hum Genet 1994;55:244 252 Quan F, Janas J, Popovich BW: A novel CAG repeat configuration in tbe SCAi gene: Implications for the molecular diagnostics of spinocerebellar ataxia type i, Hum Mol Goner 1995;4:2411-2413 Chung M-Y, Ranum LPW, Duvick L, Servadio A, Zoghbi HY, Orr HT: Evidence for a mechanism predisposing to intergenerational CAG repeat instability in spinocerebellar ataxia type I. Nat Genet 1993; 5:254-258 Matsuyama Z, Izumi Y, Kameyama M, Kawakami H, Nakamura S: Tlie effect of CAT trinucleotide interruptions on the age at onset :of spinocerebellar ataxia type 1 (SCA1). J Meal Genet 1999;36:546548 Pulst S-M: Spinocerebellar ataxia type 2. In Wells RD, Warren ST~ Genetic instabilities and hereditary neurological diseases, 1st ed. Academic Press, San Diego; 1998 Leggo J, Dalton A, Morrison PJ, et al.: Analysis of spinocerebellar ataxia types 1, 2, 3, and 6, denta, torubral-pallidoluysian atrophy, and Friedreich's ataxia genes in spinocerebellar ataxia patients in tlae UK, J Med Genet 1997;34:982,985 Malandrini A, Galli Iz, Villanova M, et ai.: CAG repeat expansion in an Italian family with spinoCerebellar ataxia type 2 (SCA2): A clinical and genetic study. Eur Nem'oi 1998;40:164-168 Babovic-Vuksanovic D, Snow K, Patterson MC, Miclaels VV: SpinocerebeHar ataxia type 2 (SCA2) in an infant with extreme CAG repeat expansion. Am:J Med Genet 1998;79:383,387 Mao R, Snow K: Screening for large CAG-repeat expansions in the SCA2 and SCA7 genes in infantile
Ataxia Technical Survey and Proficiency Test
21.
22.
23.
24.
25.
and juvenile-onset spinocerebellar ataxia (SCA). J Mol Diagn 1999;1:43 (abstr) Ishikawa K, Tanaka H, Saito M, et al.: Japanese families with autosomal dominant pure cerebellar ataxia map to chromosome 19p 13.1-p 13.2 and are strongly associated with mild CAG expansions in the spinocerebellar ataxia type 6 gene in chromosome 19p13.1. Am J Hum Genet 1997;61:336-346 Jodice C, Mantuano E, Veneziano L, et al.: Episodic ataxia type 2 (EA2) and spinocerebellar ataxia type 6 (SCA6) due to CAG repeat expansion in the CACNA1A gene on chromosome 19p. Hum Mol Genet 1997;6:1973-1978 David G, Durr A, Stevanin G, et al.: Molecular and clinical correlations in autosomal dominant cerebellar ataxia with progressive macular dystrophy (SCA7). Hum Mol Genet 1998;7:165-170 Benton CS, de Silva R, Rutledge SL, Bohlega S, Ashizawa T, Zoghbi HY: Molecular and clinical studies in SCA-7 define a broad clinical spectrum and the infantile phenotype. Neurology 1998;51: 1081-1086 Stevanin G, Giunti P, Belal GDS, et al.: De novo
26.
27.
28.
29.
30.
•
Potter and Nance
99
expansion of intermediate alleles in spinocerebellar ataxia 7. Hum Mol Genet 1998;7:1809-1813 Giunti P, Stevanin G, Worth PF, David G, Brice A, Wood NW: Molecular and clinical study of 18 families with ADCA type II: Evidence for genetic heterogeneity and de novo mutation. Am J Hum Genet 1999;64:1594-1603 Gunaratne PH, Gannavarapu A, Palmer S, Richards CS: Improvement in clinical genetic testing for SCA7: Development of a Southern to detect extremely large expansions. Am J Hum Genet 1999;65:A299 (abstr) Matsuura T, Khajavi M, de Silva R, Ashizawa T: A very large SCA7 CAG expansion is compatible with cell viability in somatic mosaicism. Am J Hum Genet 1999;65:A460 (abstr) Merz JF, Leonard DGB, Miller ER: IRB review and consent in human tissue research. Science 1999;283: 1647-1648 Clayton EW, Steinberg KK, Khoury MJ, et al.: Informed consent for genetic research on stored tissue samples. JAMA 1995;274:1786-1792