Prenatal diagnosis for risk of spinal muscular atrophy

BJOG: an International Journal of Obstetrics and Gynaecology November 2002, Vol. 109, pp. 1244 1249 Prenatal diagnosis for risk of spinal muscular atrophy I. Cuscó a, M.J. Barceló a, C. Soler a, J. Parra b, M. Baiget a, E. Tizzano a, * Objectives Prenatal diagnosis of spinal muscular atrophy is usually performed in high risk couples by detection of a homozygous deletion in the survival motor neurone gene (SMN1). However, other relatives at risk of being carriers very often request genetic counselling and the possibility of prenatal diagnosis. The aim of this study was to validate a SMN1 gene quantitative test to help the couples formed by one spinal muscular atrophy carrier and a partner of the general population (1/200 potential risk) to achieve a less ambiguous risk result for the pregnancy. Design Spinal muscular atrophy carrier studies in at-risk individuals. Setting Department of Genetics and Gynaecology and Obstetrics in a large university hospital. Population Seventy-nine obligate carriers (more than one affected child with deletion in the offspring) and 58 non-carriers (relatives of spinal muscular atrophy families defined by marker studies) were tested to set up a quantitative analysis. The method was applied in different situations in 126 members from 34 families with spinal muscular atrophy patients. Methods DNA studies of the SMN1 gene by marker analysis and quantitative assay. Main outcome measures To determine double (non-carrier) or single dose (carrier) of exon 7 of the SMN1 gene in relatives of spinal muscular atrophy patients. Bayesian calculation of risk. Results The sensitivity and specificity of the method were 96% and 100%, respectively. Studies on different couples with an a priori risk of 1/200 allowed us to reduce the final risk to 1/5000 or to increase it to 1/4. Conclusions The quantitative method can be used to achieve a less ambiguous risk in pregnancies with a 1/200 risk and in families where no sample is available to study the index case. Screening of gamete donors when the recipient is a known carrier should also be considered. INTRODUCTION Spinal muscular atrophy is an autosomal recessive disorder characterised by degeneration and loss of the alphamotor neurones in spinal cord anterior horn cells, with an incidence of approximately 1/6000 to 1/10,000 live births. Spinal muscular atrophy is the second most common severe autosomal recessive disease in Caucasians after cystic fibrosis, being one of the major hereditary causes of infant mortality. Carrier frequency is approximately 1/40 to 1/60 1,2. Spinal muscular atrophy patients can be classified into three groups based on age at onset achieved milestones and life span 3,4. Type I spinal muscular atrophy (Werdnig Hoffman disease) is the severe form with clinical onset before the age of six months. Type II spinal muscular atrophy is the intermediate form with onset before 18 months. Type III spinal muscular atrophy (Kugelberg Welander disease) is the mild form of the disease with a Genetics and Research Institute, Hospital de Sant Pau, Barcelona, Spain b Department of Obstetrics and Gynecology, Hospital de Sant Pau, Barcelona, Spain * Correspondence: Dr E. F. Tizzano, Genetics and Research Institute, Hospital of Sant Pau, Padre Claret 167, 08025 Barcelona, Spain. D RCOG 2002 BJOG: an International Journal of Obstetrics and Gynaecology PII: S1470-0328(02)02983-X onset after the age of 18 months. These three clinical forms are caused by the homozygous absence of the survival motor neurone (SMN1) gene located at 5q13. The SMN2 gene, a highly homologous copy of the SMN1 gene, is located in the same locus as part of a repeated region and is always present in spinal muscular atrophy patients. Approximately 90% of patients suffering from different forms of spinal muscular atrophy lack two copies of the SMN1 gene exons 7 and 8. A small number of spinal muscular atrophy patients show homozygous deletions of SMN1 exon 7 but not exon 8. In these patients, hybrid genes were characterised. These hybrid genes result from the fusion of exon 7 of SMN2 with exon 8 of SMN1 5. In the cases where the SMN1 gene is present, subtle mutations have been detected, indicating that the SMN1 gene is the determining gene of the disease. Given the high frequency of the homozygous absence of SMN1 observed in spinal muscular atrophy patients and in accordance with the Hardy Weinberg equilibrium, 99.7% of all spinal muscular atrophy patients must carry at least one SMN1 deletion in one chromosome 6. The availability of a useful molecular test for spinal muscular atrophy has facilitated the diagnosis of the disease. However, these routine tests do not allow detection of the hemizygous absence of the SMN1 gene, which characterises most of the cases with subtle mutations and the carriers of the disease. Relatives who are at risk of being carriers usually request genetic www.bjog-elsevier.com

CARRIER AND PRENATAL DIAGNOSES OF SMA 1245 counselling and carrier and/or prenatal diagnoses by molecular analysis. Given that the normal copy of the gene masks the deletion when routine PCR-based methods are used, carrier diagnosis in these cases relies on gene tracking with polymorphic markers of the 5V end of the SMN1 gene. However, this method is unsuitable for some families where some key members are absent. When a carrier in one family is assigned by marker analysis, his or her partner has an a priori risk of 1/50 of being carrier (carrier frequency of the general population) 1,2. According to this figure, the final risk of these couples is approximately 1/200 (1/4 1/50). In our experience, a significant proportion of women of these couples, when pregnant, ask the obstetrician for advice on carrier detection and prenatal diagnoses. We employed a quantitative method of dosage of the SMN1 exon 7 gene to allow diagnosis of hemizygous patients and carriers. The objectives were (1) to validate the spinal muscular atrophy carrier method by testing known carriers and non-carriers and (2) to study the couples formed by one spinal muscular atrophy carrier and a partner of the general population in order to achieve a less ambiguous risk for their pregnancies. Moreover, we performed molecular diagnosis in couples with a family history of Werdnig Hoffmann disease but without an affected member available for analysis. We describe a number of situations in which quantitative analysis has proved useful in taking more confident decisions about prenatal diagnosis. METHODS A total of 137 individuals were analysed to establish the sensitivity and specificity of the method. Seventy-nine were parents of multiple affected children (two or more) with homozygous deletion of SMN1 exon 7 (obligate carriers); the remaining 58 individuals were unaffected siblings or relatives of spinal muscular atrophy patients with a non-risk haplotype by linkage analysis (markers C212 and C272). Quantitative analysis was applied in 126 members from 34 families. The members studied were parents, sibs (a priori risk of being a carrier 2/3), uncles/aunts (a priori risk of 1/2), cousins (a priori risk of 1/4), grandparents (a priori risk of 1/2) and partners of carriers (a priori risk of 1/50). Furthermore, 21 prenatal diagnoses were requested in these families. Ten were performed in couples with a risk of 1/4. In the remaining 11 cases, two were couples with a history of an affected member without DNA available for study (putative risk of 1/4). The other nine were couples with a risk of 1/200. Genomic DNA was isolated from peripheral blood by the salting out method as previously described 7. Screening for deletions in the SMN gene was performed according to van der Steege et al. 8. The PCR products were visualised on ethidium bromide 1.5% agarose gel after electrophoresis. D5F149S1-S2 and D5F150S1-S2 loci 2 were typed using C272/Ag1-CA and C212 polymorphic markers as previously described 5. Amplification products were separated on a 6% polyacrylamide 8 M urea gel, transferred to a nylon membrane, hybridised with a 5V biotinylated oligonucleotide GT probe and visualised by an enhanced chemiluminescence method (ECL, Amersham). For the quantitative SMN1 exon 7 assay we used the method described by Scheffer et al. 9 with slight modifications including the use of FAM as a fluorescent dye and the processing of the samples in an ABI Prism 310 (Gene Scan software, Perkin Elmer-Applied Biosystems). Briefly, the amount of SMN exon 7 PCR product was compared with the amount of a coamplified PCR product of the retinoblastoma (RB1) exon 13 containing a DraI restriction site. Amplification efficacy was monitored by coamplifying internal standards for SMN and RB genes. The peak area for each fragment was compared multiplying the SMN1/RB ratio by RB-IS/SMN-IS (Fig. 1). The number of copies of the SMN1 gene was normalised for the mean of the ratios obtained in two standard reference samples from individuals carrying one copy of the SMN1 gene. RESULTS Of 79 parents of multiple affected children with spinal muscular atrophy, 76 presented a SMN1/Rb genomic normalised ratio of 0.95 (0.2), corresponding to a single SMN1 exon 7 dose. In 3 of 79 known carrier samples, the ratio was compatible with a double dose of SMN1 exon 7 such as 1.6, 1.8 and 1.9 (calculate sensitivity of 96.2%) (Fig. 2). All 58 non-carriers had SMN1/Rb genomic normalised ratio of 2.03 (0.3) compatible with double dose or, in a few cases, triple dose (calculate specificity 100%) (Fig. 2). Bayesian calculation To calculate the final risk, an a priori risk of being a carrier of 1/50 was used together with a sensitivity of 96% and a specificity of 100%. For an individual with a single SMN1 dose, the final risk of being a carrier was 99.98, whereas for double dose the final risk of being carrier was 0.0008 (1/1250). Applications of the method We employed the quantitative method of carrier analysis to study 126 individuals from 34 families in the following situations: 1. When a known carrier has a child with a partner from the general population (Fig. 3, Family 1): In this situation, in the absence of a family history of the disease in

1246 I. CUSCÓ ET AL. Fig. 1. Quantitative analysis of the SMN1, SMN2 and RB genes for spinal muscular atrophy. The peak area for each fragment was compared multiplying the SMN1/RB or SMN2/RB ratio by RB-IS/SMN-IS. Normalised values are indicated on the right. A ¼ example of a carrier value; B ¼ example of a non-carrier value. the partner, the a priori risk of being carrier is approximately 1/50 (general population) and the probability of having an affected child is 1/200. After quantitative analysis of the partner, the final risk of having an affected child would be as low as 1/5000 or as high as 1/4. This situation was studied in two families. By means of assisted reproductive techniques, some couples may opt for semen or oocyte donors as an alternative to lowering their risk of recurrence. The family presented here (Fig. 3, Family 2) has been discussed elsewhere 10. After a first artificial insemination (because of the husband s infertility), dizygotic twins were born. One of them had spinal muscular atrophy and a homozygous deletion of the SMN1 gene was detected. The next pregnancy was again the product of an artificial insemination but from a different semen donor. The a priori risk in this situation was 1/200, assuming that the mother was a carrier and that the new semen donor had the carrier risk of the general population. A prenatal diagnosis was requested and another affected fetus was detected. A quantitative analysis in the DNA sample of the second donor showed a single dose of SMN1, confirming that he was a spinal muscular atrophy carrier. 2. Carrier diagnosis in parents when the index case is not available (Fig. 3, Family 3): Given that type I spinal muscular atrophy is a severe disease, a proportion of couples request carrier or prenatal diagnosis without the SMN1 EXON7 COPY NUMBER 4 3 2 1 0 CARRIER NON CARRIER Fig. 2. Histogram of normalised SMN1/RB values in carriers and non-carriers of spinal muscular atrophy. The arrow indicates the three cases of carriers with a double dose (2/0 genotypes) (see text). Note the three extreme non-carrier subjects with values compatible with a triple dose of SMN1.

CARRIER AND PRENATAL DIAGNOSES OF SMA 1247 1 2 sd1 sd2 3 4 5 6 7 Fig. 3. Pedigrees of the spinal muscular atrophy families illustrating the different situations where quantitative SMN1 analysis was applied (see Applications of the method in the Results section). Arrows indicate the key members where the quantitative analysis was indicated. sd1 and sd2 ¼ semen donors 1 and 2, respectively. index case available for study. A quantitative analysis of the parents could be applied to resolve this situation. A single dose detected in both parents would indicate that the index case was the result of a homozygous deletion of SMN1. Two families requested carrier and prenatal testing in these circumstances (Table 1). 3. Carrier diagnosis in sibs when the index case is not available (Fig. 3, Family 4): Most cases of carrier diagnosis in sibs of an affected patient can be resolved by marker analysis, depending on the availability of the parent s DNA. However, quantitative analysis was particularly useful when the affected patient was dead and no DNA sample was available for study. Seven individuals belonging to six families requested analysis in this situation. 4. Carrier diagnosis in uncles/aunts (Fig. 3, Family 5): As in the case of sibs, exclusion can be achieved by marker analysis. However, a complete analysis demands a study of the grandparents, which are not always available. Moreover, once a carrier status is assigned, analysis of the partner can be performed as in situation one. Twentyeight individuals belonging to 17 families requested this analysis. 5. Carrier diagnosis in cousins (Fig. 3, Family 6): In this situation, the availability of relatives to perform marker studies may be more difficult. Then, quantitative analysis would allow further diagnosis and, depending on the results, analysis of the partners could be offered. One family requested this analysis. Table 1. Prenatal diagnoses requested by 11 couples with a priori risk of 1/4 (the index case was not available) and 1/200. A priori risk Risk after quantitative analysis Affected Carrier Non-Carrier 1/4 (n ¼ 2) 1/4 1 1 1/200 (n ¼ 9) 1/4 (n ¼ 2) 1 1 1/5000 (n ¼ 7) 3 4 Total ¼ 11 2 4 5

1248 I. CUSCÓ ET AL. 6. Determination of the branch of the family in which the mutation was transmitted (Fig. 3, Family 7): Seconddegree relatives who wanted to know whether they were carriers requested the analysis. Determination of the branch of transmission by analysis of the grandparents of the patients avoids unnecessary testing of relatives. Sixteen cases from seven families were analysed. Furthermore, prenatal diagnosis was requested by two couples with a probable 1/4 risk (the index case was not available and DNA test was not performed) and by nine couples with a risk of 1/200. A retrospective quantitative analysis was performed in these nine families resulting in two couples with a risk of 1/4 and seven couples with a risk of 1/5000 (Table 1). DISCUSSION We used quantitative analysis of the SMN1 gene, demonstrating its usefulness in different situations. The method was validated with the analysis of spinal muscular atrophy known carriers and non-carriers by haplotype analysis. All carriers defined by multiple homozygously deleted affected children might show a single dose of the SMN1 gene. However, 3 of 79 parents of multiple-affected children had a double dose compatible with two SMN1 copies. Given that de novo deletions were discarded by the fact that the mutation was transmitted more than once, the most likely explanation for this exception is the existence of two SMN1 copies in one chromosome and a null allele in the other (2/0). This phenomenon was described previously by McAndrew et al. 11, Wirth et al. 12 and Scheffer et al. 9 and accounts for approximately 3 4% of carriers. The explanation was supported by the fact that a small percentage (4 6%) of control subjects was previously reported harbouring three or four copies of SMN1 11,12. We found that 3 out of 58 control individuals (5%) were carriers of a triple dose of SMN1. The fact that some carriers may have a double SMN1 dose would bear on the sensitivity of the method, yielding a value of 96.2%. One important aspect is the fact that subtle mutations may occur in spinal muscular atrophy and these account for 3.4% of the cases 6. These mutations cannot be detected by quantitative carrier testing, which can therefore lead to false negative results. In our population, a 4 bp deletion in exon 3 was detected in approximately 2.5% of the spinal muscular atrophy cases 13. With further analysis of this 4 bp deletion, the remaining mutant alleles in our population should represent less than 1%. The sensitivity and specificity figures mentioned above are acceptable for carrier detection among random individuals without a positive history of spinal muscular atrophy, particularly in the case of partners of known spinal muscular atrophy carriers. These individuals have an a priori theoretic risk of 1/50 of being carriers. If a subject has a single dose result, then the probability of being a carrier is 0.9998. On the other hand, a double dose will result in a probability of 1/1250. Although this method has not been applied to the general population, it can be used in couples with an intermediate 1/200 a priori risk of having an affected spinal muscular atrophy child. This figure is regarded as ambiguous by most couples. However, after the test, when the partner shows a single SMN1 dose, the consequence is a high potential risk (1/4) that deserves a prenatal testing. By contrast, when the subject shows a double dose, the final risk is approximately 1/5000. This figure is close to the incidence of the disease in the general population and will influence the decision of the couple not to request a prenatal testing. It could be argued that couples with 1/200 do not merit special attention since they do not have a high risk, such as couples with 1/4. However, in our experience, the theoretic risk of these couples should not be undervalued given that in this study, two cases of single dose out of 17 individuals without family history of spinal muscular atrophy were detected (one husband and a semen donor). The analysis of a larger sample of individuals is warranted to confirm the hypothesis that the carrier frequency in Spain could be higher than in other populations. These situations are examples that favour the need for couples with a risk of 1/200 of performing a quantitative test in the non-spinal muscular atrophy family member. Furthermore, genetic screening of spinal muscular atrophy carrier status should be performed in gamete donors when the recipient is a known carrier. CONCLUSION The quantitative method for SMN gene dosage is a useful tool for couples formed by one spinal muscular atrophy carrier and a subject of the general population with 1/200 potential risk of having an affected child. The quantitative study allows us to modify the final risk to facilitate the decision of the couple. Furthermore, this analysis can be applied to families where the index case is death and no sample is available for study. The presence of a single dose in these parents would indicate that both are carriers of the deletion and, as a consequence, it would be possible to offer prenatal diagnosis and to examine other possible carriers in the family. The fact that about 3 4% of the carriers will have a 2/0 genotype must be considered a biological pitfall of this methodology. However, most of the situations can be resolved if genetic counselling is given prior to carrier testing so that the advantages and limitations of the analysis can be explained. Acknowledgements This work was supported by Telemarató Malalties Hereditaries 98/810 and FIS 00/481.

CARRIER AND PRENATAL DIAGNOSES OF SMA 1249 The authors would like to thank all pediatricians, neurologists, Spanish spinal muscular atrophy families and María Amenedo for their collaboration and E. Del Río, R. Stulp, H. Scheffer and C. Buys for help and advice with the quantitative testing. References 1. Pearn J. Classification of spinal muscular atrophies. Lancet 1980; I:919 922. 2. Melki J, Lefebvre S, Burglen L, et al. De novo and inherited deletions of the 5q13 region in spinal muscular atrophies. Science 1994;3: 1474 1477. 3. Munsat TM, Davies KE. Meeting report: International SMA Consortium meeting. Neuromuscul Disord 1992;2:423 428. 4. Zerres K, Rudnik-Schoneborn S. Natural history in proximal spinal muscular atrophy (SMA) clinical analysis of 445 patients and suggestions for modification of existing classifications. Arch Neurol 1995;52:518 523. 5. Cuscó I, Barceló MJ, del Río E, et al. Characterization of SMN hybrid genes in Spanish SMA patients: de novo, homozygous and compound heterozygous cases. Hum Genet 2001;108:222 229. 6. Wirth B. An update of the mutation spectrum of the survival motor neuron gene (SMN1) in autosomal recessive spinal muscular atrophy. Human Mutat 2000;15:228 237. 7. Miller SA, Dykes DD, Polesky H. A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1989;16:1215. 8. van der Steege G, Grootscholten PM, van der Ulics P, et al. PCR based DNA test to confirm clinical diagnosis of autosomal recessive spinal muscular atrophy [letter]. Lancet 1995;345:985 986. 9. Scheffer H, Cobben JM, Mensink RGJ, Stulp RP, van der Steege G, Buys C. SMA carrier testing validation of hemizygous SMN exon 7 deletion test for the identification of proximal spinal muscular atrophy carriers and patients with a single allele deletion. Eur J Hum Genet 2000;8:79 86. 10. Tizzano EF, Cuscó I, Barceló MJ, Parra J, Baiget M. Should gamete donors be tested for spinal muscular atrophy. Fertil Steril 2002; 77:409 411. 11. McAndrew PE, Parson DW, Simard LR, et al. Identification of proximal spinal muscular atrophy carriers and patients by analysis of SMNT and SMNC gene copy number. Am J Hum Genet 1997;60: 1411 1422. 12. Wirth B, Herz M, Wetter A, et al. Quantitative analysis of survival motor neuron copies: identification of subtle SMN1 mutations in patients with spinal muscular atrophy, genotype phenotype correlation, and implications for genetic counseling. Am J Hum Genet 1999;6:1340 1356. 13. Bussaglia E, Clermont O, Tizzano E, et al. A frameshift deletion in the survival motor neuron gene in Spanish spinal muscular atrophy patients. Nat Genet 1995;11:335 337. Accepted 18 June 2002