Summary
Background
Hexanucleotide repeat expansions in chromosome 9 open reading frame 72 (C9ORF72) are currently the major genetic cause of frontotemporal dementia (FTD) and motor neuron disease (MND). Presently, it is unknown whether expansion size affects disease severity or phenotypes.
Methods
We performed a cross-sectional Southern blot characterization study (Xpansize-72) in a cohort of subjects obtained at the Mayo Clinic and Banner Sun Health Research Institute. All subjects carried GGGGCC repeat expansions in C9ORF72, and high quality DNA was available from the frontal cortex, cerebellum and/or blood. Southern blotting techniques and densitometry were employed to estimate the repeat size of the most abundant expansion species. Comparisons of repeat sizes between tissues were made using Wilcoxon rank sum and Wilcoxon signed rank tests, and between disease subgroups using Kruskal-Wallis rank sum tests. The association of repeat size with age at onset and age at collection was evaluated using a Spearman’s test of correlation; whereas the association between repeat size and survival after disease onset was examined using Cox proportional hazards regression models.
Findings
Our cohort consisted of 84 C9ORF72 expansion carriers, including FTD patients (n=35), FTD/MND patients (n=16), MND patients (n=30), and unaffected subjects (n=3). We focused our analysis on three major tissue subgroups: frontal cortex (41 subjects [21 FTD, 11 FTD/MND, 9 MND]), cerebellum (40 subjects [20 FTD, 12 FTD/MND, 8 MND]), and blood (50 subjects [15 FTD, 9 FTD/MND, 23 MND, 3 unaffected expansion carriers]). Repeat lengths in the cerebellum were significantly smaller (median 12·3 kb [~1667 repeat units], IQR 11·1–14·3) than in the frontal cortex (median 33·8 kb [~5250 repeat units], IQR 23·5–44·9, p<0·0001), or in blood (median 18·6 kb [~2717 repeat units], IQR 13·9–28·1, p=0·0002). Within these tissues, there was no significant difference in repeat length between disease subgroups (cerebellum p=0·96, frontal cortex p=0·27, blood p=0·10). In the frontal cortex of FTD patients, repeat length correlated with age at onset (r=0·63, p=0·003) and age at collection (r=0·58, p=0·006); this correlation was not detected in the cerebellum or blood. Finally, only in the cerebellum, survival after disease onset was poorer in patients from our overall cohort with repeat lengths greater than 1467 repeat units (25th percentile, HR 3·27, 95% CI 1·34–7·95, p=0·009): the median survival was 4·8 years (IQR 3·0–7·4) in the group with longer expansions versus 7·4 years (IQR 6·3–10·9) in the group with smaller expansions.
Interpretation
Substantial variation in repeat size is observed between cerebellum, frontal cortex, and blood; relatively long repeat sizes in the cerebellum confer an important survival disadvantage. Our findings indicate that expansion size does affect disease severity, which could be relevant for genetic counseling.
Introduction
Repeat expansions in chromosome 9 open reading frame 72 (C9ORF72) are the most common known genetic cause of frontotemporal dementia (FTD) and motor neuron disease (MND).1,2 It has been proposed that these diseases result from the formation of toxic RNA foci when repeats are transcribed,1 from the generation of aggregating dipeptide-repeat proteins due to non-ATG translation,3,4 and/or from a loss-of-function of the C9ORF72 protein.5,6
While repeat-primed PCR methods are able to assess the presence or absence of C9ORF72 repeat expansions, Southern blotting techniques are needed to determine the actual size of the GGGGCC repeat. Southern blotting, however, is time-consuming and challenging, and therefore, very limited studies have thus far been performed. A recent Southern blot study did investigate a clinical cohort of 57 cases with a range of neurodegenerative diseases,7 but large scale Southern blot studies that examine repeat sizes in brain have yet to be reported. To address critical questions, such as whether repeat length is associated with the clinical phenotype, disease onset, or survival in patients with hexanucleotide repeat expansions, we performed an extensive Southern blot characterization study (Xpansize-72) of more than 200 DNA samples from C9ORF72 expansion carriers.
Methods
Participants and study design
For our pathological cohort, we selected all C9ORF72 repeat expansion carriers diagnosed with FTD and/or MND from the Mayo Clinic Florida Brain Bank and Banner Sun Health Research Institute, for whom sufficient tissue from the frontal cortex and/or cerebellum was available to allow extraction of high quality DNA for Southern blotting. For our clinical cohort, we selected one patient from each family carrying C9ORF72 repeat expansions; FTD (and FTD/MND) cases were ascertained as part of the Alzheimer’s Disease Research Center (ADRC) at Mayo Clinic Florida and Minnesota, and MND (and FTD/MND) cases were ascertained as part of the Mayo Clinic Florida ALS Center. Of those subjects for whom blood was collected at multiple time points we included all samples, and additionally, we selected one family from our FTD subgroup and one family from our MND subgroup in whom material of no less than four expansion carriers (both affected and unaffected) could be obtained. Subjects were only included in our clinical cohort when DNA of sufficient quality could be extracted from blood to allow Southern blotting (web appendix p 1).
Written informed consent was obtained before study entry from all subjects and/or their legal next of kin, and biological samples were obtained with ethical committee approval.
Procedures
Southern blotting was performed as described by DeJesus-Hernandez et al.,1 with minor modifications. DNA was isolated from frozen tissue, blood, or cell lines following standard procedures (web appendix p 2). A total of 7–10 μg of genomic DNA was digested with XbaI, and electrophoresed for approximately six hours at 100V in a 0·8% agarose gel. DNA was then transferred to a positively charged nylon membrane (Roche, Penzberg, Upper Bavaria, Germany) and cross-linked by UV irradiation. After prehybridization in digoxygenin (DIG) EasyHyb solution (Roche), hybridization with a DIG-labeled probe was performed. Anti-DIG antibody (1:10,000, Roche) was used to detect the probe, which was subsequently visualized with CDP-star substrate (Roche) on X-ray film. Multiple exposures between 30 minutes and six hours were obtained, to minimize differences in signal strength between Southern blots, and to account for a decrease in signal strength over time. Sixty-three Southern blots for a total of 495 samples (203 individual DNA samples) were performed; samples were analyzed 1–8 times, in order to obtain the most accurate repeat sizes, depending on the number of comparisons for a particular sample (between diseases, patients, and tissues), the quality of the Southern blots, and the amount of DNA that was available to us. The most abundant expansion sizes were estimated using AlphaEaseFC (version 4.0.0; Alpha Innotech, Santa Clara, CA, USA)by calculating the molecular weight of bands, based on their position relative to molecular weight markers. When samples were measured multiple times for a given tissue, the mean of these measures was calculated and used in statistical analyses. Repeat expansions above 48·5 kb were considered as expansions of 50 kb in all statistical analyses, which is a conservative prediction of their actual repeat length (web appendix p 3).
Statistical analysis
Data was summarized with median and interquartile range (IQR). For the overall cohort and for each disease subgroup separately, pair-wise comparisons of C9ORF72 repeat lengths between frontal cortex, cerebellum, blood, and temporal cortex were made using a Wilcoxon rank sum test in all instances (including all individuals with either sample available for the given comparison), except for the comparison between frontal cortex and cerebellum, where a paired Wilcoxon signed rank test was performed (including only individuals with both samples available). In the overall cohort, when comparing repeat length between the cerebellum and temporal cortex we also employed a paired Wilcoxon signed rank test, using only the 18 subjects for whom both measures were available. For those comparisons of repeat length in blood with frontal cortex and cerebellum, the overlap was small, and therefore, we utilized a stratified van Elteren Wilcoxon rank sum test to account for any potential confounding factors in our unpaired analyses.8
To determine the degree of agreement in repeat lengths obtained at two different time points, we estimated Lin’s concordance correlation coefficient, which ranges from −1 to 1 (1 indicates perfect agreement).9 Comparisons of repeat lengths between disease subgroups were made using a Kruskal-Wallis rank sum test. In the overall cohort and separately for each disease subgroup, associations of repeat lengths with clinical characteristics were assessed using a Spearman’s test of correlation, a Wilcoxon rank sum test, or a Kruskal-Wallis rank sum test. Spearman’s correlation coefficient r and 95% confidence interval (CI) were estimated. Of note, a non-parametric approach was used in the aforementioned analyses, since these tests have comparable efficiency in relation to their parametric equivalents when the assumptions of the parametric tests are met, and they are more efficient than their parametric equivalents when assumptions of the parametric tests are not met.
In addition, we investigated associations between repeat length and survival after disease onset in all subjects, and in each disease subgroup separately, using Cox proportional hazards regression models. Hazard ratios (HRs) and 95% CIs were estimated, censoring occurred at the date of last follow-up, and deaths of any cause were utilized as our survival endpoint. Both single variable and multivariable Cox regression models were utilized; multivariable models were considered when such adjustment may be reasonable, despite the relatively small sample size and correspondingly low number of patients who died.10 All models in the overall cohort were adjusted for disease type. Additionally, for models in the overall cohort involving repeat length in the cerebellum and frontal cortex, we also adjusted for age at onset. Alternatively, in the overall cohort, we adjusted for age at collection in models involving repeat length in the frontal cortex, to specifically address the confounding potential of this variable. Multivariable models in FTD patients involving repeat length in the frontal cortex and cerebellum were adjusted for age at onset. Multivariable analysis was not performed for any other models due to the small number of patients who died in the given group. Repeat length was considered as a continuous variable in Cox regression analysis, and also as a dichotomous categorical variable using three different cut-off points (median, 25th percentile, 75th percentile) in the given subject group.
In order to adjust for multiple testing and to control the family-wise error rate at 5%, we utilized a Bonferroni correction separately for each disease subgroup, and for each group of similar comparisons (web appendix p 3). All statistical analyses were performed using SAS (version 9.2; SAS Institute, Inc., Cary, NC, USA) and R Statistical Software (version 2.14.0; R Foundation for Statistical Computing, Vienna, Austria).
Role of the funding source
Sponsors of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report. The corresponding author had full access to all data in the study, and she had final responsibility for the decision to submit for publication.
Results
Participants
A total of 84 C9ORF72 expansion carriers from the Mayo Clinic (n=80) or Banner Sun Health Research Institute (n=4) were included in the Xpansize-72 study (table 1, web appendix p 1, 4, 5). Of these 84 subjects, 42 were pathologically diagnosed patients, and the remaining 42 were clinically diagnosed patients (n=39) or unaffected subjects (n=3). Thirty-five patients were diagnosed with FTD (21 pathological, 14 clinical), 16 were diagnosed with FTD/MND (12 pathological, four clinical), 30 were diagnosed with MND (nine pathological, 21 clinical), and three were unaffected, but reported a positive family history for FTD and MND (age at collection: 29, 31, and 44 years).
Table 1.
Subject characteristics, and repeat length in the overall cohort and in each disease subgroup
| Variable | Overall cohort (n=84) | FTD patients (n=35) | FTD/MND patients (n=16) | MND patients (n=30) |
|---|---|---|---|---|
| Patient characteristics | ||||
| Gender (male) | 43 (51%) | 20 (57%) | 9 (56%) | 14 (47%) |
| Age at onset (years) | 56·7 (50·5–65·5) | 60·6 (53·0–68·0) | 56·0 (52·0–66·2) | 52·3 (48·5–58·5) |
| Age at collection (years) | 62·0(51·7–71·4) | 65·4 (58·2–74·6) | 63·7 (55·5–71·7) | 54·5 (49·6–61·4) |
| Disease duration (years) | 4·8 (2·5–7·9) | 7·8 (5·9–10·9) | 5·2 (4·0–6·7) | 2·4 (1·3–3·5) |
| Family history (positive) | 49 (59%) | 20 (59%) | 9 (56%) | 17 (57%) |
| Method of diagnosis (pathological) | 42 (50%) | 21 (60%) | 12 (75%) | 9 (30%) |
| Diagnosis | ||||
| FTD | 35 (42%) | .. | .. | .. |
| FTD/MND | 16 (19%) | .. | .. | .. |
| MND | 30 (36%) | .. | .. | .. |
| Currently unaffected | 3 (4%) | .. | .. | .. |
| TDP-43 type | ||||
| 1 | 16 (44%) | 13 (81%) | 3 (27%) | 0 (0%) |
| 2 | 1 (3%) | 1 (6%) | 0 (0%) | 0 (0%) |
| 3 | 19 (53%) | 2 (13%) | 8 (73%) | 9 (100%) |
| Length of wild-type allele | ||||
| 2 | 53 (63%) | 22 (63%) | 13 (81%) | 17 (57%) |
| 5–7 | 15 (18%) | 6 (17%) | 3 (19%) | 4 (13%) |
| 8–17 | 16 (19%) | 7 (20%) | 0 (0%) | 9 (30%) |
| Repeat length (kb) | ||||
| Frontal cortex | 33·8 (23·5–44·9) | 33·9 (23·1–44·9) | 38·4 (25·6–50·0) | 25·8 (23·9–31·6) |
| Cerebellum | 12·3 (11·1–14·3) | 12·2 (10·9–14·9) | 12·6 (10·3–15·8) | 12·3 (11·3–13·2) |
| Blood | 18·6 (13·9–28·1) | 17·5 (7·3–26·5) | 15·2 (8·0–21·0) | 21·4 (17·1–29·4) |
| Repeat length - pair-wise comparison p-values | ||||
| Frontal cortex vs. cerebellum | <0·0001 | <0·0001 | 0·001 | 0·008 |
| Frontal cortex vs. blood | 0·0001 | 0·009 | 0·005 | 0·28 |
| Cerebellum vs. blood | 0·0002 | 0·61 | 0·55 | 0·0002 |
Data are sample median (IQR), number (%), or pair-wise comparison p value regarding repeat length. Pair-wise comparison p values result from a Wilcoxon rank sum test, with the exception of frontal cortex vs. cerebellum comparisons which result from a Wilcoxon signed rank test. Information was unavailable regarding age at onset (n=3; 1 FTD, 2 FTD/MND), age at collection (n=1; FTD/MND), disease duration (n=3; 1 FTD, 2 FTD/MND), family history (n=1; FTD), TDP-43 type (n=45; 19 FTD, 5 FTD/MND, 21 MND), repeat length in the frontal cortex (n=43; 14 FTD, 5 FTD/MND, 21 MND), repeat length in the cerebellum (n=44; 15 FTD, 4 FTD/MND, 22 MND), and repeat length in blood (n=34; 20 FTD, 7 FTD/MND, 7 MND).
For all expansion carriers, frontal cortex, cerebellum, and/or blood were obtained, and these tissues comprised the primary analysis of this study. Frontal cortex was available for 41 patients (21 FTD, 11 FTD/MND, 9 MND), cerebellum was available for 40 patients (20 FTD, 12 FTD/MND, 8 MND), and blood was available for 50 subjects (15 FTD, 9 FTD/MND, 23 MND, 3 currently unaffected). In order to assess the reproducibility of measurements from blood samples over time, we included a subset of 17 subjects for whom multiple time points were available (median 23·2 months [IQR 11·5–34·8]). All expansion carriers were unrelated, with the exception of two families for whom four subjects each were investigated (family one: two MND [one proband], two unaffected; family two: two FTD, one FTD/MND [proband], one unaffected; web appendix p 3).
For a subset of the aforementioned samples, DNA was also obtained from the temporal cortex (n=18), parietal cortex (n=3), occipital cortex (n=3), spinal cord (n=5), skin-derived fibroblasts (n=5), and other peripheral tissues (n=5, including spleen, heart, muscle, pancreas, liver, and testes). Owing to the relatively low number of these samples, they were not included in our primary analysis.
Small C9ORF72 expansion size in cerebellum
We performed a thorough Southern blot study to assess C9ORF72 repeat expansion lengths in all expansion carriers. Our findings demonstrated that repeat lengths were significantly longer in the frontal cortex (33·8 kb [~5250 repeat units], IQR 23·5–44·9) than in the cerebellum (12·3 kb [~1667 repeat units], IQR 11·1–14·3, p<0·0001, table 1, figure 1, web appendix p 9). These results were generally consistent when examining disease subgroups individually (table 1). To determine whether the relatively small repeat length in the cerebellum was specific for this brain region, we investigated additional regions. We assessed temporal cortex of 18 subjects (figure 1, web appendix p 9), and revealed that repeat lengths in temporal cortex (25·7 kb [~3900 repeat units], IQR 22·0–29·3) were also significantly longer than in cerebellum (p<0·0001). This was consistent when examining only the 18 patients for whom repeat lengths in both temporal cortex and cerebellum were measured (p<0·0001). When additional brain regions from a series of patients were compared, cerebellum continued to harbor the smallest repeats, with other brain regions being comparable to the frontal cortex (n=3, figure 1). Spinal cord repeat lengths were examined in five patients, and showed variable repeat lengths ranging from 16·9 kb to 33·5 kb (figure 1, web appendix p 9).
Figure 1. Repeat lengths in the central nervous system.
WT = wild-type C9ORF72 allele (web appendix p 1). On a Southern blot, frontal cortex demonstrates relatively long repeat expansions, as displayed by ten representative cases (P1–P10, A). In the cerebellum, repeat expansions are usually smaller than in the frontal cortex (P11–P20, B). The horizontal dashed lines denote 23 kb and 9·4 kb. Repeat lengths in frontal cortex (n=41), cerebellum (n=40), and blood (n=50) are displayed on a scatter plot (C). Median repeat lengths are represented by a gray line. A bar graph shows frontal cortex (black), temporal cortex (light gray) and cerebellum (dark gray) for cases for whom these tissues were all available (n=18, D). For two illustrative cases, five brain regions are shown on a Southern blot (P16, P20, E). A bar graph of patients for whom frontal cortex, spinal cord, and cerebellum were all available is presented (n=5, F).
Variability in repeat length of peripheral tissues
Blood of 50 subjects was investigated, and demonstrated that hexanucleotide repeat lengths in blood (median18·6 kb [~2717 repeat units], IQR 13·9–28·1) were significantly smaller than in the frontal cortex (p=0·0001), but longer than in the cerebellum (p=0·0002, table 1, figure 1, figure 2). These results were similar when this analysis was stratified individually by gender, diagnosis, and family history, as well as by median age at onset, age at collection, and disease duration (web appendix p 6). In order to evaluate the reproducibility of repeat length measurements in blood over time, we examined repeat lengths of 17 subjects for whom two measurements were available at different time points. In these subjects, Lin’s concordance correlation coefficient indicated a very high level of reproducibility (0·93 [95% CI 0·87–0·99]).
Figure 2. Repeat lengths in other tissues.
Expansion sizes determined in blood of ten representative subjects are displayed on a Southern blot, demonstrating variation in repeat length (P27–P36, A). In blood, repeat expansions can show roughly the same length as in the frontal cortex or in the cerebellum, but they can also show an unexpected length, as demonstrated by three illustrative cases (P16, P20, P19, B). Substantial variability in repeat length is also present in other tissues of a single patient (P20, C). The last lane contains a negative control. Repeat expansions in fibroblasts are smaller than in blood, and can show multiple crisp bands (P37–P41, D).
A direct comparison between blood, frontal cortex, and cerebellum for subjects for whom these tissues were all available (n=8), showed a similar profile in some patients, and completely different profiles in other patients (figure 2, web appendix p 10). We also assessed repeat lengths of three unaffected repeat expansion carriers in blood; two of whom showed smaller repeat lengths (18 kb and 14 kb) than their family members (22 kb and 29 kb, web appendix p 11). The third subject carried an expansion (7·8 kb) that was in line with family members (7·2 kb, 7·3 kb, and 9·3 kb, web appendix p 11).
When comparing different non-affected peripheral tissues of five individual patients, substantial variation was observed, and most of these patients showed at least one tissue with a relatively small repeat expansion of approximately 3 kb (figure 2, web appendix p 12). We also compared repeat lengths in skin-derived fibroblasts to repeat lengths in blood (figure 2). Importantly, while relatively long repeat expansions and smears were present in blood, fibroblasts of the same subjects all demonstrated smaller repeat expansions (~7 kb) and smears.
Associations of repeat length with clinical and pathological features
C9ORF72 expansion lengths did not differ significantly between FTD, FTD/MND, and MND patients in the frontal cortex (p=0·27), cerebellum (p=0·96), or blood (p=0·10, figure 3). In addition, there was no evidence of an association between repeat length and age at onset, or age at collection, in our overall cohort (table 2). When evaluating this association in disease subgroups, however, age at onset positively correlated with repeat length in the frontal cortex of FTD patients (r=0·63, p=0·003), with similar findings for age at collection (r=0·58, p=0·006, figure 3). No association was detected between repeat length and age at onset, or age at collection, for other disease subgroups (table 2). There were no noticeable differences in repeat length according to gender, TDP-43 type, or family history (web appendix p 7). Furthermore, there was no significant correlation with length of the wild-type allele (table 2).
Figure 3. Differences in age, survival, and repeat length.
Associations of repeat length with age at onset (n=20, A) and age at collection (n=21, B) in FTD patients are shown. A regression line is displayed for each plot of correlation to enhance visual display. Survival following disease onset according to 25th percentile repeat length (11·1 kb [~1467 repeat units]) in the cerebellum of the overall cohort is demonstrated (n=10≤25th percentile, n=28>25th percentile, C). In addition, survival following disease onset according to 25th percentile repeat length (11.1 kb) in the cerebellum of FTD patients is shown (n=5≤25th percentile, n=14>25th percentile, D). Repeat lengths in the frontal cortex (n=21 FTD, 11 FTD/MND, 9 MND, E), cerebellum (n=20 FTD, 12 FTD/MND, 8 MND, F), and blood (n=15 FTD, 9 FTD/MND, and 23 MND, G) are not significantly different between disease subgroups; the sample median is represented by a solid horizontal line.
Table 2.
Association of repeat length with age at onset, age at collection, and length of wild-type allele
| Overall cohort | FTD patients | FTD/MND patients | MND patients | |||||
|---|---|---|---|---|---|---|---|---|
|
| ||||||||
| Association between repeat length and: | Spearman’s r (95% CI) | p value | Spearman’s r (95% CI) | p value | Spearman’s r (95% CI) | p value | Spearman’s r (95% CI) | p value |
| Frontal cortex | n=39 | n=20 | n=10 | n=9 | ||||
| Age at onset | 0·27 (−0·04–0·58) | 0·094 | 0·63 (0·38–0·88) | 0·003 | −0·28 (−0·88–0·33) | 0·44 | −0·10 (−0·77–0·57) | 0·80 |
| Age at collection | 0·25 (−0·06–0·56) | 0·12 | 0·58 (0·32–0·83) | 0·006 | −0·23 (−0·89–0·44) | 0·53 | 0·05 (−0·58–0·68) | 0·90 |
| Length of wild-type allele | −0·03 (−0·31–0·26) | 0·87 | 0·10 (−0·31–0·51) | 0·67 | .. | .. | 0·46 (−0·07–1·00) | 0·21 |
| Cerebellum | n=38 | n=19 | n=11 | n=8 | ||||
| Age at onset | 0·24 (−0·06–0·55) | 0·14 | 0·36 (−0·07–0·79) | 0·13 | 0·07 (−0·68–0·86) | 0·83 | −0·07 (−0·75–0·61) | 0·87 |
| Age at collection | 0·11 (−0·20–0·41) | 0·52 | 0·08 (−0·38–0·55) | 0·73 | 0·05 (−0·78–0·84) | 0·87 | −0·12 (−0·75–0·53) | 0·78 |
| Length of wild-type allele | −0·06 (−0·39–0·25) | 0·69 | 0·08 (−0·43–0·59) | 0·74 | .. | .. | −0·27 (−1·00–0·52) | 0·52 |
| Blood | n=46 | n=15 | n=8 | n=23 | ||||
| Age at onset | 0·07 (−0·21–0·35) | 0·66 | 0·13 (−0·37–0·62) | 0·65 | −0·10 (−0·79–0·59) | 0·82 | 0·25 (−0·18–0·67) | 0·26 |
| Age at collection | 0·15 (−0·11–0·42) | 0·28 | 0·28 (−0·22–0·78) | 0·32 | −0·02 (−0·56–0·53) | 0·97 | 0·26 (−0·16–0·68) | 0·24 |
| Length of wild-type allele | −0·01 (−0·29–0·28) | 0·97 | 0·09 (−0·44–0·62) | 0·75 | .. | .. | −0·16 (−0·55–0·23) | 0·47 |
Data are Spearman’s correlation coefficient r (95% confidence interval [CI]), number, or p value. P values result from Spearman’s test of correlation. The sample sizes given represent the usable sample size for the given test of correlation. Associations between repeat length and length of the wild-type allele were not evaluated in FTD/MND patients owing to the lack of variation in length of the wild-type allele in this disease subgroup.
Information regarding date of onset and date of death, or last follow-up, after disease onset was available for 77 out of 81 patients included in this study (95%). The median length of follow-up in these patients was 3·9 years (IQR 2·2–6·5) and 53 patients were deceased (69%: frontal cortex 39/41 [95%], cerebellum 38/40 [95%], blood 21/47 [45%]).
In the frontal cortex and blood, we did not detect significant associations between repeat length and survival after disease onset (model adjusted for disease type, table 3). These non-significant findings did not change in the frontal cortex after adjustment for age at onset or age at collection in the overall cohort (in addition to disease type, table 3, web appendix p 8), or after adjustment for age at onset in the FTD subgroup (table 3).
Table 3.
Association of repeat length with survival after disease onset
| Overall cohort | FTD patients | FTD/MND patients | MND patients | |||||
|---|---|---|---|---|---|---|---|---|
|
| ||||||||
| Repeat length variable | HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value | HR (95% CI) | p value |
| Repeat length in frontal cortex | n=39 (39 deaths) | n=20 (20 deaths) | n=10 (10 deaths) | n=9 (9 deaths) | ||||
| Continuous variable (per 5 kb increase) | 0·91 (0·77–1·08) | 0·29 | 0·90 (0·67–1·21) | 0·48 | 0·87 (0·64–1·19) | 0·39 | 0·74 (0·46–1·19) | 0·21 |
| Categorical variable (>median) | 0·96 (0·42–2·18) | 0·93 | 1·14 (0·23–5·53) | 0·87 | 0·55 (0·14–2·22) | 0·40 | 0·78 (0·15–3·99) | 0·77 |
| Categorical variable (>25th percentile) | 0·83 (0·39–1·78) | 0·63 | 0·76 (0·23–2·50) | 0·65 | .. | .. | .. | .. |
| Categorical variable (>75th percentile) | 0·84 (0·37–1·95) | 0·69 | 0·86 (0·22–3·37) | 0·82 | .. | .. | .. | .. |
| Repeat length in cerebellum | n=38 (38 deaths) | n=19 (19 deaths) | n=11 (11 deaths) | n=8 (8 deaths) | ||||
| Continuous variable (per 5 kb increase) | 1·10 (0·66–1·83) | 0·72 | 2·06 (0·70–6·05) | 0·19 | 1·00 (0·52–1·93) | 0·99 | 1·22 (0·11–13·75) | 0·87 |
| Categorical variable (>median) | 1·62 (0·77–3·40) | 0·20 | 1·90 (0·56–6·47) | 0·30 | 1·21 (0·34–4·38) | 0·77 | 1·77 (0·38–8·20) | 0·47 |
| Categorical variable (>25th percentile) | 3·27 (1·34–7·95) | 0·009 | 4·10 (1.04–27.72) | 0·043 | .. | .. | .. | .. |
| Categorical variable (>75th percentile) | 0·86 (0·37–2·02) | 0·73 | 1·09 (0·33–3·62) | 0·89 | .. | .. | .. | .. |
| Repeat length in blood | n=45 (21 deaths) | n=14 (3 deaths) | n=8 (8 deaths) | n=23 (10 deaths) | ||||
| Continuous variable (per 5 kb increase) | 0·98 (0·78–1·23) | 0·84 | 0·41 (0·11–1·59) | 0·20 | 0·93 (0·70–1·25) | 0·63 | 1·14 (0·79–1·66) | 0·49 |
| Categorical variable (>median) | 1·00 (0·39–2·60) | 0·99 | 0·44 (0·04–5·13) | 0·51 | 0·87 (0·19–3·96) | 0·86 | 1·31 (0·26–6·57) | 0·74 |
| Categorical variable (>25th percentile) | 0·84 (0·28–2·56) | 0·76 | .. | .. | .. | .. | .. | .. |
| Categorical variable (>75th percentile) | 0·97 (0·29–3·25) | 0·96 | .. | .. | .. | .. | .. | .. |
Data are hazard ratios (HRs) with 95% confidence intervals (CIs), number, or p value. P values result from Cox proportional hazards regression models. When considering repeat length as a continuous variable, HRs correspond to a 5 kb increase. When considering repeat length as a categorical variable, HRs are given for patients with a repeat length greater than the median value, the 25th percentile value, or the 75th percentile value, in the given group in comparison to a reference group of patients with a repeat length less than or equal to these values in the given group; values for each group are provided in table 1. Models in the overall cohort of patients involving repeat length in the frontal cortex and cerebellum were adjusted for disease type and age at onset. The model in the overall cohort involving repeat length in blood was adjusted for disease type. Models in FTD patients involving repeat length in the frontal cortex and cerebellum were adjusted for age at onset. All other models were not adjusted for any variables due to the low number of deaths in the given group. Models involving the 25th and 75th percentile cut-off points for repeat length in blood of FTD patients, and also involving the 25th and 75th percentile cut-off points of FTD/MND and MND patients, were not considered due to the low number of deaths in these groups.
Importantly, in the cerebellum, we observed a significant association between repeat length and survival after disease onset: repeat lengths greater than the 25th percentile (11·1 kb, ~1467 repeat units) conferred a survival disadvantage (HR 3·71, 95% CI 1·55–8·88, p=0·003, model adjusted for disease type), and a similar association was seen in our FTD subgroup (HR 5·56, 95% CI 1·49–36·21, p=0·008). For the overall cohort, median survival was 4·8 years (IQR 3·0–7·4) in the group with longer repeat expansions as compared to 7·4 years (IQR 6·3–10·9) in the group with smaller expansions (figure 3). When adjusting for age at onset (in addition to disease type in the overall cohort), trends weakened slightly, but remained significant in the cerebellum of both the overall group (HR 3·27, 95% CI 1·34–7·95, p=0·009) and the FTD subgroup (HR 4·10, 95% CI 1·04–27·72, p=0·043, table 3).
Discussion
In the Xpansize-72 study, we sought to determine C9ORF72 repeat lengths in a comparatively large number of tissues from 84 subjects. We confirmed the presence of hexanucleotide repeat expansions in all cases. In the cerebellum, the median repeat length was significantly smaller as compared to the frontal cortex and blood. No associations between repeat length and disease subgroups were observed in any of these tissues. Furthermore, in FTD patients, we demonstrated that longer repeat expansions in the frontal cortex were associated with an increase in age at onset and age at collection. Importantly, we are the first to observe an association between repeat length and survival following disease onset (research in context panel). We would like to emphasize, however, that studies involving brain samples of C9ORF72 expansion carriers will always be hampered by a relatively low power, and therefore, all these findings should be validated by other groups and meta-analytical studies.
Our finding that C9ORF72 repeat sizes are smaller in the cerebellum than in other brain regions, has also been reported for repeat expansions in other genes. For patients with myotonic dystrophy type 1 (DM1), for instance, DMPK repeat expansions are smaller in the cerebellar cortex than in various other tissues.11–13 For other repeat expansion disorders, such as Huntington’s disease (HD), spinocerebellar ataxia type 1 (SCA1), spinocerebellar ataxia type 3 (SCA3), and dentatorubral-pallidoluysian atrophy (DRPLA), the same phenomenon has been described.14–20 It has been hypothesized that this relatively small and stable cerebellar repeat expansion is caused by the high percentage of predominantly non-dividing terminally differentiated cells,16 by reduced viability of neuronal cells harboring larger expansions,16 by differences in transcription elongation,21 by tissue-specific methylation levels,22 or by factors specific to a particular cell type or region, such as factors involved in DNA repair.11,14,23 Pathological investigations of C9ORF72 repeat expansion carriers have already shown that the cerebellum harbors characteristic TDP-43-negative, p62/sequestosome-1-positive neuronal inclusions,24 which are absent in most other brain regions, highlighting the distinctive features of this tissue type.
When comparing expanded allele sizes of various tissues within the same individual, somatic heterogeneity was noted, as reported for DM1 and HD.14,25–29 Importantly, for patients with DM1 and HD, size heterogeneity is dependent on age at collection;14,30 this dependence was also detected in our study, and was most prominent in the frontal cortex of FTD patients. Additionally, patients with DM1, DRPLA, and spinocerebellar ataxia type 10 (SCA10) have been shown to harbor an inverse correlation between age at onset and expansion size.20,31–33 Remarkably, in our FTD cohort, we observed the opposite effect: a lower age at onset and age at collection in patients with smaller GGGGCC repeat expansions. Although smaller expansions may in truth result in an earlier age at onset, an alternative explanation could be that this finding may be driven by age at collection, since repeat sizes could lengthen as an individual ages; investigations of additional pathological FTD cohorts are necessary to confirm this hypothesis.
We speculate that the age-dependent effect in the frontal cortex of FTD patients could contribute to symptoms originating from this primary affected region. In the frontal cortex, we did not detect an association between repeat length and survival after disease onset, even when considering the potential as a confounding variable of age at onset or age at collection. In the cerebellum, repeat lengths seemed more stable and less influenced by age than in the frontal cortex. Importantly, we were able to detect an association between repeat length and survival in the cerebellum of both our overall cohort and FTD subgroup, with small repeat expansions resulting in a survival benefit. Our findings remained significant when adjusting for a possible bias due to differences in age at onset, and thus, repeat lengths in the cerebellum appeared to be a consistent predictor of disease severity. This survival advantage in the cerebellum, its distinct pathological hallmarks, and the relatively small C9ORF72 repeat sizes in this region unaffected by neuronal loss, suggest that cerebellum differs from other brain regions, and factors that underlay these differences may hold clues for promising novel therapeutic targets.
A previous Southern blot study investigated the presence of C9ORF72 repeat expansions in a heterogeneous cohort of patients with neurodegenerative diseases.7 The authors performed a Southern blot for 68 C9ORF72 repeat expansion carriers (57 cases with neurodegenerative diseases), and reported correlations between repeat sizes and age at onset (r=0·38, p=0·02), comparable to our FTD patients. A correlation with age at onset could have severe implications, for instance, when a young woman of 25 years of age with a positive family history for FTD/MND visits an outpatient clinic to determine whether she has a C9ORF72 repeat expansion. Her genetic counselor may tell her that she has a small repeat expansion, and consequently, that she will most likely develop symptoms soon. Our findings, however, could point towards another potential explanation: that she may have a small repeat expansion because she is young. Large-scale studies of pathological cohorts and longitudinal studies of clinical cohorts should be performed to clarify whether repeat sizes change over time in brain tissues and/or peripheral tissues, which could provide very important information for genetic counseling.
The study cohort of the previous Southern blot paper mainly consisted of clinically diagnosed patients, and only blood or cell lines were available for most of them.7 The predominant study of blood samples may explain why they did not report associations with other clinical features, such as survival after disease onset. Notably, although our findings demonstrate that blood does not appear to be a trustworthy marker for the exact repeat size in brain and/or disease severity, we expect that blood is an acceptable basis to determine the presence or absence of C9ORF72 repeat expansions in a clinical setting.
We further demonstrated that skin-derived fibroblasts display a small crisp band on a Southern blot, much smaller than blood from the same subjects. The relatively small repeat size in fibroblasts could result from an increased viability of cells with smaller repeat sizes, which overgrow other subpopulations present in the original heterogeneous cell population. This finding emphasizes that repeat lengths in these cells do not appear to be a suitable biomarker for repeat sizes in other tissue types. Moreover, as somatic mosaicism appears to be age-dependent and tissue-specific, measured allele sizes could potentially be influenced by the age at collection of subjects, and by the tissue for which the isolated DNA is analyzed, as shown for DM1.29 This shows that repeat lengths determined in blood or fibroblasts may not represent the most adequate predictors of disease phenotype or severity. Based on our Xpansize-72 study, only hexanucleotide repeat lengths in the cerebellum, a stable region unaffected by neuronal loss, are associated with survival after disease onset.
Our study has several limitations. Although we were able to investigate a unique and relatively large cohort of C9ORF72 expansion carriers, some of our analyses included a small number of participants, and consequently, these analyses should be interpreted carefully, especially since the possibility of a Type II error (i.e. a false-negative association) cannot be excluded. Related to this, the number of MND patients included in our pathological cohort was fairly small, and precluded us from drawing any firm conclusions regarding MND patients. Furthermore, our current design was unable to assess changes over larger time intervals, and to assess potential anticipation within families. Since studies investigating brain samples from patients with C9ORF72 expansions will always be hindered by a comparatively low power, and because it is very unlikely that all tissue types will be available for all participants, other groups and meta-analytical studies should provide a larger body of research to address these issues more thoroughly. It should be taken into consideration, however, that repeat lengths provided in this study are not exact values, but estimates based on our optimized Southern blot protocol, and differences between laboratories (e.g. in Southern blotting techniques) will hamper direct comparisons.
Supplementary Material
Research in context.
Systematic review
Medline (1950 to July, 2013) was searched with the following search term “C9ORF72”. All studies on C9ORF72 were thoroughly assessed, aiming at the identification of studies that employed Southern blotting techniques to examine repeat sizes. Papers without full access and papers that were not written in English were excluded.
Interpretation
The search identified fifteen articles, all of which focused on blood or cell lines, and/or only included brain tissues from a small number of subjects (n≤3). Thus, for the first time, the Xpansize-72 study investigated repeat sizes in brain tissues of C9ORF72 expansion carriers in a relatively large and well-characterized study cohort. Moreover, this study is the first to show an association between C9ORF72 repeat length and survival after disease onset. The findings described herein could assist researches, clinicians and genetic counselors, in their quest to untangle the heterogeneity detected in C9ORF72 expansion carriers, to interpret diagnostic genetic tests, and to provide patients and their family members with more detailed information regarding repeat expansions in C9ORF72.
Acknowledgments
The research leading to these results has received funding by NIH grants R01 NS080882, R01 AG026251, P50 AG 016574, and the ALS Therapy Alliance. The Brain and Body Donation Program is supported by the National Institute of Neurological Disorders and Stroke (U24 NS072026 National Brain and Tissue Resource for Parkinson’s Disease and Related Disorders), the National Institute on Aging (P30 AG19610 Arizona Alzheimer’s Disease Core Center), the Arizona Department of Health Services (contract 211002, Arizona Alzheimer’s Research Center), the Arizona Biomedical Research Commission (contracts 4001, 0011, 05-901 and 1001 to the Arizona Parkinson’s Disease Consortium) and the Michael J. Fox Foundation for Parkinson’s Research.
Footnotes
Contributors
MvB, MDJ, and RR designed the study. MvB and RR coordinated and managed the study. MvB, MDJ, EN, PHB, and MCB did laboratory-based experiments. MvB and RR analyzed the data, and wrote the report. MGH and NND performed the statistical analysis. MEM, NAF, POB, GS, TGB, KAJ, DSK, RCP, BFB, NRG-R, KBB, LP, and DWD obtained samples from patients or clinical data. All authors participated in the interpretation of the data, and in revision of the report.
Conflicts of interest
MDJ and RR hold a patent on methods to screen for the hexanucleotide repeat expansion in the C9ORF72 gene. All other authors declare that they have no conflicts of interest.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.DeJesus-Hernandez M, Mackenzie IR, Boeve BF, Boxer AL, Baker M, Rutherford NJ, et al. Expanded GGGGCC hexanucleotide repeat in noncoding region of C9ORF72 causes chromosome 9p-linked FTD and ALS. Neuron. 2011;72(2):245–56. doi: 10.1016/j.neuron.2011.09.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Renton AE, Majounie E, Waite A, Simon-Sanchez J, Rollinson S, Gibbs JR, et al. A hexanucleotide repeat expansion in C9ORF72 is the cause of chromosome 9p21-linked ALS-FTD. Neuron. 2011;72(2):257–68. doi: 10.1016/j.neuron.2011.09.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Ash PE, Bieniek KF, Gendron TF, Caulfield T, Lin WL, Dejesus-Hernandez M, et al. Unconventional Translation of C9ORF72 GGGGCC Expansion Generates Insoluble Polypeptides Specific to c9FTD/ALS. Neuron. 2013;77(4):639–46. doi: 10.1016/j.neuron.2013.02.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Mori K, Weng SM, Arzberger T, May S, Rentzsch K, Kremmer E, et al. The C9orf72 GGGGCC Repeat Is Translated into Aggregating Dipeptide-Repeat Proteins in FTLD/ALS. Science. 2013;339(6125):1335–1338. doi: 10.1126/science.1232927. [DOI] [PubMed] [Google Scholar]
- 5.Levine TP, Daniels RD, Gatta AT, Wong LH, Hayes MJ. The product of C9orf72, a gene strongly implicated in neurodegeneration, is structurally related to DENN Rab-GEFs. Bioinformatics. 2013;29(4):499–503. doi: 10.1093/bioinformatics/bts725. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zhang D, Iyer LM, He F, Aravind L. Discovery of Novel DENN Proteins: Implications for the Evolution of Eukaryotic Intracellular Membrane Structures and Human Disease. Front Genet. 2012;3:283. doi: 10.3389/fgene.2012.00283. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Beck J, Poulter M, Hensman D, Rohrer JD, Mahoney CJ, Adamson G, et al. Large C9orf72 Hexanucleotide Repeat Expansions Are Seen in Multiple Neurodegenerative Syndromes and Are More Frequent Than Expected in the UK Population. Am J Hum Genet. 2013;92(3):345–353. doi: 10.1016/j.ajhg.2013.01.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.van Elteren P. On the combination of independent two sample tests of Wilcoxon. Bulletin of the Institute of International Statistics. 1960;37:351–361. [Google Scholar]
- 9.Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics. 1989;45(1):255–268. [PubMed] [Google Scholar]
- 10.Vittinghoff E, McCulloch CE. Relaxing the rule of ten events per variable in logistic and Cox regression. Am J Epidemiol. 2007;165(6):710–718. doi: 10.1093/aje/kwk052. [DOI] [PubMed] [Google Scholar]
- 11.Jinnai K, Mitani M, Futamura N, Kawamoto K, Funakawa I, Itoh K. Somatic instability of CTG repeats in the cerebellum of myotonic dystrophy type 1. Muscle Nerve. 2013;48(1):105–108. doi: 10.1002/mus.23717. [DOI] [PubMed] [Google Scholar]
- 12.Itoh K, Mitani M, Kawamoto K, Futamura N, Funakawa I, Jinnai K, et al. Neuropathology does not Correlate with Regional Differences in the Extent of Expansion of CTG Repeats in the Brain with Myotonic Dystrophy Type 1. Acta Histochem Cytochem. 2010;43(6):149–56. doi: 10.1267/ahc.10019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ishii S, Nishio T, Sunohara N, Yoshihara T, Takemura K, Hikiji K, et al. Small increase in triplet repeat length of cerebellum from patients with myotonic dystrophy. Hum Genet. 1996;98(2):138–40. doi: 10.1007/s004390050176. [DOI] [PubMed] [Google Scholar]
- 14.Telenius H, Kremer B, Goldberg YP, Theilmann J, Andrew SE, Zeisler J, et al. Somatic and gonadal mosaicism of the Huntington disease gene CAG repeat in brain and sperm. Nat Genet. 1994;6(4):409–14. doi: 10.1038/ng0494-409. [DOI] [PubMed] [Google Scholar]
- 15.De Rooij KE, De Koning Gans PA, Roos RA, Van Ommen GJ, Den Dunnen JT. Somatic expansion of the (CAG)n repeat in Huntington disease brains. Hum Genet. 1995;95(3):270–4. doi: 10.1007/BF00225192. [DOI] [PubMed] [Google Scholar]
- 16.Chong SS, McCall AE, Cota J, Subramony SH, Orr HT, Hughes MR, et al. Gametic and somatic tissue-specific heterogeneity of the expanded SCA1 CAG repeat in spinocerebellar ataxia type 1. Nat Genet. 1995;10(3):344–50. doi: 10.1038/ng0795-344. [DOI] [PubMed] [Google Scholar]
- 17.Lopes-Cendes I, Maciel P, Kish S, Gaspar C, Robitaille Y, Clark HB, et al. Somatic mosaicism in the central nervous system in spinocerebellar ataxia type 1 and Machado-Joseph disease. Ann Neurol. 1996;40 (2):199–206. doi: 10.1002/ana.410400211. [DOI] [PubMed] [Google Scholar]
- 18.Tanaka F, Sobue G, Doyu M, Ito Y, Yamamoto M, Shimada N, et al. Differential pattern in tissue-specific somatic mosaicism of expanded CAG trinucleotide repeats in dentatorubral-pallidoluysian atrophy, Machado-Joseph disease, and X-linked recessive spinal and bulbar muscular atrophy. J Neurol Sci. 1996;135(1):43–50. doi: 10.1016/0022-510x(95)00249-2. [DOI] [PubMed] [Google Scholar]
- 19.Cancel G, Gourfinkel-An I, Stevanin G, Didierjean O, Abbas N, Hirsch E, et al. Somatic mosaicism of the CAG repeat expansion in spinocerebellar ataxia type 3/Machado-Joseph disease. Hum Mutat. 1998;11(1):23–7. doi: 10.1002/(SICI)1098-1004(1998)11:1<23::AID-HUMU4>3.0.CO;2-M. [DOI] [PubMed] [Google Scholar]
- 20.Takano H, Onodera O, Takahashi H, Igarashi S, Yamada M, Oyake M, et al. Somatic mosaicism of expanded CAG repeats in brains of patients with dentatorubral-pallidoluysian atrophy: cellular population-dependent dynamics of mitotic instability. Am J Hum Genet. 1996;58(6):1212–22. [PMC free article] [PubMed] [Google Scholar]
- 21.Goula AV, Stys A, Chan JP, Trottier Y, Festenstein R, Merienne K. Transcription elongation and tissue-specific somatic CAG instability. PLoS Genet. 2012;8(11):e100305. doi: 10.1371/journal.pgen.1003051. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Lopez Castel A, Nakamori M, Tome S, Chitayat D, Gourdon G, Thornton CA, et al. Expanded CTG repeat demarcates a boundary for abnormal CpG methylation in myotonic dystrophy patient tissues. Hum Mol Genet. 2011;20(1):1–1. doi: 10.1093/hmg/ddq427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Tome S, Manley K, Simard JP, Clark GW, Slean MM, Swami M, et al. MSH3 polymorphisms and protein levels affect CAG repeat instability in Huntington’s disease mice. PLoS Genet. 2013;9(2):e1003280. doi: 10.1371/journal.pgen.1003280. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Al-Sarraj S, King A, Troakes C, Smith B, Maekawa S, Bodi I, et al. p62 positive, TDP-43 negative, neuronal cytoplasmic and intranuclear inclusions in the cerebellum and hippocampus define the pathology of C9orf72-linked FTLD and MND/ALS. Acta Neuropathol. 2011;122(6):691–70. doi: 10.1007/s00401-011-0911-2. [DOI] [PubMed] [Google Scholar]
- 25.Sabouri LA, Mahadevan MS, Narang M, Lee DS, Surh LC, Korneluk RG. Effect of the myotonic dystrophy (DM) mutation on mRNA levels of the DM gene. Nat Genet. 1993;4(3):233–8. doi: 10.1038/ng0793-233. [DOI] [PubMed] [Google Scholar]
- 26.Ashizawa T, Dubel JR, Harati Y. Somatic instability of CTG repeat in myotonic dystrophy. Neurology. 1993;43(12):2674–8. doi: 10.1212/wnl.43.12.2674. [DOI] [PubMed] [Google Scholar]
- 27.Jansen G, Willems P, Coerwinkel M, Nillesen W, Smeets H, Vits L, et al. Gonosomal mosaicism in myotonic dystrophy patients: involvement of mitotic events in (CTG)n repeat variation and selection against extreme expansion in sperm. Am J Hum Genet. 1994;54(4):575–85. [PMC free article] [PubMed] [Google Scholar]
- 28.Joseph JT, Richards CS, Anthony DC, Upton M, Perez-Atayde AR, Greenstein P. Congenital myotonic dystrophy pathology and somatic mosaicism. Neurology. 1997;49(5):1457–60. doi: 10.1212/wnl.49.5.1457. [DOI] [PubMed] [Google Scholar]
- 29.Radvansky J, Kadasi L. The expanding world of myotonic dystrophies: how can they be detected? Genet Test Mol Biomarkers. 2010;14(6):733–41. doi: 10.1089/gtmb.2010.0073. [DOI] [PubMed] [Google Scholar]
- 30.Wong LJ, Ashizawa T, Monckton DG, Caskey CT, Richards CS. Somatic heterogeneity of the CTG repeat in myotonic dystrophy is age and size dependent. Am J Hum Genet. 1995;56(1):114–22. [PMC free article] [PubMed] [Google Scholar]
- 31.Kim SY, Kim JY, Kim GP, Sung JJ, Lim KS, Lee KW, et al. Molecular and clinical characteristics of myotonic dystrophy type 1 in koreans. Korean J Lab Med. 2008;28(6):483–92. doi: 10.3343/kjlm.2008.28.6.483. [DOI] [PubMed] [Google Scholar]
- 32.Savic D, Rakocvic-Stojanovic V, Keckarevic D, Culjkovic B, Stojkovic O, Mladenovic J, et al. 250 CTG repeats in DMPK is a threshold for correlation of expansion size and age at onset of juvenile-adult DM1. Hum Mutat. 2002;19(2):131–9. doi: 10.1002/humu.10027. [DOI] [PubMed] [Google Scholar]
- 33.McFarland KN, Liu J, Landrian I, Gao R, Sarkar PS, Raskin S, et al. Paradoxical effects of repeat interruptions on spinocerebellar ataxia type 10 expansions and repeat instability. Eur J Hum Genet. 2013 doi: 10.1038/ejhg.2013.32. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.