. Clinical Evidence of Disease Anticipation in Families Segregating a C9orf72 Repeat Expansion. JAMA Neurol. 2017 Apr 1;74(4):445-452. PubMed.


Please login to recommend the paper.


  1. The paper demonstrates disease anticipation in a well-characterized sample of Belgian families with C9ORF72-associated FTD/ALS. Such clinical anticipation is commonly seen in diseases caused by expanded repeat mutations, such as myotonic dystrophy and Huntington's disease. Compared with the anticipation seen in myotonic dystrophy—which can amount to a generation difference in each generation, such that the grandparental generation has onset in old age, the parental generation in adulthood, but their children have congenital disease—the differences here are quite small, averaging four to five years/generation. Disease anticipation in most repeat diseases is associated with expansion of the repeat tract in the germline from generation to generation. Presenting phenotype is most likely exacerbated by somatic expansion of the repeat. Somatic expansion is more likely to occur, and the expansions that do occur are longer, the longer the original repeated tract is. However, as the authors note, small changes from generation to generation could be a result of ascertainment bias—that once a disease is known to segregate in a family, then it is more likely to be diagnosed earlier as family members and clinicians are primed to spot it. Of course, in C9ORF72 expansion disorder we can't actually measure the repeat length in most people, so whether disease anticipation is determined by genetic anticipation is unknown. Technological advances in sequencing might solve this issue.

    There is also the interesting observation that the phenotypic presentation, FTD or ALS, tends to run in families. This implies other heritable components that mediate the phenotype as well as the anticipation seen.  These could potentially arise from two sources:

    1. The nature of the repeat—generally pure repeats (not interrupted by other sequences) expand to a larger extent and faster;
    2. Modification by other genes in the genome, as in HD and the other repeat disorders (Correia  et al., 2015; Bettencourt et al., 2016). 

    In respect of option 1 above, we know that repeat interruptions can mediate the presenting phenotype in SCA2 (Kim et al., 2007). Repeat lengths below the threshold to cause SCA2 29-33 (Sproviero et al., 2017) and those with interruptions to repeats of 33-40 show Parkinsonian symptoms rather than spinocerebellar ataxia. In respect of this study it would be interesting to know whether anticipation was more marked in ALS or FTD-segregating families as one would expect greater expansions and therefore anticipation in families carrying larger or less interrupted repeats. If repeat length is a mediator of disease type, then as onset gets earlier one should see more ALS and less FTD, which the authors indicate was not seen.

    It is very plausible that other genes modify the phenotype as in HD and the SCAs, and also influence the phenotypic presentation.

    It would be most interesting to study these possibilities, but they are all hampered by our inability to accurately measure the repeat length or sequence.

    In HD we found modifiers that clustered in the DNA damage response pathways, and one hypothesis is that this modulates somatic expansion. It would be interesting to test those in C9ORF72-mediated disease but in the absence of being able to account for the repeat length in the genetic analysis, any study is likely to be underpowered to detect any effects.


    . The Genetic Modifiers of Motor OnsetAge (GeM MOA) Website: Genome-wide Association Analysis for Genetic Modifiers of Huntington's Disease. J Huntingtons Dis. 2015;4(3):279-84. PubMed.

    . DNA repair pathways underlie a common genetic mechanism modulating onset in polyglutamine diseases. Ann Neurol. 2016 Jun;79(6):983-90. Epub 2016 May 6 PubMed.

    . Importance of low-range CAG expansion and CAA interruption in SCA2 Parkinsonism. Arch Neurol. 2007 Oct;64(10):1510-8. PubMed.

    . ATXN2 trinucleotide repeat length correlates with risk of ALS. Neurobiol Aging. 2017 Mar;51:178.e1-178.e9. Epub 2016 Nov 24 PubMed.

    View all comments by Lesley Jones
  2. Genetic anticipation and the C9ORF72 repeat expansion

    Genetic anticipation is a phenomenon whereby the clinical symptoms of a genetic disorder become apparent at an earlier age as the disease is inherited from one generation to the next (Boonstra et al., 2010). One of the most well-characterized examples of genetic anticipation is the change in the repeat length of a trinucleotide repeat expansion in the Huntington’s disease (HD) gene HTT (Ridley et al., 1988). As the mutant locus is inherited across generations, meiotic recombinations in gametes can result in elongation of the repeat expansion in affected offspring.

    A hexanucleotide repeat expansion in the C9ORF72 gene has been shown to be causative of both motor neuron disease and frontotemporal dementia (DeJesus-Hernandez et al., 2011; Renton et al., 2011). In this interesting study, Van Mossevelde et al. examined whether genetic anticipation was associated with this locus. This current study is one of the largest conducted so far, comprising 244 individuals from 36 extended Belgian families. They compared clinical features of the disease across generations within each family, using data obtained from family members and clinicians. Using a mixed-effects Cox proportional hazards regression model, they demonstrated a significant generational effect on age at onset, but not disease duration, nor age at death. If this is the case, the implication is that although repeat length is a significant factor in determining when neurons start to degenerate, other genetic or environmental factors determine how fast the disease progresses.

    The authors took care to account for known ascertainment biases such as recall biases by later generations as they might recognize the disease earlier because they are more familiar with the symptoms. However, it is difficult to completely account for other factors such as change in diagnostic techniques, better recognition of the disease by clinicians, or introduction of new environmental toxicants/carcinogens that could affect specific birth cohorts (Boonstra et al., 2010). 

    Parental-gender effect is a phenomenon often observed across different disorders of unstable repeat expansion (Boonstra et al., 2010). For example, expansions from premutation to full mutation in HD, and large expansions associated with a juvenile-onset HD, occur primarily upon male transmission (Ridley et al., 1988). In contrast to HD, Mossevelde et al. reported that the difference in age at onset between father and offspring and between mother and offspring was not significant.

    Finally, it should be noted that the authors did not investigate C9ORF72 repeat length directly in this study. A prerequisite for genetic anticipation is a causal relationship between repeat length and symptom onset, duration and/or severity. Whilst a number of papers have examined these clinical parameters in relation to repeat length in C9ORF72, findings have been inconsistent (Beck et al., 2013; van Blitterswijk et al., 2013; Gijselinck et al., 2016; Suh et al., 2015). Repeat expansions of more than 80 repeat units are difficult to size accurately because the number of repeats can extend into the thousands. Other complicating factors include high GC content of the repeat expansion, which can impact on PCR-based technology, and tissue-specific mosaicism where the number of repeats can differ markedly between brain and peripheral tissue (Nordin et al., 2015). Therefore, clear demonstration of genetic anticipation in the C9ORF72 locus will be technically difficult, and will require careful use of statistical methodology to remove biases that can result in false genetic anticipation signals.


    . A review of statistical methods for testing genetic anticipation: looking for an answer in Lynch syndrome. Genet Epidemiol. 2010 Nov;34(7):756-68. PubMed.

    . Anticipation in Huntington's disease is inherited through the male line but may originate in the female. J Med Genet. 1988 Sep;25(9):589-95. PubMed.

    . Expanded GGGGCC hexanucleotide repeat in noncoding region of C9ORF72 causes chromosome 9p-linked FTD and ALS. Neuron. 2011 Oct 20;72(2):245-56. Epub 2011 Sep 21 PubMed.

    . A hexanucleotide repeat expansion in C9ORF72 is the cause of chromosome 9p21-linked ALS-FTD. Neuron. 2011 Oct 20;72(2):257-68. Epub 2011 Sep 21 PubMed.

    . 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 Mar 7;92(3):345-53. Epub 2013 Feb 21 PubMed.

    . Association between repeat sizes and clinical and pathological characteristics in carriers of C9ORF72 repeat expansions (Xpansize-72): a cross-sectional cohort study. Lancet Neurol. 2013 Oct;12(10):978-88. Epub 2013 Sep 5 PubMed.

    . The C9orf72 repeat size correlates with onset age of disease, DNA methylation and transcriptional downregulation of the promoter. Mol Psychiatry. 2016 Aug;21(8):1112-24. Epub 2015 Oct 20 PubMed.

    . Semi-automated quantification of C9orf72 expansion size reveals inverse correlation between hexanucleotide repeat number and disease duration in frontotemporal degeneration. Acta Neuropathol. 2015 Sep;130(3):363-72. Epub 2015 May 29 PubMed.

    . Extensive size variability of the GGGGCC expansion in C9orf72 in both neuronal and non-neuronal tissues in 18 patients with ALS or FTD. Hum Mol Genet. 2015 Jun 1;24(11):3133-42. Epub 2015 Feb 23 PubMed.

    View all comments by John Kwok
  3. As Carol Dobson-Stone and John Kwok mention, it is true that it is difficult to completely account for factors such as change in diagnostic techniques, better recognition of the disease by clinicians, or introduction of new environmental toxicants/carcinogens that could affect specific birth cohorts. We would like to stress that we have made a lot of effort to account as much as possible for any (known) factors of bias. We have considered correcting for birth cohort in our analyses, but as described by Boonstra et al. (2011): “Using birth cohort in the model may lead to instability in parameter estimates due to its strong correlation with generation, the primary variable of interest.” A measure we took to decrease this factor of bias was to assign generation numbers based on the earliest available generation for whom age information was present, and not to assign generation number based on year of birth. We numbered the earliest born generation in each pedigree as generation 4 irrespective of whether information was available for two, three, or four generations. Consequently, there is an overlap in the calendar years of birth between generations (especially in generation 2 and 3, which include most patients). Further, we have performed a separate analysis using only onset ages retrieved from clinical files of the patients themselves. This means onset ages were calculated based on the information the treating physician retrieved from the patient or his direct relatives/partner at the initial consult through (hetero)anamnesis. We want to point out that this age is the age at which the patient or his relatives noticed the first symptoms and not the age of diagnosis. That way, we diminish bias that might result from better diagnostic techniques and better recognition of the disease by clinicians leading to earlier/faster diagnosis in later born generations. By excluding the ages at onset retrieved from further relatives (e.g., children/grandchildren) many years after the onset of the symptoms, we further decrease ascertainment bias.

    A second remark we would like to make is about the fact that we could not reveal a significant difference in age at onset between affected father and affected offspring versus between affected mother and affected offspring. We did observe that the average difference in onset age between affected offspring and an affected mother (4.8 ± 9.4 years) was 5.1 years less than with an affected father (9.9 ± 7.5 years), though indeed not significant (p = 0.11). Although this is a large study, we still had onset age data of both affected parent and affected offspring of “only” 14 mother-offspring pairs and 17 father-offspring pairs. Possibly, this dataset was too small to have enough power to obtain a significant result. We believe further follow-up studies are definitely necessary and might provide significant evidence that paternal transmission is more prone to repeat expansion and disease anticipation than maternal transmission.  


    . Bayesian modeling for genetic anticipation in presence of mutational heterogeneity: a case study in Lynch syndrome. Biometrics. 2011 Dec;67(4):1627-37. Epub 2011 May 31 PubMed.

    View all comments by Sara Van Mossevelde

Make a Comment

To make a comment you must login or register.

This paper appears in the following:


  1. Dreaded Anticipation: ALS Strikes Earlier in Successive Generations