A crystal ball for survival for spinocerebellar ataxias?

TY - JOUR

T1 - A crystal ball for survival for spinocerebellar ataxias?

AU - Ashizawa, Tetsuo

N1 - Funding Information: Tetsuo Ashizawa a tashizawa@houstonmethodist.org a Houston Methodist Neurological Institute and Research Institute, Weill Cornell Medical College, Houston, TX 77030, USA Houston Methodist Neurological Institute and Research Institute Weill Cornell Medical College Houston TX 77030 USA Death is an unequivocal clinical endpoint. Although each of us must accept the harsh reality of 100% mortality, prolonging survival has been, and will be, one of the major goals of medicine. Spinocerebellar ataxias are a group of autosomal dominant neurodegenerative disorders, most of which show relentless progression, often resulting in total disability and premature death. 1 SCA1, SCA2, SCA3, and SCA6 are among the most common spinocerebellar ataxias and caused by expanded CAG trinucleotide repeats encoding polyglutamine (polyQ), which belong to the same class of mutations as those of Huntington's disease, spinal and bulbar muscular atrophy, and a handful of other spinocerebellar ataxias. 2 Recent advances in understanding the pathogenic mechanisms of these currently incurable spinocerebellar ataxias have raised increasing optimism for disease-modifying treatments, including those based on RNA interference. For future clinical trials of such treatments, natural history data are essential, and the age at death would be a pivotal endpoint in natural history and clinical trials. In their study reported in The Lancet Neurology , Alhassane Diallo and her colleagues 3 used data from well characterised European cohorts 4 of patients with SCA1, SCA2, SCA3, and SCA6 to establish nomograms for survival prediction, based on the Cox regression model adjusted for age at baseline. The model was internally validated by use of multiple sets of data in bootstrapping. The investigators showed impressive agreements between actual and nomogram-predicted age of death. As the authors pointed out, the nomograms need to be externally validated in independent populations, including non-European cohorts, some of which have shown variable progression rates of spinocerebellar ataxias. 5,6 Further follow-up of the cohorts used in this study would capture additional deaths, which could be compared with the predictions from the nomogram. The size of CAG repeats at other polyQ expansion loci and MRI and magnetic resonance spectroscopy data could also be possible prognostic factors of interest. Additionally, longitudinal data including progression rates, which are available for the study cohorts, 4 could be even more relevant to survival than are the baseline cross-section data. Survival was defined as the time from enrolment to death by any reason, which might lead to underestimation. One might argue that death unrelated to ataxia should be excluded. However, spinocerebellar-ataxia-related death is difficult to define. For example, accidental deaths could be due to ataxia or cognitive impairments, and decreased cancer deaths and cancer prevalence in polyQ diseases (spinocerebellar ataxias and Huntington disease) 7,8 might further confound the definition. In clinical trials that have an outcome measure of disease-related deaths, this ambiguity might pose a problem, whereas prediction of any death would be perfectly useful in stratification of study participants in clinical trials with other types of endpoint. Attrition of study participants who were lost in follow-up could also introduce a bias if a substantial proportion of dropouts are due to death. However, these issues probably have no significant effect on the overall conclusion of the study. The investigators found that physiotherapy, which had shown temporary symptomatic effects on ataxia, 9 had no effect on survival. This finding is consistent with the thought that it takes an efficacious disease-modifying therapy to affect survival. However, the effect on survival of sustained improvement of the Scale for the Assessment and Rating of Ataxia (SARA) score by continuous physiotherapy remains to be investigated. General enrichment of the patient's life has improved survival in other disorders. Likewise, therapies to improve the predictive factors identified in this study might need to be critically assessed for their benefits. Survival prediction nomograms are likely to become an important tool in clinical research for spinocerebellar ataxias. For the next several years, however, therapeutic trials are likely to focus on interventions in early stages of these disorders because current animal and human data predict decreasing efficacy of interventions as neurons continue to degenerate. 10,11 Thus, survival prediction might have a small role until therapeutics start showing either an efficacy that affects survival or the necessity for stratification by survival prediction. Conversely, the psychological and practical effects of survival prediction in each patient's life might not be trivial. As in predictive genetic testing on presymptomatic at-risk individuals, counselling might be advisable at the time of disclosure of survival prediction results. Desire and fear for knowing the day you will die are long-standing cultural themes. In spinocerebellar ataxias, the disabled life of a patient can be extremely difficult, and suicides (two were documented during the data collection for this study), 3 might become an issue that neurologists need to consider. However, with appropriate disclosure arrangements, survival prediction could also fulfil the patient's right to know, and enable them to make critical life decisions. I report grants from the National Institutes of Health, NINDS, and the National Ataxia Foundation, during the conduct of the study and outside the submitted work; and declare a patent entitled “Neurophilic nanoparticle” (drug delivery nanoparticle; serial number 14/390,584, pending). Copyright: Copyright 2019 Elsevier B.V., All rights reserved.

PY - 2018/4

Y1 - 2018/4

UR - http://www.scopus.com/inward/record.url?scp=85043578706&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85043578706&partnerID=8YFLogxK

U2 - 10.1016/S1474-4422(18)30088-7

DO - 10.1016/S1474-4422(18)30088-7

M3 - Comment/debate

C2 - 29553373

AN - SCOPUS:85043578706

SN - 1474-4422

VL - 17

SP - 292

EP - 294

JO - The Lancet Neurology

JF - The Lancet Neurology

IS - 4

ER -