Creatine kinase and prognosis in amyotrophic lateral sclerosis: a literature review and multi-centre cohort analysis

Journal of Neurology https://doi.org/10.1007/s00415-022-11195-8 ORIGINAL COMMUNICATION Creatine kinase and prognosis in amyotrophic lateral sclerosis: a literature review and multi‑centre cohort analysis Jiali Gao1 · Thanuja Dharmadasa1 · Andrea Malaspina2 · Pamela J. Shaw3 · Kevin Talbot1 · Martin R. Turner1 · Alexander G. Thompson1 Received: 18 January 2022 / Revised: 5 May 2022 / Accepted: 15 May 2022 © The Author(s) 2022 Abstract Background Amyotrophic lateral sclerosis (ALS) is a prognostically heterogeneous neurodegenerative disease. Blood creatine kinase (CK) level has been inconsistently reported as a prognostic biomarker and raised levels in some ALS patients have been presumed to reflect muscle wasting, which is also variable. Methods MEDLINE was systematically searched for papers related to CK in ALS and the relevant studies were reviewed. Using data from 222 ALS patients in a multi-centre, prospective, longitudinal cohort, survival analyses using Kaplan–Meier and Cox proportional hazards models were undertaken in relation to CK and other prognostic factors. Results Twenty-five studies investigating CK in ALS were identified, of which 10 specifically studied the link between CK and survival. Five studies observed no association, four found that higher CK levels were associated with longer survival and one, the opposite. In our cohort (n = 222), 39% of patients had a CK level above the laboratory reference range. Levels were higher in males compared to females (p < 0.001), in patients with limb versus bulbar onset of symptoms (p < 0.001) and in patients with higher lower motor neuron burden (p < 0.001). There was no significant trend in longitudinal CK values. Although a higher standardised log (CK) at first visit was associated with longer survival in univariate analysis (hazard ratio 0.75, p = 0.003), there was no significant association after adjusting for other prognostic covariates. Conclusion While raised CK levels in ALS do reflect lower motor neuron denervation to a large extent, they are not independently associated with survival when measured in the symptomatic phase of the disease. Keywords Motor neuron disease · Creatine kinase · Biomarker Introduction Amyotrophic lateral sclerosis (ALS) is a fatal neurodegenerative disease characterised by progressive muscle weakness due to degeneration of the upper and lower motor neurons, sharing clinicopathology with frontotemporal dementia. The prognosis of ALS varies greatly between individuals. * Martin R. Turner * Alexander G. Thompson 1 Nuffield Department of Clinical Neurosciences, University of Oxford, John Radcliffe Hospital, West Wing level 3 / level 6, Oxford OX3 9DU, UK 2 Institute of Neurology, University of London, London, UK 3 Sheffield Institute for Translational Neuroscience, University of Sheffield, Sheffield, UK Although most patients die within 3 years of symptom onset, 10% of patients survive beyond 10 years [1]. Predicting survival by clinical variables alone remains limited at the individual patient level [2]. The search for ALS biomarkers across a range of techniques has yielded several candidates with the potential to improve prognostic stratification [3, 4]. A blood-based biomarker applicable in the routine clinic environment would be valuable. Creatine kinase (CK) catalyses the reversible conversion of creatine and adenosine triphosphate (ATP) to phosphocreatine and adenosine diphosphate (ADP), an essential component of ATP recycling and energy metabolism [5]. An elevated level of serum CK is usually considered a marker of damage to CK-rich tissues, such as muscle [6]. CK is recognised to be raised in a proportion of ALS patients [7], with a common explanation being that levels reflect secondary muscle breakdown as a result of denervation. CK would represent an attractive biomarker, given its accessibility and 13 Vol.:(0123456789) Journal of Neurology relative low expense. However, studies of the association between blood CK and survival in ALS have yielded conflicting results. We investigated the association between CK levels and survival in ALS through a review of the published literature and analysis of a large, multi-centre, prospective UK cohort, including longitudinal and other prognostic factor data. Materials and methods Literature review The MEDLINE database was searched on 05/08/21 for articles relating to search terms: “motor neurone disease” or “amyotrophic lateral sclerosis” and “creatine kinase”. The Fig. 1  PRISMA flow diagram for study identification full search strategy can be found in Supplementary Information. Abstracts were screened (JG) and all potentially relevant studies underwent full text review. Foreign language articles, in vitro or non-human studies and reviews, commentaries and editorials reporting non-original data were excluded. Articles reporting original data regarding serum CK levels in ALS patients were included. Figure 1 indicates the flow chart for study selection. Bibliographic data, study location and timing, sample size, population demographic data, average serum CK levels and the upper laboratory limit for serum CK were extracted from the studies (JG). Studies including any analyses of CK levels and survival underwent further analyses. From these, the method of survival analysis, the outcome (e.g. hazard ratio or median survival), any confounders included in the model Identification Identification of studies via databases and registers Records identified from: PubMed/MEDLINE (n= 235) Included Screening Records screened (n = 235) 13 Records excluded (n = 206) Case report/series (n= 25) Animal/in vitro studies (n= 26) Review/Meta-analysis (n= 1) Irrelevant (n= 154) Reports sought for retrieval (n = 29) Reports not retrieved (n = 0) Reports assessed for eligibility (n = 29) Reports excluded (n=4) Case report/series (n= 3) PLS only (n= 1) Studies included (n=25) CK in MND (n = 23) CK and survival (n=10) Journal of Neurology and the overall conclusion were extracted. These articles were quality reviewed and assessed for risk of bias using the Risk of Bias Assessment tool for Non-randomized Studies (RoBANS) by JG [8]. Patients ‘AMBRoSIA’ (A Multicentre Biomarker Resource Strategy In ALS) is a prospective longitudinal cohort of those attending tertiary ALS referral clinics in Oxford, Sheffield and London, UK. Results of combined biomarker analysis have been published elsewhere [9]. Recruitment commenced in June 2017. Longitudinal data were obtained during routine follow-up clinic visits at intervals of 3–6 months, with a censorship date for survival analyses of 01/02/2020. Survival time was defined as time from symptom onset to death or censorship. Patients with a diagnosis of primary lateral sclerosis were excluded from this analysis. Venous blood was collected using the BD Vacutainer Safety-Lock set. Serum CK was measured by Clinical Biochemistry laboratories at Oxford University Hospitals NHS Foundation Trust, S (...truncated)