The MITOS system predicts long-term survival in amyotrophic lateral sclerosis

Objective: The choice of adequate proxy for long-term survival, the ultimate outcome in randomised clinical trials (RCT) assessing disease-modifying treatments for amyotrophic lateral sclerosis (ALS), is a key issue. The intrinsic limitations of the ALS Functional Rating Scale-Revised (ALSFRS-R), including non-linearity, multidimensionality and floor-effect, have emerged and its usefulness argued. The ALS Milano-Torino staging (ALS-MITOS) system was proposed as a novel tool to measure the progression of ALS and overcome these limitations. This study was performed to validate the ALS-MITOS as a 6-month proxy of survival in 200 ALS patients followed up to 18 months. Methods: Analyses were performed on data from the recombinant human erythropoietin RCT that failed to demonstrate differences between groups for both primary and secondary outcomes. The ALS-MITOS system is composed of four key domains included in the ALSFRS-R scale (walking/self-care, swallowing, communicating and breathing), each with a threshold reflecting the loss of function in the specific ALSFRS-R subscores. Sensitivity, specificity and the area under the curve of the receiver operating characteristic curves of the ALS-MITOS system stages and ALSFRS-R decline at 6 months were calculated and compared with the primary outcome (survival, tracheotomy or >23-hour non-invasive ventilation) at 12 and 18 months Predicted probabilities of the ALS-MITO system at 6 months for any event at 12 and 18 months were computed through logistic regression models. Results: Disease progression from baseline to 6 months as defined by the ALS-MITOS system predicted death, tracheotomy or >23-hour non-invasive ventilation at 12 months with 82% sensitivity (95% CI 71% to 93%, n=37/45) and 63% specificity (95% CI 55% to 71%, n=92/146), and at 18 months with 71% sensitivity (95% CI 61% to 82%, n=50/70) and 68% specificity (95% CI 60% to 77%, n=76/111). The analysis of ALS-MITOS and ALSFRS-R progression at 6-month follow-up showed that the best cut-off to predict survival at 12 and 18 months was 1 for the ALS-MITOS (ie, loss of at least one function) and a decline ranging from 6 to 9 points for the ALSFRS-R. Conclusions: The ALS-MITOS system can reliably predict the course of ALS up to 18 months and can be considered a novel and valid outcome measure in RCTs. © 2015, BMJ Publishing Group. All rights reserved.