Longitudinal modeling to predict vital capacity in amyotrophic lateral sclerosis

Abstract

Objectives: Death in amyotrophic lateral sclerosis (ALS) patients is related to respiratory failure, which is assessed in clinical settings by measuring vital capacity. We developed ALS-VC, a modeling tool for longitudinal prediction of vital capacity in ALS patients.

Methods: A gradient boosting machine (GBM) model was trained using the PRO-ACT (Pooled Resource Open-access ALS Clinical Trials) database of over 10,000 ALS patient records. We hypothesized that a reliable vital capacity predictive model could be developed using PRO-ACT.

Results: The model was used to compare FVC predictions with a 30-day run-in period to predictions made from just baseline. The internal root mean square deviations (RMSD) of the run-in and baseline models were 0.534 and 0.539, respectively, across the 7L FVC range captured in PRO-ACT. The RMSDs of the run-in and baseline models using an unrelated, contemporary external validation dataset (0.553 and 0.538, respectively) were comparable to the internal validation. The model was shown to have similar accuracy for predicting SVC (RMSD = 0.562). The most important features for both run-in and baseline models were “Baseline forced vital capacity” and “Days since baseline.”

Conclusions: We developed ALS-VC, a GBM model trained with the PRO-ACT ALS dataset that provides vital capacity predictions generalizable to external datasets. The ALS-VC model could be helpful in advising and counseling patients, and, in clinical trials, it could be used to generate virtual control arms against which observed outcomes could be compared, or used to stratify patients into slowly, average, and rapidly progressing subgroups.

Keywords:

• Predictive modeling
• PRO-ACT
• neurodegenerative disease
• ALS
• vital capacity

Acknowledgements

This work was partially supported by a grant awarded to David L. Ennist from the Amyotrophic Lateral Sclerosis Association. We thank Deborah Stull and Nicholas Stilwell of Evidence Scientific Solutions for assistance with editing the manuscript; editorial assistance was funded by Cytokinetics, Inc.

Data used in the preparation of this article were obtained from the Pooled Resource Open-Access ALS Clinical Trials (PRO-ACT) Database. As such, the following organizations within the PRO-ACT Consortium contributed to the design and implementation of the PRO-ACT Database and/or provided data, but did not participate in the analysis of the data or the writing of this report: Neurological Clinical Research Institute, Massachusetts General Hospital, Northeast ALS Consortium, Novartis, Prize4Life, Regeneron Pharmaceuticals, Inc., Sanofi, and Teva Pharmaceutical Industries, Ltd.

Declaration of interest

SJ is an employee of Origent Data Sciences. AAT is an employee of and holds stock options in Origent Data Sciences. DB is an employee of Origent Data Sciences. MK is an employee of Right Start Consulting, Inc., which has a consulting relationship with and holds stock in Origent Data Sciences, Inc. LM and AB are employees of Cytokinetics, Inc. NA is a consultant to Biogen and MT Pharma and is funded by NIH grant K23NS083715. JA is a former employee and is currently a consultant to Cytokinetics, Inc. and is funded by the ALS Association. DLE is an employee and holds stock in Origent Data Sciences, received editorial assistance funding from Cytokinetics for this manuscript, and is funded by NIH grant R43NR015721 and ALS Association grants 17-LGCA-333 and 16-IIP-254.

Supplementary material available online.

Additional information

Funding

This work was supported by the Amyotrophic Lateral Sclerosis Association under grant 17-LGCA-333.