Machine intelligence identifies soluble TNFa as a therapeutic target for spinal cord injury

J R Huie,J L Nielson,J C Gensel,P G Schupp,J C Bresnahan,R R Ratan,N Kyritsis,M C Oldham,M R Segal,A Lin,A R Ferguson,J Z Pan,K.-A Irvine,M S Beattie

doi:10.1038/s41598-021-82951-5

Abstract

Traumatic spinal cord injury (SCI) produces a complex syndrome that is expressed across multiple endpoints ranging from molecular and cellular changes to functional behavioral deficits. Effective therapeutic strategies for CNS injury are therefore likely to manifest multi-factorial effects across a broad range of biological and functional outcome measures. Thus, multivariate analytic approaches are needed to capture the linkage between biological and neurobehavioral outcomes. Injury-induced neuroinflammation (NI) presents a particularly challenging therapeutic target, since NI is involved in both degeneration and repair. Here, we used big-data integration and large-scale analytics to examine a large dataset of preclinical efficacy tests combining five different blinded, fully counter-balanced treatment trials for different acute anti-inflammatory treatments for cervical spinal cord injury in rats. Multi-dimensional discovery, using topological data analysis (TDA) and principal components analysis (PCA) revealed that only one showed consistent multidimensional syndromic benefit: intrathecal application of recombinant soluble TNFα receptor 1 (sTNFR1), which showed an inverse-U dose response efficacy. Using the optimal acute dose, we showed that clinically-relevant 90 min delayed treatment profoundly affected multiple biological indices of NI in the first 48 h after injury, including reduction in pro-inflammatory cytokines and gene expression of a coherent complex of acute inflammatory mediators and receptors. Further, a 90 min delayed bolus dose of sTNFR1 reduced the expression of NI markers in the chronic perilesional spinal cord, and consistently improved neurological function over 6 weeks post SCI. These results provide validation of a novel strategy for precision preclinical drug discovery that is likely to improve translation in the difficult landscape of CNS trauma, and confirm the importance of TNFα signaling as a therapeutic target.

Highlights

Traumatic spinal cord injury (SCI) produces a complex syndrome that is expressed across multiple endpoints ranging from molecular and cellular changes to functional behavioral deficits
We first performed multidimensional data-curation/ integration to incorporate ensemble information from preclinical health records, histology, and behavioral function from 5 different unpublished, blinded preclinical neuroinflammation trials performed over 10 years in our laboratories[22]. We subjected this high-content data to topological data analysis (TDA) for data-driven discovery, pattern-detection, and dimensionality-reduction[20,23] (Fig. 1). This approach was used in a prior post hoc analysis of historic VISION-SCI data derived from the multicenter animal spinal cord injury study (MASCIS, see24) and allowed us to discover latent predictors of outcome that characterize the entire SCI syndrome as an integrated data system
We discovered potent predictors of outcome related to physiological measures during SCI induction, and less predictive, but significant, drug effects that had not been found in univariate analyses

Summary

Introduction

Traumatic spinal cord injury (SCI) produces a complex syndrome that is expressed across multiple endpoints ranging from molecular and cellular changes to functional behavioral deficits. This approach was used in a prior post hoc analysis of historic VISION-SCI data derived from the multicenter animal spinal cord injury study (MASCIS, see24) and allowed us to discover latent predictors of outcome that characterize the entire SCI syndrome as an integrated data system.

Results

Conclusion