Our ALS Proteomics Paper is Published in Nature Medicine

I’m very excited to share that our paper, “Longitudinal plasma proteomics predict phenoconversion to clinically manifest ALS,” has been published in Nature Medicine!

This work represents a long-running collaborative effort to find early, reliable biomarkers for ALS — a disease where presymptomatic detection remains one of the biggest barriers to effective intervention.

What We Did

We analyzed 516 longitudinal plasma samples from pre-symptomatic ALS gene carriers, individuals who later progressed to clinical ALS (phenoconverters), patients with manifest disease, and healthy controls. Using a panel of 92 proteins that showed detectable changes before phenoconversion, we identified a core set of 19 proteins that collectively predict phenoconversion up to 5 years in advance.

Key results:

• Cross-validated AUCs of 0.80–0.89 across 0.5-to-5-year prediction windows
• Mean absolute error of 1.6 years for estimating time-to-phenoconversion
• The 19-protein panel substantially outperformed neurofilament light chain (NfL) alone
• Partial replication in UK Biobank data supports the panel’s generalizability

Why It Matters

ALS prevention trials face a fundamental challenge: by the time patients are diagnosed, significant motor neuron loss has already occurred. A validated biomarker panel that can identify who will convert to clinical disease — and roughly when — could completely change how we design prevention studies and allocate experimental therapies.

This work adds a multi-protein, longitudinal dimension to the biomarker toolkit, complementing existing NfL-based approaches with stronger predictive accuracy across a broader time horizon.

Acknowledgments

This was truly a team effort. I’m grateful to all co-authors and the clinical teams who made longitudinal sample collection possible. Special thanks to Dr. Michael Benatar and Dr. Zhaohui Qin for their mentorship throughout this project.

You can read the full paper here: Nature Medicine

— Ximing