Back

Proteome-wide and mendelian randomisation analyses of chronic widespread pain: diagnosis, prognosis, and drug target discovery

Chen, L.; Kelleher, E.; Meng, R.; Liu, D. k.; Guo, Y. c.; Wang, Y. h.; Gao, Y. q.; Huang, Z.; Liang, Z.; Yuan, S.; Zeng, C.; Lei, G. h.; Ma, J.; Dong, Y. h.; Irani, A.; Xie, J. q.; Alhambra, D. P.

2024-10-30 pain medicine
10.1101/2024.10.29.24316353 medRxiv
Show abstract

Chronic widespread pain (CWP) remains challenging due to its heterogenous causes and complex mechanisms. We analysed 2,923 plasma proteins from 29,254 UK Biobank participants. We first identified 811 proteins correlated with the presence of CWP cross- sectionally. We then created a sparse (top 10 proteins) and intricate (all significant proteins) proteomic-based score (ProtS) for CWP, both outperforming and improving the traditional clinical score (AUC: 0.82 and 0.88 vs 0.81 individually, and 0.87 and 0.92 in combination). Prospectively, the ProtS was associated with increased risks of a spectrum of pain traits, including dimensions from pain onset, progression and intensity, up to 13-years of follow-up; More importantly, we identified distinct proteomic signatures for nociplastic pain compared to nociceptive and neuropathic pain. For the individual proteins, carbonic anhydrase 14 (CA14) and leptin appeared as promising casual CWP biomarkers as triangulated by Mendelian randomisation and colocalization analyses. Lastly, our drug-repurposing analysis identified ten potential candidates for CWP therapies, calling for further research including randomised controlled trials. Although no CA14 agonists are currently available, CA14 remains a promising target, warranting further efforts to explore its role in pain modulation.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.