Multi-Omics: CRP Predicts HIV-AECOPD Prognosis
- Jul 20
- 4 min read
Clinicians managing an acute COPD exacerbation in a patient living with HIV face a hard question at discharge: who is about to relapse within three months? The usual severity scores were not built for this dual-inflammation phenotype. A recent multi-omics study paired serum proteomics with untargeted metabolomics across HIV-AECOPD, AECOPD-alone, and healthy control groups, then tested the strongest signals in a separate cohort. The winning marker was cheap and already on every hospital chemistry panel: C-reactive protein.
Key Takeaways
Multi-omics profiling of serum can surface a single, clinic-ready prognostic marker even when the biology spans immune, complement, and lipid pathways at once.
In HIV-associated COPD exacerbations, C-reactive protein predicted three-month poor outcomes with an AUC of 0.81, beating five other candidate proteins and metabolites.
Falling complement C3, not C4, tracked with disease severity, pointing to active complement consumption rather than broad immune shutdown.
What Multi-Omics Integration Revealed in HIV-AECOPD Serum
The discovery arm was small on purpose: five patients per group, profiled by DIA mass spectrometry and UHPLC-QTOF metabolomics. Against healthy serum, the HIV-AECOPD proteome shifted by 192 differentially expressed proteins, with humoral immunity, complement and coagulation cascades, and neutrophil extracellular trap formation leading the pathway list. Metabolomics told a parallel oxidative-stress story, flagging glutathione metabolism and a drop in the antioxidant enzyme PON1. A cross-omics correlation network then tied six hub proteins to six metabolites, so the inflammatory and metabolic signals read as one coupled network. Following the framing Hasin, Seldin, and Lusis (2017) set out for multi-omics integration, that coupling is where the interpretation lives.
Key Findings
CRP led the validation panel with an AUC of 0.812 (95% CI 0.628 to 0.997) and the highest random-forest importance (Mean Decrease Gini 1.33) for three-month prognosis.
Serum amyloid A followed closely (AUC 0.764); triglycerides, IgG, C3, and C4 each stayed below 0.63.
Complement C3 fell in an inverse gradient across HIV-AECOPD, AECOPD, and controls, while C4 barely moved.
Acute-phase proteins CRP, SAA2, LBP, and LRG1 rose together; the antioxidant PON1 and lipid carrier APOC2 moved the opposite way.
Figure 1. Study workflow. A discovery cohort of three groups (control, AECOPD, HIV-AECOPD; five each) supplied serum for LC-MS/MS proteomics and metabolomics, then correlation and protein-metabolite network analysis. An independent cohort of twenty per group was profiled on clinical analyzers and linked to three-month prognosis. Adapted from Qin et al. (2026), Frontiers in Immunology.
Complement Consumption as a Readable Disease Signal
The C3 result is the quiet standout. A dropping C3 with a flat C4 is the classic fingerprint of alternative-pathway activation, and it lines up with the humoral-immunity and NET signatures the proteomics flagged. For a coinfected population where generic inflammation markers blur together, a directional complement readout says more than another high cytokine number.
Why Untargeted Metabolomics Earned Its Place
Proteomics alone would have named CRP and SAA, but not why triglycerides climb or antioxidant defense buckles. Untargeted metabolomics surfaced glutathione metabolism and TCA-cycle intermediates, and correlating those with APOC2 and PON1 gave a plausible lipid-oxidation mechanism behind the inflammatory numbers. That is the case for measuring both layers rather than betting on one.
From One Serum Tube to an Integrated Readout
Reading proteins and metabolites from the same serum aliquot is what makes a correlation network like this believable; split the sample across labs and batch structure starts to masquerade as biology. Our group runs this kind of single-injection multi-omics workflow at Dalton, measuring the proteome and metabolome from one draw so protein-metabolite pairings survive scrutiny. For teams planning multi-omics biomarker discovery at cohort scale, that shared-sample design removes a confounder the authors would otherwise argue around.
Frequently Asked Questions
What is multi-omics analysis used for in biomarker research?
Multi-omics analysis measures several molecular layers, such as proteins and metabolites, from the same samples and integrates them, finding biomarkers a single assay would miss. Here it linked inflammatory proteins to lipid and antioxidant metabolites in one network.
Can CRP predict outcomes in HIV-associated COPD exacerbations?
In this study CRP was the strongest single predictor of three-month poor outcomes, with an AUC of 0.81, ahead of serum amyloid A, triglycerides, and complement proteins. The cohort was small, so it reads as a promising lead, not a settled rule.
How much sample does a multi-omics proteomics and metabolomics study need?
Often less than people expect. This discovery phase deeply profiled serum from five patients per group, then validated hits in twenty per group, with both layers run from a single serum tube.
Conclusion
What is believable now: CRP carries real short-term prognostic signal in HIV-AECOPD, and the complement-plus-lipid biology behind it is coherent. Whether that AUC holds outside one small, single-center cohort isn't yet proven. If you treat these patients, the practical move isn't a new assay; it's reading a CRP you already have with fresh eyes at discharge.
Related Reading
See how we run these analyses in one lab: Dalton's multi-omics CRO services.
Citation
Qin, T., Huang, W., Yang, C., Lu, T., Wang, J., Chen, J., Chen, L., Yan, T., Qian, T., Yang, H., Lu, L., Huang, D., & Zhao, M. (2026). Integrated multi-omics identifies CRP as a prognostic biomarker and reveals complement consumption in HIV-associated AECOPD. Frontiers in Immunology, 17, 1855646. https://doi.org/10.3389/fimmu.2026.1855646
Note
This blog post summarizes findings from the above-cited research. Figures are adapted from the original publication. For full details, please refer to the source article.
By Seungjun Yeo, CEO at Dalton Bioanalytics. Specializing in multi-omics mass spectrometry for drug discovery and biomarker research.