The study, conducted by Chuan-Xin Duan of the National Cancer Center in Beijing and Lang Yang of China-Japan Friendship Hospital, focused on two proteins: fibroblast growth factor 2 (FGF2) and peroxiredoxin 3 (PRDX3). Both have plausible biological connections to the processes thought to underlie Long COVID, including tissue repair, inflammation, and oxidative stress. The researchers asked a deceptively simple question: does adding measurements of these two proteins, taken during acute infection, meaningfully improve the prediction of Long COVID beyond what can already be achieved with a patient’s age and sex alone?

What distinguishes this analysis is the discipline of its design. Rather than searching through hundreds or thousands of proteins for the best-performing combination, the authors fixed their protein pair before ever looking at outcome data. This choice was made possible by a practical constraint: FGF2 and PRDX3 were the only proteins with one-to-one assay mappings between the different proteomic platforms used by the cohorts under study. In other words, these were the only markers whose measurements could be compared in a like-for-like manner across datasets generated with different technologies. By locking in the pair in advance, the researchers avoided the well-known trap of overfitting, in which models appear accurate on the data used to build them but fail on new patients.

The analytical pipeline drew on two independent cohorts. The first, Su/INCOV, served as the reference or training dataset. It contained 204 participants with complete predictor information, but only 125 of them had linked follow-up data documenting whether they developed Long COVID. The strict analysis ultimately included 122 participants, of whom 75 experienced the outcome. The second cohort, from Zurich, served as the validation set, comprising 113 participants with 40 events. Crucially, the researchers did not refit their model to the Zurich data. Instead, they applied the coefficients estimated in the Su cohort directly to the Zurich participants, standardizing protein values within each cohort beforehand. This procedure, known as coefficient transfer, is a stringent test of whether a model truly generalizes or merely memorizes the quirks of its original dataset.

To evaluate performance, the researchers relied on the Brier score, a standard metric in prediction research that combines discrimination and calibration into a single number. Lower Brier scores indicate better prediction. The key comparison was between a baseline model using only age and sex, designated C0, and an enhanced model that also incorporated FGF2 and PRDX3, designated C2. The difference between these two scores, after statistical correction for optimism, would reveal whether the proteins added genuine predictive value.

The answer, in both cohorts, was no. In the Su/INCOV reference cohort, the optimism-corrected difference between the protein model and the baseline model was +0.00342, with a 95 percent interval spanning from −0.00608 to +0.02834. Because this interval crosses zero, the difference is statistically inconclusive; the protein model might be slightly better or slightly worse, but there is no reliable evidence of improvement. The Zurich results told a similar story. The transferred model produced a difference of +0.01093 (95 percent interval, −0.01721 to +0.03882), and a strict-model transfer gave +0.00902 (−0.01809 to +0.03696). Again, the intervals included zero, and the positive point estimates suggested that adding the proteins may even have made prediction marginally worse.

Calibration emerged as a particular problem. Both models overpredicted Long COVID risk when applied to the Zurich cohort, meaning they systematically estimated higher probabilities than what actually occurred. In prediction research, calibration failures of this kind are often more damaging than modest losses of discrimination, because they distort the risk estimates that would inform clinical decisions. A patient told they have a 60 percent chance of developing Long COVID, when the true figure is closer to 35 percent, faces unnecessary anxiety and potentially inappropriate interventions. When the researchers, at reviewers’ request, updated the model’s intercept to correct for this miscalibration, the difference between the protein model and the baseline remained inconclusive, reinforcing the overall finding.

The authors are candid about the limits of their pipeline. Because protein standardization depends on the characteristics of a target cohort, the approach as implemented is not ready to predict risk for an individual patient. Standardization, which rescales protein values to a common reference distribution, is a necessary step when combining measurements from different proteomic platforms. But it introduces a dependency on the very cohort to which the model is being applied, which complicates the clean transfer of coefficients from one dataset to another. This technical subtlety is often overlooked in biomarker studies that celebrate apparent predictive success within a single dataset, only to see that success evaporate in external validation.

The study also highlights the challenges of working with modest sample sizes. With 75 events in the reference cohort and 40 in the validation cohort, the statistical intervals around the performance differences are wide. A truly helpful biomarker effect of small magnitude could hide within those intervals, just as a harmful one could. Larger, prospectively designed cohorts with complete follow-up would be needed to settle the question definitively. The researchers note that their secondary analysis used only publicly accessible, deidentified data from the source studies, and that while the original result-blind protocol and statistical analysis plan are preserved in a versioned reproducibility archive, the analysis was not registered in a public study registry.

For the broader field of Long COVID research, the message is one of methodological caution rather than outright pessimism. FGF2 and PRDX3, at least as measured in this pipeline, did not reliably improve prediction beyond age and sex, two of the most basic demographic predictors available. But the study’s framework, with its pre-specified biomarker pair, strict coefficient transfer, and honest reporting of inconclusive results, offers a template for how candidate biomarkers should be evaluated before being promoted as clinical tools. As proteomic technologies mature and larger cohorts become available, the search for acute-phase predictors of Long COVID will continue. This study demonstrates that the bar for claiming success must include not just accuracy within a dataset, but demonstrable transportability across cohorts and platforms, with calibration that holds when the model meets patients it has never seen.

Subject of Research: Evaluation of acute-phase blood proteins FGF2 and PRDX3 for predicting Long COVID across cohorts and proteomic platforms

Article Title: Evaluation of acute FGF2 and PRDX3 for long COVID prediction across cohorts and proteomic platforms

Article References: Duan, C.-X., & Yang, L. (2026). Evaluation of acute FGF2 and PRDX3 for long COVID prediction across cohorts and proteomic platforms. Scientific Reports. https://doi.org/10.1038/s41598-026-75618-6

Image Credits: AI Generated

DOI: 10.1038/s41598-026-75618-6

Keywords: Long COVID, FGF2, PRDX3, biomarkers, proteomics, prediction models, coefficient transfer, calibration, Brier score, transportability, reproducibility, SARS-CoV-2

The material at the heart of the study, published in the journal Angiogenesis, is a polymer built from 2-methacryloyloxyethyl phosphorylcholine, abbreviated MPC, whose side chains carry the same phosphorylcholine head group that decorates the outer leaflet of every human cell membrane. Phosphorylcholine-based polymers have a long and distinguished history in biomedical engineering precisely because of this biomimicry. They are zwitterionic, meaning they carry paired positive and negative charges that produce a strongly hydrated surface, which resists protein adsorption, platelet adhesion, and thrombus formation. That property has made MPC polymers the coating of choice for cardiovascular stents, contact lenses, and other blood-contacting medical devices. What the Montreal team, led by co-senior authors May Griffith and Bruno Larrivée, has done is to flip the script on this famously inert material: instead of exploiting phosphorylcholine chemistry to keep cells at arm’s length, they deploy nanoparticles of the polymer as an active biological stimulus that vessel cells willingly devour.

The nanoparticles themselves are formed from polyMPC, the polymerized form of the methacrylated monomer, and were engineered to fall below a critical size threshold of roughly 150 nanometers in diameter. That figure is not arbitrary. Work on size-dependent cellular uptake stretching back to classic studies of clathrin- and caveolae-mediated endocytosis has shown that particles in this range are internalized by cells far more efficiently than larger counterparts, which tend to remain at the surface or enter through less productive pathways. When the researchers exposed human umbilical vein endothelial cells, the workhorse HUVEC model of vascular biology, to these sub-150-nanometer particles, the cells readily swallowed them. Once inside, the nanoparticles did something remarkable: rather than sitting inertly in endosomes, they triggered a cascade of signaling events that pushed the endothelial cells into an actively angiogenic state.

Tracing that cascade led the team to protein kinase C, or PKC, a family of enzymes that sits at a major signaling crossroads in many cell types. Upon internalization of the polyMPC nanoparticles, PKC was recruited to the plasma membrane, the canonical step that converts the kinase from an inactive cytosolic form into its active, membrane-bound state. PKC activation in endothelial cells is tightly linked to the stimulation of endothelial nitric oxide synthase, which produces nitric oxide, a master regulator of vessel dilation, endothelial migration, and new vessel sprouting. The finding that a synthetic zwitterionic polymer can engage this pathway offers a mechanistic explanation for how a material long considered biologically quiet can, in nanoparticle form, become a potent pro-angiogenic agent. It also suggests that the cell membrane itself, or the membrane trafficking machinery that processes the internalized particles, may be the unexpected trigger that converts a physical delivery event into a biochemical signal.

The downstream consequences of this activation were mapped with RNA sequencing, an unbiased technique that captures the full transcriptional response of treated cells. The profiles of HUVECs exposed to polyMPC nanoparticles showed upregulation of pro-angiogenic growth factors, including VEGF and fibroblast growth factor 2 (FGF2), two of the most important drivers of blood vessel growth in both development and repair. In other words, the nanoparticles did not act as growth factors themselves; they acted upstream, flipping genetic switches inside endothelial cells so that the cells became factories for their own angiogenic signals. Functional assays confirmed the transcriptional story: treated endothelial cells proliferated more vigorously and mounted characteristic angiogenic responses in culture, forming the tube-like networks that are the in vitro hallmark of vessel-forming behavior.

Crucially, the pro-angiogenic effect was not confined to a single cell type. The researchers found that polyMPC nanoparticles also promoted the expansion of endothelial progenitor cells derived from human cord blood and bone marrow. These progenitors are the circulating seed cells that the body uses to repair and rebuild vasculature, and their mobilization and expansion are central goals of cell-based approaches to ischemic disease. A material that can simultaneously stimulate mature endothelial cells and support the growth of progenitor populations could therefore act on both arms of vascular regeneration, the local sprouting of new vessels from existing ones and the recruitment of the cellular precursors that seed new vascular networks.

To test whether the effect would survive the far messier environment of living tissue, the team implanted polyMPC nanoparticles inside Matrigel plugs under the skin of mice, a standard in vivo assay in which a basement-membrane extract serves as a neutral scaffold. Blood vessels from the surrounding host tissue grow into the plug only if something inside it attracts them, so the degree of vascularization provides a direct readout of pro-angiogenic activity. The polyMPC-laden plugs produced a significant angiogenic response, drawing in host vessels at levels that confirmed the cell culture results translated to a living organism. Equally important was a observation about persistence: the delivered nanoparticles could be retained in tissue for extended periods, which means a single implantation could provide a sustained stimulus rather than the brief pulse of activity that plagues injected protein therapeutics. Because the nanoparticles work by inducing the tissue’s own production of VEGF, FGF2, and other growth factors, the therapeutic signal is continuously renewed at the site where it is needed, in the correct proportions and in the natural context of the tissue microenvironment.

The clinical problem the researchers are aiming at is enormous. Ischemic vascular pathologies, including peripheral artery disease, critical limb ischemia, and myocardial infarction, affect hundreds of millions of people worldwide and are a leading cause of amputation and heart failure. Existing attempts at therapeutic angiogenesis have ranged from gene therapy delivering VEGF or hepatocyte growth factor genes to infusions of autologous stem cells, with mixed and often underwhelming results in controlled trials. Recombinant protein approaches face pharmacological hurdles: free growth factors have short half-lives, risk ectopic vessel growth and leaky, malformed vasculature, and are costly to produce in pharmaceutical grade. The polyMPC strategy sidesteps several of these objections at once. The particles are made from a polymer with an established safety record in approved medical devices, they are relatively inexpensive to synthesize compared with recombinant proteins or viral vectors, and their mechanism recruits the body’s own regulated programs of vessel growth rather than flooding tissue with a single supraphysiological factor.

There are also broader implications for the field of biomaterials science. Phosphorylcholine polymers have been celebrated for four decades as the archetypal non-fouling, non-activating surface chemistry, and their zwitterionic cousins are now being engineered into next-generation lipid nanoparticles for drug and RNA delivery, where their ability to modulate interactions with lipid rafts and membrane trafficking pathways is increasingly recognized. The Montreal study adds a new chapter to that story by showing that the same membrane-mimetic chemistry, presented in the right physical form and size range, can be an active participant in cell signaling rather than a passive bystander. The boundary between a biomaterial that hides from biology and one that directs biology may be thinner than the field has assumed, and size, presentation, and cell type all appear to determine which side of that boundary a given formulation occupies.

The work remains at the preclinical stage, and the path from subcutaneous Matrigel plugs in mice to treating a diabetic patient’s ischemic foot is long and demanding, requiring tests of dose, delivery route, long-term biocompatibility, and efficacy in models of true tissue ischemia. The authors have filed a provisional patent application through their hospital research centre, a signal that they see a translational pathway for the technology. Even at this early stage, the concept is striking: a cheap, stable, biocompatible polymer particle that quietly enters blood vessel cells, wakes up their internal growth programs, and leaves the tissue to heal itself with vessels built from its own blueprint. If that concept survives the gauntlet of preclinical and clinical testing, phosphorylcholine, one of the most familiar molecules in biomaterials science, may find its most consequential role not in coating devices, but in regrowing the vasculature that failing tissues so desperately need.

Subject of Research: Polymer nanoparticles that stimulate angiogenesis in ischemic tissue

Article Title: Angiogenesis-promoting properties of phosphorylcholine mimetic nanoparticles

Article References: Moradi, M., Poudel, B. K., Hooker, E., Aghajanzadeh-Kiyaseh, M., Rusu, D., Griffith, M., & Larrivée, B. (2026). Angiogenesis-promoting properties of phosphorylcholine mimetic nanoparticles. Angiogenesis, 29(4), Article 71. https://doi.org/10.1007/s10456-026-10094-0

Image Credits: AI Generated

DOI: 10.1007/s10456-026-10094-0

Keywords: angiogenesis, nanoparticles, polyMPC, phosphorylcholine, therapeutic angiogenesis, endothelial cells, protein kinase C, VEGF, FGF2, endothelial progenitor cells, RNA sequencing, ischemic disease