Thread by @ZdenekVrozina

We now have a complementary study that helps extend the picture. If the first paper suggests post-COVID biology may exist on a spectrum, this second one suggests recovery itself may also be real, prolonged, and only partial.🧵

This was a 2 year longitudinal proteomics study of hospitalized COVID survivors. The researchers profiled plasma at 6 months, 1 year, and 2 years after symptom onset and compared it with matched healthy controls.

This cohort was infected in the very early 2020 wave.

The encouraging part is that recovery clearly does happen. The number of differentially expressed proteins fell over time, from 172 at 6 months to 109 at 1 year and 73 at 2 years.

But recovery was not complete, and it was not synchronized. Different pathways normalized on different timelines, which is exactly what you would expect in a layered post infectious condition rather than a simple on/off state.

Some pathways, including focal adhesion, extracellular matrix-receptor interaction, and actin cytoskeleton regulation, improved between 6 months and 1 year.

Other systems appeared to recover more slowly. Cholesterol metabolism shifted closer to control levels between 1 and 2 years, while most immune response pathways - including complement and coagulation - also moved toward, though not necessarily fully back to, control levels before the 2-year mark.

But the reassuring reading would be a mistake. This was not a clean return to baseline. Even after 2 years, some neuron-related signaling pathways remained suppressed, and some oxidative stress-related proteins were still persistently altered. The biology was still carrying damage forward.

And even where parts of the proteomic immune signal moved closer to controls, that should not be mistaken for full recovery. What this study shows is uneven, partial recovery at best - not a complete reset.

That matters because it pushes back against two comforting myths at once. Not only the idea that people either have or don’t have long COVID, but also the idea that recovery is naturally rapid, uniform, or complete.

The picture here is not one of restoration. It is one of prolonged biological disruption, with some systems improving, others lagging, and some still measurably abnormal years later.

And this broader pattern is not without precedent. We have seen in other infections - especially HIV - that biological recovery can remain incomplete long after the acute phase, with persistent effects in immune signaling, oxidative stress, and neuroinflammatory pathways.

It does mean the basic idea is not exotic. An infection can end clinically, while parts of the biology remain dysregulated for much longer.

That is why post-COVID illness may be better understood not as a simple yes/no condition, but as a prolonged and uneven process of partial recovery.

And if 40 years of HIV research still haven’t answered every post-infectious question, the lesson should be obvious. Take non-pharmaceutical prevention seriously and avoid reinfections. Prevention matters far more than people want to admit.

Gu at al., Probing long COVID through a proteomic lens: a comprehensive two-year longitudinal cohort study of hospitalised survivors. thelancet.com/journals/ebiom…