The emergence of long COVID has fundamentally challenged our understanding of post-viral recovery, with the global patient population surging from 60 million to an estimated 400 million between 2020 and 2024. As the medical community grapples with this scale, Biopharma expert Ivan Kairatov brings a deep research perspective to the puzzling disconnect between patient symptoms and clinical data. With extensive experience in innovation and drug development, Kairatov explores why standard diagnostics often fail to capture the biological reality of those suffering from persistent fatigue and cognitive impairment nearly two years after their initial infection.
This conversation delves into the evolving clinical profile of long-term recovery, the disappearance of traditional inflammatory markers in late-stage cases, and the critical role of participant selection in clinical trials. Kairatov also discusses the potential of ultra-sensitive diagnostic technologies and shifts the focus toward how healthcare providers can manage long-term patient care when traditional blood work suggests everything is normal.
With long COVID cases reaching an estimated 400 million globally, symptoms often persist for years even as traditional inflammatory markers appear to stabilize. How does the clinical profile change for patients over a two-year period, and what specific metrics should providers monitor when routine blood work returns normal results?
The clinical landscape of long COVID is incredibly fluid, often starting as a multi-organ inflammatory storm and settling into a chronic state dominated by fatigue and cognitive impairment. In the early months, we see a wide range of fluctuating symptoms, but as we cross the one-to-two-year mark, these often crystallize into a profile that looks very much like other post-viral syndromes. When routine tests like C-reactive protein (CRP) come back normal, providers need to look toward more nuanced indicators of functional health rather than just systemic inflammation. While the 400 million people affected worldwide show that this is a systemic crisis, the “normalization” of labs at 70 weeks suggests we should monitor patient-reported outcome measures and functional capacity. Even if the inflammatory markers have stabilized, the lived experience of the patient remains the most reliable metric for the severity of the disease in its late stages.
Research conducted nearly 70 weeks post-infection suggests a lack of detectable neuroinflammation or neuronal damage markers like NfL and GFAP. Why might these biological signals vanish even when patients remain debilitated, and what alternative mechanisms, such as deep-tissue viral reservoirs, could explain this disconnect?
It is striking to see that markers like neurofilament light (NfL) and glial fibrillary acidic protein (GFAP), which are standard sentinels for brain injury, show no significant elevation in these late-stage patients. This suggests that the overt stage of neuronal damage or active neuroinflammation has likely resolved by the 69-week median point observed in recent studies. We are likely looking at a “hit-and-run” scenario where the initial viral trigger caused significant disruption, but the ongoing symptoms are driven by something other than active tissue damage. Hypotheses are now shifting toward the possibility of viral reservoirs hiding in deep tissues or extremely low-level immune activation that is simply too subtle for even advanced blood tests to catch. The symptoms persist not because the brain is currently being damaged, but perhaps because the neurological pathways haven’t regained their equilibrium after the initial insult.
Many studies include patients with preexisting autoimmune or chronic inflammatory conditions, potentially skewing data on circulating biomarkers. What is the impact of excluding these comorbidities in clinical trials, and how does isolating “pure” cases change our understanding of the disease’s unique biological footprint?
By excluding individuals with preexisting autoimmune diseases, cancer, or untreated comorbidities, we can finally strip away the “noise” that has clouded earlier long COVID research. In our study of 96 well-matched participants, removing these factors revealed that long COVID itself may not be characterized by chronic systemic inflammation in the long run. This is a pivotal realization because many earlier studies likely captured the inflammatory signatures of the patients’ preexisting conditions rather than the virus itself. When we isolate “pure” cases, we see that inflammatory markers like IL-6 and TNF-α eventually normalize, matching the profiles of healthy controls who recovered fully. This tells us that the biological footprint of long COVID at 18 months is not one of persistent high-level inflammation, which forces us to rethink our therapeutic targets.
There is often no statistical correlation between circulating biomarker levels and the actual severity of symptoms like fatigue or cognitive impairment. In a clinical setting, how do you manage patient expectations when labs are clear, and what diagnostic steps should follow to address functional limitations?
The lack of correlation between a lab value and a patient’s exhaustion can be incredibly frustrating for both the clinician and the patient, leading to a sense of being “gaslit” by the data. We must communicate clearly that a normal CRP or NfL test does not mean the symptoms are imaginary; it simply means the biological “fire” has gone out, even if the “charred remains” of the functional system haven’t been repaired. Diagnostic steps should shift away from repeated blood draws and toward comprehensive functional assessments, such as cognitive testing or metabolic stress tests. Managing expectations involves validating the patient’s physical limitations while explaining that our current tools are sometimes not sensitive enough to capture the microscopic dysfunctions occurring at the cellular level. We have to treat the patient’s functional reality rather than chasing a biomarker that may no longer be relevant to their current state.
Hypotheses for long-term symptoms range from persistent organ damage to post-viral syndromes where the initial trigger is gone. Based on the observation that inflammation eventually normalizes, what are the practical implications for treatment, and how should clinicians decide between anti-inflammatory therapies and rehabilitative care?
Since inflammation markers like IL-6 and TNF-α tend to normalize at the 70-week mark, the window for aggressive anti-inflammatory therapy may actually be quite narrow and confined to the early stages of the illness. For a patient who is nearly a year and a half into their journey, continuing with systemic corticosteroids or potent anti-inflammatories may offer little benefit and could even introduce unnecessary side effects. The data suggests we should pivot toward rehabilitative care and supportive therapies that focus on neuroplasticity and energy management. If the biological markers are no longer signaling an active “attack” on the body, the goal shifts from stopping damage to restoring function and managing the chronic physiological changes left in the virus’s wake. It is a transition from an “acute rescue” mindset to a “long-term recovery” strategy.
Ultra-sensitive assays like NULISA can detect extremely low-level signals that standard hospital tests might miss. How should the medical community integrate these high-precision tools into standard practice, and what are the step-by-step challenges of transitioning from research-grade technology to everyday diagnostic use?
The transition of tools like the NULISA platform from a research setting to a local clinic is a massive undertaking that involves rigorous standardization and cost-reduction efforts. These assays are incredibly powerful, showing nominal differences in markers like TREM2 and IL-1β that standard hospital tests miss entirely, but they currently report in “normalized protein expression units” rather than absolute concentrations. To make these tools useful for a family doctor, we first need to establish universal reference ranges so that a result in one lab means the same thing in another. Second, we must prove that detecting these ultra-low signals actually changes the treatment plan for the patient, rather than just providing more data for the sake of data. Finally, the infrastructure of hospital laboratories would need significant upgrades to handle these nucleic acid-linked immuno-sandwich assays, which is a significant financial and technical hurdle.
What is your forecast for long COVID?
I believe our understanding of long COVID will eventually merge with our broader understanding of other post-viral illnesses like Myalgic Encephalomyelitis/Chronic Fatigue Syndrome (ME/CFS), leading to a unified treatment approach. As we see inflammation normalize in the blood of 85% of our female-heavy cohorts after 69 weeks, the focus will move away from the “virus” and toward the “host response” and metabolic recovery. We will likely see a decline in the use of broad anti-inflammatory drugs and an increase in personalized, neuro-rehabilitative protocols. My forecast is that while the massive surge of new cases may slow, we will be managing the functional ripples of the 400 million already affected for at least the next decade, requiring a permanent shift in how we handle chronic post-infectious care.
