Sepsis-induced immunoparalysis is now understood as a moving target, not a fixed phase
Sepsis no longer fits neatly into the old sequence of early hyperinflammation followed by a later immunosuppressive tail. In this review, sepsis-induced immunoparalysis emerges as a dynamic biological state that can coexist with inflammation from the start, then deepen, improve, or persist across the first 1 to 2 weeks after onset. The practical consequence is stark: a single immune measurement can miss the period when host defense is most impaired and treatment windows may still be open.
The clinical stakes are high. Immunoparalysis is tied to secondary infections, prolonged organ dysfunction, longer ICU stays, and higher mortality. It is not merely a laboratory phenotype; it is part of the mechanism by which some patients fail to recover despite control of the initial infectious trigger.
Immune failure in sepsis follows several trajectories
Longitudinal studies summarized in this review point to heterogeneous immune recovery patterns rather than a single uniform course. Some patients demonstrate improving or recovering immune states, while others remain hyporesponsive. A fluctuating pattern has also been proposed, though it remains hypothesis-generating and still needs longitudinal validation.
The first 1 to 2 weeks after sepsis onset appear especially important. During this interval, mHLA-DR expression, absolute lymphocyte count, and ex vivo cytokine production capacity can move in different directions, and they do not always recover at the same pace. A patient may show partial improvement in one marker while remaining profoundly suppressed in another. In bedside terms, that means the immune system can look “better” in one compartment while still failing in another.
Persistently low immune function tracks with worse outcomes. Sustained suppression of mHLA-DR or prolonged lymphopenia is associated with increased mortality, while recovery of these measures tends to align with better clinical trajectories. The review makes a strong case that immune status in sepsis should be interpreted as a series of repeated observations rather than a single snapshot.
Monocyte HLA-DR remains the most established marker of innate immune deactivation
Among available biomarkers, monocyte HLA-DR has the deepest clinical footing. Reduced expression marks impaired antigen presentation and reflects monocyte deactivation, one of the central features of innate immune hyporesponsiveness in sepsis. The review notes that healthy reference ranges measured with Quantibrite-based assays have been reported at roughly 13,500 to 40,000 antibodies per cell, although values shift with assay method and study population.
In critically ill adults with sepsis or septic shock, an mHLA-DR below 15,000 AB/C has been linked to greater disease severity, while some intensive care immunophenotyping protocols have used values below 8,000 AB/C to identify severe immune dysfunction. Even so, these cutoffs are not interchangeable across platforms. Flow cytometry methodology, assay calibration, and the timing of measurement all influence interpretation.
A single threshold is therefore too blunt for real-world use. mHLA-DR is best used as a marker of immune phenotype and prognosis, not as a stand-alone predictor of response to a specific immunostimulatory drug. A low value on day 1 does not carry the same meaning as persistent failure to recover by days 3 to 7, when ongoing immune paralysis becomes easier to recognize.
Lymphocyte loss and checkpoint signaling deepen the immunosuppressive state
Innate dysfunction is only part of the picture. The review also places adaptive immune dysfunction at the center of sepsis-induced immunoparalysis, with persistent lymphopenia and impaired cytokine production capacity as major features. Lymphocyte depletion narrows the host’s ability to mount coordinated pathogen-specific responses, and the loss is not simply numerical. Functional exhaustion follows numerical loss.
Immune checkpoint pathways add another layer. Upregulation of PD-1 and PD-L1 is discussed as part of the negative regulatory network that restrains immune activation. In sepsis, this braking system can become overactive, blunting the response to pathogens and secondary immune stimuli. The biological logic is straightforward: when inhibitory signaling dominates, antigen presentation weakens, T-cell activity falls, and the host becomes less able to clear infection or recover from a second hit.
Metabolic and epigenetic reprogramming also contribute. These changes help explain why immune paralysis can persist even after the initial inflammatory burst subsides. Cells may remain functionally altered long after the triggering infection begins to resolve, which is one reason late sepsis can still behave like an active immunologic disorder rather than a simple convalescent phase.
Precision monitoring will depend on repeated multi-marker assessment
The review argues for a shift from static single-time-point testing to repeated, multi-marker immune monitoring. That shift is clinically important because no single biomarker fully captures the immune state of a septic patient. mHLA-DR gives information about antigen presentation, absolute lymphocyte count reflects broad adaptive immune depletion, ex vivo cytokine production offers a functional readout, and checkpoint markers such as PD-1 and PD-L1 provide insight into inhibitory signaling.
Used together, these measures may improve characterization of immune trajectories and help identify candidate treatment-eligible windows. Used alone, each can mislead. A patient with low mHLA-DR may be recovering, while another with a similar single value may be entering a persistent hyporesponsive state. The difference lies in trajectory, not just level.
The review is careful on this point: most current biomarkers are better supported for immune phenotyping and prognosis than for predicting treatment response. Reported thresholds depend on assay methodology, patient population, and measurement timing. The field still lacks a universally validated composite immune-clinical framework that can be carried into routine practice without distortion.
Biomarker-guided immunotherapy remains early proof of concept
The review discusses immunostimulatory therapy as a biologically plausible response to sepsis-induced immune paralysis. GM-CSF and IFN-γ have shown the ability to partially restore immune function in selected patients, supporting the idea that immune suppression in sepsis is not always irreversible. The implication is important: if the host-defense defect can be identified, it may also be modifiable.
Yet the evidence remains limited. The ImmunoSep trial is presented as early proof of concept for phenotype-enriched immunotherapy rather than definitive evidence for routine implementation. Its value lies in showing that immune stratification may help match therapy to biology. It does not settle the larger question of whether biomarker-guided treatment improves mortality, safety, or other patient-important outcomes across broad ICU populations.
The treatment problem is no longer whether sepsis can include immunoparalysis. The harder question is which patients have it, when it becomes persistent, and which immune intervention should follow a given biomarker pattern.
What still needs prospective validation
Several uncertainties remain. Biomarker-based treatment thresholds are not yet standardized across platforms. Composite immune-clinical models still need prospective testing. Phenotype-matched immunotherapies require stronger evidence on mortality, safety, and durability of benefit before they can be considered routine care.
The review leaves no doubt that repeated immune assessment may sharpen bedside understanding of sepsis biology. What it does not yet provide is a universal rule for action. The next step will be prospective studies that connect immune trajectories to concrete treatment decisions and patient-important outcomes, rather than to laboratory normalization alone.