Section 1.3: New Breakthroughs in Precision Immunotherapy

Sepsis management has evolved from pathogen-targeted strategies to dual targeting of pathogens and host immune dysregulation. Immune imbalance is a central driver of organ injury in sepsis, yet the extreme heterogeneity of immune responses has historically led to failed broad-spectrum immunomodulatory trials. Precision medicine approaches, particularly immune-phenotype stratification, have emerged as a viable pathway. The ImmunoSep trial represents a major milestone, demonstrating that biomarker-guided immunotherapy significantly improves organ function in sepsis.

I. ImmunoSep Trial: Overcoming Historical Trial Failures

Building on the PROVIDE trial, ImmunoSep is an international, multicenter, Phase Ib RCT that stratifies patients using serum ferritin and monocyte HLA-DR (mHLA-DR) expression:

  • MALS (Macrophage Activation-like Syndrome): Ferritin >4,420 ng/mL → Treated with IL-1 receptor antagonist (IL-1RA)

  • SIP (Sepsis-Induced Immunoparalysis): Ferritin ≤4,420 ng/mL & mHLA-DR <5,000 antibodies/cell → Treated with recombinant IFN-γ

Primary Endpoint: Proportion of patients achieving a ≥1.4-point reduction in SOFA score between days 2–9 post-randomization.
Results: 35.1% (46/131) of the immunotherapy group met the endpoint vs. 17.9% (26/145) in the placebo group (difference 17.2%, 95% CI 6.8–27.2%, P=0.002). Secondary endpoints (immune recovery, infection control) also favored targeted therapy. Safety profiles were comparable (serious adverse events: 88.5% vs. 89.0%). Notable subgroup effects included higher anemia rates with IL-1RA (48.0% vs. 17.4%) and higher bleeding events with IFN-γ (4.7% vs. 0%).

II. Clinical & Research Implications

  • Validates “Classify First, Treat Second”: Of 672 screened patients, 53% were excluded due to intermediate immune phenotypes, confirming that stratification enriches trial populations and reduces noise from immunologically ambiguous cases.

  • Identifies High-Benefit Subgroups: Post-hoc analysis revealed patients with Charlson Comorbidity Index (CCI) ≥5 or baseline SOFA ≥10 derived the greatest benefit. Higher endpoint achievement (CCI≥5: 37% vs. 9%, P=0.003; SOFA≥10: 37% vs. 16%, P=0.035) and significantly lower 28-day mortality (CCI≥5: 44% vs. 66%, P=0.02; SOFA≥10: 52% vs. 71%, P=0.045) were observed. These patients typically exhibit severe organ injury and profound immune dysregulation, making them ideal candidates for targeted immunomodulation.

  • Highlights Immune Suppression Prevalence: Among 276 analyzed patients, SIP phenotype predominated (228/276, 83%) vs. MALS (48/276, 17%). Over half of MALS patients also exhibited mHLA-DR <5,000, underscoring the clinical urgency of addressing sepsis-induced immunoparalysis. mHLA-DR remains a key monitoring metric; where flow cytometry is unavailable, peripheral lymphocyte counts serve as a practical surrogate.

  • Defines Immune Recovery Metrics:

    • MALS: Ferritin reduction ≥15% from baseline

    • SIP: mHLA-DR increase to >8,000 antibodies/cell
      Immune recovery was achieved in 78.0% of treated patients vs. 48.5% in placebo, further validating the therapeutic trajectory.

The ImmunoSep trial successfully transitions sepsis immunotherapy from a “one-size-fits-all” approach to a precision paradigm, offering a reproducible assess-treat-monitor framework for future clinical implementation and trial design.

III. Limitations of the ImmunoSep Trial

Despite the significant breakthroughs achieved by the ImmunoSep trial, its potential limitations warrant careful consideration. First, the study population was restricted to patients with sepsis originating from pulmonary infections (community-acquired pneumonia, hospital-acquired pneumonia, and ventilator-associated pneumonia) and primary bacteremia, which limits the generalizability of the findings to more heterogeneous sepsis cohorts. The investigators noted that in patients with abdominal-source sepsis, source control is frequently delayed; if the primary infectious focus is not promptly eradicated, immunotherapy alone is unlikely to effectively improve prognosis. Conversely, patients with urinary tract infection–associated sepsis typically present with infections that are easier to control and generally exhibit better prognoses compared to other sepsis phenotypes. Consequently, both of these sepsis cohorts were excluded from the study. While this design aimed to minimize confounding variables affecting the primary endpoint, from the perspective of immunoinjury mechanisms, other infection sites retain significant research value.

Second, the study defined a maximum intervention window of 72 hours from disease onset. Given the complexity and heterogeneity of the immune response in sepsis, utilizing a straightforward immune biomarker to determine the timing of immunotherapy initiation is a strength of this trial. However, this fixed time window overlooks the dynamic trajectory of the immune response in sepsis. Patient populations clinically recognized as having immune dysfunction—such as those with secondary infections or opportunistic pathogen infections—may be excluded simply because they present in the later stages of sepsis. Consequently, immune biomarker stratification based on a narrow time window may constrain the broader application of precision therapy, necessitating further validation in larger, more diverse cohorts.

Furthermore, the trial only defined two mutually opposing immune phenotypes, potentially overlooking a subset of patients who exhibit concurrent hyperinflammation and immunosuppression. This unique population may not derive benefit from monotherapy. Data from the study indicate that the number of patients with serum ferritin >4420 ng/mL and mHLA-DR <5000 antibodies/cell actually exceeded the number of patients with isolated serum ferritin >4420 ng/mL. For this subgroup, targeting a single aspect of immune dysregulation may improve organ dysfunction; however, preliminary results from early-phase clinical trials suggest that combination therapy simultaneously targeting hyperinflammatory responses and immunosuppression may yield greater clinical benefits.

IV. Future Directions for Precision Immunotherapy

Two large-scale clinical trials published in 2025 have advanced precision immunotherapy from complementary perspectives. The TESTS trial emphasized “clinical characteristic stratification + immunomodulatory therapy,” revealing the core value of individualized treatment through subgroup analyses. In contrast, the ImmunoSep trial focused on “immune phenotype stratification + targeted therapy,” achieving a major breakthrough in precision treatment via immune phenotyping. Collectively, the results of both trials demonstrate that “precisely identifying the beneficiary population” is central to precision immunotherapy in sepsis. Patients with sepsis comorbid with chronic diseases represent a potential beneficiary group, and future confirmatory clinical trials should prioritize these populations.

In summary, the successive publication of the TESTS and ImmunoSep trials marks the formal transition of immunotherapy from traditional paradigms to the precision era. Only through continuous breakthroughs in immune typing and stratification, and the precise identification of immune characteristics specific to each subtype, can the therapeutic efficacy of immunotherapy be maximized. With the synergistic advancement of biomarker technologies, artificial intelligence, and novel immunomodulators, the goal of individualized sepsis management will gradually be realized, offering new hope to tens of millions of sepsis patients worldwide.

Traditional clinical classifications and single biomarkers are insufficient to comprehensively reflect the complex immune states of sepsis patients or effectively guide immunotherapy. With the development of artificial intelligence and multi-omics technologies, the current understanding of sepsis heterogeneity is gradually shifting from clinical phenotypes to biologically defined subtypes. Advanced tools and rich datasets hold promise for driving new breakthroughs in precision immunotherapy for sepsis. However, omics-based immune typing faces limitations such as high costs, delayed detection turnaround times, and limited clinical accessibility. Furthermore, heterogeneity in study populations and methodologies across related research hinders the clinical translation of omics-derived immune subtypes. Future studies will further mine multidimensional data to promote the clinical translation of mechanism-driven immune typing. In this context, the Subtyping in Sepsis and Critical Illness (SUBSPACE) consortium was established as an international alliance aimed at sharing omics data, identifying and elucidating underlying disease pathways, and advancing precision medicine in sepsis and critical care based on molecular mechanisms rather than clinical manifestations alone. Concurrently, Chinese researchers have established the Chinese Multi-omics Advances in Sepsis (CMAISE) database, which integrates multimodal omics data (genomics, transcriptomics, proteomics, metabolomics) alongside clinical phenotypes and imaging information. Moving forward, precision treatment for sepsis will undoubtedly continue to advance along the trajectory of “precise stratification and targeted intervention.”
(Pei Fei, Wu Jianfeng; First Affiliated Hospital, Sun Yat-sen University)