The study included 126 patients with muscle-invasive bladder cancer who underwent radical cystectomy with a Bricker ileal conduit, a urinary diversion technique in which a segment of intestine is used to create a stoma. Sixty-three patients were managed with a 22-element ERAS protocol adapted from ERAS Society guidelines, while 63 historical controls received standard perioperative care. The ERAS pathway combined eight preoperative elements, including patient education, nutritional optimization, carbohydrate loading, and shortened fasting, with seven intraoperative measures such as epidural analgesia, goal-directed fluid therapy, and active maintenance of body temperature, and seven postoperative elements including early feeding, early mobilization, and milestone-based discharge criteria. Many of these components were delivered or coordinated by nursing staff, making the protocol as much a nursing intervention as a surgical one.

To track immune changes, the team drew blood at three standardized timepoints: 24 hours before surgery, 24 hours after surgery, and on postoperative day seven or at discharge. Multiplex flow cytometry quantified cytotoxic CD8-positive T cells, regulatory T cells, and natural killer cells, while a 37-analyte Luminex panel measured inflammatory cytokines and ELISA assays captured soluble immune checkpoint molecules such as soluble PD-L1. The differences between the two groups were striking. In the ERAS cohort, CD8-positive T-cell counts fell by 22.8 percent from baseline in the first 24 hours after surgery, compared with a 41.6 percent decline in the conventional group, a difference that was highly statistically significant. Serum interleukin-6, a key inflammatory cytokine released during surgical stress, averaged 43.2 picograms per milliliter in ERAS patients versus 87.4 picograms per milliliter in conventional patients.

Beyond the cellular counts, the soluble immune landscape also differed. ERAS patients showed lower concentrations of soluble PD-L1 and soluble TIM-3, molecules shed into the bloodstream that are thought to dampen antitumor immunity, and higher levels of interferon-gamma, a cytokine central to cytotoxic immune defense. Natural killer cell activity was also better preserved, at 42.3 percent versus 31.7 percent. By postoperative day seven, several of these measures remained more favorable in the ERAS group, including higher CD8 counts and lower proportions of immunosuppressive regulatory T cells. The authors are careful to note, however, that these timepoint comparisons alone cannot establish a causal effect of ERAS on immune trajectories, since no formal group-by-time mixed-effects analysis was performed.

One of the most intriguing findings is an exploratory four-marker immune index measured 24 hours after surgery, combining CD8-positive T-cell count, interleukin-6, soluble PD-L1, and natural killer cell activity. When the researchers tested this composite score against the risk of infectious complications within 30 days of surgery, it achieved an area under the receiver operating characteristic curve of 0.84, with a 95 percent confidence interval of 0.76 to 0.92. That performance substantially exceeded a reference model based only on ASA score and clinical stage, which reached an AUC of 0.61. Individual markers performed less well, with interleukin-6 alone at 0.76 and CD8 count alone at 0.73, suggesting that the integration of multiple immune dimensions carries the predictive signal.

The team also examined the tumors themselves. Using multiplex immunofluorescence on surgical specimens, they quantified immune cell infiltration in tumor nests, invasive margins, and surrounding stroma. Specimens from the ERAS cohort showed higher densities of CD8-positive tumor-infiltrating lymphocytes in both tumor nests and invasive margins, lower densities of CD68-positive CD163-positive M2-polarized macrophages, and higher ratios of CD8-positive to FoxP3-positive cells, a measure that weighs cytotoxic against regulatory immune activity. Spatial transcriptomics performed on 24 specimens revealed higher expression of cytotoxic effector genes such as GZMB, PRF1, and IFNG, and lower expression of exhaustion markers including TIGIT, LAG3, and PDCD1 in the ERAS group. The immune-inflamed tumor phenotype, characterized by T cells penetrating deep into tumor tissue, was present in 52.4 percent of ERAS specimens versus 28.6 percent of conventional ones.

Here the authors issue an important caution. Because tumor tissue was collected at the time of cystectomy, before any postoperative ERAS component such as early feeding or mobilization could have taken effect, the tumor immune differences are cross-sectional observations rather than evidence that ERAS remodels the tumor microenvironment. They may reflect pre-existing tumor biology, differences in surgical approach, calendar-period effects, or other confounders. Indeed, the ERAS group had shorter operative times, less blood loss, lower transfusion rates, and more frequent minimally invasive surgery, all of which could independently influence both systemic immunity and tumor immune features.

Clinically, the ERAS group fared better on nearly every recovery measure. Time to first flatus was 2.1 days versus 3.4 days, first oral intake occurred at 8.2 hours versus 28.7 hours, and first mobilization at 12.4 hours versus 24.6 hours. Postoperative length of stay averaged 4.8 days compared with 7.2 days, resting pain scores on the first day after surgery were 3.2 versus 5.1 on a ten-point scale, and total hospitalization costs were 22.4 percent lower. The overall rate of 30-day infectious complications was numerically lower in the ERAS group at 19.0 percent versus 33.3 percent, but this difference did not reach conventional statistical significance after recalculation, with a P value of 0.068.

Survival analyses, while exploratory, added another layer. Among patients with available follow-up, descriptive three-year overall survival proportions were 78.1 percent in the ERAS cohort and 68.3 percent in the conventional cohort, though the authors stress these are not formal Kaplan-Meier estimates for the full cohort. In multivariable Cox models, ERAS management was associated with an overall survival hazard ratio of 0.58, and an eight-gene immune signature comprising cytotoxic effectors, exhaustion markers, FOXP3, CD163, and IL6 showed an even stronger association at a hazard ratio of 0.47. Notably, the survival benefit associated with ERAS appeared concentrated in patients whose tumors had high tumor mutational burden, with a hazard ratio of 0.41 in that subgroup and a nominal interaction P value of 0.048.

The authors are unequivocal that these biomarker and survival findings are hypothesis-generating rather than validated. The four-marker index and eight-gene signature were developed and tested in the same retrospective cohort, creating a real risk of overfitting, and only 33 infectious events were available for the predictive analysis. The single-center historical-control design means ERAS exposure is tightly correlated with calendar period, and secular improvements in surgical technique, infection control, and systemic therapy could account for some of the observed differences. Mean follow-up was approximately 24 months, limiting interpretation of three-year outcomes. What the study does establish is a detailed immunological portrait of what happens to the immune system after major cancer surgery under two different care philosophies, and a roadmap for the prospective, contemporaneous trials with longitudinal immune sampling and paired tumor biopsies that will be needed before any of these signals can guide clinical practice.

Subject of Research: Perioperative immunogenomic differences between ERAS-based and conventional nursing care in muscle-invasive bladder cancer patients undergoing radical cystectomy

Article Title: Perioperative immunogenomics and clinical outcomes under ERAS-based nursing care in muscle-invasive bladder cancer

Article References: Xin, L.-L., Guo, J.-J., Zheng, X.-Y., & Li, Z.-Y. (2026). Perioperative immunogenomics and clinical outcomes under ERAS-based nursing care in muscle-invasive bladder cancer. Frontiers in Genetics, 17, Article 1967034. https://doi.org/10.3389/fgene.2026.1967034

Image Credits: AI Generated

DOI: 10.3389/fgene.2026.1967034

Keywords: bladder cancer, ERAS, radical cystectomy, perioperative immunogenomics, CD8 T cells, tumor immune microenvironment, immune biomarkers, interleukin-6, tumor mutational burden, spatial transcriptomics, nursing care, infectious complications

The central problem the researchers identify is what they call phenotypic inflation: the systematic gap between how well a biomarker performs in a discovery dataset and how much genuine clinical predictive value it actually carries. When the team plotted phenotypic performance against clinical performance across their cohorts, they found a striking and reproducible pattern. Every 0.1-unit increase in the phenotypic area under the curve, a standard measure of classification accuracy, predicted a 0.079-unit larger drop in clinical performance, a relationship that explained roughly 37 percent of the variance and held with overwhelming statistical significance. Most remarkably, above a phenotypic AUC of approximately 0.55, a threshold that 95 percent of observations in their dataset exceeded, additional phenotypic performance conferred no additional clinical predictive value whatsoever.

This finding strikes at the heart of conventional biomarker discovery practice. If most published immune biomarkers already sit above the 0.55 phenotypic AUC threshold, then the endless pursuit of ever-higher classification accuracy within discovery datasets is, according to this analysis, largely a waste of effort. The uncoupled model, which removed the systematic relationship, explained only about 4 percent of variance, confirming that the inflation phenomenon is not statistical noise but a structured, quantitatively reproducible feature of bulk transcriptomic biomarker evaluation. The relationship replicated independently across ten external RNA-sequencing cohorts and held when the analysis was stratified by cancer type, where it explained more than half of the observed variance.

To understand why this inflation occurs, the authors turned to the biology of the tumor microenvironment. Bulk RNA sequencing measures the average gene expression across a mixture of tumor cells, immune cells, fibroblasts, and other stromal components. A gene that appears to be a strong marker of antitumor immunity in bulk data may in fact be expressed by a completely different cell type than assumed, or its apparent signal may be driven entirely by shifts in the proportions of cell populations rather than by genuine changes in per-cell expression. Either scenario can produce a biomarker that looks excellent in one cohort but collapses in another where the cellular composition differs.

The team’s first diagnostic instrument addresses this problem directly. The Source Fidelity Index quantifies how faithfully a gene’s expression can be attributed to T cells and natural killer cells, drawing on single-cell RNA-sequencing data from 2.2 million individual cells across 18 datasets. The results were sobering. Of 101 immune genes commonly invoked in biomarker studies, only 35, or 34.7 percent, showed high T/NK fidelity with an SFI of 0.6 or greater. In other words, roughly two-thirds of the genes that researchers routinely treat as proxies for cytotoxic immune activity cannot be reliably attributed to the immune cells they are assumed to represent. Well-known markers such as granzyme A, perforin 1, granulysin, and NKG7 fared better than many others, but the overall picture suggests that a substantial portion of the immune biomarker literature rests on shaky cellular attribution.

The second metric, the Biomarker Fragility Index, tackles the complementary problem of composition sensitivity. Using three-dimensional simulation across 82 synthetic cellular compositions for each gene, the framework measures how much a gene’s bulk expression signal changes when the mixture of cell types shifts. A fragile biomarker is one whose bulk-level behavior is dominated by composition rather than by biology, making it inherently unreliable across patient cohorts with different immune infiltration patterns. When the researchers tested whether fragility carried information beyond what spatial methods could capture, they found that BFI provided 18.6 to 27.3 percent independent explanatory variance for clinical outcomes even after controlling for a spatially derived source-fidelity estimate, with significance in four of five phenotype definitions. Notably, BFI also correlated independently with spatial myeloid co-localization, suggesting that the local positioning of myeloid cells relative to other populations contributes to how fragile a biomarker’s bulk signal becomes.

Combining these metrics with additional risk factors, the authors assembled a six-factor Operational Risk Stratification framework and validated it in 754 samples from eight immune checkpoint blockade cohorts. The composition-boundary rule, a simple decision criterion derived from the fragility analysis, achieved 82.8 percent accuracy in classifying whether a biomarker’s phenotypic behavior was reliable. The full six-factor framework recalled 93.3 percent of phenotypic failures, and this performance remained stable under leave-one-cohort-out validation, with mean recall of 95.0 percent across held-out cohorts. Perhaps most practically, BFI-guided selection reduced the risk of phenotypic validation failure by 82 percent compared with random biomarker selection, a difference that reached permutation-based significance at p equals 0.007. For research groups deciding which candidate genes to carry forward into costly validation programs, this represents a potentially enormous saving of time and resources.

The framework’s robustness was tested against a common criticism of such approaches: sensitivity to the statistical assumptions embedded in the analysis. The authors report that BFI’s classification performance was invariant across all tested distributional assumptions, meaning the tool does not quietly depend on a particular model of how gene expression data are distributed. To demonstrate generalizability, the team applied the framework pan-cancer across The Cancer Genome Atlas, identifying 168 gene-cancer combinations in the Unreliable tier spanning 32 cancer types. This catalog provides an immediate, actionable resource: any researcher considering one of these combinations as a biomarker in a given cancer type now has a quantitative warning that the marker’s bulk-level signal is likely to be composition-driven and fragile.

The clinical context of this work is the growing reliance on immune checkpoint blockade, the class of therapies that unleashes T cells against tumors and has transformed outcomes in melanoma, lung cancer, and other malignancies. Biomarkers such as PD-L1 expression, the cytolytic activity score, and gene signatures built from T cell markers are widely used to guide patient selection, yet their performance varies frustratingly across tumor types and treatment settings. The new study does not invalidate these tools, but it reframes the question researchers should ask. Rather than asking whether a biomarker achieves high accuracy in a discovery cohort, the framework argues, investigators should first ask whether the biomarker’s signal is genuinely attributable to the intended cell type and whether it can survive changes in cellular composition. A biomarker that passes those tests may justify validation; one that fails them is likely to join the long list of phenotypically impressive markers that never reach the clinic.

The implications extend beyond immunotherapy. Bulk transcriptomics remains the workhorse of large-scale biomarker discovery across medicine, from autoimmune disease to cardiology, and the composition-driven fragility mechanism the authors describe is not unique to cancer. Any bulk measurement drawn from heterogeneous tissue is vulnerable to the same inflation phenomenon, in which mixture effects manufacture apparent signal that dissolves under cohort variation. By providing open, quantitative diagnostics, the Source Fidelity Index and Biomarker Fragility Index offer a template for pre-validation screening that could be adapted to other fields. The authors, who note that generative artificial intelligence was used only for language refinement while all scientific contributions were human-generated, funded by Taiwan’s National Science and Technology Council, present the work as a practical addition to the biomarker development pipeline. If the framework’s predictions hold as widely as its validation suggests, the era of discovering biomarkers that look brilliant on paper and fail in patients may finally be drawing to a close, replaced by a more honest accounting of what bulk gene expression can and cannot tell us about the biology of disease.

Subject of Research: Computational diagnostics for predicting translation failure of bulk RNA-sequencing immune biomarkers in cancer immunotherapy

Article Title: Phenotypic Inflation and composition-driven fragility: a pre-validation diagnostic framework for immune biomarker translation failure

Article References: Phenotypic Inflation and composition-driven fragility: a pre-validation diagnostic framework for immune biomarker translation failure. (n.d.). https://doi.org/10.1186/s12859-026-06691-x

Image Credits: AI Generated

DOI: 10.1186/s12859-026-06691-x

Keywords: immune biomarkers, phenotypic inflation, biomarker fragility, bulk RNA-sequencing, tumor microenvironment, immune checkpoint blockade, single-cell RNA sequencing, biomarker translation, cancer immunotherapy, TCGA, transcriptomics, BMC Bioinformatics

The research team enrolled 206 pediatric patients diagnosed with mycoplasma pneumoniae pneumonia and compared their peripheral blood immune and inflammatory parameters across clinically defined subgroups, distinguishing macrolide-sensitive cases from severe disease. The comparisons revealed a clear inflammatory gradient. Children with severe mycoplasma pneumonia displayed significantly elevated white blood cell counts and markedly higher circulating levels of interleukin-6, a central pro-inflammatory cytokine that drives acute-phase responses and is often used as a marker of systemic inflammation. At the same time, these severe cases showed reduced proportions of monocytes, a shift that suggests alterations in innate immune cell dynamics as the disease escalates. Taken together, these findings confirm that severe pediatric mycoplasma pneumonia is accompanied by heightened systemic inflammation, providing clinicians with measurable correlates of disease severity that go beyond symptoms and imaging.

While the blood-based immune markers painted a picture of generalized inflammation, the heart of the study lay in its analysis of the T cell receptor beta chain, specifically the complementarity-determining region 3, or CDR3. This hypervariable loop sits at the very tip of the T cell receptor and makes direct contact with peptide antigens presented by major histocompatibility complex molecules. Because the CDR3 sequence is generated through a stochastic process of V, D and J gene segment recombination, joined by random nucleotide trimming and insertion at the junctions, each naive T cell carries a essentially unique CDR3 signature. When a pathogen invades, the handful of T cells whose receptors happen to recognize its antigens undergo clonal expansion, multiplying into large populations of identical sequences that can be detected by sequencing the repertoire. The composition of these expanded clones therefore serves as a molecular record of the antigenic encounters the immune system has experienced.

To capture this record, the researchers performed TCR beta CDR3 repertoire sequencing on peripheral blood samples from eight children with mycoplasma pneumoniae pneumonia and nine age-matched healthy controls. Although the sequencing cohort was modest in size, the bioinformatic analysis was comprehensive, examining repertoire clonality, diversity indices, V and J gene segment usage, CDR3 length distributions, recombination characteristics, and the presence of disease-associated high-frequency or shared clonotypes. The results showed that children with mycoplasma pneumonia had measurably reduced CDR3 diversity compared with healthy controls, a contraction of the overall repertoire that is consistent with a strong antigen-driven response in which a limited number of pathogen-specific clones expand and crowd out the diverse background of bystander T cells.

The clonal architecture itself was also transformed. In healthy individuals, the T cell pool is dominated by a vast number of small, rare clones, with only a sparse scattering of expanded sequences. In the mycoplasma pneumonia patients, the researchers observed a shift toward more medium and large clones, indicating that a substantial fraction of the circulating T cell population had been recruited into antigen-driven expansion. This oligoclonal skewing is a hallmark of active infection and mirrors patterns previously documented in viral infections and other inflammatory lung diseases. The degree of clonal expansion, quantified through measures of clonality, could in principle serve as a quantitative readout of the intensity of the adaptive immune response, potentially helping clinicians distinguish children mounting an effective response from those at risk of immunopathology.

Beyond clonality, the study uncovered striking differences in the genetic mechanics of receptor generation. The mycoplasma pneumonia group showed significant alterations in the usage frequencies of individual TRBV and TRBJ gene segments, the variable and joining elements that form the framework of the beta chain receptor. Certain V-J pairing combinations were overrepresented or underrepresented relative to controls, suggesting that the antigenic landscape presented by mycoplasma pneumoniae selects for receptors built from particular gene segments. The patients also exhibited increased CDR3 lengths, altered amino acid composition within the hypervariable loop, and enhanced V(D)J trimming and nucleotide insertion activity at the recombination junctions. These features imply that the receptors expanded during infection tend to be longer and more compositionally distinctive than average, a pattern that may reflect the structural requirements of recognizing mycoplasma-derived peptides or, alternatively, the recruitment of T cells activated by molecular mimicry.

Perhaps the most intriguing discovery emerged from the search for pathogen-associated clonotypes. The team identified several putative mycoplasma-associated CDR3 sequences that were either present at high frequency in patients or shared across multiple individuals, suggesting convergent recognition of common mycoplasma antigens. Remarkably, some of these expanded clones showed potential cross-reactivity with respiratory viruses, including influenza and Epstein-Barr virus. This observation carries significant immunological weight. Cross-reactive T cells, activated by one pathogen but capable of recognizing structurally similar epitopes from another, are increasingly appreciated as important players in heterologous immunity, the phenomenon by which prior infections shape responses to new ones. In the context of pediatric pneumonia, such cross-reactive expansions could influence disease severity, either by accelerating viral clearance or by contributing to the immune-mediated lung damage seen in severe mycoplasma cases.

The identification of candidate immune signatures also opens a practical avenue for clinical translation. If the disease-associated clonotypes and repertoire features documented here can be validated in larger cohorts, they could form the basis of diagnostic or prognostic tools. A sequencing-based assay that detects mycoplasma-associated T cell expansions might help confirm infection in cases where culture and serology are inconclusive, since mycoplasma is notoriously difficult to grow in the laboratory and antibody responses can be delayed or unreliable in young children. Repertoire clonality measures might additionally serve as biomarkers for monitoring host immune status over the course of treatment, flagging patients whose inflammatory trajectory is worsening before clinical deterioration becomes obvious. The authors suggest that these repertoire characteristics may reflect potential biomarkers for monitoring host immune status or infection-associated T cell responses in pediatric mycoplasma pneumonia.

Several caveats temper the immediate applicability of these findings. The sequencing arm of the study included only eight patients and nine controls, a sample size sufficient for generating hypotheses but insufficient for establishing robust clinical biomarkers. Longitudinal sampling, which would track how the repertoire evolves during infection and recovery, was not performed, and the cross-reactivity assignments to influenza and Epstein-Barr virus remain putative pending functional validation with antigen-specific stimulation assays. Future studies will need to expand patient numbers, incorporate follow-up time points, and pair repertoire data with antigen-specific T cell functional testing to confirm which clonotypes genuinely recognize mycoplasma antigens and which are bystanders or cross-reactive responders.

Nevertheless, the study represents a meaningful step toward bringing immune repertoire sequencing out of the research laboratory and into pediatric infectious disease medicine. By demonstrating that mycoplasma pneumoniae leaves a detectable and characteristic imprint on the TCR beta CDR3 landscape of infected children, the work establishes a framework for understanding the immunological heterogeneity that underlies the variable clinical course of this common infection. As sequencing costs continue to fall and analytical pipelines mature, the kind of repertoire profiling demonstrated here could eventually complement conventional laboratory tests, giving clinicians a dynamic view of the adaptive immune response as it unfolds. For a pathogen that has repeatedly proven difficult to detect and even more difficult to anticipate in terms of severity, the T cell receptor repertoire may prove to be one of the most informative windows yet into how the body fights back.

Subject of Research: T cell receptor repertoire changes in pediatric Mycoplasma pneumoniae pneumonia

Article Title: TCR β CDR3 repertoire profiling identifies pathogen-associated clonal expansions and candidate immune signatures in pediatric Mycoplasma pneumoniae pneumonia

Article References: Yuan, Y., Hui, L., Zhang, X., Liu, X., Yang, C., Xu, Y., Yao, X., Chen, Y., & Li, J. (2026). TCR β CDR3 repertoire profiling identifies pathogen-associated clonal expansions and candidate immune signatures in pediatric Mycoplasma pneumoniae pneumonia. Journal of Translational Medicine. https://doi.org/10.1186/s12967-026-08992-4

Image Credits: AI Generated

DOI: 10.1186/s12967-026-08992-4

Keywords: Mycoplasma pneumoniae, pediatric pneumonia, TCR repertoire, CDR3 diversity, high-throughput sequencing, T cell receptor, clonal expansion, interleukin-6, immune biomarkers, cross-reactivity, macrolide resistance, Journal of Translational Medicine

The trial enrolled thirty patients with relapsed or refractory small cell lung cancer who had already progressed after platinum-based chemotherapy, a population where effective options are scarce and median survival is typically measured in months. The treatment sequence was deliberately simple. Each patient received 24 Gy of hypofractionated radiotherapy delivered in four fractions every other day, a schedule designed to deposit substantial radiation dose in a compressed timeframe. Immediately afterward, patients began atezolizumab at 1200 mg every three weeks. The rationale rested on a growing body of preclinical and clinical evidence suggesting that radiation can do more than kill tumor cells directly. Hypofractionated doses can trigger immunogenic cell death, releasing tumor antigens and danger signals that recruit and activate immune cells, potentially converting an irradiated tumor into something resembling an in situ vaccine that checkpoint blockade can then amplify.

The clinical outcomes told a sobering story about the difficulty of treating this disease in an unselected population. The objective response rate was 80.0 percent for the radiotherapy component itself, confirming that the tumors were radiosensitive, as small cell lung cancer characteristically is. But when atezolizumab took over as the systemic agent, the response rate fell to just 16.7 percent. Median progression-free survival was 2.5 months, with only 6.9 percent of patients remaining progression-free at six months. Median overall survival reached 10.4 months, and 41.8 percent of patients were alive at one year, figures that are respectable for relapsed disease but fall short of a breakthrough. The gap between local radiosensitivity and systemic immunotherapy benefit is precisely the puzzle the trial’s exploratory immune profiling was designed to address.

That profiling began with serial blood draws. Peripheral blood mononuclear cells were collected before radiotherapy and again after its completion, then analyzed using multiparametric flow cytometry, a technique that tags cells with fluorescent antibodies and sorts them by the proteins they display on their surfaces. The investigators focused on two cell types with very different immunological roles. The first was the natural killer T cell, identified as cells carrying both CD3, a pan-T-cell marker, and CD56, a marker shared with natural killer cells. NKT cells sit at the interface of innate and adaptive immunity, capable of rapid cytokine release and of bridging radiation-induced inflammation to broader antitumor responses. The second was the classical monocyte, defined as CD14-high and CD16-negative, with particular attention to the fraction of these cells expressing programmed death-ligand 1, the same molecule that atezolizumab blocks.

The NKT cell findings were the trial’s first revelation. Patients whose disease remained controlled for three months or longer on progression-free survival started with significantly lower circulating NKT cell frequencies at baseline than patients who progressed sooner, a difference that reached statistical significance at p equals 0.02. More striking still was what happened after radiation. In the longer-survival group, NKT cell numbers expanded after radiotherapy, with a median fold-change of 1.3, while in the shorter-survival group they contracted, with a fold-change of 0.9. The difference between these trajectories was statistically significant at p equals 0.011. In other words, the direction of change, not simply the starting point, carried prognostic information. A radiation-induced surge in these innate-like lymphocytes in the bloodstream appeared to mark patients whose immune systems could be mobilized into a durable antitumor posture.

The monocyte data told a complementary story. Classical monocytes are the most abundant monocyte subset in human blood, and when they express PD-L1, they are often interpreted as soldiers of an immunosuppressive program, cells that dampen T-cell activity and can be recruited by tumors as shields against immune attack. Yet in this trial, a post-radiotherapy increase in PD-L1-expressing classical monocytes was associated with benefit. Patients with progression-free survival of three months or longer showed a median fold-increase of 2.8 in these cells after radiation, compared with only 0.7 in the shorter-survival group, a difference significant at p equals 0.0085. One plausible interpretation is that radiation drove a wave of inflammatory monocyte mobilization, and the PD-L1 expression on those cells represented a druggable vulnerability rather than an insurmountable barrier, a target that atezolizumab could engage precisely because the cells were present and ligand-loaded.

Read together, the two signatures suggest that radiotherapy does not simply damage tumors; it measurably reprograms the innate immune compartment in the periphery, and the pattern of that reprogramming predicts who will benefit from subsequent checkpoint blockade. This is a conceptually important shift. Most biomarker research in immunotherapy has concentrated on the tumor itself, examining PD-L1 staining on biopsy specimens, tumor mutational burden, or the density of lymphocyte infiltration. Small cell lung cancer frustrates these approaches because biopsies are difficult to obtain in relapsed patients and the tumor microenvironment is often profoundly immunosuppressed. A blood test that requires only serial peripheral blood draws, analyzed by standard flow cytometry, would be far more practical for a patient population that is often too fragile for repeat invasive procedures.

The trial’s limitations deserve honest framing. Thirty patients is a small cohort, the study was single-arm without a randomized control receiving atezolizumab alone, and the immune correlates were pre-specified as exploratory rather than confirmatory objectives. Correlates identified in a phase II setting must be validated in larger, ideally randomized, cohorts before they can guide clinical decisions. There is also a biological ambiguity to resolve: whether the NKT cell expansion and monocyte PD-L1 induction are causal drivers of benefit, passive readouts of a favorable radiation response, or both. Answering that question will require mechanistic studies that connect the peripheral blood observations to what is happening inside the irradiated tumor and in distant metastatic sites.

Even so, the study adds a meaningful piece to one of the most active puzzles in modern oncology: how to make radiation and immunotherapy work as true partners rather than sequential monotherapies. The abscopal effect, in which localized radiation shrinks tumors far from the treatment field, has tantalized researchers for decades but remains rare in practice, likely because it depends on a permissive immune context that most patients’ tumors and immune systems do not spontaneously provide. Biomarkers like the ones identified here, dynamic and measured in blood, could eventually allow clinicians to identify which patients carry that permissive biology before committing them to combination therapy, and to test interventions, such as altered radiation fractionation or additional immune modulators, in those who do not. For a disease as aggressive and under-served as relapsed small cell lung cancer, even a modest systemic signal paired with a usable predictive signature represents progress worth building on, and the Korean team’s finding that innate immune reprogramming can be tracked in a simple blood sample may prove to be the trial’s most enduring legacy.

Subject of Research: Immune correlates of radiotherapy-primed atezolizumab in relapsed small cell lung cancer

Article Title: Dynamics of peripheral blood NKT cells and PD-L1-expressing classical monocytes as immune correlates of radiotherapy-primed atezolizumab in relapsed small cell lung cancer: a phase II study

Article References: Suh, Y.-G., Lee, Y., Moon, S. H., Lee, S., & Han, J.-Y. (2026). Dynamics of peripheral blood NKT cells and PD-L1-expressing classical monocytes as immune correlates of radiotherapy-primed atezolizumab in relapsed small cell lung cancer: a phase II study. Cancer Immunology, Immunotherapy. https://doi.org/10.1007/s00262-026-04570-1

Image Credits: AI Generated

DOI: 10.1007/s00262-026-04570-1

Keywords: small cell lung cancer, atezolizumab, hypofractionated radiotherapy, NKT cells, classical monocytes, PD-L1, immune checkpoint inhibitor, immune biomarkers, phase II trial, radio-immunotherapy, flow cytometry, progression-free survival

The study, led by Indri Desiati and May-Britt Tessem of the Norwegian University of Science and Technology together with colleagues at Oslo University Hospital and St. Olavs Hospital, focused on 223 men drawn from two independent cohorts: 136 patients from Trondheim and 87 from Oslo. Each patient provided a serum sample before surgery, and the researchers measured the levels of proteins involved in inflammation and immuno-oncology using the Olink platform, a proximity extension assay technology capable of quantifying hundreds of proteins from tiny volumes of blood with high sensitivity and specificity. Thirty-four proteins overlapped between the inflammation and immuno-oncology panels and were analyzed across the combined cohort.

The clinical questions at the heart of the study are among the most consequential in prostate cancer management. Biochemical recurrence was defined as a PSA level of 0.2 nanograms per milliliter or higher following surgery, a signal that prostate tissue, or possibly residual or metastatic cancer cells, is once again producing the enzyme. PSA persistence, defined as a PSA level of 0.1 nanograms per milliliter or higher, indicates that PSA never fully fell after the operation, often hinting that cancerous tissue remained in the body. Both outcomes mark turning points at which patients and clinicians must decide whether additional treatment, such as radiotherapy or androgen deprivation therapy, is warranted.

Among the 34 immune proteins measured, one stood out with striking consistency. Interleukin 18, or IL18, a cytokine known to stimulate antitumor immune activity and promote the development of cytotoxic T lymphocytes and natural killer cells, was consistently lower in patients who later experienced recurrence or PSA persistence compared with those who remained disease-free. Reduced IL18 before surgery, the authors suggest, may reflect weakened antitumor immune surveillance already in place before the tumor was removed, leaving patients biologically vulnerable to regrowth.

The PSA persistence group told an even more distinctive immunological story. Patients whose PSA never dropped to undetectable levels showed lower levels of TNFSF12, also known as TWEAK, along with reduced CCL19, a chemokine that helps guide immune cells into lymphoid tissues, and lower CD244, a receptor involved in natural killer and T cell function. Perhaps most intriguingly, the researchers observed a broader trend toward reduced protein levels across the entire panel in the persistence group in both cohorts, a pattern consistent with what oncologists call an immune-desert profile, in which the tumor environment is largely devoid of active immune cell infiltration and activity.

To translate these observations into practical risk prediction, the team built protein-based classification models using partial least squares discriminant analysis with fivefold cross-validation and recursive feature elimination, then compared their performance against models built from standard clinical parameters including age, pre-surgical PSA, biopsy Gleason Grade Group, clinical tumor stage, and PI-RADS MRI scoring. The protein signatures outperformed the clinical models in both cohorts. In Trondheim, the protein model achieved an area under the receiver operating characteristic curve, or AUC, of 0.73 for recurrence compared with 0.64 for the clinical model, and 0.83 for PSA persistence compared with 0.81. In Oslo the gap was even wider: AUC values of 0.77 versus 0.54 for recurrence and 0.89 versus 0.63 for persistence.

When the researchers combined the immune protein data with clinical parameters, discrimination improved further still. The combined models reached an AUC of 0.75 for recurrence in Trondheim and 0.75 in Oslo, while for PSA persistence the combined models achieved AUC values of 0.91 in Trondheim and 0.88 in Oslo. These numbers, while requiring validation in larger independent cohorts, indicate that a simple pre-surgical blood test capturing immune signaling could meaningfully sharpen the risk picture that clinicians currently assemble from imaging, biopsy grading, and PSA measurements alone.

The technical rigor of the study strengthens confidence in its conclusions. Group comparisons were performed using one-way analysis of variance with false discovery rate correction to guard against spurious findings, and prognostic associations were evaluated with Cox proportional hazards models for recurrence and logistic regression for PSA persistence. Analyzing two geographically and administratively separate cohorts provided an internal replication of the key findings, particularly the consistent reduction of IL18 in patients destined for recurrence or persistence, and the immune-desert pattern in the persistence group.

Beyond its immediate predictive value, the study carries broader implications for the immunobiology of prostate cancer. Prostate cancer has traditionally been considered a relatively cold tumor, less responsive to the checkpoint inhibitor immunotherapies that have transformed the treatment of melanoma and lung cancer. The observation that specific, measurable immune states, one inflamed and one suppressed, are already detectable in peripheral blood before surgery suggests that systemic immune contexture plays a role in determining which tumors evade control. If confirmed, immune-proteomic profiling could help identify patients who might benefit from immunotherapeutic strategies or more aggressive adjuvant treatment, and could enrich clinical trials designed around immune biomarkers.

The authors emphasize that their findings support the potential of pre-surgical serum immune-proteomic profiling for risk stratification and treatment planning, pending further validation in larger independent cohorts. The work was funded by the Norwegian Cancer Society and the Central Norway Regional Health Authority, and the serum samples were provided through hospital biobanks with ethics approval and written informed consent from all participants. For the roughly 1.4 million men diagnosed with prostate cancer worldwide each year, the prospect of a routine blood test performed before surgery that reveals not only the tumor’s characteristics but the body’s immunological readiness to fight it represents a compelling step toward truly individualized cancer care. As proteomic technologies grow faster and cheaper, the immune fingerprint in a vial of blood may become as standard a part of the surgical workup as the MRI scan itself.

Subject of Research: Pre-surgical serum immune-proteomic biomarkers predicting recurrence after prostate cancer surgery

Article Title: Pre-surgical immune-proteomic profiles in serum identify inflamed and suppressed immune states associated with biochemical recurrence and PSA persistence in prostate cancer

Article References: Desiati, I., Bozorgpana, S., Ramberg, H., Berge, V., Tasken, K. A., Giskeødegård, G. F., Bathen, T. F., & Tessem, M.-B. (2026). Pre-surgical immune-proteomic profiles in serum identify inflamed and suppressed immune states associated with biochemical recurrence and PSA persistence in prostate cancer. Clinical Proteomics. https://doi.org/10.1186/s12014-026-09634-z

Image Credits: AI Generated

DOI: 10.1186/s12014-026-09634-z

Keywords: prostate cancer, biochemical recurrence, PSA persistence, immune biomarkers, serum proteomics, IL18, Olink, radical prostatectomy, risk stratification, immune-desert profile, TNFSF12, CCL19

The study, led by Yuanning Jia and Weiling Chen of Dongzhimen Hospital at Beijing University of Chinese Medicine, together with colleagues including Ye Tian, followed 250 adults with alopecia areata for up to 24 months. At enrollment, the researchers measured a panel of peripheral immune markers, including the cytokines interferon-gamma, interleukin-17, interleukin-10, and interleukin-4, alongside eosinophil counts and total immunoglobulin E. The team then tracked participants over a median follow-up of 18 months to see who would experience disease progression or relapse, the study’s primary outcome. The work was supported by the Young Teachers Project of Beijing University of Chinese Medicine and approved by the hospital’s ethics committee, with written informed consent obtained from all participants.

The results were striking. During follow-up, 90 patients, or 36 percent of the cohort, experienced progression or relapse of their disease. After adjusting for other variables in multivariable Cox regression models, the researchers found that each one-standard-deviation increase in baseline interferon-gamma was associated with a 58 percent higher risk of progression or relapse, with a hazard ratio of 1.58 and a confidence interval of 1.25 to 2.00, a highly significant association. Interleukin-17, a signature cytokine of the Th17 immune pathway, told a similar story: higher baseline levels corresponded to a 46 percent increase in risk, with a hazard ratio of 1.46. Eosinophil counts, a marker of innate immune and allergic-type activity, were also linked to worse outcomes, with each standard-deviation increase raising the risk by 33 percent.

Not all immune signals pointed in the same direction. Interleukin-10, an anti-inflammatory cytokine that helps restrain immune responses, was inversely associated with disease progression, with a hazard ratio of 0.79, meaning higher levels appeared protective. This finding aligns with a growing body of evidence that regulatory immune activity, including interleukin-10 production by blood B cells, correlates with a more favorable course in alopecia areata. Taken together, the data suggest that the balance between pro-inflammatory drivers such as interferon-gamma and interleukin-17 on one side, and anti-inflammatory regulators such as interleukin-10 on the other, may encode critical information about the future behavior of the disease in an individual patient.

Perhaps the most compelling findings came from the longitudinal analysis. The researchers observed that interferon-gamma and interleukin-17 levels rose over time in patients who went on to progress or relapse, while the same cytokines declined in those who remained stable. The group-by-time interactions were statistically significant, indicating that the divergent trajectories were unlikely to be chance findings. This dynamic pattern suggests that peripheral blood does not merely reflect a static snapshot of immune activity but tracks the evolving immunological battle occurring around the hair follicles, offering clinicians a potential window into disease activity before clinical hair loss becomes visible.

To translate these biological signals into clinical utility, the team built predictive models and evaluated them with rigorous statistical methods, including time-dependent area under the curve, Harrell’s C-index, calibration plots, and bootstrap validation. A clinical model based on standard patient characteristics alone achieved a 24-month AUC of 0.75, indicating moderate discrimination. When baseline interferon-gamma and interleukin-17 were added, the AUC rose significantly to 0.84, a statistically meaningful improvement with a p-value of 0.004. The combined model achieved an optimism-corrected C-index of 0.81 and showed acceptable calibration, meaning its predicted risks aligned reasonably well with observed outcomes across the cohort.

The mechanistic backdrop for these findings is well established. Alopecia areata has long been understood as a T-cell-mediated autoimmune disease in which the collapse of immune privilege around the hair follicle allows cytotoxic CD8-positive T cells and helper T cells to attack the follicle. Interferon-gamma, the hallmark cytokine of Th1 immunity, is considered a central driver of this attack, promoting chemokine signaling that recruits more immune cells into the follicle. Preclinical work in the C3H/HeJ mouse model has shown the functional relevance of interferon-gamma, and recent studies have demonstrated that Th1 effector CD4 T cells rely on interferon-gamma production to induce alopecia areata. Interleukin-17 and the broader IL-23 axis have also been implicated in lesional inflammation, although their precise role has been debated.

What sets the new study apart is its longitudinal design. Much of the existing evidence on peripheral immune profiles in alopecia areata comes from cross-sectional studies, which can show that patients with severe disease have different cytokine levels but cannot establish whether those differences precede and predict worsening. By measuring immune markers at baseline and following patients forward in time, the Beijing team directly addressed the question that matters most clinically: can blood immune profiles identify patients at risk before the disease progresses? Their affirmative answer, supported by rigorous survival analysis and model validation, moves the field closer to prognostic rather than merely descriptive immunology.

The clinical implications are considerable. JAK inhibitors such as tofacitinib and baricitinib have revolutionized alopecia areata treatment by damping interferon-gamma and related cytokine signaling, but treatment decisions are currently made largely on disease extent and duration rather than biological risk. If validated, a biomarker panel combining interferon-gamma, interleukin-17, eosinophil counts, and interleukin-10 could help clinicians identify high-risk patients early, justify more aggressive or earlier intervention, tailor monitoring frequency, and stratify patients in clinical trials. The finding that interleukin-17 adds prognostic value may also renew interest in IL-17-targeted therapies, which have shown mixed results in alopecia areata to date but could theoretically benefit the subset of patients with a strong Th17 signature.

The authors are appropriately cautious, noting that external validation in independent cohorts is required before such models enter routine practice. Questions remain about how these peripheral markers relate to intracellular and lesional immune activity, whether the findings generalize across ancestries, ages, and disease severities, and whether serial cytokine measurements could guide treatment response monitoring as well as prognosis. Nevertheless, this prospective cohort study represents a significant step toward personalized medicine in a disease that has long defied prediction. For the millions of people living with the unpredictable course of alopecia areata, the prospect that a routine blood draw could reveal what the future holds, and allow clinicians to act on that knowledge before hair is lost, marks an encouraging advance at the intersection of immunology, dermatology, and predictive analytics.

Subject of Research: Peripheral immune profiles as prognostic biomarkers for disease progression in alopecia areata

Article Title: Longitudinal cohort study on peripheral immune profiles and risk of disease progression in alopecia areata

Article References: Jia, Y., Chen, W., Zhao, Y., Long, Y., & Tian, Y. (2026). Longitudinal cohort study on peripheral immune profiles and risk of disease progression in alopecia areata. Archives of Dermatological Research, 318(1), Article 411. https://doi.org/10.1007/s00403-026-04865-4

Image Credits: AI Generated

DOI: 10.1007/s00403-026-04865-4

Keywords: alopecia areata, IFN-γ, IL-17, IL-10, eosinophils, immune biomarkers, disease progression, relapse, risk prediction, prospective cohort, cytokines, Longitudinal