Salivary microbial signature highlighting actinomyces as a predictor of immune-checkpoint inhibitor monotherapy response in advanced non–small cell lung cancer Silvia Cavaliere, Marta Fogolari, Michele Iuliani, Simone Foderaro, Alessio Cortellini, et al. Journal of Translational Medicine, 2026 Background Immune checkpoint inhibitors (ICIs) have improved survival in advanced non-small cell lung cancer (NSCLC), yet reliable biomarkers beyond programmed death-ligand 1 (PD-L1) expression remain limited. Increasing evidence links the gut microbiome to ICI activity, but the predictive value of the salivary microbiome is poorly defined. Methods We prospectively analyzed baseline saliva from 71 stage IV NSCLC patients treated with anti–PD-1/PD-L1 (ICI) monotherapy. After quality control, 70 samples underwent 16 S rRNA gene sequencing of the V1–V3 region. Microbial diversity, differential abundance (LEfSe, Mann-Whitney/Kruskal-Wallis with false discovery rate correction) and survival associations (Kaplan-Meier; Cox proportional-hazards with LASSO-based variable selection and 1000-fold bootstrap validation) were examined. In this cohort, an exploratory genus-level cut-off was derived by receiver operating characteristic (ROC) analysis. Results α-diversity and β-diversity did not differ between responders (progression-free survival (PFS) ≥ 12 months; n = 18) and non-responders ( n = 52). Differential‑abundance profiling revealed a graded enrichment of the phylum Actinobacteria across all lower ranks, class Actinobacteria, order Actinomycetales, family Actinomycetaceae and genus Actinomyces ,in non‑responders (LEfSe LDA > 3.5; p = 0.001 for each level; FDR ≤ 0.049). ROC analysis suggested an Actinomyces relative abundance of 11% (AUC = 0.768; sensitivity 0.94; specificity 0.44) as a data-driven threshold, classifying patients into low (≤ 11%, n = 46) and high (> 11%, n = 24) groups. High abundance was associated with shorter PFS (median 3 vs. 4 months; HR = 2.16, 95% CI 1.21–3.88, p = 0.009) and overall survival (OS) (median 5 vs. 9 months; HR = 2.61, 95% CI 1.48–4.61, p < 0.001) after multivariable adjustment for ECOG status, treatment line, corticosteroid and opioid use, smoking, histology and metastatic sites. Bootstrap validation supported model stability, with median bootstrap HRs of 2.56 (PFS) and 2.63 (OS), with narrow percentile CIs (PFS 1.57–4.49; OS 1.40–6.34) overlapping the original estimates. Conclusions In this exploratory cohort, salivary microbiome signature characterized by high Actinomyces abundance was independently associated with poorer ICI outcomes in NSCLC. Saliva profiling is non-invasive and, if validated in larger and independent cohorts, may complement tumour PD-L1 and clinical factors to refine patient stratification.
Pancreatic Steatosis as a Risk Phenotype for Pancreatic Ductal Adenocarcinoma: A Narrative Review Roberto Cammarata, Vincenzo La Vaccara, Lucrezia Bani, Federica Giordano, Pierpaolo Castagliuolo, et al. Medicina Lithuania, 2026 Background and Objectives: Pancreatic ductal adenocarcinoma (PDAC) is one of the leading causes of cancer-related mortality, largely due to late-stage diagnosis and the absence of effective population-based screening. Intrapancreatic fat deposition (IPFD) has emerged as a potential risk phenotype. This narrative review critically appraises the clinical, metabolic, epidemiologic, and mechanistic evidence linking IPFD to PDAC and discusses its implications for risk stratification and prevention. Materials and Methods: A structured literature search was conducted in PubMed/MEDLINE and Scopus for studies published between 2007 and 2025 using predefined terms related to pancreatic steatosis and pancreatic cancer. After duplicate removal and screening according to predefined inclusion and exclusion criteria, 42 articles were included. Evidence was synthesized focusing on epidemiologic associations, mechanistic pathways, and imaging-based quantification methods. Results: A strong association between IPFD and PDAC was found. Although definitive causality remains unproven, some studies support temporal correlation between IPFD and PDAC, suggesting that IPFD precedes PDAC. A possible pathophysiological explanation to this correlation has been advanced in experimental models indicating IPFD as a pro-inflammatory factor cooperating with oncogenic KRAS to facilitate neoplastic progression. Finally, variability in IPFD definitions and heterogeneity in imaging assessment limit interpretability. Conclusions: Current evidence links IPFD to PDAC risk, suggesting a strong suspicion that pancreatic steatosis may represent an independent risk factor for PDAC. Still robust causal inference remains unproven. Well-designed prospective studies, standardized imaging protocols, and mechanistic investigations are required to clarify causality and determine whether pancreatic steatosis can be incorporated into risk-based screening and preventive strategies.
Machine Learning Models for Sepsis: From Early Detection to Short- and Long-Term Prognosis Maria Vittoria Ristori, Filippo Ruffini, Silvia Spoto, Roberto Cammarata, Vincenzo La Vaccara, et al. International Journal of Molecular Sciences, 2026 Sepsis is a leading cause of morbidity and mortality worldwide, and its outcomes depend on early recognition and timely intervention. Conventional clinical scores and biomarkers provide prognostic value but often lack accuracy for individualized prediction. Machine learning (ML) offers the ability to integrate multidimensional data to improve risk stratification. We analyzed 477 patients admitted to our hospital, including 251 with sepsis, 100 with septic shock, and 126 controls. Demographic, clinical, and laboratory data were collected. Univariate correlation analyses explored associations with sepsis severity and mortality (in-hospital, 30-day, and 90-day). Several ML models were tested, with performance assessed by area under the receiver operating characteristic curve (AUC-ROC) and Matthews’s correlation coefficient (MCC). Model interpretability was evaluated using SHAP (SHapley Additive exPlanations). Sepsis severity and mortality correlated with biomarkers (procalcitonin, mid-regional pro-adrenomedullin, lactate) and clinical scores (SOFA, qSOFA). In-hospital mortality was associated with ADM, catecholamine use, and SOFA, while 90-day mortality involved smoking and Gram-negative or polymicrobial infections. Different machine learning models were evaluated, and the model achieving the highest performance on the validation set was selected. The selected model either outperformed or demonstrated comparable performance to logistic regression, depending on the specific prediction task (AUC 0.99 for sepsis, 0.96 for septic shock, 0.70 for ICU admission; 0.90, 0.72, and 0.87 for in-hospital, 30-day, and 90-day mortality). SHAP confirmed the clinical relevance of these predictors. ML models integrating clinical and biochemical data outperform conventional methods in predicting sepsis progression and mortality, while maintaining interpretability. These findings support the use of ML-based tools for early diagnosis and personalized risk stratification in sepsis, though external validation is required before clinical application.
Genomic Surveillance and Resistance Profiling of Multidrug-Resistant Acinetobacter baumannii Clinical Isolates: Clonal Diversity and Virulence Insights Maria Vittoria Ristori, Ilaria Pirona, Lucia De Florio, Sara Elsa Aita, Gabriele Macari, et al. Microorganisms, 2025 Acinetobacter baumannii is a multidrug-resistant opportunistic pathogen that poses critical challenges in hospital settings due to its environmental resilience and high resistance to antibiotics. Genomic surveillance has become essential for identifying transmission patterns, guiding antimicrobial stewardship, and informing infection control policies. We conducted whole-genome sequencing on 44 A. baumannii isolates collected between 2022 and 2023 from diverse wards in an Italian hospital. Illumina-based sequencing was followed by a comprehensive bioinformatics pipeline, including genome assembly, taxonomic validation, MLST, SNP-based phylogeny, pan-genome analysis, antimicrobial resistance (AMR) gene profiling, and virulence factor prediction. Most isolates were classified as ST2; SAMPLE-34 was ST1 and genetically distinct. Phylogenetic analysis revealed four clonal clusters with cluster-specific AMR and accessory gene content. The pan-genome included 5050 genes, with notable variation linked to hospital ward origin. ICU and internal medicine strains carried higher loads of AMR genes, especially against aminoglycosides, β-lactams, and quinolones. Virulence profiling highlighted widespread immune evasion mechanisms; “Acenovactin” was predominant, while some isolates lacked key adhesion or toxin factors. Our findings underscore the clinical relevance of integrating genomic epidemiology into routine hospital surveillance. Identifying clonal clusters and resistance signatures supports real-time outbreak detection, risk stratification, and targeted infection prevention strategies.
Investigating the evolutionary dynamics and mutational pattern of SARS-CoV-2 spike gene on selected SARS-CoV-2 variants Bachir Balech, Alessandra Lo Presti, Claudia Telegrafo, Lucia Maisto, Emanuela Giombini, et al. Plos One, 2025 The continuous evolution of SARS-CoV-2 has led to the emergence of several variants representing significant challenges for public health. Many studies highlight the relevance of phylogenetic inference or mutational pattern analysis to understand the evolutionary relatedness of viral variants and to estimate the potential effect of new mutations on viral transmission, virulence and antigenicity. Here we describe an evolutionary investigation approach combined with mutational analyses of SARS-CoV-2 Spike gene to annotate and potentially track important amino acid site variation of specific functional domain relevant for viral survival. This approach was applied on XBB*, EG* and BA* and their sub-lineages (see materials and methods) available from GISAID. In addition, we considered the major variants of concern (Alpha, Delta, Omicron) and Wuhan-Hu-1 strain as references. Maximum likelihood phylogenetic tree was constructed from the complete dataset while selection pressure and mutational analyses were conducted on single variants separately. The obtained phylogenetic tree of Spike amino acid gene sequence showed a clear separation of viral variants as well as their expected appearance order. This result supported the significance of selection pressure analyses outcomes combined with amino acid mutational frequencies where in many cases they showed a linear and parallel trend. This allowed also to hypothesize the potential importance of low-frequency mutations in new potential virus variants. This study constitutes an asset of important insights to be considered in regular monitoring programs. In addition, the analysis framework described here introduces a starting point for further standardization, optimization and application on different data types and in large-scale studies.
Proteomic Insights into Bacterial Responses to Antibiotics: A Narrative Review Sara Elsa Aita, Maria Vittoria Ristori, Antonio Cristiano, Tiziana Marfoli, Marina De Cesaris, et al. International Journal of Molecular Sciences, 2025 Antimicrobial resistance is an escalating global threat that undermines the efficacy of modern antibiotics and places a substantial economic burden on healthcare systems—costing Europe alone over EUR 11.7 billion each year due to rising medical expenses and productivity losses. While genomics and transcriptomics have significantly advanced our understanding of the genetic foundations of resistance, they often fail to capture the dynamic, real-time adaptations that enable bacterial survival. Proteomics, particularly mass spectrometry-based strategies, bridges this gap by uncovering the functional protein-level changes that drive resistance, persistence, and tolerance under antibiotic pressure. In this review, we examine how proteomic approaches provide new insights into resistance mechanisms across various antibiotic classes, with a particular focus on β-lactams, aminoglycosides, and fluoroquinolones, highlighting clinically relevant pathogens, especially members of the ESKAPE group. Finally, we examine future directions, including the integration of proteomics with other omic technologies and the growing role of artificial intelligence in resistance prediction, paving the way for more predictive, personalized, and effective solutions to combat antimicrobial resistance.
The Use of Self-Sampling Devices via a Smartphone Application to Encourage Participation in Cervical Cancer Screening: A Pilot Study Francesco Plotti, Fernando Ficarola, Giuseppina Fais, Carlo De Cicco Nardone, Roberto Montera, et al. Journal of Clinical Medicine, 2025 Background: Cervical cancer ranks among the most prevalent tumors in low-income countries, with the Pap test as one of the primary screening tools. The Pap smear detects abnormal cells, the CLART test identifies specific HPV genotypes, and HPV self-sampling allows for self-collected HPV testing. This study aimed to evaluate the feasibility of the first smartphone-based health device for home-collection HPV testing. Methods: Enrolled patients during the gynecological examination underwent three different samplings: Pap smear, HPV DNA genotyping test CLART, and vaginal HPV-Selfy swab. Each patient received a kit including an activation code, vaginal swab, and instructions. After performing the self-sample, patients returned the kit to our laboratory. Both the samples collected by the gynecologist and those collected by the patients themselves were analyzed. Results: A total of 277 patients were enrolled, with 226 self-collected swabs received for analysis. The assay yielded valid results for both self-collected and clinician-collected swabs in 190 patients. When comparing these results with paired clinician-taken vaginal swabs, we observed an agreement of 95.2% (Cohen’s Kappa: 0.845). We report an agreement of 93.7% (Cohen’s Kappa: 0.798). Conclusions: The study demonstrated the feasibility of HPV-Selfy as a complementary tool in cervical cancer screening, especially where adherence to traditional surveillance is low.
An Antibiotic Stewardship Program in Pancreatic Surgery Matteo De Pastena, Salvatore Paiella, Erica Secchettin, Damiano Caputo, Luca Moraldi, et al. JAMA Network Open, 2025 ImportanceAntimicrobial stewardship (AMS) programs optimize antibiotic use and mitigate antimicrobial resistance. The literature on the efficacy of AMS programs in pancreatic surgery is limited.ObjectiveTo investigate the association of a multifaceted AMS intervention targeting surgical antibiotic prophylaxis (SAP) with the rate of surgical site infections (SSIs) following pancreatic surgery.Design, Setting, and ParticipantsThis cross-sectional study was a multicenter, before-and-after analysis conducted at 3 Italian centers. The intervention cohort included adult patients aged 18 years or older who underwent pancreatectomy between January 1, 2020, and December 31, 2022, while the historical cohort included patients from January 1, 2015, to December 31, 2019.ExposureA multiprofessional, multidimensional ASM program that included a bundle of interventions and pivoted on preoperative rectal screening for multidrug-resistant bacteria and targeted SAP.Main Outcomes and MeasuresThe primary outcomes were SSI incidence and SAP appropriateness, assessed through the coverage rate of rectal and biliary isolates. Data were analyzed using propensity score weighting. Secondary outcomes evaluated were other postoperative outcomes (eg, pancreatic fistula rate, length of stay), antibiotic use, and costs.ResultsA total of 3387 patients (median [IQR] age, 66 [66-73] years; 1788 male [52.8%]) were included, with 1219 in the intervention cohort and 2168 in the historical cohort. After implementing the AMS program, a statistically significant reduction was found in rates of overall (30.1% vs 20.6%), superficial (5.8% vs 2.5%), deep (0.9% vs 0.3%), and organ-space (26.3% vs 19.3%) SSIs. After propensity score weighting, the odds ratios for the estimated mean treatment effect were 0.92 (95% CI, 0.89-0.96) for overall, 0.85 (95% CI, 0.78-0.93) for superficial, and 0.95 (95% CI, 0.92-0.99) for organ-space SSIs. Surgical antibiotic prophylaxis coverage increased significantly for rectal screening (87.2% vs 100%) and biliary bacterial colonization (59.7% vs 68.7%). Complications, infections, length of stay, and antibiotic consumption also decreased, with an overall cost savings of 247 460 euros.Conclusions and RelevanceThese findings suggest that a multifaceted, pancreatic surgery–specific AMS program is associated with decreased rates of SSIs, increased coverage of isolated bacteria, improved clinical outcomes, more judicious antibiotic use, and lower costs.
Machine Learning for Predicting the Low Risk of Postoperative Pancreatic Fistula After Pancreaticoduodenectomy: Toward a Dynamic and Personalized Postoperative Management Strategy Roberto Cammarata, Filippo Ruffini, Alberto Catamerò, Gennaro Melone, Gianluca Costa, et al. Cancers, 2025 Background. Postoperative pancreatic fistula (POPF) remains one of the most relevant complications following pancreaticoduodenectomy (PD), significantly impacting short-term outcomes and delaying adjuvant therapies. Current predictive models offer limited accuracy, often failing to incorporate early postoperative data. This retrospective study aimed to develop and validate machine learning (ML) models to predict the absence and severity of POPF using clinical, surgical, and early postoperative variables. Methods. Data from 216 patients undergoing PD were analyzed. A total of twenty-four machine learning (ML) algorithms were systematically evaluated using the Matthews Correlation Coefficient (MCC) and AUC-ROC metrics. Among these, the GradientBoostingClassifier consistently outperformed all other models, demonstrating the best predictive performance, particularly in identifying patients at low risk of postoperative pancreatic fistula (POPF) during the early postoperative period. To enhance transparency and interpretability, a SHAP (SHapley Additive exPlanations) analysis was applied, highlighting the key role of early postoperative biomarkers in the model predictions. Results. The performance of the GradientBoostingClassifier was also directly compared to that of a traditional logistic regression model, confirming the superior predictive performance over conventional approaches. This study demonstrates that ML can effectively stratify POPF risk, potentially supporting early drain removal and optimizing postoperative management. Conclusions. While the model showed promising performance in a single-center cohort, external validation across different surgical settings will be essential to confirm its generalizability and clinical utility. The integration of ML into clinical workflows may represent a step forward in delivering personalized and dynamic care after pancreatic surgery.
