ABSTRACT Background Bulky disease represents a clinically aggressive subset of diffuse large B‐cell lymphoma (DLBCL) associated with adverse clinical outcomes. The aim of this study was to investigate the influence of oncogenic mutations and tumor microenvironment alterations on bulky disease in DLBCL. Methods We analyzed a cohort of 939 patients with newly diagnosed DLBCL. Using DNA (n = 934) and RNA (n = 524) sequencing, we compared oncogenic mutations and tumor microenvironment (TME) alterations based on tumor diameter, with cutoff values at 5.0 cm and 10.0 cm. Further stratification by mutations in key genes (CD58, STAT6, EBF1) correlated with tumor diameter revealed distinct transcriptomic and immunologic profiles. Subsequent single‐cell RNA sequencing, guided by these mutational signatures, resolved the cellular heterogeneity within the TME. Results Integrative analysis revealed that tumor diameter correlated with increased incidence of mutations in CD58, STAT6, and EBF1; adverse genetic subtypes such as EZB‐like MYC+ and TP53Mut; activation of oncogenic pathways (JAK/STAT, BCR, PI3K, and MYC); and an immunosuppressive tumor microenvironment. Notably, immune checkpoint molecules varied across the bulky stages, with CTLA‐4, TIGIT, ICOS, and CD28 expression inversely correlated with tumor diameter, while CD70 and 4‐1BBL expression positively correlated. Single‐cell RNA sequencing further revealed mutation‐specific tumor microenvironment insights. CD58‐mutated tumor exhibited a profoundly immune‐deserted microenvironment dominated by malignant B cells with minimal immune infiltration, whereas STAT6‐mutated tumor was associated with increased fibroblasts and CD4 + T cells, particularly regulatory T cells (Treg) and Th1‐like cells; EBF1‐mutated tumor was characterized by increased proportions of malignant B cells. Conclusions Collectively, our findings highlight the biological complexity of bulky disease, identifying candidate molecular targets and providing a biological framework for future therapeutic hypothesis generation in this clinically aggressive subset of DLBCL.
ABSTRACT Background Metabolic dysfunction‐associated steatotic liver disease (MASLD) is a growing public health concern, particularly among young adults. While MASLD is a well‐established risk factor for liver‐related complications, growing evidence suggests an association with several extrahepatic cancers. However, data regarding cancer risk among young adults with MASLD remain limited. This study investigates the association between MASLD and cancer risk in young adults in South Korea. Methods A nationwide cohort study was conducted using the Korean National Health Insurance Service database, including individuals aged 20–39 years who underwent health screenings (2004–2007). Clinically diagnosed fatty liver disease was identified using ICD‐10 code K76.0, and incident cancers were identified using ICD‐10 malignant neoplasm codes recorded in the National Health Insurance Service claims database through 2022. Cox proportional hazards models estimated hazard ratios (HRs) for cancer risk, adjusting for confounders. Results Among 4,094,241 participants, 254,897 (6.22%) had clinically diagnosed fatty liver disease. Clinically diagnosed fatty liver disease was significantly associated with increased all‐cancer incidence in both men (HR: 1.48, 95% CI 1.45–1.51) and women (HR: 1.23, 95% CI 1.20–1.27). The highest risks were observed for liver cancer (men, HR: 2.39, 95% CI 2.28–2.50, women, HR: 2.88, 95% CI 2.56–3.23), thyroid cancer (men, HR: 1.49, 95% CI 1.43–1.55, women, HR: 1.18, 95% CI 1.13–1.24). BMI‐based fatty liver disease (BMI < 25 kg/m2) showed a higher liver cancer risk. Non‐drinkers with clinically diagnosed fatty liver disease also had increased cancer risk. Conclusions Clinically diagnosed fatty liver disease in young adults was associated with a higher risk of both hepatic and extrahepatic cancers. Further studies are needed to determine appropriate risk‐stratified prevention and surveillance strategies in this population.
ABSTRACT Cervical adenocarcinoma accounts for 15%–20% of cervical cancers and is associated with poorer survival and reduced response to screening and immunotherapy compared with squamous cell carcinoma (SCC). The genomic drivers underlying this molecular subgroup remain incompletely characterized. Whole‐exome sequencing was performed on 302 invasive cervical cancers from Guatemala and Venezuela. Structural variation analysis was conducted using SNP‐array and whole‐genome sequencing data. Findings were replicated in more than 4600 additional cervical cancer samples from TCGA, AACR Project GENIE, MSKCC, and Caris datasets. TP53 mutations were more frequent in adenocarcinoma than SCC, particularly in HPV‐negative tumors. STK11 alterations, including mutations and focal deletions, were significantly enriched in HPV‐positive adenocarcinomas compared with SCC and affected 23% of adenocarcinomas overall. Whole‐genome analyses identified recurrent focal deletions, inversions, chromosomal rearrangements, and breakage‐fusion‐bridge events involving chromosome 19p and STK11 that were not detected by exome sequencing alone. STK11 alterations were associated with younger age at diagnosis, poorer overall survival, and inferior outcomes following immune checkpoint inhibitor (ICI) therapy. STK11 alterations significantly co‐occurred with YAP1 amplification but were largely mutually exclusive with PIK3CA mutation. Cervical adenocarcinomas also demonstrated significantly lower CD274 (PD‐L1) expression than SCC. STK11 alterations define a distinct molecular subgroup of cervical adenocarcinoma characterized by structural disruption of chromosome 19p, younger age at onset, and poorer clinical outcomes. These findings have implications for molecular classification and future targeted therapeutic approaches in cervical cancer.
ABSTRACT Background Breast cancer is the most commonly diagnosed malignancy in women and a leading cause of cancer‐related mortality. Beyond the physical burden, the disease significantly impacts emotional well‐being and quality of life (QoL). Telenursing, as a nurse‐led telemedicine intervention, has emerged as a promising strategy to support patients remotely. Objective To evaluate the effectiveness of nurse‐led telenursing interventions on QoL in women with breast cancer and to compare the impact of different technological delivery modes. Methods A systematic review and meta‐analysis was conducted following PRISMA 2020 and the Cochrane Handbook. Studies were identified through five databases. Randomized controlled trials (RCTs) and quasi‐experimental studies comparing nurse‐led telenursing to usual care were included. The primary outcome was QoL. Risk of bias was assessed using RoB 2.0 and ROBINS‐I; GRADE was used to rate the certainty of evidence. Results Seventeen studies (n = 4199) were included in the systematic review. Seven studies provided sufficient quantitative data for meta‐analysis. Overall, telenursing interventions showed a small to moderate pooled effect on quality of life (SMD = 0.40, 95% CI [0.09, 0.70]; p = 0.019); however, after trim‐and‐fill correction for potential publication bias, the adjusted estimate was attenuated and did not reach statistical significance (adjusted SMD = 0.14, 95% CI [−0.17, 0.45]; p = 0.376). Subgroup analyses—to be interpreted as exploratory—suggested a larger effect for mobile application–based interventions (SMD = 0.63, 95% CI [0.46, 0.80], I2 = 0%, k = 3) compared to telephone follow‐up (SMD = 0.21, 95% CI [−0.34, 0.76], I2 = 85.2%, k = 4), though the small number of contributing studies limits the reliability of this comparison. Conclusions These findings provide preliminary evidence that nurse‐led telenursing may be associated with improvements in quality of life in women with breast cancer, though the certainty of the evidence is low (GRADE: Low). Results should be interpreted with caution given the attenuation of the pooled effect to non‐significance after trim‐and‐fill adjustment, substantial methodological heterogeneity, and the exploratory nature of subgroup comparisons. High‐quality randomized trials are needed to confirm these findings.
ABSTRACT Anaplastic thyroid cancer (ATC) is a rare but extremely aggressive malignancy responsible for most thyroid cancer–related deaths. To elucidate molecular and immunological characteristics distinguishing ATC from aggressive papillary thyroid cancer (PTC), we performed single‐cell RNA sequencing on three ATC and three lethal PTC samples. ATC tumor cells exhibited enhanced inflammatory and immune‐related gene expression, and co‐activation of MAPK and PI3K pathways. Notably, p53 pathway activity was suppressed in ATC, alongside activation of angiogenic and immune‐evasion programs, indicating a shift from proliferation‐driven to immune‐evasion–oriented signaling. Transcriptional network analysis revealed E2F family activation, including E2F1, E2F7, and E2F8, suggesting CDK–RB–E2F axis dysregulation drives tumor dedifferentiation and proliferation, whereby p53 suppression unleashes E2F‐dependent transcriptional reprogramming. Immunoprofiling revealed CD8+ T cells in ATC showed elevated exhaustion signatures and high expression of inhibitory immune checkpoint molecules, accompanied by enhanced Treg–CD8+ T cell interactions, indicating a dual‐layered immunosuppressive mechanism involving both T‐cell exhaustion and enhanced Treg‐mediated suppression. In addition, we identified antigen‐presenting CAFs with altered CD4+ T‐cell interactions. These findings delineate a comprehensive single‐cell landscape of ATC and demonstrate coordinated alterations in tumor‐intrinsic signaling and the tumor microenvironment during dedifferentiation, provide a foundation for developing rational combination immunotherapies tailored to the molecular and immunological landscape of thyroid cancer.
ABSTRACT Background Renal impairment (RI) increases mortality risk in multiple myeloma (MM), yet RI represents a relatively late stage of kidney dysfunction. Whether early indicators of renal dysfunction, such as the estimated glomerular filtration rate (eGFR slope), predict mortality risk in MM remains unknown. Methods This single center, retrospective study included 974 newly diagnosed MM (NDMM) patients (2008–2022), with prospectively monitored creatinine levels and survival status over a median follow‐up period of 40 months. The first‐year eGFR slope, representing the average annual rate of eGFR change, was calculated by least‐squares regression using all monitored eGFR values during the first year. Cox proportional‐hazards regression and restricted cubic splines (RCS) analyses were used to explore the association of first‐year eGFR slope with all‐cause mortality. Subgroup analyses were stratified by first‐year treatment response (partial response (PR) or better response group and less than PR group), and baseline eGFR strata (< 40, 40–90, and ≥ 90 mL/min/1.73 m2). Results In the overall population, the first‐year eGFR slope‐mortality association exhibited a U‐shaped pattern (p for nonlinearity 0.0145), indicating that excessive declines and increases in eGFR were linked to elevated mortality. This pattern was consistent across baseline renal function strata. Notably, the association varied substantially by treatment response: in the PR or better group, a rise in eGFR slope conferred a higher mortality risk (HR > 1 when annual increment exceeded 4.3 mL/min/1.73 m2), whereas in the less than PR group, mortality risk was predominantly driven by eGFR slope declines (HR > 1 when slope fell below 1.18 mL/min/1.73 m2/year). Conclusions Overall, first‐year eGFR slope demonstrates a U‐shaped association with mortality in NDMM. Importantly, excessive eGFR elevation, particularly in patients achieving PR or better, is paradoxically associated with increased death risk, highlighting the need for response‐stratified interpretation of renal dynamics to guide early clinical decision‐making.
OBJECTIVE:To investigate the clinical characteristics of newly diagnosed multiple myeloma (NDMM) patients with 1q21 gain/amplification (1q21+), construct and validate a progression-free survival (PFS) prediction model, and explore the clinical implications of model-based risk stratification. METHODS:We retrospectively analyzed 186 newly diagnosed multiple myeloma patients with 1q21+ treated between 2018 and 2024. Patients were randomly assigned at a 7:3 ratio to a training cohort (n = 131) and an internal split-sample validation cohort (n = 55). Independent prognostic factors for PFS were identified using multivariate Cox regression to construct a nomogram (the HLBP model). Model performance was evaluated through bootstrap resampling, time-dependent receiver operating characteristic (ROC) curves, calibration plots, and Kaplan-Meier survival analysis. Furthermore, the model's predictive accuracy was compared with conventional staging systems and established prognostic models. Finally, risk-stratified exploratory analyses were conducted to assess treatment-associated outcomes across different risk groups. RESULTS:Higher hemoglobin was protective, whereas elevated lactate dehydrogenase, increased β2-microglobulin, and TP53 deletion were adverse prognostic factors. The HLBP model yielded AUCs of 0.805, 0.885, and 0.847 at 6, 12, and 24 months in the training cohort, and 0.747, 0.827, and 0.712 in the internal split-sample validation cohort. Patients stratified into high- and low-risk groups showed significantly different PFS outcomes. Risk-stratified analyses suggested heterogeneous prognostic patterns associated with induction treatment regimens and autologous stem cell transplantation across risk groups. Overall, the HLBP model demonstrated superior predictive performance compared with existing staging systems and previously reported prognostic models. CONCLUSION:The HLBP model showed promising preliminary performance for predicting PFS in patients with 1q21+ NDMM and may provide an exploratory framework for individualized risk stratification.
BACKGROUND:Brazil has substantially expanded public oncology investment over the past decade, particularly within the Brazilian Unified Health System (SUS). However, whether increased spending has translated into measurable population-level improvements in cancer outcomes remains uncertain. METHODS:We conducted a retrospective ecological time-series analysis using national data from 2008 to 2023. Public expenditures for female breast and prostate cancer were extracted from DATASUS and inflation-adjusted, while mortality data were obtained from INCA and age-standardized. Key policy milestones-RDC 55/2010, creation of Conitec in 2011, Ordinance No. 874/2013, and Constitutional Amendment 95/2016-were mapped against expenditure and mortality trends. Findings therefore represent system-level associations rather than causal estimates of patient-level therapeutic effectiveness. RESULTS:Oncology expenditures increased substantially, with more pronounced growth after 2011, whereas national age-standardized mortality rates for female breast and prostate cancer changed comparatively little. This pattern represents a descriptive temporal divergence between expenditure growth and mortality trends, not evidence of therapeutic inefficacy or a quantified causal effect. Regional differences in diagnostic capacity, healthcare infrastructure, continuity of care, and treatment timeliness may influence how financial inputs translate into population-level outcomes. CONCLUSIONS:Brazil's experience supports a value-oriented oncology strategy linking investment not only to technology incorporation, but also to early diagnosis, implementation capacity, regional equity, and effective care pathways. These ecological findings do not support individual-level causal inference, particularly for prostate cancer, for which mortality is a delayed and relatively insensitive medium-term endpoint. Future studies should incorporate stage at diagnosis, treatment pathways, time-to-treatment indicators, and regional survival data.
ABSTRACT Background Predicting clear cell renal cell carcinoma (ccRCC) pathological grade preoperatively is critical for clinical management. This study aims to evaluate the diagnostic accuracy and clinical utility of machine learning (ML)‐based imaging models. Methods The Cochrane Library, PubMed, Embase, and Scopus databases were searched systematically for studies published before January 2026. Study quality was assessed using the Quality Assessment of Diagnostic Accuracy Studies 2 and the Radiomics Quality Score. Pooled sensitivity, specificity, positive/negative likelihood ratios (PLRs/NLRs), diagnostic scores, diagnostic odds ratios (DORs), and summary receiver operating characteristic (SROC) curves were calculated. Decision curve analysis (DCA) and Fagan nomogram analysis were performed to evaluate clinical utility. Subgroup analyses were conducted to further explore sources of heterogeneity. Results A total of 43 studies involving 12,675 patients were included. The area under the SROC curve was 0.89, with a sensitivity of 0.79, specificity of 0.85, PLR of 5.27, NLR of 0.25, diagnostic score of 3.07, and DOR of 21.52. Fagan analysis revealed a positive prediction increased the posttest probability of high‐grade disease to 70%, whereas a negative prediction decreased it to 10%. DCA demonstrated a net benefit over standard strategies across a 0.10–0.70 threshold range. Subgroup analyses revealed significantly greater sensitivity for the deep learning (DL) models than for the radiomics (0.91 vs. 0.75; p < 0.01) and automatic models compared with manual segmentation (0.86 vs. 0.76; p = 0.03). Notably, single‐center independent validation (0.92) outperformed both multicenter external (0.79) and internal validation (0.72) strategies (p < 0.01). No significant performance differences were observed across imaging modalities, phase protocols, clinical variable integration, geographic regions, or sample sizes. Conclusion This study confirms the significant potential of radiomics and DL models for the preoperative prediction of the pathological grade of ccRCC. Nevertheless, future multicenter validation is essential to address the performance gap observed in external datasets. Trial Registration Prospero: CRD42023455847
BACKGROUND:Ferritin elevation is frequently observed during cytokine release syndrome (CRS) following CAR T-cell therapy; however, its independent clinical utility as a predictive or prognostic biomarker remains uncertain. We performed a systematic review critically evaluating the timing-specific and context-dependent role of ferritin in CRS risk stratification and outcomes. METHODS:We systematically searched PubMed, Web of Science, and Scopus for studies evaluating ferritin in CAR T-cell recipients. We performed structured qualitative synthesis with semi-quantitative comparison across studies, including direction-of-effect analysis and threshold stratification. We also performed a meta-analysis on the association of progression-free survival (PFS) and overall survival (OS) based on pre-infusion ferritin levels using a random-effects model. RESULTS:Fifteen studies (n = 1671) were included. Elevated pre-infusion ferritin (commonly ≥ 400 ng/mL) was associated with higher CRS incidence and severity in most studies (10/12), but its independent predictive value was inconsistent. Post-infusion ferritin demonstrated a consistent association with CRS severity across all studies (9/9), with extreme elevations (> 10,000 ng/mL) observed in high-grade CRS. However, ferritin lacked specificity as a standalone biomarker and performed more robustly when integrated into multimarker models (e.g., cytokines, EASIX score). Survival associations were heterogeneous and likely confounded by disease burden and systemic inflammation. Interestingly, meta-analysis showed that pre-infusion ferritin levels were significantly associated with worse PFS [HR: 2.18 (1.74-2.73), p < 0.00001, I2 = 0%] and worse OS [HR: 2.97 (2.22-3.97), p < 0.00001, I2 = 0%]. Potential publication bias was not identified in this analysis. CONCLUSION:Ferritin is best interpreted as a dynamic inflammatory correlate rather than an independent predictor of CRS. Its clinical utility in CRS prediction lies in risk enrichment when combined with other biomarkers and clinical scores, rather than as a standalone decision tool. Nevertheless, pre-infusion levels may be useful in predicting worse PFS and OS given the standardization of timing, thresholds, and integration into predictive models.
OBJECTIVE:To investigate the prognostic value of the hemoglobin, albumin, lymphocyte, and platelet (HALP) score in patients with locally advanced nasopharyngeal carcinoma (LANPC) receiving concurrent chemoradiotherapy (CCRT), and to clarify its predictive significance for overall survival (OS), progression-free survival (PFS), locoregional failure-free survival (LRFFS), and distant metastasis-free survival (DMFS). METHODS:Clinical data of 295 patients with LANPC who received CCRT between April 2021 and April 2022 were retrospectively analyzed. Routine blood and biochemical indices were collected within 1 week before CCRT. The HALP score was calculated as follows: HALP = [hemoglobin (g/L) × albumin (g/L) × lymphocytes (109/L)]/platelets (109/L). Patients were divided into high and low HALP groups with the median as the cutoff point. Kaplan-Meier analysis was used to draw survival curves. The log-rank test was used to compare the survival differences between the high and low HALP groups. Univariate and multivariate Cox regression models were applied to identify independent prognostic factors for patients with LANPC. RESULTS:Survival analysis showed that the 3-year OS of the high and low HALP groups were 95.27% and 93.20%, the 3-year PFS were 87.84% and 74.15%, and the 3-year DMFS were 87.84% and 78.91%, respectively. The differences between the groups were statistically significant (p < 0.05). While the 3-year LRFFS were 97.30% vs. 96.60%, there were no statistically significant differences between the groups. Multivariate Cox regression analysis confirmed that a low HALP score was an independent risk factor for OS and PFS. CONCLUSION:The HALP group is an effective predictor of prognosis in patients with LANPC. A low HALP group indicated higher disease progression and death. This measurement can be used for prognostic stratification and individualized treatment decision-making, both simply and economically, and provides a reference for clinical decision-making and individualized treatment.
BACKGROUND:ER stress (ERS) influences tumor behavior through the unfolded protein response (UPR), yet its role in glioma is not fully defined. This study investigated the function of the ERS-related regulator QRICH1 in glioma progression. METHODS:Public databases (TCGA and GTEx) were used to evaluate QRICH1 expression, survival, and prognostic significance in glioma. Lentiviral QRICH1 overexpression or knockdown was established in U87 and U251 cells, followed by functional and mechanistic assays. A subcutaneous xenograft model, histological analyses, and immunohistochemistry were performed to validate the biological function of QRICH1 in vivo. RESULTS:QRICH1 was markedly upregulated in glioma and associated with better patient survival. QRICH1 overexpression suppressed glioma proliferation, migration, and invasion while promoting apoptosis, whereas silencing QRICH1 enhanced malignancy. Mechanistically, QRICH1 activated the PERK-ATF4-CHOP pathway, increased caspase-12 cleavage, and strengthened ERS-induced apoptosis. In vivo, QRICH1 overexpression reduced tumor size and increased apoptotic markers. CONCLUSION:QRICH1 shifts QRICH1 shifts the UPR toward its pro-apoptotic branch, thereby inhibiting glioma progression, and represents a promising prognostic biomarker and therapeutic target.
PURPOSE:With AI tools being increasingly utilized for medical inquiries, this study evaluated the agreement between clinicians, text-only clinicians, DeepSeek-V3, and ChatGPT-4o in TNM staging and preferred treatment recommendations for newly diagnosed nasopharyngeal carcinoma (NPC) patients. METHODS:A retrospective study analyzed 322 consecutive NPC patients treated at our institution from January 2023 to February 2025. TNM staging (AJCC 8th edition) and preferred treatment recommendations were independently assessed by text-only clinicians, DeepSeek-V3, and ChatGPT-4o. Interrater agreement was quantified using Cohen's kappa coefficient and Fleiss' kappa analysis (κ), with the range of κ = 0.81-1.00 considered almost perfect agreement. RESULTS:The cohort comprised 244 males and 78 females (median age, 52 years, range, 18-77 years). Cohen's kappa analysis showed that for clinical staging, moderate agreement was exhibited in clinician-AI comparisons, while substantial agreement was observed between clinicians and text-only clinicians. For preferred treatment recommendations, slight agreement was noted in clinician-AI comparisons, while moderate agreement was found between clinicians and text-only clinicians. Fleiss' kappa analysis demonstrated moderate agreement among the 4 raters for T stage, N stage, and clinical staging, while M stage achieved almost perfect agreement. However, overall agreement for treatment recommendations was fair. CONCLUSIONS:The AI tools demonstrated moderate agreement with clinicians in overall clinical staging for NPC, with heterogeneous performance across T, N, and M categories (almost perfect for M stage, moderate for T and N stages), whereas their preferred treatment recommendations showed only slight agreement with clinical decision-making.
ABSTRACT Tertiary lymphoid structures (TLSs) modulate immune responses in various solid tumors, but their comprehensive role in lung adenocarcinoma (LUAD) remains unclear. In this study, we analyzed RNA‐seq data from 539 LUAD patients in The Cancer Genome Atlas (TCGA) and microarray data from 223 samples from the Gene Expression Omnibus (GEO, GSE13213, and GSE37745). TLS signatures were evaluated via unsupervised consensus clustering based on 12 chemokine transcriptome signatures. The relationships between TLS and clinical characteristics, tumor microenvironment (TME) cell infiltration, and prognosis were assessed using ESTIMATE and CIBERSORT. A prognostic model was established using LASSO regression and validated with external datasets. Additionally, H&E and IHC analyses were performed to explore associations between intratumoral TLS density, immune‐related molecular expression, and patient prognosis in LUAD. Consensus clustering of the TCGA cohort revealed two distinct LUAD patient clusters according to TLS abundance. Cluster 1 exhibited greater immune cell infiltration, more favorable prognosis, and increased expression of immune checkpoint molecules. We developed a prognostic model comprising eight survival‐associated genes that act as independent prognostic factors for patient survival. H&E/IHC analyses revealed that TLS density—regardless of pathological stage—was associated with better prognosis; higher intratumoral TLS density/proportion was also related to more favorable outcomes. IHC confirmed that survival‐associated genes (CD5, HLA‐DMB, and P2RY13) are independent prognostic indicators in LUAD. Our study demonstrated the close relationship between TLS signatures and an active immune microenvironment, highlighting their potential as independent prognostic indicators in LUAD.
ABSTRACT Enfortumab vedotin (EV) is a crucial treatment for patients with metastatic urothelial carcinoma (mUC). However, a significant proportion of patients experience adverse events (AEs). The identification of biomarkers for AEs is imperative for the detection and treatment of AEs at an early stage. In this exploratory study, we aimed to identify biomarkers of AEs in patients with mUC treated with EV. We retrospectively examined 10 factors identified from the data of 116 patients with mUC treated with EV to identify biomarkers (age, body mass index, C‐reactive protein level, Eastern Cooperative Oncology Group performance status, eosinophil proportion, history of diabetes, lymphocyte proportion, neutrophil proportion, neutrophil‐to‐lymphocyte ratio, and platelet count) associated with the occurrence of AEs of any grade. The candidate biomarkers were measured at the start of EV treatment. The least absolute shrinkage and selection operator method was used to select the most useful parameters for predicting AE occurrence. Among the 10 factors, eosinophil proportion was identified as the only potential biomarker. The optimal cutoff value for eosinophil proportion against the occurrence of AEs of any grade was 2.5% (area under the curve = 0.625). Univariable logistic regression analyses showed that an eosinophil proportion of ≥ 2.5% was a risk factor for AE development (odds ratio = 4.35, 95% confidence interval = 1.35–14.0). Therefore, the results of this exploratory study indicated that an eosinophil proportion of ≥ 2.5% at the start of EV treatment may be a candidate biomarker for the occurrence of AEs of any grade.
White blood cell differential (WBC-Diff) scattergrams are used to screen for hematological abnormalities in potential hematopoietic malignancies and abnormal conditions. However, these potential indications can be easily missed. We applied deep learning models to develop a new and rapid workflow by learning the characteristics of WBC-Diff scattergram images for early hematological abnormality screening. A total of 4297 WBC-Diff scattergram images, including healthy controls, were included and divided into 19 categories. Eleven models encompassing both CNNs (ResNet18, ResNet50, VGG16, InceptionV3, DenseNet121, MobileNetV2, EfficientNet-B0, and EfficientNet-B1) and Vision Transformers (Swin-Tiny, DeiT-Base, and ConvNeXt-Tiny) were trained and compared. Grad-CAM interpretability analysis, hierarchical clustering, and multivariate analysis of variance (MANOVA) were used to characterize the learned feature space. Among all models, Swin-Tiny achieved the highest accuracy (70.44%), while EfficientNet-B0 was selected as the primary model due to its favorable balance of classification performance (67.90% accuracy) and computational efficiency (4.0 M parameters, 0.42 G floating-point operations (GFLOPs), 2.8 × faster CPU inference than Swin-Tiny). The 5-run stability evaluation confirmed reproducible performance, and evaluation on an independent hold-out dataset confirmed patient-independent generalizability. As an upstream triage endpoint (normal vs. abnormal), the model achieved a sensitivity of 0.975 (test) and 0.984 (independent hold-out set) with a false-negative rate of 2.5% and 1.6%, respectively. At the screening level, only 2.2% of abnormal images were mis-routed to "normal," reflecting a sensitivity-first, specificity-as-safeguard design. Grad-CAM analysis revealed disease-specific activation patterns that mapped to established cell population zones on the WBC-Diff scattergram (e.g., myeloblast zone for M1 and neutrophil-basophil cluster for CML), providing support for the biological plausibility of model decisions, with significant differences among disease categories in the visual feature space (MANOVA, p < 0.01). This study demonstrated that a new automated hematology workflow incorporating EfficientNet-B0 could provide early and rapid screening of potential hematological abnormalities, particularly suitable for deployment in resource-constrained clinical settings.
Studies have demonstrated that the nervous system plays an important role in cancer progression. Nevertheless, the effect of mesenchymal stem cells (MSCs) on cancer has yielded conflicting results. In this study, we discovered that cancer-derived MSCs (CA-MSCs) not only exhibit the properties of neural stem cells (NSCs) but also express a significant amount of neural-related products. These products include neurotrophic factors, neuropeptides, synapse-related products, and axon guidance factors. These identified products exhibit similar or different expression status among human NSCs, cancer and fat-derived MSCs. CA-MSCs expressed more BDNF, GDNF, and NGF than NSCs. Based on these neural-related products expressed by CA-MSCs, expression-prognostic analysis was conducted. It revealed that the expression of MDK and MANF was higher in cancer, and the expression of BDNF, NPTX1, and NGF increased as the tumor stage advanced. For ten neuropeptides, their expression levels were lower in cancer. Thirty synapse-related products displayed their respective expression status between cancer and normal tissues. SYPL1, SYBP2/3, and STX4 had a significant impact on tumor prognosis. Higher levels of syntaxin-4, neugrin, and other axogenesis products were associated with a worse prognosis in cancer. Moreover, immunohistochemistry revealed that NSC-like CA-MSCs were abundant in cancer and exhibited a clear growth-promoting effect on cancer. Briefly, CA-MSCs, which are densely distributed in cancer and have the characteristics of NSCs, express a large number of nerve-related products. These products may have significant effects on tumor progression and prognosis.
Somatostatin receptor 2 is expressed in nasopharyngeal carcinoma (NPC). We report genomic and transcriptomic analysis results of 163 NPC cases, demonstrating that somatostatin receptor 2 (SSTR2) gene expression in EBV-positive and in EBV-negative NPC correlated with genomic alterations and an inflamed microenvironment. Median SSTR2 expression was 6.52 transcripts per million (TPM), ranging from 0.59 TPM (Quartile 1 [Q1], 18.2% EBV-positive) to 35.79 TPM (Q4, 82.9% EBV-positive). PIK3CA (20.51%) and CYLD (10.53%)/NFKBIA (12.82%) mutations were enriched in SSTR2-Q1 versus -Q4. Tumor mutation burden was negatively correlated (R = -0.27, p < 0.001) with SSTR2, while PD-L1 tumor proportion score (2% vs. 92.5% [SSTR2:Q1 vs. Q4], p < 0.001) and T-cell inflamed score correlated with SSTR2 expression (R = 0.53, p < 0.001). B-cells, M1 macrophages, CD8+ T-cells, Treg cells, and dendritic cells were enriched in SSTR2-Q4, while neutrophils were prominent in SSTR2-Q1. These results demonstrate a significant positive correlation between SSTR2 expression and an inflamed tumor microenvironment in NPC. This suggests that SSTR2 expression in NPC may be a clinically useful biomarker, correlating with the tumor microenvironment (including, potentially, sensitivity to immunotherapy) and its genetic profile.
PURPOSE:To evaluate the significance of persistent arterial-phase hyperenhancement (APHE) on contrast-enhanced CT/MRI at 6 and 12 months after proton beam therapy (PBT) for hepatocellular carcinoma (HCC). METHODS:We retrospectively analyzed 108 patients with solitary HCC without vascular invasion who underwent definitive PBT as initial treatment between January 2008 and December 2020. APHE was assessed on contrast-enhanced CT/MRI at approximately 6 and 12 months. Overall survival (OS), local control (LC), and progression-free survival (PFS) were estimated using the Kaplan-Meier method. Associations between APHE and outcomes were evaluated using landmark analyses and Cox models. RESULTS:The median follow-up time was 4.64 years. APHE assessment was available in 87 patients at 6 months and 89 patients at 12 months; APHE was observed in 28 (32.2%) and 26 (29.2%), respectively. At the 6-month landmark, APHE was associated with worse subsequent OS and PFS, whereas LC did not differ significantly. At the 12-month landmark, APHE was associated with worse subsequent LC and PFS and borderline worse OS. The 3-year LC and PFS rates from the 12-month landmark were 69.1% and 25.3% in patients with APHE versus 89.6% and 54.3% in those without APHE (LC: p = 0.015; PFS: p = 0.025). However, after adjustment for clinical and treatment-related factors, APHE status was not independently associated with OS, LC, or PFS in the multivariate landmark analyses. CONCLUSION:Persistent intratumoral APHE at 12 months after PBT was associated with an increased subsequent risk of local failure in landmark analysis, but was not an independent prognostic factor. Persistent APHE may warrant careful imaging follow-up and clinical reassessment but should not be regarded as definitive evidence of viable tumor.
