Dynamic ultrasound indices assess preload-dependent changes in stroke volume via the Frank-Starling mechanism and guide fluid therapy. This study aimed to determine optimal cutoff values for ultrasound-derived peak aortic (ΔVAo) and carotid (ΔVCa) velocity variations to predict fluid responsiveness in critically ill pediatric oncology patients. In this prospective cohort, 83 children underwent 88 fluid challenges with 10 mL/kg saline. Fluid responsiveness was defined as a >15% increase in cardiac index, measured by left ventricular outflow tract Doppler after volume expansion. Fluid responsiveness occurred in 54.5% of assessments. ΔVAo demonstrated the highest predictive accuracy, with a 16.3% cutoff (sensitivity 91.6%, specificity 80%, AUC 0.89). ΔVCa showed moderate performance (cutoff 14.2%; sensitivity 79.1%, specificity 65%, AUC 0.75), while ΔIVC was not predictive (AUC 0.56). In mechanically ventilated patients (n = 60), ΔVAo remained accurate (cutoff 16.3%; AUC 0.90), whereas ΔVCa was modest (cutoff 16.5%; AUC 0.74). In spontaneously breathing patients (n = 28), ΔVAo cutoff was 15.5% (sensitivity 95%, specificity 87.5%, AUC 0.89), and ΔVCa was 13.2% (sensitivity 100%, specificity 50%, AUC 0.69). ΔVAo is a reliable predictor of fluid responsiveness in critically ill pediatric oncology patients. ΔVCa may serve as an alternative, though with lower accuracy.

The incidence of pancreatic neuroendocrine neoplasms (NENs) is rising; whether this reflects a true increase in disease occurrence or improved detection remains uncertain. We conducted a retrospective, population-based study using data from the Surveillance, Epidemiology, and End Results (SEER) Program (1975-2021) to examine temporal trends in the incidence of pancreatic NENs and assess whether changes reflect improved detection versus a true increase. Incidence trends were stratified by demographic and socioeconomic proxies of healthcare access, including income, residential setting, and race as recorded in SEER at the county level. We identified 16,253 cases of pancreatic NENs (44.6% women; median age 62 years). Incidence increased 7.75-fold between 1975 and 2021, rising from 0.21 cases per 100,000 population in 1975 to 1.58 per 100,000 in 2021. Median tumour size at diagnosis decreased significantly, with an average annual reduction of 0.73 mm (R² = 0.765; p < 0.001). After adjustment, incidence increased more steeply among men, individuals aged 40-65 years and >65 years (vs. <40 years), those recorded as White (vs. Black and other races), individuals with higher income, and those residing in urban (vs. rural) counties. Incidence also rose more steeply for tumours located in the pancreatic tail, for grade 1 tumours (vs. grades 2 and 3), and for smaller tumours (vs. larger ones). The rising incidence of pancreatic NENs is probably explained by improved detection, particularly among populations with greater access to healthcare, rather than by a true increase in disease occurrence.

Falls are a leading cause of injury in older adults, making risk prediction a clinical priority. While many machine learning (ML) models exist, they typically provide a static assessment, predicting risk for a single, fixed timeframe. This approach overlooks how the clinical drivers of risk evolve over time, failing to distinguish between acute factors that signal an imminent fall and chronic conditions that confer long-term vulnerability. This study uses an explainable ML framework across multiple time horizons to unlock deeper clinical insights into the temporal nature of fall risk. We conducted a retrospective, matched case-control study using electronic health record (EHR) data from 99,078 patients who fell and 99,078 matched controls. Seven ML models were trained to predict fall risk across seven distinct prediction windows (3, 6, 12, 24, 36, 48, and 60 months). The best-performing model for each horizon, consistently XGBoost, was interpreted using SHAP (SHapley Additive exPlanations) to identify how the importance of clinical and demographic predictors changed over time. A clear performance trade-off emerged across time horizons. Short-term models (3-12 months) delivered balanced discrimination (best model XGBoost, AUC ≈ 0.75), while long-term models became progressively better at identifying eventual fallers (recall ≈ 80% at 60 months) at the cost of lower specificity (≈ 46%). SHAP analysis revealed distinct temporal patterns: short-term risk was driven by acute conditions like syncope, respiratory symptoms, and urinary tract infections, while long-term risk was predicted by chronic, cumulative factors such as spondylopathies, nutritional deficiencies, and benign neoplasms. Three primary risk trajectories (increasing, steady, and decreasing) were identified, each corresponding to distinct underlying clinical profiles. Fall risk is a dynamic process, not a static state. By analyzing risk across multiple timeframes, we can distinguish between acute triggers requiring immediate intervention and chronic vulnerabilities demanding long-term management. This multi-horizon framework provides a data-driven foundation for a new paradigm in fall prevention: moving beyond generic "high-risk" labels to personalized, temporally aware strategies that align the type and timing of interventions with the specific nature of a patient's evolving risk.

The lymph node ratio (LNR) is gaining recognition as a prognostic biomarker for various malignant neoplasms. However, its prognostic role in postoperative chemotherapy for high-risk muscle-invasive bladder cancer (MIBC) remains unclear.

This study included 2,320 patients with lymph node-positive T2-4N1-3M0 MIBC receiving postoperative chemotherapy from the Surveillance, Epidemiology, and End Results (SEER) database, and a validation cohort of 273 patients from Sun Yat-sen University Cancer Center (SYSUCC). Optimal LNR thresholds were identified by using X-tile software. The relationship between LNR and overall survival (OS) was evaluated via Kaplan-Meier analysis, Cox regression, subgroup analyses, and restricted cubic splines. Nomograms and 101 machine learning (ML) algorithms were developed, incorporating LNR and clinical covariates for enhanced prognosis.

The Kaplan-Meier and restricted cubic splines analyses revealed that LNR was an independent risk factor of OS in post-radical cystectomy and chemotherapy patients with lymph node-positive MIBC. Furthermore, the logistic regression model demonstrated that a lower LNR was significantly associated with a reduced incidence of second primary malignancies. After adjustment for potential confounders-including age, sex, histology, radiotherapy, and TNM staging-patients with elevated LNR values exhibited markedly inferior OS compared with those with lower LNR values, a finding consistent across both the SEER database and SYSUCC cohort. Additionally, the nomogram and machine learning algorithms, integrating LNR with clinical parameters, exhibited robust discriminative performance, with calibration curves confirming excellent concordance between predicted and observed outcomes.

LNR is an independent prognostic factor for OS in post-radical cystectomy and chemotherapy patients with lymph node-positive MIBC.

Myelodysplastic syndromes (MDS) comprise clonal hematopoietic stem cell disorders characterized by heterogeneous clinical manifestations. Approximately 20-30% of MDS cases progress to acute myeloid leukemia, whereas transformation to acute lymphoblastic leukemia (ALL) is extremely rare. In this report, we present the case of a Chinese male patient who presented with MDS-refractory anemia with ringed sideroblasts, which developed into B-cell ALL. During disease transformation, the patient acquired novel gene mutations. By comparing the gene mutations identified at the initial MDS diagnosis with those observed at the time of transformation to ALL, we aim to elucidate the genetic alterations associated with disease progression. Furthermore, we provide a comprehensive review of 60 MDS and MDS/MPN cases reported in the literature so far.

Trastuzumab used for the treatment of patients with HER2-positive breast cancer induces cardiotoxicity. The NRG1/HER pathway plays a central role in human cardiovascular physiology; however, the link between exercise, NRG1, and cardiotoxicity is unclear.

Patients were randomized to receive adjuvant trastuzumab in combination with a training program (TG) or trastuzumab alone (CG). The aim of this study was to assess the effect of a 12-week supervised exercise training on the circulating level of NRG1. Secondary objectives were to assess the correlation between NRG1 level and cardiotoxicity.

89 patients were randomized (TG; n = 46; CG; n = 43), 76 have a NRG1 concentration available at baseline. After the exercise program, plasma levels of NRG1 decreased significantly in the TG (mean difference -0.20 ng/ml; 95% CI, -0.32, - 0.07) whereas they remained stable in the CG (mean difference - 0.05 ng/ml; 95% CI, - 0.20, 0.10). Notably, baseline NRG1 concentrations were higher in the TG group. However, no correlation between NRG1 changes and either cardiorespiratory fitness (V̇O₂ max) and left ventricular ejection fraction (LVEF) was observed (R = 0.087, p = 0.53; R = -0.157, p = 0.26; and R = -0.131, p = 0.33, respectively).

A 12-week interval training program significantly decreased NRG1 concentration in HER2-positive patients with breast cancer treated with adjuvant trastuzumab therapy, despite the trained group presenting higher baseline NRG1 values compared to the control group. In addition, this change was neither associated with V̇O₂ max nor with LVEF.

This trial was registered with ClinicalTrials.gov under the number NCT02433067.

The incorporation of organoids with immune cells in co-culture systems signifies a groundbreaking advancement in the fields of cancer research and immunology. These three-dimensional models, derived from primary tumor specimens or stem cells, provide a more accurate representation of the tumor microenvironment (TME) than conventional two-dimensional cultures or animal models. This enhanced model allows for a thorough examination of the intricate interactions between cancer cells and the immune system. Although the success rates for organoid initiation can vary, averaging 36.8% across 13 different tumor types, successful organoid establishment enables the co-culture with a variety of immune cells, such as T cells, tumor-infiltrating lymphocytes (TILs), peripheral blood mononuclear cells (PBMCs), macrophages, dendritic cells, and natural killer (NK) cells. This platform enables the study of immune responses to cancer, mechanisms of immune evasion, and the influence of the TME on immune activation and suppression. The review emphasizes research involving intestinal, pancreatic, brain, liver, and cervical organoids, highlighting their role in elucidating disease mechanisms, assessing the effectiveness of immunotherapies (including checkpoint inhibitors and therapeutic vaccines), and conducting preclinical drug evaluations. Notable examples include modeling graft-versus-host disease with intestinal organoids, investigating the influence of DCLK1 on immunosuppression in pancreatic cancer, evaluating the effectiveness of engineered T cells against neuroblastoma using brain organoids, and analyzing the effects of cancer-associated fibroblasts on drug responses in colon cancer. Additionally, the potential of organoids in vaccine development and testing, particularly for influenza and other viral infections, is examined, demonstrating their utility in assessing immune responses and vaccine effectiveness. Despite existing challenges, such as the relatively low efficiency of organoid generation and the complexities involved in fully mimicking the TME, ongoing technological innovations, including tumor-on-chip systems and enhanced matrix materials, are expected to improve the functionality and clinical applicability of these advanced in vitro models.

Ferritin heavy chain 1 (FTH1) is abnormally expressed in various cancers, but its role and mechanism in endometrial cancer (EC) remain unclear. This study aims to explore the clinical significance, biological functions and potential inhibitor of FTH1 in EC. The expression, prognosis and clinical correlation of FTH1 in EC were analyzed using TIMER, GEO, kaplan-meier plotter and UALCAN databases. Virtual screening and molecular docking were conducted for identifying potential inhibitors of FTH1. In vitro experiments were conducted using human EC cells HEC-1A and RL95-2. Cell proliferation, cell cycle and apoptosis were detected by CCK-8 assay and flow cytometry. The level of reactive oxygen species (ROS) was detected by using the DCFH-DA probe. The levels of malondialdehyde (MDA) and glutathione (GSH) were detected using the corresponding test kits. Western blot was used to detect the expression level of FTH1, AKT and p-AKT. FTH1 was highly expressed in EC tissues and was associated with a shorter overall survival time of patients. Functional enrichment analysis revealed that FTH1 was mainly involved in the iron homeostasis and ferroptosis pathways. FTH1 knockdown inhibits the proliferation of EC cells, induces cell cycle arrest at the G0/G1 phase and triggers cell apoptosis. In EC cells with FTH1 knockdown, the levels of ROS and MDA were significantly increased, accompanied by a decrease in GSH levels. Furthermore, morin had a high binding affinity with FTH1, which also inhibited the malignant phenotypes of EC cells. Morin triggered ferroptosis of EC cells by down-regulating FTH1 expression and inhibiting the PI3K/AKT pathway. FTH1 is a potential prognostic biomarker and therapeutic target for EC. Morin induces ferroptosis of EC cells by regulating FTH1-PI3K/AKT axis, providing a new candidate drug and theoretical basis for the treatment of EC.

The clinical success of FLT3 inhibitors has led to their steadily increasing use in the treatment of acute myeloid leukemia (AML), both in the relapsed/refractory setting and as post-transplant maintenance. Despite their expanding application, there is currently no guidance on the optimal duration of therapy or the feasibility of discontinuation. In the maintenance context, current practice is largely based on trial protocols with predefined treatment periods, yet relapses after cessation have been documented. Similarly, in the relapsed/refractory setting, the management of long-term responders to FLT3-directed monotherapy lacks evidence-based guidance.We report two cases of FLT3-ITD AML patients with relapse after discontinuation of prolonged FLT3 inhibitor therapy, despite sustained remission prior to withdrawal. As such scenarios remain insufficiently characterized in the literature, these case vignettes are presented to highlight the unresolved challenge of defining the appropriate duration of FLT3 inhibitor therapy and to underscore the need for systematic evaluation to establish evidence-based strategies for safe discontinuation or extended administration.

The Geriatric-8 (G8) is used for the functional status of older adult patients with cancer. However, its role in treatment decision-making for gynecological malignancies has not been established.

We retrospectively analyzed the data of 180 women aged ≥75 years with gynecological malignancies who underwent initial treatment at our institution between January 2019 and December 2023. Pre-treatment G8 scores were assessed and patients were categorized as fit (G8 > 14) or frail (G8 ≤ 14). Associations between the G8 score and patient background, disease characteristics, treatment options, and treatment tolerability were examined.

Of the 180 women, 53 (29.4%) were classified as fit and 127 (70.6%) as frail. Frail patients required long-term care (P = .008) and used anticoagulants more frequently than fit patients (P = .019). Median G8 scores were highest in endometrial cancer (14) and lowest in vulvar cancer (10). Best supportive care (8) and neoadjuvant chemotherapy (10) had lower G8 scores than surgery and concurrent chemoradiotherapy (14) (P < .001). Postoperative complications occurred in 10/96 surgical cases; these cases had lower scores than those without complications (12 vs. 14, P = .044). During chemotherapy, median scores were lower in women with ≥ grade 3 (12 vs. 14, P = .008) and grade ≥ 4 adverse events (10 vs. 14, P = .002).

The G8 score is associated with patient background, cancer type, and treatment options, and is associated with treatment tolerability in women aged ≥75 years with gynecological malignancies.