Cancer Science最新文献

This multicenter longitudinal study was conducted across 24 institutions in Japan to examine the impact of urinary diversion on health-related quality of life (HRQOL) among bladder cancer patients who underwent radical cystectomy (RC). We evaluated bladder cancer-specific HRQOL and general HRQOL via the bladder cancer index (BCI) and the QOL General (QGEN-8), respectively, before the operation and at 3, 6, and 12 months postoperatively. The scores were compared across urinary diversion groups as well as across different time points within each urinary diversion group with linear mixed-effects models. Data from 227 patients were analyzed (151 with ileal conduits, 45 with ureterostomy, and 31 with neobladders). Neobladder patients were more likely to experience longitudinal impacts of their urinary diversion on urinary function than ileal conduit or ureterostomy patients were. Compared with that at baseline, the bowel function of neobladder patients remained impaired 12 months after surgery. All urinary diversion groups had worse sexual function scores at 3 and 6 months than at baseline, and the ileal conduit and neobladder groups had significantly worse sexual function scores at 12 months than at baseline. On the other hand, there was no significant difference in bother scores in the urinary, bowel, or sexual domain. The generic HRQOL was maintained from the preoperative to the postoperative period in all urinary diversion groups. This study explored longitudinal changes in HRQOL after RC, and the findings may help inform patient counseling regarding possible QOL trajectories.

Chimeric antigen receptor T (CAR-T) cell therapy has demonstrated remarkable success in treating hematological malignancies; however, antigen-positive relapse remains a significant obstacle to achieving sustained remission in B-cell acute lymphoblastic leukemia (B-ALL). To enhance the therapeutic efficacy of anti-CD19 CAR (CAR19)-T cells, we overexpressed CXCR4 in CAR19-T cells to improve their trafficking to bone marrow (BM), a key sanctuary for minimal residual disease. We engineered CAR19-T cells with overexpression of wild-type CXCR4 (CAR19/CXCR4^WT-T) or gain-of-function CXCR4 S338X mutant (CAR19/CXCR4^S338X-T). Both CAR19/CXCR4^WT-T and CAR19/CXCR4^S338X-T cells exhibited enhanced CXCR4 surface expression in vitro and in vivo compared to control CAR19-T cells, with the latter showing significantly superior improvements under all tested conditions, including engagement with CAR-specific antigens or CXCR4 ligand CXCL12. Upon engagement with CXCL12, CAR19/CXCR4^S338X-T cells, but not CAR19/CXCR4^WT-T cells, displayed significantly increased activation of ERK1/2 and AKT signaling pathways, as well as elevated transcription of TNF-α, IFN-γ, granzyme B, CDK6, and BCL2A1, along with strengthened effector functions, chemotaxis, and activation of anti-apoptotic pathways. Furthermore, CAR19/CXCR4^S338X-T cells demonstrated significantly improved migration to and retention in the BM accompanied by increased CD45RA⁺CCR7⁺ memory T cell populations, which correlated with enhanced anti-leukemic effects following injection into B-ALL-bearing mice. This study offers a potentially effective strategy to improve the functionality and durability of CAR-T cell responses in hematological malignancies.

Li-Fraumeni syndrome (LFS) is a cancer predisposition syndrome caused by germline pathogenic variants in the TP53 gene. With the increasing use of multi-gene panel testing, TP53 variants have been identified in individuals who do not meet established TP53 testing criteria, such as the Chompret criteria. The term "attenuated LFS" has been proposed for some of these cases, particularly those with adult-onset cancer. We analyzed participants of the Japanese nationwide prospective clinical trial of the cancer surveillance program (Japan Children's Cancer Group LFS-20), along with clinical information including their family histories, to better understand their genotypic and phenotypic characteristics. We identified 32 distinct TP53 variants from 41 families (45 participants), including four missense variants with conflicting classifications of pathogenicity in ClinVar. Among these families, 36 (88%) met the LFS criteria (hereafter referred to as "LFS" in contrast to attenuated LFS), while 5 (12%) were classified as attenuated LFS. Including 30 additional family members carrying the same variant, we analyzed 75 individuals with TP53 variants. Of these, 40 with LFS and 6 with attenuated LFS had cancer. Multiple primary cancers occurred in 22 individuals (21 LFS, 1 attenuated LFS). LFS-core tumors accounted for 66% (58/88) of cancers in the LFS group and 63% (5/8) in the attenuated LFS group; of note, all core tumors in the attenuated group were limited to breast cancer. Hotspot missense variants were detected in 11 of 36 LFS families and in none of 5 attenuated LFS families, and non-hotspot null variants were found in 14 and 1, respectively. Our study revealed genotype-phenotype correlations in several respects. UMIN-CTR: UMIN000045855.

Precision oncology, based on the genomic profile of tumors, is increasing in cancer treatment. It is important in pediatric and adolescent and young adult (AYA) malignancies, a group of rare cancers that are difficult to diagnose and progress quickly. We evaluated the usefulness of in-hospital molecular profiling using a specific sequencing panel for pediatric and AYA malignancies in clarifying the diagnosis, predicting the prognosis, and identifying therapeutic targets. We evaluated the Oncomine Childhood Cancer Research Assay (203 genes, 1700 fusions) and analyzed 153 samples at diagnosis and 34 samples at relapse or refractory from 165 pediatric and AYA patients with various solid malignancies, including neuroblastoma, hepatoblastoma, brain tumor, and sarcoma. We simulated in-hospital molecular profiling. In total, 320 reportable variants were identified in 153 samples (81.8%). Variants with clinical significance were identified in 62 samples (33.1%), including diagnostic variants in 26 (13.9%), prognostic variants in 17 (9%), targetable variants in 41 (21.9%), and variants with drug resistance in 18 (9.6%). Additionally, 62.0% and 94.1% of identified diagnostic and prognostic variants, respectively, had high levels of evidence. In five patients with cancer predisposition syndromes, all pathogenic germline mutations were validated. The turnaround time (TAT) from sample collection to reporting of the molecular profile was within seven working days. In-hospital molecular profiling using a sequencing panel specified tailored for pediatric and AYA malignancies has a high rate of identifying reportable variants and enables accurate diagnosis, malignancy grading, and treatment selection, as well as cancer predisposition syndromes with a short TAT.

Advanced hepatic fibrosis is a major risk factor for cirrhosis and hepatocellular carcinoma (HCC), and it is required to identify a key mediator involved in intratumoral fibrosis and HCC development. Transcriptomic analysis of 372 HCC samples using publicly available datasets revealed that SPP1 was significantly upregulated in fibrotic HCC tissues and associated with unfavorable outcomes. Immunohistochemical analysis of 103 HCC tissues and single-cell RNA sequencing (scRNA-seq) analysis of 228,564 live cells identified SPP1 overexpression in HCC cells, which strongly correlated with intratumoral fibrosis. In xenograft models, HCC cells with SPP1 overexpression (SPP1-OE) exhibited enhanced fibrosis and tumor growth. Coculture assay demonstrated that SPP1-OE cells stimulated hepatic stellate cells (HSCs), and gene set enrichment analysis and differential gene expression analysis elucidated the activation of the Hedgehog signaling pathway and upregulation of GLI1 in HSCs. Cell-cell interaction prediction analysis using scRNA-seq data suggested that SPP1-CD44 signaling transduction might contribute to HSC activation. Pharmacological inhibition of GLI1 with the SMO inhibitor vismodegib suppressed HSC activation in vitro and reduced fibrosis and tumor growth in vivo. These findings indicate that SPP1 promotes intratumoral fibrosis and HCC progression through the SPP1-CD44-GLI1 axis, highlighting its potential as a prognostic biomarker and therapeutic target. Inhibition of SPP1-CD44-Hedgehog signaling may provide a promising strategy to mitigate fibrosis and improve HCC patient outcomes.

Clinical development of crizotinib in children with ALK-positive anaplastic large cell lymphoma (ALCL) was advanced in the US but not in Japan. The objectives of phase 1 part of a clinical study in Japan were to assess the safety and pharmacokinetics and determine the recommended phase 2 dose (RP2D) of crizotinib in children with relapsed/refractory (R/R) ALK-positive ALCL or neuroblastoma (NB). Two dose levels (165 and 280 mg/m² twice daily (BID)) were planned to be evaluated. The aim of phase 2 part of the study was to assess the antitumor activity of crizotinib at the RP2D in children with R/R ALK-positive ALCL. Seven patients were eligible for phase 1 part, 5 with ALCL and 2 with NB. All 7 patients received crizotinib at a starting dose of 165 mg/m² BID, with 1 dose-limiting toxicity observed (grade 3 gamma-glutamyltransferase increased). The most common grade 3 or 4 adverse event was neutropenia (71%). The steady-state C_max and AUC_tau of crizotinib at 165 mg/m² BID were higher than expected. Considering the risk of a greater incidence of severe neutropenia, we decided that the 165 mg/m² BID be the RP2D without evaluation at 280 mg/m² BID. A total of 11 patients with ALK-positive ALCL were enrolled in the phase 2 part. The objective response rate was 72.7% (90% CI, 43.6%-92.1%). One patient permanently discontinued crizotinib due to disease progression. Crizotinib at 165 mg/m² BID was well tolerated and showed favorable antitumor activity in children with R/R ALK-positive ALCL. Trial Registration: UMIN-CTR: UMIN000028075.

Early-onset colorectal cancer (EOCRC), diagnosed in patients under 50, is particularly aggressive, yet lacks targeted therapeutic strategies. This study aimed to explore the role of cellular senescence in driving EOCRC's malignancy and developed a senescence scoring system (EO-Senscore) to guide precision oncology. Through a multi-omics analysis of 2961 patients, we discovered that cellular senescence is more pronounced and varied in EOCRC and is not tied to chronological age. Two distinct senescence subtypes were identified: Cluster 1 (low-senescence tumors) showed prolonged survival, enhanced immunogenicity, and cell cycle activation, while Cluster 2 (high-senescence tumors) exhibited aggressive phenotypes and an immunosuppressive microenvironment. We further developed a machine learning model, the EO-Senscore, to quantify a tumor's senescence level. This score effectively stratified patients by prognosis and potential treatment response. Patients with a low EO-Senscore were predicted to respond well to immunotherapy and chemotherapy. In contrast, those with a high score had more invasive tumors but showed significant sensitivity to senolytic drugs (like ABT-263) in lab-based experiments. In conclusion, this research establishes cellular senescence as a crucial factor in EOCRC's aggressiveness. The EO-Senscore provides a practical, quantitative tool to guide clinical decisions, suggesting that patients could be directed toward immunotherapy or novel senolytic-based combination therapies for more personalized and effective cancer care.

Repeat transurethral resection of bladder tumor (re-TURBT) is commonly recommended for patients with non-muscle-invasive bladder cancer (NMIBC) with high-risk features, although many individuals may undergo unnecessary procedures that increase morbidity without clear benefit. Urinary tumor DNA (utDNA) has emerged as a promising biomarker, and we evaluated the performance of a multidimensional utDNA assay (utLIFE) in this context. In a retrospective cohort of 161 patients who underwent re-TURBT between May 2020 and May 2025, urine samples collected 2-6 weeks after initial TURBT were analyzed using utLIFE, which integrates shallow whole-genome sequencing and targeted sequencing of 155 genes with machine learning-based classification. Residual tumor was identified in 35% of patients, and utLIFE demonstrated high diagnostic accuracy, with sensitivity of 80.7%, specificity of 96.2%, and overall accuracy of 90.7%, markedly outperforming urine cytology. A positive utLIFE result was the strongest independent predictor of residual disease (OR = 77.5, 95% CI: 22.4-268.2, p < 0.001). During a median follow-up of 32.5 months, recurrence occurred in 20.5% of patients; recurrence was defined as urothelial tumor relapse within the urinary tract and did not include distant metastasis. utLIFE positivity was significantly associated with shorter recurrence-free survival (HR = 16.4, 95% CI: 2.7-101.8, p = 0.003). Longitudinal testing further revealed that utDNA positivity frequently preceded cystoscopic evidence of recurrence by several months. These findings highlight the strong diagnostic and prognostic value of utDNA in identifying residual disease and predicting recurrence, supporting its potential role in refining postoperative management strategies for NMIBC.

Artificial intelligence (AI) has transformed medical imaging, notably in radiology and endoscopy. Semantic segmentation, a pixel-level technique crucial for delineating pathological features, has become pivotal in digital pathology. Pathology segmentation AI models are often trained using annotations generated by pathologists. Despite the meticulous care typically exercised, pathologist-generated annotations often contain label noise whose types and effects on model training remain underexplored. This study combined a survey of public datasets with the synthesis of artificial label noise to evaluate its effects on pathology segmentation models. Using publicly available datasets and a breast cancer semantic segmentation dataset, modules were developed to simulate four types of artificial label noise at varying intensity levels. These datasets were used to train deep learning models and their performance was evaluated. The results indicated that models were highly susceptible to overfitting label noise, particularly boundary-dependent noise, such as dilation and shrinkage. Discrepancies were identified between apparent performance scores obtained under real-world conditions and true performance scores derived using clean test data. This overestimation risk was most pronounced for datasets containing boundary-altering noise. Furthermore, random noise combinations further degraded generalization. This study underscores the critical importance of addressing label noise in pathology datasets. It is proposed that future efforts focus on developing standardized methods for quantifying and mitigating label noise, along with creating robust benchmarks using noise-inclusive datasets. Enhancing annotation quality and addressing label noise can improve the reliability and generalizability of AI in pathology, facilitating broader clinical adoption.

Accurate primary tumor (pT) staging in muscle-invasive bladder cancer (MIBC) is crucial for treatment and prognosis. Current methods require time-consuming, labor-intensive microscopic evaluation by pathologists, with inherent interobserver variability. There is a need for AI-driven automated diagnosis of whole-slide images (WSI) to improve diagnostic efficiency while maintaining accuracy in pT staging. We obtained 281 H&E-stained WSI samples from the TCGA dataset for developing and validating a graph neural network (GNN)-based diagnostic model, and 83 additional samples from a hospital for external validation. The GNN method integrated multi-scale WSI data, evaluated using areas under the curve (AUC), accuracy, sensitivity, and specificity. A multi-scale attention mechanism was added to enhance model interpretability by capturing pT staging infiltration patterns. Diagnostic results were compared with those of three pathologists of varying expertise. We developed the multi-scale WSI-integrated GNN model for histopathological staging (T2/T3/T4) of MIBC. The model demonstrated excellent performance on external validation, achieving an AUC of 0.911 and an accuracy of 0.905. Interpretability analysis revealed distinct infiltration patterns for each T-stage, while diagnostic comparisons against both ground truth and three independent pathologists showed strong agreement, with a Cohen's kappa coefficient exceeding 0.876. The model developed based on graph neural network methods can integrate multi-scale information from whole-slide tissue images, allowing it to capture key infiltration patterns for muscle-invasive bladder cancer pT staging. This enables precise pT staging and visualizes the multi-scale tumor infiltration regions through attention scores, with accuracy showing strong consistency with expert pathologists.