Frontiers in Oncology最新文献

{"title":"New insights into tumor size in colorectal cancer: smaller size, higher risk?","authors":"Ruipeng Ding, Tianyu Li, Cong Li, Wang Song, Xianhua Gao, Midie Xu, Sheng Wang","doi":"10.3389/fonc.2026.1908104","DOIUrl":"10.3389/fonc.2026.1908104","url":null,"abstract":"<p><strong>Background: </strong>In colorectal cancer, the AJCC T category is defined by the depth of bowel wall invasion rather than by the absolute size of the primary tumor. The clinical value of tumor size in assessing tumor aggressiveness, metastatic risk, and prognosis remains insufficiently investigated.</p><p><strong>Aim: </strong>To evaluate the clinical value of tumor size in assessing potential metastatic risk and prognostic stratification in colorectal cancer.</p><p><strong>Methods: </strong>We retrospectively included 10,513 patients with colorectal cancer who underwent surgical treatment at FUSCC between 2008 and 2020. We integrated tumor size with clinicopathological characteristics to assess metastatic risk using machine-learning models and SHAP-based interpretation. By modeling tumor size as both a categorical and continuous variable, we further investigated its prognostic relevance using Kaplan-Meier survival analysis and Cox proportional hazards regression incorporating restricted cubic splines. Additionally, an independent cohort of 6,739 patients from Changhai Hospital was used for external validation.</p><p><strong>Results: </strong>Machine-learning identified lymph node metastasis as the primary predictive endpoint, with the CatBoost model achieving the highest AUROC of 0.800. SHAP analysis showed that smaller tumor diameter contributed positively to predicted metastatic risk. Multivariable restricted cubic spline analysis further demonstrated a significant nonlinear association between tumor diameter and OS, with mortality risk increasing as tumor diameter decreased within the approximately 0-3 cm range. Among younger male patients with locally advanced disease, tumors ≤3 cm were associated with significantly worse OS and DFS than tumors >5 cm (both <i>P</i> < 0.001). External validation yielded similar findings.</p><p><strong>Conclusion: </strong>Smaller tumor size may paradoxically indicate higher lymph node metastatic risk and poorer prognosis in a subset of patients with colorectal cancer, particularly younger male patients with locally advanced disease. Tumor size may therefore serve as a complementary factor for postoperative risk stratification beyond conventional TNM staging.</p>","PeriodicalId":12482,"journal":{"name":"Frontiers in Oncology","volume":"16 ","pages":"1908104"},"PeriodicalIF":3.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13671944/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149215487","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Case Report: Multiple radiotherapy courses for extensive recurrent cutaneous metastases from colorectal cancer.","authors":"Mehdi Belbachir, Nicolas Danthez, Cécile Torregrosa, Gian Guyer, Rémy Kinj","doi":"10.3389/fonc.2026.1910929","DOIUrl":"10.3389/fonc.2026.1910929","url":null,"abstract":"<p><p>We report the management of a 36-year-old man presenting with rapidly progressive and highly symptomatic cutaneous metastases (CMs) from metastatic colorectal cancer (mCRC). He was referred to palliative radiotherapy because of painful and pruritic skin lesions. His history included two synchronous colorectal adenocarcinomas. Comprehensive next-generation sequencing identified the rectosigmoid tumor as the metastatic source and revealed a pathogenic germline BRCA1 mutation, supporting a precision systemic strategy (olaparib plus pembrolizumab). CMs were the patient's dominant source of morbidity and progressively involved new skin sites; the patient therefore, received three sequential courses of helical tomotherapy (Radixact^®), each delivering 30 Gy in 10 fractions to extended superficial volumes exceeding 50 cm. Each course treated adjacent fields with limited high-dose overlap. Deformable co-registration with the prior plans and cumulative-dose review confirmed that the high-dose overlap was limited, distinguishing this approach from re-irradiation. A posterior skin strip was consistently spared to preserve lymphatic drainage and reduce the risk of compartment toxicity. All three courses produced a very good partial response (VGPR) at six weeks, with toxicity limited to grade 1 dermatitis (no grade ≥2). Crucially, repeated irradiation maintained durable symptom control, avoided escalation of opioid analgesia for approximately six months, and allowed continuation of the same systemic regimen for nine months until lung progression. This case illustrates that iterative skin-directed radiotherapy can deliver safe, repeatable, high-quality palliation in extensive CMs.</p>","PeriodicalId":12482,"journal":{"name":"Frontiers in Oncology","volume":"16 ","pages":"1910929"},"PeriodicalIF":3.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13672187/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149215783","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Self-management support interventions for patients with cancer treatment-related oral mucositis: a scoping review.","authors":"Han Yan, Xu Liu, Jiaying Li, Dongchi Ma, Wenjie Wang, Yuxing Xie, Li Wang, Li Ning","doi":"10.3389/fonc.2026.1950737","DOIUrl":"10.3389/fonc.2026.1950737","url":null,"abstract":"<p><strong>Objective: </strong>Cancer treatment-related oral mucositis (OM) is one of the most common adverse effects of cancer treatment and can substantially impair patients' quality of life. Its effective management depends heavily on patients' continuous self-management. This scoping review aimed to systematically map the intervention components, delivery modes, and existing limitations of self-management support interventions, thereby providing evidence to inform the optimization of intervention strategies.</p><p><strong>Methods: </strong>A scoping review was conducted in accordance with the PRISMA Extension for Scoping Reviews. Seven databases, including PubMed, Cochrane Library, Web of Science, CINAHL, Embase, China National Knowledge Infrastructure, and WanFang, were searched from inception to May 10, 2026. Two reviewers independently screened titles, abstracts, and full texts according to predefined inclusion and exclusion criteria. Data were charted using a standardized form and summarized using descriptive numerical and content analyses.</p><p><strong>Results: </strong>A total of 7,132 records were identified, and 16 studies were ultimately included, comprising 9 quasi-experimental studies, 6 randomized controlled trials, and 1 mixed-methods study. The interventions were primarily hospital-based and nurse-led, with some incorporating digital tools or written materials. Reported outcomes generally included OM-related clinical outcomes, quality of life, self-management-related outcomes, and nutritional status, although the direction and magnitude of findings varied across studies.</p><p><strong>Conclusions: </strong>Existing studies have reported potential improvements in OM-related and self-management-related outcomes following self-management support interventions. However, these findings should be interpreted cautiously because of heterogeneity in study designs, intervention components, and outcome measures, together with the absence of critical appraisal. Further well-designed studies using standardized outcome measures are needed to strengthen the evidence base.</p>","PeriodicalId":12482,"journal":{"name":"Frontiers in Oncology","volume":"16 ","pages":"1950737"},"PeriodicalIF":3.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13673534/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149215591","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A case report: axillary ALK-positive anaplastic large-cell lymphoma mimicking an infectious abscess in an adolescent.","authors":"Zhanwei Wang, Liuhong Yuan, Xudong Guan, Xiaohui Liu, Yuanting Song, Kun Tao, Tao Dai","doi":"10.3389/fonc.2026.1915361","DOIUrl":"10.3389/fonc.2026.1915361","url":null,"abstract":"<p><strong>Background: </strong>Abscess-like presentations are common in surgical practice, but malignant soft-tissue and lymphoid disorders can occasionally imitate infection and delay definitive treatment. Anaplastic large-cell lymphoma (ALCL) is a CD30-positive T-cell lymphoma that may involve skin, lymph nodes, or soft tissue and can show extensive necrosis, inflammatory change, and systemic symptoms.</p><p><strong>Case presentation: </strong>A 17-year-old female student presented with a 3-week history of progressive painful erythema and swelling of the left axilla after ineffective intravenous antibiotic treatment elsewhere. On admission, the lesion was large, indurated, tender, and clinically suspicious for a soft-tissue infection with possible abscess formation. Incision and drainage yielded only scant purulent fluid but abundant friable nodular tissue, and cultures were negative. Persistent fever, local progression, lymphadenopathy, and poor response to antimicrobial therapy prompted multidisciplinary reassessment. Histopathology of the solid tissue demonstrated large pleomorphic lymphoid cells, and immunohistochemistry showed strong CD30 and ALK expression, establishing ALK-positive ALCL. The patient was transferred to hematology and received initial cytoreductive therapy with brentuximab vedotin and short-course zeprumetostat, after which the local tumor burden decreased and delayed wound closure was performed. She subsequently received five cycles of brentuximab vedotin plus CHP and chidamide maintenance after complete remission.</p><p><strong>Conclusion: </strong>A presumed abscess that yields little pus but contains solid friable tissue should not be managed as infection alone. In refractory, culture-negative, or biologically discordant soft-tissue lesions, early biopsy of the cavity wall and solid components may help reduce diagnostic delay and redirect care toward malignancy-specific treatment.</p>","PeriodicalId":12482,"journal":{"name":"Frontiers in Oncology","volume":"16 ","pages":"1915361"},"PeriodicalIF":3.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13671954/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149215698","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A multimodal deep learning radiomics nomogram integrating clinicopathological features for predicting pathological complete response after neoadjuvant therapy in triple-negative breast cancer: a multicenter retrospective study.","authors":"Chuanchuan Qi","doi":"10.3389/fonc.2026.1893983","DOIUrl":"10.3389/fonc.2026.1893983","url":null,"abstract":"&lt;p&gt;&lt;strong&gt;Background: &lt;/strong&gt;Accurate noninvasive estimation of pathological complete response (pCR) after neoadjuvant therapy (NAT) in triple-negative breast cancer (TNBC) may support multidisciplinary discussions, but multimodal MRI-ultrasound evidence remains limited. We developed and externally validated a clinicopathological deep learning radiomics nomogram and explored its potential auxiliary value in a strictly hypothetical post-hoc surgical-planning scenario.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Methods: &lt;/strong&gt;With institutional review-board approval and waived consent, this retrospective multicenter diagnostic accuracy study included 436 biopsy-proven TNBC patients who underwent NAT and surgery: Center 1 training cohort (n = 268) and internal validation cohort (n = 86), plus historical cases from cooperating Centers 2-3 for external validation (n = 82). pCR was defined as ypT0/is ypN0. Pretreatment dynamic contrast-enhanced MRI (DCE-MRI), diffusion-weighted imaging (DWI)/apparent diffusion coefficient (ADC) maps, T2-weighted imaging (T2WI), and B-mode ultrasound were analyzed. Intratumoral and 5-mm peritumoral ring radiomics features were extracted and selected by LASSO. DenseNet-121 extracted modality-specific deep features, fused using multi-head self-attention. Age, tumor size, Ki-67, tumor-infiltrating lymphocytes (TILs), cT/N stage, and histological grade were integrated with imaging scores in a logistic-regression nomogram. Parameters and thresholds were locked after Center 1 development and applied once to external data. No model output was used clinically. Feature standardization and ComBat harmonization parameters were strictly locked to the training cohort before application to validation data.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Results: &lt;/strong&gt;pCR occurred in 182/436 patients (41.7%). The nomogram achieved areas under the receiver operating characteristic curve (AUCs) of 0.917 (95% CI, 0.881-0.947) in training, 0.883 (0.814-0.937) in internal validation, and 0.856 (0.773-0.921) in external validation, exceeding clinical (external AUC, 0.721), radiomics (0.786), and deep learning (0.819) models. External sensitivity, specificity, and accuracy at the locked threshold were 0.765, 0.813, and 0.793, respectively. Calibration was acceptable (slope, 0.93; intercept, -0.06; Brier score, 0.16). Decision curve analysis showed net benefit for thresholds of 0.20-0.70. Multi-threshold analysis revealed that a high-specificity threshold (0.65) yielded a positive predictive value of 0.85, potentially identifying a subset of patients with very high confidence of achieving pCR. Compared with the clinical model, external reclassification improved (continuous net reclassification improvement [NRI], 0.34; integrated discrimination improvement [IDI], 0.11; P = 0.003 and P = 0.008, respectively).&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Conclusion: &lt;/strong&gt;Among the few multicenter TNBC studies combining MRI, ultrasound, intratumoral/peritumoral radiomics, and attention-based deep learning, this nomogram show","PeriodicalId":12482,"journal":{"name":"Frontiers in Oncology","volume":"16 ","pages":"1893983"},"PeriodicalIF":3.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13671933/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149215822","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated genome-wide DNA methylation and transcriptomic analysis prioritizes <i>EPHA5</i> as a candidate non-invasive epigenetic biomarker for oral squamous cell carcinoma.","authors":"Semra Demokan, Begüm Özemek, Seval Turna, Sevde Cömert, Cömert Şen, Kağan Avcı, Ahmet Özkan, Sena Şen, Mehveş Poda, Murat Ulusan, Uğur Sezerman, Gülsüm Ak, Canan Alatlı, Nejat Dalay","doi":"10.3389/fonc.2026.1932336","DOIUrl":"10.3389/fonc.2026.1932336","url":null,"abstract":"<p><strong>Purpose: </strong>Oral squamous cell carcinoma (OSCC) remains a major clinical challenge because of delayed diagnosis and the lack of validated non-invasive biomarkers for OSCC detection. Epigenetic alterations during oral carcinogenesis represent promising biomarker candidates. This study aimed to prioritize clinically relevant epigenetic biomarkers and validate <i>EPHA5</i> using an integrated genome-wide DNA methylation and transcriptomic approach.</p><p><strong>Experimental design: </strong>Genome-wide DNA methylation and transcriptomic profiling were performed to identify epigenetically deregulated genes in OSCC. Candidate genes were prioritized by integrating DNA methylation, transcriptomic data and biological relevance. <i>EPHA5</i> was subsequently selected for validation and evaluated by quantitative methylation-specific PCR (QMSP) and quantitative real-time PCR (QRT-PCR) in tissue, saliva and serum samples from independent cohorts comprising patients with OSCC, oral premalignant lesions (OPML) and healthy controls.</p><p><strong>Results: </strong>Genome-wide analysis identified 661 upregulated and 577 downregulated genes together with 2,431 differentially methylated regions corresponding to 1,776 genes. Integrated analysis generated a panel of 51 candidate genes, from which <i>EPHA5</i> was prioritized using integrated DNA methylation, gene expression, pathway enrichment and biological relevance. <i>EPHA5</i> promoter hypermethylation was detected in 66.7% of OSCC tumors, whereas concurrent promoter hypermethylation and reduced expression were observed in 28.6% of tumors. Methylation was also detected in 54.5% of OPML lesions and in the saliva of 47.6% of OSCC patients but was rarely detected in healthy controls. Tumor <i>EPHA5</i> methylation significantly discriminated OSCC from healthy controls (AUC = 0.780, p = 0.004; sensitivity, 71.4%; specificity, 75%). Salivary <i>EPHA5</i> methylation also significantly discriminated OSCC from healthy controls (AUC = 0.717, p = 0.025; sensitivity, 52.4%; specificity, 87.5%), whereas serum methylation showed no significant diagnostic utility (AUC = 0.571, p = 0.462). Discrimination between OSCC and OPML was limited and did not reach statistical significance (tumor AUC = 0.662, p = 0.068; salivary AUC = 0.600, p = 0.264).</p><p><strong>Conclusions: </strong>Frequent tumor hypermethylation together with detectable salivary methylation supports further evaluation of <i>EPHA5</i> as a candidate non-invasive epigenetic biomarker for OSCC. However, its moderate diagnostic performance, limited discrimination between OSCC and OPML, and lack of external validation indicate that larger prospective studies and evaluation within multimarker methylation panels are required before clinical implementation.</p>","PeriodicalId":12482,"journal":{"name":"Frontiers in Oncology","volume":"16 ","pages":"1932336"},"PeriodicalIF":3.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13671974/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149215191","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Transcription factor-driven metabolic stratification reveals glycolytic and lipogenic subtypes with distinct prognostic and functional features in pancreatic cancer.","authors":"Jie Ke, Shuai Liu, Junming Xu, Fenfang Wu","doi":"10.3389/fonc.2026.1900370","DOIUrl":"10.3389/fonc.2026.1900370","url":null,"abstract":"<p><strong>Objective: </strong>Pancreatic ductal adenocarcinoma (PDAC) exhibits profound molecular and metabolic heterogeneity associated with distinct phenotypic states. However, the contribution of transcription factor (TF)-driven regulatory programs to metabolic subtype stratification remains poorly understood. This study aims to determine whether TF-based classification can identify prognostically relevant glycolytic and lipogenic PDAC subtypes.</p><p><strong>Methods: </strong>Bulk transcriptomic data from multiple PDAC cohorts and single-cell RNA sequencing data were integrated to characterize TF-associated metabolic heterogeneity. Metabolic signatures and TFs were quantified using single-sample gene set enrichment analysis (ssGSEA) and AddModule Score (AMS). Differential expression, enrichment analysis, integrative bulk and single-cell transcriptomic analysis, and deep neural network-based metabolic flux modeling were performed to investigate metabolic programs and functional differences between TF-defined subtypes. Additionally, RNA interference, RT-qPCR, lipid droplets staining, lactate assay, and transwell assays were used to evaluate the roles of TP63 in metabolic regulation.</p><p><strong>Results: </strong>Glycolytic and lipogenic TF-defined metabolic subtypes were identified across multiple independent PDAC cohorts. These subtypes exhibited distinct metabolic programs, differentiation states, and prognostic outcomes. The glycolytic TF-defined subtype was associated with worse prognosis and differentiation, whereas the lipogenic TF-defined subtype exhibited more favorable clinical features. Integration of public bulk and single-cell transcriptomic analyses, together with deep neural network-based metabolic flux modeling, revealed consistent metabolic heterogeneity at both cellular and functional levels. Notably, TP63 was identified as a potential regulator of glycolytic-lipogenic metabolic reprogramming in PDAC.</p><p><strong>Conclusions: </strong>TF-based classification delineates glycolytic and lipogenic subtypes in PDAC, underscoring the critical role of TF-regulated metabolic programs in patient stratification. These findings highlight the prognostic significance of glycolytic and lipogenic metabolic heterogeneity and support the development of metabolism-targeted therapeutic strategies.</p>","PeriodicalId":12482,"journal":{"name":"Frontiers in Oncology","volume":"16 ","pages":"1900370"},"PeriodicalIF":3.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13672013/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149215573","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Stage IA undifferentiated uterine sarcoma with SMARCA4 and SMARCA2 co-deficiency: a case report.","authors":"Rong Ji, Xialin Chen, Juan Lang, Zhongkui Xiong","doi":"10.3389/fonc.2026.1939651","DOIUrl":"10.3389/fonc.2026.1939651","url":null,"abstract":"<p><p>Uterine sarcomas represent a rare and heterogeneous group of malignancies whose management remains clinically challenging. They account for approximately 3%-5% of all uterine malignancies. Undifferentiated uterine sarcoma (UUS) is a subtype of uterine sarcomas. <i>SMARCA4</i> is a tumor suppressor gene that is aberrantly altered in approximately 5%-7% of human malignancies. <i>SMARCA4</i>-deficient UUS is a highly aggressive, undifferentiated mesenchymal neoplasm associated with a median overall survival (OS) of 7 months. During tumorigenesis, loss of function of either <i>SMARCA4</i> or <i>SMARCA2</i> is frequently observed; however, concurrent loss of both genes has been reported to be synthetically lethal and is therefore considered unexpected. In this case report, a 52-year-old Chinese woman with early-stage UUS and <i>SMARCA4</i> and <i>SMARCA2</i> co-deficiency underwent surgical resection followed by adjuvant chemotherapy and achieved a disease-free survival (DFS) of 32 months. This case suggests that patients with UUS harboring <i>SMARCA4</i> and <i>SMARCA2</i> co-deficiency may derive a survival benefit from early diagnosis and surgical resection. However, owing to the extreme rarity of UUS with co-deficiency of <i>SMARCA4</i> and <i>SMARCA2</i>, conducting a high-level phase III randomized controlled trial to validate this observation remains impractical.</p>","PeriodicalId":12482,"journal":{"name":"Frontiers in Oncology","volume":"16 ","pages":"1939651"},"PeriodicalIF":3.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13671925/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149215623","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning with interactive segmentation for risk stratification of cystic renal lesions on tri-phase CT: a multicenter study.","authors":"Wenlong Yuan, Pinglin He, Quanhao He, Xuan Zhang, Qing Jiang, Siwen Yin, Fajin Lv, Jie Liu, Weiyang He, Mingzhao Xiao","doi":"10.3389/fonc.2026.1832969","DOIUrl":"10.3389/fonc.2026.1832969","url":null,"abstract":"<p><strong>Background: </strong>Accurate preoperative risk stratification of cystic renal lesions (CRLs) remains a clinical challenge, primarily due to morphological overlap which often leads to the overtreatment of indolent cysts. To address this, we developed and validated an integrated deep learning framework that combines interactive segmentation with tri-phase CT feature integration to refine malignancy prediction.</p><p><strong>Methods: </strong>This retrospective multicenter study enrolled 434 patients with CRLs from four institutions, partitioned into training (n=195), validation (n=100), and external test (n=139) cohorts. The reference standard was histopathological diagnosis or ≥4 years of radiological stability. For volumetric segmentation, we implemented an interactive 'human-in-the-loop' Swin UNETR model. To integrate complementary information across the three CT phases, we then deployed a Tri-Phase-Channel (TPC) Perceiver architecture to integrate features extracted by parallel SEResNet50 backbones from unenhanced, corticomedullary, and nephrographic CT phases. Diagnostic performance was evaluated against the single-phase SEResNet50 models and the Bosniak classification version 2019 using the Area Under the Receiver Operating Characteristic Curve (AUC) and Decision Curve Analysis (DCA).</p><p><strong>Results: </strong>In the external test cohort, the interactive segmentation model yielded a Dice score of 0.890. For malignancy prediction, the TPC-Perceiver attained an AUC of 0.969 (95% CI: 0.911-1.000), significantly outperforming the Bosniak classification version 2019 (AUC 0.884; <i>p =0.016</i>). At the fixed probability threshold of 0.5, the model achieved a sensitivity of 94.4% and a specificity of 96.7%, compared with 100.0% and 76.9%, respectively, for the clinical standard. Furthermore, decision curve analysis indicated that the deep learning framework conferred greater net benefit across a broad range of threshold probabilities.</p><p><strong>Conclusion: </strong>The integrated framework showed favorable performance for cystic renal lesion risk stratification. Exploratory pairwise AUC comparisons favored the TPC Perceiver over the single-phase models and the clinical Bosniak v2019 assessment, while its AUC was comparable to that of the previous SETD model. The sensitivity-specificity profile observed at the fixed threshold suggests potential value in reducing false-positive risk stratification. Larger prospective multicenter studies are required before clinical implementation.</p>","PeriodicalId":12482,"journal":{"name":"Frontiers in Oncology","volume":"16 ","pages":"1832969"},"PeriodicalIF":3.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13671972/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149215757","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Association of dynamic changes in platelet-to-lymphocyte ratio during chemotherapy with survival in patients with non-small cell lung cancer.","authors":"Tingshan Xu, Zihang Wu, Jingshan Li, Lingxuan Wang, Baoqiang Weng, Luyang Chen, Xiaohui Chen, Qingyuan Zhu, Peng Liu, Fei He, Yuan Xu","doi":"10.3389/fonc.2026.1934583","DOIUrl":"10.3389/fonc.2026.1934583","url":null,"abstract":"<p><strong>Background: </strong>Inflammatory markers have been reported to be associated with outcomes in non-small cell lung cancer (NSCLC). However, whether early treatment-related changes in these markers provide additional prognostic information remains unclear. This study evaluated the association between dynamic changes in peripheral blood inflammatory markers during chemotherapy and survival outcomes in patients with NSCLC.</p><p><strong>Methods: </strong>We retrospectively analyzed 479 patients with NSCLC who received chemotherapy. Changes in lymphocyte, platelet, and neutrophil counts, as well as the neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio (PLR), were calculated as absolute differences (D-values) between baseline measurements obtained before the first chemotherapy session and post-chemotherapy measurements obtained within 7 days after the first session. Overall survival (OS) was assessed using Kaplan-Meier analysis and Cox proportional hazards models. Additional Cox models were used to examine the association of PLR D-value with OS after accounting for baseline PLR and clinical covariates. A prognostic nomogram was constructed as an exploratory risk-estimation model.</p><p><strong>Results: </strong>Greater changes in the evaluated hematological markers were associated with shorter OS in Kaplan-Meier analysis. In the primary multivariable Cox regression, high PLR D-value was associated with shorter OS (hazard ratio, 1.422; 95% confidence interval, 1.001-2.020; <i>P</i> = 0.049), together with sex, anemia status, and tumor extent. After adjustment for baseline PLR and clinical covariates, however, PLR D-value was no longer significantly associated with OS when analyzed either as a continuous variable (HR per 50-unit increase, 1.010; 95% CI, 0.947-1.078; <i>P</i> = 0.762) or using the original high/low classification (HR, 1.31; 95% CI, 0.91-1.89; <i>P</i> = 0.145). The nomogram had an apparent C-index of 0.630 and an optimism-corrected C-index of 0.618 after 1,000 bootstrap resamples.</p><p><strong>Conclusions: </strong>Greater changes in PLR during chemotherapy were associated with shorter OS in the primary categorical analysis. However, this association was attenuated after baseline PLR and clinical factors were taken into account, indicating that the additional prognostic contribution of PLR D-value may be limited. The nomogram should be considered exploratory and requires external validation before clinical application.</p>","PeriodicalId":12482,"journal":{"name":"Frontiers in Oncology","volume":"16 ","pages":"1934583"},"PeriodicalIF":3.4,"publicationDate":"2026-10-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC13672012/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149215795","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}