AI and Cancer Research ── 2026-10-12
New papers on cancer and AI, collected from journal feeds, PubMed (searched across all journals) and OpenAlex (a public index used instead of Google Scholar), judged by Jev (an AI that makes language judgements): original research, AI central, which of 13 themes, treatment or trials, and then ranked by importance up to a daily limit. Journals are not filtered by impact factor; each paper shows its journal's IF-equivalent (OpenAlex two-year mean citedness) and the date of that value. Each entry shows the full original abstract.
Read by cancer type: Cancer biology（1） · Breast（0） · Lung（1） · Pancreatic（1） · Prostate（2） · ACC（0） · Brain（2） · Ovarian（0） · Endometrial（0） · Gastric（0） · Liver（1） · Kidney（0） · Bladder（0）
Cancer molecular biology and AI
- Abstract
BACKGROUND AND PURPOSE: Magnetic resonance-guided radiotherapy (MRgRT) offers high-quality soft tissue contrast and treatment monitoring, supporting longitudinal radiomics analysis. We investigated treatment-related radiomic changes during stereotactic body radiotherapy (SBRT) for liver metastases and healthy liver. MATERIALS AND METHODS: Features were extracted from the gross tumour volume (GTV) and five 2-cm liver rings around the planning target volume (PTV) in 34 patients treated with five-fraction MRgSBRT. Features showing changes below baseline variability or high correlation (Spearman's |r| > 0.9) were excluded, with pre-treatment values averaged into a robust baseline. Linear regression slopes informed uniform manifold approximation and projection (UMAP) followed by hierarchical density-based spatial clustering of applications with noise (HDBSCAN) for patient stratification, with false discovery rate-corrected Kruskal-Wallis testing identifying representative features. Correlations with GTV volume and mean cumulative liver-ring dose were evaluated. RESULTS: Of 160 GTV features, 55 uncorrelated features remained. Three patient groups ( n = 4, 10, and 20) with distinct longitudinal patterns were identified. Intensity-based energy showed significant differences across patient groups (adjusted p = 0.001), weakly correlated (|r| volume in the two larger groups. In the innermost liver ring, 49 uncorrelated features were analysed. Two patient groups ( n = 19 and 15) were identified, with intensity-based energy , mean intensity , and maximum intensity showing significant group differences, weakly correlated with mean cumulative dose. CONCLUSIONS: This longitudinal MR-based radiomics framework identified GTV and healthy liver patient groups, with distinct treatment-related feature changes. This pipeline could be extended to integrate treatment response assessment in future studies.
Journal IF-equivalent: 1.9 (OpenAlex 2-year mean citedness, value as of 2026-10-11, retrieved 2026-10-12; not the official Clarivate IF)Reference: Paunoiu A, Fritsak M, Gabryś HS, Christ SM, Guckenberger M, Tanadini-Lang S, et al. Longitudinal radiomics of liver metastases during magnetic resonance-guided radiotherapy. Phys Imaging Radiat Oncol. 2026 Sep;41:101094. doi:10.1016/j.phro.2026.101094; PMCID: PMC13651652.Checked: Full text checked
See all 1 Cancer biology papers →
Breast cancer: treatment, trials and AI
No new qualifying paper for this issue.
Lung cancer: treatment, trials and AI
- Abstract
BACKGROUND: Lung cancer remains a leading cause of cancer-related mortality, and postoperative risk stratification may support individualized surveillance after pulmonary resection. This study aimed to develop and internally validate an exploratory prediction model for 3-year postoperative mortality using routinely available clinical, pathological, and laboratory variables, while explicitly addressing outcome ascertainment, model calibration, and the limitations of a single-center retrospective cohort. METHODS: LASSO regression was used for feature selection, followed by a comparison of seven algorithms: Logistic Regression, Decision Tree, Random Forest, Support Vector Machine, Artificial Neural Network, XGBoost, and LightGBM. Model discrimination, calibration, and decision-curve performance were evaluated in a 7:3 internal hold-out split. The final model was selected primarily according to an internal-validation AUC, with calibration and classification metrics considered jointly. A web-based calculator and a nomogram were developed from the final model. RESULTS: In the final outcome-ascertained analytic sample, 166 of 350 patients (47.43%) were classified as having died within 36 months and 184 (52.57%) were alive beyond 36 months. This proportion describes the event composition of the selected analytic cohort and should not be interpreted as a population-level postoperative mortality estimate. Six predictors were retained by LASSO: T stage, airway dissemination, pleural invasion, Ki-67 index, BMI, and direct bilirubin. Random Forest showed perfect apparent discrimination in the training set (AUC = 1.00) but decreased to 0.73 in validation. Logistic regression achieved the highest validation AUC (0.79; 95% CI, 0.70-0.87), with an accuracy of 0.75, an F1 score of 0.73, and a Brier score of 0.187, and was therefore selected as the final model. CONCLUSION: The final logistic regression model provided moderate internal discrimination and an interpretable framework for exploratory 3-year mortality risk estimation after lung cancer resection. The nomogram and web-based calculator facilitated model visualization, but the outcome-ascertained analytic cohort, potential event enrichment, and absence of external validation limited the interpretation of absolute risk and direct clinical application. Therefore, an independent multicenter and prospective validation with appropriate recalibration is required.
Journal IF-equivalent: 2.6 (OpenAlex 2-year mean citedness, value as of 2026-10-08, retrieved 2026-10-09; not the official Clarivate IF)Reference: Liu C, Wu Y, Wang J, Lei K, Xing Y, Kang M, et al. Development and internal validation of a machine-learning model for 3-year mortality after lung cancer resection: an online risk calculator and nomogram. Front Med (Lausanne). 2026 Sep 25;13:1815915. doi:10.3389/fmed.2026.1815915; PMCID: PMC13649404.Checked: Full text checked
Pancreatic cancer: treatment, trials and AI
- Abstract
Pancreatic ductal adenocarcinoma (PDAC) exhibits profound stromal heterogeneity that influences tumor progression, recurrence risk, and therapeutic response; however, stromal phenotypes are currently assessed by postoperative histopathology. Here, we developed a super-resolution MRI (srMRI)-based radiomics model to enable noninvasive tumor-stroma ratio (TSR) phenotyping in PDAC. Using multicenter cohorts comprising 737 patients, we constructed a srMRI-derived TSR-radscore capable of predicting pathological TSR with high accuracy (AUC = 0.870 and 0.791 in training and external validation cohorts). The TSR-radscore independently stratified recurrence-free survival and, when integrated with clinicopathological variables, improved early recurrence prediction in an independent surgical cohort. In a locally advanced PDAC cohort, srMRI-derived TSR was associated with distinct therapeutic outcomes, with H-TSR tumors showing improved survival under systemic therapy and an association with outcomes following immune checkpoint inhibitor-based combination therapy. Transcriptomic analyses revealed that L-TSR tumors were enriched in extracellular matrix remodeling and TGF-β-related signaling pathways. Single-cell profiling further uncovered distinct cancer-associated fibroblast and T-cell states across TSR phenotypes, linking imaging-derived TSR patterns to underlying stromal and immune microenvironmental programs. Collectively, srMRI-derived TSR phenotyping provides a noninvasive and biologically interpretable biomarker for stromal characterization in PDAC, enabling improved recurrence-risk stratification and pre-treatment clinical assessment.
Journal IF-equivalent: 10.8 (OpenAlex 2-year mean citedness, value as of 2026-10-11, retrieved 2026-10-12; not the official Clarivate IF)Reference: Wang X, Xie Z, Li H, Tang Q, Zhou J, Huang J, et al. Super‐Resolution MRI–Derived Tumor‐Stroma Ratio Phenotyping for Prognostic and Therapeutic Stratification in Pancreatic Ductal Adenocarcinoma. MedComm (2020). 2026 Oct 9;7(10):e71039. doi:10.1002/mco2.71039; PMCID: PMC13650606.Checked: Full text checked
Prostate cancer: treatment, trials and AI
- Abstract
PURPOSE: To systematically review MRI-derived radiomic and artificial intelligence models predicting outcomes in men on or eligible for active surveillance (AS) for prostate cancer, and whether reported endpoints permit quantitative synthesis. METHOD: Following PRISMA 2020, we searched PubMed, Scopus, Embase, and Web of Science (2000-September 2026). Two reviewers independently screened, extracted data, and assessed risk of bias (QUADAS-2, PROBAST) and methodological quality (METRICS). Three subgroups were defined: AS selection at diagnosis, baseline detection of clinically significant cancer, and progression during follow-up. Pooling was pre-specified for subgroups with at least four independent cohorts; endpoint equivalence was assessed. RESULTS: Seventeen studies across 12 unique cohorts were included. Seven publications derived from two cohorts; 12 independent cohort-level estimates (n = 2703) remained after overlap resolution. Nine publications (four independent cohorts) addressed progression (AUC 0.70 to 0.95), but progression was defined in five non-equivalent ways, from any grade-group increase to composite core-count criteria, precluding quantitative synthesis. Risk of bias was high or unclear in 16 of 17 studies; the median METRICS score was 32.3 % (range 18.9-52.3 %). No study reported external validation or calibration for progression. CONCLUSIONS: Progression endpoints were heterogeneous and several were of limited clinical relevance. A core outcome set for AS (at least ISUP grade group ≥3 and/or unequivocal radiological progression), followed by multi-centre external validation with a head-to-head comparison against the PRECISE v2 (Prostate Cancer Radiological Estimation of Change in Sequential Evaluation) score, is the priority before clinical use can be considered.
Journal IF-equivalent: 3.6 (OpenAlex 2-year mean citedness, value as of 2026-10-11, retrieved 2026-10-12; not the official Clarivate IF)Reference: Kyle ET, Sweeney KJ, McCabe DH, O’Sullivan NJ, Corr A, Sheehy N, et al. MRI-derived radiomics for prostate cancer active surveillance: a systematic review. European Journal of Radiology. 2026 Dec;205:113291. doi:10.1016/j.ejrad.2026.113291.Checked: Abstract only - Abstract
Foundation models are trained on massive amounts of data to capture complex patterns in images. Subsequently, a wide range of downstream tasks can be adopted with minimal computational resources. We have developed HistoEncoder, a foundation model for prostate cancer digital pathology by pre-training on 48 million prostate tissue tile images. HistoEncoder allows automated extraction of histological features highly predictive of Gleason patterns achieving comparable performance with substantially larger pan-cancer foundation models while being much more efficient. By fine-tuning the model with a small amount of data and computational resources, we describe two clinical use cases for HistoEncoder. First, HistoEncoder can be used to automatically annotate large-scale datasets with high accuracy. Second, we show that HistoEncoder-derived histology clusters contain prognostic information of a similar magnitude to Gleason grading in internal cross-validation. Lightweight foundation models such as HistoEncoder allow organizations to build effective clinical software tools without the need for extensive datasets and heavy computing.
Journal IF-equivalent: 3.3 (OpenAlex 2-year mean citedness, value as of 2026-10-11, retrieved 2026-10-12; not the official Clarivate IF)Reference: Pohjonen J, Batouche O, Kantola J, Rannikko A, Sandeman K, Erickson A, et al. HistoEncoder: A digital pathology foundation model for prostate cancer. Journal of Pathology Informatics. 2026 Nov;23:100715. doi:10.1016/j.jpi.2026.100715.Checked: Abstract only
Adenoid cystic carcinoma: treatment, trials and AI
No new qualifying paper for this issue.
Brain tumours: treatment, trials and AI
- Abstract
BACKGROUND: Molecular features are fundamental to both the diagnosis and prognosis of gliomas. Specific molecular features are generally unavailable at the time of index surgery; however, upfront knowledge of molecular characteristics could meaningfully inform surgical strategy. AIMS/OBJECTIVES: To determine the feasibility of a machine learning model trained on fluorescence spectral features to classify IDH mutation and MGMT promoter methylation status in glioma tissue. METHODS: Tissue samples were collected during 5-ALA-guided surgery and interrogated with a fibre probe to record fluorescence spectra. The fluorescence data were blinded and randomised and were trimmed to include only relevant wavelengths. Molecular data (IDH status and MGMT methylation) were recorded for each tumour specimen and incorporated into a machine learning algorithm. The algorithm was trained on the fluorescence data of half the patients exhibiting each label, and the model was then tested on the other half. RESULTS: Twenty-seven patients were recruited to the study, in whom over 8000 spectra were measured. IDH-mutant samples were identified with 87% accuracy. The algorithm was unable to reliably classify MGMT methylation. CONCLUSIONS: In this proof-of-concept study, machine learning applied to fluorescence spectra shows promise for real-time intraoperative prediction of IDH status. Further work integrating IDH and 1p/19q status in larger cohorts is warranted.
Journal IF-equivalent: 0.9 (OpenAlex 2-year mean citedness, value as of 2026-10-11, retrieved 2026-10-12; not the official Clarivate IF)Reference: Dablouk MO, Buckley K, Faul S, O’Sullivan MGJ. Intraoperative fluorescence spectroscopy for IDH classification in glioma: a feasibility study. British Journal of Neurosurgery. 2026 Oct 9 [Epub ahead of print]. doi:10.1080/02688697.2026.2746372.Checked: Abstract only - Abstract
BACKGROUND AND PURPOSE: Segmentation of the cerebral ventricular system (VS) is central to radiotherapy, both for whole-ventricular irradiation in intracranial germ cell tumors (GCTs) and protection of the periventricular region. Complex, variable VS anatomy makes manual delineation challenging and hampers auto-segmentation model development. We reported the evaluation and workflow integration of a deep-learning (DL) model for VS and brainstem segmentation on pediatric magnetic resonance imaging (MRI). MATERIALS AND METHODS: An nnU-Net was trained on MRIs from 76 unique brain tumor cases ( n = 73/71 for T1/T2). The evaluation used two independent retrospective cohorts: (i) GCTs evaluated on planning computed tomography (CT) images after MRI-to-CT registration mapping ( n = 13/14), and (ii) brain tumors evaluated on MRI ( n = 14). Performance was assessed using the Dice similarity coefficient (DSC) and Hausdorff distances (HD) against manual contours. Prospective deployment in routine clinical planning assessed usability and workflow integration. RESULTS: In the GCT-CT cohort, brainstem segmentations achieved median DSCs of >0.85 and HDs 95 of ≤5.0/6.4 mm, while VS segmentations achieved DSCs of >0.70 and HDs 95 of ≤5.3/4.3 mm, after rigid/deformable mapping to planning CT. In the MR cohort, performance on T1-/T2-weighted MRI reached DSCs of ≥0.90 for both structures, and HDs 95 of ≤3.0/3.4 mm and ≤ 1.3/2.2 mm for the brainstem and VS, respectively. The DL-segmentation workflow reduced workflow time by >50% compared with manual contouring. CONCLUSIONS: Lower performance in the GCT-CT cohort likely reflects noisier CT-based references, interobserver variability, and uncorrected geometric/anatomical mismatch. The model's accuracy supports radiotherapy planning in pediatric brain tumors through risk-structure delineation and target-volume segmentation in GCTs.
Journal IF-equivalent: 1.9 (OpenAlex 2-year mean citedness, value as of 2026-10-11, retrieved 2026-10-12; not the official Clarivate IF)Reference: Heinzelmann F, Hörst F, Heine L, Fragemann J, Rempe M, Luijten G, et al. Deep learning segmentation of the cerebral ventricular system and brainstem for pediatric radiotherapy planning. Phys Imaging Radiat Oncol. 2026 Sep;41:101086. doi:10.1016/j.phro.2026.101086; PMCID: PMC13651686.Checked: Full text checked
Ovarian cancer: treatment, trials and AI
No new qualifying paper for this issue.
Endometrial cancer: treatment, trials and AI
No new qualifying paper for this issue.
Gastric cancer: treatment, trials and AI
No new qualifying paper for this issue.
Liver cancer: treatment, trials and AI
- Abstract
BACKGROUND: Nuclear β-catenin accumulation reflects activation of the Wnt/β-catenin pathway in hepatocellular carcinoma (HCC), but noninvasive preoperative prediction of this immunohistochemical phenotype remains incompletely established. PURPOSE: To develop and externally validate CT radiomics and topology-based intratumoral heterogeneity (ITH) models for preoperative prediction of nuclear β-catenin expression in HCC. MATERIALS AND METHODS: This multicenter retrospective study included 1029 patients with histopathologically confirmed HCC. Patients from two participating institutions formed the development cohort and were split into training (n=620; nuclear β-catenin-positive, 293) and internal validation (n=267; nuclear β-catenin-positive, 126) sets. Patients from an independent third institution formed the external validation cohort (n=142; nuclear β-catenin-positive, 43). Whole-tumor portal venous phase CT radiomics features and topology-based ITH features were extracted from the audited image-mask cohort. Model discrimination, calibration, and clinical utility were evaluated using AUC, bootstrap 95% confidence intervals, calibration summaries, Brier scores, and decision curve analysis. RESULTS: In external validation, AUCs were 0.677 for the clinical model, 0.862 for the radiomics model, 0.857 for the ITH model, 0.831 for the core ITH model, 0.863 for the radiomics-ITH model, and 0.886 for the stacking model. Compared with radiomics alone, stacking showed a numerical ΔAUC of 0.024 (95% CI, -0.010 to 0.060; P=0.180). CONCLUSION: CT radiomics and topology-based ITH features showed external predictive value for nuclear β-catenin expression in HCC. The stacking model had the highest external AUC, but its improvement over radiomics alone was not statistically significant. If prospectively validated, these CT-based approaches could complement tissue assessment for pretreatment biological stratification.
Journal IF-equivalent: 2.5 (OpenAlex 2-year mean citedness, value as of 2026-10-11, retrieved 2026-10-12; not the official Clarivate IF)Reference: Tan Y, Xu S, Liu X, Liao W, Zhu R, Shen M, et al. CT radiomics and topology-based intratumoral heterogeneity for preoperative prediction of nuclear β-catenin expression in hepatocellular carcinoma. Abdom Radiol. 2026 Oct 10 [Epub ahead of print]. doi:10.1007/s00261-026-05788-4.Checked: Abstract only
Kidney cancer: treatment, trials and AI
No new qualifying paper for this issue.
Bladder cancer: treatment, trials and AI
No new qualifying paper for this issue.