AI and Cancer Research ── 2026-11-19
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.
All cancers: Cancer biology(0)
Hematologic malignancies: AML(0) · CML(0) · ALL(0) · CLL(0) · Lymphoma(0) · Myeloma(0)
Solid tumors: Breast(2) · Lung(2) · Colorectal(2) · Pancreatic(1) · Prostate(0) · Brain(2) · Ovarian(0) · Endometrial(0) · Gastric(1) · Liver(2) · Kidney(0) · Bladder(0)
Rare cancers: ACC(0)
Cancer molecular biology and AI
No new qualifying paper for this issue.
Hematologic malignancies
Acute myeloid leukemia (AML): treatment, trials and AI
No new qualifying paper for this issue.
Chronic myeloid leukemia (CML): treatment, trials and AI
No new qualifying paper for this issue.
Acute lymphoblastic leukemia (ALL): treatment, trials and AI
No new qualifying paper for this issue.
Chronic lymphocytic leukemia (CLL): treatment, trials and AI
No new qualifying paper for this issue.
Lymphoma: treatment, trials and AI
No new qualifying paper for this issue.
Multiple myeloma: treatment, trials and AI
No new qualifying paper for this issue.
Solid tumors
Breast cancer: treatment, trials and AI
- Abstract
Accurate risk stratification is essential for guiding treatment decisions and postoperative management in breast cancer. However, unimodal models are limited in their ability to fully exploit complementary information from clinical, radiological, and histopathological data associated with metastatic progression. We aimed to develop and validate Breast Cancer Bone Metastasis Network (BCBM-Net), a transformer-based multimodal deep learning network for predicting postoperative bone metastasis in patients with estrogen receptor (ER)-positive, human epidermal growth factor receptor 2 (HER2)-negative breast cancer. In this multi-institutional retrospective study, 1,240 patients from five clinical centers were included in training, internal test, and independent external test cohorts. BCBM-Net integrated selected clinicopathological variables, ultrasound (US)-derived features, and pathomics features extracted from whole-slide images (WSIs). Modality-specific fully connected encoders were used to learn nonlinear representations and project heterogeneous features into a shared latent space, followed by adaptive fusion using a multi-head attention transformer. Model performance was evaluated using receiver operating characteristic analysis, DeLong tests, and net reclassification improvement (NRI) analysis and was compared with clinical, US-based, and pathomics unimodal models. BCBM-Net achieved consistently higher discriminative performance than unimodal models across all cohorts, with area under the receiver operating characteristic curve (AUC) of 0.873, 0.866, and 0.855 in the training, internal test, and external test sets, respectively. DeLong tests and NRI analyses further supported the added predictive value of multimodal fusion. BCBM-Net effectively integrates routinely available clinical, US, and histopathological information to provide individualized assessment of 5-year postoperative bone metastasis risk. Its consistent performance in independent external centers supports the value of transformer-based multimodal fusion for cancer prognostication. BCBM-Net could provide a practical framework for risk-adapted surveillance and postoperative management, although prospective and multinational validation is required before clinical implementation.
Journal IF-equivalent: 7.9 (OpenAlex 2-year mean citedness, value as of 2026-11-03, retrieved 2026-11-05; not the official Clarivate IF)Reference: Zhang Z, Huang G, Deng Z, Ouyang L, Wu Z, Zhang K, et al. Transformer-based multimodal deep learning network for predicting postoperative bone metastasis in ER-positive/HER2-negative breast cancer: a multi-institutional retrospective study. BMC Med. 2026 Nov 16 [Epub ahead of print]. doi:10.1186/s12916-026-05352-3.Checked: Abstract only - Abstract
HER2-low breast cancer has emerged as a distinct subtype with unique biological characteristics. Habitat imaging can characterize intratumoral heterogeneity by partitioning tumors into subregions with distinct imaging phenotypes. However, the comparative prognostic value of habitats derived from diffusion-weighted imaging (DWI) and dynamic contrast-enhanced MRI (DCE-MRI) remains unexplored. This study aimed to develop and compare DWI and DCE habitat radiomics models for predicting disease-free survival (DFS) in HER2-low breast cancer. A total of 401 patients from four centers were retrospectively included. DWI and DCE images were independently clustered into three habitats using K-means, relabeled by signal intensity as low (L), medium (M), and high (H). From each habitat, 107 three-dimensional radiomics features were extracted. LASSO-Cox regression with 5-fold cross-validation was used for model construction. Model performance was assessed by C-index, time-dependent AUC, Brier score, and log-rank test. Among pure imaging models, the DCE high-signal habitat (DCE_H) demonstrated the best performance (C-index = 0.642, P = 0.033). When combined with clinical variables, the DCE_H model achieved modestly higher time-dependent AUC on nested cross-validation (3-year AUC 0.699→0.720; 5-year AUC 0.728→0.754), although this incremental difference did not reach statistical significance in paired comparison ( P = 0.056 and 0.080, respectively). Standard multivariable Cox regression suggested that DCE_H Sphericity was associated with DFS (hazard ratio [HR] = 0.067, 95% confidence interval [CI]: 0.010–0.520, P = 0.0096). In contrast, DWI habitat features were not selected by LASSO when clinical variables were present. DCE high-signal habitat radiomics—particularly Sphericity—was associated with DFS in HER2-low breast cancer, although its incremental value over clinical variables was modest and not statistically significant. DWI features added little. External validation is required. The study was retrospectively registered in the Chinese Clinical Trial Registry (ChiCTR2400085188, registered June 03, 2024, http//www.chictr.org.cn/).
Journal IF-equivalent: 4.8 (OpenAlex 2-year mean citedness, value as of 2026-11-12, retrieved 2026-11-13; not the official Clarivate IF)Reference: Meng M, Liu C, Peng Y, Wang F, Huang R, Zhu M, et al. MRI habitat radiomics with clinical features for prognosis prediction in HER2-low breast cancer: a multicenter study. BMC Cancer. 2026 Nov 16 [Epub ahead of print]. doi:10.1186/s12885-026-17380-6.Checked: Abstract only
Lung cancer: treatment, trials and AI
- Abstract
AbstractAI models using medical images are increasingly proposed to support clinical decision-making in lung cancer, motivating assessment of their methodological rigor and transparency. To assess how these models are developed, evaluated, and reported, and to identify gaps in best statistical practices, we conducted a cross-sectional meta-research review of OpenAlex-indexed studies (January 1, 2023, to June 30, 2025) that developed image-based AI tools for lung cancer risk prediction, detection, or prognosis. Thirty-six studies met our inclusion criteria. Convolutional neural networks were used in most studies (27 studies, 75%), and area under the curve (AUC) was the most frequently reported evaluation metric (29 studies, 81%). Although 23 studies (64%) reported some form of uncertainty quantification, most were limited to presenting confidence intervals for global performance metrics (e.g., AUC, sensitivity, and accuracy) rather than individual prediction- or risk-level uncertainty. Common limitations included limited external test set use (14 studies, 39%), insufficient subgroup analyses (10 studies, 28%), and limited adherence to established prediction-model reporting guidelines. Overall, future research should prioritize standardized reporting, use of external test sets, and model performance assessment across clinically relevant subpopulations. The coexistence of increasingly powerful AI-based prediction models with inconsistent study design, evaluation, and reporting practices emphasizes the urgent need to implement more rigorous and transparent methodological standards in this field. (Funded by Yale School of Public Health.).
Journal IF-equivalent: 27.2 (OpenAlex 2-year mean citedness, value as of 2026-11-16, retrieved 2026-11-18; not the official Clarivate IF)Reference: Hou Y, Ward T, Yang CH, McGrath S, Jernigan E, Caturegli G, et al. A Meta-Research Review of Statistical Practices for Image-Reliant AI Tools Predicting Lung Cancer Outcomes. NEJM AI. 2026 Nov 16 [Epub ahead of print]. doi:10.1056/aira2600064. PMID: 43037489.Checked: Abstract only - Abstract
RATIONALE: Chronic obstructive pulmonary disease (COPD) affects 40-60% of non-small cell lung cancer (NSCLC) patients but remains frequently underdiagnosed and undertreated, contributing to adverse outcomes [1]. METHODS: Our objective was to develop an automated machine learning (ML) classifier to identify potential COPD, and potentially untreated COPD among NSCLC patients. We performed a retrospective cohort study of 4,604 NSCLC patients. Manual chart review (development set; DS; [n = 600]; test set; TS; [n = 294]), using Global Initiative for Obstructive Lung Disease (GOLD) criteria, served as the reference standard for COPD diagnosis. Predictor selection used least-absolute-shrinkage-and-selection-operator (LASSO) logistic regression. Two logit thresholds were chosen to create binary classifiers: a ≥ 95% specificity quality-improvement threshold (QIT) for COPD identification, and Youden's index threshold (YT) for survival analysis and association with radiographic emphysema. RESULTS: The classifier achieved an area-under-the-receiver-operating-characteristic-curve of 0.93 in the DS and TS (95% CI:0.91-0.95; 95% CI:0.91-0.96, respectively). The QIT yielded logit cutoffs ≥ 0.9242 (DS) and ≥ 0.9625 (TS), with sensitivities 69% and 72%, specificities 95% and 96%, positive-predictive-value (PPV) 93%, and negative-predictive-value (NPV) 77% and 80%, respectively. In the DS and TS, potential COPD prevalence was 48% and 46%, with 41% and 46% potentially untreated; the QIT-classifier identified 69% and 72% of potential COPD cases and 40% and 50% of potentially untreated cases, respectively (PPV 85%, 95% CI:73-93% and 86%, 95% CI:71-95%). YT yielded a sensitivity and specificity of 88%. In Cox-proportional hazards models (n = 4,501, 1,959 deaths), the YT-COPD classifier was associated with increased mortality, with significant effect measure modification by clinical stage (p = 0.003; stratified HR YT-COPD: stage I: 1.71, 95% CI:1.30-2.44, p < 0.001; Stage II: 1.72, 95% CI:1.15-2.59, p = 0.009; Stage III: 1.67, 95% CI:1.34-2.08, p < 0.001; Stage IV: 1.29, 95% CI:1.15-1.44, p < 0.001). Furthermore, classifier-predicted COPD (YT) was strongly associated with radiographic emphysema (p < 0.001). CONCLUSION: This ML classifier effectively emulates structured chart review to identify potential COPD and potentially untreated COPD in NSCLC, with prognostic validity and potential for scalable, low-cost quality improvement and research. External validation in other institutions and EHR systems is warranted before broader implementation.
Journal IF-equivalent: 3.4 (OpenAlex 2-year mean citedness, value as of 2026-11-11, retrieved 2026-11-12; not the official Clarivate IF)Reference: Halawa AR, Robles SE, Ferguson J, Young C, Sainz PV, Lu SC, et al. Development of an automated machine learning classifier for chronic obstructive pulmonary disease in non-small cell lung cancer patients using electronic health records. BMC Pulm Med. 2026 Aug 21;26(1):573. doi:10.1186/s12890-026-04448-z. PMID: 43040280.Checked: Abstract only
Colorectal cancer: screening, treatment, trials and AI
- Abstract
Homologous recombination deficiency (HRD), caused by impaired homologous recombination repair (HRR), represents a clinically relevant molecular phenotype associated with genomic instability and therapeutic vulnerability. However, current HRD assessment relies mainly on tissue-based molecular assays, limiting its application in routine clinical practice. This study aimed to develop and validate a computed tomography (CT)-based deep radiomics signature (DRS) for noninvasive prediction of HRD status and to investigate its prognostic value and its association with outcomes after adjuvant chemotherapy in colorectal cancer (CRC). This multicenter retrospective study included 2065 patients with CRC from three hospitals in China. Among the 210 patients used for model training, 73 were classified as HRD and 137 as homologous recombination proficient (HRP). Radiomic and deep learning features were extracted from preoperative CT images. After feature selection, seven machine learning models were developed and compared. The optimal model was used to establish the DRS, which was further evaluated for HRD prediction and survival stratification in independent cohorts. Exploratory observational analyses examined whether associations between adjuvant chemotherapy and survival differed by DRS classification. Transcriptomic analysis was performed to explore the biological characteristics associated with the DRS-defined HRD phenotype. The DRS model demonstrated favorable discrimination for HRD status in the training, internal validation, and external validation cohorts, with AUCs of 0.873, 0.858, and 0.826, respectively. DRS classification effectively stratified overall survival (OS) and disease-free survival (DFS) across all cohorts. Multivariable Cox regression analysis showed that DRS remained independently associated with prognosis in CRC patients (all P < 0.001). Among patients with stage IIB-III CRC, adjuvant chemotherapy was associated with improved survival in the DRS-HRD subgroup, whereas no similar association was observed in the DRS-HRP subgroup. Transcriptomic analysis further showed that genes upregulated in DRS-HRD tumors were enriched in pathways related to cell cycle, DNA replication, and homologous recombination. The CT-based DRS noninvasively predicted the study-specific reference HRD label and stratified prognosis. In exploratory observational analyses, associations between adjuvant chemotherapy and survival differed by DRS group among patients with stage IIB-III disease.
Journal IF-equivalent: 8.6 (OpenAlex 2-year mean citedness, value as of 2026-11-03, retrieved 2026-11-05; not the official Clarivate IF)Reference: Lei X, Wang R, Li Z, Zhang R, Chen B, Zhou K, et al. Noninvasive imaging-based assessment of homologous recombination deficiency for risk stratification and adjuvant chemotherapy guidance in colorectal cancer: a multicenter study. J Transl Med. 2026 Nov 16 [Epub ahead of print]. doi:10.1186/s12967-026-09237-0.Checked: Abstract only - Abstract
Chemoresistance remains a major obstacle in colorectal cancer (CRC) treatment. In this study, we performed an integrative multi-omics analysis of The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) datasets, combined with machine learning approaches, to systematically investigate the UFMylation pathway in CRC. The E2 conjugating enzyme UFC1 was identified as a key determinant of chemotherapy sensitivity, with copy number amplification and elevated expression in tumors, particularly in chemotherapy-responsive patients. Single-cell transcriptomics revealed UFC1 enrichment in malignant epithelial cells, and its expression correlated with DNA damage response and apoptosis pathways. A LASSO-Cox risk model incorporating UFM1 pathway genes effectively stratified patient prognosis. Based on the UFC1 crystal structure, structure-based virtual screening was performed using TransformerCPI 2.0 against a million-scale compound libraries, followed by molecular docking, molecular dynamics simulations, and in vitro validation. The lead compound AK968 exhibited stable binding to UFC1 (ΔG = -25.07 kcal/mol), suppressed CRC cell proliferation, and synergistically enhanced 5-fluorouracil efficacy in a UFC1-dependent manner, with increased γ-H2AX foci formation indicating enhanced DNA damage accumulation. These findings identify UFC1 as a druggable effector of the UFMylation pathway in CRC chemotherapy stress and provide a framework integrating computational target identification, structure-based drug screening, and experimental validation for precision oncology.
Journal IF-equivalent: 9.8 (OpenAlex 2-year mean citedness, value as of 2026-11-04, retrieved 2026-11-05; not the official Clarivate IF)
Pancreatic cancer: treatment, trials and AI
- Abstract
Background: KRAS mutations occur in over 90% of pancreatic ductal adenocarcinomas (PDAC), and KRASG12D is the most frequent variant. G12D-selective inhibitors are now in clinical development, but computational pathology has so far predicted mutation status only at the gene level, leaving allele-specific prediction unexplored. Because genomic testing can be time-consuming and costly, a tool that predicts KRASG12D mutation status from routine hematoxylin and eosin (H&E) images could help identify patients for confirmatory molecular testing and potential treatment with KRASG12D inhibitors. Here we present a multi-scale framework that predicts KRASG12D status from H&E whole-slide images, with integration of computed tomography (CT).
Methods: Two public cohorts (TCGA, CPTAC) formed the training set, and an independent external cohort (Osaka Metropolitan University Hospital) served as validation. We developed a dual-magnification pipeline. In the high-magnification branch, a pathology foundation model (CONCH v1.5) was fine-tuned with low-rank adaptation (LoRA) and an attention-based multiple-instance learning head on 20× patches; in the low-magnification branch, a generative image model encoded a whole-slide overview. The two branches were integrated by late fusion into a multi-scale pathomics model, which was further combined with a radiomics model of CT.
Results: The multi-scale pathomics model reached a pooled internal test area under the receiver operating characteristic curve (AUC) of 0.611 and an external AUC of 0.701. In the external patients with CT, the combined model reached an AUC of 0.772, compared with 0.753 for pathomics alone.
Conclusions: KRASG12D status leaves a signal in routine H&E images and CT of PDAC that deep learning can detect and that transfers to an independent institution. After prospective validation, the framework might help prioritize patients for KRAS testing.
Journal IF-equivalent: 4.1 (OpenAlex 2-year mean citedness, value as of 2026-11-16, retrieved 2026-11-18; not the official Clarivate IF)Reference: Ma D, Sano T, Nishikubo H, Furukawa Y, Uchiyama K, Imanishi D, et al. Dual-Magnification Deep Learning Predicts KRASG12D Status from H&E Images of Pancreatic Cancer: A Multi-Cohort Study. Genes. 2026 Nov 15;17(11):1432. doi:10.3390/genes17111432.Checked: Abstract only
Prostate cancer: treatment, trials and AI
No new qualifying paper for this issue.
Brain tumours: treatment, trials and AI
- Abstract
Preoperative determination of O6-methylguanine-DNA methyltransferase (MGMT) promoter methylation status is of major prognostic and predictive relevance in IDH-wildtype glioblastoma, yet currently depends on invasive tissue sampling. This study investigates deep learning-based strategies for MGMT prediction from multiparametric MRIs, with particular emphasis on robustness and generalizability under heterogeneous, multicenter conditions. A total of 949 glioblastoma cases were analyzed, integrating public benchmark data from the Brain Tumor Segmentation Challenge (BraTS) with multi-institutional private cohorts acquired across diverse scanners and imaging protocols. Four complementary experimental settings were designed to compare reproduced literature baselines, private-only training, whole-brain end-to-end models, and segmentation-guided region-of-interest (ROI) pipelines. Tumor delineation was performed using an nnU-Net framework with expert quality control, and multiple three-dimensional convolutional neural network architectures were evaluated. Model performance was highest on the homogeneous BraTS benchmark, where a whole-brain DenseNet121 achieved an area under the receiver operating characteristic curve of 0.80. However, marked performance deterioration was observed when models were evaluated on independent held-out multicenter test set. In contrast, segmentation-guided ROI-based approaches showed a more stable performance profile in this preliminary independent held-out multicenter test. Explicit lesion localization was associated with reduction in false-positive classifications, suggesting a potential benefit for performance stability across heterogeneous datasets, although validation in external cohorts from institutions not represented during training is required. Collectively, these results demonstrate that upstream methodological choices, especially tumor segmentation and ROI definition, remain critical determinants of performance in MRI-based radiogenomic modeling. Future studies should emphasize prospective multicenter validation, domain-shift harmonization, multimodal data integration, and uncertainty-aware deep learning to enhance robustness and interpretability.
Journal IF-equivalent: 2.1 (OpenAlex 2-year mean citedness, value as of 2026-11-16, retrieved 2026-11-18; not the official Clarivate IF) - Abstract
Accurate, interpretable classification of brain tumours from magnetic resonance imaging (MRI) is central to treatment planning, yet high-performing deep models are often poorly calibrated and rarely tested beyond a single dataset. We present QSVQ-Net, a hybrid quantum-classical framework that couples an EfficientNet-B0 backbone and a complex-valued spiking neural network to a morphology-conditioned variational quantum circuit, in which eighteen morphological and radiomic descriptors parameterise the qubit rotations through a learnable mapping. Evaluation covers five-fold cross-validation with confidence intervals, calibration, a matched multi-seed ablation, intrinsic circuit characterisation, auxiliary localisation for mask-free inference and external validation against classical baselines. On the BRISC-2025 test partition, QSVQ-Net attains 98.8% accuracy with an expected calibration error of 0.006; under five-fold cross-validation it reaches 99.48% (95% CI 98.91–100.0), statistically indistinguishable from EfficientNet-B0 (99.57%; paired t-test p = 0.59). The quantum branch performs within confidence-interval overlap of a parameter-matched classical control, so we claim no quantum accuracy advantage. On an independent mask-free dataset, accuracy falls to 72.6% while classical baselines retain 90–93%, a limitation we trace to the morphology branch. QSVQ-Net thus offers a reproducible, transparently characterised quantum-classical design and a candid account of where morphology-conditioned quantum encoding helps and where it does not.
Journal IF-equivalent: 5.9 (OpenAlex 2-year mean citedness, value as of 2026-11-04, retrieved 2026-11-05; not the official Clarivate IF)Reference: Abinash P, Krishnamoorthy N. QSVQ-Net: A morphology-conditioned spiking–variational quantum network for interpretable brain tumour MRI classification. Sci Rep. 2026 Nov 16 [Epub ahead of print]. doi:10.1038/s41598-026-76936-5.Checked: Abstract only
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
- Abstract
RATIONALE AND OBJECTIVES: To develop and validate a computed tomography-based model integrating multichannel deep learning and radiomics for predicting pathologic significant response (PSR) to neoadjuvant chemoimmunotherapy (NACI) in locally advanced gastric cancer (LAGC), and to assess whether the model specifically captures immunotherapy-related phenotypes by testing it on a chemotherapy-alone cohort. MATERIALS AND METHODS: This multicenter study included 455 patients with LAGC: 411 receiving NACI (split into training, internal validation, and external test set 1) and 44 receiving chemotherapy alone (test set 2). From three-dimensional (3D) segmentations, the largest axial, coronal, and sagittal slices were stacked as three-channel input for a pretrained ResNet50 to extract deep features; radiomic features were extracted from the 3D volume. Features were fused, reduced by the least absolute shrinkage and selection operator, and modeled with a support vector machine to construct a deep learning radiomics (DLR) signature. A combined model further incorporated mismatch repair status, programmed death-ligand 1 combined positive score, and tumor grade. Model performance was evaluated by area under the curve (AUC), calibration, and decision curve analysis, with SHAP for interpretability. RESULTS: The multichannel model outperformed single-plane models (AUCs: 0.783, 0.749, 0.706). The DLR model achieved the highest AUCs (0.827, 0.802, 0.776), surpassing radiomics-only (0.756, 0.712, 0.691) and multichannel deep models. Critically, its AUC dropped to 0.597 in the chemotherapy-alone cohort. SHAP provided global and individual feature attributions. CONCLUSION: The multichannel DLR model accurately predicts PSR to NACI in LAGC. The substantial performance decline in the chemotherapy-alone cohort suggests that the captured imaging phenotypes may be more closely associated with immunotherapy-related response, rather than with general chemosensitivity, potentially offering a noninvasive, interpretable tool for pretreatment stratification.
Journal IF-equivalent: 3.5 (OpenAlex 2-year mean citedness, value as of 2026-11-04, retrieved 2026-11-05; not the official Clarivate IF)Reference: Ran C, Wang C, Chu G, Kong D. Multichannel Deep Learning Radiomics for Predicting Chemoimmunotherapy Response in Gastric Cancer: Multicenter Validation and Chemotherapy-alone Specificity Assessment. Acad Radiol. 2026 Nov [Epub ahead of print]. doi:10.1016/j.acra.2026.10.035. PMID: 43039808.Checked: Abstract only
Liver cancer: treatment, trials and AI
- Abstract
Background/Objectives: Hepatic arterial infusion chemotherapy (HAIC) is increasingly combined with tyrosine kinase inhibitors, immune checkpoint inhibitors or transarterial chemoembolisation, yet radiomics models seldom distinguish HAIC alone from HAIC within a multi-agent regimen. We reviewed handcrafted or semi-automated radiomics models in HAIC, stratified by regimen.
Methods: Following PRISMA 2020 (PROSPERO CRD420261487449), MEDLINE, Embase and Web of Science were searched to 6 November 2026 for English-language studies combining radiomic feature extraction with a machine-learning classifier to predict response, survival or treatment selection; end-to-end deep-learning studies were excluded. Studies were classified as monotherapy, combination regimen or mixed, appraised with the Radiomics Quality Score, PROBAST+AI and METRICS, and synthesised narratively.
Results: Eleven studies (2032 patients) were included: five monotherapy, four combination-regimen and two mixed or non-separable-regimen studies. Validation, test or bootstrap AUCs were 0.75–0.90 for monotherapy and 0.86–0.97 for combination-regimen studies; the highest values (0.99) were training-set estimates. Excluding three studies with uncertain regimen widened the gap. Only two had external validation, none shared data or code, and most had high or unclear risk of bias.
Conclusions: Handcrafted radiomics models show high but methodologically immature discrimination. Higher combination-regimen AUCs are confounded by tumour type, cohort and pipeline and do not demonstrate a HAIC-specific effect.
Funding: none.
Journal IF-equivalent: 6.1 (OpenAlex 2-year mean citedness, value as of 2026-11-07, retrieved 2026-11-08; not the official Clarivate IF)Reference: Tan YZ, Temperley HC, O’Sullivan NJ, Gallagher D, Gillham C, Harrold E, et al. Handcrafted and Semi-Automated Radiomics Models for Hepatic Arterial Infusion Chemotherapy: A Systematic Review Stratified by Monotherapy and Combination Regimens. Cancers. 2026 Nov 15;18(22):3655. doi:10.3390/cancers18223655.Checked: Abstract only - Abstract
BACKGROUND AND AIMS: The five-year post-resection recurrence rate of hepatocellular carcinoma (HCC) remains 50-70%, and satellite nodules constitute a well-established independent high-risk factor for early recurrence. The therapeutic benefit of adjuvant transarterial chemoembolization (TACE) in this specific subpopulation has not been rigorously evaluated in large cohorts. We therefore aimed to quantify the impact of postoperative adjuvant TACE and to develop a machine learning-based nomogram for individualized survival prediction in HCC patients with pathologically confirmed satellite nodules following radical hepatectomy. METHODS: This retrospective, non-randomized, multi-center cohort enrolled 273 HCC patients with pathologically confirmed satellite nodules who underwent radical hepatectomy between January 2015 and March 2024. Assignment to the adjuvant TACE group versus the radical-resection-only control group was determined by shared decision-making between the treating multidisciplinary team and the patient (or family) at the first postoperative outpatient review, rather than by randomization. Recurrence-free survival (RFS) and overall survival (OS) were compared by Kaplan-Meier analysis with log-rank tests. A 1:1 propensity-score-matched sensitivity analysis (caliper 0.2 SD; covariates: sex, age, BCLC stage, cirrhosis, HBV infection, ALBI grade, MVI, Edmondson grade, satellite multiplicity, resection type, and surgical margin) was performed to adjust for residual baseline imbalance. Independent prognostic factors were identified by univariable and multivariable Cox proportional hazards regression. A peritumoral delta-radiomics score (peritumoral_delta_radscore), derived from contrast-enhanced CT by three-dimensional volumetric segmentation, was generated. Three machine learning-based Cox models-XGBoost-Cox, Cox-LASSO, and CoxBoost-were constructed and evaluated by time-dependent AUC, decision curve analysis (DCA), calibration, and concordance index (C-index) under ten-fold cross-validation. SHAP (SHapley Additive exPlanations) values were applied to quantify feature contributions, and a clinical nomogram was constructed. RESULTS: Adjuvant TACE markedly improved both RFS and OS. In the full cohort, Median RFS was 18.0 months in the TACE group versus 10.6 months in the control group (log-rank P < 0.001), with 1- and 2-year RFS rates of 72.8% vs. 44.6% and 39.2% vs. 8.8%, respectively. Median OS was 36.5 months in the TACE group versus 19.5 months in controls (log-rank P < 0.001), with a 3-year OS of 53.6% vs. 13.5%. After 1:1 propensity-score matching (n = 248; 124 matched pairs), the survival advantage of adjuvant TACE remained statistically significant for both RFS (median 18.0 vs. 9.8 months; log-rank P < 0.001; 1-year RFS 72.6% vs. 42.7%; 2-year RFS 39.5% vs. 9.7%) and OS (median 36.9 vs. 22.9 months; log-rank P < 0.001; 1-year OS 100% vs. 100%; 3-year OS 54.0% vs. 16.1%). Multivariable Cox analysis identified treatment group, female sex, cirrhosis, BCLC stage, and peritumoral_delta_radscore as independent prognostic factors for both RFS and OS. Among the three machine learning models, XGBoost-Cox achieved the highest predictive performance, with time-dependent AUCs of 0.977/0.987/0.984 (1/3/5-year) in the training set (n = 163, events = 155) and 0.900/0.880/0.900 in the independent test set (n = 110, events = 105), consistently outperforming Cox-LASSO and CoxBoost. SHAP analysis identified treatment assignment, sex, BCLC stage C, and peritumoral_delta_radscore as the four dominant contributors. A 4-variable nomogram integrating these features demonstrated excellent calibration and net clinical benefit for 1-, 3-, and 5-year OS prediction. CONCLUSIONS: Adjuvant TACE might significantly prolong both RFS and OS in HCC patients with satellite nodules after radical hepatectomy. The XGBoost-Cox nomogram provides accurate individualized 1-, 3-, and 5-year survival prediction and may serve as an evidence-based decision-support tool for this high-risk subpopulation. As a retrospective, non-randomized study, the present findings require confirmation by prospective multicenter validation, ideally incorporating a randomized or propensity-score-stratified design to mitigate residual selection bias.
Journal IF-equivalent: 4.2 (OpenAlex 2-year mean citedness, value as of 2026-11-16, retrieved 2026-11-18; not the official Clarivate IF)Reference: Wang Y, Li W, Han C, Cao J, Xu H, Ye Y, et al. Machine learning-augmented Cox regression for evaluating adjuvant transarterial chemoembolization efficacy and predicting survival in hepatocellular carcinoma with satellite nodules following radical hepatectomy. BMC Gastroenterol. 2026 Aug 20;26(1):764. doi:10.1186/s12876-026-05235-0. PMID: 43040233.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.
Rare cancers
Adenoid cystic carcinoma: treatment, trials and AI
No new qualifying paper for this issue.