AI and Cancer Research ── 2026-11-20
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(1)
Hematologic malignancies: AML(0) · CML(0) · ALL(0) · CLL(0) · Lymphoma(0) · Myeloma(0)
Solid tumors: Breast(0) · Lung(1) · Colorectal(1) · Pancreatic(0) · Prostate(1) · Brain(1) · Ovarian(0) · Endometrial(0) · Gastric(0) · Liver(1) · Kidney(1) · Bladder(0)
Rare cancers: ACC(0)
Cancer molecular biology and AI
- Abstract
BACKGROUND: Metabolic dysfunction-associated steatotic liver disease (MASLD) has become a global metabolic epidemic and represents a leading cause of progressive liver injury, ranging from steatosis to advanced fibrosis and even hepatocellular carcinoma. Despite its rising prevalence, the molecular mechanisms driving MASLD progression remain incompletely elucidated. Emerging evidence suggests that dysregulation of mitochondrial homeostasis and aberrant immune cell infiltration are critical contributors to disease pathogenesis, yet the interplay between these two processes and their regulatory networks in MASLD is poorly defined. METHODS: Transcriptomic data of MASLD and normal liver tissues were obtained from the Gene Expression Omnibus (GEO) database. Differential expression analysis, weighted gene coexpression network analysis (WGCNA), consensus clustering, and machine learning algorithms were integrated to identify hub genes associated with mitochondrial homeostasis in MASLD. The diagnostic performance of these hub genes was evaluated using receiver operating characteristic (ROC) curves. Furthermore, immune infiltration analysis was performed to characterize alterations in the immune microenvironment and to explore the correlations between hub genes and infiltrating immune cells. RESULTS: Five hub genes related to mitochondrial homeostasis were identified, including ZFP1, ZNF446, KCND3, RNF152, and OR13C9, which had a high diagnostic efficiency. The immune microenvironment of the MASLD group differed from that of the control group. The proportion of activated dendritic cells and myeloid-derived suppressor cells in MASLD group was higher, and the proportion of natural killer cells was lower. Hub genes were significantly associated with the infiltration of immune cells. CONCLUSION: Our findings underscore the critical role of mitochondrial homeostasis in shaping the immune microenvironment during MASLD progression. The identified hub genes may serve as promising diagnostic biomarkers and potential therapeutic targets, offering new avenues for precision intervention in MASLD.
Journal IF-equivalent: 1.9 (OpenAlex 2-year mean citedness, value as of 2026-11-17, retrieved 2026-11-19; not the official Clarivate IF)Reference: Jiang S, Li H. Bioinformatics and Machine Learning Identify Key Genes in Mitochondrial Homeostasis and Immune Infiltration in Metabolic Dysfunction–Associated Steatotic Liver Disease Progression. Biomed Res Int. 2026 Nov 17;2026(1):8770857. doi:10.1155/bmri/8770857. PMID: 43041520.Checked: Abstract only
See all 1 Cancer biology papers →
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
No new qualifying paper for this issue.
Lung cancer: treatment, trials and AI
- Abstract
BACKGROUND: Malignant pleural effusion (MPE) is a common complication of lung cancer and is associated with poor prognosis and substantial clinical burden. Differentiating malignant from non-malignant pleural effusion remains challenging, particularly when definitive pleural fluid cytology or pleural biopsy results are pending, inconclusive, or delayed. This study aimed to develop and validate a machine learning-based model for auxiliary classification of MPE in lung cancer patients after lung cancer diagnosis and partial tumor-related pathological or staging information had been obtained. METHODS: In this retrospective multicenter study, a total of 715 lung cancer patients with pleural effusion were included from two tertiary hospitals. The Shanxi cohort from Heping Hospital Affiliated to Changzhi Medical College included 420 patients and was used as the development cohort, whereas the Anhui cohort from Tongling People's Hospital included 295 patients and served as the independent external validation cohort. The development cohort was randomly divided into a training set and an internal test set at a ratio of 7:3, resulting in 295 patients in the training set and 125 patients in the internal test set. Clinical, laboratory, and pathological variables were collected. Feature selection was performed using least absolute shrinkage and selection operator regression within the training set. Seven machine-learning algorithms were developed and evaluated. Model performance was assessed using receiver operating characteristic curves, calibration analysis, and decision curve analysis. External validation was conducted using the independent Anhui cohort. RESULTS: Six predictors were selected, including age, Ki-67 index, lymph node metastasis, tumor size, total cholesterol, and serum albumin. Multiple machine learning models demonstrated comparable performance in internal validation. The XGBoost model achieved an area under the curve (AUC) of 0.97 in the internal test set and 0.91 in the independent external validation cohort. At the training-derived operating threshold, the external validation sensitivity was 0.35 and the specificity was 0.98, indicating favorable overall discrimination but limited rule-out performance. SHAP analysis suggested that tumor burden-related variables and systemic condition had substantial influence on model output. CONCLUSION: A machine learning-based model was developed for auxiliary classification of malignant and non-malignant pleural effusion in lung cancer patients after lung cancer diagnosis and partial tumor-related pathological or staging information had been obtained. The model showed favorable discrimination but limited sensitivity at the training-derived operating threshold in external validation and should therefore be regarded only as a candidate auxiliary tool. Further prospective and multicenter validation and recalibration are required before clinical implementation, and the model is not intended to replace cytological or histopathological confirmation.
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: Cai Q, Wu Y, Wang J, Gao Q, Xing Y, Kang M, et al. Machine learning–based differentiation of malignant and non-malignant pleural effusion in lung cancer patients: a multicenter retrospective study. BMC Cancer. 2026 Aug 21;26(1):1451. doi:10.1186/s12885-026-16755-z.Checked: Abstract only
Colorectal cancer: screening, treatment, trials and AI
- Abstract
Colorectal cancer (CRC) requires robust biomarkers and therapeutic targets. This study aimed to identify secretome-associated genes with diagnostic, prognostic, and tumor-suppressive relevance in CRC, focusing on GUCA2B (uroguanylin). We integrated single-cell transcriptomics, pseudobulk differential expression, machine learning (ML), multi-omics, and in vitro functional assays (HCT-116 cells). ML consensus feature selection across four algorithms identified 20 shared features from 850 downregulated genes; eight genes, including GUCA2B, were selected by all four models, forming the finalized signature. GUCA2B was markedly reduced in CRC epithelium. A frozen eight-gene logistic model showed an apparent (training-set) AUC of 0.998 in TCGA-COADREAD and maintained discrimination in three independent cohorts (AUC = 0.881-0.974). Reduced GUCA2B expression correlated with adverse survival, promoter hypermethylation, and altered immune infiltration. In vitro, GUCA2B restoration induced intrinsic apoptosis (increased BAX, decreased BCL2), reduced viability, and impaired migration; because GC-C/cGMP dependence was not tested, these effects constitute receptor-mechanism-agnostic proof of concept. In conclusion, GUCA2B represents a candidate epithelial-restricted CRC biomarker and tumor-suppressive gene. While the frozen multi-gene model shows promise for diagnostic stratification and GUCA2B restoration demonstrates preliminary tumor-suppressive activity, the small discovery set (n = 23 pseudobulk samples), moderate feature-selection stability, and single-cell-line validation necessitate cautious interpretation. Prospective clinical validation and in vivo multi-model studies are required before diagnostic or therapeutic application.
Journal IF-equivalent: 4.9 (OpenAlex 2-year mean citedness, value as of 2026-11-17, retrieved 2026-11-19; not the official Clarivate IF)Reference: Samadi P, Maghool F, Nouri F, Jalali A. Machine learning pipeline for secretome biomarker discovery in colorectal cancer: from single-cell transcriptomics to functional validation. Naunyn-Schmiedeberg's Arch Pharmacol. 2026 Nov 18 [Epub ahead of print]. doi:10.1007/s00210-026-06059-1.Checked: Abstract only
Pancreatic cancer: treatment, trials and AI
No new qualifying paper for this issue.
Prostate cancer: treatment, trials and AI
- Abstract
The International Society of Urological Pathology (ISUP) grade upgrading from biopsy to radical prostatectomy (RP) occurs in approximately 30%-40% of prostate cancer patients and may compromise treatment planning. Although multiparametric MRI (mpMRI) improves lesion characterization, prostatitis-like signal changes can mimic clinically significant cancer and confound radiological interpretation; however, no predictive model has incorporated mpMRI-diagnosed prostatitis as a variable for ISUP upgrading. This single-center retrospective study enrolled 503 consecutive men who underwent "12+1"-core transperineal ultrasound-guided systematic biopsy supplemented by cognitive mpMRI-targeted cores followed by laparoscopic RP (from May 2021 to October 2024). Least absolute shrinkage and selection operator (LASSO) logistic regression with 10-fold cross-validation was used for feature selection, and six machine learning classifiers were trained and evaluated on an independent test set (7:3 split). ISUP upgrading occurred in 195 patients (38.8%), and mpMRI-diagnosed prostatitis was significantly more prevalent in upgraded cases (27.2% vs 11.4%, P < 0.001). The random forest model achieved the highest weighted composite score (area under the receiver operating characteristic curve [ROC-AUC]: 0.823; average precision: 0.799) and was selected as the final model. SHapley Additive exPlanations (SHAP) analysis ranked feature importance as: basal tumor location (mean |SHAP|: 0.145), Prostate Imaging Reporting and Data System (PI-RADS) ≥4 (mean |SHAP|: 0.121), percentage of positive cores (mean |SHAP|: 0.100), maximum tumor diameter (mean |SHAP|: 0.062), and mpMRI-diagnosed prostatitis (mean |SHAP|: 0.040). As one of the first models to incorporate mpMRI-diagnosed prostatitis, this SHAP-interpretable framework showed that such prostatitis adds predictive value beyond PI-RADS scoring. This may help identify patients at heightened risk of undergrading on biopsy.
Journal IF-equivalent: 2.8 (OpenAlex 2-year mean citedness, value as of 2026-10-21, retrieved 2026-10-22; not the official Clarivate IF)Reference: Wang SY, Jin LC, Tian Z, Jia ZM, Li GZ, Chen X, et al. Predicting ISUP upgrading from biopsy to radical prostatectomy using an interpretable machine learning model with mpMRI-diagnosed prostatitis: a retrospective study. Asian J Androl. 2026 Nov 17 [Epub ahead of print]. doi:10.4103/aja202643. PMID: 43042515.Checked: Abstract only
Brain tumours: treatment, trials and AI
- Abstract
Accurate segmentation of brain tumor sub-regions from multimodal MRI is essential for diagnosis and treatment planning, but most deep learning methods process millions of voxels, which is computationally demanding. We propose TransformerGNN, a supervoxel graph-Transformer framework that reformulates volumetric segmentation as node classification on a sparse graph. Each multimodal volume (FLAIR, T1, T1ce, T2) is partitioned into approximately 2,000 SLIC supervoxels, reducing about 8.9 million voxels to a region adjacency graph whose nodes carry multimodal intensity and spatial features. Stacked TransformerConv layers combine graph connectivity with multi-head attention. On BraTS 2020, the model reaches a mean Dice of 0.9002 (WT 0.9579, TC 0.8393, ET 0.9033), and 0.893 } 0.005 across repeated runs. Paired ablation tests confirm the contribution of graph connectivity and, for whole-tumor and enhancing-tumor Dice, of attention. The model has 1.36 M parameters; its network forward pass takes 0.10 s on a 16 GB laptop and uses about 1.5 GB of memory, although end-to-end processing (about 32 s per volume) is dominated by supervoxel generation. Post-hoc gradient-based saliency provides regional, modality-level attributions whose clinical value remains to be validated.
Journal IF-equivalent: 2.0 (OpenAlex 2-year mean citedness, value as of 2026-11-09, retrieved 2026-11-11; not the official Clarivate IF)Reference: Ondo Minko AR, Tchagna Kouanou A, Mbuyi Bidikuchance V, Camara B, Kamguia H, Balde C. A Supervoxel Graph-Transformer Framework for Lightweight and Interpretable Brain Tumor Segmentation from Multimodal Magnetic Resonance Imaging. Biomed Phys Eng Express. 2026 Nov 17 [Epub ahead of print]. doi:10.1088/2057-1976/aec2f6.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
No new qualifying paper for this issue.
Liver cancer: treatment, trials and AI
- Abstract
BACKGROUND: Combined hepatocellular-cholangiocarcinoma (CHC) is a rare primary liver cancer with low incidence and poor prognosis. CHC can develop distant metastasis (DM) at an early stage, which severely shortens the survival time of patients. We aimed to use machine learning (ML) to construct a decision-making system to assess risk factors for the development of DM in CHC patients. METHODS: 1180 CHC patients from the SEER database between 2000 and 2020 were collected and randomized into the train set and internal test group in a 7:3 ratio. Patients diagnosed with CHC at our hospital between 2011 and 2018 were also collected as the external validation set (N = 125). Patients with CHC were divided into a metastasis group and a non-metastasis group according to the occurrence of DM. Univariate and multivariate logistic regression analyses were conducted to evaluate the risk factors influencing DM of CHC patients in the training set. We screened the feature variables based on random forest (RF) and forward feature importance sequences. Then we incorporated them into six ML algorithms for constructing machine learning models and used 10-fold cross-validation for internal validation. Receiver operating characteristic (ROC) curves, precision-recall curve (PRC), calibration curve (CC), and confusion Matrix (CM) were used to evaluate the predictive ability of the models. We determined the importance ranking of risk factors for DM in CHC patients using Shapley additive explanations (SHAP) and further used SHAP to analyze the interpretability of the black-box model. Finally, we built a web risk calculator based on the optimal performance model to facilitate its clinical application. RESULTS: A total of six variables were selected for constructing the decision model. Extreme gradient boosting (XGB) obtained the most significant recognition ability, with ROCAUC = 0.863, accuracy of 0.802, sensitivity of 0.875, and PRAUC of 0.642 in the internal test set. 10-fold cross-validation results showed that the ROCAUC of XGB was 0.989, with a standard error of 0.019. The SHAP method revealed that node, surgery, age, grade, primary site, and race are the top 6 key variables that contribute to the occurrence of DM in CHC. In addition, the analysis of two typical cases proved the reliability of the model. CONCLUSIONS: The XGB model outperforms other machine learning methods in recognizing the occurrence of distant metastases in CHC patients, and has a high degree of utility and reliability to inform clinical treatment decisions for patients.
Journal IF-equivalent: 7.8 (OpenAlex 2-year mean citedness, value as of 2026-11-17, retrieved 2026-11-19; not the official Clarivate IF)Reference: Xu L, Liu R, Wu S, Wang S, Cheng L, Wen T, et al. Decision system for combined hepatocellular-cholangiocarcinoma distant metastasis: a retrospective cohort study based on machine learning. Cancer Cell Int. 2026 Aug 10;26(1):365. doi:10.1186/s12935-026-04422-z.Checked: Abstract only
Kidney cancer: treatment, trials and AI
- Abstract
Pathology classification of cancer tumor subtypes and benign tumors can be difficult due to cellular heterogeneity and similar morphological features. Renal Cell Cancer (RCC) poses a major challenge for uropathologists as advanced RCC is often incidentally diagnosed with complex histological characteristics. We hypothesised that molecular profiling of RCC and benign tumor specimen by mass spectrometry with machine learning methods enable accurate tumor classification and identification of candidate RCC biomarkers. MALDI-MSI and quantitative proteomics (LC-MS) provided peptide and protein profiles of RCC neoplasms and benign tumors presented in tissue microarray (TMA) format of 541 tissue cores from 126 patients. Machine learning models classified RCC subtypes: Clear cell RCC (ccRCC), Chromophobe RCC (chRCC), papillary RCC (pRCC), and oncocytoma (OC). Known and novel candidate protein biomarker panels for discrimination of RCC subtypes were identified. RCC peptide mapping, evaluated using patient-level cross-validation, reliably classified ccRCC and healthy tissue, but did not reliably discriminate the diagnostically challenging RCC subtypes chRCC vs OC. Quantitative proteomics identified over 5000 proteins and classified all four RCC subtypes with 95.5% patient-level accuracy, including robust discrimination of chRCC from oncocytoma. We report new candidate protein biomarkers for ccRCC and pRCC and distinguish histologically similar subtypes OC (benign) and chRCC (malignant), as demonstrated by mass spectrometry-based immunohistochemistry. Candidate prognostic biomarkers for ccRCC contribute to risk of relapse (CDK6, ATM) and risk of metastasis (DDB2, MK03). Digital histo-molecular differentiation of cancer tumor subtypes and benign tumors by mass spectrometry and macbine learning enables retrospective and prospective studies. Identified candidate protein biomarkers complement existing protocols in tumor pathology.
Journal IF-equivalent: 7.2 (OpenAlex 2-year mean citedness, value as of 2026-11-17, retrieved 2026-11-19; not the official Clarivate IF)Reference: Sparre AA, Rabjerg M, Marcussen N, Lund L, Jensen ON. Mass spectrometry and machine learning for classification and molecular phenotyping of renal cell carcinoma and benign tumors. Molecular & Cellular Proteomics. 2026 Nov [Epub ahead of print]. doi:10.1016/j.mcpro.2026.101708.Checked: Abstract only
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.