AI and Cancer Research ── Cancer biology ── 2026-10-22
Cancer molecular biology and AI
- Abstract
PURPOSE: In urological practice, the decision to resect or preserve a kidney with obstruction, tumor or atrophy is based on split renal function, although SPECT renal dynamic scintigraphy as the reference standard is resource-intensive. We developed and externally validated a non-contrast CT radiomics model to identify nonfunctioning kidneys and support urological decision-making in this setting. METHODS: In this three-center retrospective study (TRIPOD+AI), adults underwent non-contrast CT and SPECT renal dynamic scintigraphy within 72 h. IBSI-compliant radiomics features were extracted with PyRadiomics, and between-kidney asymmetry features were constructed. After three-stage selection (Spearman, recursive feature elimination, and LASSO), 12 machine learning models were developed and externally validated in 2 independent cohorts. A Rad-score and clinical nomogram were also derived. RESULTS: Among 1029 patients, the internal cohort included 418 patients. Twenty-three of features were retained. The random forest model yielded an internal test AUC of 0.9426, sensitivity of 85.5%, specificity of 88.3%, and negative predictive value of 95.6%. Its external AUCs were 0.875 and 0.913. The nomogram attained external AUCs of 0.920 and 0.881. At the thresholds fixed in the training data, the random forest achieved external sensitivity of 66.0 and 76.4% with negative predictive values of 94.8 and 93.3%, and the nomogram 92.5 and 93.7% with 98.3 and 97.2%. Both radiomics models outperformed logistic-regression baselines built from clinical variables and simple CT size and asymmetry measurements, although the margin over the combined baseline was not statistically significant in one external cohort. The Rad-score was correlated with single-kidney GFR (ρ = -0.738, P < 0.001) and facilitated risk stratification. CONCLUSION: Non-contrast CT radiomics identified nonfunctioning kidneys without contrast agents or radiotracers. Because non-contrast CT is already part of routine urological workup, this externally validated rule-out tool may reduce unnecessary SPECT examinations and support decisions between nephrectomy and nephron-sparing management.
Journal IF-equivalent: 1.9 (OpenAlex 2-year mean citedness, value as of 2026-10-21, retrieved 2026-10-22; not the official Clarivate IF)Reference: Jiang Y, Zheng Y, Feng Z, Bao L, Wei Q. Non-contrast CT radiomics for identifying nonfunctioning kidneys: a multicenter development and external validation study. Int Urol Nephrol. 2026 Oct 20 [Epub ahead of print]. doi:10.1007/s11255-026-05451-3. PMID: 42902104.Checked: Abstract only - Abstract
Conventional radiomics is hindered by manual feature engineering and poor cross-tumor generalization. We introduce SAFARI (Self-supervised Tumor-Agnostic Framework for Radiomics Representation Learning), a novel, self-supervised model for radiomics that learns disease agnostic tumor characteristics. Built on a transformer architecture with a dual-objective self-supervised strategy, combining masked feature reconstruction and contrastive learning across tumor subvolumes, SAFARI was trained on 1,218 radiomic features from 27,144 tumor subvolumes across 1,086 patients with six major cancer types (lung, liver, colorectal, kidney, pancreatic, and colorectal liver metastases). In transfer learning experiments on independent datasets comprising 762 patients, fixed SAFARI embeddings outperformed our matched conventional radiomics baselines across multiple downstream tasks. Performance gains were substantial for predicting HER2 status in gallbladder cancer (AUC: 0.758 vs. 0.610), NSCLC histopathology (accuracy: 71.3% vs. 62.1%), and EGFR mutation status (AUC: 0.805 vs. 0.690). By demonstrating that a single pretrained model outperforms matched task-specific baselines on the majority of endpoints, SAFARI establishes a new, more robust paradigm for radiomic analysis in oncology.
Journal IF-equivalent: 3.2 (OpenAlex 2-year mean citedness, value as of 2026-10-20, retrieved 2026-10-21; not the official Clarivate IF)Reference: Gupta P, Bose K, Dutta N, Mudgil P, Arora C. SAFARI: A Self-supervised Tumor-Agnostic Framework for Radiomics Representation Learning in Oncology Practice. ACM Trans Comput Healthcare. 2026 Oct 19 [Epub ahead of print]. doi:10.1145/3859479.Checked: Abstract only - Abstract
Considering that clinical data for 68 Ga-labeled poly(ADP-ribose) polymerase (PARP)-targeted probes have not yet been reported and integrated artificial intelligence (AI) workflows for radiotracer development are still lacking, to support radiotracer development and clinical translation, we constructed an AI-assisted multi-stage prioritization workflow and then synthesized two 68 Ga-labeled PARP-1-targeted probes, validating their performance through in vitro , in vivo , and clinical evaluations. An AI-guided multi-objective workflow (encoder-decoder recurrent neural network (RNN) generation, convolutional neural network (CNN) docking scoring, Synthetic Complexity Learned from a Reaction Corpus (SCScore)-based tractability triage, and deep neural network (DNN)) was used to prioritize a derivatization-ready Fuzuloparib analog (NH 2 -FZPF) and to guide chelator installation, yielding DOTA-FZPF and NOTA-FZPF precursors for subsequent 68 Ga radiolabeling. Both [ 68 Ga]Ga-DOTA-FZPF and [ 68 Ga]Ga-NOTA-FZPF exhibited high radiochemical purity, stability, and PARP-1-targeted specificity. In tumor models, [ 68 Ga]Ga-DOTA-FZPF displayed favorable tumor uptake and rapid background clearance, supporting further evaluation. Exploratory positron emission tomography (PET) assessments of Fuzuloparib target engagement show the specificity and functionality of [ 68 Ga]Ga-DOTA-FZPF, supporting in vivo evaluation of PARP inhibitor target engagement and pharmacodynamics. This study recruited 4 cancer patients and performed positron emission tomography/computed tomography (PET/CT) scans using [ 68 Ga]Ga-DOTA-FZPF. Biopsy specimens showed that PET/CT-positive lesions corresponded to regions of high PARP-1 expression. This study demonstrates how an AI rational design workflow can support decision-making in radiotracer design while maintaining synthetic feasibility and translational potential. These preliminary findings support the clinical feasibility of [ 68 Ga]Ga-DOTA-FZPF and suggest that it is promising for non-invasive assessment of PARP-1 expression and dynamic monitoring of in vivo target occupancy and pharmacodynamics.
Journal IF-equivalent: 6.6 (OpenAlex 2-year mean citedness, value as of 2026-10-20, retrieved 2026-10-21; not the official Clarivate IF)Reference: Ji M, Jiang C, Xi Y, Liu S, He S, Zhang J, et al. An artificial intelligence-assisted PARP-1-targeted PET imaging agent [68Ga]Ga-DOTA-FZPF: From bench to clinic. Materials Today Bio. 2026 Dec;41:103728. doi:10.1016/j.mtbio.2026.103728.Checked: Abstract only - Abstract
Chemical-induced transcriptional profiles facilitate systematic characterization of compound-driven perturbations, offering critical insights into mechanisms of action, especially for compounds with poorly defined molecular targets. However, experimentally profiling gene expression across all combinations of cell lines and compounds is impractical, limiting the large-scale application of comparative transcriptomics in drug screening. To address this, we developed a deep learning framework named WAVE (Wave Action-of-Drug with Variational Encoder), which predicts gene expression profiles from chemical structures and the basal states of cell lines and single cells. By effectively integrating drug molecular representations and cellular basal transcriptional profile using a β-variational autoencoder (β-VAE) framework, WAVE accurately predicts untested chemical-induced transcriptional profiles. Leveraging this integration of cellular insights and molecular characteristics, WAVE excels in drug discovery and drug repurposing based on in silico perturbation experiment. It's application in lung adenocarcinoma (LUAD) drug discovery showcased its potential to identify potentially effective compounds. Our results demonstrate the utility of WAVE for elucidating mechanisms of action for potential drugs and enabling large-scale drug screening.
Journal IF-equivalent: 18.1 (OpenAlex 2-year mean citedness, value as of 2026-10-06, retrieved 2026-10-07; not the official Clarivate IF)Reference: Lv T, Chen B, Dai X, Gao M, Zhu B, Shen M, et al. Learning chemical-induced gene expression perturbations with WAVE. Nat Commun. 2026 Sep 18;17(1):10995. doi:10.1038/s41467-026-77645-3. PMID: 42901891.Checked: Abstract only