Search Results (67)

Non-small cell lung cancer (NSCLC) is the leading cause of cancer-related mortality worldwide. Accurate determination of epidermal growth factor receptor (EGFR) mutation status is essential for selecting patients eligible for tyrosine kinase inhibitors (TKIs). However, invasive genotyping is often limited by tissue accessibility and sample quality. This study presents a non-invasive machine learning model combining clinical data, CT morphological features, and radiomic descriptors to predict EGFR mutation status. A retrospective cohort of 138 patients with confirmed EGFR status and pre-treatment CT scans was analyzed. Radiomic features were extracted with PyRadiomics, and feature selection applied mutual information, Spearman correlation, and wrapper-based methods. Five Random Forest models were trained with different feature sets. The best-performing model, based on 11 selected variables, achieved an AUC of 0.91 (95% CI: 0.81–1.00) under stratified five-fold cross-validation, with an accuracy of 0.88 ± 0.03. Subgroup analysis showed that EGFR-WT had a performance of precision 0.93 ± 0.04, recall 0.92 ± 0.03, F1-score 0.91 ± 0.02, and EGFR-Mutant had a performance of precision 0.76 ± 0.05, recall 0.71 ± 0.05, F1-score 0.68 ± 0.04. SHapley Additive exPlanations (SHAP) analysis identified tobacco use, enhancement pattern, and gray-level-zone entropy as key predictors. Decision curve analysis confirmed clinical utility, supporting its role as a non-invasive tool for EGFR-screening. Full article

Precision oncology is witnessing an increasing number of molecular targets fueled by the continuous improvement of cancer genomics and drug development. Tumor genomic profiling is nowadays (August 2025) part of routine cancer patient care, guiding therapeutic decisions day by day. Nevertheless, implementing and distilling the increasing number of potential gene targets and possible precision drugs into therapeutically relevant actions is a challenge. The availability of prescreening programs for clinical trials has expanded the description of the genomic landscape of gastrointestinal tumors. The selection of the genomic test to use in each clinical situation, the correct interpretation of the results, and ensuring clinically meaningful implications in the context of diverse geographical drug accessibility, economic cost, and access to clinical trials are daily challenges of personalized medicine. In this context, well-established negative predictive biomarkers, such as extended RAS extended mutations for anti-EGFR therapy in colorectal cancer, and positive predictive biomarkers, such as MSI status, BRAF p.V600E hotspot mutation, ERBB2 amplification, or even NTRK1, NTRK2, NTRK3, RET, and NRG1 fusions across gastrointestinal cancers, are mandatory to provide tailored clinical care, improve patient selection for treatment and clinical trials, maximize therapeutic benefit, and minimize unnecessary toxicity. In this review, we provide an updated overview of actionable genomic alterations in GI cancers and discuss their implications for clinical decision making. Full article

The 2021 WHO classification of brain tumours revolutionised the oncological field by emphasising the role of molecular, genetic and pathogenetic advances in classifying brain tumours. In this context, incidental gliomas have been increasingly identified due to the widespread performance of standard and advanced MRI sequences and represent a diagnostic and therapeutic challenge. The impactful decision to perform a surgical procedure deeply relies on the non-invasive identification of features or parameters that may correlate with brain tumour genetic profile and grading. Therefore, it is paramount to reach an early and proper diagnosis through neuroradiological techniques, such as MRI. Standard MRI sequences are the cornerstone of diagnosis, while consolidated and emerging roles have been awarded to advanced sequences such as Diffusion-Weighted Imaging/Apparent Diffusion Coefficient (DWI/ADC), Perfusion-Weighted Imaging (PWI), Magnetic Resonance Spectroscopy (MRS), Diffusion Tensor Imaging (DTI) and functional MRI (fMRI). The current novelty relies on the application of AI in brain neuro-oncology, mainly based on radiomics and radiogenomics models, which enhance standard and advanced MRI sequences in predicting glioma genetic status by identifying the mutation of multiple key biomarkers deeply impacting patients’ diagnosis, prognosis and treatment, such as IDH, EGFR, TERT, MGMT promoter, p53, H3-K27M, ATRX, Ki67 and 1p19. AI-driven models demonstrated high accuracy in glioma detection, grading, prognostication, and pre-surgical planning and appear to be a promising frontier in the neuroradiological field. On the other hand, standardisation challenges in image acquisition, segmentation and feature extraction variability, data scarcity and single-omics analysis, model reproducibility and generalizability, the black box nature and interpretability concerns, as well as ethical and privacy challenges remain key issues to address. Future directions, rooted in enhanced standardisation and multi-institutional validation, advancements in multi-omics integration, and explainable AI and federated learning, may effectively overcome these challenges and promote efficient AI-based models in glioma management. The aims of our multidisciplinary review are to: (1) extensively present the role of standard and advanced MRI sequences in the differential diagnosis of iLGGs as compared to HGGs (High-Grade Gliomas); (2) give an overview of the current and main applications of AI tools in the differential diagnosis of iLGGs as compared to HGGs (High-Grade Gliomas); (3) show the role of MRI, radiomics and radiogenomics in unravelling glioma genetic profiles. Standard and advanced MRI, radiomics and radiogenomics are key to unveiling the grading and genetic profile of gliomas and supporting the pre-operative planning, with significant impact on patients’ differential diagnosis, prognosis prediction and treatment strategies. Today, neuroradiologists are called to efficiently use AI tools for the in vivo, non-invasive, and comprehensive assessment of gliomas in the path towards patients’ personalised medicine. Full article

Background: Colorectal liver metastasis (CRLM) represents a major clinical challenge in oncology, affecting 25–50% of colorectal cancer patients and significantly impacting survival. While multimodal therapies—including surgical resection, systemic chemotherapy, and local ablative techniques—have improved outcomes, prognosis remains heterogeneous due to variations in tumor biology, patient factors, and institutional practices. Methods: This review synthesizes current evidence on prognostic factors influencing CRLM management, encompassing clinical (e.g., tumor burden, anatomic distribution, timing of metastases), biological (e.g., CEA levels, inflammatory markers), and molecular (e.g., RAS/BRAF mutations, MSI status, HER2 alterations) determinants. Results: Key findings highlight the critical role of molecular profiling in guiding therapeutic decisions, with RAS/BRAF mutations predicting resistance to anti-EGFR therapies and MSI-H status indicating potential responsiveness to immunotherapy. Emerging tools like circulating tumor DNA (ctDNA) and radiomics offer promise for dynamic risk stratification and early recurrence detection, while the gut microbiome is increasingly recognized as a modulator of treatment response. Conclusions: Despite advancements, challenges persist in standardizing resectability criteria and integrating multidisciplinary approaches. Current guidelines (NCCN, ESMO, ASCO) emphasize personalized strategies but lack granularity in terms of incorporating novel biomarkers. This exhaustive review underscores the imperative for the development of a unified, biomarker-integrated framework to refine CRLM management and improve long-term outcomes. Full article

Background: Tyrosine kinase inhibitors (TKIs) have significantly improved outcomes in non-small cell lung cancer (NSCLC), especially among patients with actionable genetic mutations. However, the influence of family and personal medical history (FPMH) on clinical and treatment outcomes with TKI therapy remains underexplored. Methods: We conducted a retrospective cohort study involving 136 NSCLC patients receiving TKIs, categorized into two groups based on the presence or absence of documented FPMH. Clinical variables assessed included demographic data, comorbidities, Eastern Cooperative Oncology Group (ECOG) performance status, tumor characteristics, genetic mutations (EGFR, ALK, ROS1), treatment responses, toxicity profiles, and survival outcomes. Statistical analyses included Chi-square tests, t-tests, Mann–Whitney U tests, Spearman correlation, and univariate logistic regression (p < 0.05 threshold for significance). Results: Patients with FPMH (n = 34) had a significantly higher burden of chronic diseases (58.8% vs. 15.7%), poorer ECOG scores (≥3: 8.8% vs. 1.0%), increased recurrence (41.2% vs. 20.6%), and greater chemotherapy-related toxicity (50.0% vs. 28.4%) compared to those without FPMH (n = 102). However, there were no significant differences in survival duration or mutation status between the two groups. Conclusions: FPMH may be a predictive factor for treatment complications and recurrence in NSCLC patients receiving TKIs, although it does not appear to influence survival or genetic mutation status. These findings support the need for personalized clinical monitoring strategies based on medical history. Full article

Background: This study aimed to develop a nomogram based on the most relevant clinical, CT, and radiomic features comprising 11 key signatures (2 clinical, 2 CT-based, and 7 radiomic) for the non-invasive prediction of the EGFR mutation status and to support the timely initiation of tyrosine kinase inhibitor (TKI) therapy in patients with non-small cell lung cancer (NSCLC) adenocarcinoma. Methods: Retrospective real-world data were collected from 521 patients with histologically confirmed NSCLC adenocarcinoma who underwent CT imaging and either surgical resection or pathological biopsy for EGFR mutation testing. Five Random Forest classification models were developed and trained on various datasets constructed by combining clinical, CT, and radiomic features extracted from CT image regions of interest (ROIs), with and without feature preselection. Results: The model trained exclusively on the most relevant clinical, CT, and radiomic features demonstrated superior predictive performance compared to the other models, with strong discrimination between EGFR-mutant and wild-type cases (AUC = 0.88; macro-average = 0.90; micro-average = 0.89; precision = 0.90; recall = 0.94; F1-score = 0.91; and accuracy = 0.87). Conclusions: A nomogram constructed using a Random Forest model trained solely on the most informative clinical, CT, and radiomic features outperformed alternative approaches in the non-invasive prediction of the EGFR mutation status, offering a promising decision-support tool for precision treatment planning in NSCLC. Full article

Background/Objectives: Treatment of locally advanced rectal cancer (LARC) very often requires a neoadjuvant multimodal approach. Neoadjuvant treatment (NAT) encompasses treatments like chemoradiotherapy (CRT), short-course radiotherapy (SCRT), radiotherapy (RT) or a combination of either of these two with additional induction or consolidation chemotherapy, namely total neoadjuvant treatment (TNT). In case of complete radiological and clinical response, the non-operative watch-and-wait strategy can be adopted in selected patients. This strategy is impacted by a regrowth rate of approximately 30%. Predicting biomarkers of tumor response to NAT could improve guidance of clinicians during clinical decision making, improving treatment outcomes and decreasing unnecessary treatment exposure. To this day, there is no validated biomarker to predict tumor response to any NAT strategies in clinical use. Most research focused on CRT neglects the study of other regimens. Methods: We conducted a narrative literature review which aimed at summarizing the status of biomarkers predicting tumor response to NAT other than CRT in LARC. Results: Two hundred and fourteen articles were identified. After screening, twenty-one full-text articles were included. Statistically significant markers associated with improved tumor response pre-treatment were as follows: low circulating CEA levels; BCL-2 expression; high cellular expression of Ku70, MIB-1(Ki-67) and EGFR; low cellular expression of VEGF, hPEBP4 and nuclear β-catenin; the absence of TP53, SMAD4, KRAS and LRP1B mutations; the presence of the G-allel of LCS-6; and MRI features such as the conventional biexponential fitting pseudodiffusion (Dp) mean value and standard deviation (SD), the variable projection Dp mean value and lymph node characteristics (short axis, smooth contour, homogeneity and Zhang et al. radiomic score). In the interval post-treatment and before surgery, significant markers were as follows: a reduction in the median value of circulating free DNA, higher presence of monocytic myeloid-derived suppressor cells, lower presence of CTLA4+ or PD1+ regulatory T cells and standardized index of shape changes on MRI. Conclusions: Responders to neoadjuvant SCRT and RT tended to have a tumor microenvironment with an immune–active phenotype, whereas responders to TNT tended to have a less active tumor profile. Although some biomarkers hold great promise, scarce publications, inconsistent results, low statistical power, and low reproducibility prevent them from reliably predicting tumor response following NAT. Full article

Lung adenocarcinoma (LUAD) is driven by epidermal growth factor receptor (EGFR) mutations, making it a key therapeutic target. ADAM9, a member of the A disintegrin and metalloproteinase (ADAM) family, facilitates the release of growth factors and was implicated in activating the EGFR-mediated progression in several cancer types. In this study, we explored potential associations among ADAM9 single-nucleotide polymorphisms (SNPs), the EGFR mutation status, and the clinicopathological progression of LUAD in a Taiwanese population. In total, 535 LUAD patients with various EGFR statuses were enrolled, and allelic distributions of ADAM9 SNPs—located in promoter and intron regions, including rs78451751 (T/C), rs6474526 (T/G), rs7006414 (T/C), and rs10105311 (C/T)—were analyzed using a TaqMan allelic discrimination assay. We found that LUAD patients with at least one polymorphic G allele in ADAM9 rs6474526 had a lower risk of developing EGFR mutations compared to those with the wild-type (WT) TT genotype. Furthermore, G-allele carriers (TG + GG) of rs6474526 were associated with an increased likelihood of developing larger tumors (T3 or T4), particularly among patients with mutant EGFR. Conversely, in patients with WT EGFR, carriers of the T allele in rs10105311 had a lower risk of progressing to advanced stages (stage III or IV). Among females or non-smokers, G-allele carriers of rs6474526 demonstrated a higher risk of advanced tumor stages and distant metastases. In clinical data from the Genotype-Tissue Expression (GTEx) database, individuals with the polymorphic T allele in rs6474526 showed reduced ADAM9 expression in lung and whole blood tissues. Screening the genotype of rs6474526 in a set of LUAD cell lines revealed that cells carrying at least one minor G allele exhibited higher ADAM9 levels compared to those with the TT genotype. Additionally, analyses using TCGA and CPTAC databases revealed elevated ADAM9 expression in LUAD specimens compared to normal tissues. Elevated protein levels were correlated with advanced T stages, pathological stages, and worse prognoses. In summary, our results suggest that ADAM9 genetic variants of rs6474526 may affect ADAM9 expression and are associated with the EGFR mutation status. Both rs6474526 and rs10105311 were correlated with disease progression in LUAD patients. These variants could serve as potential biomarkers for predicting clinical outcomes. Full article

Colorectal cancer (CRC) is the third-most common cancer globally and a leading cause of cancer-related deaths. While the prognostic and predictive roles of RAS mutations in advanced CRC are well-established, their significance in early-stage CRC remains a topic of debate. Studies have been conducted for many years on clinical and pathological parameters that may be associated with RAS mutation, and there are inconsistent results in this regard. Currently, the only biomarker used in early-stage CRC is microsatellite status. KRAS mutations are detected in 40–50% of patients with colorectal cancer. RAS activating mutations cause loss of EGFR regulation by acting on the RAS/RAF/MAPK signaling pathways. In advanced colorectal cancer, these mechanisms cause a decrease in the effectiveness of EGFR inhibitors. However, studies on patients with early-stage colorectal cancer have inconsistent results. This review highlights the prognostic and clinical significance of KRAS mutations in early-stage CRC, particularly in MSS tumors. In the MSS group, KRAS mutations were associated with shorter TTR and OS compared to DWT patients. In contrast, in the MSI-H group, KRAS mutations showed no prognostic effect in TTR and OS. However. KRAS mutations were associated with shorter SAR in both MSI-H and MSS groups of patients. The findings underscore the need for routine molecular profiling, including KRAS and MSI status, to refine risk stratification and guide adjuvant therapy decisions. Further studies are warranted to explore targeted therapeutic approaches for KRAS-mutant CRC in the adjuvant setting. Full article

Background: Detecting molecular drivers is crucial in the management of non-small-cell lung cancer (NSCLC). This study aimed to evaluate the use of pretreatment chest computed tomography (CT) radiomics features for predicting epidermal growth factor receptor (EGFR) mutation status in NSCLC. Materials and Methods: CT images were used to develop a radiomics-based model for predicting EGFR mutation status. Two different groups were formed from the dataset, namely groups for training (n = 380) and testing (n = 86). Feature selection was performed using the Least Absolute Shrinkage and Selection Operator (LASSO) algorithm on a radiomics platform. Machine learning methods were then applied to construct the radiomics models. Receiver operating characteristic curve analysis was conducted to assess the performance of the radiomics signature across different datasets and methods. Results: The frequency of EGFR mutation was 13.5% (58/430). A total of 1409 quantitative imaging features were extracted from CT images using the Radcloud platform. Among the six radiomics-based classifiers (k-Nearest Neighbor, Support Vector Machine (SVM), eXtreme Gradient Boosting, Random Forest, Logistic Regression, and Decision Tree), SVM demonstrated the highest area under the curve values in both the testing and training groups, reaching 0.87 and 0.98, respectively. Our model, which incorporated both clinical and radiomics data, successfully predicted EGFR mutation status with an accuracy rate of 86.9%. Conclusion: Our findings highlight the potential of radiomics features as a non-invasive predictive imaging biomarker for EGFR mutation status, which could enhance personalized treatment in NSCLC. Radiomics emerges as a valuable tool for identifying driver mutations, although further studies are necessary to validate its clinical utility in NSCLC. Full article

Lung cancer is a major global health concern with a low survival rate, often due to late-stage diagnosis. Liquid biopsy offers a non-invasive approach to cancer detection and monitoring, utilizing various features of circulating cell-free DNA (cfDNA). In this study, we established two models based on cfDNA coverage patterns at the transcription start sites (TSSs) from 6X whole-genome sequencing: an Early Cancer Screening Model and an EGFR mutation status prediction model. The Early Cancer Screening Model showed encouraging prediction ability, especially for early-stage lung cancer. The EGFR mutation status prediction model exhibited high accuracy in distinguishing between EGFR-positive and wild-type cases. Additionally, cfDNA coverage patterns at TSSs also reflect gene expression patterns at the pathway level in lung cancer patients. These findings demonstrate the potential applications of cfDNA coverage patterns at TSSs in early cancer screening and in cancer subtyping. Full article

Background: Spheroids generated by tumor cells collected from malignant pleural effusion (MPE) were shown to retain the characteristics of the original tumors. This ex vivo model might be used to predict the response of non-small cell lung cancer (NSCLC) to anticancer treatments. Methods: The characteristics, epidermal growth factor receptor (EGFR) mutation status, and clinical response to EGFR-TKIs treatment of enrolled patients were recorded. The viability of the spheroids generated from MPE of enrolled patients were evaluated by visualization of the formazan product of the MTT assay. Results: Spheroids were generated from 14 patients with NSCLC-related MPE. Patients with EGFR L861Q, L858R, or Exon 19 deletion all received EGFR-TKIs, and five of these seven patients responded to treatment. The viability of the spheroids generated from MPE of these five patients who responded to EGFR-TKIs treatment was significantly reduced after gefitinib treatment. On the other hand, gefitinib treatment did not reduce the viability of the spheroids generated from MPE of patients with EGFR wild type, Exon 20 insertion, or patients with sensitive EGFR mutation but did not respond to EGFR-TKIs treatment. Conclusion: Multicellular spheroids generated from NSCLC-related MPE might be used to predict the response of NSCLC to treatment. Full article

Background: Isocitrate Dehydrogenase 1/2 (IDH 1/2)-wildtype (WT) astrocytomas constitute a heterogeneous group of tumors and have undergone a series of diagnostic reclassifications over time. This study aimed to investigate molecular markers, clinical, imaging, and treatment factors predictive of outcomes in WHO grade 2/3 IDH-WT astrocytomas (‘early glioblastoma’). Methodology: Patients with WHO grade 2/3 IDH-WT astrocytomas were identified from the hospital archives. They were cross-referenced with the electronic medical records systems, including neuroimaging. The expert neuro-pathology team retrieved data on molecular markers—MGMT, TERT, IDH, and EGFR. Tumors with a TERT mutation and/or EGFR amplification were reclassified as glioblastoma. Results: Fifty-four patients were identified. Sixty-three percent of the patients could be conclusively reclassified as glioblastoma based on either TERT mutation, EGFR amplification, or both. On imaging, 65% showed gadolinium enhancement on MRI. Thirty-nine patients (72%) received long-course radiotherapy, of whom 64% received concurrent chemotherapy. The median follow-up of the group was 16 months (range: 2–90), and the median overall survival (OS) was 17.3 months. The 2-year OS of the whole cohort was 31%. On univariate analysis, older age, worse performance status (PS), and presence versus absence of contrast enhancement on diagnostic MRI were statistically significant for poorer OS. Conclusion: IDH-WT WHO grade 2/3 astrocytomas are a heterogeneous group of tumors with poor clinical outcomes. The majority can be reclassified as glioblastoma, based on current WHO classification criteria, but further understanding of the underlying biology of these tumors and the discovery of novel targeted agents are needed for better outcomes. Full article

Purpose: The authors aimed to develop and validate deep-learning-based radiogenomic (DLR) models and radiomic signatures to predict the EGFR mutation in patients with NSCLC, and to assess the semantic and clinical features that can contribute to detecting EGFR mutations. Methods: Using 990 patients from two NSCLC trials, we employed an end-to-end pipeline analyzing CT images without precise segmentation. Two 3D convolutional neural networks segmented lung masses and nodules. Results: The combined radiomics and DLR model achieved an AUC of 0.88 ± 0.03 in predicting EGFR mutation status, outperforming individual models. Semantic features further improved the model’s accuracy, with an AUC of 0.88 ± 0.05. CT semantic features that were found to be significantly associated with EGFR mutations were pure solid tumours with no associated ground glass component (p < 0.03), the absence of peripheral emphysema (p < 0.03), the presence of pleural retraction (p = 0.004), the presence of fissure attachment (p = 0.001), the presence of metastatic nodules in both the tumour-containing lobe (p = 0.001) and the non-tumour-containing lobe (p = 0.001), the presence of ipsilateral pleural effusion (p = 0.04), and average enhancement of the tumour mass above 54 HU (p < 0.001). Conclusions: This AI-based radiomics and DLR model demonstrated high accuracy in predicting EGFR mutation, serving as a non-invasive and user-friendly imaging biomarker for EGFR mutation status prediction. Full article

Non-invasive methods to assess mutational status, as well as novel prognostic biomarkers, are warranted to foster therapy personalization of patients with advanced non-small cell lung cancer (NSCLC). This study investigated the association of contrast-enhanced Computed Tomography (CT) radiomic features of lung adenocarcinoma lesions, alone or integrated with clinical parameters, with tumor mutational status (EGFR, KRAS, ALK alterations) and Overall Survival (OS). In total, 261 retrospective and 48 prospective patients were enrolled. A Radiomic Score (RS) was created with LASSO-Logistic regression models to predict mutational status. Radiomic, clinical and clinical-radiomic models were trained on retrospective data and tested (Area Under the Curve, AUC) on prospective data. OS prediction models were trained and tested on retrospective data with internal cross-validation (C-index). RS significantly predicted each alteration at training (radiomic and clinical-radiomic AUC 0.95–0.98); validation performance was good for EGFR (AUC 0.86), moderate for KRAS and ALK (AUC 0.61–0.65). RS was also associated with OS at univariate and multivariable analysis, in the latter with stage and type of treatment. The validation C-index was 0.63, 0.79, and 0.80 for clinical, radiomic, and clinical-radiomic models. The study supports the potential role of CT radiomics for non-invasive identification of gene alterations and prognosis prediction in patients with advanced lung adenocarcinoma, to be confirmed with independent studies. Full article