Search Results (1,459)

Non-small cell lung cancer (NSCLC) remains one of the leading causes of cancer-related mortality, with resistance to chemotherapy representing a major therapeutic challenge. In this study, we investigated the effects of conventional chemotherapeutics, Cisplatin and 5-fluorouracil (5-FU), in combination with small molecule inhibitors (SMIs) targeting the PI3K/AKT signaling pathway, on NSCLC cell viability. Two NSCLC cell lines, H460 (large cell lung carcinoma) and A549 (adenocarcinoma), both characterized by constitutive activation of PI3K/AKT signaling, were evaluated. A normal human lung fibroblast cell line, MRC-5, was used as a non-cancer control to assess selectivity and exclude cytotoxic effects. Dose–response analyses were performed to determine the optimal concentrations of Cisplatin, 5-FU, the AKT inhibitor MK2206, and the PI3K inhibitor BKM120, both as monotherapies and in combination treatments. We identified a synergistic combination of 5-FU and BKM120 that significantly reduced viability and induced apoptosis in NSCLC cells while sparing MRC-5 cells. Mechanistic studies revealed that apoptosis induction was mediated through the apoptotic pathway regulated by the Bcl-2 family and activation of caspase-3 and caspase-6. These findings highlight the therapeutic potential of combining PI3K/AKT inhibitors with conventional chemotherapy to overcome resistance mechanisms in NSCLC. Full article

The growing adoption of deep learning (DL) in early-stage cancer diagnosis has demonstrated remarkable performance across multiple imaging tasks. Yet, the lack of transparency in these models (“black-box” problem) limits their adoption in clinical environments. This study proposes a methodological framework for developing interpretable DL models to support the early histopathological diagnosis of lung cancer, with a focus on adenocarcinoma and squamous cell carcinoma. The approach leverages publicly available datasets (TCGA-LUAD, TCGA-LUSC, LC25000) and employs high-performing architectures such as EfficientNet, along with post hoc explainability techniques including Grad-CAM and SHAP. Data will be pre-processed and sampled using stratified purposeful strategies to ensure diversity and balance across subtypes and stages. Model evaluation will combine standard performance metrics with clinician feedback and the spatial alignment of visual explanations with ground-truth annotations. While implementation remains a future step, this paper proposes a methodological framework designed to guide the development of DL systems that are not only accurate but also interpretable and clinically meaningful. Full article

Background/Objectives: Despite the significant advancements made in the diagnosis of lung cancer, the traditional diagnostic methods remain limited because they are often invasive, expensive, and not suitable for regular screening, creating a need for more accessible and non-invasive alternatives. In this context, the analysis of miRNAs in EVs and free circulating microRNA may be used as liquid biopsies in lung cancer to identify individuals at risk. This study aimed to compare miRNA profiles in the serum and EVs derived from lung cancer patients by focusing on Let-7a-5p and miR-21-3p. Materials and Methods: Serum and EVs were isolated from lung cancer patients and healthy controls. EVs were characterized using nanoparticle tracking analysis, electron microscopy, and Western blotting for surface markers (CD63, CD81, TSG101). Total miRNA levels were quantified in the serum and EVs, and specific miRNAs (hsa-let-7a-5p and hsa-miR-21-3p) were analyzed using RT-qPCR. Statistical analysis evaluated miRNA expression across clinicopathological features, including age, gender, smoking status, tumor stage, cancer type, and EGFR mutation status. Results: Total miRNA levels were significantly enriched in EVs compared to the serum. Let-7a-5p was downregulated in EVs from patients with advanced-stage lung cancer (Stage III–IV) compared to those with early-stage cancer and controls (p < 0.05), while no differences were observed in the serum. Conversely, miR-21-3p was significantly upregulated in EVs and serum from advanced-stage patients (p < 0.01) and in adenocarcinoma compared to squamous cell carcinoma (p < 0.05). No significant differences were observed for age, gender, or smoking status. Conclusions: Our findings highlight the differential expression of miRNAs in EVs and the serum, emphasizing the diagnostic potential of EV-associated Let-7a-5p and miR-21-3p in lung cancer. These results suggest that EVs are a more robust source for miRNA biomarkers compared to the serum. Full article

Background/Objectives: Medicinal plants are an abundant source of bioactive molecules, particularly in arid environments, such as Saudi Arabia, where Ocimum basilicum L. (Saudi basil) has long been used for its therapeutic properties. This study aimed to examine the phytochemical profile and bioactivities of non-ozonated (untreated) and ozonated methanolic extracts of O. basilicum and to determine whether ozonation enhances their biological effects, with a focus on antidiabetic, anti-Alzheimer, anti-inflammatory, antimicrobial, and cytotoxic properties. Methods: Fresh leaves of O. basilicum were extracted with methanol, subjected to ozonation, and analyzed by HPLC. In vitro assays were conducted to evaluate α-amylase, α-glucosidase, and BChE inhibition, RBC membrane stabilization, antibacterial activity against Helicobacter pylori and cytotoxicity using normal lung fibroblasts (WI-38) and human colorectal adenocarcinoma cell line (Caco-2). Results: Ozonation modified the phytochemical profile, enriching chlorogenic and rosmarinic acids. Ozonated extracts exhibited stronger inhibition of α-amylase with an IC₅₀ of 5.09 µg/mL compared to 13.6 µg/mL of untreated Saudi basil and α-glucosidase (IC₅₀ 6.15 µg/mL vs. 9.42 µg/mL). They also showed enhanced BChE inhibition with an IC₅₀ of 13.4 µg/mL compared to 31.8 µg/mL of non-ozonated extract. In addition, ozonated extracts produced significant anti-inflammatory effects by stabilizing RBCs, with an IC₅₀ of 8.04 µg/mL compared to 8.44 µg/mL for untreated extracts and 4.41 µg/mL for indomethacin. Ozonated extracts produced larger H. pylori inhibition zones (26.7 mm) and an MBC/MIC ratio of 1. Cytotoxicity testing revealed that ozonated extracts were less toxic to WI-38 cells, with IC₅₀ values of 437.89 µg/mL versus 191.06 µg/mL, and 149.14 µg/mL compared to 103.7 µg/mL of untreated Saudi basil in Caco-2 cells. Conclusions: Ozonation enriches the phytochemical composition of O. basilicum, enhancing antidiabetic, neuroprotective, anti-inflammatory, and antibacterial activities while reducing cytotoxicity on normal cells. These findings support the potential of ozonated O. basilicum as a safe and promising natural therapeutic candidate for metabolic, neurodegenerative, and infectious diseases. Full article

Background/Objectives: PROX1 (prospero homeobox 1) is a transcription factor involved in lymphangiogenesis and cellular differentiation. Its role in cancer biology appears to be highly context-dependent, with it exhibiting both tumor-promoting and -suppressive functions across various malignancies. Nonetheless, the clinical significance of PROX1 expression in non-small cell lung cancer (NSCLC) remains poorly elucidated. The objective of this study is to evaluate the immunohistochemical expression of PROX1 in NSCLC, specifically in the adenocarcinoma and squamous cell carcinoma subtypes, and to assess its correlation with clinicopathologic features and overall survival (OS). Methods: This retrospective study included surgically resected specimens from 121 patients with histologically confirmed NSCLC. PROX1 expression was assessed via immunohistochemistry on formalin-fixed, paraffin-embedded specimens. Staining intensity (graded 0– National and Kapodistrian University of Athens 3) and the percentage of positive tumor cells were recorded. Correlations with histological subtype, tumor characteristics, and OS were analyzed using chi-square tests, one-way ANOVA, and Kaplan–Meier survival analysis with log-rank testing. Results: Low PROX1 intensity (level 1) was significantly associated with P63 positivity (p = 0.028), while high PROX1 intensity (level 3) correlated with nodal metastasis to station 3 (S3+) (p = 0.025). Additionally, alveolar-pattern adenocarcinomas exhibited intermediate PROX1 expression (26–50%) (p = 0.010). Although PROX1 positivity did not differ among mucinous and non-mucinous adenocarcinomas (p = 0.152), its distribution across defined expression subgroups was statistically significant (p = 0.002). Tumors with low PROX1 expression (0–24%) were associated with a larger maximum tumor diameter (p = 0.026). PROX1 expression was not independently associated with OS (p > 0.05). Factors significantly associated with improved survival included an age < 50 years, female sex, the absence of necrosis, fewer than 10 positive lymph nodes, a lymph node ratio < 0.5, and the absence of extensive nodal involvement in stations 5, 10, 11, and 12. Conclusions: Although PROX1 expression is variably associated with specific histologic subtypes and nodal metastases in NSCLC, it does not independently predict overall survival. Its expression patterns suggest a potential role in tumor differentiation and lymphatic spread. Further mechanistic and immunologic studies are warranted to elucidate the functional significance of PROX1 in lung cancer biology. Full article

Chlorpyrifos (CPF), an organophosphate pesticide, is known to induce pulmonary toxicity through oxidative stress, mitochondrial dysfunction, and inflammation. Astaxanthin (ASX), a xanthophyll carotenoid derived primarily from marine microalgae (Haematococcus pluvialis), possesses strong antioxidant properties and has demonstrated cellular protective effects in numerous oxidative stress studies. However, its efficacy against CPF-induced lung cell damage remains uncharacterized. This study revealed the protective role of ASX, as a pretreatment and co-treatment, against CPF-induced cytotoxicity in human A549 lung adenocarcinoma cells by assessing cell viability, intracellular reactive oxygen species (IROS), total oxidative status (TOS), total antioxidant capacity (TAC), mitochondrial membrane potential (MMP), intracellular calcium ions (Ca²⁺), lactate dehydrogenase (LDH) release, malondialdehyde (MDA) levels, glutathione peroxidase (GPx) activity, superoxide dismutase (SOD) activity, DNA fragmentation, and apoptosis/inflammation-associated gene expression. CPF treatment significantly decreased cell viability and TAC, while elevating IROS, TOS, MMP, intracellular Ca²⁺, and LDH release. CPF also increased MDA levels and suppressed GPx and SOD activities. DNA fragmentation and quantitative polymerase chain reaction (qPCR) analysis revealed upregulation of pro-apoptotic and inflammatory markers such as BCL2-associated X protein (BAX), caspase-3 (CASP3), tumor protein p53 (TP53), tumor necrosis factor-alpha (TNF-α), interleukin-1 beta (IL-1β), nuclear factor kappa B (NFκB), and voltage-dependent anion-selective channel protein 1 (VDAC1) and suppression of anti-apoptotic B-cell lymphoma 2 (BCL2) and antioxidant defense genes nuclear factor erythroid 2-related factor 2 (Nrf2) and heme oxygenase-1 (HO-1). ASX treatment, particularly when administered as a pretreatment, significantly reversed CPF-induced oxidative and inflammatory responses by restoring SOD, GPx, and TAC levels, reducing IROS, TOS, MDA, and LDH release, and downregulating apoptotic and inflammatory gene expressions. ASX pretreatment notably decreased MMP and intracellular Ca²⁺ levels, indicating protection against mitochondrial dysfunction and calcium dysregulation. ASX upregulated Nrf2 and HO-1 expression and restored the BCL2/BAX balance, suggesting inhibition of mitochondrial-mediated apoptosis. Additionally, ASX significantly attenuated CPF-induced anti-angiogenic effects in the in ovo Hen’s Egg Test Chorioallantoic Membrane (HET-CAM) assay. These findings demonstrate, for the first time, that ASX exerts a broad spectrum of protective effects against CPF-induced cytotoxicity in lung cells, mainly through the stabilization of mitochondrial redox status and modulation of apoptosis- and inflammation-related gene pathways, highlighting ASX as a promising candidate for further therapeutic development. Furthermore, the pronounced efficacy observed in the pretreatment regimen suggests that ASX can be evaluated as a potential nutritional preventive strategy in high-risk populations with occupational or environmental CPF exposure. Full article

Objectives: Non-small cell lung cancer (NSCLC), the most prevalent type of lung cancer, includes subtypes such as adenocarcinoma (ADC) and squamous cell carcinoma (SCC), which require distinct management approaches. Accurately differentiating NSCLC subtypes based on diagnostic imaging remains challenging. However, the extraction of radiomic features—such as first-order statistics (FOS), second-order statistics (SOS), and fractal dimension texture analysis (FDTA) features—from magnetic resonance (MR) images supports the development of quantitative NSCLC assessments. Methods: This study aims to evaluate whether the integration of FDTA features with FOS and SOS texture features in MR image analysis improves machine learning classification of NSCLC into ADC and SCC subtypes. The study was conducted on 274 MR images, comprising ADC (n = 122) and SCC (n = 152) cases. From the segmented MR images, 93 texture features were extracted. The random forest algorithm was used to identify informative features from both FOS/SOS and combined FOS/SOS/FDTA datasets. Subsequently, the k-nearest neighbors (kNN) algorithm was applied to classify MR images as ADC or SCC. Results: The highest performance (accuracy = 0.78, precision = 0.81, AUC = 0.89) was achieved using 37 texture features selected from the combined FOS/SOS/FDTA dataset. Conclusions: Incorporating fractal descriptors into the texture-based classification of lung MR images enhances the differentiation of NSCLC subtypes. Full article

Background and Objectives: Non-small-cell lung cancer (NSCLC) often has epidermal growth factor receptor (EGFR) mutations, which are key targets for therapy. EGFR mutation subtypes, especially exon 19 deletions and exon 21 L858R mutations, influence responses to EGFR tyrosine kinase inhibitors (TKIs) and patient survival. Despite progress in TKI treatments, resistance and different responses remain challenges. This study explores the relationship between EGFR mutation subtypes, PD-L1 expression, and patient outcomes after anti-EGFR therapy. Materials and Methods: We studied 176 cases of EGFR mutation-positive NSCLC. Next-generation sequencing was used to analyze EGFR and other mutations, while PD-L1 expression was evaluated through immunohistochemistry. We analyzed EGFR mutation subtypes, PD-L1 status, treatments, and survival outcomes. Results: Among 176 cases, 88.6% were adenocarcinomas. Within the EGFR mutation spectrum, exon 19 deletions were the most common subtype, accounting for 40.9% of cases, followed by the point mutation in exon 21, which occurred in 35.8% of cases. Less frequent alterations, making up 23.3% of all detected mutations, included mutations in exon 18, insertions, and point mutations such as S768I and T790M in exon 20, as well as changes in exon 2, exon 7, and other less frequently affected regions. Exon 19 mutations were associated with older age, female sex, adenocarcinoma, and bone metastasis (p < 0.05). TP53 was the most common concurrent mutation (44.3%). PD-L1 positivity (TPS ≥ 1%) was observed in 48.3%, with high expression (TPS ≥ 50%) in 25.9%. Exon 21 mutations were significantly linked to PD-L1 negativity (p = 0.008). The median overall survival was longest with TKI therapy (51 months), and this was also observed in PD-L1-positive patients, although the difference was not statistically significant. The median progression-free survival for patients treated with TKIs and those with EGFR mutations was 14 months. PD-L1-positive patients receiving TKIs had significantly longer survival than those who did not (51 vs. 17 months, p = 0.003). Conclusions: EGFR mutation subtypes and PD-L1 expression seem to affect treatment outcomes and survival in NSCLC. The observed links emphasize the potential value of combining molecular and immunological markers to guide therapy choices. Full article

The objective of the present study was to develop injectable solutions of curcumin (CUR) and resveratrol (RES) for intravenous administration as a strategy to increase their solubility and stability, as well as to evaluate their cytotoxic potential, individually and in combination, on human lung non-small adenocarcinoma cells (A549 cells) and non-tumoral cells isolated from normal human bronchial epithelium (BEAS cells) to establish possible synergistic effects and potential therapeutic alternatives for lung cancer. Using factorial experimental designs, the components of the injectable CUR and RES solutions were selected, and their hemolytic potential was evaluated by a static method. In addition, combinations of injectable CUR:RES solutions (25:75, 50:50 and 75:25) were prepared from the individual ones, and their stability under refrigeration conditions and cytotoxic potential on A549 and BEAS cells were evaluated. The stability of the injectable solutions of CUR, RES and their different combinations was maintained for 3 months, except for the 25:75 combination of CUR:RES. Furthermore, the cytotoxic potential of CUR and RES on tumoral cells (A549) and non-tumoral (BEAS) cells was evaluated, indicating a dose-dependent effect; the combination of CUR:RES 50:50 and the combination of CUR:RES 75:25 presented synergistic effects in reducing cell viability. This study suggests that injectable solutions of CUR, RES and their combination for intravenous administration could be potential viable candidates and should be evaluated for their efficacy in animal models of lung cancer to establish new possible treatments. Full article

RNA-binding proteins (RBPs), particularly IGF2BP3, play critical but underexplored roles in lung adenocarcinoma (LUAD). This study investigated IGF2BP3′s clinical and functional significance using single-cell/RNA sequencing, validated by qPCR, Western blot, and immunohistochemistry. The results show IGF2BP3 was significantly upregulated in LUAD tissues and associated with advanced-stage, larger tumors, lymph node metastasis, and poor prognosis. A prognostic nomogram confirmed its independent predictive value. Functionally, IGF2BP3 knockdown suppressed proliferation, and induced G2/M arrest and apoptosis. GSEA linked high IGF2BP3 to cell cycle activation and low expression to metabolic pathways. Notably, high IGF2BP3 correlated with immune evasion markers (downregulated CD4+ effector T cells, upregulated Th2 cells), while TIDE analysis suggested a better immunotherapy response in low-expressing patients. Drug screening identified BI-2536 as a potential therapy for low-IGF2BP3 cases, supported by strong molecular docking affinity (−7.55 kcal/mol). These findings establish IGF2BP3 as a key driver of LUAD progression and a promising target for immunotherapy and precision medicine. Full article

In this study, a series of novel β-phenylalanine derivatives were synthesised and evaluated for their anticancer activity. The 3-(4-methylbenzene-1-sulfonamido)-3-phenylpropanoic acid (2) was prepared using β-phenylalanine as a core scaffold. The β-amino acid derivative 2 was converted to the corresponding hydrazide 4, which enabled the development of structurally diverse heterocyclic derivatives including pyrrole 5, pyrazole 6, thiadiazole 8, oxadiazole 11, triazoles 9 and 12 with Schiff base analogues 13 and series1,2,4-triazolo [3,4-b][1,3,4]thiadiazines 14. These modifications were designed to enhance chemical stability, solubility, and biological activity. All compounds were initially screened for cytotoxicity against the A549 human lung adenocarcinoma cell line, identifying N-[3-(3,5-dimethyl-1H-pyrazol-1-yl)-3-oxo-1-phenylpropyl]-4-methylbenzenesulfonamide (5) and (E)-N-{2-[4-[(4-chlorobenzylidene)amino]-5-thioxo-4,5-dihydro-1H-1,2,4-triazol-3-yl]-1-phenylethyl}-4-methylbenzenesulfonamide (13b) as the most active. The two lead candidates were further evaluated in H69 and H69AR small cell lung cancer lines to assess activity in drug-sensitive and multidrug-resistant models. Schiff base 13b containing a 4-chlorophenyl moiety, retained potent antiproliferative activity in both H69 and H69AR cells, comparable to cisplatin, while compound 5 lost efficacy in the resistant phenotype. These findings suggest Schiff base derivative 13b may overcome drug resistance mechanisms, a limitation commonly encountered with standard chemotherapeutics such as doxorubicin. These results demonstrate the potential role of β-phenylalanine derivatives, azole-containing sulphonamides, as promising scaffolds for the development of novel anticancer agents, particularly in the context of lung cancer and drug-resistant tumours. Full article

Background/Objectives: Brain metastasis (BM) is a common and often early manifestation in lung adenocarcinoma (LUAD), yet its tumor microenvironment remains poorly defined at the time of initial diagnosis. This study aims to characterize early immune microenvironmental alterations in synchronous BM using spatial proteomic profiling. Methods: We performed digital spatial proteomic profiling using the NanoString GeoMx platform on formalin-fixed paraffin-embedded tissues from five treatment-naïve LUAD patients in whom BM was the initial presenting lesion. Paired primary lung and brain metastatic samples were analyzed across tumor and stromal compartments using 68 immune- and tumor-related protein markers. Results: Spatial profiling revealed distinct expression patterns between primary tumors and brain metastases. Immune regulatory proteins—including IDO-1, PD-1, PD-L1, STAT3, PTEN, and CD44—were significantly reduced in brain metastases (p < 0.01), whereas pS6, a marker of activation-induced T-cell death, was significantly upregulated (p < 0.01). These alterations were observed in both tumor and stromal regions, suggesting a more immunosuppressive and apoptotic microenvironment in brain lesions. Conclusions: This study provides one of the first spatially resolved proteomic characterizations of synchronous BM at initial LUAD diagnosis. Our findings highlight early immune escape mechanisms and suggest the need for site-specific immunotherapeutic strategies in patients with brain metastasis. Full article

Most patients with non-small-cell lung cancer (NSCLC) present with advanced/metastatic disease at diagnosis, and molecular profiling is critical in guiding treatment decisions. This retrospective cohort study aimed to characterize EGFR mutations (EGFRm) in advanced/metastatic NSCLC patients, treatment patterns, and real-world outcomes. Adults diagnosed between 2018 and 2021 and treated at a Comprehensive Care Center were included. Time-to-event outcomes were analyzed using the Kaplan–Meier method. A total of 110 patients were included, with a median age of 69.0 years (range, 37–93), 76.4% female, and 83.2% non-smokers. About 97.3% had adenocarcinomas, with 93.6% at stage IV, 40.9% with ≥ three metastatic sites (brain metastases in 24.5%), 33.6% ECOG 2–4, and 58.2% with an EGFR exon-19 deletion. A minority started supportive care or curative-intent treatment, and 81.8% underwent first-line palliative systemic therapy (TKIs, 91.1%; chemotherapy, 8.9%). Median real-world overall survival (rwOS) was 18.9 months (95% CI, 13.8–28.1). Worse rwOS was observed in patients with ECOG 2–4 versus ECOG 0–1 (10.3 vs. 22.8 months; HR 1.82, 95% CI 1.17–2.85; p = 0.008) and in patients with exon-21 L858R versus exon 19 deletions (15.8 vs. 24.2 months; HR 1.59, 95% CI 1.00–2.54; p = 0.048). In patients treated with palliative systemic treatment, median progression-free survival was 10.9 months (95% CI, 8.8–13.6). This study provides important insights regarding real-world characteristics, treatment patterns, and outcomes from a cohort of EGFRm advanced/metastatic NSCLC patients. Full article

Plant flowers are recognized as a rich source of bioactive phenolic compounds. In this study, for the first time, the recovery of antioxidant phenolic compounds from Cytisus striatus flowers (CF) was optimized using microwave-assisted extraction (MAE). The variables (% of ethanol, temperature, and time) were studied using a response surface methodology (RSM). Extraction efficiency was assessed by total phenol content, total flavonoid content, and the antioxidant capacity through DPPH, ABTS, FRAP, and CUPRAC assays. Additionally, cytotoxicity and anti-inflammatory properties were evaluated in different cell lines. The optimal extraction conditions (87.6% ethanol, 160.8 °C and 8.76 min) yielded extracts rich in phenolics (85.9 mg GAE/g CF) and flavonoids (120.3 mg RE/g CF), with strong antioxidant capacity. LC-MS/MS analysis identified 27 phenolic compounds, including chrysin, apigenin, and quercetin derivatives. Cytotoxicity tests showed that CF extract maintained high viability (>80%) in human embryonic kidney (HEK293T) and human lung adenocarcinoma (A549) cells up to 2000 µg/mL, indicating low cytotoxicity. The anti-inflammatory potential was evidenced by a decrease in IL-1β levels and an increase in IL-10 cytokine production in LPS-stimulated macrophages. These results highlight the great potential of CF as a promising bioresource to obtain value-added compounds for the development of functional foods, nutraceuticals, and cosmetic products. 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