Search Results (102)

Epidermal growth factor receptor (EGFR) mutations occur in approximately 10–20% of Caucasian and up to 50% of Asian patients with oncogene-addicted non-small cell lung cancer (NSCLC). Most frequently, alterations include exon 19 deletions and exon 21 L858R mutations, which confer sensitivity to EGFR tyrosine kinase inhibitors (TKIs). In the last decade, the third-generation EGFR-TKI osimertinib has represented the first-line standard of care for EGFR-mutant NSCLC. However, the development of acquired mechanisms of resistance significantly impacts long-term outcomes and represents a major therapeutic challenge. The mesenchymal–epithelial transition (MET) gene amplification and MET protein overexpression have emerged as prominent EGFR-independent (off-target) resistance mechanisms, detected in approximately 25% of osimertinib-resistant NSCLC. Noteworthy, variability in diagnostic thresholds, which differ between fluorescence in situ hybridization (FISH) and next-generation sequencing (NGS) platforms, complicates its interpretation and clinical applicability. To address MET-driven resistance, several therapeutic strategies have been explored, including MET-TKIs, antibody–drug conjugates (ADCs), and bispecific monoclonal antibodies, and dual EGFR/MET inhibition has emerged as the most promising strategy. In this context, the bispecific EGFR/MET antibody amivantamab has demonstrated encouraging efficacy, regardless of MET alterations. Furthermore, the combination of the ADC telisotuzumab vedotin and osimertinib has been associated with activity in EGFR-mutant, c-MET protein-overexpressing, osimertinib-resistant NSCLC. Of note, several novel agents and combinations are currently under clinical development. The success of these targeted approaches relies on tissue re-biopsy at progression and accurate molecular profiling. Yet, tumor heterogeneity and procedural limitations may challenge the feasibility of re-biopsy, making biomarker-agnostic strategies viable alternatives. Full article

The human genome has a profound impact on human health and disease detection. Carcinoma (cancer) is one of the prominent diseases that majorly affect human health and requires the development of different treatment strategies and targeted therapies based on effective disease detection. Therefore, our research aims to identify biomarkers associated with distinct cancer types (gastric, lung, and breast) using machine learning. In the current study, we have analyzed the human genomic data of gastric cancer, breast cancer, and lung cancer patients using XGB-BIF (i.e., XGBoost-Driven Biomarker Identification Framework for detecting cancer). The proposed framework utilizes feature selection via XGBoost (eXtreme Gradient Boosting), which captures feature interactions efficiently and takes care of the non-linear effects in the genomic data. The research progressed by training XGBoost on the full dataset, ranking the features based on the Gain measure (importance), followed by the classification phase, which employed support vector machines (SVM), logistic regression (LR), and random forest (RF) models for classifying cancer-diseased and non-diseased states. To ensure interpretability and transparency, we also applied SHapley Additive exPlanations (SHAP) and Local Interpretable Model-agnostic Explanations (LIME), enabling the identification of high-impact biomarkers contributing to risk stratification. Biomarker significance is discussed primarily via pathway enrichment and by studying survival analysis (Kaplan–Meier curves, Cox regression) for identified biomarkers to strengthen translational value. Our models achieved high predictive performance, with an accuracy of more than 90%, to classify and link genomic data into diseased (cancer) and non-diseased states. Furthermore, we evaluated the models using Cohen’s Kappa statistic, which confirmed strong agreement between predicted and actual risk categories, with Kappa scores ranging from 0.80 to 0.99. Our proposed framework also achieved strong predictions on the METABRIC dataset during external validation, attaining an AUC-ROC of 93%, accuracy of 0.79%, and Kappa of 74%. Through extensive experimentation, XGB-BIF identified the top biomarker genes for different cancer datasets (gastric, lung, and breast). CBX2, CLDN1, SDC2, PGF, FOXS1, ADAMTS18, POLR1B, and PYCR3 were identified as important biomarkers to identify diseased and non-diseased states of gastric cancer; CAVIN2, ADAMTS5, SCARA5, CD300LG, and GIPC2 were identified as important biomarkers for breast cancer; and CLDN18, MYBL2, ASPA, AQP4, FOLR1, and SLC39A8 were identified as important biomarkers for lung cancer. XGB-BIF could be utilized for identifying biomarkers of different cancer types using genetic data, which can further help clinicians in developing targeted therapies for cancer patients. Full article

Background/Objectives: Lung cancer is the leading cause of cancer-related deaths worldwide. Non-Small-Cell Lung Cancer (NSCLC) accounts for 80–90% of all lung cancers. Antibody-Drug Conjugates (ADCs) represent an expanding targeted therapy option for the treatment of NSCLC. The aim is to perform a systematic literature review to evaluate the efficacy and safety profiles of ADCs currently undergoing clinical trials for the treatment of NSCLC. Methods: The study adhered to the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) statement. Literature searches were conducted in PubMed, ClinicalTrial.gov and Web of Science databases, covering the period from 2014 to 2024. Only randomized and non-randomized phase II-IV clinical trials focusing on ADC-based therapies for adult patients affected by NSCLC were selected. The Revised Cochrane Risk-of-Bias Tool for Randomized Trials (RoB 2.0) and the Risk of Bias in Non-randomized Studies of Interventions (ROBINS-I) were used to evaluate the overall risk of bias in the included randomized and non-randomized studies, respectively. While GRADE (Grading of Recommendations, Assessment, Development and Evaluations) methodology was used to assess the certainty of the evidence. Efficacy endpoints were categorized based on primary outcomes while safety was assessed through the frequency and severity of Treatment-Emergent Adverse Events (TEAEs), and a qualitative summary of the findings was conducted. Results: A total of seven studies, including three randomized, three non-randomized, and one without specific allocation, were included, comprising 1287 patients, with 693 (54%) men, and an average age of 63 years old. Two studies were deemed to have a low risk of bias, while six had a moderate risk or some concerns. Five ADCs were evaluated: trastuzumab deruxtecan (T-DXd), trastuzumab emtansine (T-DM1), telisotuzumab vedotin, patritumab deruxtecan, and datopotamab deruxtecan (Dato-DXd). T-DXd demonstrated superior efficacy in HER2-overexpressing and HER2-mutant NSCLC, with an ORR of 52.9% and 49.0%, respectively. However, HER2-mutant patients exhibited a longer median DOR (16.8 vs. 6.2 months) but a higher incidence of grade ≥ 3 TEAEs (38.6% vs. 22%). T-DM1 showed modest efficacy, with an ORR of 20% in HER2-overexpressing NSCLC and 6.7% in HER2-mutant patients. Dato-DXd demonstrated improved ORR (26.4% vs. 12.8%) and PFS (4.4 vs. 3.7 months) compared to docetaxel. Patritumab deruxtecan achieved an ORR of 39% in EGFR-mutant NSCLC, while telisotuzumab vedotin exhibited limited activity in c-MET-positive NSCLC (ORR 9%, median DOR 7.5 months). Frequency and severity of TEAEs varied across ADCs, with ILD being a major concern, highlighting the need for strict patient monitoring and early intervention to mitigate severe adverse events. Conclusions: ADCs represent a promising advancement in NSCLC treatment, offering targeted therapeutic options beyond conventional chemotherapy and immunotherapy. T-DXd has emerged as the most effective ADC for HER2-mutant NSCLC with manageable safety profile, whereas Dato-DXd provides a viable alternative for TROP2-expressing tumors. While ADCs offer significant clinical benefits, careful patient selection and proactive management of adverse events remain crucial. Ongoing and future trials will further refine the role of ADCs in personalized NSCLC treatment, potentially expanding their tumor-agnostic use to broader patient populations. Full article

Breast cancer is recognized as an aggressive cancer with the highest rate of mortality. Ultrasound imaging is a non-invasive and cost-effective strategy which is most frequently utilized in clinical methods. Especially, in ultrasound scan, breast tumors may appear in blurred and unclear boundaries. Thus, there is a necessity to improve the quality of breast ultrasound images. In this work, we introduce a cycle generative adversarial network (GAN) for translating noisy breast ultrasound images to denoised images. Furthermore, translating denoised images to reconstructed images helps in preserving breast tumor boundaries for better efficacy. To accurately identify the augmented breast tumor images, we consider an ensemble model of pre-trained transfer learning models such as Inception-v3, Densenet121, and XceptionLike. Furthermore, we present an automated boundary extraction using Local Interpretable Model-agnostic Explanations (LIME), providing interpretability for boundary extraction in breast lesions from ultrasound images. Through experimentation, we have achieved 93% of accuracy for the proposed model, and LIME provides better interpretability for each pre-trained model. Furthermore, the proposed model outperforms Vison Transformer (ViT) models. Full article

This review evaluates the current evidence on the use of the Neck Imaging Reporting and Data System (NI-RADS) for the surveillance and early detection of recurrent head and neck cancers. NI-RADS offers a standardized, structured framework specifically tailored for post-treatment imaging, aiding radiologists in differentiating between residual tumors, scar tissue, and post-surgical changes. NI-RADS demonstrated a strong diagnostic performance across multiple studies, with high sensitivity and specificity reported in detecting recurrent tumors at primary and neck sites. Despite these strengths, limitations persist, including a high frequency of indeterminate results and variability in di-agnostic concordance across imaging modalities (computed tomography, magnetic resonance imaging (MRI), positron emission tomography(PET)). The review also highlights the NI-RADS’s reproducibility, showing high inter- and intra-reader agreements across different imaging techniques, although some modality-specific differences were observed. While it demonstrates strong diagnostic performance and good reproducibility across imaging modalities, attention is required to address indeterminate imaging findings and the limitations of modality-specific variations. Future studies should focus on integrating advanced imaging characteristics, such as diffusion-weighted imaging and PET/MRI fusion techniques, to further enhance NI-RADS’s diagnostic accuracy. Continuous efforts to refine NI-RADS protocols and imaging interpretations will ensure more consistent detection of recurrences, ultimately improving clinical outcomes in head and neck cancer management. Full article

The increasing complexity and importance of medical data in improving patient care, advancing research, and optimizing healthcare systems led to the proposal of this study, which presents a novel methodology by evaluating the sensitivity of artificial intelligence (AI) algorithms when provided with real data, synthetic data, a mix of both, and synthetic features. Two medical datasets, the Pima Indians Diabetes Database (PIDD) and the Breast Cancer Wisconsin Dataset (BCWD), were used, employing the Gaussian Copula Synthesizer (GCS) and the Synthetic Minority Oversampling Technique (SMOTE) to generate synthetic data. We classified the new datasets using fourteen machine learning (ML) models incorporated into PyCaret AutoML (Automated Machine Learning) and two deep neural networks, evaluating performance using accuracy (ACC), F1-score, Area Under the Curve (AUC), Matthews Correlation Coefficient (MCC), and Kappa metrics. Local Interpretable Model-agnostic Explanations (LIME) provided the explanation and justification for classification results. The quality and content of the medical data are very important; thus, when the classification of original data is unsatisfactory, a good recommendation is to create synthetic data with the SMOTE technique, where an accuracy of 0.924 is obtained, and supply the AI algorithms with a combination of original and synthetic data. Full article

Accurate skin cancer classification is essential for early diagnosis and effective treatment planning, enabling timely interventions and improved patient outcomes. In this paper, the performance of four pre-trained models—two convolutional neural networks (ResNet50 and VGG19) and two vision transformers (ViT-b16 and ViT-b32)—is evaluated in distinguishing malignant from benign skin cancers using a publicly available dermoscopic dataset. Among these models, ResNet50 achieved the highest performance across all the evaluation metrics, with accuracy, precision, and recall of 89.09% and an F1 score of 89.08%, demonstrating its ability to effectively capture complex patterns in skin lesion images. While the other models produced competitive results, ResNet50 exhibited superior robustness and consistency. To enhance model interpretability, two eXplainable Artificial Intelligence (XAI) techniques, Local Interpretable Model-Agnostic Explanations (LIME) and integrated gradients, were employed to provide insights into the decision-making process, fostering trust in automated diagnostic systems. These findings underscore the potential of deep learning for automated skin cancer classification and highlight the importance of model transparency for clinical adoption. As AI technology continues to evolve, its integration into clinical workflows could improve diagnostic accuracy, reduce the workload of healthcare professionals, and enhance patient outcomes. Full article

Transcription factors (TFs) and microRNAs (miRNAs) are fundamental regulators of gene expression, cell state, and biological processes. This study investigated whether a small subset of TFs and miRNAs could accurately predict genome-wide gene expression. We analyzed 8895 samples across 31 cancer types from The Cancer Genome Atlas and identified 28 miRNA and 28 TF clusters using unsupervised learning. Medoids of these clusters could differentiate tissues of origin with 92.8% accuracy, demonstrating their biological relevance. We developed Tissue-Agnostic and Tissue-Aware models to predict 20,000 gene expressions using the 56 selected medoid miRNAs and TFs. The Tissue-Aware model attained an $R^2$ of 0.70 by incorporating tissue-specific information. Despite measuring only 1/400th of the transcriptome, the prediction accuracy was comparable to that achieved by the 1000 landmark genes. This suggests the transcriptome has an intrinsically low-dimensional structure that can be captured by a few regulatory molecules. Our approach could enable cheaper transcriptome assays and analysis of low-quality samples. It also provides insights into genes that are heavily regulated by miRNAs/TFs versus alternative mechanisms. However, model transportability was impacted by dataset discrepancies, especially in miRNA distribution. Overall, this study demonstrates the potential of a biology-guided approach for robust transcriptome representation. Full article

Early diagnosis is critical for improving outcomes in cancer patients; however, the application of diagnostic markers derived from serum proteomic screening remains challenging. Artificial intelligence (AI), encompassing deep learning and machine learning (ML), has gained increasing prominence across various scientific disciplines. In this study, we utilized cervical cancer (CC) as a model to develop an AI-driven pipeline for the identification and validation of serum biomarkers for early cancer diagnosis, leveraging mass spectrometry-based proteomics data. By processing and normalizing serum polypeptide differential peaks from 240 patients, we employed eight distinct ML algorithms to classify and analyze these differential polypeptide peaks, subsequently constructing receiver operating characteristic (ROC) curves and confusion matrices. Key performance metrics, including accuracy, precision, recall, and F1 score, were systematically evaluated. Furthermore, by integrating feature importance values, Shapley values, and local interpretable model-agnostic explanation (LIME) values, we demonstrated that the diagnostic area under the curve (AUC) achieved by our multi-dimensional learning models approached 1, significantly outperforming the diagnostic AUC of single markers derived from the PRIDE database. These findings underscore the potential of proteomics-driven integrated machine learning as a robust strategy to enhance early cancer diagnosis, offering a promising avenue for clinical translation. Full article

This review comprehensively analyzes the current landscape of tumor-agnostic therapies in oncology. Tumor-agnostic therapies are designed to target specific molecular alterations rather than the primary site of the tumor, representing a shift in cancer treatment. We discuss recent approvals by regulatory agencies such as the FDA and EMA, highlighting therapies that have demonstrated efficacy across multiple cancer types sharing common alterations. We delve into the trial methodologies that underpin these approvals, emphasizing innovative designs such as basket trials and umbrella trials. These methodologies present unique advantages, including increased efficiency in patient recruitment and the ability to assess drug efficacy in diverse populations rapidly. However, they also entail certain challenges, including the need for robust biomarkers and the complexities of regulatory requirements. Moreover, we examine the promising prospects for developing therapies for rare cancers that exhibit common molecular targets typically associated with more prevalent malignancies. By synthesizing these insights, this review underscores the transformative potential of tumor-agnostic therapies in oncology. It offers a pathway for personalized cancer treatment that transcends conventional histology-based classification. Full article

Background: Early-phase clinical trials offer a unique opportunity for patients with cancer. These trials often mandate biopsies to collect tumor tissue for research purposes, requiring patients to undergo invasive procedures. Some trials mandate molecular prescreening, but the success of these analyses relies on the quality and quantity of the tested materials. Additionally, bioptic procedures may result in complications. Methods: We retrospectively examined the records of patients referred to the Early Drug Development (EDD) Unit of the European Institute of Oncology who underwent biopsies for research purposes between January 2014 and December 2022. Our objective was to assess the safety of biopsy procedures and adequacy of the samples for NGS testing. Results: In total, 355 out of 731 patients (48.6%) underwent protocol-mandated biopsies. The most frequent sites of biopsy were the liver, lymph nodes, skin, and breast. Histological diagnosis was achieved in 349 (98%) patients, and NGS testing was successfully conducted in 111/127 (88.4%) cases. Of the 16 unsuccessful NGS attempts, 9 were performed on liver tissue. Unsuccessful NGS testing was attributed to poor sample quality and/or quantity, and the success rate varied significantly based on the specific tests attempted. Complications occurred in a small proportion of patients (4.8%), and none were serious. Conclusions: The non-negligible failure rate of NGS testing highlights the crucial need for implementing specific guidelines and Standard Operating Procedures for samples intended for NGS. With the use of a risk-based biopsy framework to guide clinical decisions, procedure-related complications may be minimized. Full article

Circulating tumor DNA (ctDNA) is a promising cancer biomarker, but accurately detecting tumor mutations in cell-free DNA (cfDNA) is challenging due to their low frequency and sequencing errors. Our study benchmarked Mutect2, VarScan2, shearwater, and DREAMS-vc using deep targeted sequencing of cfDNA with Unique Molecular Identifiers (UMIs) from 111 colorectal cancer patients. Performance was assessed at both the mutation level (distinguish tumor variants from errors) and the sample level (detect if an individual has cancer). Additionally, we investigated the effects of various UMI grouping and consensus strategies. The shearwater-AND variant calling method demonstrated the highest precision in detecting tumor-derived mutations from plasma, and reached the highest ROC-AUC of 0.984 for sample classification in tumor-informed cfDNA analyses. DREAMS-vc exhibited the highest ROC-AUC of 0.808 for sample classification in tumor-agnostic studies. We also found that sequencing depth differences in PBMCs could lead to false positives, particularly with VarScan2 and Mutect2, which was addressed by downsampling to equivalent mean depths. Additionally, network-based UMI grouping methods outperformed those using identical UMIs when all reads were retained. Our findings emphasize that the optimal variant caller depends on the study context—whether focused on mutation or sample classification, and whether conducted under tumor-informed or tumor-agnostic conditions. Full article

NTRK fusions are oncogenic drivers for multiple tumor types. Therefore, the development of selective tropomyosin receptor kinase (TRK) inhibitors, including larotrectinib and entrectinib, has been transformative in the context of clinical management, given the high rates of responses to these drugs, including intracranial responses in patients with brain metastases. Given their promising activity in pan-cancer cohorts, larotrectinib and entrectinib received U.S. Food and Drug Administration (FDA) and European Medicines Agency (EMA) approval for tissue-agnostic indications in patients with advanced solid tumors harboring NTRK fusions. The safety profiles for both drugs are quite manageable, although neurotoxicity driven by the on-target inhibition of normal NTRK can be a concern. Also, on- and off-target resistance mechanisms can arise during therapy with TRK inhibitors, but they can be addressed with the use of combination therapy and next-generation TRK inhibitors. More recently, the FDA approved the use of repotrectinib, a second-generation TRK inhibitor, in patients with NTRK fusions, based on data suggesting clinical efficacy and safety, which could offer another tool for the treatment of NTRK-altered cancers. In this review, we summarize the current evidence related to the use of TRK inhibitors in the tissue-agnostic setting. We also elaborate on the safety profiles and resistance mechanisms from a practical perspective. Full article

The regulatory approvals of tumor-agnostic therapies have led to the re-evaluation of the drug development process. The conventional models of drug development are histology-based. On the other hand, the tumor-agnostic drug development of a new drug (or combination) focuses on targeting a common genomic biomarker in multiple cancers, regardless of histology. The basket-like clinical trials with multiple cohorts allow clinicians to evaluate pan-cancer efficacy and toxicity. There are currently eight tumor agnostic approvals granted by the Food and Drug Administration (FDA). This includes two immune checkpoint inhibitors, and five targeted therapy agents. Pembrolizumab is an anti-programmed cell death protein-1 (PD-1) antibody that was the first FDA-approved tumor-agnostic treatment for unresectable or metastatic microsatellite instability-high (MSI-H) or deficient mismatch repair (dMMR) solid tumors in 2017. It was later approved for tumor mutational burden-high (TMB-H) solid tumors, although the TMB cut-off used is still debated. Subsequently, in 2021, another anti-PD-1 antibody, dostarlimab, was also approved for dMMR solid tumors in the refractory setting. Patients with fusion-positive cancers are typically difficult to treat due to their rare prevalence and distribution. Gene rearrangements or fusions are present in a variety of tumors. Neurotrophic tyrosine kinase (NTRK) fusions are present in a range of pediatric and adult solid tumors in varying frequency. Larotrectinib and entrectinib were approved for neurotrophic tyrosine kinase (NTRK) fusion-positive cancers. Similarly, selpercatinib was approved for rearranged during transfection (RET) fusion-positive solid tumors. The FDA approved the first combination therapy of dabrafenib, a B-Raf proto-oncogene serine/threonine kinase (BRAF) inhibitor, plus trametinib, a mitogen-activated protein kinase (MEK) inhibitor for patients 6 months or older with unresectable or metastatic tumors (except colorectal cancer) carrying a BRAF^V600E mutation. The most recent FDA tumor-agnostic approval is of fam-trastuzumab deruxtecan-nxki (T-Dxd) for HER2-positive solid tumors. It is important to identify and expeditiously develop drugs that have the potential to provide clinical benefit across tumor types. Full article

Cancer accounted for 10 million deaths in 2020, nearly one in every six deaths annually. Despite advancements, the contemporary clinical management of human neoplasms faces a number of challenges. Surgical removal of tumor tissues is often not possible technically, while radiation and chemotherapy pose the risk of damaging healthy cells, tissues, and organs, presenting complex clinical challenges. These require a paradigm shift in developing new therapeutic modalities moving towards a more personalized and targeted approach. The tumor-agnostic philosophy, one of these new modalities, focuses on characteristic molecular signatures of transformed cells independently of their traditional histopathological classification. These include commonly occurring DNA aberrations in cancer cells, shared metabolic features of their homeostasis or immune evasion measures of the tumor tissues. The first dedicated, FDA-approved tumor-agnostic agent’s profound progression-free survival of 78% in mismatch repair-deficient colorectal cancer paved the way for the accelerated FDA approvals of novel tumor-agnostic therapeutic compounds. Here, we review the historical background, current status, and future perspectives of this new era of clinical oncology. Full article