Scope: This comprehensive review published in Signal Transduction and Targeted Therapy covers the full landscape of non-small cell lung cancer (NSCLC) molecular pathology, from epidemiology and genomic drivers to epigenetics, spatial omics, and the latest targeted therapies.
Clinical Burden: NSCLC accounts for 85-90% of all lung cancers and remains the leading cause of cancer death worldwide, with approximately 2 million new cases annually and an overall 5-year survival below 20%.
Molecular Revolution: Over the past two decades, the identification of actionable driver mutations - including EGFR, KRAS, ALK, ROS1, BRAF, MET, RET, and NTRK - has transformed NSCLC from a histology-driven to a molecularly stratified disease.
Treatment Paradigm Shift: Targeted therapies and immunotherapy have replaced cytotoxic chemotherapy as first-line treatment for many NSCLC subtypes, requiring comprehensive molecular profiling of every newly diagnosed patient.
EGFR Mutations: EGFR mutations (exon 19 deletions, L858R) occur in 10-35% of NSCLC cases (higher in Asian populations) and are treated with tyrosine kinase inhibitors (TKIs) including erlotinib, gefitinib, afatinib, and the third-generation osimertinib, which also covers T790M resistance mutations.
KRAS Mutations: KRAS mutations (particularly G12C) are the most common driver in Western NSCLC populations (25-30%). Sotorasib and adagrasib are the first approved KRAS G12C inhibitors, representing a breakthrough after decades of KRAS being considered undruggable.
ALK and ROS1 Rearrangements: ALK rearrangements (3-5% of NSCLC) and ROS1 rearrangements (1-2%) define patient subgroups highly responsive to ALK/ROS1 inhibitors like crizotinib, alectinib, brigatinib, and lorlatinib, often achieving durable responses.
Emerging Drivers: BRAF V600E mutations (2-3%), MET exon 14 skipping (3-4%), RET fusions (1-2%), and NTRK fusions (rare) each define distinct molecular subgroups with approved or emerging targeted therapies, underscoring the importance of comprehensive molecular profiling.
PD-1/PD-L1 Blockade: Anti-PD-1 (pembrolizumab, nivolumab) and anti-PD-L1 (atezolizumab, durvalumab) therapies have become standard of care for NSCLC patients without actionable driver mutations, achieving durable responses in a subset of patients.
PD-L1 Expression as Biomarker: PD-L1 tumor proportion score (TPS) predicts response to pembrolizumab monotherapy, with TPS >/= 50% identifying patients likely to benefit from first-line single-agent pembrolizumab.
TMB and Emerging Predictors: Tumor mutational burden (TMB), microsatellite instability (MSI), and STK11/KEAP1 co-mutations influence immunotherapy response, though PD-L1 remains the dominant clinical biomarker for treatment decisions.
Combination Strategies: Combining checkpoint inhibitors with platinum-based chemotherapy improves outcomes across PD-L1 expression levels, and dual checkpoint blockade (PD-1 + CTLA-4) shows benefit in selected patient subsets.
Epigenetic Alterations: DNA methylation, histone modification, and non-coding RNA dysregulation contribute to NSCLC pathogenesis, influence tumor heterogeneity, and can drive resistance to targeted therapies through epigenetic reprogramming.
Epigenetic Therapies: HDAC inhibitors, DNMT inhibitors, and EZH2 inhibitors are being evaluated in combination with targeted agents and immunotherapy to overcome epigenetic resistance mechanisms in advanced NSCLC.
Spatial Transcriptomics: Spatial omics technologies enable simultaneous measurement of gene expression and spatial location within the tumor microenvironment, revealing how tumor cell neighborhoods influence treatment response and immune evasion.
Single-Cell Analysis: Single-cell RNA sequencing and proteomics have uncovered intratumoral heterogeneity - the coexistence of distinct cancer cell subpopulations - as a key driver of treatment resistance and recurrence after initial targeted therapy response.
On-Target Resistance: Point mutations in the drug target itself - such as EGFR T790M (conferring erlotinib/gefitinib resistance) and EGFR C797S (conferring osimertinib resistance) - represent a primary resistance mechanism that can be countered by next-generation inhibitors.
Bypass Signaling: Activation of alternative signaling pathways (MET amplification, HER2 amplification, KRAS mutations) can bypass inhibited oncogenic drivers, rendering targeted therapies ineffective.
Histological Transformation: A subset of EGFR-mutant NSCLC transforms to small-cell lung cancer upon acquiring resistance to EGFR TKIs, a phenomenon that requires re-biopsy and represents a distinct treatment challenge.
Combating Resistance: Liquid biopsy (cfDNA analysis) enables non-invasive detection of resistance mutations at relapse, informing the choice of subsequent-line therapy without requiring re-biopsy of potentially inaccessible metastatic lesions.
Circulating Tumor DNA: ctDNA in blood plasma can detect NSCLC driver mutations, monitor treatment response, and identify resistance mechanisms without tissue biopsy, enabling dynamic molecular profiling throughout the treatment course.
Minimal Residual Disease: Post-surgical ctDNA detection predicts relapse risk in early-stage NSCLC, identifying patients who may benefit from adjuvant therapy even when imaging remains normal.
Circulating Tumor Cells: CTCs provide phenotypic information about circulating cancer cells, complementing ctDNA-based genotypic analysis and potentially revealing mechanisms of hematogenous metastasis.
Exosomes and Other Biomarkers: Tumor-derived exosomes, cell-free RNA, and methylation biomarkers in blood and other biofluids are being evaluated as complementary or orthogonal liquid biopsy analytes with distinct biological information.
Antibody-Drug Conjugates: ADCs like trastuzumab deruxtecan (HER2-targeted) and patritumab deruxtecan (HER3-targeted) are emerging as effective options for NSCLC patients whose tumors express these targets, independent of mutation status.
Bispecific Antibodies: Bispecific T-cell engagers and other bispecific antibodies are in early trials for NSCLC, aiming to enhance tumor-directed immune killing by simultaneously targeting both tumor antigens and immune effector cells.
AI-Assisted Pathology: Machine learning algorithms applied to whole-slide images and multi-omics data are improving molecular subtype classification, biomarker quantification, and response prediction, beginning to augment routine pathological analysis.
Personalized Combinations: The convergence of molecular profiling, spatial omics, liquid biopsy, and AI is enabling increasingly precise treatment selection, pointing toward a future of individualized combination regimens tailored to each patient's unique tumor biology.