Non-small cell lung cancer has seen dramatic treatment expansion beyond chemotherapy. The current landscape includes antibody-drug conjugates, next-generation tyrosine kinase inhibitors, immunotherapy, liquid biopsy, and AI-assisted diagnostics, each representing a distinct frontier in the effort to improve outcomes for patients with this leading cause of cancer death.
This review organizes developments around three themes: therapeutic innovations and resistance mechanisms, diagnostic and biomarker advances, and implementation challenges and equity gaps. Together these themes reveal a field at an inflection point where scientific progress is outpacing the ability to deliver it equitably to patients.
Despite remarkable innovation, only 11 to 34 percent of eligible patients ultimately receive a genomically matched therapy, underscoring how systemic and biological barriers continue to limit the real-world impact of precision oncology.
Antibody-drug conjugates represent a major new treatment class for refractory NSCLC. Agents like HER3-DXd deliver potent cytotoxic payloads directly to EGFR-TKI resistant tumor cells, while TROP-2-targeting agents such as sacituzumab govitecan and datopotamab deruxtecan are under active investigation for tumors that have failed chemotherapy and immunotherapy.
Next-generation tyrosine kinase inhibitors continue to redefine treatment standards for oncogene-driven lung cancer. Lorlatinib for ALK-positive disease achieved a five-year progression-free survival of 60 percent versus only 8 percent for crizotinib in the CROWN trial, while tepotinib targets MET exon 14 skipping and glecirasib addresses the previously undruggable KRAS G12C mutation with an objective response rate of 47.9 percent in a phase 2b trial.
Each new agent class carries a distinct toxicity profile that must be weighed in treatment planning. Lorlatinib causes lipid abnormalities, MET inhibitors produce edema and creatinine elevation, and HER2-targeting ADCs carry a risk of interstitial lung disease, requiring specialized monitoring and informing how these drugs can be combined.
Emerging platforms including bispecific antibodies targeting EGFR and c-MET simultaneously, antisense oligonucleotides to dampen regulatory T-cell activity, nanoparticle drug delivery systems, and autologous T-cell therapies are all in early clinical testing, expanding the future therapeutic toolkit beyond small molecules and standard antibodies.
Comprehensive genomic profiling using next-generation sequencing now identifies actionable mutations across a wide range of drivers. Current guidelines mandate testing for EGFR, ALK, ROS1, BRAF, NTRK, MET exon 14, RET, and KRAS G12C mutations alongside PD-L1 expression in newly diagnosed advanced NSCLC before selecting first-line therapy.
Liquid biopsy for circulating tumor DNA is increasingly integrated into clinical workflows as a non-invasive alternative or complement to tissue biopsy, particularly valuable when tissue is scarce or to monitor treatment response over time. Studies have shown that early ctDNA clearance predicts better outcomes, while high ctDNA variant allele frequency and TP53 co-mutations correlate with worse survival in patients receiving targeted therapy.
AI applied to histopathology slides shows early promise in predicting genomic mutations from routine hematoxylin and eosin stained tissue sections, potentially allowing gene status prediction without expensive molecular testing. However, current performance remains suboptimal for clinical adoption, lacking robust external validation across diverse populations.
Beyond genomics, transcriptomic and proteomic biomarkers are emerging as predictive tools. High natural killer cell counts and specific monocyte populations predicted longer progression-free survival with pembrolizumab in patients with PD-L1 below 50 percent, and low plasma L-arginine levels correlated with worse overall survival in patients receiving immune checkpoint inhibitors, with a hazard ratio of 3.03 in one analysis.
Acquired resistance is the most fundamental barrier to durable benefit from targeted therapies. On-target resistance mutations such as EGFR C797S and ALK G1202R directly inhibit drug binding, while off-target bypass pathway activation through MET, HER2, or AXL amplification allows tumor cells to maintain growth signaling through alternative routes.
Phenotypic transformation from non-small cell to small cell lung cancer represents another resistance mechanism that cannot be overcome by continued targeted therapy but requires an entirely different treatment approach, illustrating the biological plasticity of lung cancer under therapeutic pressure.
Immunotherapy resistance involves a different set of mechanisms centered on the tumor microenvironment. STK11 and KEAP1 mutations create an immunosuppressive milieu, and low L-arginine levels further suppress anti-tumor immune activity, with STK11/LKB1 co-mutations shown to worsen prognosis dramatically in KRAS-mutated early-stage disease with a hazard ratio of 3.85 for disease-free survival.
Strategies to overcome resistance include proactive combinations such as osimertinib plus ramucirumab, which improved median progression-free survival from 15.6 to 24.8 months, and the use of ADCs which can deliver cytotoxic payloads to tumor cells independently of the original oncogenic signaling pathway that has been bypassed.
First-line therapy in stage IV NSCLC is now rigidly stratified by molecular subtype. Osimertinib is standard for EGFR mutations, lorlatinib or alectinib for ALK rearrangements, entrectinib for ROS1 fusions, and pembrolizumab monotherapy for tumors with PD-L1 expression of 50 percent or higher without an actionable driver.
KRAS G12C mutations, found in approximately 13 percent of NSCLC patients, are now addressable with approved covalent inhibitors sotorasib and adagrasib, though first-line data are still maturing. STK11 and KEAP1 co-mutations in this group predict poorer immunotherapy response, complicating treatment sequencing decisions.
For early-stage resectable disease, neoadjuvant and perioperative immunotherapy combinations are emerging as new standards alongside surgery, with efficacy potentially modulated by actionable genomic alterations. Unresectable stage III disease is managed with concurrent chemoradiation followed by consolidative durvalumab as the established benchmark.
The positioning of emerging agents within these established algorithms is continuously evolving. HER3-DXd is specifically under investigation after EGFR TKI failure, T-DXd combined with immunotherapy is being evaluated for HER2-mutant advanced disease, and TROP-2-targeting ADCs are being developed for refractory non-oncogene-addicted tumors after chemoimmunotherapy.
Precision oncology is failing to reach the majority of eligible patients in practice. In real-world cohorts, only about half of advanced NSCLC patients receive complete biomarker profiling, and an estimated 66 to 89 percent of those with actionable alterations do not receive a matched therapy due to drug access issues, restrictive trial eligibility, or clinical deterioration.
The cost of comprehensive genomic profiling, ranging from approximately 1,500 to 3,900 euros per patient, combined with a lack of next-generation sequencing infrastructure in resource-limited settings creates significant geographic and socioeconomic disparities. Molecular tumor boards improve target identification but succeed in recommending matched therapy in only 19 to 61 percent of cases while facing lengthy turnaround times and complex variant interpretation challenges.
Elderly patients, who constitute a large proportion of the NSCLC population, are frequently underrepresented in clinical trials and require tailored assessment beyond chronological age. Comprehensive geriatric assessment evaluating functional status, comorbidities, and polypharmacy is crucial. Evidence suggests targeted therapies like osimertinib are effective and tolerable in older adults, but combination regimens may incur excessive toxicity.
Ethnic differences in mutation prevalence also matter. EGFR mutations affect 40 to 55 percent of Asian patients but only 10 to 15 percent of Caucasian patients, justifying different empirical approaches and trial designs. Rare subtypes such as hepatoid adenocarcinoma of the lung lack prospective trial data entirely, relying on retrospective case series to inform management.
The field must prioritize overcoming resistance through combinatorial strategies. Bispecific antibodies targeting dual pathways, next-generation ADCs, and cellular therapies must be tested in pragmatic combination and sequencing studies designed around the specific resistance mechanisms they aim to circumvent.
Biomarker refinement is essential, with AI-driven multi-omics integration needed to discover validated predictive signatures that go beyond imperfect markers like PD-L1 expression. ctDNA dynamics for minimal residual disease detection and early resistance monitoring need prospective validation across diverse clinical pathways before they can guide routine treatment decisions.
Operationally, deploying cost-effective next-generation sequencing panels, expanding molecular tumor board access, developing equitable reimbursement models, and broadening liquid biopsy use globally are all necessary to close the gap between scientific progress and patient access.
The review concludes that while the NSCLC therapeutic arsenal has never been more potent or precise, realizing its full potential requires dismantling biological, technological, and systemic barriers simultaneously. Success depends on integrating deep molecular science with pragmatic solutions for accessible profiling, tailored toxicity management, and inclusive trial design that encompasses elderly patients and rare subtypes.