Background Osimertinib is a standard first-line treatment for EGFR-mutant Non-Small Cell Lung Cancer (NSCLC), but acquired resistance remains a clinical challenge. Resistance involves genetic mechanisms and non-genetic cell-state transitions, including epithelial-to-mesenchymal (EMT) plasticity and drug-tolerant persister phenotypes. Patient-derived models established at disease progression may better capture clinically relevant resistance mechanisms. Methods We established LIBM-ADK-11 from the pleural effusion of a patient with EGFR exon 19p.Leu747_Pro753delinsSer-mutant NSCLC progressing on osimertinib. Cells were studied at low and high passages and fractionated by CD44 expression into CD44-positive/P2 and CD44-negative/P4 subpopulations. EGFR mutation status, CD44 expression, transcriptomics, growth, osimertinib sensitivity, EGFR signaling, sphere formation, cell-cycle distribution and pemetrexed response were evaluated. Results LIBM-ADK-11 retained the EGFR mutation during long-term culture. CD44 sorting identified two phenotypically distinct cell states. CD44-positive P2 cells were enriched for extracellular matrix remodeling, EMT-like, inflammatory, CAF-like and macrophage/TAM-mimicry programs, with increased glycolysis-related signatures, slower growth, higher metabolic activity and osimertinib resistance. CD44-negative P4 cells displayed a more epithelial profile, cholesterol-associated gene expression and greater osimertinib sensitivity. Osimertinib inhibited EGFR phosphorylation in both populations, indicating that P2 resistance was not explained by persistent EGFR signaling. Long-term culture favored loss of CD44-positive cells, whereas osimertinib delayed this loss and preferentially depleted the CD44-negative fraction. Pemetrexed partially reduced both compartments. Conclusions LIBM-ADK-11 is a clinically relevant model of osimertinib-resistant EGFR-mutant NSCLC in which CD44-positive drug-tolerant and CD44-negative drug-sensitive states coexist. These findings highlight phenotypic heterogeneity as a potential determinant of osimertinib resistance and support strategies addressing distinct cellular states at disease progression.
Background Adverse events (AEs) are routinely reported in placebo arms of lung cancer trials, yet these data have seldom been systematically synthesized. This meta-analysis aimed to quantify the incidence of AEs in placebo arms of phase III randomized controlled trials (RCTs) for non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC). Methods We systematically searched PubMed, Embase, Ovid Medline, and Cochrane Library for phase III RCTs evaluating targeted therapy and/or immunotherapy in NSCLC or SCLC that included a placebo arm. Pooled incidence was estimated using random-effects models for any-grade and grade 3 or 4 AEs, treatment modifications (interruption, dose reduction, or discontinuation) due to AEs, and common AEs. Results Twenty-six RCTs comprising 7641 placebo-treated patients were included. The pooled incidence of any-grade AEs was 84% (95%CI: 78–90%, 95%PI: 54–100%), with grade 3 or 4 AEs occurring in 25% (95%CI: 19–32%, 95%PI: 4–57%). The most frequent any-grade AEs were fatigue (20%), dyspnea (20%), and cough (20%). Grade 3 or 4 AEs most commonly included fatigue (4%), dyspnea (3%), pneumonia (2%), and hypertension (2%). Treatment interruption or dose reduction due to AEs occurred in 10% (95%CI: 6–15%, 95%PI: 0–36%), and discontinuation in 5% (95%CI: 4–8%, 95%PI: 0–17%). Conclusion Placebo-treated patients with lung cancer experience substantial AE burden, may attributable to disease itself and its prior treatment. The scarcity of nocebo-related data underscores the urgent need for future studies to investigate its mechanisms and systematically assess negative expectancy in patients.
Objectives The eighth edition TNM classification measures part-solid lung adenocarcinoma by the solid component alone. We examined whether CT-measured size carries the same pathologic and prognostic meaning in part-solid and solid adenocarcinoma. Materials and methods We retrospectively analyzed 3857 consecutive patients with resected non-mucinous lung adenocarcinoma (2011–2022), using phenotype and size from preoperative CT reports. Shape-constrained additive models estimated the size-dependent prevalence of seven adverse features, including high-grade (HG) patterns, micropapillary component and spread through air spaces (STAS). Five-year cumulative incidence of recurrence (CIR) was estimated with death as a competing risk. Results Tumors were part-solid in 2100 and solid in 1757 patients. In part-solid tumors, HG patterns, micropapillary component and STAS rose most steeply at 1.25–1.36 cm of total size (0.85–0.93 cm of solid component); in solid tumors, HG patterns and STAS were already frequent at the smallest measured sizes. Lymphovascular, pleural and nodal involvement became prevalent at larger sizes in both phenotypes. Among 3580 patients analyzed for recurrence (467 recurrences; median follow-up, 56.6 months), solid tumors had higher CIR than part-solid tumors within each matched solid-component size stratum (feature-present tumors > 2.5 cm: 42.7% vs 19.6%). In part-solid tumors, 5-year CIR was 1.4% when HG morphology was the most advanced feature, 11.0% with local invasion and 50.0% with nodal involvement. Conclusions Measuring the solid component narrows but does not close the difference in pathologic risk between part-solid and solid adenocarcinoma of the same size. These associations warrant external validation and evaluation in preoperative risk models.
