EGFR exon 20 insertion mutations represent a distinct subset of non-small cell lung cancer (NSCLC) with limited sensitivity to earlier-generation EGFR tyrosine kinase inhibitors (TKIs). CLN081 (zipalertinib; TAS6417) is a novel covalent EGFR TKI under development with activity against EGFR exon 20 insertion variants. We report the case of a 44-year-old man with stage IVB lung adenocarcinoma harboring an EGFR exon 20 insertion mutation who received multiple lines of systemic therapy, including CLN081. After eight months of treatment with CLN081, liquid biopsy revealed the emergence of a secondary EGFR C797S mutation (p.Cys797Ser; variant allele frequency [VAF] 0.05%), concurrent with a reduction in the exon 20 insertion allele frequency from 35.1% to 0.5%, accompanied by disease progression. Although C797S-mediated resistance has been hypothesized and demonstrated in preclinical models of CLN081 exposure, this represents, to our knowledge, the first reported in vivo identification of an EGFR C797S resistance mutation following CLN081 therapy. This case provides clinically relevant insight into resistance mechanisms associated with emerging EGFR exon 20-targeted therapies.

BackgroundOsimertinib is an approved first-line therapy for epidermal growth factor receptor (EGFR)-mutated non-small cell lung cancer (NSCLC). However, beyond the identification of common EGFR mutations, additional pathological or molecular factors that predict treatment response, survival outcomes, or toxicity remain limited.Materials and MethodsThis retrospective study analyzed data from a registry of NSCLC patients with EGFR mutations treated with first-line osimertinib between March 2017 and December 2024. Variant allele frequency (VAF) was evaluated as a potential predictive factor for overall survival (OS), progression-free survival (PFS), and adverse events (AEs).ResultsAmong 147 eligible patients, the mean OS was 25.5 months and the mean PFS was 21.4 months. Patients with VAF ≥30% exhibited improved outcomes compared to those with VAF <30%, with mean OS of 31.4 months versus 19.7 months (p = 0.022), and mean PFS of 25.0 months versus 18.2 months (p = 0.234). Similar trends were observed across EGFR exon 19 deletion and exon 21 L858R subgroups (p = 0.056). When comparing toxicity profiles, the overall AE rates were similar between high-VAF and low-VAF patients. However, several statistically significant differences were noted: diarrhea (21.3% vs. 5.7%, p = 0.005) and dyspnea (16.4% vs. 3.4%, p = 0.0085) were more frequent in the high-VAF group, while anemia (9.2% vs. 3.3%, p = 0.03) and creatinine elevation (5.7% vs. 1.6%, p = 0.01) occurred more commonly in the low-VAF group.ConclusionHigher EGFR VAF was significantly associated with improved overall survival in patients with EGFR-mutated NSCLC treated with first-line osimertinib and showed a numerical trend toward longer progression-free survival. Similar patterns were observed across key molecular subgroups. Additionally, this study is the first to report potential VAF-related differences in adverse event patterns, suggesting that VAF may have relevance not only for efficacy but also for toxicity characterization. These findings support the potential role of VAF as a prognostic biomarker in EGFR-mutant NSCLC and warrant further prospective validation.

Yaacov, Levi, and Peled argue that AI-powered clinical trial matching has achieved near-expert technical accuracy yet fails to increase patient enrollment because the dominant bottleneck is systemic—involving logistics, workflow misalignment, and consent burden—not informational. They propose reframing AI’s role from matching engine to trial facilitation system.

BACKGROUND:∼60%-70% of patients with metastatic non-small cell lung cancer (NSCLC) have relatively low tumor mutational burden (TMB), with limited biomarker-guided immunotherapy options beyond programmed death-ligand 1 (PD-L1) expression. APOBEC (apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like) mutational signatures, defined by characteristic genomic footprints, are associated with higher TMB and a better response to immune checkpoint inhibitors (ICIs). However, whether APOBEC independently predicts ICI response in TMB-low tumors has not been established. MATERIALS AND METHODS:We analyzed 857 ICI-treated and 1328 ICI-untreated TMB-low (<10 mut/Mb) patients with metastatic NSCLC from the MSK-CHORD cohort. APOBEC status was classified using MESiCA, a machine-learning algorithm for mutational signature detection from targeted gene panels. Predictive value was established through treatment interaction testing, multivariate adjustment, propensity score weighting, and landmark analyses. External validation included 82 TMB-low patients from published whole-exome sequencing studies. RESULTS:APOBEC-positive patients (n = 52, 6.1%) had significantly improved overall survival [median 33.7 versus 17.4 months; hazard ratio (HR) 0.60, 95% confidence interval (CI) 0.42-0.85, P = 0.004]. No such benefit was observed in ICI-untreated patients (HR 0.98, P = 0.85), with significant treatment interaction (P = 0.032), confirming a predictive rather than prognostic effect. APOBEC remained independently predictive after adjustment for PD-L1 status and age (adjusted HR 0.57, P = 0.002). The benefit was particularly pronounced in PD-L1-negative ICI-treated patients (HR 0.51, P = 0.002). The effect was robust across propensity score, landmark, and bootstrap analyses. External validation confirmed independent APOBEC benefit (adjusted HR 0.23, P = 0.020). CONCLUSIONS:APOBEC mutational signatures represent a robust predictive biomarker for ICI response in TMB-low metastatic NSCLC, identifying responders particularly among PD-L1-negative patients where biomarker guidance is most needed, detectable from targeted gene panels used in routine clinical practice.

INTRODUCTION:HER2 mutations define a distinct, therapeutically actionable subset of NSCLC. Trastuzumab deruxtecan (T-DXd) has demonstrated strong activity in clinical trials, but real-world data are limited. METHODS:We conducted a retrospective, multinational cohort study of patients with advanced HER2-mutant NSCLC treated with T-DXd outside clinical trials between August 2021 and January 2025 across 68 centers in Europe and Israel. The primary end point was objective response rate (ORR). Secondary end points included progression-free survival (PFS), overall survival (OS), intracranial efficacy, and safety. RESULTS:Among 168 patients (median age 62 y; 59% female; 56% never-smokers), the ORR was 54.8% (95% confidence interval [CI]: 46.9-62.4) and disease control rate was 88.7% (95% CI: 82.9-93.1). Median PFS was 7.2 months (95% CI: 6.2-9.7) and median OS was 18.3 months (95% CI: 13.3-24.8). Treatment-naive patients (n = 18) achieved an ORR of 72.2% (95% CI: 46.5-90.3) and a median OS of 22.1 months (95% CI: 10.0-not evaluable). Patients with measurable brain metastases (n = 27) had an intracranial ORR of 74.1% (95% CI: 53.7-88.9), including 25.9% complete responses. Grade 3 or higher treatment-related adverse events occurred in 32% of patients. Interstitial lung disease/pneumonitis occurred in 14% (four fatal cases) of patients, without consistent predictors. CONCLUSIONS:In the largest real-world cohort reported to date, T-DXd demonstrated robust systemic and intracranial activity in HER2-mutant NSCLC, including treatment-naive patients and those with active brain metastases who were largely excluded from prior studies. Toxicity was consistent with previous reports, with interstitial lung disease remaining the main safety concern. These findings support integration of HER2-targeted therapies into evolving treatment algorithms.

Non-small cell lung cancer (NSCLC) patient management relies on molecular analysis to determine eligibility for targeted therapy. Furthermore, neoadjuvant immunotherapy is primarily suitable in the absence of specific genomic alterations. However, significant challenges remain, including suboptimal molecular testing and patients being assigned to non-optimal treatment strategies. Here, we present AI classifiers for the identification of EGFR, ALK, BRAF and MET alterations directly from hematoxylin and eosin (H&E)-stained tissue using CanvOI 1.1, a digital pathology foundation model. Their performance was evaluated on an independent validation dataset of 968 NSCLC samples. The classifiers achieved AUCs of 0.87 for EGFR, 0.96 for ALK, 0.88 for BRAF and 0.83 for MET. Moreover, they demonstrated high accuracy in identifying cases lacking alterations. Our results highlight the potential of deep-learning tools for the detection of NSCLC biomarkers and specifically the identification of tumors without EGFR or ALK driver alterations, supporting more informed clinical decision-making.

Despite a decade of immunotherapy, treatment selection in non-small cell lung cancer (NSCLC) still relies on subgroup analyses and clinical scores. I3LUNG ( NCT05537922 ) is currently the largest international, real-world, multimodal, artificial intelligence (AI)-based trial, enrolling 2365 patients. We integrated real-world clinical data (RWD), computed tomography (CT) images, digital pathology (DP), and genomics (G) into machine learning early-fusion (MLEF) and deep-learning intermediate-fusion (DLIF) models. MLEF achieved consistent performance across outcomes (AUC≈0.74), with improved results in first-line patients (AUC up to 0.82). Multimodal models outperformed RWD in clinical-specific subgroups (AUCs up to 0.86). In the test set, AI models surpassed PD-L1, ECOG PS, NLR, LDH (all with p <0.01) and the LIPI score. The clinical usability study showed that expert and non-expert physicians could improve their prediction with the explainable AI (XAI) tool. The I3LUNG tool emerges as a clinically relevant decision-support system and is currently under prospective validation in >2,000 patients.

Background: Minimal residual disease (MRD) assessment is an emerging tool for refining the risk of relapse following definitive therapy in non-small-cell lung cancer (NSCLC). However, data regarding its clinical impact on the decision-making process remain limited. We evaluated MRD feasibility and its impact in the real-world setting. Methods: A pooled retrospective analysis of longitudinal MRD data in NSCLC patients (n = 34: Signatera™ (Exome), n = 25, Guardant Reveal™, n = 9) was implemented. Co-primary endpoints: MRD feasibility and clinical impact on management (changes in surveillance intensity or therapy escalation/de-escalation). Secondary endpoints: sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy for recurrence detection, and MRD lead time. Results: MRD was feasible in 32/34 patients (94.1%); longitudinal testing included two samples in 15 patients (44.1%) and three samples in 2 patients (5.9%). Signatera™ (Exome) failed in 2/25 (8.0%) due to insufficient tissue. MRD influenced management in 20/34 (58.8%) patients, most commonly supporting therapy de-escalation (15/34, 44.1%), followed by imaging surveillance modification (3/34, 8.8%) and therapy escalation (2/34, 5.9%). In univariable analysis, tumor grade and STAS were associated with MRD-driven management impact, but neither remained significant in multivariable analysis. With a median follow-up of 18.9 months (IQR 8.5-30.7), MRD was positive in 6/32 (18.8%), while recurrence/progression occurred in 10/32 (31.3%) patients. MRD yielded 21 true negatives, five true positives, five false negatives (including two isolated brain recurrences), and one false positive, corresponding to a sensitivity of 50.0%, specificity of 95.5%, PPV of 83.3%, NPV of 80.8%, and an accuracy of 81.3%. The median MRD lead time (n = 5) was 1.31 months (range, 0.46-5.52). Conclusions: In this real-world cohort, MRD testing was feasible and frequently guided clinical decisions, mainly supporting treatment de-escalation. MRD was highly specific but less sensitive. Prospective studies are needed to define optimal testing intervals and validate MRD-guided strategies.

Background:Amivantamab is an approved dual epidermal growth factor receptor (EGFR) and mesenchymal-epithelial transition (MET) inhibitor for the treatment of EGFR exon 20 insertion (EGFRex20ins) mutations. Recent data support the use of amivantamab for both common and uncommon EGFR mutations after previous therapies. In this study, we investigated the role of adding amivantamab to the ongoing EGFR-tyrosine kinase inhibitor (TKI) in later lines of therapy upon progression. Methods:Patients treated at Shaare Zedek Medical Center (SZMC) from October 2021 to May 2024 who received amivantamab plus a previous EGFR-TKI. Cohort A included nine patients with common EGFR mutations [four exon 19 deletions (ex19dels), one G719C, four L858R]. Cohort B included six patients with exon 20 insertions. Safety and preliminary efficacy were evaluated according to Response Evaluation Criteria in Solid Tumors (RECIST) 1.1. Results:In cohort A, the objective response rate (ORR) was 22% (L858R 50%, exon 19 0%), disease control rate (DCR) 44% (L858R 100%, exon 19 0%), median duration of treatment (mDoT) 3 months (L858R 7.5 months, exon 19 2.3 months), and median overall survival (mOS) 6.7 months (L858R 14.4 months, exon 19 4.6 months). In cohort B, ORR was 17%, DCR 83%, mDoT 5.5 months, and mOS 16.2 months. Grade ≥3 toxicities included nausea, diarrhea, rash, infusion reactions, and thromboembolism. Conclusions:This pilot study suggests that adding late-line amivantamab to an ongoing EGFR-TKI may have potential benefits in selected non-small cell lung cancer (NSCLC) patients with EGFR mutations, but resulted in high skin toxicity. Patients with EGFR L858R mutations appeared to show improved responses to amivantamab compared to the lack of response with ex19dels, while acquired resistance was associated with loss of the original EGFR driver mutation and MET alterations. However, these preliminary findings lack robust evidence due to study limitations, and larger, prospective, multi-center trials are needed to validate these results.

Machine learning models require large, diverse datasets which can be challenging to acquire, even more so for multimodal and paired histology data. Within the I3LUNG European Funded project (NCT05537922), we evaluated multimodal synthetic data generation as a solution to enable domain-specific pretraining and imputation in NSCLC patients treated with immunotherapy (IO) using multimodal data. Our two-stage method included multimodal data simulation and AI-enabled data evaluation. First, a cross-modal autoencoder jointly embedded histology foundation model features with key clinical features: PD-L1 expression, smoking status, baseline ECOG performance status, histologic subtype, gender, metastatic sites, progression and survival events, LDH, BMI, neutrophil-lymphocyte ratio (NLR), and progression free survival (PFS). The joint latent spaced was sampled using a Gaussian Copula model to generate synthetic patients with coherent multimodal features. Second, to evaluate the fidelity of synthetic clinical representations, we trained a deep neural network models using Cox proportional hazards endpoints on real and simulated data to predict PFS, validating on held-out real patient data. Additionally, we used HistoXGAN to generate paired histology tile images for each synthetic patient. We analyzed NSCLC patients (N=1813) treated with immunotherapy from five centers, split into training (n=1630) and test (n=183) cohorts. The synthetic data matched the original distributions, with minimal differences in continuous features (t-test p > 0.05 and mean differences: BMI -1.72%, PFS -0.37%, LDH -9.18%) and categorical ones (chi-square p > 0.05 and maximum class proportion differences of 3.9%, 15.5%, and 1.2% for bone metastasis, PD-L1 expression, and smoking history respectively). Models trained on synthetic data (N=1000) performed similarly to real data. In validation, the Cox model trained on synthetic data achieved a c-index of 0.683, versus 0.679 for real data (0.6% relative difference). Both synthetic and real data identified consistent prognostic factors (HR [95% CI]): bone metastases (real: 2.36 [1.39-4.80], synthetic: 2.46 [1.39-3.77]), LDH (real: 1.60 [1.16-2.69], synthetic: 1.48 [1.22-2.39]), and liver metastases (real: 1.52 [1.24-3.69], synthetic: 1.46 [1.11-2.80]). Our multimodal synthetic data successfully captured complex multi-feature relationships predictive of PFS in NSCLC patients treated with IO. Synthetic data enables cross-institutional model development while increasing patient privacy, with minimal impact on model performance. This approach paves the way for data democratization, fostering rapid collaboration and mutual validation of AI algorithms. Hanna M. Hieromnimon, Vanja Miskovic, Matteo Sacco, Alberto Ferrarin, Laura Mazzeo, Andrea Spagnoletti, Monica Ganzinelli, Cecilia Silvestri, Leonardo Provenzano, Claudia Proto, Nir Peled, Enriqueta Felip, Helena Linardou, Martin Reck, Francesco Trovo, Giuseppe Lo Russo, Marina Chiara. Garassino, Samantha J. Riesenfeld, Alexander T. Pearson, Arsela Prelaj. Multimodal generative AI jointly learns pathology and clinical data to synthesize a multinational lung cancer cohort [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Artificial Intelligence and Machine Learning; 2025 Jul 10-12; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(13_Suppl):Abstract nr A058.

Patients (pts) with non-small-cell lung cancer (NSCLC) who do not achieve pathologic complete response (pCR) after neoadjuvant chemoimmunotherapy and surgery have a worse prognosis than those who achieve pCR, and escalation of therapy for improved outcomes may be warranted. Pembrolizumab (pembro) is approved in several countries as monotherapy for the adjuvant treatment of pts with NSCLC following neoadjuvant pembro plus platinum-based chemotherapy (chemo) and resection. V940 (mRNA-4157) is a novel, mRNA-based, individualized neoantigen therapy that encodes up to 34 neoantigens specific to each pt’s unique tumor mutanome and is encapsulated in a lipid nanoparticle that is administered intramuscularly (IM). Preliminary antitumor activity for V940 as monotherapy and in combination with pembro was shown in the phase 1 KEYNOTE-603 study in solid tumors, including NSCLC, and in KEYNOTE-942 in melanoma. The INTerpath-009 study (NCT06623422) evaluates adjuvant pembro with and without V940 in pts with resected stage II-IIIB (N2) NSCLC that did not achieve pCR after neoadjuvant pembro plus chemo. This phase 3, multicenter, double-blind study is enrolling pts ≥18 years old with histologically-confirmed stage II-IIIB (N2) squamous (sq) or nonsquamous (nonsq) NSCLC (AJCC v8) with no tumor-activating EGFR and no known ALK alterations. Eligible pts are able to undergo protocol therapy, including surgery, and have ECOG PS 0 or 1. Pts eligible for randomization must have resected (R0 or R1) tumors that did not achieve pCR after ≤4 cycles of neoadjuvant pembro plus chemo, and surgical tumor sample available for biomarker analysis and next-generation sequencing. Pts who received neoadjuvant pembro plus chemo prior to enrollment are eligible provided other eligibility criteria are met. Approximately 680 pts will be randomized 1:1 to V940 1 mg IM or placebo Q3W for 9 doses, both in combination with pembro (400 mg IV Q6W for 7 cycles). Treatment will continue until PD, pt withdrawal, unacceptable toxicity, or an additional malignancy requiring treatment. Randomization will be stratified by histology (sq vs nonsq), PD-L1 expression (TPS <1% vs ≥1%), disease stage (II vs III), and geographic location (North America/Western Europe/Australia vs rest of world). Tumor imaging will occur at baseline and every 12 wk until wk 48, every 24 wk through year 3, and every 48 wk thereafter from randomization until distant recurrence, death, withdrawal of consent, or end of study. The primary endpoint is disease-free survival (DFS) per investigator assessment. Secondary endpoints include OS, distant metastasis-free survival, DFS after next-line therapy, lung cancer-specific survival, patient-reported outcomes, and safety. The first pt was screened on October 21, 2024, and recruitment is ongoing. Tina Cascone, Nir Peled, Alona Zer, David Chism, Danko Martincic, Laureen S. Ojalvo, Joydeep Banerjee, Zheng Wang, Steven M. Keller, Jonathan D. Spicer. Phase 3 INTerpath-009 study: Individualized neoantigen therapy V940 (mRNA-4157) plus pembrolizumab for resected stage II-IIIB (N2) NSCLC with incomplete pathological response to neoadjuvant immunochemotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_2):Abstract nr CT251.

Background/Objectives: The Lung Immune Prognostic Index (LIPI) has emerged as a promising biomarker for predicting outcomes in advanced non-small cell lung cancer (aNSCLC). We assessed whether LIPI, in combination with baseline clinical characteristics, can guide first-line treatment selection between pembrolizumab (P) and pembrolizumab plus platinum-based chemotherapy (PCT) in patients with PD-L1 tumor proportion score (TPS) ≥ 50% and EGFR/ALK/ROS1 wild-type. Methods: A predictive score was developed using baseline clinical variables, including age, sex, smoking status, and LIPI, in a proof-of-concept cohort (n = 241). This model was then validated in an independent cohort of 409 patients. OS was compared between patients treated with P versus PCT, stratified by predictive score. Results: In the proof-of-concept cohort, the median OS was 18.3 months for P and 26.6 months for PCT (p = 0.001). In the validation cohort, the median OS was 28.0 months for P and 22.2 months for PCT (p = 0.062). Stratification using the predictive score showed that patients with high scores (3–5) had improved OS with PCT compared to P (31.2 vs. 25.5 months, p = 0.001), while those with low scores (0–2) derived similar benefits from both treatments. Conclusions: This LIPI-based predictive score may assist in identifying aNSCLC patients who derive greater benefit from chemo-immunotherapy over immunotherapy. Its simplicity and clinical relevance support integration into treatment decision-making, pending prospective validation.

INTRODUCTION:Despite recent advances in immunotherapy combinations for extensive-stage small cell lung cancer (ES-SCLC), rapid disease progression following chemotherapy discontinuation remains a significant challenge. While the addition of pembrolizumab to platinum-etoposide has demonstrated a modest improvement in progression-free survival (PFS), there is an urgent need for more effective maintenance strategies. Sacituzumab govitecan (SG), an antibody-drug conjugate targeting Trop-2, has shown promising activity in pretreated ES-SCLC. This phase II study evaluates the efficacy and safety of adding SG to pembrolizumab maintenance therapy following chemoimmunotherapy induction in treatment-naïve ES-SCLC patients. METHODS:In the PESGA trial, a prospective, open-label, single-arm phase II trial, patients with previously untreated ES-SCLC will receive induction therapy consisting of pembrolizumab (200 mg Q3 W) plus carboplatin (AUC 5) and etoposide (100 mg/m² Days 1-3) for 4 cycles. This will be followed by maintenance therapy combining pembrolizumab (200 mg Q3 W) with SG (10 mg/kg on Days 1 and 8 of 21-day cycles) for up to 31 cycles. The primary endpoint is PFS from the start of induction treatment. Secondary endpoints include overall survival, duration of response, and safety. Exploratory analyses will investigate molecular resistance mechanisms through sequential liquid and tissue biopsies and evaluate correlations between tumor Trop-2 expression and clinical outcomes. The study plans to enroll 21 patients over 18 months, with an estimated total study duration of 54 months. Results will be analyzed after 50% of patients have achieved PFS. CONCLUSIONS:The PESGA study design builds upon the KEYNOTE-604 regimen by incorporating SG into the maintenance phase, potentially addressing the challenge of early progression in ES-SCLC. The study may provide valuable insights into novel maintenance strategies and molecular mechanisms of treatment resistance in ES-SCLC.

BACKGROUND:Small cell lung cancer (SCLC) remains a therapeutic challenge with limited treatment options and poor survival outcomes. Lurbinectedin has been approved for the post-platinum setting, but there is a lack of real-world evidence and information related to its optimal sequencing and effectiveness across different lines of therapy. This study aimed to analyze line-dependent outcomes and response patterns of lurbinectedin in SCLC patients. METHODS:In this retrospective multi-center study analysis, we enrolled SCLC patients treated with lurbinectedin between January 2020 and December 2024. Patient demographics, treatment histories, and outcomes were collected from electronic medical records. Response assessment was performed using RECIST v1.1 criteria. Overall survival, progression-free survival, and treatment duration were analyzed using Kaplan-Meier methodology, with specific attention to outcomes across different lines of therapy. RESULTS:A total of 64 patients were enrolled, with 59 evaluable for analysis (36 males [56.3 %], median age 65 years [range 37-81]). Lurbinectedin was given as second-line therapy in 39 patients (66.1 %) and as third-line or beyond in 20 patients (33.9 %). The overall response rate (ORR) was 37.3 % (n = 22), including 1 complete response (1.7 %) and 21 partial responses (35.6 %). By line of therapy, the ORR was 38.5 % (n = 15) in second-line and 35.0 % (n = 7) in third-line or later. The median overall survival (mOS) for all evaluable patients was 7.7 months, and the median duration of treatment (mDoT) was 4.3 months. At data cutoff, 13 patients (22 %) were alive and censored in the OS analysis, with a median follow-up of 8.5 months among survivors. Prior immune checkpoint inhibitor (IO) exposure was documented in 71.9 % of patients (n = 46); although those with prior IO had numerically longer mOS (8.8 vs. 6.0 months, p = 0.06) and mDoT (4.6 vs. 2.3 months, p = 0.14), neither difference reached statistical significance. This survival analysis included 43 patients in the prior-IO group (3 patients with prior IO were excluded due to incomplete follow-up) and 16 patients in the no-IO group. The presence of baseline brain metastases (47 %) did not preclude clinical benefit. Treatment was generally well tolerated: the most common hematologic toxicities observed were anemia (44.1 %), thrombocytopenia (37.3 %), and neutropenia (15.3 %), predominantly grade 1-2, and no grade 4 adverse events were observed. CONCLUSIONS:Lurbinectedin demonstrated meaningful clinical activity in second-line therapy for SCLC, while also showing durable responses in heavily pretreated cases. These findings support the consideration of lurbinectedin also beyond second-line therapy in selected SCLC patients, though larger prospective studies are needed to validate these response patterns.

Background:Peripheral blood mononuclear cell (PBMC) humanized mouse models are essential for researching non-small cell lung cancer (NSCLC) treatments. However, these models are prone to xeno-graft versus host disease (xeno-GVHD), hampering their utility and requiring further investigation. This study examined xeno-GVHD responses from PBMCs of advanced-stage NSCLC patients compared to healthy donors (HDs) in a humanized peripheral blood lymphocyte (hu-PBL) model. Methods:PBMCs from NSCLC patients and HDs were injected into immunocompromised NSG-SGM3 mice and monitored for eight weeks. xeno-GVHD progression was assessed through clinical examinations and flow cytometry of human T-cell levels in various tissues. Results:Mice injected with PBMCs from HDs showed xeno-GVHD signs as early as 28 days post-injection, whereas those from NSCLC patients exhibited minimal signs, with only one model showing delayed responses by day 42. Clinical symptoms in mice included weight loss, anemia, low platelet counts, fur changes, and behavioral modifications. Flow cytometry of human PBMCs in mice indicated dominant CD8+ effector memory T cells in peripheral blood. In contrast, CD4+ effector memory T cells were predominant in the organs, with overall T cell levels lower in NSCLC models. Conclusions:This study demonstrates significant differences in xeno-GVHD progression between advance-stage NSCLC patients and HDs, likely influenced by the patient's treatment histories. These findings improve our understanding of hu-PBL models for NSCLC research and may inform future treatment studies and strategies.