Decoding B-cell Signatures of Complete Pathologic Response to Perioperative Chemoimmunotherapy in Non-Small Cell Lung Cancer.

Decoding B-cell Signatures of Complete Pathologic Response to Perioperative Chemoimmunotherapy in Non–Small Cell Lung Cancer - PMC Skip to main content An official website of the United States government Here's how you know Here's how you know Official websites use .gov A .gov website belongs to an official government organization in the United States. Secure .gov websites use HTTPS A lock ( Lock Locked padlock icon ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites. Search Log in Dashboard Publications Account settings Log out Search… Search NCBI Primary site navigation Search Logged in as: Dashboard Publications Account settings Log in Search PMC Full-Text Archive Search in PMC Journal List User Guide PERMALINK Copy As a library, NLM provides access to scientific literature. Inclusion in an NLM database does not imply endorsement of, or agreement with, the contents by NLM or the National Institutes of Health. Learn more: PMC Disclaimer | PMC Copyright Notice Clin Cancer Res . 2026 Mar 11;32(8):1499–1512. doi: 10.1158/1078-0432.CCR-25-3315 Search in PMC Search in PubMed View in NLM Catalog Add to search Decoding B-cell Signatures of Complete Pathologic Response to Perioperative Chemoimmunotherapy in Non–Small Cell Lung Cancer Belén Sierra-Rodero Belén Sierra-Rodero 1 Medical Oncology, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana (IDIPHISA), Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. Find articles by Belén Sierra-Rodero 1, # , Ángeles Gil-González Ángeles Gil-González 1 Medical Oncology, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana (IDIPHISA), Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. Find articles by Ángeles Gil-González 1, # , Marta Molina-Alejandre Marta Molina-Alejandre 1 Medical Oncology, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana (IDIPHISA), Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. Find articles by Marta Molina-Alejandre 1 , Ernest Nadal Ernest Nadal 2 Institut Català d’Oncologia (ICO), IDIBELL, L’Hospitalet de Llobregat, Barcelona, Spain. Find articles by Ernest Nadal 2 , Virginia Calvo Virginia Calvo 1 Medical Oncology, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana (IDIPHISA), Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. Find articles by Virginia Calvo 1 , Martín Lázaro Martín Lázaro 3 Medical Oncology, Hospital Alvaro Cunqueiro, Complexo Hospitalario Universitario de Vigo, Vigo, Spain. Find articles by Martín Lázaro 3 , Amelia Insa Amelia Insa 4 Fundación INCLIVA, Medical Oncology, Hospital Clínico Universitario de Valencia, Valencia, Spain. Find articles by Amelia Insa 4 , Bartomeu Massuti Bartomeu Massuti 5 Medical Oncology, Hospital General Dr. Balmis de Alicante, ISABIAL, Alicante, Spain. Find articles by Bartomeu Massuti 5 , Alex Martinez Marti Alex Martinez Marti 6 Medical Oncology, Vall Hebron Institute of Oncology (VHIO), Vall d’Hebron Hospital Universitari, Barcelona, Spain. Find articles by Alex Martinez Marti 6 , Javier de Castro Javier de Castro 7 Medical Oncology, Hospital Universitario La Paz, Madrid, Spain. Find articles by Javier de Castro 7 , Rosario García Campelo Rosario García Campelo 8 Medical Oncology, Hospital Universitario A Coruña, A Coruña, Spain. Find articles by Rosario García Campelo 8 , Jose Luis González Larriba Jose Luis González Larriba 9 Medical Oncology, Hospital Clínico San Carlos, Madrid, Spain. Find articles by Jose Luis González Larriba 9 , Reyes Bernabé Reyes Bernabé 10 Medical Oncology, Hospital Universitario Virgen del Rocio, Sevilla, Spain. Find articles by Reyes Bernabé 10 , Manuel Dómine Manuel Dómine 11 Medical Oncology, Hospital Universitario Fundación Jiménez Díaz, Madrid, Spain. Find articles by Manuel Dómine 11 , Santiago Ponce Aix Santiago Ponce Aix 12 Medical Oncology, Hospital Universitario 12 de Octubre, Madrid, Spain. Find articles by Santiago Ponce Aix 12 , Manuel Cobo Manuel Cobo 13 Medical Oncology Intercenter Unit, Regional and Virgen de la Victoria University Hospitals, Instituto de Investigación Biomédica de Málaga, Málaga, Spain. Find articles by Manuel Cobo 13 , Carlos Camps Carlos Camps 14 Medical Oncology, Hospital General Universitario de Valencia, Universidad de Valencia and Centro de Investigación Biomédica en Red Cáncer, Valencia, Spain. Find articles by Carlos Camps 14 , Noemi Reguart Noemi Reguart 15 Medical Oncology, Hospital Clinic and Translational Genomics and Targeted Therapies in Solid Tumors, Institut de Investigacions Biomèdiques, Barcelona, Spain. Find articles by Noemi Reguart 15 , Joaquím Bosch-Barrera Joaquím Bosch-Barrera 16 Institut Català d’Oncologia, Hospital Universitari Dr. Josep Trueta and Precision Oncology Group (OncoGIR-Pro), Intitut d’Investigacions Biomèdiques de Girona (IDIBGI-CERCA), Girona, Spain. Find articles by Joaquím Bosch-Barrera 16 , Margarita Majem Margarita Majem 17 Medical Oncology, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain. Find articles by Margarita Majem 17 , Andres Aguilar Andres Aguilar 18 Medical Oncology, Instituto Oncológico Dr. Rosell, Dexeus University Hospital, Barcelona, Spain. Find articles by Andres Aguilar 18 , Ramón Palmero Ramón Palmero 2 Institut Català d’Oncologia (ICO), IDIBELL, L’Hospitalet de Llobregat, Barcelona, Spain. Find articles by Ramón Palmero 2 , Mariola Blanco Clemente Mariola Blanco Clemente 1 Medical Oncology, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana (IDIPHISA), Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. Find articles by Mariola Blanco Clemente 1 , Javier Martín-López Javier Martín-López 19 Pathology Department, Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. Find articles by Javier Martín-López 19 , Rafael Muñoz-Viana Rafael Muñoz-Viana 20 Bioinformatics Unit, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana, Madrid, Spain. Find articles by Rafael Muñoz-Viana 20 , Diego Megías Diego Megías 21 Advanced Optical Microscopy Unit, Instituto de Salud Carlos III, Majadahonda, Spain. Find articles by Diego Megías 21 , Juan Manuel Gutiérrez-Escobedo Juan Manuel Gutiérrez-Escobedo 1 Medical Oncology, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana (IDIPHISA), Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. Find articles by Juan Manuel Gutiérrez-Escobedo 1 , Cristina Martínez-Toledo Cristina Martínez-Toledo 1 Medical Oncology, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana (IDIPHISA), Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. Find articles by Cristina Martínez-Toledo 1 , Alberto Cruz-Bermúdez Alberto Cruz-Bermúdez 1 Medical Oncology, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana (IDIPHISA), Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. Find articles by Alberto Cruz-Bermúdez 1, *, ‡ , Mariano Provencio Mariano Provencio 1 Medical Oncology, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana (IDIPHISA), Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. Find articles by Mariano Provencio 1, *, ‡ Author information Article notes Copyright and License information 1 Medical Oncology, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana (IDIPHISA), Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. 2 Institut Català d’Oncologia (ICO), IDIBELL, L’Hospitalet de Llobregat, Barcelona, Spain. 3 Medical Oncology, Hospital Alvaro Cunqueiro, Complexo Hospitalario Universitario de Vigo, Vigo, Spain. 4 Fundación INCLIVA, Medical Oncology, Hospital Clínico Universitario de Valencia, Valencia, Spain. 5 Medical Oncology, Hospital General Dr. Balmis de Alicante, ISABIAL, Alicante, Spain. 6 Medical Oncology, Vall Hebron Institute of Oncology (VHIO), Vall d’Hebron Hospital Universitari, Barcelona, Spain. 7 Medical Oncology, Hospital Universitario La Paz, Madrid, Spain. 8 Medical Oncology, Hospital Universitario A Coruña, A Coruña, Spain. 9 Medical Oncology, Hospital Clínico San Carlos, Madrid, Spain. 10 Medical Oncology, Hospital Universitario Virgen del Rocio, Sevilla, Spain. 11 Medical Oncology, Hospital Universitario Fundación Jiménez Díaz, Madrid, Spain. 12 Medical Oncology, Hospital Universitario 12 de Octubre, Madrid, Spain. 13 Medical Oncology Intercenter Unit, Regional and Virgen de la Victoria University Hospitals, Instituto de Investigación Biomédica de Málaga, Málaga, Spain. 14 Medical Oncology, Hospital General Universitario de Valencia, Universidad de Valencia and Centro de Investigación Biomédica en Red Cáncer, Valencia, Spain. 15 Medical Oncology, Hospital Clinic and Translational Genomics and Targeted Therapies in Solid Tumors, Institut de Investigacions Biomèdiques, Barcelona, Spain. 16 Institut Català d’Oncologia, Hospital Universitari Dr. Josep Trueta and Precision Oncology Group (OncoGIR-Pro), Intitut d’Investigacions Biomèdiques de Girona (IDIBGI-CERCA), Girona, Spain. 17 Medical Oncology, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain. 18 Medical Oncology, Instituto Oncológico Dr. Rosell, Dexeus University Hospital, Barcelona, Spain. 19 Pathology Department, Hospital Universitario Puerta de Hierro-Majadahonda, Madrid, Spain. 20 Bioinformatics Unit, Instituto de Investigación Sanitaria Puerta de Hierro-Segovia de Arana, Madrid, Spain. 21 Advanced Optical Microscopy Unit, Instituto de Salud Carlos III, Majadahonda, Spain. * Corresponding Authors: Alberto Cruz-Bermúdez, Medical Oncology, Hospital Universitario Puerta de Hierro Majadahonda, Majadahonda 28222, Spain. E-mail: [email protected] ; and Mariano Provencio, Medical Oncology, Hospital Universitario Puerta de Hierro Majadahonda, Majadahonda 28222, Spain. E-mail: [email protected] Clin Cancer Res 2026;32:1499–512 # B. Sierra-Rodero and Á. Gil-González contributed equally as co-first authors of this article. ‡ A. Cruz-Bermúdez and M. Provencio contributed equally as the co-senior authors of this article. Received 2025 Sep 1; Revised 2025 Nov 12; Accepted 2026 Feb 6; Issue date 2026 Apr 15. ©2026 The Authors; Published by the American Association for Cancer Research This open access article is distributed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International (CC BY-NC-ND 4.0) license. PMC Copyright notice PMCID: PMC13080320  PMID: 41805895 Abstract Purpose: Complete pathologic response (CPR) correlates with long-term survival after perioperative chemoimmunotherapy (ChIO) in resectable non–small cell lung cancer (NSCLC). We provide a multiomic characterization of B cells and tertiary lymphoid structures (TLS) to dissect the immune landscape associated with CPR. Experimental Design: We integrated B-cell receptor (BCR) repertoire profiling ( n = 87 tissue, n = 25 blood), multiplex immunofluorescence ( n = 67), and bulk ( n = 15), spatial ( n = 12), and single-cell transcriptomics ( n = 15) from tumor tissue and blood (baseline, surgery, and at 6 months of adjuvant therapy) in 123 patients (NADIM/NADIM II trials, NCT03081689 / NCT03838159 ). Results: CPR tumors exhibited a more clonal baseline BCR repertoire (AUC 0.775; P = 0.030) that was better conserved and reinvigorated during neoadjuvant ChIO. In blood, patients with CPR tumors displayed a repertoire enriched in class-switched clones (AUC 0.833; P = 0.008), characterized by higher diversity, lower clonality, and upregulation of activation-related transcriptional programs. Neoadjuvant ChIO was associated with the induction of B-cell–related genes within TLS regions and with higher TLS density at surgery compared with Ch ( P = 0.034). TLS density was not associated with CPR ( P = 0.129); however, mature TLS in CPR tumors were enriched in immune activation and antigen-presenting pathways, estimated T follicular helper cells, plasmacytoid dendritic cells, and plasma cells, whereas low B-cell regions from CPR tumors displayed higher inferred infiltration of CD8 + T cells, NK cells, and macrophages, with reduced neutrophils and Tregs. Conclusions: Patients with CPR tumors exhibit a preexisting and more mature B-cell response that develops further during neoadjuvant ChIO. Our findings link B-cell–related features to CPR and highlight BCR metrics as promising predictive biomarkers in NSCLC. Translational Relevance. Perioperative chemoimmunotherapy (ChIO) has significantly improved complete pathologic response (CPR) rates in resectable non–small cell lung cancer (NSCLC). However, the biological determinants driving differential responses remain poorly defined, limiting the development of reliable predictive biomarkers. In this study, we provide a comprehensive multiomic characterization of B cells and tertiary lymphoid structures (TLS) in the context of perioperative ChIO in NSCLC. Patients with CPR tumors exhibit at baseline a preexisting and more mature B-cell immune response that is also better developed during neoadjuvant ChIO, as evidenced by mature TLS (mTLS) and low B-cell regions displaying enhanced immune activation pathways at surgery, along with more mature and active circulating B-cell populations. Collectively, these findings link B-cell– and TLS-mediated immunity to therapeutic response and support BCR metrics as promising biomarkers for patient stratification in perioperative immunotherapy. Introduction Lung cancer is the leading cause of cancer-related death worldwide and the second most commonly diagnosed malignancy (1), with non–small cell lung cancer (NSCLC) accounting for approximately 85% of cases (2). Recently, the introduction of neoadjuvant and perioperative chemoimmunotherapy (ChIO) in locally advanced NSCLC has led to improved survival outcomes ( 3 – 7 ), establishing a new standard of treatment (FDA and European Medicines Agency). In this context, achieving a complete pathologic response (CPR; defined as 0% viable tumor cells) is associated with long-term survival and may identify patients who are potentially cured ( 4 , 8 ). However, a substantial proportion of patients exhibit a non-CPR (NCPR), which is associated with a higher risk of recurrence ( 4 , 8 , 9 ). Several biomarkers, such as programmed cell death ligand 1 (PD-L1) expression and tumor mutational burden (TMB), have been proposed as predictors of response. Nonetheless, the mechanisms underlying differential treatment outcomes remain poorly understood, contributing to the current lack of effective predictive biomarkers—an unmet clinical need that is critical for guiding patient selection and therapeutic decision-making ( 10 ). A comprehensive understanding of how antitumor immune responses develop both within the tumor microenvironment (TME) and at the systemic level is essential to elucidate the biological differences between CPR and NCPR. Among the various components of the immune response, T cells have been extensively studied, with features such as their phenotypic profiles and T-cell receptor (TCR) repertoire proposed as potential biomarkers of treatment response and survival ( 11 – 13 ). Recently, B lymphocytes have emerged as key players in the orchestration of antitumor immunity, particularly through the formation of intratumoral tertiary lymphoid structures (TLS) and their potential involvement in peripheral immunosurveillance ( 14 ). TLS are ectopic lymphoid aggregates that develop in chronically inflamed, antigen-rich non-lymphoid tissues and have been associated with improved clinical outcomes to immune checkpoint blockade, including in NSCLC ( 15 – 18 ). However, the induction of TLS and the specific role of B cells in mediating responses to neoadjuvant ChIO in NSCLC remain largely unexplored. Here, we present the most comprehensive translational study to date investigating the role of B cells in pathologic responses among patients with potentially resectable NSCLC treated with perioperative ChIO. This work leverages the unique opportunity to analyze paired tumor and blood samples collected before and during treatment from the NADIM ( NCT03081689 ) and NADIM II ( NCT03838159 ) trials. Using a multiomic approach—including B-cell receptor (BCR) repertoire profiling, spatial transcriptomics, multiplex immunofluorescence (mIF), and single-cell RNA sequencing (scRNA-seq)—we provide an integrated view of the B-cell–mediated immune landscape and its association with treatment response. Materials and Methods Study cohort and design This study included 123 patients from the NADIM ( n = 44; refs. 3 , 4 ) and NADIM II ( n = 79; ref. 5 ) clinical trials ( NCT03081689 , NCT03838159 ) promoted by the Spanish Lung Cancer Group (SLCG) and approved by the Puerta de Hierro Hospital Research Ethics Committee for Medicinal Products (references 20.16 NADIM and 05.19 NADIM II). Studies were conducted in accordance with the Declaration of Helsinki, and informed written consent was obtained from each subject. Both were multicenter phase II trials enrolling patients with potentially resectable stage IIIA/B NSCLC treated with perioperative ChIO based on carboplatin–paclitaxel plus nivolumab. In NADIM, all patients received ChIO, whereas in NADIM II, patients were randomized (2:1) to ChIO versus chemotherapy alone. After neoadjuvant therapy and surgical resection, patients in the experimental arm received adjuvant nivolumab. Pathologic response was centrally reviewed and categorized as CPR (0% viable tumor) or NCPR (>0% viable tumor). Progression-free survival (PFS) was defined as the time from diagnosis to tumor progression or cancer-related death, and overall survival (OS) was defined as the time from diagnosis to death due to cancer. Causes of death not related to disease progression were considered as loss of follow-up at the time of death. Sample collection and nucleic acid extraction Formalin-fixed, paraffin-embedded (FFPE) tumor tissue was collected at baseline and at surgery. Peripheral blood mononuclear cells (PBMC) were collected at baseline, following neoadjuvant therapy, and 6 months after the initiation of adjuvant therapy. Cells were isolated using a density gradient with Lymphoprep (STEMCELL Technologies) and stored at ‒180°C. Nucleic acids were isolated using standard kits (truXTRAC FFPE total DNA kit, Covaris, for tissue, and Maxwell RSC simplyRNA Cells Kit, Promega, for PBMC) and quality-controlled before downstream applications. RNA and DNA concentrations were measured with the Qubit RNA HS Assay and Qubit 1× dsDNA High Sensitivity kits (Thermo Fisher Scientific) and stored at −80°C. PD-L1 and TMB assessment PD-L1 expression was quantified in pretreatment samples using the Dako 22C3 pharmDx assay and reported as tumor proportion score (TPS; refs. 3 – 5 ). TMB was determined from 20 ng of pretreatment tumor DNA with a minimum of 20% tumor cells using the Oncomine Tumor Mutation Load Assay (Thermo Fisher Scientific), as previously described ( 3 – 5 ). TMB calculation was performed using the Oncomine Tumor Mutation Load algorithm (version 3.2). Variants were detected and annotated for potential clinical significance using the Oncomine Variants 5.12 filter. Only mutations with an allele frequency ≥5%, coverage ≥60×, and P value ≤ 0.05 were included in the analysis. BCR and TCR repertoire sequencing and analysis BCR repertoires were profiled from both tumor tissue (Oncomine BCR IGH SR Assay, RNA kit, Thermo Fisher Scientific) and PBMC (BCR IGH LR Assay, RNA kit, Thermo Fisher Scientific) at baseline, after neoadjuvant therapy, and at 6 months of adjuvant treatment. Libraries were prepared from 100 ng (tissue) or 25 ng (PBMC) of RNA. Library concentration was measured with the Ion Library TaqMan Quantitation kit (Thermo Fisher Scientific) in a LightCycler 480 (Roche). The quality of RNA and BCR libraries was evaluated with the RNA 6000 Pico Kit and High Sensitivity DNA Kit (Agilent) on the Bioanalyzer 2100 (Agilent). Thirty-two tissue and eight PBMC libraries were mixed at 25 pmol/L in Ion 540 and 530 chips (Thermo Fisher Scientific), respectively, and sequenced on the Ion GeneStudio S5 System. TCR libraries were prepared from the same RNA samples using the Oncomine TCR Beta-SR (100-ng input, tissue) and Oncomine TCR Beta-LR (25-ng input, PBMC) kits and sequenced on the Ion GeneStudio S5 System, as previously described ( 12 ). Data were analyzed with Ion Reporter software using the 1.2 workflow 5.12 version for tissue and the 1.4 workflow 5.20 version for PBMC. Samples with productive and rescued productive read frequencies below 0.2 were discarded. BCR metrics included the number of clones, clonality (1 – evenness), and Shannon diversity. The top 10 clones were selected as the 10 clones with the highest frequencies for each sample, and the clonal space (CS) occupied by the top 10 clones (Top 10 CS) was calculated by adding the frequencies of these clones. Percentages of non-switched ( IGHD + IGHM ) and switched ( IGHG1–4 , IGHA1–2 , IGHE ) clones were calculated by adding the respective isotype proportions. The Vegan R package (RRID:SCR_011950) was used to calculate Shannon diversity and clonality for each isotype and grouped isotypes. Bulk and spatial transcriptomic analyses Bulk gene expression profiling of tumor tissue was performed and analyzed with the Oncomine Immune Response Research Assay (Thermo Fisher Scientific), as previously described ( 13 ). Spatial transcriptomic analysis was performed on 12 surgical tissue samples using the Visium Spatial for FFPE Gene Expression Kit for Human Transcriptome (1000336, 10x Genomics) following the manufacturer’s instructions. Pathologist-selected 6- × 6-mm regions from four CPR and eight NCPR patients ensured RNA quality criteria (DV200 > 50%). FFPE sections were processed on Visium slides, hematoxylin and eosin (H&E) stained, imaged (Leica THUNDER wide-field microscope), and libraries prepared following 10x Genomics protocols. Libraries were quality-controlled (Bioanalyzer High Sensitivity DNA Analysis, Agilent), quantified (KAPA Library Quantification Kit for Illumina platforms, Roche), and sequenced on a NovaSeq X Plus at ∼25,000 reads per tissue-covered spot. Before performing differential gene expression analysis, gene set enrichment analysis (GSEA; RRID:SCR_003199), and deconvolution, B-cell–rich regions and TLS were annotated by a pathologist using H&E staining features together with the gene expression of a set of well-established markers ( CD19 , MS4A1 , CD3 , CXCL13 , CR2 ; ref. 19 ) in Loupe Browser version 8.1.2 (RRID:SCR_018555, Supplementary Fig. S5A and S5B). Four spatial categories were defined: mTLS, non-mTLS, dispersed B cells, and low B-cell areas. TLS (both mTLS and non-mTLS) were identified as CD19 + or MS4A1 + and CXCL13 + regions colocalizing with adjacent CD3 + areas. TLS maturity was determined by the expression of CR2 , a germinal center marker. Dispersed B-cell regions were defined as CD19 + or MS4A1 + areas that did not overlap with TLS, representing diffuse B-cell infiltration outside structured aggregates. The remaining tissue regions, characterized by CD19 − and MS4A1 − , were categorized as low B-cell areas. Quality control was performed with the Giotto R package (RRID:SCR_027369); samples were normalized together, filtering out genes in <5 spots and spots with <300 genes. Differential gene expression at the spot level between groups of response or regions of interest was analyzed using SCRAN (RRID:SCR_016944) via Giotto, with FDR <0.001 and |Log 2 FC| > 0.5, where FC means fold change, as significance thresholds. GSEA (RRID:SCR_003199) was performed with clusterProfiler (RRID:SCR_016884; ref. 20 ) on differentially expressed genes (DEG) ranked by Log 2 FC, selecting the top 15 pathways (normalized enrichment score, FDR <0.01, ≥5 core genes). Deconvolution was carried out with SpaCET ( 21 ). MCP counter (RRID:SCR_027449; ref. 22 ) was used to establish a neoadjuvant induction score. For visualization, data were log 10 -transformed, with zeros set to 10 times below the smallest nonzero value. mIF and TLS quantification mIF was used to quantify TLS in 67 post-neoadjuvant surgical samples. FFPE sections (4 μm) were incubated overnight at 37°C, deparaffinized, rehydrated, and permeabilized with PBS containing 0.3% Triton X-100. Antigen retrieval was carried out in 10 mmol/L citric acid for 20 minutes, followed by blocking with 1% BSA for 1 hour. Slides were incubated overnight at 4°C with anti-CD23 (DAK-CD23, Ref:GA781, Agilent Dako), then with anti-CD20 (Clone L26; Ref:GA604; Agilent Dako) and conjugated anti-CD3 (Clone: SP162; Ref:ab135372; Abcam) antibodies, followed by secondary antibodies and DAPI. Sections were mounted with ProLong Diamond and scanned at zoom 10× on a Leica STELLARIS 8 confocal microscope. TLS were defined as organized B- and T-cell aggregates (>20 lymphocytes) and considered mature when displaying CD23 + germinal centers. TLS density was calculated as TLS per mm 2 . scRNA-seq of PBMC Single-cell gene expression profiling was performed on 30 PBMC samples (15 patients with paired blood samples, BS1 and BS2) using the Chromium Next GEM Single Cell 5′ platform (10x Genomics). PBMC were thawed, and gel beads-in-emulsion were generated by the Chromium X instrument (10x Genomics). Approximately 10,000 cells per sample were recovered, and libraries were prepared according to the manufacturer’s protocol. Sequencing was performed on a NovaSeq X Plus using a paired-end, dual-indexing workflow (read 1: 26 cycles; i7 index: 10 cycles; i5 index: 10 cycles; read 2: 90 cycles), targeting a depth of ∼20,000 reads per cell for the 5′ gene expression libraries (∼200 million reads per sample). After quality control and filtering, 26 samples remained for analysis. Reads were processed with CellRanger (version 6.0.1, RRID:SCR_023221), and doublets were removed using Scrublet (RRID:SCR_018098; ref. 23 ). Data were analyzed in R with Seurat (RRID:SCR_007322; 1,000–40,000 total counts, 400–5,500 detected features, <20% mitochondrial genes; ref. 24 ). Samples were normalized with SCTransform (RRID:SCR_022146; ref. 25 ), followed by PCA, clustering, and Uniform Manifold Approximation and Projection for Dimension Reduction visualization. Cell types were annotated using the Monaco Immune dataset via the SingleR package (RRID:SCR_023120; ref. 26 ). B and T cells were extracted and reprocessed similarly, yielding 11,053 B cells and 91,754 T cells for downstream analyses. Differential gene expression between CPR and NCPR groups was assessed in each cell subtype, followed by GSEA (RRID:SCR_003199). Statistical analysis Survival analysis of the entire cohort was performed with Kaplan–Meier analysis and the log-rank test. Given the low sample size, we presumed a non-normal distribution; Mann–Whitney U and Kruskal–Wallis tests were performed to compare independent groups of patients classified by pathologic response. The Wilcoxon signed-rank test was used to compare paired samples from the same patient at different time points. Bivariate correlations using Spearman’s rho were calculated between BCR metrics and PD-L1, TMB, and TCR. Two-tailed P values < 0.05 were considered statistically significant. Areas under the curve (AUC) > 0.70 with significant P values in receiver operating characteristic (ROC) analysis were considered relevant biomarkers. Statistical analyses were performed using R (version 4.4.1, RRID:SCR_001905). Graphs were generated using GraphPad Prism (version 8, RRID:SCR_002798) and R (version 4.4.1, RRID:SCR_001905). Results Patient outcomes and established biomarkers All patients from the NADIM and NADIM II clinical trials with at least one available sample for the analysis of the role of B cells in treatment response were included (Supplementary Table S1). In total, 123 patients were analyzed: 98 received perioperative ChIO, and 25 received neoadjuvant chemotherapy alone and served as controls. The clinicopathologic characteristics of all patients are summarized in Table 1 . Table 1. Cohort clinical characteristics according to treatment regimen. ​ Experimental ( N = 98) Control ( N = 25) Age median, years (IQR) 64 (58–70) 63 (56–68) Sex, n (%) ​ ​ Male 66 (67) 14 (56) Female 32 (33) 11 (44) Smoking status, n (%) ​ ​ Former 42 (43) 8 (32) Current 51 (52) 17 (68) Never 5 (5) 0 (0) ECOG, n (%) ​ ​ 0 54 (55) 14 (56) 1 44 (45) 11 (44) Histology, n (%) ​ ​ Adenocarcinoma 50 (51) 10 (40) Squamous carcinoma 35 (36) 12 (48) Other 13 (13) 3 (12) PD-L1, n (%) ​ ​ >50% 37 (38) 6 (24) 1%–49% 12 (12) 9 (36) <1% 30 (31) 8 (32) Undetermined 19 (19) 2 (8) Pathologic response, n (%) ​ ​ Complete pathologic response 45 (46) 1 (4) Non-complete pathologic response 46 (47) 16 (64) Not resected 7 (7) 8 (32) Open in a new tab With a median follow-up of 32.69 months [95% confidence interval (CI), 29.83–36.24], a significant PFS benefit was observed in the experimental group compared with the control arm [ P < 0.001; hazard ratio (HR), 3.693; 95% CI, 1.60–8.53; Fig. 1A ]. OS was also improved in the experimental arm ( P = 0.003; HR, 3.347; 95% CI, 1.02–11; Fig. 1A ). Following neoadjuvant treatment, 45 out of 98 patients (46%) in the experimental group achieved a CPR (0% viable tumor cells), compared with only one out of 25 patients (4%) in the control arm ( Fig. 1B ). Within the experimental arm, patients with CPR tumors showed significantly longer PFS and OS than those with NCPR (PFS: P < 0.001; HR, 13.89; 95% CI, 5.61–30.28; OS: P < 0.001; HR, 15.08; 95% CI, 5.05–45.04; Fig. 1C ). Figure 1. Open in a new tab Treatment response, survival outcomes, and biomarker analysis. A, Kaplan–Meier survival curves for PFS and OS for patients treated with ChIO ( n = 98) and chemotherapy ( n = 25). B, Percentages of pathologic response within each treatment arm. C, Kaplan–Meier survival curves for PFS and OS for ChIO-treated patients with CPR ( n = 44) and NCPR ( n = 47). D, PD-L1 TPS comparison between tumors with CPR and NCPR in the experimental arm and its ROC curve. E, TMB comparison between tumors with CPR and NCPR in the experimental arm and its ROC curve analysis. F, Schematic of treatment, techniques, and analyses performed. Dotted lines for Kaplan–Meier curves indicate 95% CI. Ch, chemotherapy; NR, not resected. Bold text denotes statistical significance (*, P < 0.05; **, P < 0.01; ***, P < 0.001). We then evaluated the prognostic value of established biomarkers—PD-L1 expression and TMB—for predicting pathologic response in the experimental group. At baseline, CPR tumors showed higher PD-L1 TPS compared with NCPR tumors ( P = 0.001, Fig. 1D ); however, responses were observed in both PD-L1–negative and PD-L1–positive tumors, with an AUC of 0.697 for CPR prediction ( Fig. 1D ). Pretreatment TMB did not significantly differ between CPR and NCPR tumors ( P = 0.382, Fig. 1E ), with a limited AUC of 0.595 ( Fig. 1E ). In this context, where established biomarkers show limited predictive capacity, and considering prior evidence suggesting that B cells may play a key role in orchestrating antitumor immunity, exploring B-cell–related mechanisms and biomarkers becomes particularly relevant. Accordingly, Fig. 1F outlines the overall translational study design aimed at dissecting the contribution of B cells to treatment response. Tissue and blood BCR repertoire as biomarkers of complete pathologic response To characterize the tumor BCR repertoire and assess its potential as a predictor of pathologic response, we performed bulk BCR sequencing on 107 FFPE tumor samples (Supplementary Table S2) and analyzed key ecologic metrics (Supplementary Table S3). At baseline, there were no significant differences in clone count between tumors with CPR and those with NCPR. However, CPR tumors exhibited significantly higher BCR clonality ( P = 0.030) and greater Top 10 CS ( P = 0.043), along with a trend toward lower diversity ( P = 0.069), compared with NCPR tumors ( Fig. 2A ). These baseline differences were no longer evident after neoadjuvant treatment, primarily due to an increase in diversity ( P = 0.004) and a decrease in clonality ( P = 0.055) and Top 10 CS ( P = 0.027) exclusive of CPR tumors (Supplementary Fig. S1A). Figure 2. Open in a new tab Tumor and blood BCR repertoire metrics and clonal tracking differ according to pathologic response. A, Comparison of tumor tissue clone counts, Shannon diversity, clonality, and Top 10 CS between NCPR and CPR tumors before and after neoadjuvant treatment. B, Tumor tissue BCR clonal tracking from pretreatment to post-neoadjuvant comparisons between NCPR and CPR. CS occupied by shared clones before and after neoadjuvant treatment (FFPE shared clones); CS of pretreatment top 10 clones at pretreatment and post-neoadjuvant (pretreatment tissue top 10 follow-up in tissue), and CS of post-neoadjuvant top 10 clones occupied by clones that were already present at baseline (preexisting clones in posttreatment top 10). C, Number of shared clones between pretreatment or post-neoadjuvant tumor tissue and blood samples collected before treatment, after neoadjuvant, and after 6 months of adjuvant treatment. D, Clone count, Shannon diversity, clonality, and Top 10 CS comparisons between patients with NCPR and CPR tumors in blood collected before treatment, after neoadjuvant treatment, and after 6 months of adjuvant treatment. E, Heatmap with the percentage of clones, Shannon diversity, and clonality of blood BCR repertoire differences between patients with NCPR and CPR tumors according to isotypes before treatment, after neoadjuvant treatment, and after 6 months of adjuvant treatment PBMC. F, Pretreatment blood percentage of clones, Shannon diversity, and clonality comparisons between patients with NCPR and CPR tumors according to grouped non-switched clones ( IGHD and IGHM ) and switched clones ( IGHG1 , IGHG2 , IGHG3 , IGHG4 , IGHA1 , IGHA2 , and IGHE ). 6M, 6 months of adjuvant treatment; IGH, immunoglobulin heavy chain; post, post-neoadjuvant treatment; pre, pretreatment. Bold text denotes statistical significance (*, P < 0.05; **, P < 0.01; ***, P < 0.001). We next conducted clone tracking analysis on paired baseline and surgery tumor samples. CPR tumors exhibited a greater CS occupied by clones shared between both time points compared with NCPR tumors, both at baseline ( P = 0.020) and after neoadjuvant treatment [ P = 0.031; Fig. 2B (left)]. Given the association between baseline BCR clonality and treatment response, we further analyzed the dynamics of the top 10 clones identified at baseline. Following neoadjuvant therapy, these top 10 clones diminished their dominance within the repertoire, and no significant differences were observed between CPR and NCPR tumors in the CS occupied by the pretreatment top 10 clones in surgical tissue [ P = 0.503; Fig. 2B (central)]. Finally, we sought to further characterize the top 10 clones present after neoadjuvant treatment in CPR tumors. We found that in CPR cases, a significantly larger proportion of the posttreatment Top 10 CS was occupied by clones already present in the baseline repertoire compared with NCPR tumors [ P = 0.025; Fig. 2B (right); Supplementary Fig. S2A]. To further explore the potential systemic role of tumor-associated clones, we tracked all these tumoral clones in peripheral blood at baseline, at the time of surgery, and after 6 months of adjuvant therapy. No significant differences were observed in the number of tumor-associated clones (either baseline- or surgery-derived) detected in the blood during neoadjuvant treatment. However, a marked reduction in these clones was observed after 6 months of adjuvant therapy ( Fig. 2C ). Finally, we investigated the potential association between BCR repertoire metrics and PD-L1 TPS, TMB, histology, TP53 mutational status, and TCR. Neither BCR diversity nor clonality showed any correlation with pretreatment PD-L1, TMB, histology, or TP53 , whether analyzed as continuous or categorical variables (Supplementary Fig. S1B). A weak but statistically significant correlation was observed between BCR and TCR clonality in tumor samples (r = 0.268, P = 0.027), but not for diversity (Supplementary Fig. S1C). Next, we explored associations of BCR repertoire in blood and its potential as a source for predictive biomarkers of pathologic response. We performed bulk RNA-seq of 67 PBMC samples during perioperative treatment (Supplementary Table S4) and studied BCR repertoire metrics (Supplementary Table S5), including isotype subanalyses. No significant differences were observed in the number of clones, diversity, or Top 10 CS at any of the analyzed time points. However, patients whose tumors achieved CPR exhibited a less clonal BCR repertoire in peripheral blood both at baseline ( P = 0.013) and at 6 months of adjuvant therapy ( P = 0.001; Fig. 2D ). No significant differences between paired samples at consecutive time points were observed (Supplementary Fig. S2B). Differences between tissue and blood BCR metrics are shown in Supplementary Fig. S2C. Isotype-specific analyses revealed BCR repertoire features associated with response not apparent in global metrics ( Fig. 2E ). At baseline, differences between CPR and NCPR were observed across different isotypes in terms of relative percentage, diversity, and clonality. Notably, differences between CPR and NCPR patients were inverted depending on whether the isotype had undergone class switching (IgG and IgA) or not (IgD and IgM). Specifically, at baseline, patients with CPR tumors exhibited a higher proportion of class-switched clones, which displayed increased diversity and lower clonality compared with those class-switched clones from patients with NCPR tumors. In contrast, patients with CPR tumors showed a lower proportion of non–class-switched clones, characterized by reduced diversity and increased clonality ( Fig. 2F ; Supplementary Fig. S3A). Similar patterns were observed after neoadjuvant therapy and persisted at 6 months of adjuvant treatment, reinforcing these isotype-associated features as a stable characteristic of response at the systemic level (Supplementary Fig. S3B and S3C). Finally, there was no association between BCR and TCR diversity nor clonality in blood (Supplementary Fig. S1D). BCR-derived metrics from both tissue and blood showed potential as predictive biomarkers of pathologic response (Supplementary Fig. S1E). Tissue baseline diversity (AUC = 0.733), clonality (AUC = 0.775), and Top 10 CS (AUC = 0.758) demonstrated potential as predictors of pathologic response. Similarly, in blood, some of these isotype-specific metrics demonstrated potential as predictive biomarkers of response, with significant AUC values > 0.75. Neoadjuvant ChIO induces TLS formation, with enhanced functionality associated with CPR We first compared bulk gene expression profiles between pre- and posttreatment tumor samples from patients treated with neoadjuvant ChIO. This analysis identified 35 DEG ( Fig. 3A ; Supplementary Table S6). Neoadjuvant ChIO led to the upregulation of 17 genes, several of which are associated with B cells, including JCHAIN , CD79A , and MS4A1 . Using spatial transcriptomics data (Visium version 1, FFPE, 10x Genomics, 55 μm spot resolution), we observed that these induced genes were significantly enriched in mTLS regions ( Fig. 3B ; Supplementary Fig. S4A; B-cell– and TLS-related regions defined in Supplementary Figs. S5A, S5B, and S6; Supplementary Table S7), suggesting that neoadjuvant ChIO may promote TLS formation within the TME. Figure 3. Open in a new tab Neoadjuvant treatment promotes the formation of TLS within the TME. A, Volcano plot for pretreatment and post-neoadjuvant bulk RNA-seq samples. B, H&E stain, distribution of areas of interest, and induction scores for mTLS ( N = 223), non-mTLS ( N = 113), dispersed B cells ( N = 240), and low B-cell areas ( N = 3,291) within a representative post-neoadjuvant surgery sample. Induction score by area of interest in all 12 posttreatment surgery samples pooled: mTLS ( N = 1,125), non-mTLS ( N = 762), dispersed B cells ( N = 4,243), and low B-cell areas ( N = 36,667). N indicates the number of spots analyzed for each area of interest. C, Representative immunofluorescence at 10× magnification of an mTLS in a post-neoadjuvant surgery sample. T cells (CD3), green; B cells (CD20), blue; germinal center (CD23), cyan. D, TLS density for all post-neoadjuvant tumor surgery samples, for NCPR tumors, and by response in the experimental arm. Bold text denotes statistical significance (*, P < 0.05; **, P < 0.01; ***, P < 0.001). mIF assays confirmed a higher density of TLS in tumors treated with ChIO compared with those receiving chemotherapy alone ( P = 0.034), even among tumors that did not achieve a CPR ( P = 0.012). However, no significant differences in TLS density were observed between CPR and NCPR tumors within the ChIO-treated cohort ( Fig. 3C and D ). Similar results were obtained considering only CD23 + TLS in Supplementary Fig. S7. Despite the lack of differences in TLS density between CPR and NCPR tumors, we hypothesized that functional differences might exist. We performed a differential gene expression analysis using spatial transcriptomics data (Visium version 1, FFPE, 10x Genomics, 55 μm spot resolution) from B-cell– and TLS-related regions between CPR and NCPR tumors (Supplementary Fig. S5C; Supplementary Table S8). GSEA revealed that CPR-associated mTLS exhibited upregulation of several biological processes compared with NCPR mTLS, including enhanced antigen presentation and increased antibody-mediated immune responses ( Fig. 4A ; Supplementary Table S9). Figure 4. Open in a new tab The functionality and composition of CPR-achieving tumors are different. A, GSEA top 15 upregulated pathways in CPR mTLS ( N = 686) and in CPR low B-cell areas ( N = 11,317). Pathways displayed have P adjusted < 0.01 and core enrichment >20 genes. B, Deconvolution scores of B-cell, plasma, T CD4 + , and T CD8 + phenotypes by area of interest: mTLS ( N = 1,125), non-mTLS ( N = 762), dispersed B cells ( N = 4,243), and low B-cell areas ( N = 36,667). N indicates the number of spots analyzed for each area of interest. For violin plots, the y -axis is displayed on a log 10 scale. C, Deconvolution scores by area of interest and response. The y -axis is displayed on a log 10 scale. NES, normalized enrichment score. Bold text denotes statistical significance (*, P < 0.05; **, P < 0.01; ***, P < 0.001). Other tissue compartments outside TLS—such as dispersed B cells and even low B-cell areas—also exhibited increased B-cell activity and humoral responses in CPR tumors. Notably, these TLS-excluded regions additionally showed enrichment of pathways associated with enhanced T cell– and NK cell–mediated antitumor responses, supporting a broader landscape of immune activation beyond B cells in CPR patients (Supplementary Fig. S5D; Supplementary Table S9). To further explore spatial patterns potentially associated with treatment response, we performed cell type deconvolution using the SpaCet algorithm to infer the relative proportions and distribution of different immune cell populations within the tissue ( Fig. 4B ; Supplementary Fig. S8). The deconvolution results revealed distinct spatial patterns for each cell type compatible with H&E microanatomy. As expected, B cells were primarily localized within TLS regions; however, plasma cells exhibited a more diffuse distribution, with higher densities observed outside TLS. Similarly, CD4 + T cells were predominantly enriched in TLS-associated areas, whereas CD8 + T cells showed a more widespread, dispersed spatial pattern across the tissue. Consistent with the GSEA findings, inferred immune cell composition also differed between CPR and NCPR tumors, exhibiting region-specific differential response patterns (Supplementary Fig. S9; Supplementary Table S10). In CPR tumors, mTLS showed increased proportions of T follicular helper (Tfh) cells, plasmacytoid dendritic cells (pDC), and plasma cells, along with decreased frequencies of total B cells and naïve B cells compared with NCPR mTLS. Dispersed B-cell regions from CPR tumors showed elevated levels of switched memory B cells and plasma cells compared with NCPR tumors. B-cell–low areas in CPR tumors showed higher levels of CD8 + T cells, NK cells, and macrophages while showing reduced levels of neutrophils and regulatory T cells compared with NCPR tumors (representative plots in Fig. 4C ; Supplementary Fig. S9). Finally, we assessed the spatial colocalization patterns of immune cell populations inferred by SpaCET, focusing on whether these interactions differed between mTLS from CPR and NCPR tumors. Notably, mTLS regions from CPR tumors exhibited a modest but significant positive correlation between B cells and CD4 + T cells, a relationship that was diffuse in mTLS from NCPR tumors (Supplementary Fig. S10; Supplementary Table S11). Peripheral blood B cells display distinct transcriptional features in patients achieving a CPR A total of 228,465 single cells were analyzed from 13 patients at two paired time points—baseline (BS1) and after neoadjuvant treatment (BS2)—yielding 26 PBMC samples in total. Among them, approximately 11,053 B cells were identified and classified into naïve B cells, memory B cells (further subdivided into non–class-switched and class-switched), and plasmablasts ( Fig. 5A ). Figure 5. Open in a new tab Peripheral blood B cells display distinct transcriptional features in patients whose tumors achieve CPR. A, Uniform Manifold Approximation and Projection for Dimension Reduction (UMAP) of scRNA-seq data showing all cells ( N = 228,465 cells) and B cells ( N = 12,651 cells) across all samples. B, Proportion of each B-cell subtype relative to its direct parent by pathologic response (CPR, N = 6 cases; NCPR, N = 7 cases) for pre- and posttreatment samples. C, Volcano plot and GSEA of B cells and switched memory B cells comparison between CPR and NCPR in pre- and post-neoadjuvant samples. B, B cell; CS-MB, class-switched memory B cell; MB, memory B cell; NB, naïve B cell; NCS-MB, non–class-switched memory B cell; NES, normalized enrichment score; PB, plasmablasts. No significant differences were observed in the proportion of B-cell subsets between patients with CPR and NCPR tumors at either time point. However, a trend for a higher percentage of class-switch memory B cells after neoadjuvant treatment in patients whose tumors achieved CPR compared with NCPR is observed ( P = 0.101; Fig. 5B ). Nevertheless, DEG and GSEA revealed subtype- and time point–specific transcriptional differences associated with treatment response ( Fig. 5C ; Supplementary Fig. S11; Supplementary Table S12). At baseline, total B cells from CPR patients showed increased enrichment of pathways related to adaptive immune responses. Notably, posttreatment samples revealed enrichment of immunoglobulin-mediated responses, suggesting higher activation of humoral immunity after neoadjuvant treatment in patients whose tumors achieved CPR. The reduced number of plasmablasts prevented DEG and GSEA. However, similar results to those observed in total B cells were seen across individual B-cell subpopulations, including naïve, memory, non–class-switched memory, and class-switched memory B cells (each with >200 cells per group). Particularly in class-switched memory B cells from CPR patients, we observed strong enrichment of pathways related to adaptive immune activation and immunoglobulin somatic recombination specifically after ChIO. Comparable findings were observed for T cells. Although no significant differences were detected in the relative proportions of major T-cell subsets between CPR and NCPR patients, transcriptional profiling revealed that patients whose tumors achieved CPR exhibited higher enrichment of adaptive immune activation pathways both at baseline and after neoadjuvant treatment (Supplementary Fig. S12; Supplementary Table S12). Discussion To our knowledge, this is the first study to provide a comprehensive, multiomic characterization of B-cell–mediated immune responses in patients with potentially resectable NSCLC treated with perioperative ChIO. By integrating tumor and peripheral B-cell repertoire analyses, bulk and spatial tumor transcriptomics, scRNA-seq, and mIF, we demonstrate that B cells and TLS are associated with the antitumor immune response in this setting. These findings have important implications for our understanding of response mechanisms and for the development of predictive biomarkers. Although not specifically designed for this objective, this translational cohort—including patients from the NADIM and NADIM II trials—provides additional support for the superiority of ChIO over chemotherapy alone, both in terms of pathologic response rates and survival outcomes. Our findings further confirm the long-term survival advantage associated with achieving CPR while also highlighting the limitations of current biomarkers such as TMB and, to a lesser extent, PD-L1 in CPR prediction. These results emphasize the clinical relevance of the early identification of patients whose tumors will achieve CPR and the need to unravel the distinct immune mechanisms driving durable responses in a potentially curable disease setting. In parallel with previous findings from our group and others on TCR repertoire as a potential source for CPR prediction ( 12 , 27 ), baseline BCR repertoire also exhibited higher clonality in tumors that ultimately achieved CPR. Notably, both tissue- and blood-based BCR metrics showed predictive accuracies (AUC 0.7–0.85) exceeding those of PD-L1 expression or TMB in our cohort. Importantly, these B-cell–related features did not correlate with PD-L1, TMB, and poorly with TCR metrics, suggesting that they may provide complementary predictive information. This is the first time BCR metrics have been associated with pathologic response in the context of perioperative ChIO in NSCLC. If validated in larger cohorts, these findings support the potential clinical value of BCR-derived biomarkers—particularly given their feasibility for minimally invasive assessment in peripheral blood. Other studies have also associated lower BCR diversity and higher clonality at baseline tumor samples with improved responses to immunotherapy or survival outcomes ( 28 – 30 ). This suggests that CPR tumors are characterized by the baseline expansion of specific T- and B-cell clones, likely reflecting specific neoantigen recognition and greater intrinsic immunogenicity. However, based on our data, we cannot confirm that these dominant baseline clones directly mediate the antitumor response, as they tend to contract by the time of surgery and show no significant differences in abundance between response groups at that point. Nonetheless, further supporting the notion of a more robust preexisting immunity in CPR tumors, we observed that these tumors retain a larger proportion of their initial BCR repertoire throughout treatment. This conservation suggests greater functional relevance and stability of baseline clones in CPR cases compared with NCPR tumors, which exhibit more extensive clonal turnover—likely reflecting a de novo immune response that is, by the definition of NCPR, less efficient at eliminating the tumor. Conversely, CPR tumors showed a greater CS occupied by reinvigorated baseline clones at the time of surgery, highlighting the importance of these clones in CPR compared with NCPR tumors. Altogether, the higher baseline clonality, repertoire conservation, and the prominent presence of baseline clones among the most expanded posttreatment clones support the existence of a preestablished and functionally superior B-cell response in tumors achieving CPR. With respect to the presence of tumor-derived B-cell clones in peripheral blood, we observed that both pre- and posttreatment tumor clones were detectable in circulation during neoadjuvant therapy. However, in both cases, the number of these clones markedly decreased after surgery. This finding suggests that the maintenance of circulating tumor-associated B-cell clones depends on local immune structures involved in clonal activation and proliferation—such as tumor-associated TLS, regional lymph nodes, and the presence of tumor-derived neoantigens—all of which are removed during surgical resection. Similar observations, including a postsurgical decline in circulating memory B cells, have previously been reported by our group in this setting ( 14 ) and by others in preclinical models ( 31 – 33 ). Beyond tumor-derived clones, analysis of the overall circulating B-cell repertoire demonstrated predictive value for pathologic response. In particular, patients with CPR tumors exhibited more developed B-cell repertoires, characterized by a higher proportion of class-switched clones, with greater diversity and lower clonality compared with patients with NCPR tumors. Notably, class switching requires functional germinal centers and Tfh cells, a condition we observed to be enhanced within the mTLS of patients achieving CPR. These findings further support the concept of a more advanced immune status at baseline in patients whose tumors achieve CPR, reinforcing the notion of preestablished immunity ( 34 , 35 ). Importantly, this distinct repertoire profile remained largely stable throughout the neoadjuvant phase and into adjuvant immunotherapy, indicating robust biological differences that could serve as a source for strong biomarkers, as these differences were consistently observed across three different time points. Mechanistically, these blood-based differences suggest an active, antigen-experienced B-cell response in patients with CPR tumors throughout treatment, potentially reflecting more effective immune priming and memory formation ( 34 , 35 ). Consistent with these findings, analysis of scRNA-seq data revealed a trend toward higher proportions of peripheral class-switched memory B cells in patients whose tumors achieve CPR, along with significantly upregulated pathways related to increased functional activity. Finally, the opposing dynamics observed between class-switched and non–class-switched clones help explain the lack of differences in the total B-cell pool, underscoring the need to analyze class-switched and non–class-switched B-cell populations separately. With respect to TLS, our data show that neoadjuvant ChIO treatment induces the expression of B-cell-related genes predominantly localized within mTLS regions, supporting their role as key structures orchestrating a coordinated B- and T-cell response to therapy ( 15 – 18 ). In this context, the addition of immunotherapy to neoadjuvant chemotherapy seems to promote an increase in TLS density and size at the time of surgery, consistent with our findings ( 36 – 38 ). However, this feature alone did not distinguish between tumors that achieved CPR and those that did not after ChIO in our cohort, which could be explained by the possibility that the peak TLS response occurs earlier during treatment ( 38 ). Nonetheless, this does not preclude a potential prognostic or predictive role for TLS in pretreatment samples ( 39 ). The absence of a direct association between TLS density and pathologic response after three cycles of neoadjuvant ChIO underscores the need to move beyond quantitative assessments and focus instead on the functional and spatial characteristics of the immune microenvironment ( 40 , 41 ). In this regard, mTLS from CPR tumors exhibited upregulation of pathways related to enhanced B-cell activity and antigen presentation, along with a higher estimated proportion of pDC and Tfh cells, and reduced levels of naïve B cells. These findings point to a more functionally mature and antigen-experienced mTLS compartment in complete responders ( 16 , 18 ). Moreover, cell–cell colocalization analysis revealed a significant spatial correlation between B cells and CD4 + T cells specifically within mTLS of CPR tumors—a pattern diminished in NCPR tumors. This finding may reflect enhanced structural organization of these immune niches and potentially facilitate T‐cell–B‐cell interactions within mTLS of CPR tumors ( 42 , 43 ). Outside TLS regions, dispersed B cells and low B-cell areas in CPR tumors showed increased pathways related to improved B, T, and NK cell–mediated immunity, with higher cytotoxicity and antibody responses. Cell type deconvolution further revealed increased levels of class-switched memory B cells, plasma cells, CD8 + T cells, NK cells, and macrophages, along with reduced frequencies of neutrophils, which have been associated with poor responses to immunotherapy through neutrophil extracellular trap formation ( 44 ). These findings support both an irradiated effector immune response from mTLS regions, as previously described ( 45 ) and the notion of a more effective antitumor immune activation beyond mTLS areas in CPR tumors. Remarkably, the increased cell activity observed in the TME also occurs at the systemic level, with an upregulation of gene expression pathways associated with enhanced adaptive immunity across the B- and T-cell subpopulations analyzed. Interestingly, non-mTLS regions in CPR tumors exhibited enrichment of both immunostimulatory and immunosuppressive pathways. This apparent paradox likely reflects spatial and temporal heterogeneity within these areas, capturing a mixture of early TLS formation and regions of mTLS resolution—recently described as involuted TLS ( 46 ). Given the current spatial transcriptomic resolution and the lack of specific markers to distinguish between eTLS and involuted TLS, these states remain indistinguishable in our dataset. This represents a limitation of our study and reinforces our focus on mTLS, which constitute a more defined and homogeneous immune structure for mechanistic exploration. Limitations of our study include the following: First, although our integrative multiomic approach supports a functional role for B cells and TLS in mediating the response to neoadjuvant ChIO, the study is observational in nature and does not provide direct mechanistic evidence. However, a key strength lies in the use of patient-derived samples rather than animal models, thus enhancing clinical relevance. Second, the relatively small sample size in certain high-resolution analyses limits subgroup analyses and underscores the need for validation in larger, independent cohorts. Third, the absence of single cell–level spatial resolution, inherent to the Visium platform, precludes a more granular characterization of cell compositions and interactions within the TME. Nonetheless, these limitations do not compromise the value of this study as an exploratory, hypothesis-generating analysis, which, to our knowledge, represents the largest and most comprehensive effort to date in the perioperative NSCLC setting. Collectively, our findings identify multiple layers of B-cell–related immune activity associated with CPR to neoadjuvant ChIO in NSCLC. Patients whose tumors achieve a complete response exhibit a preexisting and more mature B-cell response, detectable both in the tumor and in peripheral blood, a response that is better preserved and reactivated during perioperative treatment. This reactivation is reflected in the TME by the induction of more organized mTLS displaying increased activity of antitumor immune-related pathways and, systemically, by more mature and active circulating B-cell subpopulations. Taken together, these results support the use of B-cell–related parameters as early predictive biomarkers—that outperform PD-L1 and TMB—and suggest that strategies aimed at enhancing B-cell maturation and TLS functionality may improve clinical outcomes in NSCLC. Future work should explore the mechanistic underpinnings of these observations and assess their relevance across other tumor types and treatment regimens. Supplementary Material Table S1 Techniques performed for each patient and sample ccr-25-3315_table_s1_suppts1.xlsx (24.1KB, xlsx) Table S2 and S3 BCR tissue ccr-25-3315_table_s2_and_s3_suppts2-3.xlsx (10.8KB, xlsx) Table S4 and S5 Blood BCR ccr-25-3315_table_s4_and_s5_suppts4-5.xlsx (10.9KB, xlsx) Table S6 DEG bulk RNA-seq ChIO ccr-25-3315_table_s6_suppts6.xlsx (39.5KB, xlsx) Table S7 Visium DEG by main phenotype ccr-25-3315_table_s7_suppts7.xlsx (2.8MB, xlsx) Table S8 Visium DEG per main phenotype by response ccr-25-3315_table_s8_suppts8.xlsx (2.6MB, xlsx) Table S9 Visium GSEA by response results ccr-25-3315_table_s9_suppts9.xlsx (78.3KB, xlsx) Table S10 SpaCET deconvolution ccr-25-3315_table_s10_suppts10.xlsx (32.7KB, xlsx) Table S11 Phenotype-location correlations by response ccr-25-3315_table_s11_suppts11.xlsx (62.4KB, xlsx) Table S12 Single-cell DEG and GSEA by cell type and extraction B&Tcells ccr-25-3315_table_s12_suppts12.xlsx (5MB, xlsx) Figure S1 Supplementary Figure 1. Tumor BCR repertoire dynamics, correlations and potential as biomarker. ccr-25-3315_figure_s1_suppfs1.pdf (1.7MB, pdf) Figure S2 Supplementary Figure 2. BCR repertoire dynamics. ccr-25-3315_figure_s2_suppfs2.pdf (1.2MB, pdf) Figure S3 Supplementary Figure 3. Blood BCR repertoire according to isotype distribution. ccr-25-3315_figure_s3_suppfs3.pdf (2.9MB, pdf) Figure S4 Supplementary Figure 4. Normalized expression of B cell related ChIO-induced genes. ccr-25-3315_figure_s4_suppfs4.pdf (7.4MB, pdf) Figure S5 Supplementary Figure 5. Areas of interest represent different biological entities and differ by pathological response. ccr-25-3315_figure_s5_suppfs5.pdf (44.5MB, pdf) Figure S6 Supplementary Figure 6. B-cell regions analyzed by Visium spatial transcriptomics. ccr-25-3315_figure_s6_suppfs6.pdf (13.4MB, pdf) Figure S7 Supplementary Figure S7. Neoadjuvant Chemoimmunotherapy is associated to increased TLS density. ccr-25-3315_figure_s7_suppfs7.pdf (20MB, pdf) Figure S8 Supplementary Figure 8. Deconvolution scores reflect spatial tissue architecture. ccr-25-3315_figure_s8_suppfs8.pdf (38.2MB, pdf) Figure S9 Supplementary Figure 9. Deconvolution scores differ by pathological response in the areas of interest. ccr-25-3315_figure_s9_suppfs9.pdf (18.2MB, pdf) Figure S10 Supplementary Figure 10. Correlation analysis of deconvolution scores by pathological response. ccr-25-3315_figure_s10_suppfs10.pdf (13MB, pdf) Figure S11 Supplementary Figure 11. Comparison between Complete Pathological Response (CPR) (N=6) and Non-Complete Pathological Response (NCPR) (N=7) in pre- and post-treatment samples across all other cell types. ccr-25-3315_figure_s11_suppfs11.pdf (16.1MB, pdf) Figure S12 Supplementary Figure 12. Peripheral blood T-cells display distinct transcriptional features in patients whose tumors achieve CPR. ccr-25-3315_figure_s12_suppfs12.pdf (11.7MB, pdf) Acknowledgments Work in the authors’ laboratories was supported by the Instituto de Salud Carlos III (ISCIII) grants PI19/01652 and PI22/01223, co-funded by the European Regional Development Fund (ERDF); Bristol Myers Squibb; the Ministry of Science and Innovation (grants RTC2019-007359-1 BLI-O and CPP2022-009545 STRAGEN-IO); and the European Union’s Horizon 2020 research and innovation programme, P4-LUCAT project (PERME19002RODR and AC19/00026), to M. Provencio. A. Cruz-Bermúdez is supported by a Miguel Servet contract (CP23/00044) from the ISCIII and received an ISCIII project grant (PI23/01054), both co-funded by the European Union. C. Martínez-Toledo is supported by the Comunidad de Madrid PIPF-2022/SAL-GL-25283 contract granted to M. Provencio. M. Molina-Alejandre is supported by the Ayuda Predoctoral Asociación Española Contra el Cáncer (AECC) Madrid 2023 contract granted to M. Provencio. A. Gil-González is supported by the Comunidad de Madrid and the European Social Fund (PEJ-2023-AI/SAL-GL-27634) contract and is currently supported by the ISCIII FI24/00270 predoctoral contract granted to A. Cruz-Bermúdez. J.M. Gutiérrez-Escobedo is supported by the Comunidad de Madrid and the European Social Fund (PEJ-2023-AI/SAL-GL-27634) contract. We thank the patients and their families, the clinical teams, and the Spanish Lung Cancer Group for their essential contributions. We also acknowledge Bristol Myers Squibb for their support and the IDIPHISA Confocal Unit for assistance with the multiplex immunofluorescence panel. We thank technicians Ana Isabel Rodríguez, Daniel Lobato, and Sergio Espartero (Translational Lung Cancer Research Group, Department of Medical Oncology, IDIPHISA) for their excellent technical assistance. Footnotes Note: Supplementary data for this article are available at Clinical Cancer Research Online ( http://clincancerres.aacrjournals.org/ ). Contributor Information Alberto Cruz-Bermúdez, Email: [email protected]. Mariano Provencio, Email: [email protected]. Data Availability Raw sequencing data are not freely available due to the lack of specific authorization present in the original consent form signed by the patients during trial enrollment. Deidentified participant and molecular data will be made available upon reasonable request through the corresponding authors, requiring the approval of the SLCG and the institutional ethics committee. Authors’ Disclosures Á. Gil-González reports grants from the National Institute of Health Carlos III (ISCIII) during the conduct of the study and grants from ISCIII outside the submitted work. M. Molina-Alejandre reports grants from the Spanish Association Against Cancer (AECC) during the conduct of the study and grants from AECC outside the submitted work. E. Nadal reports personal fees from Amgen, AstraZeneca, BeOne, Boehringer Ingelheim, Daiichi Sankyo, Genmab, GSK, Illumina, Johnson & Johnson, Lilly, Merck Sharp & Dohme, PharmaMar, Pierre Fabre, QIAGEN, Regeneron, Sanofi, and Takeda and grants and personal fees from Bristol Myers Squibb, Merck Serono, Pfizer, and Roche during the conduct of the study. V. Calvo reports personal fees from Roche, AstraZeneca, MSD, Bristol Myers Squibb, Takeda, Regeneron, Amgen, GSK, Boehringer Ingelheim, Johnson & Johnson, BeOne, Pierre Fabre, and Pfizer outside the submitted work. M. Lázaro reports personal fees from Lilly, Ipsen, Bristol Myers Squibb, MSD, Astellas, Bayer, Merck, BeOne, AstraZeneca, Roche, and Pierre Fabre outside the submitted work. A. Insa reports other support from Amgen, Regeneron, Roche, Takeda, AstraZeneca, MSD, and Pfizer outside the submitted work. B. Massuti reports grants and personal fees from Roche, personal fees from Bristol Myers Squibb and BeOne, grants and nonfinancial support from AstraZeneca and Pfizer, and nonfinancial support from Merck MSD outside the submitted work. J. de Castro reports personal fees and nonfinancial support from AstraZeneca, Hoffmann-La Roche, and Merck Sharp & Dohme and personal fees from Bristol Myers Squibb, Boehringer Ingelheim, Janssen, Lilly, Sanofi, Takeda, Pfizer, Glaxo, Gilead, and BeiGene outside the submitted work. R. García Campelo reports personal fees from Bristol Myers Squibb during the conduct of the study and personal fees from AstraZeneca, Janssen, Lilly, Pfizer, Roche, and Takeda outside the submitted work. N. Reguart reports personal fees from AbbVie, Amgen, Arrivent Biopharma, AstraZeneca, Bayer, Bristol Myers Squibb, Gilead, Harpoon, Janssen, Johnson & Johnson, Novartis, Pharmamar, Regeneron, Revolution Medicines, Roche, Sanofi, Summit Therapeutics, and Tubulis and grants and personal fees from MSD during the conduct of the study. J. Bosch-Barrera reports personal fees from Regeneron, AstraZeneca, MSD, Bristol Myers Squibb, Pierre Fabre, Roche, Pfizer, Merck, Takeda, and Johnson & Johnson outside the submitted work. A. Aguilar reports nonfinancial support from Bristol Myers Squibb and Merck Sharp & Dohme-MSD, personal fees and nonfinancial support from Johnson & Johnson and Roche-Farma S.A., and personal fees from Takeda Oncology outside the submitted work. R. Palmero reports personal fees from AstraZeneca, Guardant Health, and Pfizer and nonfinancial support from Merck Sharp & Dohme outside the submitted work. J. Martín-López reports grants from Roche outside the submitted work. J.M. Gutiérrez-Escobedo reports grants from the Community of Madrid during the conduct of the study. C. Martínez-Toledo reports grants from the Comunidad de Madrid during the conduct of the study. M. Provencio reports grants, personal fees, and nonfinancial support from Bristol Myers Squibb, Roche, and AstraZeneca and personal fees from MSD and Takeda outside the submitted work. No disclosures were reported by the other authors. Authors’ Contributions B. Sierra-Rodero: Data curation, software, formal analysis, investigation, visualization, methodology, writing–original draft, writing–review and editing. Á. Gil-González: Data curation, software, formal analysis, investigation, visualization, methodology, writing–original draft, writing–review and editing. M. Molina-Alejandre: Data curation, formal analysis, investigation, visualization, methodology, writing–original draft, writing–review and editing. E. Nadal: Resources, data curation, investigation, writing–review and editing. V. Calvo: Resources, data curation, investigation, writing–review and editing. M. Lázaro: Resources, data curation, investigation, writing–review and editing. A. Insa: Resources, data curation, investigation, writing–review and editing. B. Massuti: Resources, data curation, investigation, project administration, writing–review and editing. A. Martinez Marti: Resources, data curation, investigation, writing–review and editing. J. de Castro: Resources, data curation, investigation, writing–review and editing. R. García Campelo: Resources, data curation, investigation, writing–review and editing. J.L. González Larriba: Resources, data curation, investigation, writing–review and editing. R. Bernabé: Resources, data curation, investigation, writing–review and editing. M. Dómine: Resources, data curation, investigation, writing–review and editing. S. Ponce Aix: Resources, data curation, investigation, writing–review and editing. M. Cobo: Resources, data curation, investigation, writing–review and editing. C. Camps: Resources, data curation, investigation, writing–review and editing. N. Reguart: Resources, data curation, investigation, writing–review and editing. J. Bosch-Barrera: Resources, data curation, investigation, writing–review and editing. M. Majem: Resources, data curation, investigation, writing–review and editing. A. Aguilar: Resources, data curation, investigation, writing–review and editing. R. Palmero: Resources, data curation, investigation, writing–review and editing. M. Blanco Clemente: Resources, data curation, investigation, writing–review and editing. J. Martín-López: Resources, data curation, investigation, methodology, writing–review and editing. R. Muñoz-Viana: Data curation, software, formal analysis, investigation, visualization, methodology, writing–review and editing. D. Megías: Data curation, software, formal analysis, investigation, visualization, methodology, writing–review and editing. J.M. Gutiérrez-Escobedo: Data curation, software, formal analysis, investigation, visualization, methodology, writing–original draft, writing–review and editing. C. Martínez-Toledo: Data curation, software, formal analysis, investigation, visualization, methodology, writing–original draft, writing–review and editing. A. Cruz-Bermúdez: Conceptualization, resources, data curation, software, formal analysis, supervision, funding acquisition, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. M. Provencio: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. References 1. Siegel RL, Kratzer TB, Giaquinto AN, Sung H, Jemal A. Cancer statistics, 2025. CA Cancer J Clin 2025;75:10–45. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 2. Thai AA, Solomon BJ, Sequist LV, Gainor JF, Heist RS. Lung cancer. Lancet 2021;398:535–54. [ DOI ] [ PubMed ] [ Google Scholar ] 3. Provencio M, Nadal E, Insa A, García-Campelo MR, Casal-Rubio J, Dómine M, et al. Neoadjuvant chemotherapy and nivolumab in resectable non-small-cell lung cancer (NADIM): an open-label, multicentre, single-arm, phase 2 trial. Lancet Oncol 2020;21:1413–22. [ DOI ] [ PubMed ] [ Google Scholar ] 4. Provencio M, Nadal E, Insa A, García Campelo R, Casal J, Dómine M, et al. Perioperative chemotherapy and nivolumab in non-small-cell lung cancer (NADIM): 5-year clinical outcomes from a multicentre, single-arm, phase 2 trial. Lancet Oncol 2024;25:1453–64. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 5. Provencio M, Nadal E, González-Larriba JL, Martínez-Martí A, Bernabé R, Bosch-Barrera J, et al. Perioperative nivolumab and chemotherapy in stage III non–small-cell lung cancer. N Engl J Med 2023;389:504–13. [ DOI ] [ PubMed ] [ Google Scholar ] 6. Forde PM, Spicer J, Lu S, Provencio M, Mitsudomi T, Awad MM, et al. Neoadjuvant nivolumab plus chemotherapy in resectable lung cancer. N Engl J Med 2022;386:1973–85. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 7. Cascone T, Awad MM, Spicer JD, He J, Lu S, Sepesi B, et al. Perioperative nivolumab in resectable lung cancer. N Engl J Med 2024;390:1756–69. [ DOI ] [ PubMed ] [ Google Scholar ] 8. Deutsch JS, Cimino-Mathews A, Thompson E, Provencio M, Forde PM, Spicer J, et al. Association between pathologic response and survival after neoadjuvant therapy in lung cancer. Nat Med 2024;30:218–28. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 9. Forde PM, Spicer JD, Provencio M, Mitsudomi T, Awad MM, Wang C, et al. Overall survival with neoadjuvant nivolumab plus chemotherapy in lung cancer. N Engl J Med 2025;393:741–52. [ DOI ] [ PubMed ] [ Google Scholar ] 10. Provencio M, Calvo V, Romero A, Spicer JD, Cruz-Bermúdez A. Treatment sequencing in resectable lung cancer: the good and the bad of adjuvant versus neoadjuvant therapy. Am Soc Clin Oncol Educ B 2022;42:1–18. [ DOI ] [ PubMed ] [ Google Scholar ] 11. Laza-Briviesca R, Cruz-Bermúdez A, Nadal E, Insa A, Del Rosario García-Campelo M, Huidobro G, et al. Blood biomarkers associated to complete pathological response on NSCLC patients treated with neoadjuvant chemoimmunotherapy included in NADIM clinical trial. Clin Transl Med 2021;11:e491. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 12. Casarrubios M, Cruz-Bermúdez A, Nadal E, Insa A, Del Rosario García Campelo M, Lázaro M, et al. Pretreatment tissue TCR repertoire evenness is associated with complete pathologic response in patients with NSCLC receiving neoadjuvant chemoimmunotherapy. Clin Cancer Res 2021;27:5878–90. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 13. Casarrubios M, Provencio M, Nadal E, Insa A, del Rosario García-Campelo M, Lázaro-Quintela M, et al. Tumor microenvironment gene expression profiles associated to complete pathological response and disease progression in resectable NSCLC patients treated with neoadjuvant chemoimmunotherapy. J Immunother Cancer 2022;10:e005320. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 14. Rodero BS, Martínez-toledo C, Nadal E, Molina M, Campelo RG, Gil-gonzález Á, et al. Peripheral memory B cell population maintenance and long-term survival after perioperative chemoimmunotherapy in NSCLC (NADIM trial). Oncoimmunology 2025;14:2513109. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 15. Molina-Alejandre M, Perea F, Calvo V, Martinez-Toledo C, Nadal E, Sierra-Rodero B, et al. Perioperative chemoimmunotherapy induces strong immune responses and long-term survival in patients with HLA class I-deficient non-small cell lung cancer. J Immunother Cancer 2024;12:e009762. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 16. Teillaud JL, Houel A, Panouillot M, Riffard C, Dieu-Nosjean MC. Tertiary lymphoid structures in anticancer immunity. Nat Rev Cancer 2024;24:629–46. [ DOI ] [ PubMed ] [ Google Scholar ] 17. Flippot R, Teixeira M, Rey-Cardenas M, Carril-Ajuria L, Rainho L, Naoun N, et al. B cells and the coordination of immune checkpoint inhibitor response in patients with solid tumors. J Immunother Cancer 2024;12:e008636. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 18. Fridman WH, Meylan M, Pupier G, Calvez A, Hernandez I, Sautès-Fridman C. Tertiary lymphoid structures and B cells: an intratumoral immunity cycle. Immunity 2023;56:2254–69. [ DOI ] [ PubMed ] [ Google Scholar ] 19. de Chaisemartin L, Goc J, Damotte D, Validire P, Magdeleinat P, Alifano M, et al. Characterization of chemokines and adhesion molecules associated with T cell presence in tertiary lymphoid structures in human lung cancer. Cancer Res 2011;71:6391–9. [ DOI ] [ PubMed ] [ Google Scholar ] 20. Xu S, Hu E, Cai Y, Xie Z, Luo X, Zhan L, et al. Using clusterProfiler to characterize multiomics data. Nat Protoc 2024;19:3292–320. [ DOI ] [ PubMed ] [ Google Scholar ] 21. Ru B, Huang J, Zhang Y, Aldape K, Jiang P. Estimation of cell lineages in tumors from spatial transcriptomics data. Nat Commun 2023;14:568. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 22. Becht E, Giraldo NA, Lacroix L, Buttard B, Elarouci N, Petitprez F, et al. Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression. Genome Biol 2016;17:218. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 23. Wolock SL, Lopez R, Klein AM. Scrublet: computational identification of cell doublets in single-cell transcriptomic data. Cell Syst 2019;8:281–91.e9. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 24. Stuart T, Butler A, Hoffman P, Hafemeister C, Papalexi E, Mauck WM 3rd, et al. Comprehensive integration of single-cell data. Cell 2019;177:1888–902.e21. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 25. Hafemeister C, Satija R. Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression. Genome Biol 2019;20:296. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 26. Monaco G, Lee B, Xu W, Mustafah S, Hwang YY, Carré C, et al. RNA-seq signatures normalized by mRNA abundance allow absolute deconvolution of human immune cell types. Cell Rep 2019;26:1627–40.e7. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 27. Zhang J, Ji Z, Caushi JX, El Asmar M, Anagnostou V, Cottrell TR, et al. Compartmental analysis of T-cell clonal dynamics as a function of pathologic response to neoadjuvant PD-1 blockade in resectable non-small cell lung cancer. Clin Cancer Res 2020;26:1327–37. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 28. Helmink BA, Reddy SM, Gao J, Zhang S, Basar R, Thakur R, et al. B cells and tertiary lymphoid structures promote immunotherapy response. Nature 2020;577:549–55. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 29. Selitsky SR, Mose LE, Smith CC, Chai S, Hoadley KA, Dittmer DP, et al. Prognostic value of B cells in cutaneous melanoma. Genome Med 2019;11:36. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 30. Song L, Ouyang Z, Cohen D, Cao Y, Altreuter J, Bai G, et al. Comprehensive characterizations of immune receptor repertoire in tumors and cancer immunotherapy studies. Cancer Immunol Res 2022;10:788–99. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 31. Liu J, O’Donnell JS, Yan J, Madore J, Allen S, Smyth MJ, et al. Timing of neoadjuvant immunotherapy in relation to surgery is crucial for outcome. Oncoimmunology 2019;8:e1581530. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 32. Dammeijer F, van Gulijk M, Mulder EE, Lukkes M, Klaase L, van den Bosch T, et al. The PD-1/PD-L1-checkpoint restrains T cell immunity in tumor-draining lymph nodes. Cancer Cell 2020;38:685–700.e8. [ DOI ] [ PubMed ] [ Google Scholar ] 33. Fear VS, Forbes CA, Neeve SA, Fisher SA, Chee J, Waithman J, et al. Tumour draining lymph node-generated CD8 T cells play a role in controlling lung metastases after a primary tumour is removed but not when adjuvant immunotherapy is used. Cancer Immunol Immunother 2021;70:3249–58. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 34. Victora GD, Nussenzweig MC. Germinal centers. Annu Rev Immunol 2022;40:413–42. [ DOI ] [ PubMed ] [ Google Scholar ] 35. Inoue T, Kurosaki T. Memory B cells. Nat Rev Immunol 2024;24:5–17. [ DOI ] [ PubMed ] [ Google Scholar ] 36. Sun X, Liu W, Sun L, Mo H, Feng Y, Wu X, et al. Maturation and abundance of tertiary lymphoid structures are associated with the efficacy of neoadjuvant chemoimmunotherapy in resectable non-small cell lung cancer. J Immunother Cancer 2022;10:5531. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 37. Schmid D, Sobottka B, Manzo M, Trüb M, Leonards K, Herzig P, et al. Tumor immune dynamics and long-term clinical outcome of stage IIIA NSCLC patients treated with neoadjuvant chemoimmunotherapy. Nat Commun 2025;16:8673. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 38. Jiang L, Li S, Zhou P, Huang Y, Chen M, Yang C. The germinal center-tertiary lymphoid structure after neoadjuvant chemo-immunotherapy for locally advanced lung squamous cell carcinoma can predict the disease progression. Front Immunol 2025;16:1579840. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 39. Peyraud F, Guégan J-P, Rey C, Lara O, Odin O, Del Castillo M, et al. Spatially resolved transcriptomics reveal the determinants of primary resistance to immunotherapy in NSCLC with mature tertiary lymphoid structures. Cell Rep Med 2025;6:101934. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 40. Rahal Z, Zhou T, Serrano AG, Feng J, Vaccaro A, Sinjab A, et al. Abstract 5820: spatial and cellular immune landscapes of non-small cell lung cancer in association with response to neoadjuvant chemoimmunotherapy. Cancer Res 2025;85:5820. [ Google Scholar ] 41. Liu Z, Yang Z, Wu J, Zhang W, Sun Y, Zhang C, et al. A single-cell atlas reveals immune heterogeneity in anti-PD-1-treated non-small cell lung cancer. Cell 2025;188:3081–96.e19. [ DOI ] [ PubMed ] [ Google Scholar ] 42. Kinker GS, Vitiello GAF, Ferreira WAS, Chaves AS, Cordeiro de Lima VC, da Silva Medina T. B cell orchestration of anti-tumor immune responses: a matter of cell localization and communication. Front Cell Dev Biol 2021;9:678127. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 43. Garaud S, Dieu-Nosjean M-C, Willard-Gallo K. T follicular helper and B cell crosstalk in tertiary lymphoid structures and cancer immunotherapy. Nat Commun 2022;13:2259. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 44. Shahzad MH, Feng L, Su X, Brassard A, Dhoparee-Doomah I, Ferri LE, et al. Neutrophil extracellular traps in cancer therapy resistance. Cancers (Basel) 2022;14:1359. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] 45. Meylan M, Petitprez F, Becht E, Bougoüin A, Pupier G, Calvez A, et al. Tertiary lymphoid structures generate and propagate anti-tumor antibody-producing plasma cells in renal cell cancer. Immunity 2022;55:527–41.e5. [ DOI ] [ PubMed ] [ Google Scholar ] 46. Shu DH, Ho WJ, Kagohara LT, Girgis A, Shin SM, Danilova L, et al. Immunotherapy response induces divergent tertiary lymphoid structure morphologies in hepatocellular carcinoma. Nat Immunol 2024;25:2110–23. [ DOI ] [ PMC free article ] [ PubMed ] [ Google Scholar ] Associated Data This section collects any data citations, data availability statements, or supplementary materials included in this article. Supplementary Materials Table S1 Techniques performed for each patient and sample ccr-25-3315_table_s1_suppts1.xlsx (24.1KB, xlsx) Table S2 and S3 BCR tissue ccr-25-3315_table_s2_and_s3_suppts2-3.xlsx (10.8KB, xlsx) Table S4 and S5 Blood BCR ccr-25-3315_table_s4_and_s5_suppts4-5.xlsx (10.9KB, xlsx) Table S6 DEG bulk RNA-seq ChIO ccr-25-3315_table_s6_suppts6.xlsx (39.5KB, xlsx) Table S7 Visium DEG by main phenotype ccr-25-3315_table_s7_suppts7.xlsx (2.8MB, xlsx) Table S8 Visium DEG per main phenotype by response ccr-25-3315_table_s8_suppts8.xlsx (2.6MB, xlsx) Table S9 Visium GSEA by response results ccr-25-3315_table_s9_suppts9.xlsx (78.3KB, xlsx) Table S10 SpaCET deconvolution ccr-25-3315_table_s10_suppts10.xlsx (32.7KB, xlsx) Table S11 Phenotype-location correlations by response ccr-25-3315_table_s11_suppts11.xlsx (62.4KB, xlsx) Table S12 Single-cell DEG and GSEA by cell type and extraction B&Tcells ccr-25-3315_table_s12_suppts12.xlsx (5MB, xlsx) Figure S1 Supplementary Figure 1. Tumor BCR repertoire dynamics, correlations and potential as biomarker. ccr-25-3315_figure_s1_suppfs1.pdf (1.7MB, pdf) Figure S2 Supplementary Figure 2. BCR repertoire dynamics. ccr-25-3315_figure_s2_suppfs2.pdf (1.2MB, pdf) Figure S3 Supplementary Figure 3. Blood BCR repertoire according to isotype distribution. ccr-25-3315_figure_s3_suppfs3.pdf (2.9MB, pdf) Figure S4 Supplementary Figure 4. Normalized expression of B cell related ChIO-induced genes. ccr-25-3315_figure_s4_suppfs4.pdf (7.4MB, pdf) Figure S5 Supplementary Figure 5. Areas of interest represent different biological entities and differ by pathological response. ccr-25-3315_figure_s5_suppfs5.pdf (44.5MB, pdf) Figure S6 Supplementary Figure 6. B-cell regions analyzed by Visium spatial transcriptomics. ccr-25-3315_figure_s6_suppfs6.pdf (13.4MB, pdf) Figure S7 Supplementary Figure S7. Neoadjuvant Chemoimmunotherapy is associated to increased TLS density. ccr-25-3315_figure_s7_suppfs7.pdf (20MB, pdf) Figure S8 Supplementary Figure 8. Deconvolution scores reflect spatial tissue architecture. ccr-25-3315_figure_s8_suppfs8.pdf (38.2MB, pdf) Figure S9 Supplementary Figure 9. Deconvolution scores differ by pathological response in the areas of interest. ccr-25-3315_figure_s9_suppfs9.pdf (18.2MB, pdf) Figure S10 Supplementary Figure 10. Correlation analysis of deconvolution scores by pathological response. ccr-25-3315_figure_s10_suppfs10.pdf (13MB, pdf) Figure S11 Supplementary Figure 11. Comparison between Complete Pathological Response (CPR) (N=6) and Non-Complete Pathological Response (NCPR) (N=7) in pre- and post-treatment samples across all other cell types. ccr-25-3315_figure_s11_suppfs11.pdf (16.1MB, pdf) Figure S12 Supplementary Figure 12. Peripheral blood T-cells display distinct transcriptional features in patients whose tumors achieve CPR. ccr-25-3315_figure_s12_suppfs12.pdf (11.7MB, pdf) Data Availability Statement Raw sequencing data are not freely available due to the lack of specific authorization present in the original consent form signed by the patients during trial enrollment. Deidentified participant and molecular data will be made available upon reasonable request through the corresponding authors, requiring the approval of the SLCG and the institutional ethics committee. Articles from Clinical Cancer Research are provided here courtesy of American Association for Cancer Research ACTIONS View on publisher site PDF (18.0 MB) Cite Collections Permalink PERMALINK Copy RESOURCES Similar articles Cited by other articles Links to NCBI Databases Cite Copy Download .nbib .nbib Format: AMA APA MLA NLM Add to Collections Create a new collection Add to an existing collection Name your collection * Choose a collection Unable to load your collection due to an error Please try again Add Cancel Follow NCBI NCBI on X (formerly known as Twitter) NCBI on Facebook NCBI on LinkedIn NCBI on GitHub NCBI RSS feed Connect with NLM NLM on X (formerly known as Twitter) NLM on Facebook NLM on YouTube National Library of Medicine 8600 Rockville Pike Bethesda, MD 20894 Web Policies FOIA HHS Vulnerability Disclosure Help Accessibility Careers NLM NIH HHS USA.gov Back to Top