Abstract LB123: Spatial immune architecture at single-cell resolution predicts response to atezolizumab plus bevacizumab in advanced HCC

Abstract Background: Atezolizumab+bevacizumab (atezo+bev) is a first-line therapy for advanced hepatocellular carcinoma (aHCC) with ∼30% objective responses. Single-cell studies identified responses linked to CD8⁺ T effector cells (CD8_Tex and Temra) and CXCL10⁺ macrophages (Macro_CXCL10), and resistance to TREM2⁺ macrophages (Macro_TREM2) and CD14⁺ monocytes (Mono_CD14). We hypothesized that, beyond cell composition, immune spatial organization within the tumor microenvironment (TME) determines response to atezo+bev. Methods: Single-cell spatial transcriptomics (ST) was performed on 27 aHCCs (9 atezo+bev responders and 18 resistant) using the Xenium Prime 5.000-gene panel (10x Genomics) customized with 100 additional genes. For local tissue organization, distances between cells were computed using the dbscan package [radius(r)=300µm]. Immune cell clusters were defined as groups of ≥3 immune cells within 100 µm, and were identified through a density-based algorithm. This spatial clustering was assessed using both ST and multiplexed immunohistochemistry (mIHC) data. Cell-to-cell interactions (neighborhoods) were analyzed (r=25µm). Ligand-receptor interactions were evaluated using a spatial neighborhood-based approach. Results: ST of 27 aHCCs yielded 4.6 million cells. In responders, CD8_Tex and Temra, and Macro_CXCL10 cells localized within the tumor and were closer to cancer cells (p=0.02, p=0.08, and p=0.1, respectively). Furthermore, in responders, tumor hepatocytes exhibited a significantly enriched lymphoid neighborhood composition and significantly higher rates of immune-recruting interactions, including PDL1-PD1, CXCL9-CXCR3 and CXCL10-CXCR3 (p<0.05). Non-responders were enriched in TREM2⁺ macrophage-containing clusters. Specifically, they showed Macro_TREM2-pure clusters (10% vs 0%, p=0.009), mixed Macro_TREM2-Mono_CD14 clusters (27% vs 5%, p=0.03), and mixed Macro_TREM2-CD8⁺ T-cell clusters (38% vs 22%, p=0.04). This was accompanied by enrichment of the CXCL12-CXCR4 chemokine axis, consistent with an immunosuppressive TME. Using mIHC, we confirmed enrichment of Macro_TREM2-pure (p=0.02) and Macro_TREM2-CD8⁺ mixed clusters (p=0.03). Conclusions: Atezo+bev response in HCC is marked by intratumoral effector CD8⁺ T cells and CXCL10_Macro, tumor hepatocytes enriched in lymphoid neighborhoods, and immune-recruiting ligand-receptor interactions, whereas resistance is driven by TREM2_Macro clusters and CXCL12-CXCR4 signaling. These findings identify spatial immune organization as a key determinant of response to atezo+bev. Citation Format: Anna Vila-Escoda, Marta Piqué-Gili, Marta Casado-Pelaez, Roser Pinyol, Ana Hernández de Sande, Verónica Davalos, Albert Gris-Oliver, Carla Montironi, Judit Peix, Daniela Grases, Ezequiel Mauro, Guillem Cano-Segarra, Sarah Cappuyns, Igor Figueiredo, Giorgio Ioannou, Edgar Gonzalez-Kozlova, Tim Meyer, Anja Lachenmayer, Jens U Marquardt, Helen L Reeves, Julien Edeline, Fabian Finkelmeier, Jörg Trojan, Sacha Gnjatic, Jean-Frederic Blanc, Richard A Hubner, Matthias Pinter, Tom Luedde, Arndt Vogel, Daniela Sia, Vincenzo Mazzaferro, Manel Esteller, Jeroen Dekervel, Eduard Porta-Pardo, Josep M Llovet. Spatial immune architecture at single-cell resolution predicts response to atezolizumab plus bevacizumab in advanced HCC [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(8_Suppl):Abstract nr LB123.

Authors

Institutions

Publication Details

Journal
Cancer Research
Published
2026-04-17
DOI
https://doi.org/10.1158/1538-7445.am2026-lb123
Primary Topic
Single-cell and spatial transcriptomics
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Abstract LB123: Spatial immune architecture at single-cell resolution predicts response to atezolizumab plus bevacizumab in advanced HCC

Carla Montironi, Sarah Cappuyns, Albert Gris‐Oliver, Fabian Finkelmeier et al.
Cancer Research
Single-cell and spatial transcriptomics
article

Abstract LB123: Spatial immune architecture at single-cell resolution predicts response to atezolizumab plus bevacizumab in advanced HCC

Carla Montironi, Sarah Cappuyns, Albert Gris‐Oliver, Fabian Finkelmeier, Marta Casado-Peláez, Roser Pinyol, Ana Hernández de Sande, Judit Peix, Josep Maria Llovet, Matthias Pinter, Vincenzo Mazzaferro, Julien Edeline, Jeroen Dekervel, Verónica Dávalos, D Sia, Anja Lachenmayer, Helen L. Reeves, Ezequiel Mauro, Guillem Cano-Segarra, Anna Vila-Escoda, J U Marquardt, Manel Esteller, Sacha Gnjatic, Daniela Grases, Arndt Vogel, Jean-Frederic Blanc, Eduard Porta-Pardo, Edgar Gonzalez-Kozlova, Jörg Trojan, Igor Figueiredo, Richard A Hubner, Tom Luedde, Giorgio Ioannou, Tim Meyer, Marta Piqué-Gili
article en

Abstract

Abstract Background: Atezolizumab+bevacizumab (atezo+bev) is a first-line therapy for advanced hepatocellular carcinoma (aHCC) with ∼30% objective responses. Single-cell studies identified responses linked to CD8⁺ T effector cells (CD8_Tex and Temra) and CXCL10⁺ macrophages (Macro_CXCL10), and resistance to TREM2⁺ macrophages (Macro_TREM2) and CD14⁺ monocytes (Mono_CD14). We hypothesized that, beyond cell composition, immune spatial organization within the tumor microenvironment (TME) determines response to atezo+bev. Methods: Single-cell spatial transcriptomics (ST) was performed on 27 aHCCs (9 atezo+bev responders and 18 resistant) using the Xenium Prime 5.000-gene panel (10x Genomics) customized with 100 additional genes. For local tissue organization, distances between cells were computed using the dbscan package [radius(r)=300µm]. Immune cell clusters were defined as groups of ≥3 immune cells within 100 µm, and were identified through a density-based algorithm. This spatial clustering was assessed using both ST and multiplexed immunohistochemistry (mIHC) data. Cell-to-cell interactions (neighborhoods) were analyzed (r=25µm). Ligand-receptor interactions were evaluated using a spatial neighborhood-based approach. Results: ST of 27 aHCCs yielded 4.6 million cells. In responders, CD8_Tex and Temra, and Macro_CXCL10 cells localized within the tumor and were closer to cancer cells (p=0.02, p=0.08, and p=0.1, respectively). Furthermore, in responders, tumor hepatocytes exhibited a significantly enriched lymphoid neighborhood composition and significantly higher rates of immune-recruting interactions, including PDL1-PD1, CXCL9-CXCR3 and CXCL10-CXCR3 (p<0.05). Non-responders were enriched in TREM2⁺ macrophage-containing clusters. Specifically, they showed Macro_TREM2-pure clusters (10% vs 0%, p=0.009), mixed Macro_TREM2-Mono_CD14 clusters (27% vs 5%, p=0.03), and mixed Macro_TREM2-CD8⁺ T-cell clusters (38% vs 22%, p=0.04). This was accompanied by enrichment of the CXCL12-CXCR4 chemokine axis, consistent with an immunosuppressive TME. Using mIHC, we confirmed enrichment of Macro_TREM2-pure (p=0.02) and Macro_TREM2-CD8⁺ mixed clusters (p=0.03). Conclusions: Atezo+bev response in HCC is marked by intratumoral effector CD8⁺ T cells and CXCL10_Macro, tumor hepatocytes enriched in lymphoid neighborhoods, and immune-recruiting ligand-receptor interactions, whereas resistance is driven by TREM2_Macro clusters and CXCL12-CXCR4 signaling. These findings identify spatial immune organization as a key determinant of response to atezo+bev. Citation Format: Anna Vila-Escoda, Marta Piqué-Gili, Marta Casado-Pelaez, Roser Pinyol, Ana Hernández de Sande, Verónica Davalos, Albert Gris-Oliver, Carla Montironi, Judit Peix, Daniela Grases, Ezequiel Mauro, Guillem Cano-Segarra, Sarah Cappuyns, Igor Figueiredo, Giorgio Ioannou, Edgar Gonzalez-Kozlova, Tim Meyer, Anja Lachenmayer, Jens U Marquardt, Helen L Reeves, Julien Edeline, Fabian Finkelmeier, Jörg Trojan, Sacha Gnjatic, Jean-Frederic Blanc, Richard A Hubner, Matthias Pinter, Tom Luedde, Arndt Vogel, Daniela Sia, Vincenzo Mazzaferro, Manel Esteller, Jeroen Dekervel, Eduard Porta-Pardo, Josep M Llovet. Spatial immune architecture at single-cell resolution predicts response to atezolizumab plus bevacizumab in advanced HCC [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(8_Suppl):Abstract nr LB123.

Cancer ResearchVol. 86(8_Supplement)
Institució Catalana de Recerca i Estudis Avançats (ES), University Hospital of Bern (CH), The Royal Free Hospital (GB), Medizinische Hochschule Hannover (DE), Düsseldorf University Hospital (DE), Josep Carreras Leukaemia Research Institute (ES), Mount Sinai Hospital (US), Centre Eugène Marquis (FR), Hospital Clínic de Barcelona (ES), Hôpital Cardiologique du Haut-Lévêque (FR), Technical University of Applied Sciences Lübeck (DE), Tisch Hospital (US), Centro de Investigación Biomédica en Red de Enfermedades Hepáticas y Digestivas (ES), NIHR Newcastle Biomedical Research Centre (GB), The Christie NHS Foundation Trust (GB), Fondazione IRCCS Istituto Nazionale dei Tumori (IT), Fundació Clínic per a la Recerca Biomèdica (ES), Consorci Institut D'Investigacions Biomediques August Pi I Sunyer (ES), Frankfurt Cancer Institute (DE), Heinrich Heine University Düsseldorf (DE), University College London (GB), Medical University of Vienna (AT), University of Lübeck (DE), Universitat Politècnica de Catalunya (ES), Icahn School of Medicine at Mount Sinai (US), KU Leuven (BE)
Sustainable cities and communities
Openalex Percentile: Top 8%
Single-cell and spatial transcriptomics
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.