Gastrointestinal stromal tumors (GISTs), the most common gastrointestinal sarcomas, experience a high rate of resistance to standard treatment tyrosine kinase inhibitor (TKI) imatinib, leading to tumor progression. Identifying non-responders and offering precise alternatives are crucial, however, multi-omics studies of GIST are limited, which hinders GIST biology understanding and novel drug development. We performed whole-exome and transcriptomic sequencing on 106 primary GIST patients. Subtype-specific molecular features and therapeutic responses were further validated in vitro and in vivo. Unsupervised clustering analysis of RNA data classified them into four subtypes, including immune (G1), stromal (G2), proliferative (G3), and metabolic (G4) subtypes. The G4 subtype had more patients with primary non-gastric tumors, a poor response to neoadjuvant imatinib treatment and synchronous metastasis, with an overall poor prognosis. Notably, G4 subtype exhibited significantly enhanced aerobic metabolism. Moreover, G4 subtype harbored homologous recombination repair deficiency (HRD) signatures, with significantly reduced homologous recombination activity and upregulation of non-homologous repair pathways. By stratifying GIST cell lines into subtypes, we confirmed that the HRD-targeting poly ADP-ribose polymerase (PARP) inhibitor Olaparib suppressed tumor growth in the G4 subtype-representative GIST882 cell line. In vivo experiments demonstrated that Olaparib monotherapy significantly reduced tumor burden in GIST882 xenograft-bearing mice and synergized with imatinib. We systematically deconstructed the molecular subtypes of primary GISTs by integrated genomic and transcriptomic analysis. A specific GIST subtype characterized by poor treatment responses and prognosis, marked by the activation of aerobic metabolism and HRD features, may be a potential candidate for PARP inhibitors.

The existence of intratumoral tertiary lymphoid structures (TLS) has been reported to be correlated with reduced recurrence of hepatocellular carcinoma (HCC). However, the cellular characteristics and driving mechanisms of TLSs in HCC remain largely unknown. In this study, we compared the clinical outcomes of TLSs in HCC using whole-exome sequencing, bulk RNA sequencing, and single-cell RNA sequencing on a cohort of 339 patients with HCC belonging to different TLS groups. Intratumoral TLSs were significantly associated with improved recurrence-free survival in HCC (P = 0.00013), with higher maturity of TLSs correlating with better prognosis (P = 0.00033). A B cell-related seven-gene signature effectively predicted TLS presence (area under the curve = 0.78) and patient prognosis, outperforming previously reported signatures, which were validated in situ by spatial transcriptomic data. Bulk and single-cell transcriptomic analyses revealed that TLS-positive (TLS+) tumors were immunologically active and strongly associated with immunotherapy response signatures. IgG-producing plasma cells, identified as key effector subsets enriched in TLS+ tumors, exhibited clonal expansion, somatic hypermutation, and high-affinity antibody production. Among potential tumor-enriched TLS-associated genes, HAPLN3 was overexpressed in TLS+ HCC and induced high serum antibody titers (P = 0.0032). Spatial transcriptomics and in vivo experiments confirmed that HAPLN3 promotes B-cell activation, leading to suppressed tumor growth. Administration of HAPLN3 protein displayed immunostimulatory and antitumor effects in an orthotopic mouse model. These findings reveal that targeting TLS-associated B-cell responses or leveraging HAPLN3-specific immunity may offer therapeutic avenues for improving immunotherapy outcomes in HCC. See related Spotlight, p. 716.

Supplementary Table S4 Marker genes for B and Plasma cell subclusters.

Supplementary Table S3 List of gene signatures for scoring.

Supplementary Figure 5. HAPLN3 is positively associated with immune activation but has no direct impact on tumor cells, related to Figure 7.

Supplementary Figure 3. Annotation and analysis of scRNA-seq data, related to Figure 5.

Supplementary Table S1 Clinicopathological information of 314 patients from FAH-SYSU cohort.

Recent studies indicated that insufficient radiofrequency ablation (RFA) could endow hepatocellular carcinoma (HCC) with higher aggressive potential. Stress-induced phosphoprotein 1 (STIP1), which was found highly expressed in HCC, is a chaperone molecule mediating cell homeostasis under thermal stress. We aimed to explore the role of STIP1 on the metastasis of residual HCC after RFA. Mice model with orthotopic HCC implants or caudal vein injection were employed to assess potential of lung metastasis and/or intrahepatic metastasis (IHM) of HCC cells. Cell culture model was used to determine cell invasion, mesenchymal marker genes expression, and underlying molecular mechanisms. Clinical specimens were collected to analyze the relationship between STIP1 and clinical outcome. We found that insufficient RFA elicited more IHM of HCCLM3 tumors, which could be reduced by silencing STIP1. Knockdown of STIP1 also significantly decreased lung metastatic potential of HCCLM3 cells. In vitro, HCCLM3 and HepG2 displayed a spindle-shaped morphology with upregulation of STIP1 and mesenchymal markers after sublethal heat exposure. Mechanistically, heat exposure induced the formation of STIP1-heat shock protein 90 (HSP90) complex, which could shuttle epithelial transcription repressor Snail1 into nucleus and regulate mesenchymal gene transcription. Blocking the HSP90–STIP1 complex reduced the invasive potential of HCC cells after heat exposure. Using clinical specimen, we found that STIP1 was expressed significantly higher in metastatic tumor tissues and in sera from metastatic HCC patients (p < 0.05). The high expression of STIP1 was significantly linked to shorter recurrence-free survival (p < 0.05). To sum up, our study found that STIP1 is positively associated with the sublethal heat-induced cancer cell metastasis through mediating the mesenchymal gene transcription. Blocking STIP1 activity may suppress HCC cell metastatic potential after RFA.

BACKGROUND & AIMS:Local ablation triggers anti-tumor response and is regarded as an encouraging treatment combined with immunotherapy for hepatocellular carcinoma (HCC). Irreversible electroporation (IRE) is a novel ablative technique eliminating tumor cells by electroporation, however, the characteristics of IRE-induced immune microenvironment and underlying mechanism remain unclear. METHODS:We developed an orthotopic immunocompetent HCC mouse model and performed incomplete IRE-ablation. The post-IRE immune microenvironment was characterized by RNA sequencing, single-cell RNA sequencing, flow cytometry, and multiplex immunofluorescence. Cytokine-chemokine array and organoid-immune cell co-culture were used to investigate the functions and underlying mechanism. The combination therapy of IRE and anti-PD-1 was examined for HCC treatment in mice. RESULTS:IRE initially inhibited HCC growth, but tumors rapidly regrew by day 14 after ablation, with increased Ly6G+ Polymorphonuclear (PMN) Myeloid-Derived Suppressor Cells (MDSCs) infiltration and CD8+ T cell exhaustion. In patient-derived HCC organoids, IRE-induced CD8+ T cell cytotoxicity was suppressed by MDSCs. Ly6G antibody-mediated PMN-MDSC depletion repressed HCC regrowth after IRE-ablation in vivo (1.29±0.12 vs 0.49±0.09 g, p<0.001). Mechanistically, HCC cells degraded CXCL10 mRNA via YTH N6-methyladenosine RNA binding protein F2 (YTHDF2), which activated MDSCs that led to CD8+ T cells exhaustion. Combined anti-PD-1 and IRE reduced tumor burden by 84.3% compared with IRE alone (0.22±0.08 vs 1.43±0.13 g, p<0.01), prolonged survival (day-60 survival: 93.33% vs 40%, p<0.001), and induced long-term anti-tumor immunity in mice. Four unresectable recurrent HCC patients treated with IRE plus anti-PD-1 therapy showed improved recurrence-free time than patients receiving IRE alone. CONCLUSIONS:PMN-MDSCs mediated immunosuppression that promoted HCC regrowth after incomplete IRE ablation. Combining anti-PD-1 therapy is a promising approach to eliminate post-ablation residual HCC, supporting evaluation in prospecitve clinical trials. IMPACT AND IMPLICATIONS:Post-ablation recurrence remains a major challenge in HCC. Here, we show that although IRE initially activates CD8+ T cells but subsequently promotes PMN-MDSCs activation and CD8+ T-cell exhaustion through YTHDF2-mediated CXCL10 mRNA degradation, resulting in a suppressive tumor immune microenvironment (TIME). These findings identify a previously unrecognized mechanism of post-IRE tumor regrowth and provide a rationale for early combination with anti-PD-1 therapy. Although further prospective clinical validation is required, this work suggests that monitoring post-ablation immune changes and applying rational immunotherapy combinations may help optimize treatment strategies for patients with HCC undergoing local ablation.

Supplementary Figure 4. HAPLN3 is positively associated with immune activation but has no direct impact on tumor cells, related to Figure 6.

Supplementary Table S5 Marker genes for T and NK cell subclusters.

Supplementary Table S6. Full sequence of Hapln3 expression lentiviral vector

Supplementary Table S2 Clinicopathological information of 25 patients for single-cell RNA sequencing.

The prevailing notion is that effector T cell activation mediates anti-PD-1 efficacy in cancer. Here, we conducted a mechanistic study parallel to our phase 2 trial of perioperative anti-PD-1 therapy in patients with resectable recurrent hepatocellular carcinoma (HCC) (NCT04615143) to study its mechanism of action. Late-recurrence patients present two distinct subtypes characterized by T cell or B cell dominant responses in the tumor microenvironment by dynamic single-cell multi-omics analysis. Clonal antibody repertoire analysis and spatially paired scRNA-seq/BCR-seq reveal somatic hypermutation promoting antibody binding against hepatitis B virus core antigen (HBcAg) within tumor tertiary lymphoid structures (TLSs) in these type B-late recurrence patients. Mechanistically, HBcAg is exported into the extracellular space, triggering local B cell and antibody responses and complement activation. In mice, these high-affinity HBcAg-reactive antibodies lead to complement-mediated antitumor activity with enhanced anti-PD-1 efficacy. Thus, we uncover enhanced anti-virus B cell immunity within the TLS as a mechanism to anti-PD-1 in HCC.

Supplementary Figure 2. Validation of TLS signature and enrichment of plasma cells in TLS+ patients, related to Figure 4.

Microvascular invasion (MVI) is a key prognostic factor in hepatocellular carcinoma but is currently only detectable after surgery. Here, we develop MAPUSE, a deep learning model using contrast-enhanced ultrasound (CEUS) to predict MVI non-invasively. We train and test the model on 5148 CEUS videos from 1716 patients across multiple centers. Results show that MAPUSE achieves accurate MVI prediction (AUCs 0.835-0.978) across different tumor sizes, contrast agents, and prospective validations. Transcriptomic analysis links the model's predictions to CD8 + T cell immune infiltration, confirmed via the model's attention maps. In a clinical cohort, patients predicted as MVI-positive can benefit from post-ablation immunotherapy. MAPUSE thus enables preoperative, non-invasive MVI assessment and provides insights into the tumor immune microenvironment, offering a valuable tool for clinical decision-making.

Supplementary Figure 1. Validation of TLS significance in TCGA-LIHC cohort, related to Figure 1.