AI-Powered Spatial Biology Platforms Advance Predictive Biomarker Discovery in Cancer Immunotherapy
Key Insights
Bio-Techne and Nucleai (search) demonstrated AI-driven spatial biology workflow identifying predictive biomarkers in melanoma (search) patients treated with immunotherapy and targeted therapy combinations.
The SECOMBIT clinical trial data revealed distinct immune cell interactions correlating with survival outcomes across different treatment sequences in 42 metastatic melanoma (search) patients.
PredxBio (search) showcased multi-omic spatial analytics at SITC 2025, presenting findings on immune metabolism and stromal architectures that predict therapeutic response across multiple cancer types.
Two leading spatial biology companies presented breakthrough findings at the Society for Immunotherapy of Cancer (SITC) 2025 Annual Meeting, demonstrating how artificial intelligence-powered platforms can identify predictive biomarkers to guide cancer immunotherapy decisions.
SECOMBIT Trial Reveals Spatial Immune Signatures
Bio-Techne Corporation (search) and Nucleai (search) announced pivotal data from the SECOMBIT clinical trial, conducted in collaboration with Professor Paolo Ascierto from the University of Napoli Federico II (search) and Istituto Nazionale Tumori IRCCS Fondazione Pascale (search). The study was selected as one of the top 150 abstracts from over 1,200 submissions at SITC 2025.
Using Bio-Techne's COMET platform and a 28-plex multiplex immunofluorescence panel, researchers profiled 42 pre-treatment biopsies from patients with metastatic melanoma (search). Nucleai (search)'s multimodal spatial operating system integrated high-plex imaging, histopathology, and clinical outcome data to identify distinct immune cell interactions correlating with progression-free survival, overall survival, and clinical benefit across three treatment arms incorporating Immune Checkpoint Blockade.
Treatment-Specific Biomarker Patterns
The study revealed distinct predictive signatures for each treatment sequence:
Arm A (MAPKi → ICB): Immune activation markers including PD-L1 (search)+ CD8 (search) T-cells and ICOS (search)+ CD4 (search) T-cells linked to better outcomes.
Arm B (ICB → MAPKi): PD-1 (search)+ CD8 (search) T-cells in the tumor invasive margin and their interactions with PD-L1 (search)+ CD4 (search) T-cells correlated with improved survival.
Arm C (MAPKi → ICB → MAPKi): APC-T-cell interactions in tumor margins associated with better outcomes, while macrophage interactions in outer tumor microenvironment indicated poorer prognosis.
"This study exemplifies how our innovative spatial imaging and analysis workflow can be applied broadly to clinical research to ultimately transform clinical decision-making in immuno-oncology," said Matt McManus, President of the Diagnostics and Spatial Biology Segment at Bio-Techne.
Multi-Omic Spatial Analytics Expand Precision Oncology
PredxBio (search) presented complementary findings through four collaborative studies spanning multiple disease contexts, including head and neck cancer (search), non-small cell lung cancer (search), pulmonary fibrosis (search), and combination chemo-immunotherapy trials.
The company's research revealed how immune metabolism, spatial signaling, and molecular interactions within the tumor microenvironment determine therapeutic outcomes. Key findings included markers of Wnt/β-catenin signaling and glutamine utilization in head and neck squamous cell carcinoma (search) predicting cancer recurrence, and metabolic reprogramming patterns in macrophages predicting immunotherapy response in non-small cell lung cancer (search).
"By mapping immune and metabolic programs within the tumor microenvironment, we can start to predict which patients will benefit from specific immunotherapies and why," said Arutha Kulasinghe, Professor at Queensland Spatial Biology Centre (search), who co-authored two of the studies.
Technology Integration Drives Clinical Translation
Both platforms demonstrate the clinical potential of integrating artificial intelligence with high-resolution spatial biology data. Nucleai (search)'s approach combines high-plex spatial proteomics with histopathology and clinical information, while PredxBio (search)'s SpaceIQ platform uses explainable AI to decode tumor immune microenvironment architecture.
"Our multimodal spatial operating system enables integration of high-plex imaging, data, and clinical information to identify predictive biomarkers in clinical settings," said Avi Veidman, CEO of Nucleai (search). "This collaboration shows how precision medicine products can become more accurate, explainable, and differentiated when powered by high-plex spatial proteomics."
The research demonstrates that immune cell location and interactions within tumors significantly impact treatment success. By mapping these immune niches using AI and spatial biology, researchers can better predict which patients will benefit from specific therapies.
"Our explainable AI models reveal the microdomain architecture and cellular network wiring that define how tumors evolve and respond to therapy," said S. Chakra Chennubhotla, Co-Founder and Chief of AI at PredxBio (search). "This integration of network biology and spatial intelligence moves precision oncology beyond observation toward true biological discovery."
Professor Ascierto emphasized the clinical significance: "The SECOMBIT trial is a milestone in demonstrating the possible predictive power of spatial biomarkers in patients enrolled in a clinical study."