How Bio-Techne and Nucleai’s AI spatial biology breakthrough could redefine melanoma biomarker discovery

A groundbreaking collaboration between Bio-Techne Corporation and Nucleai has revealed new potential in predictive biomarker discovery for metastatic melanoma. By combining high-plex spatial imaging with artificial intelligence, the partners demonstrated how mapping immune-cell interactions in the tumor microenvironment could forecast which patients respond best to immunotherapy. The findings, presented at the Society for Immunotherapy of Cancer (SITC) 2025 Annual Meeting, have positioned spatial biology as one of the most disruptive forces in next-generation oncology.

The study examined forty-two pre-treatment biopsies from the SECOMBIT clinical trial and correlated the spatial organization of immune cells with progression-free survival and overall survival. Instead of relying solely on the presence of biomarkers such as PD-L1, the research emphasized spatial proximity, orientation, and communication among immune cell types within tumor regions. The resulting patterns formed an interactive map that could reveal a patient’s likely therapeutic trajectory before treatment even begins.

How the AI-enabled COMET and Nucleai workflow connects high-plex imaging with spatial intelligence to identify predictive biomarkers in melanoma

Bio-Techne’s COMET platform, a fully automated multiplex immunofluorescence (mIF) system, serves as the physical and analytical foundation of the workflow. Capable of simultaneous detection of up to forty biomarkers, COMET captures tissue-level complexity in vivid detail. Once the imaging is complete, Nucleai’s multimodal spatial AI operating system interprets the data by integrating immunofluorescence, digital pathology, and clinical parameters into a single model.

This end-to-end workflow enables precise segmentation of immune and tumor cells while identifying higher-order spatial relationships—such as T-cell clustering, immune-cell exclusion zones, and macrophage infiltration. Through these relationships, researchers can distinguish between responders and non-responders with greater accuracy than any single-marker assay could achieve.

In melanoma, where checkpoint inhibitor response rates remain highly variable, such multi-dimensional insights are critical. The study revealed that specific immune neighborhoods—combinations of PD-1+ CD8 T-cells, PD-L1+ CD4 T-cells, and ICOS+ helper T-cells—aligned closely with improved survival. The proximity of these cells to the invasive tumor margin emerged as a particularly strong predictor of durable benefit. This reinforces the idea that not only immune composition but spatial architecture defines tumor immunogenicity.

By automating this analysis, the Bio-Techne–Nucleai workflow transforms what was once a slow, subjective pathologist-driven process into a reproducible, quantitative spatial metric. The resulting dataset functions as a new generation of biomarker—one rooted in the language of spatial interaction rather than expression level alone.

What the SECOMBIT trial reveals about therapy sequencing, immune-cell coordination, and melanoma treatment outcomes

The SECOMBIT study was designed to assess how different treatment sequences—targeted therapy followed by immune checkpoint blockade, immunotherapy first, or alternating regimens—affect patient outcomes. Within this framework, the Bio-Techne–Nucleai workflow provided a granular view of the immune ecosystem under each sequence.

In Arm A (MAPKi → ICB), the data showed that tumors rich in PD-L1+ CD8 cytotoxic T-cells and ICOS+ CD4 T-helper cells responded more favorably. In Arm B (ICB → MAPKi), successful outcomes were associated with spatial crosstalk between PD-1+ CD8 T-cells and PD-L1+ CD4 T-cells located at the invasive edge of the tumor, implying that early immune engagement at this frontier could predict checkpoint inhibitor success.

Arm C (MAPKi → ICB → MAPKi) produced yet another insight: antigen-presenting cell and T-cell interactions correlated with survival, while dense macrophage populations in the outer tumor microenvironment were linked to poorer prognosis. Across all three sequences, spatial relationships were stronger predictors of response than bulk expression or total immune-cell count, underscoring the clinical significance of spatial context in immuno-oncology.

The SITC selection committee’s decision to recognize this study among its top 150 abstracts—out of more than 1,200 submissions—highlights a turning point for the field. It signals that AI-driven spatial biology is evolving from exploratory science to translational relevance capable of influencing future trial design.

How spatial biology and AI integration could transform clinical trial design, data interpretation, and biomarker validation in oncology research

Spatial biology’s growing integration into translational medicine is changing how pharmaceutical companies design studies and interpret patient data. Bio-Techne’s COMET platform allows walk-away automation and simultaneous visualization of proteins and RNA molecules in tissue context, while Nucleai’s AI software uses deep learning to identify interaction motifs across millions of cells.

The result is a dataset that merges biology, histology, and data science—a true multimodal profile of tumor ecosystems. When applied at scale, this technology can shorten biomarker discovery cycles from months to weeks by reducing manual image annotation and analysis time. It also enables longitudinal assessment of immune architecture before and after therapy, offering real-time insight into drug efficacy.

For clinical trial sponsors, such tools could revolutionize patient stratification. Rather than recruiting broadly based on PD-L1 positivity or tumor mutation burden, sponsors could enroll patients exhibiting spatial patterns known to correlate with response, thereby improving statistical power and reducing failure rates in expensive phase 2 and 3 studies.

As the industry shifts toward adaptive trial models and AI-assisted biomarker endpoints, the Bio-Techne–Nucleai workflow could become a template for precision enrollment and on-therapy monitoring. Regulatory agencies are increasingly receptive to digital pathology and AI-enabled metrics when backed by robust validation, making this workflow’s scalability an attractive advantage for drug developers.

Why investor sentiment sees spatial biology as the next major growth driver within life-science tools and diagnostics

From a market perspective, Bio-Techne’s expansion into spatial biology fits a broader investor narrative that positions this field as one of the highest-growth verticals in life-science instrumentation. As of November 2025, Bio-Techne’s shares (NASDAQ: TECH) traded around US $61 per share—about 18 percent lower year-over-year amid broader biotech volatility. However, analysts remain upbeat about the company’s diagnostics and spatial-biology portfolio, viewing it as a structural growth engine supported by automation and partnership-driven innovation.

For Nucleai, a private company, the collaboration offers both validation and visibility. Its spatial AI platform is already being evaluated across lung, colorectal, and breast cancers, suggesting potential scalability far beyond melanoma. Together, the companies have positioned themselves as complementary leaders—Bio-Techne providing the imaging hardware and reagents, and Nucleai supplying the AI interpretation layer that unlocks biological meaning from complex tissue data.

Investor sentiment around these developments leans positive, particularly given that spatial analytics addresses one of oncology’s most persistent challenges: why similar patients respond differently to identical therapies. As the field matures, spatially defined biomarkers could form part of companion diagnostics for immunotherapies, introducing new recurring-revenue models for technology providers.

How the Bio-Techne–Nucleai partnership is reshaping the trajectory of precision medicine and predictive oncology

The collaboration between Bio-Techne and Nucleai represents more than a single research milestone—it reflects a broader shift in how oncology innovation is conceptualized and commercialized. The capacity to visualize and quantify the cellular choreography within tumors provides researchers with an unprecedented lens into the biology of response and resistance.

For physicians, the promise lies in tailoring therapy based not on static genomic or protein markers but on dynamic, spatially contextualized signatures. For biopharma developers, it means access to faster, data-rich feedback loops that can refine dosing regimens, combination strategies, and trial eligibility criteria.

While the dataset remains relatively small, its proof-of-concept strength lies in demonstrating that AI-enhanced spatial profiling can generate reproducible, clinically relevant correlations. Larger, prospective validation studies will determine whether these spatial metrics can be incorporated into regulatory submissions or companion diagnostic programs.

Still, the momentum is undeniable. As spatial biology moves closer to clinical integration, it could redefine how efficacy endpoints are established and how therapies are personalized. The Bio-Techne–Nucleai partnership exemplifies this transformation, bridging the gap between image and insight, between pathology and prediction, and between technology and tangible clinical value.