How AI and Spatial Biomarkers Are Transforming Antibody–Drug Conjugate Development
ADCs are a promising therapeutic modality for oncology.
Complete the form below to unlock access to ALL audio articles.
Spatial biomarkers are making their mark on oncology drug development.
By quantifying the spatial heterogeneity within tumor tissues, predictive spatial scores can help identify relationships between biomarker expression patterns and therapeutic outcomes. The scores also enable researchers to assess the spatial relationships critical to the mechanisms of antibody–drug conjugates (ADCs), such as the bystander effect, where a drug impacts not only the target cells but also those in their microenvironment.
Dr. Jason Reeves, director of application science at Nucleai , detailed all these relationships during Technology Networks’ “Spatial Biology Revolution 2024”. But what really got Reeves’ excited was the potential of using artificial intelligence (AI) to address critical challenges in ADC development.
The challenges in ADC development
ADCs are a promising therapeutic modality for oncology, harnessing the ability of antibodies to deliver cytotoxic agents directly to target cells. But their success hinges on spatial interactions within the tumor microenvironment.
A key challenge, according to Reeves, is the need to understand how ADCs interact spatially with antigen-positive cells and their neighbors.
“Antibody drug conjugates have a known effect that’s called a bystander effect,” he said during his presentation, “where they kill not only antigen-specific cells but also surrounding cells by delivering the payload to a local microenvironment.”
Another challenge is targeting co-expression. Some ADCs require multiple targets to be present within the same spatial context to activate properly. Traditional biomarker assays often fail to capture these complex spatial dynamics, limiting their utility in ADC research and clinical trials. Tumor heterogeneity compounds these difficulties.
“The complex spatial relationships between individual cells and even subcellular localization of proteins are critical to the response to ADCs,” Reeves explained.
The Spatial Biology Revolution 2024
Edit or remove this text before publishing content piece.View Online Symposium
Case example: AstraZeneca’s spatial biomarker research
AstraZeneca has been at the forefront of spatial biomarker research, particularly in ADC development. During his presentation, Reeves highlighted how the company has employed predictive spatial scores to assess tumor heterogeneity and its impact on patient response to ADCs. These spatial scores quantify tumor complexity by analyzing patterns of biomarker expression, cellular interactions and tissue architecture.
“AstraZeneca’s findings revealed that spatial heterogeneity – variations in the tumor microenvironment – plays a pivotal role in determining ADC efficacy,” Reeves said.
“For example, tumors with higher spatial heterogeneity exhibited differences in patient response rates, driven by the distribution of antigen-positive cells and their proximity to neighboring cells. By integrating predictive spatial scores into their clinical trial designs, AstraZeneca improved patient stratification and optimized therapeutic targeting.”
The role of AI in spatial biomarker analysis
AI will undoubtedly be a transformative tool in spatial biomarker research, enhancing the precision and scalability of data analysis. Reeves detailed how these new kinds of machine learning facilitate both the expansion of existing assays and the creation of novel analytical methods.
“AI can help expand what we can do with existing assays,” he said, “and create novel scores that are impossible for the human eye to calculate on their own.”
AI’s ability to process complex imaging data enables researchers like Reeves to capture spatial relationships that traditional methods cannot. His algorithms at Nucleai help quantify spatial proximity between target cells and their microenvironments.
In addition, machine learning supports clinical trial workflows by enabling rapid, automated processing of spatial biomarker data.
“We have the first AI prospective trial enrollment assay currently being used in the clinic for oncology,” Reeves noted.
Benefits of spatial biomarkers in ADC development
Integrating spatial biomarkers into ADC development provides multiple advantages:
- Improved predictive accuracy: Predictive spatial scores enhance the ability to forecast patient response, particularly for complex ADC mechanisms.
- Optimized clinical trials: Real-time spatial analysis enables better patient stratification, reducing costs and accelerating trial timelines.
- Enhanced therapeutic precision: Understanding spatial dynamics allows for more personalized treatment approaches, maximizing efficacy while minimizing adverse effects.
For example, spatial scoring can reveal that tumors with low antigen positivity still respond to ADCs due to favorable spatial arrangements, such as antigen-positive cells clustered near critical tumor regions. This insight could enable more patients to qualify for trials and receive potentially effective treatments.
Challenges and future directions
While spatial biomarkers and AI offer transformative potential, there are hurdles to overcome. Multiplex immunofluorescence, a leading spatial analysis tool, requires robust pipelines to handle the growing data complexity.
Reeves acknowledged this scalability challenge and emphasized the need for validation and regulatory readiness, noting: “We need to be building our algorithms in a way that they’re ready for regulatory agency review.”
Collaboration among stakeholders – pharmaceutical companies, AI developers and clinical researchers – is also critical. Future innovations, such as spatial transcriptomics and higher-plex assays, are expected to expand the applications of spatial biomarkers, offering even deeper insights into tumor biology and drug interactions.
Toward more effective and personalized treatments
Spatial biomarkers are redefining ADC development by addressing critical challenges in understanding tumor biology and drug mechanisms. Armed with advanced AI-driven tools and informed by the spatial scores they predict, companies like AstraZeneca and Nucleai are leading the charge in integrating spatial biomarkers into drug development. These innovations not only enhance therapeutic precision but also pave the way for more effective and personalized cancer treatments.
