Biomarker discovery is currently undergoing a technological renaissance, driven by breakthroughs in multi-omics, spatial biology, AI, and high-throughput analytics, which offer higher resolutions, faster speed, and more translational relevance than ever before. This is reshaping how research teams identify, validate, and translate biomarkers, transforming the entire biomarker discovery pipeline and elevating biomarkers from mere diagnostic tools to indispensable orchestrators of personalized treatment paradigms.
However, success depends on choosing the right technology for the research objective, disease context, and stage of development. This article introduces emerging technologies in biomarker discovery, explores how they are helping to transform the future of cancer research and treatment, and explains the importance of matching these new technologies to your research goals.
The evolving role of biomarkers in translational research
Cancer biomarkers (biological indicators that can be found in a patient's bodily fluids or tissues) can point to the presence of cancer and provide insights into its current and likely progression, including the potential for recurrence and what outcome can be expected. They play a central role in oncology research, with different biomarker types serving distinct purposes. Predictive biomarkers, for example, help research teams forecast how different patients will respond to therapeutics, while diagnostic biomarkers can be used to accurately identify cancer types and stages. Integrating these biomarkers into oncology research has revolutionized cancer treatment, driving advancements in both therapeutics and prognoses.
However, as clinical research problems become more complex, more adaptive and personalized trial designs are emerging. This means research teams are modifying trial designs based on the accumulation of data, rather than simply following a traditional, fixed design. During an adaptive trial, decisions are made in real-time and at numerous points, so they are reliant on accurate, actionable data. To support these new trial forms, biomarker development must become more precise and scalable. This will require a fundamental shift within biomarker analysis, from single-modality testing to an integrated, high-resolution analysis of disease biology.
Isolated measurements will not be sufficient; comprehensive biological signatures that capture the complexity of different cancers will be essential. Elevating cancer biomarker research will require shifting toward multiparameter approaches, incorporating dynamic processes, and immune signatures. Central to this approach is the integration of new technologies, including multi-omics, standardized assay platforms and protocols, integrative data analysis, and machine learning.
Spatial biology and multi-omic profiling in biomarker discovery
The complex heterogeneity of tumors (including their many different cell types and their dynamic local environments) makes it challenging to identify new biomarker candidates. In light of this, the emergence of spatial biology techniques has been one of the most significant advances in biomarker discovery as they can reveal the spatial context of dozens (or more) markers within a single tissue, enabling the full characterization of the complex and heterogeneous tumor microenvironment (TME).
Unlike with traditional approaches, spatial transcriptomics and multiplex immunohistochemistry (IHC) allow researchers to study gene and protein expression in situ without altering the spatial relationships or interactions between cells. This provides principle information about the physical distance between cells, which type of cells are present, the shape and size of cells, and they way these cells are organized. Using this information, researchers can not only identify which biomarkers are present, but also understand how they are organized within the tumor. Rather than simply measuring average gene expression, for example, these technologies allow them to identify novel biomarkers based on location, pattern, or gradient.
These spatial contexts are particularly important for biomarker identification, as the distribution of expression throughout the tumor is an important factor when considering the utility of a predictive biomarker. For instance, a biomarker may only indicate the presence of cancer when expressed in a specific region, different microenvironments may express different biomarkers that are relevant to different aspects of disease progression or therapeutic response, and cell interaction may itself be a useful marker. As an illustration, studies suggest that the distribution (rather than simply the absence or presence) of a spatial interaction can actually impact response.
When paired with multi-omic profiling (including genomic, epigenomic, and proteomic data), these new technologies provide an effective, holistic approach to biomarker discovery. By combining different types of data, multi-omics can reveal novel insights into the molecular basis of diseases and drug responses, identify new biomarkers and therapeutic targets, and predict and optimize individualized treatments. Notably, an integrated multi-omic approach played a central role in identifying the functional role of two genes, TRAF7 and KLF4, which are frequently mutated in meningioma.
Artificial intelligence and machine learning in biomarker analytics
In oncology, the use of artificial intelligence (AI) represents a transformative advancement in cancer diagnosis, prognosis, and treatment. As its analytical capabilities exceed human capabilities, AI is essential for analyzing the large volume of complex data generated by new technologies. It is capable of pinpointing subtle biomarker patterns in high-dimensional multi-omic and imaging datasets that conventional methods may miss, aiding not only the early discovery and treatment of cancer but the identification of new biomarkers.
Predictive models could ultimately facilitate a paradigm shift within oncology as they go further than just identifying biomarkers, to actually forecast future outcomes, enabling more personalized and effective therapies. These models make use of patient data to predict patient responses, the risk of recurrence, and likelihood of survival. AI-powered biosensors (which detect biomarkers) are already being used to not only process fluorescence imaging data to detect circulating tumor cells, but predict how these cancers will progress and suggest how different patients will respond to specific treatments.
Natural language processing (NLP) is also revolutionizing how researchers extract insights from clinical data, helping them annotate complex clinical data and identify novel therapeutic targets hidden in electronic health records. As these models can process vast amounts of information, they are able to identify links between biomarkers and patient outcomes which would be impossible to identify manually. As cancer remains a leading cause of death worldwide, cancer registers have been seen as a priority for NLP application. One such example is ongoing AI-powered research focused on gathering numerical data related to cancer, including tumor grade, size, and behavior.
Crown Bioscience integrates AI-powered analytics to enhance the discovery of clinically relevant biomarkers from complex datasets.
Advanced models: organoids and humanized systems
Advanced models, including organoids and humanized systems, represent another advance in biomarker discovery as these platforms can better mimic human biology and drug responses compared to conventional 2D or animal models. Specifically, organoids excel at recapitulating the complex architectures and functions of human tissues when compared to traditional 2D cell line models. While humanized mouse models mimic complex human tumor-immune interactions, overcoming the limitations of traditional animal models which cannot provide as reliable a reference for treatments in patients.
This means organoids are well suited to functional biomarker screening, target validation, and exploration of resistance mechanisms. They have been shown to play a key role in the identification of biomarkers for drug screening and can reveal how biomarker expression may change during treatment or as cancer progresses. In contrast, humanized mouse models allow research teams to complete studies in the context of human immune responses. They have been used in the development of predictive biomarkers and are particularly beneficial for research teams investigating response and resistance to immunotherapies.
However, these models become even more valuable for biomarker discovery and validation when used in conjunction, and integrated with, multi-omic technologies. By combining data from various models, research teams can enhance the robustness and predictive accuracy of their studies, paving the way for more personalized and effective treatments. Adopting a strategic, holistic approach can help researchers maximize the utility of every platform, amplifying the insights that can be extracted from them and bridging the gap between bench research and clinical application.
Choosing the right technology for your research goals
While these new technologies are creating additional opportunities and directions in biomarker discovery, not every technology is suitable for every study. Choosing the right platforms and approaches depends on the research objective, disease context, and stage of development, as well as more practical considerations like timelines and budgets. For example, research teams in the early stages of discovery work can make best use of AI-powered high-throughput approaches, while teams looking to validate early findings would benefit from spatial biology technologies that reveal how biomarkers function within the TME, or organoid models that confirm the functional relationships between biomarkers and different therapeutics.
Learn More
To find out more about which technologies, approaches, or platforms could benefit your research project, speak to the Crown Bioscience team
