Optimizing the Next Generation of Targeted Cancer Treatments
From multiplex assays to artificial intelligence, new technology is improving how patients are matched to targeted treatments.
Next-generation targeted therapeutics continue to revolutionize the treatment of many cancers by making it possible to personalize the plan of attack to each patient’s unique pattern of disease. Among the oncology treatments growing in popularity are antibody-drug conjugates (ADCs) and multi-specific antibodies that target tumor-driving biomarkers, and immune checkpoint inhibitors (ICIs) that improve the ability of the immune system to recognize and attack tumors.
Selecting the patients who are most likely to respond well to these treatments is an ongoing challenge for oncologists—and a major focus of diagnostics innovation. The development of tissue-based companion diagnostics that leverage immunohistochemistry approaches continues to grow, driven by new technologies designed to optimize the task of targeting new medicines to the patients most likely to respond. In fact, the market for oncology companion diagnostics is forecast to grow at a compound annual rate of 8.7% to $8.4 billion in 2030.1
Reaching those lofty expectations won’t be easy, however. Technologies such as multiplexing, spatial profiling, and artificial intelligence (AI) are already proving to be potentially game-changing in research settings. To successfully integrate these tools into companion diagnostics that meaningfully change the prognosis for cancer patients will require unprecedented collaboration between drug developers, technology innovators, and regulators.
Multiplexing for more precise patient matching
As cancer therapeutics evolve, the task of predicting which patients are likely to respond positively to targeted drugs becomes significantly more complicated. For example, PD-L1 expression is routinely assessed in selecting patients for treatment with ICIs.2 The challenge with PD-L1 as a standalone biomarker is that it doesn’t tell the entire story—other factors in the tumor microenvironment, such as the types, locations, and numbers of immune cells, also influence the response of the tumor to ICIs.
Multiplex technology has the potential to improve the ability of companion diagnostics to predict response by measuring several biomarkers simultaneously. Multiplexing can reveal details about the spatial context in which biomarkers are expressed within the tumor and its microenvironment.3 This spatial information enhances our ability to predict the behavior of the tumor, and by extension, the likelihood that the cancer will respond to specific targeted drugs.
For example, a tumor with a similar population of T cells encircling the tumor versus infiltrating the tumor is likely to have a different response to checkpoint inhibition. Hence, a diagnostic that assesses both the presence as well as the location of multiple indicators of a tumor’s susceptibility to checkpoint inhibition, including PD-L1 and T-cell markers, could simplify the challenge of identifying which patients are most likely to respond to an ICI. What’s more, a growing interest in combining ADCs or bispecific antibodies with ICIs is increasing the need for diagnostics that can simultaneously detect target antigens, such as HER2 and TROP2, along with immune checkpoint biomarkers like PD-L1 and PD-1.
That’s where spatial profiling could play an important role. Spatial profiling is a discovery-focused form of multiplexing that makes it possible to assess hundreds or even thousands of biomarkers in a single assay, including RNA and proteins, revealing details about the context in which they're expressed in the tumor and the tumor microenvironment. Analysis of this rich spatial expression data enables the identification of subsets of biomarkers that influence the behavior of the tumor, and by extension, the likelihood of response to targeted drugs.
For translational research studies, panels in the range of 10–50 biomarkers can be performed on research platforms by specialized R&D personnel. However, such panels and platforms are highly complex and lack the reproducibility and robustness for further development as in vitro diagnostic assays and devices that can be deployed broadly in clinical labs.
To realize the potential of spatial profiling, simplification of the platforms and downsizing of the panels is necessary. To ensure that the assays can be deployed broadly across clinical labs and provide robust and reproducible results, the number of biomarkers should be less than 10 and, ideally, no more than 3–5.4 The testing reagents, including both chromogenic- and fluorescence-based, should be compatible with standard automated immunohistochemistry stainers that are available in most labs.
Beyond the smaller panel size and automated staining requirements, diagnostic spatial assays will require imaging and analysis. Depending on the type of assay, either brightfield or fluorescence digital pathology scanners can be used for imaging the multiplex assays. Following imaging, computational analysis with AI approaches is needed for assessment of the multiplexed biomarkers and determination of results or scores linked, for example, to therapy response.
In summary, the translation of spatial profiling technology to the clinic requires downsizing of biomarker panels, readily available and reliable multiplex reagents, automated staining, brightfield or fluorescent imaging, and computational analysis, all of which should be linked in an end-to-end seamless and easy-to-perform workflow. How do we get to such an end-to-end solution?
Close collaboration within diagnostics companies between assay development (multiplex reagents and antibodies), platform development (automated staining and scanning), and software development (computational pathology and AI) is needed to bring spatial multiplex tests to the clinic. Diagnostics companies will also need to partner with regulators. To gain regulatory approval for multiplex diagnostics tests, developers will need clear guidance from the FDA and other regulatory bodies on the requirements for validating these complex biomarkers with associated computational analysis.
There have been some early successes in spatial multiplex development, including Acrivon’s Oncosignature® companion diagnostic for its targeted therapy ACR-368, a CHK1/CHK2 inhibitor for ovarian, endometrial, and urothelial cancers, being developed in partnership with Quanterix.5 The OncoSignature® assay received FDA Breakthrough Device Designation for ovarian cancer (2023) and endometrial cancer (2025), with Phase 2 trials showing a 62.5% response rate in OncoSignature-positive patients, supporting its predictive utility.6 The OncoSignature® assay could be the first FDA-approved spatial multiplex assay if ACR-368 is approved for ovarian and/or endometrial cancer in the coming years.
The future of AI in diagnostics
Interpreting multiplex assays is inherently complex. Pathologists and researchers must distinguish multiple biomarkers based on chromogenic or fluorescent signals generated by reagents—a task that becomes exponentially harder as the number of biomarkers increases.7 In the near future, AI will be integrated into digital pathology workflows, assisting pathologists in interpreting multiplex assays with greater ease, accuracy, and reproducibility.8 These AI systems will identify which cells express target biomarkers, quantify expression levels, and detect overlapping biomarker patterns that are difficult to discern manually.
AI-Enabled Imaging Solutions for Complex 3D Organoid Screening
This application note demonstrates how next-generation confocal imaging technology combined with AI-powered analysis tools enables researchers to capture high-quality images from 3D organoids.View App Note / Case Study
Beyond interpretation, AI is poised to transform patient selection through precise, quantitative scoring of biomarkers using novel approaches that cannot be performed via conventional manual microscopy.9 For example, AI can assist in the simultaneous assessment of multiple ADC targets with overlapping expression patterns and subsequent selection of the most appropriate ADC for treatment for tumor types with multiple ADC approvals.
In addition, AI will be used to assess biomarkers on a quantitative continuous scale.10 This approach is critical for some biomarker assessments because many ADCs remain effective even in tumors with very low biomarker expression—levels that traditional semi-quantitative scoring might overlook.11 AI-powered analysis tools will enable precise measurement of biomarker concentrations, allowing diagnostics companies in partnership with Pharma to develop biomarker assays with quantitative continuous scoring systems that correlate expression levels with the likelihood of therapeutic response.12 This will ensure that patients who could benefit from targeted agents are not missed and patients who are unlikely to respond are not treated unnecessarily.
The advantages of multiplexing extend far beyond matching patients to a single therapy. In many cases, oncologists must consider multiple targeted treatment options that can be administered sequentially or in combination. As AI-enabled multiplex assays advance, clinicians will gain the ability to optimize treatment sequencing, improving outcomes and potentially extending survival.
Thanks to these innovations in diagnostics in association with targeted cancer therapies, precision oncology will continue to evolve, delivering increasingly personalized care.13 To realize this vision, stakeholders across the oncology space—translational researchers, diagnostic developers, pharmaceutical companies, regulators, and healthcare providers—must continue to collaborate to accelerate the development and adoption of the next-generation diagnostic tools discussed here. This collaboration will ensure that the right medicines reach the right patients at precisely the right time.
Add Technology Networks as a preferred Google source to see more of our trusted coverage.
