Connecting target engagement to treatment response with multimodal data
How a clinical-stage pharmaceutical company integrated imaging, omics and clinical data to investigate target engagement, downstream biological effects and treatment response.
Did target engagement produce the expected biological effect, and was that effect connected to treatment response?
The client had generated rich molecular, imaging and clinical data through its oncology programme. The challenge was no longer simply establishing whether the drug engaged its intended target. The team needed to understand what happened downstream: which biological processes changed following treatment, how those changes related across different data modalities and biological compartments, and whether any of those signals were associated with patient response. Sonrai worked alongside the client's scientific team to integrate these data and build a coherent translational picture of treatment activity.
A small drug developer with complex multi-modal data and an end-of-year trial deadline
A Boston-based, clinical-stage pharmaceutical company with over 200 employees needed a deeper biological understanding of their multi-modal data to stratify patients for an upcoming clinical trial. The company's R&D team comprised scientists, clinical researchers, bioinformaticians, and support staff, with a strong emphasis on translational research and the drug development pipeline.
"We'd been stitching imaging and omics together quite manually for months. Sonrai helped streamline the whole process, and we had stratification-ready insights in just a few days. It changed how we planned the trial."
Connecting biological signals across modalities, compartments and outcomes
The central challenge was not simply analysing each dataset independently. It was determining whether signals observed across different types of data formed a consistent biological story. The programme focused on three related scientific questions.
The analytical challenge
Answering these questions required the integration of data generated using different technologies and from different biological compartments. Transcriptomic measurements from tissue, proteomic measurements from plasma and urine, digital pathology and clinical outcome data each captured a different aspect of the treatment response. Relevant public datasets also provided an opportunity to place findings in a broader biological context. These analyses needed to be performed reproducibly, with clear provenance from the source data through to the final results.
Translational science supported by integrated, reproducible analysis
Sonrai scientists worked with the client's R&D team to define the analytical strategy around the programme's biological and clinical questions. Rather than treating each modality as an independent analysis, the work focused on connecting evidence across data types and determining where multiple signals supported the same biological interpretation.
Multi-Omic Factor Analysis (MOFA) within Sonrai Discovery — integrating and analysing multi-omic data to support clinical trial decisions.
A clearer translational picture of treatment activity
By bringing molecular, imaging and clinical evidence together, the client was able to evaluate treatment activity as a connected biological process rather than as a collection of separate datasets. The work provided:
- A clearer view of the biological effects associated with treatment
- Evidence linking molecular and tissue changes with clinical outcomes
- Candidate pharmacodynamic and predictive biomarkers for further investigation
- Greater biological context through integration with external datasets
- A reproducible analytical foundation that could support subsequent analyses as the clinical programme progressed
Most importantly, the programme helped move the scientific question from "did the drug engage its target?" towards "what happened biologically after target engagement, and does that help explain treatment response?" That evidence could then be used to inform biomarker strategy and future clinical development decisions.
Does your proposed mechanism translate into patient response?
Tell us what you need to understand about target engagement, biological effect and treatment response. Our scientific team can help define the evidence needed.