Patients with late-stage melanoma, having exhausted standard treatment options, often find themselves with no other course except palliative care.
Now, a study published in Nature Medicine shows that a multiomics approach including a spatial single-cell proteomics test can inform treatment decisions for advanced melanoma patients – achieving a three-fold increase in progression-free survival compared to patients treated with standard of care.
Here, members of the clinical study team explain the approach, findings, and their implications:
Stéphane Chevrier is Chief Scientific Officer and co-founder of Navignostics and lead of the spatial proteomics study arm.
Andreas Wicki is Vice Director of the Department of Medical Oncology and Hematology at the University Hospital Zürich, and clinical study lead.
Bernd Bodenmiller is co-founder of Navignostics, Director of Technologies at the Comprehensive Cancer Center Zurich, and Dual Professor at the University of Zurich and ETH Zurich.
The study showed that multiomics data can be delivered within a 48-hour turnaround. What were the biggest logistical and technical challenges to that?
SC: Delivering molecular tumor boards with clinically relevant multiomics data within just a few weeks (and even 72 hours for some reports) meant solving challenges on multiple fronts. First, we had to work closely with clinical teams to establish efficient logistics, particularly for fresh tissue handling. To speed up execution, we automated many of the experimental and computational processes, and rigorously standardized the full pipeline to ensure consistent and comparable data across more than two years of patient samples.
The final challenge was translating this complexity into intuitive reports that oncologists could interpret in minutes for treatment decisions. We’ve incorporated these learnings in our work at Navignostics, enabling us to achieve a 48-hour turnaround while maintaining clinical rigor and scalability.
Among the nine technologies used, which provided the most actionable insights, and how did their value differ across patient groups?
AW: The selection of biomarkers for therapy prediction is highly dependent on the specific tumor entity and the therapeutic options available for that context. In our cohort, a central clinical question, particularly for patients harboring BRAF mutations, was whether to prioritize treatment with targeted therapies (BRAF and MEK inhibitors), or to consider immunotherapeutic approaches with immune checkpoint inhibitors.
To address this, read-outs from molecular technologies that are well-suited to detecting predictive markers for these therapeutic classes were preferentially selected. This included next-generation sequencing for identifying BRAF mutations or TMB, imaging mass cytometry and cytometry by time-of-flight for high-dimensional immune and tumor microenvironment profiling as well as activation of the MAPK pathway (pERK), and digital pathology to assess immunological biomarker expression patterns at the tissue level.
In clinical scenarios that extend beyond standard of care decision-making, where prediction of therapy response requires consideration of investigational or less well-established treatments for the specific entity, the biomarker panel was typically further broadened. In these cases, markers measured by functional assays, such as ex vivo drug sensitivity testing (Pharmacoscopy and 4iDRP), were often used to evaluate tumor responsiveness to a broader range of therapeutic agents, including off-label or combination treatments.
What are the key factors that enabled an objective response rate of 38 percent and disease control rate of 54 percent in the difficult-to-treat beyond standard of care group?
AW: Key factors contributing to improvement of clinical outcomes, such as response rate and disease control rate, are multiple. However, in the context of our approach, two aspects are particularly critical.
First, the extension beyond conventional markers incorporating novel read-outs enables a more precise and individualized treatment selection. Such comprehensive profiling allows oncologists to tailor the therapies with the unique molecular and cellular characteristics of each patient’s tumor.
This not only supports clinical decision-making in the beyond standard of care setting but can also strengthen the confidence of routine diagnostic findings, which often are limited to very few predictive markers, in patients receiving standard of care. However, the direct causal relationship between extended profiling and improved clinical outcomes needs to be further investigated in randomized clinical trials.
Second, while seemingly obvious, the availability of drugs for off-label use is a crucial determinant in the practical implementation of personalized therapy projects. This factor becomes especially important when reimbursement policies and/or institutional access are taken into account, as they can significantly influence whether a recommended off-label treatment is actually feasible for a given patient.
Given that reimbursement is a critical bottleneck in implementing multiomics-guided therapy, how close do you think we are to routine integration of such multi-platform diagnostics in standard care?
AW: As mentioned above, access to drugs is a critical element of success for a multiomics program. In Switzerland, off-label drug access has a specific regulatory framework (article 71), which offers opportunities for securing drug access and reimbursement if there is data supporting the therapeutic intervention.
On a diagnostic level, when we set out with the project, the total cost of testing across all nine technologies was approximately 140,000 Swiss francs per single patient. Today, the cost has dropped to less than one-tenth of the previous amount, and further reductions can be anticipated for several technologies as they continue to evolve for clinical applications.
Moreover, our analyses demonstrated that by refining the combination of technologies – specifically by minimizing redundancies between platforms – the overall cost can be further reduced. It could reach a level that approaches or potentially falls within the range of reimbursement provided today by health insurance for diagnostic procedures.
What role do you foresee for AI and machine learning in future iterations of the platform?
AW: The integration of machine learning and artificial intelligence will be pivotal if we truly aim to harness the full potential of omics technologies. Currently, we are only able to utilize a small fraction of the data generated. For example, in our study, we typically used up to five biomarkers for an individual patient, often selected in a biased approach, out of more than 40,000 measured features and approximately 500 GB of data generated per patient.
These figures clearly demonstrate that only computational approaches can enable the comprehensive interpretation and application of omics data for clinical decision-making. Beyond identifying optimal biomarkers or molecular signatures associated with therapeutic vulnerabilities, AI and machine learning will also play a key role in streamlining and optimizing diagnostic workflow.
What practical steps can clinical laboratories take today to begin aligning their practices with the precision oncology paradigm you envision?
BB: The field of precision oncology is rapidly advancing, but the infrastructure to support it still needs to be built. There are several concrete steps clinical laboratories can take today, such as investing in training teams to interpret multidimensional diagnostics like spatial proteomics. Pathologists can also take the lead in establishing standard operating procedures for how these results are reviewed and used in tumor boards to make this data actionable.
Beyond that, labs can work closely with oncologists to define clinical use criteria for these new tests – particularly in late-stage or guideline-exhausted cases. They can also support the development of appropriate patient consent procedures and push for standardized collection of clinical data, which is essential for interpreting diagnostics in context.
At a broader level, labs can advocate for reimbursement pathways, both for advanced diagnostics like spatial proteomics and for off-label treatments informed by precision data. This includes contributing real-world evidence and working with institutional stakeholders to demonstrate clinical value.
Newsletters
Receive the latest pathologist news, personalities, education, and career development – weekly to your inbox.
