Expanding liquid biopsy with proteomics

Executive summary

Circulating tumour DNA (ctDNA) has become a major focus of liquid biopsy research, providing information about the tumour genome and tumour-derived material in circulation. Yet it represents only one layer of cancer biology. Circulating proteins provide a complementary view, reflecting interactions between the tumour, its microenvironment and the host, including immune activity, inflammation, tissue injury and treatment-induced biological change. The studies reviewed here illustrate how proteomics can contribute information beyond ctDNA and how integrating these orthogonal signals may support the development of multimodal research models and adaptive clinical research strategies.

From tumour DNA to tumour biology

The appeal of ctDNA is clear: tumour-derived DNA fragments in blood can be studied to characterise mutations, estimate molecular burden and follow molecular changes over time. However, ctDNA is a selective measurement. It is well suited to describing tumour-derived genomic information but is less able to capture immune response, inflammatory state, stromal activity or wider systemic biology. In addition, small, early-stage and intrinsically low-shedding tumours may release limited DNA into circulation, creating challenges for ctDNA-based research in settings where circulating tumour material is scarce.

Proteomics offers a different perspective. Circulating proteins can reflect tumour activity alongside host response, immune engagement, tissue context and treatment-induced change. CancerSEEK provided an early research example of this principle. Cohen et al. (2018) combined mutations in 16 cancer-associated genes with eight circulating proteins in research on eight common cancers and tissue-of-origin information. The study demonstrated the potential value of combining orthogonal biological signals rather than relying on increasingly sensitive measurement of a single analyte.

Challenges associated with ctDNA-only liquid biopsies

Since then, a growing body of research work has further demonstrated the value of incorporating circulating proteins into liquid biopsy. The studies highlighted here illustrate how proteomics is contributing biological information that complements and extends beyond that provided by ctDNA alone.

Proteomics provides information beyond ctDNA

Strong evidence that circulating proteins provide biological information beyond ctDNA came from Arffman et al. (2024). By profiling 1,463 serum proteins from 109 individuals with high-risk large B-cell lymphoma, they found that:

• An inflammatory proteomic signature, characterised by cytokines, immune-checkpoint molecules and cytotoxic mediators, closely reflected the tumour microenvironment.
• Tissue analysis revealed transcriptional and histological evidence of an inflamed, yet functionally exhausted, immune landscape, demonstrating that the circulating proteome can serve as a window into both the tumour microenvironment and systemic host response.
• The “inflamed” signature predicted worse overall and progression-free survival independently of ctDNA, indicating that it captured biological information beyond tumour-derived DNA.
• Inflammatory protein signatures declined during treatment and rose again at relapse.

A similar principle emerged in metastatic colorectal cancer. In individuals treated with tislelizumab, cetuximab and irinotecan, Xu et al. (2024) utilized the Olink™ Target 96 Immuno-Oncology panel to identify a circulating immune signature that described what they termed the peripheral macroenvironment.

The research study showed that:

·       Individuals with a “hot” immune macroenvironment experienced better clinical outcomes.

·       A composite proteomic score predicted overall survival independently of ctDNA in a research setting.

·       Subjects with favourable immune-protein profiles generally exhibited lower ctDNA burden and greater ctDNA clearance during therapy, creating what the authors described as a biological seesaw between tumour dynamics and host immunity.

Toney et al. (2025) provide a further example in residual triple-negative breast cancer, where:

·       Exploratory treatment-specific proteomic signatures were associated with recurrence and shorter invasive disease-free survival.

·       These signatures also correlated with baseline ctDNA positivity.

·       Incorporating ctDNA into the proteomic models provided little additional predictive value.

Complementary biology can lead to better models

Recognizing that proteomics and ctDNA capture different dimensions of cancer biology creates an opportunity to use them together in ways that neither biomarker can achieve alone.

This principle is illustrated by Stensgaard et al (2023) in advanced non-small-cell lung cancer. The application of Olink Target 96 Immuno-Oncology panel identified baseline concentrations of the immune proteins FAS ligand and ICOS ligand as potential predictors of likely benefit from pembrolizumab, while early ctDNA clearance provided an independent measure of molecular response after treatment had begun. The greatest separation in progression-free and overall survival was achieved when baseline proteomic stratification and on-treatment ctDNA dynamics were interpreted together.

A similar approach was used to distinguish benign from malignant pulmonary nodules. Rather than relying on a single biomarker, Yang et al. (2024) integrated three complementary views of the lesion:

• plasma cfDNA methylation;
• plasma proteomics using Olink; and
• CT radiomic features.

Each modality captured a different aspect of tumour biology. The proteomic model outperformed the methylation model alone, but the greatest discriminatory power was achieved when all three signals were combined, particularly for nodules measuring 5–10 mm.

Proteomics predicts. ctDNA confirms.

The complementary strengths of proteomics and ctDNA can also be leveraged at different stages of a trial. The DANCER trial by Zhou and colleagues provides an example in a clinical research setting. In participants receiving neoadjuvant dalpiciclib and endocrine therapy for luminal B breast cancer:

• Baseline circulating immune proteins measured with Olink, integrated with tumour pRb expression, formed a baseline response index (BRI) that identified individuals most likely to respond before treatment began.
• Serial ctDNA measurements then assessed whether those predicted responses translated into molecular tumour regression during therapy.

The two analytes formed a two-step adaptive strategy capable of identifying both participants likely to benefit from therapy and those with persistent molecular disease despite an apparent biological response.

Although this approach requires validation in larger prospective studies, it illustrates how combining proteomics and ctDNA at different stages of a trial may support more adaptive clinical research strategies.

A broader future for liquid biopsy

Taken together, these studies suggest that the next advance in liquid biopsy will come from integrating complementary layers of biology. ctDNA remains unparalleled for measuring tumour burden, molecular response and the emergence of resistance. Proteomics contributes a different dimension, capturing the immune landscape, tumour microenvironment and systemic host response that shape cancer behaviour and treatment response. The future of liquid biopsy will therefore be defined not by a single biomarker, but by its ability to measure, and ultimately interpret, the dynamic dialogue between tumour and host.

Olink products and services are For Research Use Only. Not for use in diagnostic procedures.

References

1. Cohen, J. D., Li, L., Wang, Y. et al. Detection and localization of surgically resectable cancers with a multi-analyte blood test. Science 359, 926–930 (2018). https://doi.org/10.1126/science.aar3247
2. Stensgaard, S., Thomsen, A., Helstrup, S., Meldgaard, P. & Sørensen, B. S. Plasma immune proteins and circulating tumor DNA predict the clinical outcome for non-small-cell lung cancer treated with an immune checkpoint inhibitor. Cancers 15, 5628 (2023). https://doi.org/10.3390/cancers15235628
3. Arffman, M., Meriranta, L., Autio, M. et al. Inflammatory and subtype-dependent serum protein signatures predict survival beyond the ctDNA in aggressive B-cell lymphomas. Med 5, 583–602 (2024). https://doi.org/10.1016/j.medj.2024.03.007
4. Xu, X., Ai, L., Hu, K. et al. Tislelizumab plus cetuximab and irinotecan in refractory microsatellite stable and RAS wild-type metastatic colorectal cancer: a single-arm phase 2 study. Nature Communications 15, 7255 (2024). https://doi.org/10.1038/s41467-024-51536-x
5. Yang, M., Yu, H., Feng, H. et al. Enhancing the differential diagnosis of small pulmonary nodules: a comprehensive model integrating plasma methylation, protein biomarkers, and LDCT imaging features. Journal of Translational Medicine 22, 984 (2024). https://doi.org/10.1186/s12967-024-05723-5
6. Zhou, Y., Zhang, Z., Chen, H. et al. Efficacy, safety, and biomarkers of neoadjuvant dalpiciclib plus aromatase inhibitors in operable HER2-negative luminal B breast cancer: a prospective phase II study (DANCER). MedComm 6, e70402 (2025). https://doi.org/10.1002/mco2.70402
7. Toney, N. J., Lynch, M. T., Lynce, F. et al. Serum analytes as predictors of disease recurrence and the duration of invasive disease-free survival in patients with triple-negative breast cancer enrolled in the OXEL trial treated with immunotherapy, chemotherapy, or chemoimmunotherapy. Journal for ImmunoTherapy of Cancer 13, e011379 (2025). https://doi.org/10.1136/jitc-2024-011379