Spatial multi-omics analysis in single cells for precision oncology

Multiplex bioimaging of single-cell profiles in patient biopsies and artificial intelligence analysis of spatially distinct patient cohorts inform personalized cancer treatments.

Cancers are heterogeneous in distinct diseases and even among patients with the same cancer type. Tumors contain distinct microenvironments that often contain crucial information about the pathological state of the patient. These solid cancer masses are surrounded by a complex ecosystem that is composed of immune cells, stromal cells, and other extracellular proteins.  Despite intricate molecular compositions of cancers, histological analysis of these tumor ecosystems is still limited to a few markers in current medical practice. Thus, conventional imaging techniques are insufficient to design patient-specific treatment plans. Thereby, cancer patients often get “one size fits all” treatments which results in a significant portion of patients being under/over treated1.

In this paper2 (npj Precision Oncology on May 1, 2020), we envision that multiplex bioimaging is key to detect many biomarkers from the same tissue sample, allowing us to quantify expression levels of biomarkers and spatial formation of tissue architectures in biopsy samples. As a unique molecular solution, multiplex imaging can now achieve proteomic3, transcriptomic4, and metabolic5 profiling of tumor microenvironments from distinct patients at the single cell level (Figure 1). This spatial multi-omics information is critical for clinicians to design the personalized treatment regiments by choosing the appropriate chemotherapeutic agent/s, targeted therapy, or immunotherapy, providing ample opportunities for future drug design and development.

Figure 1. Precision oncology framework by spatial multi-omics. Distinct cancer patients (PT1-6: Patients 1-6) exhibit different pathophysiological signatures. Individuals exhibit color-codes (dark-red, red, pink, yellow, magenta, and cyan) for their tumor compositions. Cancer patients are composed of a unique molecular network across their bodies, providing a conceptual depiction of aberrant changes in diseases.  Biopsy samples from these patients are then analyzed by multiplex imaging platforms that comprise automated microscopes and microfluidic imaging instruments, wherein proteomic, transcriptomic, and metabolic profiles will be mapped out to reveal patient-specific tumor microenvironments (TME-1 for patient 1 and TME-2 for patient 2). Such a rich cellular big data inform diagnosis and therapeutic options of individual patients, providing quantitative metrics for clinicians in designing effective treatment regimens for future patients to reach to enable personalized therapies.

Furthermore, cancer masses disseminate circulating tumor cells (CTCs) that are found in the bloodstream of a patient in the form of “cell clusters” and “single cells”6. Capturing and analyzing these CTCs allow tracking disease progression and therapeutic response. While CTCs are currently incorporated into clinical decision making, molecular details of intracellular responses in CTCs have inefficiently been characterized. Since CTCs are rare cells, multiplex bioimaging offers unique advantages with their high-sensitivity even using low-input samples. Therefore, we proposed a “hierarchical microenvironments” viewpoint from solid tumors to “cell-cluster microenvironments”, and “subcellular microenvironments” in liquid biopsies. Spatially resolved, multiplexed cellular profiles in cell clusters and subcellular granularity will help decipher the tumor’s evolution and response dynamics of emerging therapeutic combinations.  

As another advancement in cancer technologies, our work describes the need for “multi-scale” spatial bioengineering of tumors that uniquely combines “bottom-up” systematic experimental design in cell cultures and “top-down” studies in sectioned biopsies from patients that leverage the rich data that is generated by the multiplex bioimaging platforms. Image-based drug discovery creates a library for therapeutic targets in patient-derived cells for personalized treatment design in two-cell, three-cell, and multi-cell cultures and organoid platforms. On the other hand, the complexity of three-dimensional tumor structures is elucidated from a well-associated transcript, protein, and metabolite profiles in single cells. This integrated framework across multiple scales shed light on molecular mechanisms of cancers.

Finally, multiplex bioimaging technologies have the potential to impact precision medicine7. Using training data from sets of patients’ image-based data along with their response to therapies, tumors’ multi-omics profiles can be quantitatively modeled using a cascade of high parameter imaging maps to decipher the disease mechanisms. The resulting computational framework can then be feed into artificial intelligence systems to categorize patients based on their disease state and drug response. This data-driven profiling will further aid physicians in analyzing new patients’ samples to reach precision therapies.


Ahmet F. Coskun, Ph.D.

Assistant Professor

Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, GA, USA

Mayar Allam and Shuangyi Cai

Ph.D. Students

Wallace H. Coulter Department of Biomedical Engineering, Georgia Institute of Technology and Emory University, Atlanta, GA, USA

For more information, please visit Single Cell Biotechnology Laboratory ( )


1. Shin, S. H., Bode, A. M. & Dong, Z. Precision medicine: the foundation of future cancer therapeutics. npj Precision Onc 1, 1–3 (2017).

2. Allam, M., Cai, S. & Coskun, A. F. Multiplex bioimaging of single-cell spatial profiles for precision cancer diagnostics and therapeutics. npj Precis. Onc. 4, 1–14 (2020).

3. Jackson, H. W. et al. The single-cell pathology landscape of breast cancer. Nature (2020) doi:10.1038/s41586-019-1876-x.

4. Xia, C., Fan, J., Emanuel, G., Hao, J. & Zhuang, X. Spatial transcriptome profiling by MERFISH reveals subcellular RNA compartmentalization and cell cycle-dependent gene expression. PNAS 116, 19490–19499 (2019).

5. Sun, C. et al. Spatially resolved metabolomics to discover tumor-associated metabolic alterations. PNAS 116, 52–57 (2019).

6. Lim, S. B., Lee, W. D., Vasudevan, J., Lim, W.-T. & Lim, C. T. Liquid biopsy: one cell at a time. npj Precis. Onc. 3, 1–9 (2019).

7. Schork, N. J. Personalized medicine: Time for one-person trials. Nature 520, 609–611 (2015).

Please sign in or register for FREE

If you are a registered user on Cancer Community, please sign in