SANOFI [EPA: SAN]

Dr. Virginia Savova

Will single-cell technology trigger a revolution in the world of drug discovery? While academia is convinced, ambivalence persists among those closer to the clinic. Despite increased availability of high-through-put commercial platforms, single-cell experiments at the bedside are still cumbersome (and expensive!) endeavors with long turn-around times. Their analytic value is largely exploratory, and the availability of reproducible workflows relating them to testable hypotheses is limited. Should we curb the enthusiasm?

To gain clarity on the potential of single-cell technology, let's take a historical perspective. In the not-so-distant past, drug discovery relied on a largely haphazard process, combining intuition, serendipitous observation, and trial-and-error experimentation. Therapeutic hypotheses were often derived from non-scientific paradigms or lacked explicit understanding of underlying mechanisms. Even today, drugs occasionally reach the market based solely on demonstrated efficacy, without a specific grasp of the underlying mechanism.

Genetics brought the first major disruption to this paradigm, establishing a mechanistic link between genes and diseases that extended beyond monogenic conditions. Quantitative analysis has shown that targets supported by genetic evidence are more likely to succeed in the clinic. However, genetics alone falls short in explaining the pathogenic process, and many diseases lack a genetic link altogether. Drug discovery now requires a new level of mechanistic explanation to improve success rates and address complex and heterogeneous disorders that represent areas of high unmet medical need.

“Single-cell technologies offer a cell-centric perspective on therapeutic hypotheses -- diseases are driven by disruptions in cellular phenotypes and intercellular interactions.”

Single-cell biology might just be the answer. Single-cell technologies offer a cell-centric perspective on therapeutic hypotheses -- diseases are driven by disruptions in cellular phenotypes and intercellular interactions. What does this mean in practice? It means that multifaceted diseases like Lupus Nephritis, Rheumatoid arthritis, ALS, and Parkinson's can be deconstructed into cell type-specific driving mechanisms and cell-cell communication networks. This approach uncovers common mechanisms underlying different clinical manifestations and identifies distinct patient endotypes with similar manifestations. In this paradigm, the target becomes the pathogenic cell itself and the gene merely the mark on the target. And while the idea of targeting the cell is not new – think B-cell depleting antibodies against CD19, CD20 (rituximab, ocrelizumab), and CD22 (epratuzumab), single-cell technologies allow for precision strikes on much narrower subsets. Drugs can now be designed to modify specific cellular pathogenic states, minimizing impact on the rest of the system. By discovering combination codes of surface molecule expression, therapeutic agents could be delivered to their precise cellular addresses by antibody drug conjugates or viral capsids with optimized specificity. We can also dissect the direct and indirect effects of drugs in the multicellular context, continuously refining targeting approaches and better understanding why they do or do not work in some patients.

Are we there yet? No. To fully harness the potential of single-cell technology, we must establish logistical and computational infrastructure and develop new operational frameworks for its routine use in clinical studies. This will accelerate the development and positioning of successful therapeutics and allow us to learn from failures by translating high-dimensional observations back to the laboratory. Creating large-scale cellular maps of diseases and therapeutic interventions, coupled with high-quality patient data and AI-powered algorithms, will enable a more rational and data-driven drug discovery process, mitigating risks and fostering innovation. We must remember that every time an investigational drug goes to the clinic, it becomes an experiment on a human being. Only through the lens of high-dimensional technologies can we fully utilize the valuable participation of our patients.