The Immuno-Revolution: Resolving “Complexity Exemplified”
A perfect storm of unsiloed data, constructive computation, and pioneering biotechnology is transforming our perception of the immune system.
How is this happening, and what traits will define the leaders Reading, Mapping, and Modulating immune biology (AKA “complexity exemplified”)?
“What we think we know about our immune system changes every 10 years” is a common refrain in the field. It’s not surprising. From the early days of Emil Behring’s discovery of antibodies we have haltingly pieced together aspects of immune biology, often taking one step forward and two back. Given the seemingly unmanageable diversity of cell types, cell states, recognition receptors, and interdependent biochemical signaling pathways, unraveling the complexity of the immune system is not for the faint of heart.
Yet, the enormous importance of doing so remains. With applications in treating, diagnosing, and preventing disease across cancer, infection, immune deficiency, allergy, aging, and more, you would be hard-pressed to find more valuable and pervasive biological knowledge than that of the immune system. In the current time of COVID-19, the limelight only shines brighter on the importance of our body’s line of defense.
Any claims of a “revolution” should understandably be met with skepticism. After all, “biology consumes everything”, and most of our attempts to transform our understanding of it have been met with staunch resistance in the form of indecipherable convolution and redundancy. However, we have some truths. Firstly, it is undeniable that the massive, archived body of immunology data and expertise is poised to be integrated with novel computational tools, leading to unanticipated insights. Additionally, it is also undeniable that we are on the cusp of generating massive experimental immune biology data through technologies in genome engineering, microscopy, multi-omics, cell-cell interaction screening, and more, at lower cost and greater efficiency than ever before.
Given the challenges of interrogating delicate immune cells and networks, these advancements in rapid computation and experimentation are necessary. Such progress will certainly help uncover insights regarding varied concepts such as the role of genetic heterogeneity in immune response, crosstalk between immune and cancer cells, and the duties of the prodigious numbers of inputs to migratory immune cells. Yet, “more” is not sufficient.
More and faster data and analysis is not revolution.
Lasting transformation does not solely come from “rapid” — it requires “valid”.
We are now executing biology more reproducibly, with reagent, process, and hardware automation control. We are generating data at greater resolution with tools such as high-dimensional imaging, mass cytometry and RNA-seq, and creating data pipelines characterizing the immune system down to single-cell and spatiotemporal levels. We are better able to detect the subtle, by collecting relevant data and metadata to build machine learning-driven models of biology (as opposed to simply testing rough hypotheses of it). Importantly, much of this data is being generated in more physiologically representative models, as platforms for primary cell research and organoids begin to demonstrate scalability. Given the temperamental behavior of immune networks, and historic challenges in working with models of the immune system, all these advancements in data validity are exceptionally impactful.
Essentially, the ability to generate and interpret reproducible, high-resolution, biologically relevant data will take us leaps forward in validating the disparate immune biology data we have amassed to date. We will move from deconstructed individual components to probabilistic models of systems, defined by second and third-order rules. Once we have developed confidence in our models of immune biology, we may be better able to use novel experimental data to fill the gaps.
This is the underpinning of the Immuno-Revolution.
Beyond the technological landscape, there is immense market momentum demanding the Immuno-Revolution. The import of the applications requiring an understanding of immune biology is reaching a crescendo, with approaches such as CAR-T, multi-specific antibodies, immuno-oncology combination therapy, allergy therapies, and immune-based diagnostics having unheralded initial results. However, truly scaling the impact of these approaches across disease areas, targets, and individual patients, will depend on a rich understanding of the immune system. Overall, there is a massive glut of immunology-related research (with nearly 4000 ongoing I-O trials in 2019 alone!), and the current COVID-19 pandemic is only adding fuel to the fire on the infectious disease front.
To summarize — we now have a merging of technological factors…
- Amassed (but siloed) immunology data
- Distinct, constructive computational advances
- Experimental platforms for valid immune data generation
…whose coalescence for immune research is demanded by the market.
These will spur an “Immuno-Revolution”, vastly enhancing our understanding of immune biology, and dramatically accelerating our ability to develop effective diagnostics and therapies targeting elements of immune biology.
This Immuno-Revolution will create immense opportunities for Immuno-Technology companies Reading, Mapping, and Modulating immune system elements.
Such Immuno-Technology companies will often cover 2, or even all 3 of these categories, and their focus will vary between tools, software, diagnostics, and therapeutics. These companies will drive the Immuno-Revolution, powered by the intersection of data science, computation, and life sciences technology, and buoyed by market demand across the healthcare value-chain and across therapeutic areas.
But who are these companies? And what will be the characteristics of Immuno-Revolution Leaders?
The Readers
As the immune system is founded on unparalleled genomic units of diversity — with antibody, T-cell receptor, and other identified genetic combinations accounting for over 47 billion immune proteins — it is here where the reading starts. Companies such as Adaptive Biotechnologies are charging forward with the genomic sequencing of B and T cell receptors, providing deep insights into immune response and leveraging the reads to identify antigen relationships and optimal receptors.
As the central dogma would suggest, RNA and protein-focused Immuno-Tech companies are not far behind. Cofactor Genomics’s “ImmunoPrism” assay is dedicated specifically to developing AI-driven models of immune cell gene expression profiles, and early-stage companies such as Nplex are developing nanotechnology-based, multiplex proteomic readouts of the immune system.
Lastly, mastery of the characterization of whole immune cells and patient samples themselves is gaining steam. Berkeley Lights is leaping into the limelight, demonstrating strong technological capabilities through an automated, optofluidic cell processing approach enabling the real-time characterization of individual cell antibody and cytokine secretion, cell-cell interaction, and much more. On the therapeutic side, academic spin-outs such as Tscan are driving platforms by which cell-cell interactions between T cells and target peptides can be rapidly identified and deconvoluted .
In a recurring theme across the immune-revolution, Readers will not content to remain under their classification, and will expand their footprint commensurate to the value they add.
Groups like Adaptive are already co-developing TCR therapies with established players such as Genentech, and Berkeley Lights is entering strategic partnerships with synthetic biology leaders such as Ginkgo. Never before have we had so many groups profiling the components of the immune system, and capturing the value such progress deserves.
The Mappers
But, what is to be done with the rich data generated by reading the disorganized codes of the immune system? Taking these codes, assembling them in a hierarchy, and effectively mapping or otherwise predicting immune biology. Gross oversimplification — discerning true “cause and effect”.
These are the data aggregators, the network biologists, the computational experts. Names like Immunai and Cytoreason come to mind, leveraging existing clinical data, multi-omic profiling data, and machine learning methodologies to elucidate and “un-silo” immune biology networks, patterns, and intelligence. They may generate or leverage the DNA, RNA, proteome, and cellular-level data of the Readers, but their value is in the application of this data to predictive models of immune relationships and function. Through the Mappers we may “translate” into a whole picture the vast data that both exists today and will be generated tomorrow.
Without Mappers, we are stuck with siloed knowledge, redundant experimentation, and a general feeling that the more we know, the less we understand.
Every player in the Immuno-Revolution must be at least somewhat of a Mapper. From the tools companies engineering novel formulations of antibody, to the cell therapy experts designing the next generation of CAR-T therapies, understanding the network biology of the immune system will be a must to designing optimal tools, services, diagnostics, and drugs.
The Modulators
Who will design the technologies by which we therapeutically modulate the immune system? The batch of companies enabling the the iteration of modulating technologies at unparalleled rates, across antibody drug discovery, immuno-oncology, cell therapy, and more. In the antibody space, groups like A-Alpha Bio are enabling unprecedented “library on library” screening of antibody-protein interactions, and companies such as Distributed Bio are democratizing access to “computational immuno-engineering” of the antibodies themselves. Of course, these antibodies are oft being targeted to modulate the immune system itself, as demonstrated by Harpoon Therapeutics’ T cell-engaging bispecific antibodies, or Verseau Therapeutics’ macrophage-targeting approach.
However, the immune modulation revolution is by far the most evident in the immune cell therapy space. The massive diversity of approaches is evident; groups are performing unprecedented rates of cell engineering, both in the academic research space at the UCSF Marson Lab, and commercially at Synthego. This cell engineering is occurring in a wealth of cell types, from approaches in natural killer cells (Nkarta), to macrophages (Carisma), and much more. The engineering of these cells is being enhanced by delivery platform players (Kytopen), and complexity of their engineering is being taken to the next level through the use of genetic circuits (Senti Biosciences).
This modulation is even extending to the cell therapy manufacturing space. Given the high expense of cell therapy manufacturing, innovators are finding new way to drive down the costs of the manufacturing process. Technologies such as acoustic cell processing at FloDesign Sonics have attracted much attention, and specialization has occurred in the area of development of automated, closed manufacturing processes for cell therapies at groups such as through Lonza’s Cocoon Platform.
The technologies employed by modulators to modify, condition, and manufacture immune components and cells will be as diverse as those components and cells themselves.
While armed with powerful technologies, the Modulators should be reminded of the adage “to a man with a hammer, everything looks like a nail”.
Focused deployment of technology based on sound insights from Reading and Mapping will enable the Modulators to achieve true therapeutic impact leveraging the immune system.
What Will Define Immuno-Revolution Leaders?
The above Readers, Mappers, and Modulators span the entire biotechnology value chain of applications. In any case, many of the front-runners in the Immuno-Revolution will have a number of similar empowering traits.
- Integration of clinical/amassed data sets, high-throughput data generation, and ideally in silico approaches. Historical genomic, phenotypic, and clinical data related to the immune system, while plentiful, is disparate, unstandardized, and suffers from batch effects. Validation of immune system biology will come not just from mapping this data, but being able to integrate such data with high throughput experimental immune biology. Additionally, while the role of AI in biology is still nascent, even completely in silico approaches may play a role. For example, companies like Lyell Immunopharma are leveraging Amazon Web Services compute power for the purpose of in silico design of cell therapy constructs. Success in predicting and networking immune biology will require orthogonal approaches, and those positioned to implement such methods will take the lead.
- Feedback Loop-Oriented Development (“Design-Build-Test-Learn”). The immune system is exceptionally finicky and lives in a continuum of levels of responsiveness. This essentially compounds the difficulty of predicting even “ordinary biology”. It is unlikely that the first attempts at predicting a model of network biology, an immune-related biomarker, or a cell therapy target will be correct. Success will come to those who learn as they go, continuously iterating to obtain finer resolution of the immune biology they are interrogating.
- Integration of immunology with complementary fields of biology. From its inception, immunology has been thought of as the study of defense against pathogens. We now know that the immune system interacts in intricate ways with many bodily systems, and that there are critical overlaps between immunology and microbiology, neuroscience, metabolism, and more. As a brief example, consider how a gene knockout in a particular T cell may influence cell metabolism — such a modification could enable a more persistent and effective cellular immunotherapy. Advancing the field of immunology will require an inclusive understanding of complementary fields of biology, and groups demonstrating such a holistic approach will be on the frontiers of discovery.
- Unity of biologists, engineers, and data scientists. Immune biology has historically been the domain of the tenured biologist, if for no other reason than the fact that its complexity was indecipherable to anyone else. With the advent of available compute power, better model-building ML algorithms, and the tools to generate data to fuel these algorithms, the relevance of tenure is not diminished, but enhanced. Only through intelligent integration of existing expertise with novel techniques will breakthroughs be made.
- Reducing risk in immune cell therapy development. While extremely promising, cell therapy development is still a risky process. This risk is across targeting, designing, and manufacturing the therapies. At this time (and for the foreseeable future) there will be incredible opportunities for companies demonstrating the ability to reduce risk in the development process. These may be groups iterating rapidly through cell therapy candidate evaluation or viral vector development. They may be groups with automated and closed manufacturing processes driving down the cost of therapy, Regardless, this is a window of opportunity across all stages of the value chain.
Future of the Immuno-Revolution
How will we measure the future progress of such a revolution? The most obvious (and most distant) markers may be improved rates of therapeutic success in fields such as immuno-oncology, the coronation of (immune) cell therapies as another pillar of therapeutics, or the clinical paradigm-shifting early detection of cancer. Initially though, evidence may come in the form of consortiums sharing immune biology data and expertise, and the continued emergence of ambitious and well-capitalized Immuno-Tech startups tackling BHAGs (Big Hairy Audacious Goals).
Progress in biological science often feels akin to small streams slowly trickling together to form a river. Immune biology is not an exception. But here we have been damming streams of knowledge for too long, and they have been unable to connect. These dams are now being removed, by the amassed data, computational capabilities, and tools to read, map, and modulate the immune system.
The needs of our healthcare system, from diagnostics to therapies addressing the most draining diseases of our generation, provide the additional wave that will just perhaps get the Immuno-Revolution flowing. Perhaps then we can push through, break down, and resolve complexity exemplified.
