How Targeted Nanoparticles And Circular RNA Are Advancing CAR-T Development

A conversation with Michael Lam Ph.D., vice president, strategy & scientific external affairs, Sail Biomedicines

The field of RNA therapeutics has grown rapidly in recent years. Despite several notable successes, limitations still remain. In an effort to overcome the short duration of therapeutic protein expression associated with mRNA, Sail Biomedicine has developed novel circular, or endless, RNA constructs that can be delivered using targeted nanoparticles. Together, these Endless RNA (eRNA) molecules and the targeted delivery approach expand the potential of RNA-based therapeutics.

In this Q&A, Life Science Connect’s Izzy Dininny sits down with Michael Lam from Sail Biomedicines to discuss their recent developments, delivery mechanisms, and a future outlook for broadened application of these therapeutics.

The therapy your company is developing combines circular eRNA with targeted nanoparticles (TNPs). Why eRNA? What limitations of conventional mRNA does the platform overcome?

Conventional mRNA has changed what is possible for medicine, but its design also creates important limitations for therapeutic applications. Linear mRNA has open ends that make it susceptible to exonuclease degradation, which can limit the duration of protein expression. For vaccines, a short burst of expression may be sufficient. For many therapeutic applications, including immune cell programming, the timing, level, and duration of protein expression can directly shape the pharmacology of the medicine.

Sail’s Endless RNA (eRNA) constructs are circular, translatable RNA designed to support more durable expression. Because eRNA does not have free 5’ or 3’ ends, it is less vulnerable to the degradation pathways that can limit conventional mRNA. This enables an extended expression window while preserving the transient nature of RNA-based medicines.

That distinction is important. Our goal is not permanent expression. In the context of in vivo CAR-T, we want enough CAR expression to program T cells, drive meaningful B-cell depletion, and support an immune reset, but we also want that activity to resolve. eRNA gives us a way to extend expression long enough to support the desired pharmacology, without the permanent genomic integration associated with viral approaches.

TNPs are capable of selectively reaching CD4 and CD8 T cells in vivo. Can you discuss why you chose nanoparticle delivery over viral delivery? What are the benefits and limitations of using TNPs?

For autoimmune diseases, we believe the product profile matters as much as the biological target. Autologous lentiviral CAR-T has generated compelling clinical data in autoimmune disease, but it is complex, highly individualized, and requires specialized infrastructure, cell collection, ex vivo manipulation, manufacturing, lymphodepleting conditioning, and careful monitoring. Viral delivery also raises important questions when applied outside oncology, particularly because autoimmune disease patients may be younger, have longer expected treatment horizons, and may not have the same risk-benefit profile as patients with advanced cancer.

We chose targeted nanoparticle delivery because it offers a non-integrating, transient, and systemically administered approach to immune cell programming. With TNPs, the RNA payload can be delivered directly in vivo to the patient’s own immune cells. The nanoparticle can be engineered for cell selectivity, the RNA payload can be engineered for the desired expression profile, and the overall product can be designed around dose efficiency and therapeutic index.

In Sail’s In Vivo eRNA-TNP CAR-T program, the TNP is designed to selectively reach both CD4 and CD8 T cells. That is important because both populations can contribute to the depth of B-cell depletion needed for immune reset. Our preclinical data have shown efficient and selective delivery to CD4 and CD8 T cells, including in blood and lymphoid tissues, supporting the potential for a pharmacologic approach that reaches the relevant immune compartments.

The limitations are also real. TNPs require careful optimization of multiple components, including particle composition, targeting ligand, conjugation chemistry, endosomal escape, biodistribution, tolerability, and payload expression. Delivery is not simply a vehicle problem. It is a systems design problem. The benefit is that, if the system is engineered correctly, TNPs can provide a level of control that is difficult to achieve with permanent or genome-integrating approaches.

In Sail’s eRNA-TNP CAR-T program, the goal is to achieve B-cell depletion in the right places for the right amount of time to achieve a full immune reset in autoimmune diseases. Can you elaborate on how the eRNA-TNP succeeds at specific and transient therapeutic integration?

In autoimmune diseases driven by pathogenic B-cell biology, the therapeutic goal is not simply to reduce circulating B cells. The deeper question is whether a therapy can reach the compartments where disease-relevant B cells reside, deplete them sufficiently, and allow the B-cell compartment to repopulate in a healthier state.

Sail’s In Vivo eRNA-TNP CAR-T program is designed around that biology. The TNP component is engineered to deliver the eRNA payload selectively to T cells, including both CD4 and CD8 T-cell subsets. The eRNA payload encodes an anti-CD19 CAR, enabling those T cells to transiently express the CAR and mediate B-cell depletion. Because the payload is RNA, this programming is non-integrating and temporary.

That transient activity is a key part of the intended therapeutic profile. We are not trying to create permanently modified T cells. We are trying to create a short, controlled window of CAR expression that is sufficient to drive deep B-cell depletion across relevant compartments, including blood, lymphoid tissues, and bone marrow progenitor populations.

In preclinical models, Sail has shown deep depletion of B cells in circulation and tissues, including lymph nodes and bone marrow. In a B-cell repopulating humanized mouse model, repopulating B cells showed an immature phenotype, which is consistent with the type of biological reset we believe may be needed to produce durable benefit in autoimmune disease. These findings support our view that the integration of eRNA pharmacology and targeted nanoparticle delivery can create a differentiated product profile: transient exposure, deep tissue pharmacology, and dose-efficient immune cell programming.

How is AI influencing the design of both your eRNA constructs and TNP delivery vehicles?

AI is important because the design space for RNA medicines is too large and too interconnected to optimize through intuition alone. In our platform, both the eRNA payload and the targeted nanoparticle vehicle contain many variables that can influence performance. For eRNA, those variables may include sequence architecture, translation elements, stability, expression level, and duration of expression. For nanoparticles, they include lipid composition, particle properties, targeting ligands, biodistribution, endosomal escape, and tolerability.

Sail uses large-scale experimental data generation together with AI and machine learning to move from empirical screening toward systematic engineering. For eRNA, this allows us to design and evaluate large libraries of sequence variants and identify features associated with stronger, more durable expression. For TNPs, AI helps us learn from in vivo and in vitro data sets to better predict how changes in composition and targeting affect delivery performance.

The larger point is that eRNA and delivery cannot be optimized independently. A payload that performs well in one cell type or tissue may not be optimal in another. A nanoparticle that reaches the right cell must also release its payload efficiently and support the right level and duration of expression. AI helps us connect those variables and design toward the desired biological outcome, rather than optimizing isolated components.

Beyond autoimmune diseases, where do you see the greatest opportunity for programmable eRNA delivery platforms?

Autoimmune disease is an important first application because the biology is well suited to a transient, dose-efficient immune cell programming approach. However, the broader opportunity extends to any therapeutic area where a protein, antibody, receptor, or multi-component biologic could be encoded and expressed in the body for a defined period of time.

We see significant potential across immunology, oncology, rare diseases, metabolic disease, and respiratory disease. In each case, the value comes from matching the expression profile and delivery strategy to the biology of the disease. Some applications may benefit from longer protein expression. Others may require tissue-selective delivery, redosing, lower peak exposure, or expression in specific immune cell populations.

In oncology, for example, transient immune cell programming could potentially be used to direct immune activity without permanently modifying the genome. In rare or metabolic diseases, eRNA could be used to express therapeutic proteins with more durable exposure than conventional mRNA. In respiratory disease, tissue-selective delivery and controlled expression may open opportunities that have been difficult for earlier RNA approaches.

The common theme is control. eRNA allows us to tune expression, and targeted nanoparticles allow us to tune delivery. Together, they create a platform that can be adapted to many therapeutic contexts where the right protein needs to be expressed in the right cell, at the right level, and for the right amount of time.

What hurdles remain before in vivo immune cell programming can become a broadly applicable therapeutic modality?

The field has made meaningful progress, but several hurdles remain. First, delivery must continue to improve. In vivo immune cell programming requires selective delivery to the right cell populations, efficient uptake, endosomal escape, and limited off-target expression. Achieving that consistently across species and, ultimately, across patients is essential.

Second, the field needs to define the right pharmacology for each disease. In autoimmune disease, for example, the relevant question is not simply whether a CAR can be expressed in vivo. It is whether that expression can drive the depth, tissue reach, and duration of B-cell depletion needed for immune reset, while maintaining an appropriate safety profile.

Third, transient programming requires precise control of expression. Too little expression may not produce the intended effect. Too much, or expression in the wrong cells, could create tolerability issues. That is why payload design, delivery design, and dosing strategy need to be developed together.

Finally, translation into the clinic will require careful attention to manufacturing, scalability, safety monitoring, regulatory expectations, and patient selection. The promise of in vivo immune cell programming is that it could make powerful cell therapy-like biology more accessible through a drug-like product profile. To realize that promise, the field must show that these therapies can be delivered safely, reproducibly, and with the depth of biological effect needed to change the course of disease.

About The Expert

Michael Lam, Ph.D., is a strategic and scientific leader who shapes Sail’s R&D narrative, partnership strategy, and external positioning, bridging science and business to drive innovation and value creation across the company’s genetic medicine programs. Mike brings over 16 years of experience across CRISPR-Cas9 and base editing, RNA therapeutics, small molecules, and cell therapy modalities. Before Sail, he held senior leadership roles at Intellia Therapeutics, where he led the ex vivo cell therapy portfolio and delivered the company’s first allogeneic CAR-T development candidate. He earned his Ph.D. in cell biology from Baylor College of Medicine and completed postdoctoral training at the Whitehead Institute for Biomedical Research (MIT) as an American Cancer Society Fellow.