Chimeric antigen receptor (CAR) T cell therapy is in the clinic, performing impressively against blood cancers such as leukemia. It is also the subject of many clinical trials, which are assessing its efficacy against cancers such as melanoma and solid tumors. Its therapeutic promise has also motivated biotech start-ups, large pharmaceutical companies, and governments to devote significant investments to the cell therapy space.1

Despite the experimental and clinical successes of CAR T-cell therapy, a major technical limitation is the way in which gene cargo is delivered into T cells for genetic transformation. The most popular way is using viruses such as adeno-associated viruses, which encapsulate gene cargo in viral capsids before infecting cells and delivering genes into the nuclei.

“Viruses are popular among clinicians who have been working with viral vectors for decades and are most familiar with this method,” says Alexander Marson, MD, PhD, an associate professor at the University of California, San Francisco, who has been pursuing the use of ribonucleoprotein (RNP) electroporation2 and nonviral DNA templates for genome editing3 in immune cells. “A drawback of virus is that its production is labor intensive and its clinical applications are expensive because each step of clinical-grade viral vector manufacturing must be compliant with good manufacturing practices (GMP),” he adds.

Another disadvantage of viral vectors is lingering uncertainty over their safety profile. Although progress has been made to reduce the toxicity of viruses by directed evolution and engineering,4 most viruses that infect cells and deliver genes will inevitably integrate their own genetic elements into the host genome. These elements can pose long-term safety risks.

“Virus-mediated transgene integration is not site specific. This could theoretically lead to oncogenic insertion,5 but there has been no evidence of that occurring in CAR T-cell therapy to date,” says Yvonne Chen, PhD, an associate professor at the University of California, Los Angeles, and a member researcher of the Parker Institute for Cancer Immunotherapy.

Consequently, the high manufacturing costs and safety concerns of viral vectors have motivated the cell therapy field to search for nonviral gene delivery alternatives using microtechnology and nanotechnology.

Cost-efficient nanoparticle gene delivery

Nanoparticles can be manufactured cheaply and in bulk, and their use may bring down the prices of CAR T-cell therapy to make this promising treatment accessible to more patients. There are also well-established scientific protocols to synthesize and functionalize nanoparticles for gene delivery applications.

A team of scientists led by Matthias T. Stephan, MD, PhD, at the Fred Hutchinson Cancer Research Center, generated CAR T cells from circulating lymphocytes in situ using biodegradable poly(β-amino ester)-based nanoparticles.6 To ensure that these gene carriers were taken up by T cells, the investigators coupled the nanoparticles’ polymer surfaces with T-cell-targeting anti-CD3e fragments. To enhance transport of gene cargo to the T cells’ nuclei, the investigators functionalized the polymer with peptides containing microtubule-associated sequences and nuclear localization signals.

To program their nanoparticles for treating leukemia, which is characterized by the presence of B cells with high levels of CD19 surface antigens, the investigators made use of DNA plasmid encoding for CAR T-cell receptor (TCR) with specificity toward CD19+ cells. The investigators found that T cells rapidly endocytosed the polymeric nanoparticles. Also, as a result of somatic integration, the T cells expressed high and stable levels of the CAR transgene for at least 2 weeks. When the nanoparticles were injected into the bloodstream, 34% of the T cells were labeled with the nanoparticles. This intervention increased survival rate of 58 days on average in mouse model of leukemia.

Despite their elaborate functionalization, Stephan and colleagues encountered off-target effects. Circulating cells, including neutrophils and monocytes, endocytosed the nanoparticles. This is a huge safety concern as nontarget cells that are capable of proliferating also became genetically engineered.

In fact, Carl H. June, MD, Simon F. Lacey, PhD, J. Joseph Melenhorst, PhD, and colleagues at the University of Pennsylvania reported that leukemia recurred in a patient after unintentional introduction of a CAR gene into a single leukemia B cell, highlighting the danger of off-target effects.7

The team at Fred Hutchinson Cancer Research Center also reported that a significant percentage (~20%) of the injected nanoparticles remained in the blood or were cleared in the liver. Nanoparticles that miss their mark, the researchers warned, can cause tissue and systemic toxicity.

Another limitation of nanoparticle-based transfection method is that cell population is inherently heterogenous. Different numbers of nanoparticles may be internalized by different cells, depending on the cells’ variable expressions of endocytic receptors. CAR T-cells that internalize large numbers of nanoparticles may suffer from cytotoxicity and suboptimal biological functions. This topic deserves greater study before nanoparticles can become part of clinical CAR T-cell engineering.

Minimally perturbative gene delivery with nanostructures

Owing to advances in nanofabrication techniques such as atomic layer deposition and plasma etching, high-aspect-ratio nanostructures have been developed for nonviral gene delivery to immune cells. These nanostructures include hollow nanochannels.

In general, nanostructures provide lower cytotoxicity than viruses or nanoparticles because nanostructures deliver only the gene cargo needed. No exogenous viral proteins or chemicals are passed. Furthermore, nanostructures may be a universal delivery tool for various types of immune cells because gene delivery via nanostructures does not depend on biologically variable properties such as membrane receptor density or lipid composition/charge. Instead, gene delivery via nanostructures depends on physical properties such as electric fields and mechanical forces. Consequently, deliver mechanisms can be engineered and optimized to provide consistent performance.

At Harvard University, researchers led by Hongkun Park, PhD, demonstrated the use of biomolecule-functionalized nanowires to transfect diverse types of immune cells.8 Nanowire-mediated delivery, the researchers suggested, could outperform viral delivery or conventional electroporation. To improve intracellular access during transfection, researchers at Ohio State University led by Wu Lu, PhD, and L. James Lee, PhD, used dielectrophoresis-assisted 3D nanoelectroporation,9 a means of coupling electrical control and nanostructures, to achieve close to 70% transfection efficiency.

Nanostructure transfection may also better preserve the long-term biological functions of CAR T cells. “It is important that the method of transgene delivery does not alter the functional fitness and long-term activity of T cells to ensure therapeutic efficacy,” says Chen. “This is more than simply maintaining high viability during ex vivo culture.”

Recently, two Stanford University researchers—this article’s author and Nicholas Melosh, PhD—demonstrated that nanostructure-localized electroporation can deliver genes more efficiently and preserve critical biological attributes (such as proliferation and gene expression of immune cells) better than viruses, bulk electroporation, and chemical polymers.10

Nanostructures, however, should come with a caveat. They require a manufacturing process more complex than the one for nanoparticles. Nonetheless, nanostructures can comply with GMP more easily.

Recently, Peidong Yang, PhD, and other researchers at the University of California, Berkeley, introduced a technique that takes advantage of localized cell interactions while avoiding laborious nanofabrication. The technique incorporates nanopore electroporation and a water-filter nanoporous membrane.11 The idea, which was demonstrated in adherent and suspension cells, is to induce localized electroporation on a nanosized area of the cell membrane and thereby preserve cell viability. However, the lack of control over membrane pore properties such as pore density and diameters may lead to significant batch-to-batch variability in gene delivery efficiency.

Another weakness of nanostructure is their low throughput. This is because most of the equipment used for nanofabrication is optimized for silicon wafers with diameters between 6 and 12 inches. This limits the scalability of nanostructure fabrication. As CAR T-cell therapy requires many cells (107 to 109 for adult patients), innovations are necessary to enhance the throughput of nanostructure-based gene delivery before this platform can be integrated into commercial cell manufacturing.

Science concept; Experimental scientists in chemical laboratory. Development and scientific research in laboratory
Credit: Anchalee Phanmaha/Getty Images

High-throughput microfluidics platforms

Like nanostructure technology, microfluidics technology can be integrated with physical forces or fields to create pores on immune cells for gene delivery.12 One distinct advantage of microfluidics technology is the availability of continuous-flow and parallelized operation. This advantage allows microfluidics technology to offer much higher throughput than other nonviral gene delivery methods.

At the University of California, Los Angeles, Paul S. Weiss, PhD, Steven J. Jonas, MD, PhD, and colleagues recently developed an acoustofluidic sonoporation system that makes use of ultrasound waves to physically shear the membrane of immune cells.13 Using optimized parameters, the team showed that primary peripheral blood mononuclear cells can be transfected at a rate of 200,000 cells/min, with the cells retaining green fluorescent protein expression over 72 hours, demonstrating probable stable gene integration due to ultrasound-induced nuclear membrane rupture.

A similar system has been described by scientists from SQZ Biotechnologies and the Roche Innovation Center. The scientists, including SQZ’s director of application development, Tia DiTommaso, PhD, and SQZ’s CEO, Armon Sharei, PhD, indicated that the system, a microfluidic squeezing platform, can disrupt the membrane of immune T cells to knock out the PD-1 gene, which suppresses T-cell functions when in contact with PDL-1 molecules expressed on cancer cells.14 The SQZ-Roche team reported that compared to bulk electroporation, microfluidic squeezing better preserved the functionalities of T cells. In comparison to bulk electroporation, cells that were treated with microfluidic squeezing also suffered from fewer gene expression changes and negligible changes in cytokine production. Furthermore, squeeze-transfected T cells demonstrated better in vivo tumor-killing efficacy.

Nevertheless, to achieve high-throughput transfection, microfluidic channels must never be clogged, or the entire transfection operation has to pause.15 However, this can be challenging to achieve practically because proteins from cell media may nonspecifically adsorb onto microchannels, causing cells to stick and aggregate. Furthermore, for gene cargo to enter cells during transient cell membrane rupture, the cargo must be available throughout the entire microchannel at relatively high concentrations, which can significantly raise the costs of biological materials such as DNA plasmids.

Moving forward

The field of nonviral alternative for CAR T-cell transformation is growing rapidly with increasing innovation and investments. Ideally, researchers will be able to capitalize on the respective strengths of nanoparticles, nanostructures, and microfluidics technology to create a cost-efficient, minimally perturbative, and high-throughput technique to engineer CAR T cells for clinical applications.

As we learn more about immuno-oncology, the functions of CAR T cells are also expected to become more complex: multiple CAR transgenes to target multiple cancer antigens, gene for faster growth to reduce manufacturing time and cost, and programmable genetic switches for on-demand activation and inhibition to minimize toxicity.16 If these possibilities are realized, nonviral gene delivery techniques may become even more useful for CAR T-cell therapy, particularly since nonviral delivery options, unlike viral delivery options, lack strict size limits for gene cargo.



  1. Palada B. Bright Outlook for the Global Cell and Gene Therapy Market. GEN 2020; 40(5, suppl.): 6–11.
  2. Roth TL, Puig-Saus C, Yu R, et al. Reprogramming human T cell function and specificity with non-viral genome targeting. Nature 2018; 559: 405–409.
  3. Nguyen DN, Roth TL, Li PJ, et al. Polymer-stabilized Cas9 nanoparticles and modified repair templates increase genome editing efficiency. Nature 2020; 8: 44–49.
  4. LeMieux J. Going Viral: The Next Generation of AAV Vectors. GEN 39(9): 22–27.
  5. Kaiser J. Virus used in gene therapies may pose cancer risk, dog study hints. Science 2020; DOI: 10.1126/science.aba7696.
  6. Smith TT, Stephan SB, Moffett HF, et al. In situ programming of leukaemia-specific T cells using synthetic DNA nanocarriers. Nature Nanotech. 2017; 12(8): 813–820.
  7. Ruella M, Xu J, Barrett DM, et al. Induction of resistance to chimeric antigen receptor T cell therapy by transduction of a single leukemic B cell. Nature Med. 2018; 24:1499–1503.
  8. Shalek AK, Gaublomme JT, Wang L, et al. Nanowire-Mediated Delivery Enables Functional Interrogation of Primary Immune Cells: Application to the Analysis of Chronic Lymphocytic Leukemia. Nano Lett. 2012: 12(12): 6498–6504.
  9. Chang L, Gallego-Perez D, Zhao X, et al. Dielectrophoresis-assisted 3D Nanoelectroporation for Non-Viral Cell Transfection in Adoptive Immunotherapy. 2015: 15(15): 3147–3153.
  10. Tay A, Melosh N. Transfection with Nanostructure Electro‐Injection is Minimally Perturbative. Adv. Ther. 2019; 2(12): 1900133.
  11. Cao Y, Ma E, Cestellos-Blanco S, et al. Nontoxic nanopore electroporation for effective intracellular delivery of biological macromolecules. Proc. Natl. Acad. Sci. USA 2019; 116(16): 7899–7904.
  12. Ding X, Stewart MP, Sharei A, et al. High-throughput nuclear delivery and rapid expression of DNA via mechanical and electrical cell-membrane disruption. Nature Biomed. Eng. 2017; 1: 0039.
  13. Belling JN, Heidenreich LK, Tian Z, et al. Acoustofluidic sonoporation for gene delivery to human hematopoietic stem and progenitor cells. Proc. Natl. Acad. Sci. USA 2020; 117(20): 10976–10982.
  14. DiTommaso T, Cole JM, Cassereau L, et al. Cell engineering with microfluidic squeezing preserves functionality of primary immune cells in vivo. Proc. Natl. Acad. Sci. USA 2018; 115(46): E10907–E10914.
  15. Dressaire E, Sauret A. Clogging of microfluidic systems. Soft Matter 2017;13: 37–48.
  16. Guedan S, Calderon H, Posey AD, Maus MV. Engineering and Design of Chimeric Antigen Receptors. Mol. Ther. Methods Clin. Dev. 2018;12: 145–156.
Previous articleCould Treating Hypertension Without Side Effects be as Simple as Eating a Spoonful of Rice?
Next articleCancer’s Extreme Genomic Complexity Explained