A Field Guide To In Vivo Cell Therapy Companies

Written By Paul Rennert

Astra Zeneca’s purchase of Esobiotec with a $425M upfront has rekindled interest in CAR-T cell therapy technologies, and in particular, the in vivo cell therapy companies. Here we’ll take a hard look at the landscape and highlight several trends.

It seems wise to acknowledge the abysmal state of public and private financing for biotechnology, recently worsened by the administration’s egregious attacks on the nation’s scientific infrastructure and the turmoil caused by a reckless trade policy. That said, interest and opportunity in advanced therapeutic modalities for oncology – cell therapy, bispecific antibodies, antibody drug conjugates, vaccine strategies – remains high.

And a disclosure – I’ve served as Acting Chair of the Esobiotec Board of Directors over the last 4 years – service that will come to an end when Astra Zeneca closes on their 1 billion dollar acquisition. As such I can’t say anything about that deal that isn’t already public - but we can see this deal as an important benchmark for the field.

Let’s dive in.

Part 1 – A Technology Field Guide for In Vivo Cell Therapy

This is a complex scientific landscape featuring diverse methods of gene delivery, enhancements like immune modulation, an every-expanding array of indications, and complexity in the cell types targeted. Happily, the technological aspects have recently been reviewed in depth¹. CAR-T cell therapeutics that are created ex vivo – thus, genetically engineered outside of the patient, then reinfused - have been approved for use in treating blood cancers. In vivo CAR-T cell therapy requires that genetic material be delivered to the desired immune cells while those cells are still inside the patient. This means you need to choose the genetic material to be delivered, the immune cell type(s) you wish to target, and the method of delivery.

The most advanced method uses the same viral vector system – lentivirus – that is used for ex vivo cell therapy, as in the production of the CD19-targeted CAR-T cell therapeutic Yescarta² that is used to treat advanced B cell lymphomas. Lentivirus particles readily package genetic sequences for delivery into cells, particularly if the genetic construct is small, generally less than 5500 nucleotides (5.5 kB). That’s enough size to encode an expression construct containing a chimeric antigen receptor (CAR), perhaps along with an enhancement or 2, eg. an expressed cytokine or a secreted BiTE-like protein³. Since the lentiviral particles are themselves encoded by a 4 plasmid “packaging” system, additional features such as a targeting antibody domains can be expressed by one of the production plasmids, giving even more genetic “room” to modify the surface of the particle. Vive Biotech and other lentivirus particle CDMOs are rapidly advancing such technologies⁴˒⁵.

All of which is to say that with lentivirus particles you can package the genetic construct of interest and place cell-targeting antibody domains on the particle surface (more on this below). You have to add a virus-derived “cell fusion” protein that delivers the genetic payload to the bound cell, but that’s easy, as virus particles have evolved cell fusion as an infection strategy. One more useful feature of lentivirus is that once released into the targeted cell, the viral expression sequence integrates into the cell genome, leading to stable gene expression that is “inheritable” by cell division. That last bit sounds a scary but the risks associated with genomic integration are very small⁶.

Back to the “cell-targeting antibody domains” of the prior paragraph: most often used are the T cell targeting antibody domains that recognize CD3, that is expressed on all T cells, and CD8 that is expressed only on CD8-positive T cells. There are also programs directed to CD7 (expressed on T cells and their distant cousins, NK cells), CD16 expressed on myeloid lineage cells and so on. Of note, as you move away from CD3 and CD8 there are fewer examples in the preclinical and clinical scientific literature. Other viral methodologies use AAV “capsids”, a truncated particle that has been used in the gene editing field, and CMV1. There has been limited commercial development of these methods.

After lentiviral particles, the second common platform for in vivo CAR -T delivery uses lipid nanoparticle (LNP) technology. The most typical genetic cargo for LNP-mediated in vivo CAR T cell production is messenger RNA (mRNA)¹. Stability of mRNA is naturally limited therefore LNP encapsulation strategies typically seek to improve that stability, however, mRNA translation remains quite transient after LNP are injected, and this limits the persistence of expression of the genetically encoded CAR. Further, mRNA cannot be integrated into the genome, so expression diminishes rapidly within the rapidly dividing CAR T cell population. Current development efforts that focus on DNA delivery by LNP in vivo are advancing rapidly; large payloads can be accommodated which is very useful⁷. Finally, the relative simplicity of LNP manufacturing has attracted multiple companies into this space including for use with in vivo CAR T generation.

Part 2 – Companies, Technologies, Prospects 

Starting with the lentivirus delivery companies seems most reasonable as these are the furthest advanced. Many companies already have established relationships with large pharma (Table 1).

Table 1. Biotech companies using viral-based delivery for in vivo CAR T production

A notable trend here is the use of China and Australia clinical sites to initiate studies: both countries have friendly regulatory settings for Investigator-Initiated Trials and early phase clinical studies. Also notable is that financing support has been robust through 2024 (Interius) and into 2025 (Umoja).

Based on the early clinical result from Esobiotec⁸ we can view LV technology for CAR T cell therapy as derisked, and this should stimulate further investment and perhaps additional acquisitions. Pharma deals shown in Table 1 – AZ, Abbvie, Astellas, Novartis – illustrate the high level of interest. CAR-T-capable pharmas that have not done deals either have internal programs (Sanofi) or are likely pursuing deals within the in vivo biotech landscape. Such pharmas include JNJ, BMS, Gilead, Legend, Biontech, etc.

In contrast to the evident progress being made with LV technologies, the LNP delivery landscape is a bit behind (Table 2). We’re going to note right up front this week’s collapse of LNP manufacturing specialist Amplify Bio – suggesting a softening of interest in this space⁹. Whether the shrill anti-RNA stance taken by the new head of FDA will impact RNA-based therapeutics is unclear.

Table 2. Biotech companies using LNP-based delivery for in vivo CAR T production.

The most advanced program, from Carisma, had tested a Her2-targeting CAR-Macrophage in a Phase 1 clinical trial. Although the engineered macrophages trafficked to the Her2-positive tumors, no meaningful activity was observed¹⁰. The next clinical data should come from Myeloid Therapeutics who is testing a GPC3-targeting CAR-macrophage in a Phase 1 liver cancer study.

We’ll be watching this space as data appear during the upcoming conference season: AACR this month, ASGCT in May, ASCO and EHA in June, ESMO in October, SITC in November, ASH in December.

Stay tuned.

_____________________________________


Paul Rennert
Next

Aleta’s Novel T Cell Engagers