On 25 March, detailed Phase I trial data was published for ESO-T01, AstraZeneca’s in vivo CAR-T therapy, in patients with multiple myeloma (MM) who have been treated with at least three previous lines of therapy. This highly anticipated data is some of the first published regarding the use of in vivo CAR-Ts in a clinical setting, and covers a separate cohort of patients from that reported in The Lancet in July 2025. ESO-T01 became part of AstraZeneca’s portfolio as part of its acquisition of EsoBiotec, which involved an upfront payment of $425m with potential for $575m in milestone payments. The novel therapy comprises a lentiviral vector encoding an anti-B cell maturation antigen (BCMA) chimeric antigen receptor (CAR). The genetic payload integrates into the host’s genome, and the CAR is subsequently expressed using a synthetic T–cell–specific promoter. This therapy is designed to overcome some of the limitations of traditional ex vivo CAR-T therapies, such as long vein-to-vein times and adverse events related to lymphodepletion chemotherapy. Ex vivo CAR-Ts such as J&J’s Carvykti (ciltacabtagene autoleucel) have established themselves in the late-line MM setting, and Carvykti is currently in a Phase III trial in first-line MM.
The single-centre, single-arm study involved five patients who were all infused with the therapy within 24 hours of enrollment. All five patients experienced Grade 3 or above adverse events, many of which were cytopenias. There were four cases of any grade cytokine release syndrome, three of which were Grade 3. The first patient dosed experienced a hyperacute inflammatory reaction to the therapy, which resulted in a mid-trial protocol change to include prophylactic steroids, and this successfully mitigated CRS in subsequent patients and did not impair CAR-T proliferation. One patient experienced acute neurological deterioration that coincided with peak CAR-T expansion, and this patient died of a cardiac arrest 19 days after infusion; the death was attributed to pre-existing extramedullary disease by the study authors, and they suggest stricter anatomical risk stratification for future in vivo CAR-T studies. Despite concerning safety signals, efficacy looked promising: all four evaluable patients were negative for minimal residual disease (MRD) while three had stringent complete responses (sCR) and the fourth had a very good partial response (VGPr).
Go deeper with GlobalData
Access deeper industry intelligence
Experience unmatched clarity with a single platform that combines unique data, AI, and human expertise.
The study was stopped in March 2025 with no further patients enrolled, and with such a small number of patients with only a short follow-up, clear safety and efficacy trends can not be established; however, headline results are comparable to those seen in early-stage trials for Carvykti. The official reason cited for stopping the trial was the restructuring of the sponsor. One of the findings of the study that contrasts this therapy against traditional ex vivo CAR-Ts is the pattern of immune responses. In this trial, significant immune responses triggered an inflammatory surge within 24 hours of infusion, suggesting the viral vector was the cause itself, despite EsoBiotec engineering the vector with “immune-shielding.”
This will concern others who have acquired in vivo CAR-T biotechs with lentiviral vector platforms, such as Gilead, which acquired Interius BioTherapeutics in October 2025. It will be difficult to engineer viral vectors to provide effective treatment while reducing the significant safety effects seen in this study. What could be attempted would be to reduce the viral dose, but the dose used in this study is already low, and it may reduce the therapeutic effect if it is lowered further. Steroids should be used prophylactically in future studies of this modality, and the vectors should be engineered further to reduce the immune response. In contrast, AbbVie acquired Capstan Therapeutics with an mRNA-loaded lipid nanoparticle platform, which could sidestep some of these safety issues, but comes with potential effectiveness pitfalls. This is not the end for lentiviral-based in vivo CAR-T, but significant progress needs to be made on safety if these therapies are going to compete with the existing standards of care.
