Ting-En Huang
School of Medicine, College of Medicine, Taipei Medical University, Taipei, Taiwan

Introduction

Q1 2026 marked a slower period for regulatory approvals across gene, cell, and RNA therapies, alongside a growing emphasis on evidence-based development and de-risking. As of 2026 Q1, 40 gene therapies, 37 RNA therapies, and 75 non-genetically modified cell therapies had received approval worldwide. For gene therapies and RNA therapies, no new approvals in Q1. As for non-genetically modified cell therapies, there were two new approvals in Japan: Sumitomo’s raguneprocel for Parkinson’s disease and Cuorips’ ReHeart for heart failure. [1]

Across the cell, gene, and RNA therapy pipeline, 2132 gene therapies were in development, accounting for 50% of the total pipeline. CAR-T remained the leading technology, accounting for 52% of genetically modified cell therapies. Oncology and rare diseases remained major areas of gene therapy development. As for common indications, acute lymphocytic leukemia, systemic lupus erythematosus, and brain cancer are the top three among therapies with specified indications.[1]

CRISPR (two 2026 clinical trials)

Against this backdrop, advances in liver-directed gene editing in 2026 illustrate the broader shift toward evidence-driven development and de-risking. The following two clinical trials published in 2026 used lipid nanoparticles (LNPs) for the systemic delivery of CRISPR-based gene-editing systems to the liver.

One representative example is the phase 1 trial of YOLT-101, an in vivo adenine base-editing investigational therapy for heterozygous familial hypercholesterolemia (HeFH). This trial aimed to develop a potentially one-time treatment that could reduce the need for repeated administration of therapies such as inclisiran, a PCSK9-targeting siRNA therapy which silences expression of PCSK9 in hepatocytes.

Via GalNAc-modified LNPs, high-precision adenine base editor (hpABE5) mRNA and sgRNA were delivered to hepatocytes. By doing so, the base-editing system targets PCSK9, disrupts normal PCSK9 mRNA splicing, and thereby reduces circulating PCSK9 and LDL-C levels. [2] Six participants received escalating doses (0.2, 0.4, 0.6 mg/kg) of YOLT-101. Five of the six participants experienced treatment-related adverse events (AEs), all of which were mild to moderate. Fever, myalgia, vomiting, and other infusion-related reactions were the most common AEs (83.3%); alanine aminotransferase (ALT) or aspartate aminotransferase (AST) elevations occurred in 50% of participants, and no grade 3 or higher-level AEs were reported. At week 24, the 0.6mg/kg group had a mean 74.4% reduction in PCSK9 level and a mean 52.3% reduction in LDL-C. For context, a previous study of inclisiran reported a 60.7% PCSK9 reduction and a 38.1% LDL-C reduction. In addition to these early efficacy signals, the study included preclinical off-target evaluations and a planned long-term follow-up, highlighting the integration of therapeutic innovation with systematic risk evaluation in gene-editing trials. [2]

A second example is a phase 3, double-blind, randomized, placebo-controlled trial published in The New England Journal of Medicine. Lonvoguran Ziclumeran (lonvo-z) is an investigational, single-dose, in vivo gene-editing therapy for hereditary angioedema. LNPs deliver Streptococcus pyogenes Cas9 mRNA and an sgRNA targeting KLKB1. After translation, Cas9 protein forms a complex with the sgRNA and is directed to the KLKB1 gene, which encodes prekallikrein, to inactivate it. Once KLKB1 is inactivated, prekallikrein and plasma kallikrein levels decrease, further reducing excessive bradykinin production and vascular permeability. [3]

A total of 80 patients from multiple countries were randomly assigned to lonvo-z (50mg) or placebo group at a 2:1 ratio and stratified by baseline attack rates of ≤3 or >3 attacks per month. Efficacy was evaluated by the monthly HAE attack rate from weeks 5 to 28 after infusion. The mean monthly attack rate was 0.26 attacks/month in the lonvo-z group compared with 2.10 attacks/month in the placebo group, corresponding to an 87% relative reduction (P<0.001). Long-term follow-up remains important to evaluate the risk of off-target editing. However, compared with the YOLT-101 phase 1 clinical trial, this randomized phase 3 trial provides higher-level clinical evidence that liver-directed in vivo gene editing has progressed to late-stage clinical validation. Together, these two trials demonstrate different stages of evidence generation in current gene-editing development. [3]

AAV (two 2026 clinical trials)

Alongside CRISPR-based approaches, AAV gene therapy remains essential to clinical development, particularly for long-term gene replacement. One example is the phase 1 clinical trial of NGGT006. Like YOLT-101, which edits the PCSK9 gene to treat heterozygous familial hypercholesterolemia (HeFH), NGGT006 is also for familial hypercholesterolemia. However, NGGT006 is for homozygous familial hypercholesterolemia (HoFH), and it uses a recombinant AAV serotype 8-based vector to carry codon-optimized LDLR complementary DNA. To promote liver-specific expression, a human alpha 1-antitrypsin (hAAT) promoter was inserted. Restoring LDLR expression enhances hepatic uptake of LDL particles and lowers circulating LDL-C. [4]

In HoFH hamsters, a single dose of NGGT006 was injected at either a high (1.25×1013 vg/kg) or a low (4.17×1012 vg/kg) dose. At 28 weeks, LDL-C levels were reduced by 72% and 62% in the high- and low-dose groups, respectively. In rhesus monkeys, NGGT006 was also injected to evaluate the toxicology and immunogenicity. Both high (5×1013 vg/kg) and low (1.5×1013 vg/kg) dose groups achieved sustained liver transduction and dose-dependent human LDLR expression during the 10-week follow-up, with only transient ALT elevations and no adverse clinical observations or major organ abnormalities. [4]

Three patients received escalated doses (7.5×1012, 1.5×1013, or 3×1013 vg/kg) of NGGT006. Among them, the high-dose participant showed a marked decrease in LDL-C level: from 11.17 to 0.26 mmol/L and maintained a >90% reduction after week 6. The medium-dose participant showed a contrasting response: although there was a reduction at the beginning, the LDL-C level was 1.61 mmol/L higher than baseline after 52 weeks. The low-dose participant showed a modest response: the LDL-C level was 1.2 mmol/L lower than baseline after 52 weeks. The responses varied across participants, highlighting the uncertainty of dose–response relationships in this three-participant first-in-human study. All three patients had increased ALT/AST levels but no severe adverse event.

Although the relationship between treatment-induced anti-AAV8 neutralizing antibodies, adverse events, and therapeutic efficacy remained unclear in this small cohort, immune responses remain a major translational challenge because of the possibility of affecting the efficacy of therapy. Furthermore, patients with pre-existing AAV8 neutralizing antibodies were excluded unless antibody levels could be reduced to negative, which means the treatment eligibility may be limited. [4]

Another 2026 study, the phase 1–2 clinical trial of AAV9-GLB1, further illustrates both the therapeutic potential and immune challenges in AAV therapy. Type II GM1 gangliosidosis is caused by biallelic variants in GLB1, resulting in lysosomal β-galactosidase deficiency and abnormal GM1 ganglioside degradation. Impaired degradation of GM1 ganglioside leads to lysosomal substrate accumulation and progressive injury to the central nervous system. AAV9-GLB1 delivers a functional GLB1 gene to restore β-galactosidase activity and promote GM1 ganglioside degradation. [5]

During the trial, a single dose of AAV9-GLB1 was injected into twelve children with type II GM1 gangliosidosis through IV. The reported analysis included nine participants with at least two years of follow-up. Five participants received 1.5×10¹³ vg/kg and four received 4.5×10¹³ vg/kg. Over up to three years of follow-up, CSF β-galactosidase activity increased and GM1 ganglioside levels decreased. Functional outcomes were mixed: expressive communication and gross motor scores appeared stable, whereas fine motor and receptive communication scores declined. Most adverse events were mild to moderate, but liver-enzyme elevations occurred in all participants and one serious event of vomiting requiring hospitalization was attributed to the gene therapy. However, AAV9 activated anti-vector immune reactions in all participants and potentially limited the future readministration of the same vector. Different from NGGT006, which restores LDLR expression in hepatocytes, AAV9-GLB1 restores β-galactosidase activity and reduces GM1 ganglioside accumulation by delivering a functional GLB1 gene. [5]

Taken together, while immune responses, variable clinical efficacy, and patient eligibility remain major barriers to broader clinical application, these two studies show that AAV-based gene therapy continues to expand across distinct disease areas, from inherited metabolic disorders to progressive neurological diseases. [4,5]

Conclusion

Despite its therapeutic potential, gene therapy remains associated with considerable clinical uncertainty. Both efficacy and safety require careful consideration. The 2026 CIFFREO trial provides an important negative example. Fordadistrogene movaparvovec is an investigational rAAV9-based gene therapy encoding a mini-dystrophin transgene protein for Duchenne muscular dystrophy. However, further clinical development of the therapy was discontinued because the trial failed to meet the primary efficacy endpoint and the overall benefit–risk profile was negative based on both efficacy and safety data. The outcome emphasized that a compelling biological rationale doesn’t represent the success of translating into a meaningful clinical outcome. [6]

Taken together, clinical trials published in 2026 suggest gene therapy is transitioning from expansion to more evidence-driven and risk-aware development. Therefore, clinical validation, long-term safety monitoring, and systematic de-risking will be critical to future advancement.

References

• Quarterly Landscape Report. (2026). American Society of Gene & Cell Therapy. https://www.asgct.org/news-publications/landscape-report
• Wan, P., Tang, S., Ling, D., Lu, yuming , Long, mei , Xiao, L., Jiang, Y., Liao, J., Ma, X., Liu, Y., Yu, wensu , Ott, M., Wang, Z. J., Wu, Y., Yang, taihua , & Xia , qiang . (2026). In Vivo Base Editing Gene Therapy for Heterozygous Familial Hypercholesterolemia: A Phase 1 Trial. Nature Medicine, 32, 1045–1051. https://www.nature.com/articles/s41591-026-04254-4?utm#citeas
• Cohn., Gurugam, P., Longhurst, H. J., Pürsün, E. A., Craig, T. J., Farkas, H., Jacobs, J., Lumry, W. R., Magerl, M., Peter, J., Riedl, Maag, D., Golden, A., Shah, M. Y., Sutherland, A., Miller, … Haelo investigators. (2026). Lonvoguran Ziclumeran — In Vivo CRISPR Gene Editing in Hereditary Angioedema. The New England Journal of Medicine. https://www.nejm.org/doi/abs/10.1056/NEJMoa2600931
• Zheng, T., Gao, G., Xu, C., Li, R., Shao, J., Wang, Xiaoke, Han, X., Li, J., Liu, P., Xiong, Y., Song, Lailing, Wang, J., Wen, Shengmei, Zhuo, Xiaozhen, Wang, Gang, Jiang, L., Qu, G., Chen, T., She, Jianqing, … Wu, Yue. (2026). AAV Gene Therapy for Homozygous Familial Hypercholesterolemia: A Phase 1 Trial. Nature Medicine, 32, 2137–2146. https://www.nature.com/articles/s41591-026-04441-3?utm#citeas
• Lewis, C. J., Johnston, J. M., S., M., Acosta, M. T., Farmer, C., Baker, E. H., Crowell, A., Mojica, Y., Joseph, L., Vézina, G., Quezado, Z., Yousef, M. H., Vardar, Z., Shazeeb, Corti, M., Blackwood, M., … Tifft, C. J. (2026). AAV9 Gene Therapy in Type II GM1 Gangliosidosis — A Phase 1–2 Trial. The New England Journal of Medicine, 394, 1184–1194. https://www.nejm.org/doi/full/10.1056/NEJMoa2510935?utm
• Muntoni, F., Nascimento, A., Shin, Jinhong, Guglieri, M., Stettner, Georg M., Veerapandiyan, Aravindhan, Gallo, S., Shi, H., Gundapaneni, B., Neelakantan, S., Lobello, K., Shen, Qi, Levy, Daniel I, Mercuri, E., & Ciffreo study group. (2026). Safety and Efficacy of Fordadistrogene Movaparvovec in Ambulatory Participants with Duchenne Muscular Dystrophy (CIFFREO): A Phase 3, Double-Blind, Randomised, Placebo-Controlled Study. THE LANCET Neurology. https://www.thelancet.com/journals/laneur/article/PIIS1474-4422%2826%2900036-0/abstract?utm