Cardiotropic Gene Therapy for Heart Failure: Translational Lessons From AB‐1002

In a recent study published in Nature Medicine, Henry et al. 1 reported the first-in-human trial of AB-1002, a cardiotropic adeno-associated viral (AAV) vector encoding constitutively active inhibitor-1 (I-1c) for the treatment of advanced non-ischemic heart failure. The study demonstrates that a single intracoronary administration of AB-1002 is feasible, well tolerated and provides early evidence that targeted gene therapy may offer a practical translational strategy for restoring cardiac function in patients with limited treatment options. Heart failure remains a pervasive global challenge, with current therapies largely slowing progression rather than restoring lost myocardial contractile function. The failing heart is characterized by impaired calcium handling, downregulated sarcoplasmic–endoplasmic reticulum calcium ions (Ca2+) ATPase (SERCA2a) pump activity, and maladaptive remodeling that perpetuates systolic dysfunction 2, 3. Earlier gene therapy trials, including the CUPID trial targeting SERCA2a, were hindered by limited vector tropism and immune barriers 2, 4. Yet advances in capsid engineering and cardiac-specific delivery have reignited the field 3, culminating in the current clinical translation. I-1c, the central effector gene in AB-1002, encodes a truncated and constitutively active form of the endogenous protein phosphatase 1 (PP1) inhibitor 3. In normal cardiomyocytes, β-adrenergic stimulation phosphorylates phospholamban (PLN) and activates the SERCA2a, enabling calcium reuptake during relaxation 5. In heart failure, overactive PP1 dephosphorylates PLN, suppressing SERCA2a and causing cytosolic calcium overload and contractile weakness 5. By maintaining PLN phosphorylation, I-1c restores SERCA2a activity, re-establishes Ca2+ homeostasis, and improves relaxation and contractility 3, 5. Preclinical work demonstrated that cardiac overexpression of I-1c using a re-engineered AAV2/8 vector improved ejection fraction and reduced left-ventricular dilation in a swine model of ischemic heart failure 3. These studies laid the foundation for the cardiotropic AAV2i8 vector used in AB-1002—a hybrid of AAV2 and AAV8 sequences designed to enhance myocardial transduction while minimizing hepatic exposure 3. In the first-in-human, open-label phase 1 trial 1, 11 patients with non-ischemic cardiomyopathy, New York Heart Association (NYHA) class III symptoms, and a baseline left ventricular ejection fraction (LVEF) of 15%–35% received a single intracoronary infusion of AB-1002, a chimeric cardiotropic AAV2i8 vector encoding constitutively active inhibitor-1. Two escalating doses were evaluated—3.25 × 1013 vg (cohort 1) and 1.08 × 1014 vg (cohort 2)—across three study centers in the United States. Participants were predominantly male (82%) with advanced, medically refractory non-ischemic cardiomyopathy; median age was 73.5 years in cohort 1 and 57 years in cohort 2. The infusion procedure was completed without major complications. During 12 months of follow-up, 83 treatment-emergent adverse events occurred in nine patients, with no treatment-related serious adverse events or predefined dose-limiting toxicities (myocarditis, myositis, neutropenia, or cardiac tamponade). The most frequent events were mild hypotension and increased blood creatinine (each in three patients). Transient, self-limiting alanine aminotransferase (ALT) and aspartate aminotransferase (AST) elevations (≤ 2× the upper limit of normal) occurred in four of five high-dose recipients and resolved without sequelae. One death (cardiomyopathy, 204 days post-infusion) was adjudicated as unrelated to treatment. No clinically relevant abnormalities were noted in laboratory, electrocardiographic, or vital-sign parameters. At 12 months, mean LVEF increased from 30.5% to 42.2% in cohort 1 and from 24.5% to 36.7% in cohort 2, corresponding to absolute gains of about 12 percentage points; all evaluable patients improved by ≥ 5 points. Seven of eleven patients improved by ≥ 1 NYHA class, while cohort 1 demonstrated an 18.8% increase in peak oxygen consumption (pVO2) and a 36% increase in 6-min walk test (6MWT). Quality-of-life metrics improved or remained stable (Minnesota Living with Heart Failure and Kansas City Cardiomyopathy questionnaires), and cardiac biomarkers (NT-proBNP, troponin I, complement C3/C4) remained stable throughout the study. Transient T-cell responses to the AAV2i8 capsid or I-1c were observed between weeks 4 and 12, coinciding with mild liver enzyme elevations, but waned to undetectable levels by month 12. In one patient who underwent left-ventricular assist device implantation 13 months post-infusion, myocardial tissue analysis confirmed high AAV2i8-I-1c transduction efficiency (1.19 vg dg-1) and restored phospholamban Ser16 phosphorylation comparable to non-failing controls, demonstrating in vivo target engagement in humans. These results demonstrate the feasibility, safety, and early biological activity of a one-time, intracoronary, cardiotropic gene therapy for advanced heart failure, providing the first clinical proof-of-concept for durable molecular restoration of calcium handling in the failing human heart (Figure 1). The vector's cardiotropism is the key to its success. Conventional AAV1 or AAV9 serotypes exhibit partial cardiac tropism but also infect liver and skeletal muscle, raising safety concerns. AAV2i8 achieves preferential uptake in cardiomyocytes through rational capsid engineering, combining the binding specificity of AAV2 with the efficient transduction of AAV8 1. This selective delivery concentrates therapeutic expression in the heart, enabling lower vector doses and minimizing systemic exposure. The evolution of recombinant AAVs toward tissue-specific precision underpins the renaissance of gene therapy in cardiovascular medicine 2. Beyond technical advances, AB-1002 exemplifies the translational trajectory from bench to bedside. Integrating such one-time genetic interventions with precision diagnostics and multi-omics profiling could revolutionize how heart failure is managed 4. The AB-1002 trial, though small, demonstrates the feasibility of in-situ myocardial gene repair as a complement, or even an alternative, to lifelong pharmacological therapy. At the same time, these findings should be interpreted in the context of a deliberately selected patient population typical of early-phase studies, which prioritizes safety and assessment of myocardial target engagement over broad generalizability. While the pathway targeted by AB-1002 is not expected to be strongly influenced by age or sex, the exclusion of ischemic cardiomyopathy defines an important boundary for extrapolation, given potential effects of myocardial scar and perfusion heterogeneity on gene delivery. Alongside these considerations, several challenges temper enthusiasm. Pre-existing neutralizing antibodies against AAV capsids could limit patient eligibility 2. Vector manufacturing remains costly and capacity-constrained, threatening scalability. Long-term durability of transgene expression and potential immune responses to I-1c or the vector require extended observation. Moreover, while short-term contractile improvement is promising, demonstration of survival or hospitalization benefit will demand larger randomized trials. These caveats underscore the importance of continued mechanistic studies and parallel innovation in vector design and immune modulation. Future iterations may incorporate self-regulating or tissue-responsive promoters, enabling titratable gene expression and personalized dosing. The same cardiotropic vector could also be adapted for other myocardial targets—SERCA2a re-delivery, S100A1 augmentation, or CRISPR-based correction of inherited cardiomyopathies 2. The modularity of the AAV2i8 platform thus positions it as a cornerstone for the next generation of regenerative cardiology. This transformation aligns with the mission of precision and translational medicine, which seeks to design therapies that not only prolong life but also restore physiological function. Nghia Phu Nguyen: conceptualization, writing – original draft. Phillip Tran: writing – review and editing, supervision. Both authors have read and approved the final manuscript. The authors have nothing to report. The authors received no specific funding for this work. The authors have nothing to report. The authors declare no conflicts of interest. The authors have nothing to report.